From Acute To Chronic Back Pain

From Acute to Chronic
Back Pain
This page intentionally left blank
From Acute to Chronic
Back Pain
Risk Factors, Mechanisms,
and Clinical Implications
Edited by
Monika I. Hasenbring
Department of Medical Psychology and Medical Sociology,
Faculty of Medicine,
Ruhr-University,
Bochum, Germany
Adina C. Rusu
Department of Medical Psychology and Medical Sociology,
Faculty of Medicine,
Ruhr-University,
Bochum, Germany,
Department of Psychology, Royal Holloway,
University of London,
London, UK
Dennis C. Turk
School of Medicine,
University of Washington,
Seattle, Washington, USA
1
1
Great Clarendon Street, Oxford ox2 6dp
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide in
Oxford New York
Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam
Oxford is a registered trade mark of Oxford University Press
in the UK and in certain other countries
Published in the United States
by Oxford University Press Inc., New York
© Oxford University Press, 2012
The moral rights of the authors have been asserted
Database right Oxford University Press (maker)
First published 2012
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
without the prior permission in writing of Oxford University Press,
or as expressly permitted by law, or under terms agreed with the appropriate
reprographics rights organization. Enquiries concerning reproduction
outside the scope of the above should be sent to the Rights Department,
Oxford University Press, at the address above
You must not circulate this book in any other binding or cover
and you must impose the same condition on any acquirer
British Library Cataloguing in Publication Data
Data available
Library of Congress Cataloging in Publication Data
Library of Congress Control Number: 2011943637
Typeset in Minion by Cenveo, Bangalore, India
Printed and bound by
CPI Group (UK) Ltd, Croydon, CR0 4YY
ISBN 978–0–19–955890–2
10 9 8 7 6 5 4 3 2 1
Whilst every effort has been made to ensure that the contents of this book are as complete, accurate and up-to-date
as possible at the date of writing, Oxford University Press is not able to give any guarantee or assurance that such is
the case. Readers are urged to take appropriately qualified medical advice in all cases. The information in this book is
intended to be useful to the general reader, but should not be used as a means of self-diagnosis or for the prescription
of medication.
Preface
Chronic back pain has been and continues to be a major cause of distress (both to people with
persistent pain and their significant others), disability, work loss, and costs to society. Moreover,
with the aging population, it is becoming even more prevalent and as a consequence is having an
escalating impact upon the healthcare systems and society as a whole worldwide. A significant
issue concerns understanding why, although the majority of people with acute back symptoms
recover in a reasonable time, a significant minority evolve into patients with chronic pain and
prolonged pain-related disability. Understanding the variables that contribute to chronicity could
serve as a basis for early intervention to prevent the downward spiral. A growing number of stud-
ies have been conducted designed to discover predictors of chronicity and clinical trials have been
initiated in an attempt to identify targets for intervention. To date there have been no volumes
that have attempted to compile this research in a single source or that integrate the results of
available studies in order to facilitate prevention and intervention in practice.
The identification of clinically relevant risk factors in low back pain has broad practical impli-
cations for the healthcare system globally. During the past 15 years, psychosocial risk factors and
psychobiological mechanisms have been identified as important risk factors and have led to the
development of early screening methods (e.g. ‘yellow flag’ diagnostics) and new psychosocial
interventions by targeting treatment modalities to patients’ particular characteristics and needs
(risk factor-based interventions for pain and pain-related disability). Research is evolving from
asking ‘What treatments work?’ to a set of inter-related questions: ‘What treatments are most
effective to people with what set of characteristics, provided when, on what outcome measures,
compared to what alternatives, and at what costs?’. However, substantial aspects of the pathway
from acute to chronic pain still remain unexplained. Recent neurobiological paradigms investi-
gating genetic, neurophysiological, and biomechanical processes elucidate important mecha-
nisms of chronic back pain, which represent important pathways from acute to chronic pain.
We hope that these paradigms will lead to the development of new pharmacological and non-
pharmacological treatment approaches, which might establish evidence that supports a compre-
hensive approach to assessment and treatment of back pain spanning the entire spectrum
from acute through prevention and treatment of chronic pain and disability. Early and more
appropriate interventions are needed to prevent long-term, disabling back pain with accompany-
ing socioeconomic consequences.
This book was conceived following a series of discussions at international conferences and
symposia about the future of evidence-based pain science and research between the three of us
several years ago. We felt that there was an absence of a single volume that integrated the large but
disparate body of knowledge of numerous specialties—medicine, psychology, and physiotherapy.
The major aim of the symposia that preceded the development of the current publication was to
present advances in basic pain research with a view to their relevance for the transition from acute
to chronic pain. Thus, the meetings presented an opportunity for some of the most prominent
pain scientists to present and critically discuss their current findings in an interdisciplinary set-
ting. These conferences proved to be extremely stimulating to all parties and it revealed that much
knowledge needs to be synthesized and transferred from research to the clinical practice. We hope
that this book will fill a void by translating basic pain research to clinical practice in back pain.
vi PREFACE
The volume should be of equal interest to clinicians from multiple and diverse specialties who are
involved in the treatment of back pain patients, as well as to pain researchers. Clinicians working
with back pain patients and wanting to understand more about the basic mechanisms underlying
back pain as well as novel developments in the clinical science will find a wealth of information in
this book. To our knowledge, no other book has exactly this focus, and we hope that it may
contribute to further increase the collaboration and the exchange of information between back
pain experts and basic pain scientists.
We focused on three main themes in conceptualizing this text: (1) the mechanisms involved in
the transition from acute to persistent pain; (2) the concept of treatment-relevant subgroups; and
(3) how available research evidence can inform the prevention management of acute and chronic
back pain. The volume brings together an internationally renowned group of contributors who
are recognized as experts in their fields. It is organized in six inter-related sections. In Part 1, we
introduced current developments in pain epidemiology. We also included chapters on recent
neurobiological paradigms investigating genetic, neurophysiological (Part 2), biomechanical
processes (Part 3), and psychosocial mechanisms including fear-avoidance, endurance, cognitive
processing, and significant others/behavioral (Part 4) that may represent important pathways
from acute to chronic pain. Part 5 delineates important advances in the practitioner’s role in the
process of care. Parts 6 to 9 summarize important advances to diagnostics and treatment of acute,
subacute, and chronic back pain. In these final sections we extended the approach of treatment-
relevant subgroups further and provide ideas on how to relate those findings to the prevention
management of acute, sub-acute, and chronic back pain and disability.
Multi-author books take some years from initial conceptualization to publication. Hopefully,
that wait will have been worthwhile, both for those readers who have been aware of impending
completion of the text and for the authors themselves. When drafting the first outline of the
current edited volume, we intended to develop a book that would stimulate discussion and offers
avenues of future investigation and collaboration. Publication of this volume should be viewed as
a status report and serve as a stimulus for additional research as there remains much that is not
understood and much needs to be learned to prevent chronic pain and pain-related disability.
Clinicians should find the materials as a useful guide for approach patients with acute, subacute,
and chronic pain but they need to continue to follow research developments to enhance their
approaches as evidence becomes available.
Adina C. Rusu
Dennis C. Turk
Bochum, Germany and Seattle, Washington, USA March 2011
Contents
List of Contributors xi
List of Abbreviations xv
Part 1 Current Developments in Epidemiology

1 Epidemiology of Back Pain, from the Laboratory to the Bus Stop:
Psychosocial Risk Factors, Biological Mechanisms, and Interventions in Population-
Based Research 3
Blair H. Smith, Nicola Torrance, and Gary J. Macfarlane
2 Defining Chronic Pain by Prognosis 21
Kate M. Dunn, Michael Von Korff, and Peter R. Croft
Part 2 Risk Factors of Chronic Back Pain and Disability:

Biological Mechanisms
3 Genetic Factors Modulating Chronic Back Pain 43
Julia Metzner and Irmgard Tegeder
4 Peripheral and Central Sensitization as Risk Factors of Low Back Pain 55
Hermann O. Handwerker
5 Dysfunction of the Hypothalamic–Pituitary–Adrenal Axis and Associated Stress
Axes in the Development of Chronic Low Back Pain 69
John McBeth and Andrea Power
6 Central Imaging of Pain and the Process of Chronicity 89
Sandra Kamping and Herta Flor
7 Structural Brain Changes in Patients with Chronic Back Pain 105
Arne May and A. Vania Apkarian
8 The Psychophysiology of Chronic Back Pain Patients 115
Kati Thieme and Richard H. Gracely

Biomechanical Mechanisms
9 Electromyographically-Determined Muscular Fatigue in Low Back Pain 155
Anne F. Mannion and David O’Riordan
10 Unmasking the Deconditioning Paradigm for Chronic Low Back Pain
Patients 185
Jeanine Verbunt, Rob J.E.M. Smeets, and Harriet Wittink
viii CONTENTS

Sociodemographic and Psychosocial Mechanisms
11 Screening of Psychosocial Risk Factors (Yellow Flags) for Chronic Back Pain
and Disability 203
Chris J. Main, Nicholas A.S. Kendall, and Monika I. Hasenbring
12 Dispositional Fear, Anxiety Sensitivity, and Hypervigilance 231
R. Nicholas Carleton and Gordon J.G. Asmundson
13 Processes Underlying the Relation between Catastrophizing and
Chronic Pain: Implications for Intervention 251
Michael J.L. Sullivan and Marc O. Martel
14 Fear-Avoidance as a Risk Factor for the Development of Chronic Back
Pain and Disability 269
Linda Vancleef, Ida Flink, Steven J. Linton, and Johan Vlaeyen
15 Endurance-Related Pain Responses in the Development of
Chronic Back Pain 295
Monika I. Hasenbring, Dirk Hallner, and Adina C. Rusu
16 Cognitive Processing and Self-Pain Enmeshment in Chronic Back Pain 315
Adina C. Rusu and Tamar Pincus
17 Significant Others in the Chronicity of Pain and Disability 339
Annmarie Cano and Laura Leong
18 Effects of Workers’ Compensation Systems on Recovery from
Disabling Injuries 355
James P. Robinson and John D. Loeser
19 Work-Related Risk Factors for Transition to Chronic Back Pain
and Disability 377
William S. Shaw, Glenn S. Pransky, and Chris J. Main
Part 5 Practitioner’s Role in the Process of Care

20 The Physician as Disability Advisor for Back Pain Patients 391
James Rainville, Glenn S. Pransky, Sarah Gibson, and Pradeep Suri
21 The Attitudes and Beliefs of Clinicians Treating Back Pain: Do They
Affect Patients’ Outcome? 405
Tamar Pincus, Rita Santos, and Steven Vogel
Part 6 Clinical Implications: New Approaches to

Diagnostics and Treatment
22 International Guidelines for the Diagnostics and Treatment of Acute,
Subacute, and Chronic Back Pain 419
Maurits van Tulder and Bart Koes
CONTENTS ix
Part 7 Clinical Approaches for Patients with Acute

and Subacute Low Back Pain
23 Engaging Patients in their Own Care for Back Care: The Role of Education
and Advice in the Prevention of Chronic Pain and Disability 433
Chris J. Main and Kim Burton
24 Motivational Issues in Pain Management 453
Robert D. Kerns, Mark P. Jensen, and Warren R. Nielson
25 Pharmacotherapy of Low Back Pain 471
Kay Brune and Bertold Renner
Part 8 Subgroup-Specific Approaches for Patients at

Risk For or With Chronic Pain
26 Reviewing the Concept of Subgroups in Subacute and Chronic Pain and
the Potential of Customizing Treatments 485
Adina C. Rusu, Katja Boersma, and Dennis C. Turk
27 Risk Factor-Based Cognitive-Behavioural Therapy for Acute and Subacute
Back Pain 513
Monika I. Hasenbring, Bernhard W. Klasen, and Adina C. Rusu
Part 9 Clinical Approaches for Patients with Established

Pain and Disability
28 Physical Exercise Interventions and Low Back Pain 531
J. Bart Staal, Chris G. Maher, and William S. Shaw
29 Contextual Cognitive-Behavioural Therapy for Chronic Pain (Including
Back Pain) 547
Lance M. McCracken
30 Rehabilitation Programmes to Prevent Severely Disabling Chronic
Back Pain 565
Michael K. Nicholas and Rob J.E.M. Smeets
Index 581
List of Contributors
A. Vania Apkarian Ida Flink

Department of Physiology Center for Health and Medical
Northwestern University Psychology
Feinberg School of Medicine Örebro University
Chicago, IL, USA Örebro, Sweden
Gordon J.G. Asmundson Herta Flor
Department of Psychology Department of Cognitive and
University of Regina Clinical Neuroscience
Regina, SK, Canada Central Institute of Mental Health
Katja Boersma University of Heidelberg
Örebro University Mannheim, Germany
Center for Health and Medical Sarah Gibson
Psychology (CHAMP) Meli Orhtopedics
Örebro, Sweden Ft Lauderdale, FL, USA
Kay Brune Richard H. Gracely
Department of Experimental and Clinical Center for Neurosensory Disorders
Pharmacology and Toxicology University of North Carolina
FAU Erlangen-Nuremberg Chapel Hill, NC, USA
Erlangen, Germany Dirk Hallner
Kim Burton Department of Medical Psychology
Spinal Research Unit and Medical Sociology
University of Huddersfield Faculty of Medicine
Huddersfield, UK Ruhr-University of Bochum
Annmarie Cano Bochum, Germany
Department of Psychology Hermann O. Handwerker
Wayne State University Department of Physiology &
Detroit, MI, USA Pathophysiology
R. Nicholas Carleton Friedrich-Alexander Universität
Department of Psychology Erlangen/Nürnberg
University of Regina Erlangen, Germany
Regina, SK, Canada Monika I. Hasenbring
Peter R. Croft Department of Medical Psychology
Arthritis Research Campaign and Medical Sociology
National Primary Care Centre Faculty of Medicine
Keele University Ruhr-University of Bochum
Keele, UK Bochum, Germany
Kate M. Dunn Mark P. Jensen

Arthritis Research Campaign Department of Rehabilitation
National Primary Care Centre Medicine
Keele University University of Washington
Keele, UK Seattle, WA, USA
xii LIST OF CONTRIBUTORS
Sandra Kamping Gary J. Macfarlane

Department of Cognitive and Aberdeen Pain Research Collaboration
Clinical Neuroscience Epidemiology Group
Central Institute of Mental Health University of Aberdeen
University of Heidelberg Aberdeen, UK
Mannheim, Germany Chris G. Maher
Nicholas A.S. Kendall The George Institute for
Health Services Consultant International Health
Surbiton, Surrey, UK The University of Sydney
Robert D. Kerns Sydney, Australia
Departments of Psychiatry, Neurology, Chris J. Main
and Psychology Arthritis Research Campaign
Yale University National Primary Care Centre
West Haven, CT, USA; and Keele University
VA Central Office, and Keele, UK
Psychology Service Anne F. Mannion
VA Connecticut Healthcare System Spine Center Division
West Haven, CT, USA Department of Research and Development
Bernhard W. Klasen Schulthess Klinik
Department of Medical Psychology Zürich, Switzerland
and Medical Sociology Marc O. Martel
Faculty of Medicine Department of Psychology
Ruhr-University of Bochum McGill University
Bochum; and Montreal, QC, Canada
Algesiologikum
Munich, Germany Arne May
Department of Systems Neuroscience
Bart Koes University of Hamburg
Department of General Practice Hamburg, Germany
ErasmusMC-University Medical
Center Rotterdam John McBeth
Rotterdam, The Netherlands Arthritis Research UK
Epidemiology Unit
Laura Leong University of Manchester
Department of Psychology Manchester, UK
Wayne State University
Detroit, MI, USA Lance M. McCracken
Health Psychology Section
Steven J. Linton Department of Psychology
Center for Health and Medical Psychology Institute of Psychiatry
Örebro University King's College London
Örebro, Sweden London, UK
John D. Loeser Julia Metzner
Departments of Neurological Surgery and Institute of Pharmaceutical Chemistry
Anesthesiology and Pain Medicine, Goethe-University,
University of Washington, Frankfurt, Germany
Seattle, WA, USA
LIST OF CONTRIBUTORS xiii
Michael K. Nicholas Adina C. Rusu

Pain Management Research Institute Department of Medical Psychology
University of Sydney at Royal North and Medical Sociology
Shore Hospital Faculty of Medicine
St Leonards, NSW, Australia Ruhr-University
Warren R. Nielson Bochum, Germany;
Department of Medicine (Rheumatology) Department of Psychology,
University of Western Ontario and Royal Holloway,
St Joseph’s Health Care University of London,
London, ON, Canada London, UK
David O’Riordan Rita Santos

Spine Center Division Department of Psychology
Department of Research and Development Royal Holloway
Schulthess Klinik University of London
Zürich, Switzerland London, UK
Tamar Pincus William S. Shaw

Department of Psychology Liberty Mutual Research Institute
Royal Holloway for Safety
University of London Hopkinton; and
London, UK University of Massachusetts
Medical School
Andrea Power Worcester, MA, USA
Human Pain Research Group
Hope Hospital Rob J.E.M. Smeets
Salford, UK Centre of Expertise in Rehabilitation and
Audiology
Glenn S. Pransky Adelante, Hoensbroek; and
Liberty Mutual Research Institute for Safety Department of Rehabilitation Medicine
Hopkinton; and Caphri, Maastricht University,
University of Massachusetts Medical School Maastricht University Medical Centre
Worcester, MA, USA Maastricht, The Netherlands
James Rainville Blair H. Smith
Department of Physical Medicine and Medical Research Institute
Rehabilitation University of Dundee
Harvard Medical School, Boston; Dundee, UK
New England Baptist Hospital
Boston, MA, USA J. Bart Staal
Scientific Institute for Quality of Healthcare
Bertold Renner Radboud University Medical Centre
Department of Experimental and Nijmegen, The Netherlands
Clinical Pharmacology and Toxicology
FAU Erlangen-Nuremberg Michael J.L. Sullivan
Erlangen, Germany Departments of Psychology, Medicine
and Neurology
James P. Robinson McGill University
Department of Rehabilitation Medicine Montreal, QC, Canada
University of Washington
Seattle, WA, USA
xiv LIST OF CONTRIBUTORS
Pradeep Suri Jeanine Verbunt

Department of Physical Medicine and Adelante, Center of Expertise in
Rehabilitation Rehabilitation and Audiology,
Harvard Medical School Adelante, Hoensbroek; and
Boston, MA USA; and Department of Rehabilitation Medicine,
VA Boston Healthcare System Maastricht University and Maastricht
Boston, MA, USA University Medical Center, Maastricht, the
Irmgard Tegeder Netherlands
Institute of Clinical Pharmacology Johan Vlaeyen
Goethe-University, Pain and Disability Research Program
Frankfurt, Germany University of Leuven
Kati Thieme Leuven, Belgium; and
Department of Medical Psychology Department Clinical Psychological Science
Philipps-University Marburg Faculty of Psychology and Neuroscience
Marburg, Germany Maastricht University
Maastricht, The Netherlands
Nicola Torrance
Medical Research Institute Steven Vogel
University of Dundee British School of Osteopathy,
Dundee, UK London, UK
Dennis C. Turk Michael Von Korff

Department of Anesthesiology Group Health Research Institute
University of Washington Seattle, WA, USA
Seattle, WA, USA Harriet Wittink
Linda Vancleef Research group Lifestyle and Health
Department Clinical Psychological Science Utrecht University of Applied Sciences
Faculty of Psychology and Neuroscience Utrecht, The Netherlands
Maastricht University
Maastricht, The Netherlands
Maurits van Tulder
Department of Health Sciences and EMGO
Institute for Health and Care Research
Faculty of Earth & Life Sciences
VU University
Amsterdam, The Netherlands
List of abbreviations
ABPS-MP Attitudes to Back Pain Scale for DTI diffusion tensor imaging
Musculoskeletal Practitioners EEG electroencephalography
ACT Acceptance and Commitment EER eustress-endurance response
Therapy
EMG electromyographic
ACC anterior cingulate cortex
FABQ Fear Avoidance Beliefs
ACTH adrenocorticotropin hormone Questionnaire for Health Care
ADL activity of daily living Practitioners
AEM avoidance-endurance model FAM fear-avoidance model
AEQ Avoidance-Endurance FAR fear-avoidance response
Questionnaire fMRI functional magnetic resonance
ALBP acute low back pain imaging
APAP acetaminophen (paracetamol) fMRI fibromyalgia
AS anxiety sensitivity FNE fear of negative evaluation
ASA acetylicsalicylacid FR flexion-relaxation
BBQ Back Beliefs Questionnaire GI gastrointestinal
BCI brain–computer interface GivE graded in vivo exposure
BDI Beck Depression Inventory GnIH gonadotropin inhibitory hormone
CA central augmentation GnRH gonadotropin-releasing hormone
CBA cognitive-behavioural approach GP general practitioner
CBP chronic back pain GWAS genome-wide association studies
CBT cognitive-behavioural therapy HCP healthcare professional
CCBT contextual cognitive-behavioural HC-PAIRS Health Care Providers Pain and
therapy Impairment Relationship Scale
CI confidence interval HPA hypothalamic–pituitary–adrenal
CLBP chronic low back pain HPG hypothalamic–pituitary–gonadal
COX cyclo-oxygenase HPGH hypothalamic–pituitary–growth
CPAQ Chronic Pain Acceptance hormone
Questionnaire IBCT integrative behavioural couple
CRH corticotrophin releasing hormone therapy
CSQ Coping Strategies Questionnaire IIS illness/injury sensitivity
CST coping skills training IMF initial median frequency
CT computed tomography IU intolerance of uncertainty
CV cardiovascular IW injured worker
CWP chronic widespread pain KPI Kiel Pain Inventory
DER distress-endurance response LP lumbopelvic
DHEA-s dehydroepiandrosterone-sulphate MAAS Mindful Attention Awareness Scale
DLBP disabling low back pain MEG magnetoencephalographic
DLPFC dorsolateral prefrontal cortex MET metabolic equivalent
DNIC diffuse noxious inhibitory control MF median frequency
DRAM Distress Risk Assessment Method MHPG 3-methoxy-4-hydroxyphenylglycol
xvi LIST OF ABBREVIATIONS
MMPI Minnesota Multiphasic Personality PPI proton-pump inhibitors

Inventory PPV positive predictive value
MPF mean power frequency PSOCQ Pain Stages of Change
mPFC medial prefrontal cortex Questionnaire
MPI Multidimensional Pain Inventory RCBI risk factor-based cognitive-
MPRCQ Multidimensional Pain Readiness to behavioural interventions
Change Questionnaire RCT randomized controlled trial
MRI magnetic resonance imaging RER respiratory exchange ratio
NPV negative predictive value RISC-BP RIsk SCreening of Back Pain
NSAID non-steroidal anti-inflammatory ROC receiver operator curve
drug ROM range of motion
NSAID non-steroidal, antiphlogistic, RTW return to work
inflammatory drugs
SBDT STarT Back Decision Tool
ÖMPSQ Örebro Musculoskeletal Pain
S-CST spouse-assisted coping skills training
Screening Questionnaire
sEMG surface electromyography
PAG periaqueductal grey
SEMP Schema Enmeshment Model
PAIRS Pain and Impairment Relationship
of Pain
Scale
SLC sick leave certification
PAL physical activity in daily life
SNAP Schedule for Nonadaptive and
PAL physical activity level
Adaptive Personality
PARIS-CBA pain risk screening-based
SSRI serotonin reuptake inhibitors
cognitive-behavioural approaches
SST Self-System Therapy
PASS Pain Anxiety Symptoms Scale
TMD temporomandibular disorder
PASS-20 Pain Anxiety Symptoms Scale-20
TSK Tampa Scale for Kinesiophobia
PCS Pain Catastrophizing Scale
TSS Thought Suppression Scale
PDP Pain-Disability Prevention
[Programme] TTM Transtheoretical Model of
Behaviour Change
PET positron emission tomography
UDE unwanted drug effects
PFC prefrontal cortex
VO2 maximum oxygen intake
PHODA Photograph Series of Daily Activities
WC workers' compensation
PMS Positive Mood Scale
WHO World Health Organization
POMC proopiomelanocortin
Part 1
Current Developments in
Epidemiology
Chapter 1
Epidemiology of Back Pain, from

the Laboratory to the Bus Stop:
Psychosocial Risk Factors, Biological
Mechanisms, and Interventions in
Population-Based Research
Blair H. Smith, Nicola Torrance, and
Gary J. Macfarlane
1.1 Introduction
1.1.1 Goals of epidemiology—improving health
Epidemiology is defined by MacMahon and Pugh (1970) as ‘the study of the distribution and
determinants of disease frequency in man’. It has its origins in the early days of modern medicine
and continues to develop its science and expand its roles. In recognition of this, Porta et al.’s
Dictionary of Epidemiology (2008) extends the definition to imply greater clinical relevance, and
describes ‘the study of the occurrence and distribution of health-related states or events in specific
populations, including the study of the determinants influencing such states, and the application
of this knowledge to control health problems’ (Porta et al. 2008). It is the latter part of this defini-
tion that makes epidemiology an important clinical discipline, particularly for chronic low back
pain (CLBP), an important purpose being to inform the provision of health services. Rothman
and Greenland (1998) suggest that the ultimate goal of epidemiology is to identify the causes of
disease, but good epidemiological data on chronic pain will also provide valuable information on
a wide range of important areas (Box 1.1).
Epidemiological research should, therefore, have as one of its ultimate goals the development
of effective healthcare interventions, through an understanding of the distribution and determi-
nants of illnesses (Porta et al. 2008). The last 20 years has seen a large number of population-
based epidemiological studies of chronic pain, including back pain, and these have provided
important information for educational and clinical service resource requirements, and for pre-
vention and management strategies (Verhaak et al. 1998; Manchikanti 2000; Harstall and Ospina
2003). These studies have consistently demonstrated that chronic back pain is a highly prevalent
condition, with an important detrimental impact on individuals’ health, the healthcare services
and society, and for which successful treatment outcomes are difficult to achieve.
CLBP is a heterogeneous group of clinical conditions, a minority of which are associated with
a specific, demonstrable structural disorder (such as lumbar disc degeneration or ankylosing
spondylitis), and many of which coexist with chronic pain at other anatomical sites. (Smith et al.
2004a) The International Association for the Study of Pain classifies well over 100 different diag-
nostic categories of (low) back pain, though good evidence for the epidemiology of most of these
4 FROM ACUTE TO CHRONIC BACK PAIN
Box 1.1 Beneficial outcomes of epidemiological study of

chronic low back pain (CLBP)*
Good epidemiological data on CLBP will provide information on:
◆ Its distribution—this will allow an understanding of the profile of CLBP in the community.
◆ Its aetiology (determinants)—clinical or demographic factors that are associated with the
presence of CLBP will be revealed by epidemiological study, as will the identification of
factors which lead to or favour chronicity of pain.
◆ Possible preventive measures—having an awareness of aetiological factors or associations
will allow consideration of interventions at an early stage of the development of CLBP,
either to prevent chronicity or to minimize its impact.
◆ Improving prognosis—even if CLBP cannot be prevented, knowledge of the factors associ-
ated with the development of severe CLBP should inform the clinical management of
the condition, thereby possibly limiting severity and minimizing disability. This would be
best informed by longitudinal studies of CLBP to distinguish between determinants
and effects.
◆ The impact on quality of life—this will provide an understanding of the importance of
CLBP, as well as knowledge of the factors which are associated with high levels of physical,
psychological or social adverse effects.
◆ Definition and classification—if CLBP can be classified in a clinically relevant way, manage-
ment strategies can be targeted at subgroups in greatest need of treatment or with the
greatest likelihood of improvement, while individuals with less severe CLBP can be identi-
fied with a view to prevention of exacerbation.
◆ The evaluation of treatment strategies—until the distribution, determinants, impact, and
natural history of CLBP are understood, it is impossible to properly evaluate any interven-
tion aimed at improving chronic pain.
◆ Allocation of health service resources—ideally this should be informed by robust epidemio-
logical data. With a condition of the importance of CLBP, it is therefore crucial that
research information is available for health service planning.
◆ Allocation of educational resources—as with financial and clinical resources, appropriate
education of professionals and patients can be greatly assisted by epidemiological study.
* Smith et al. 1996.
remains elusive, and the great majority (90%) of back pain cases have no identifiable cause in any
case (Manek and MacGregor 2005). Given that the majority of CLBP is managed in primary care,
if anywhere (Croft et al. 1998), where neither the facilities nor expertise generally exist to render
such specific diagnosis feasible or appropriate, it is often practically necessary to consider low
back pain (LBP) as a single global entity (after excluding ‘red flag’ diagnoses such as fracture and
malignancy (Royal College of General Practitioners 1999)). This is also true of population-based
epidemiological research, which generally relies on questionnaires or interviews administered by
non-specialists. A site-specific case definition, perhaps enhanced by a criterion referring to back
pain-related disability (Dionne et al. 2008) is therefore the most frequently encountered classifi-
cation of back pain, both clinically and epidemiologically. Despite this lack of specificity in case
definition, there is remarkable consistency in studies across the world and in different population
EPIDEMIOLOGY OF BACK PAIN 5
groups, on the risk factors (particularly mechanical, psychological, and social) and therefore
prospective intervention targets for CLBP.
We therefore need generic treatment strategies, including pharmacological and non-
pharmacological approaches that can address CLBP as a global entity. This should be reflected in
the epidemiological approach to seeking aetiological and risk factors, while also seeking a deeper
understanding of sub-groups of CLBP that might benefit from more specific, targeted therapeutic
approaches. This chapter will review the outcomes of this approach to the chronic pain contin-
uum, and consider the potential rewards of adding measurement of biological factors to the
knowledge and methods developed to date.
1.1.2Recent success of epidemiology in improving understanding

of CLBP
In studies of chronic pain in the general population and in primary care, prevalence estimates
of back pain vary, and it is generally acknowledged that the quality of the data is variable, based
on the primary purpose of the investigation, the population sampled, and the definition of
back pain.
From cross-sectional surveys, point prevalence estimates of LBP range from 12–35% in the
general adult population, with lifetime incidence ranging from 49–70% (Andersson 1999). The
majority of people who experience an episode of back pain recover quickly without residual func-
tional loss, and most of these episodes, appropriately, are never brought to the consulting room
(Papageorgiou et al. 1996). Overall it is estimated that 60–70% of patients recover by 6 weeks, and
80–90% by 12 weeks (Andersson 1999). However, recovery after 12 weeks is slow and uncertain,
and fewer than half of those individuals disabled for longer than 12 months return to work, and
after 2 years of absence from work, the return to work rate is close to zero (Andersson 1999).
CLBP is not just ‘the same as acute back pain lasting longer’, but the result of a complex interplay
of physical, psychological, and social factors in tandem with the onset of neck pain (Jayson 1997).
Despite the above data on recovery, Croft et al. (1998) found that the great majority (75%) of
those consulting with LBP in primary care remain disabled and in pain one year after consultation.
Although an inception cohort in a more recent Australian study found a higher recovery rate,
around one-third still had not recovered (Henschke et al. 2008) after one year, and therefore back
pain for which healthcare is sought must be considered a serious condition, with epidemiological
approaches that address this as CLBP.
1.1.3 Defining LBP in epidemiology studies

Standardized definitions of LBP, as distinct, global clinical conditions, have been proposed to
allow comparisons of data derived from epidemiological prevalence studies (Dionne et al. 2008)
(Box 1.2), recognizing the above argument that, although LBP encompasses a heterogeneous
group of conditions, there is merit in grouping them together for epidemiological study. The
main aim in reaching consensus on definitions is to improve the validity of future comparison of
LBP prevalence figures for different countries, age groups, settings and occupational groups,
among others. (Dionne et al. 2008).
1.2 Risk factors for LBP already established in population-based

research
Large epidemiological studies have contributed considerably to our knowledge about potentially
modifiable factors associated with the onset and outcome of developing a new episode of back
Box 1.2 Proposed standardized definitions of low

back pain (LBP)*
The ‘minimal’ definition for use in epidemiological prevalence studies, where there are likely
to be many constraints, is ascertained by positive responses to two questions:
◆ ‘In the past four weeks, have you had any pain in your lower back?’
◆ ‘If yes, was this pain bad enough to limit your usual activities or change your daily routine
for more than one day?’
The ‘optimal’ definition, for use in focused studies where the investigators have space or time
for multiple questions, is derived from the minimal definition plus add-on questions covering:
◆ Frequency and duration of symptoms.
◆ Additional measures of severity, sciatica, and exclusions.
* Dionne et al. 2008.
pain, and our understanding of the influences on whether such an episode becomes chronic. At a
population level, these known risk factors include physical, psychological, and social variables
and some (though not all) are amenable to medical intervention. The presence of LBP is associ-
ated with demographic factors including female gender, older age, lower social class, as well as
lifestyle factors, including lack of physical activity, obesity, and smoking (Elliott et al. 1999;
Manchikanti 2000; Macfarlane 2006). Employment status and occupational factors have also
identified the role of perceived workload (Linton 2005), mechanical load, posture and whole
body vibration (Lis et al. 2007; see also Chapter 19). Factors that are not modifiable can nonethe-
less point to other productive lines of research. For example, the consistent finding of female
preponderance in all chronic pain conditions has led to hypothesized endocrine (Macfarlane
et al. 2002) and genetic (Mogil 2003, 2004) explanations that illuminate our understanding of
pain pathways in general, and may inform interventions. The most important clinical risk factor
for chronic pain at a specific site appears to be pain: either non-chronic pain (Croft et al. 1998;
McBeth et al. 2001; Smith et al. 2004b) or chronic pain at other anatomical sites (Bergman et al.
2002; Papageorgiou et al. 2002; Smith et al. 2004b). The more severe the pain, and the greater the
number of pain sites, the more likely is subsequent severe chronic pain (Bergman et al. 2002;
Elliott et al. 2002). This highlights the need to intervene effectively with all presentations of pain.
Poor general health (Elliott et al. 2002; Smith et al. 2004b), and a family or personal history of
other related conditions (‘functional somatic syndromes’) (Clauw and Chrousos 1997; Wessely
et al. 1999; Gran 2003; Bergman et al. 2002) are also risk factors for chronic pain. Known impor-
tant psychosocial risk factors implicated in the transition to CLBP include psychological distress,
pain catastrophizing, depressive mood, and more severe, persistent, or disabling symptoms at
presentation, and to a lesser extent somatization (Pincus et al. 2002; Linton 2005; Henschke et al.
2008). Similarly, coping style, in particular passive coping, has been shown to be associated with
increased risk of chronicity (Pincus et al. 2002; Jones et al. 2006). One of the strongest markers for
developing chronic pain is illness behaviour, specifically people who have a history of frequent
consultation and onward referral to specialist care. It could be hypothesized that this identifies
individuals with a predisposition to somatization with high health anxiety. However, health
anxiety has not been demonstrated to be a good predictor of the onset of chronic pain and the
association with illness behaviour has been reported to be independent of markers of somatiza-
tion (McBeth et al. 2003). It is likely that many of these risk factors or risk markers are caused by
earlier experiences of chronic pain, (Jones et al. 2007) as well as causing future chronic pain,
(Magni et al. 1993) and the concept of ‘bi-directional aetiology’ is important (Von Korff 1999), as
is the possibility of a shared common aetiological mechanism
In summary, we now have considerable knowledge (albeit incomplete) about the aetiology of
LBP—in particular the role of mechanical factors, psychological factors, and the role of the social
environment as well as beliefs and behaviours. Accordingly, there have been efforts to address
these with the development and testing of clinical interventions specifically targeting those we
believe to be risk factors. We will now consider how this knowledge has been translated into
management strategies in primary care.
1.3 Evidence on management for LBP in primary care

Community-based psychosocial interventions and physical therapies modelled on established
risk factors have been found, at best, to be of marginal benefit in reducing LBP (UK BEAM Trial
Team 2004; Hay et al. 2005; Heymans et al. 2005; Jellema et al. 2005b; Von Korff et al. 2005)
though some have been effective in highly selected secondary care samples, and there are exam-
ples of interventions that have improved attitudes towards chronic back pain (Williams et al.
1996; Morley et al. 1999; van Tulder et al. 2000; Buchbinder and Jolley 2004).
For non-pharmaceutical interventions, the best evidence for effective management is that
advice to remain active improves both short-term and long-term functions (Hagen et al. 2005).
However, a systematic review found that strengthening and stabilizing exercises alone with other
conservative interventions showed only small benefits in either short-term pain relief or func-
tional outcomes (Hayden et al. 2005). Similarly, trials of spinal manipulation, (Assendelft et al.
2004) when compared to other traditionally advocated therapies, showed no differences in effec-
tiveness of treatment (for more information see Chapter 25).
Research over the past decade which has emphasized the importance of psychosocial factors in
the development of chronic or recurrent back pain (Pincus et al. 2002; Linton 2005; van der
Windt et al. 2008) has led to recent interventions looking to test the effectiveness of either behav-
ioural or multidisciplinary treatment programmes aimed at tackling these psychological and
psychosocial factors. A number of trials have been conducted in primary care settings, in which
primary healthcare professionals were trained to provide psychosocial interventions (Frost et al.
2004; Hay et al. 2005; Jellema et al. 2005b). All of these studies have included an active ‘control’
treatment arm to represent ‘usual or best care’ and include: a trial of a brief pain management
programme in primary care compared with physical treatment (so called hands-on versus hands-
off physiotherapy) (Hay et al. 2005); one session of assessment and advice versus routine physi-
otherapy (Frost et al. 2004); and another trial that compared a 20-minute consultation with
general practitioners (GPs) who were trained to identify and discuss any potential psychosocial
barriers to recovery with usual care (Jellema et al. 2005b). In a number of trials in primary care,
providers (physiotherapists, nurses, or GPs) have been trained to deliver an intervention that uses
a cognitive behavioural approach to the psychosocial management of back pain (Frost et al. 2004;
UK BEAM Trial Team 2004; Hay et al. 2005; Jellema et al. 2005b; van der Windt et al 2008) with,
at best, only marginal benefit.
Despite the considerable knowledge accrued from large population-based studies, the
results from trials of management in the community/primary have proved disappointing, with no
or only very modest improvements in patient outcome when compared with usual care/active
control treatments. Before concluding that previously reported interventions did not work
we have to understand better why current trials appear not to be working. A number of possi-
ble explanations for these ‘negative’ findings have been proposed. This could be because our
information on psychosocial factors is still insufficient, because the interventions need better
targeting, and/or because we need to consider other definitions and measures of successful
outcomes of treatment (such as return to work, or patient-determined outcomes), and evalua-
tions of the delivery of interventions (Macfarlane et al. 2006). For example, Jellema et al (2005a)
showed some changes in GP behaviour and patient perceptions as a result of a psychosocial
intervention, but that GPs rarely identified the psychosocial factors at which to target the inter-
vention. In a systematic review of prospective studies, Hartvigsen et al. (2004) found relatively
weak evidence for any single type of psychosocial factor being an important determinant of future
onset of LBP. Research could look to identify which clusters of psychosocial factors, rather than
picking out individual factors, influence LBP onset and outcome and their mechanism(s)
of influence, in order to target intervention studies to population subgroups exhibiting these
clusters.
Recent evidence suggests that it may be difficult to distinguish specific treatment effects in trials
of complex interventions for chronic pain (i.e. the precise intervention under evaluation) from
non-specific treatment effects (such as the duration of consultation, or the interest shown in the
patient) (Newton-John and Geddes 2008). The latter may actually be more important than the
former, with the result that any specific effect is overshadowed by the effects of treatment given to
both the control arm and the treatment arm of a randomized controlled trial. Further research is
beginning to suggest useful ways of identifying and harnessing these non-specific effects (Newton-
John and Geddes 2008; van der Windt et al. 2008).
In interventions involving specific techniques, such as cognitive behavioural approaches, it is
important that consistent messages are delivered by all care providers involved. Future trials
should look to implement methods to measure skills and competencies of care providers, the
actual content and delivery of the provided treatment/intervention, and adherence to protocols
by patients and care providers. One way of tackling these issues may involve utilizing a ‘mixed
methods’ approach, for example, combining a detailed qualitative content analysis of video-taped
recordings of the intervention, together with conventional quantitative outcome measures (Lamb
et al. 2007). Further investigation of what participants understood about the programme is also
warranted (Lamb et al. 2007; Lambeek et al. 2007).
The outcome of an intervention is likely to be dependent upon several important, pre-existing
patient factors, and it is important to consider the tailoring of interventions to individuals, or
targeting subgroups of individuals with specific interventions, depending on either clinical factors,
sociodemographic characteristics, or the type and extent of risk factors (Hay et al. 2008). A retro-
spective attempt to address this by secondary analysis of a large clinical trial of physical therapy,
however, found the intervention still to be only marginally effective (Underwood et al. 2007).
Patient preference for different interventions may also exert an effect on outcome and there is
some evidence that patients’ expectations and preferences may modify the outcome of treatment
(Linde et al. 2007). New trial designs, such as ‘enriched enrolment, randomised withdrawal’
(McQuay et al. 2008) may be appropriate for addressing this in the future.
Further epidemiological, methodological, and clinical research would be valuable in each
of these areas, as well as in refining and validating the measures we have for identifying high-
risk groups, particularly psychosocial measures. Alternatively, or in addition, attention could
focus on the identification and assessment of other risk factors for CLBP, or on approaches
to assessing known variables that have been identified as potential risk factors. Taken along
with our existing knowledge, this new work could inform the design or development of
new interventions for application in the community, where most CLBP arises and is managed
(Smith 2001).
1.4 Potential biological risk factors hypothesized from

existing research
Epidemiological study has therefore been most successful in identifying the psycho-social
components of the bio-psycho-social model of CLBP, though this has yet to be converted into
effective interventions at a population level. We should turn our attention to the biological
component, therefore, with a view to enhancing our aetiological understanding of CLBP in the
community, and, potentially, the outcome of interventions. Studies in animals, and relatively
small studies in humans, have proposed important biological mechanism in the development of
CLBP that warrant further exploration in population-based research.
1.4.1 Neurological
Our understanding of the neurological mechanisms of acute and chronic (back) pain continues
to advance, rapidly in recent years as a result of developing technology. Neurological imaging
techniques, including functional magnetic resonance imaging (fMRI), now allow new views of
the central processing of pain, and have demonstrated features specific to individuals who have
chronic pain conditions (Gracely et al. 2002; Borsook and Becerra 2003; Cook et al. 2004). Perhaps
most importantly for the current discussion, objective neurological evidence of lower pain thresh-
olds and higher sensitivities has been found in individuals with chronic pain compared to those
without (Sorensen et al. 1998; Banic et al. 2004). A detailed systematic review of studies of brain
mechanisms in pain perception and regulation demonstrated the existence of a ‘brain network’
for acute pain perception, with many studies revealing the anatomical locations of network com-
ponents, and postulated the nociceptive system as a separate sensory system (Apkarian et al. 2005,
see also Chapters 7 and 8 for more detail). It also found consistent differences in the activation of
this network, in response to acute pain stimuli, between those who had chronic pain and those
who had not (Apkarian et al. 2005). Important similarities and differences were demonstrated in
the parts of network involved in different chronic pain conditions, such as fibromyalgia, cardiac
pain, neuropathic pain, and different types of headache. With reference to chronic back pain,
imaging and neurochemistry studies have identified reduced activity and grey matter density in
the prefrontal cortex and thalamus areas of the brain compared with those without chronic back
pain (Apkarian et al. 2004; Apkarian et al. 2005; Baliki et al. 2006, 2008; Grachev et al. 2000). It is
claimed that these variations account for up to 80% of the variance in duration and intensity of
chronic pack pain (Baliki et al. 2008), whereas the social and demographic factors described
above account for a much smaller proportion. The reduction in grey matter volume was equiva-
lent to between 10 and 20 years of normal ageing, with the amount of loss correlating positively
with the duration of back pain (Apkarian et al. 2004). Different patterns of brain involvement
were found between neuropathic and non-neuropathic back pain (Apkarian et al. 2004) and
between back pain and other causes of chronic pain (Apkarian et al. 2005; Baliki et al. 2008).
Extending this further, some of these activity patterns were reversed with analgesic treatment
(Baliki et al. 2008), corresponding with symptomatic relief of chronic back pain. These findings
provide some clues as to mechanisms of CLBP, and to its comorbidities, and may be helpful in
the design of targeted treatment strategies. The prefrontal cortex areas involved are also those
associated with certain emotional experiences, such as anxiety and depression, in whose presence
the patterns of activation differ (Grachev et al. 2001, 2002, 2003). They are also co-located
with regions concerned with the emotional view of the self, where a persistently negative response
to suffering, and consequent maladaptive behaviour can be initiated (Melzack and Casey
1968; Baliki et al. 2006). These findings support the hypothesis (above) of shared aetiological
mechanisms between these clinical conditions. The neurological findings also raise the possibility
of biomarkers for chronic back pain or objective markers of response to treatment. These could
be in the form of imaging (Baliki et al. 2008) or neurochemistry (Grachev et al. 2000).
It is important to note that the studies contributing to our understanding of neural mecha-
nisms of CLBP have, for excellent practical reasons, been cross-sectional, on small sample sizes.
As such, they need confirmation in other populations and in larger samples, where their impor-
tance at a population level can also be assessed. Furthermore, they cannot distinguish between
cause and effect, and we cannot tell whether altered brain activity patterns precede the onset of
back pain, whether they arise as a result of chronic back pain, or whether other factor(s), such as
lifestyle are the cause of both. This understanding can only come from well-designed longitudinal
studies.
1.4.2 Endocrine and immunological

Disorders of the hypothalamic–pituitary–adrenal (HPA) axis have been implicated in both
regional musculoskeletal and chronic widespread pain (CWP) (Vaeroy 1989; Griep et al. 1998;
Haddad et al. 2002) although the precise aspects of function associated with symptoms vary
between studies. Other research has demonstrated that, amongst people free of CWP but at high
risk of its development (based on known risk markers), aspects of HPA axis function, such as loss
of diurnal rhythm and failure to suppress cortisol secretion after an overnight dexamethasone
test, increased the risk of future onset of CWP (McBeth et al. 2005). It is not clear how this may
relate to back pain, but patients with CLBP and an abnormal dexamethasone suppression test
responded better to a high-intensity aerobic exercise programme in a recent randomized control-
led trial (Chatzitheodorou et al. 2008). Abnormalities of HPA axis function have also been found
in conditions such as depression and fatigue, (Gaab et al. 2005; Jerjes et al. 2005, 2006; Smith et al.
2006) both common comorbidities among people with chronic (back) pain. An intriguing
hypothesis linking these data with previous observations of the possible role of early life factors
(Jones et al. 2007) is that such adverse or traumatic life events may alter the stress response, which
then makes individuals vulnerable to future adverse events. Testing such a hypothesis requires
large long-term studies.
Cytokines and neuropeptides have been implicated in chronic pain conditions including back
pain (MacGregor et al. 2004; Manek and MacGregor 2005) neuropathic pain (Abbadie et al.
2003), polymyalgia rheumatica and rheumatoid arthritis (Pountain et al. 1998), and fibromyalgia
(Staud 2004). These generally reflect inflammatory processes, and show certain consistency in
the biochemical profile elicited between these pain conditions, including important roles
for substance P and the interleukins (Omoigui 2007). Although, in theory, this information
could lead to the development of biomarkers to promote an understanding and diagnosis of con-
ditions such as disc degeneration, results have so far been disappointing, with a comprehensive
search for candidate markers in spinal fluid lavage yielding no measurable cytokines (Scuderi
et al. 2006).
Noradrenaline, dopamine, and serotonin metabolites have also been found to have important
roles in chronic widespread pain and fibromyalgia, as have neuropeptides such as substance P,
nerve growth factor, and dynorphin A (Russell et al. 1992, 1998). In a prospective study of 67
adults with acute back pain, Hasselhorn et al. (2001) found that low serum levels of 3-methoxy-
4-hydroxyphenylglycol (MHPG), dehydroepiandrosterone-sulphate (DHEA-s), and beta-
endorphin each predicted persistent disability after 6 months. They postulated that these links
with stress response, anabolism, and the opioid system could help to explain the known associa-
tions with some of psychological variables described above. However, as with studies of
neurological mechanisms of back pain, these findings need to be confirmed and assessed in large,
population-based studies, which would ideally be prospective in order to distinguish the cause/
effect relationship. The precise nature of the role, and the overall importance of these endocrine
factors in the development of chronic back pain therefore remain poorly understood, but
initial results suggest that this could be an important area for further research (Staud 2004), for
understanding the biology of pain, for developing biomarkers and treatment biomarkers, and
measuring the response to these treatments (see also Chapter 5).
1.4.3 Genetic
There are three questions to address, and therefore areas of research in the consideration of
genetic risk factors for CLBP: (1) is CLBP a heritable condition? (2) Is there a genetic basis to pain
mechanisms generally? (3) Is there a genetic basis to specific causes of CLBP? Despite many recent
advances in both genetics and pain research, approaches to these questions are still in their rela-
tive infancy. When addressing the last of these, and despite the comments above, this is one area
when the consideration of CLBP as a single global entity may be less helpful than its consideration
as a group of specific disorders, each of which has distinct pathophysiology, a distinct molecular
origin, and therefore distinct genetic associations. However, there is likely to be an important
interaction between the molecular origins of any back pain condition, and the perception of this
as an individually important and disabling condition. The latter perception may be more vulner-
able to generic factors, including genetic and environmental factors, and the picture is likely to be
complex (see also Chapter 3 in this volume).
There is some evidence that chronic pain (Edwards et al. 1985) and back pain (Matsui et al.
1997; Postacchini et al. 1998) cluster in families, suggesting a heritable component. While much
of this may be due to social, psychological, and behavioural factors within families (a ‘maternal
effect’), (Lau et al. 1990; Terre and Ghiselli 1997; Hotopf et al. 1998; MacGregor et al. 2004), stud-
ies comparing monozygotic and dizygotic twins have identified a specific genetic contribution to
back pain (Sambrook et al. 1999). Depending on the definition of back pain applied, heritability
estimates were found in more recent twin studies to range from 30–46% (Battié et al. 2007), with
perhaps as much as 57% of the variance in population liability to severe disabling back pain
attributable to genetic factors (MacGregor et al. 2004). Heritability was found to be greater
in more severe back pain (Battié et al. 2007), but the genetic contribution to decrease with increas-
ing age (Hartvigsen et al. 2004). MacGregor et al. (2004) found considerable overlap between the
genetic predictors of back pain and of psychological well-being, suggesting a partially common
genetic aetiology that requires further exploration. In contrast to these studies, however, an ear-
lier twin study (MacGregor et al. 1997) found no evidence of a genetic component to pain thresh-
old, ascribing the variability to shared environmental influences. Other research has found no
evidence of familial aggregation of recurrent pain complaints (Koutantji et al. 1998; Borge and
Nordhagen 2000), or of back pain (Hartvigsen et al. 2004).
Biologically, there is evidence in mice of genetic transmission of nociception (Mogil et al.
1999), and there is growing evidence of specific genetic contributions to some pain conditions in
humans (Mogil 2004; Diatchenko et al. 2006; Limer et al. 2008; Max and Stewart 2008). This
suggests that at least some components of the pain pathway are subject to the effects of measur-
able genetic variation, irrespective of the cause of the pain. This makes sense given the central role
for molecular mechanisms, either as originators of pain sensations or as their modulators in
interaction with external, environmental factors (Diatchenko et al. 2006). In addition to provid-
ing valuable insights into mechanisms of pain processing, specific polymorphisms in some of
these genes may be associated with the development or perception of chronic pain. These include
the genes coding for μ-opioid receptors (Yu 2004), serotonin receptors 1A and 1B (Bondy et al.
1999; Pata et al. 2004), and transporter (5-HTT) (Herken et al. 2001; Szilagyi et al. 2006), and
val-158-met catecholamine-o-methyltransferase (COMT) (Gursoy et al. 2003; Zubieta et al.
2003). The list of genes whose variants are purported to be involved in human pain conditions,
either as single nucleotide polymorphisms or with haplotype variants, is constantly growing, but
includes those listed in Box 1.3. In addition, genetic polymorphisms of the cytochrome P450
enzyme system (Wolf et al. 2000; Fishbain et al. 2004) and of opioid receptors (Mogil 2004) are
likely to be important in determining responses to pharmacological pain treatment, and warrant
epidemiological study. The latter are examples of how genetic research might directly influence
the treatment of chronic pain conditions, for example, by targeting drug treatment on an indi-
vidual basis, optimizing its effectiveness on the basis of a known genetically-tested profile relating
to the relevant cell receptors and metabolic pathways (‘personalized medicine’). This is an impor-
tant aim of research in this area (Max and Stewart 2008), though its transfer to clinical practice is
still some way off.
With respect to the genetics of conditions that cause CLBP, most evidence relates to causes of
lumbar disc degeneration. Inherited lumbar disc degeneration was recently confirmed to be an
important cause of CLBP (Battié et al. 2007), with up to one-quarter of the genetic effect on back
pain explicable by genetic causes of disc degeneration. The mode of inheritance is likely to be
complex, involving multiple factors including multiple genes (Kalichman and Hunter 2008).
Box 1.3 Genes implicated in human pain conditions*

◆ 5-HTR2A ◆ IL10
◆ ABCB1 or MDR1 ◆ IL1A, IL1B, and IL1RN
◆ ACE ◆ IL4
◆ ADRA2A, ADRA2C ◆ IL6
◆ ADRB2 ◆ LTA
◆ ATP1A2 ◆ MAOA
◆ CACNA1A ◆ MC1R
◆ COMT ◆ MTHFR
◆ CYP2D6 ◆ NGFB
◆ DRD2 ◆ NOS3
◆ DRD4 ◆ NTRK1 or TRKA
◆ GCH1 ◆ OPRM1 or MOR
◆ HSN2 ◆ SCN1A
◆ HTR2A ◆ SCN9A
◆ HTR3A ◆ SERPINA1
◆ HTR3B ◆ SLC6A3 or DAT1
◆ IFNB2 ◆ SLC6A4, or 5-HTT, or SERT
◆ IKBKAP or IKAP ◆ SPTLC1 or SPT1
◆ TNF
* Diatchenko et al. 2007; Limer et al. 2008.
Polymorphisms of the genes coding for interleukin 1 (IL-1) (Solovieva et al. 2004a, 2004b),
vitamin D receptor (Jones et al. 1998; Videman et al. 1998; Kawaguchi et al. 2002), IX collagen
(Annunen et al. 1999; Paassilta et al. 2001), and aggrecan (Kawaguchi et al. 1999) have been
found, among other polymorphisms, to be associated with lumbar disc degeneration and chronic
back pain (Ala-Kokko 2002). The association between ankylosing spondylitis and the HLA B27
gene has been long established, and links between back pain and other genes coding for inflam-
matory markers are also apparent (Valdes et al. 2005), but the biological mechanisms behind
these associations remain unclear.
Most of the evidence associating (back) pain conditions with genetic polymorphisms is based
on small twin or case–control studies, and further confirmatory evidence is called for (Max 2004),
ideally on large population samples. A range of epidemiological approaches is required to deter-
mine the prevalence of candidate polymorphisms, and their interaction with other, environmen-
tal factors. In addition, a continuing search for polymorphisms at other sites that may be associated
with chronic pain is important (Solovieva et al. 2004b, Eisenach 2004). The COMT gene was the
first, and perhaps the most researched gene in pain, with single nucleotide and haplotype variants
being found in small studies to be associated with pain perception or development, yet with con-
flicting evidence found in confirmatory studies (Max and Stewart 2008). Larger sample sizes still
are required for definitive evidence. The increasing use and falling costs of genome-wide associa-
tion studies (GWAS), perhaps using collaborative consortia of samples derived from different
studies offer further exciting opportunity for identifying multiple candidate genes and testing
them in large samples (http://www.wtccc.org.uk/).
If any of these genetic factors are found to be important at a population level, rather than just
in the small samples so far tested, developing work in the field of proteomics could potentially
provide the basis for identifying measurable markers of risk for future chronic pain, with the
prospect of early intervention or prevention.
1.5 The need and potential benefit of population-based studies

of hypothesized biological risk factors
These biological factors may be important as variables on the path to development of chronic
pain (as in possible HPA dysfunction), as effect modifiers (as in some of the fMRI changes, or
genetic polymorphisms) and/or as risk markers (as in cerebral spinal fluid substance P levels, or
proteomic measurement based on genetic polymorphisms). The proposed factors do not operate
in isolation, and many will interact with each other, as well as with existing known factors
(Diatchenko et al. 2006). For example, brain imaging studies demonstrate that psychological
modulation of pain processing contributes to chronic pain, completing a hypothetically demon-
strable pathway from genes to prefrontal cortex in CLBP (Apkarian et al. 2005). The task is not,
therefore, to isolate new risk factors so much as to identify important ones, and assess their inter-
action with each other. Studying the biology of CLBP will both extend our knowledge of risk
factors and potential intervention points, and improve our understanding of the biological
mechanisms by which other risk factors exert their effects on the pain pathways.
The importance of the proposed biological factors in the community, particularly of those that
may inform interventions, needs now to be tested. This will require population-based studies to
determine: (1) whether any observed associations between proposed risk factors and CLBP
extrapolate to larger, representative samples; (2) the strength of these associations in CLBP as it
occurs in the community; and (3) the prevalence of the proposed factors. The technology and
opportunity now exist to conduct such studies, addressing a number of potential risk factors that
have been demonstrated. Studies should include and compare distinct diagnostic or structural
disorders, as well as the global entity of CLBP, including unexplained causes; these will present
different challenges in terms of sample size requirements, recruitment, and interpretation, but
will all contribute to the developing picture. Evidence from research on psychosocial factors
suggests that there will be some biological markers found for specific back conditions (such as the
genes associated with degenerative back pain), but that others will relate to several chronic pain
conditions (such as the COMT gene), or to chronic pain as a single entity (such as altered neuro-
nal pathways).
Epidemiologists will need a new set of tools in their box for this. In addition to continuing to
measure demographic and psychosocial risk factors in population-based studies, they will need
to take biological samples for DNA testing and storage and biochemical measurements, and to
consider, where appropriate, the use of brain imaging techniques such as fMRI (Borsook and
Becerra 2003). Ideally, in order to assess the predictive importance of clusters of risk factors, pro-
spective cohort studies are required. These may either be established de novo, or, with addi-
tional funding represent added value to existing or planned cohorts such as the UK Biobank
(www.ukbiobank.ac.uk) or Generation Scotland (Smith et al. 2006). These studies are expensive,
however and will not produce useful data for several years. Other study designs (such as cross-
sectional and case–control studies) will also be valuable though will lack the ability to distinguish
cause from effect with dynamic (non-genetic) markers, and will therefore be less helpful in
informing interventions. In all cases, close collaborations will be required between epidemiolo-
gists, pain and generalist clinicians, radiologists, laboratory scientists, and geneticists, as well as
with the psychologists and sociologists with whom pain epidemiologists have more traditionally
consorted. Importantly, these collaborations should be bi-directional, with the epidemiology
research informing the focus of laboratory and imaging research, as well as vice versa.
1.5.1 Beyond the bedside

The results of this new effort will include new understandings of the development and persistence
of chronic back pain, informing the development of new prevention and treatment strategies for
the population. These can either be applied on a large scale, if important, prevalent risk factors
can be addressed, or targeted accurately for maximum benefit using the knowledge of distribu-
tion that the research will provide. In the UK, the Medical Research Council aims to translate
research ‘from the laboratory to the bedside and back again’ (www.mrc.ac.uk/index/current-
research/current-clinical_research.htm). Similarly the National Institutes of Health in the USA
promote and reward ‘bench-to-bedside’ research (www.nih.gov/news/pr/jun2006/cc-07.htm).
However, in the case of CLBP, where only a small proportion of sufferers are treated in hospital
and many do not even attend a family practitioner, this translation must reach beyond the bedside,
into community-based consulting rooms, and beyond—from the laboratory to the bus stop, and
back again (Smith et al. 2007).
References
Abbadie, C., Lindia, J.A., Cumiskey, A.M., et al. (2003). Impaired neuropathic pain responses in mice
lacking the chemokine receptor CCR2. Proceedings of the National Academy of Sciences of the
United States of America, 100, 7947–52.
Ala-kokko, L. (2002). Genetic risk factors for lumbar disc disease. Annals of Medicine, 34, 42–7.
Andersson, G.B. (1999). Epidemiological features of chronic low-back pain. Lancet, 354, 581–85.
Annunen, S., Passilta, P., Lohiniva, J., et al.(1999). An allele of COL9A2 associated with intervertebral disc
disease. Science, 285, 409–12.
Apkarian, A.V., Sosa, Y., Sonty, S., et al. (2004). Chronic back pain is associated with decreased prefrontal
and thalamic gray matter density. The Journal of Neuroscience, 24, 10410–15.
Apkarian, A.V., Bushnell, M.C., Treede, R.D. and Zubieta, J.K. (2005). Human brain mechanisms of pain
perception and regulation in health and disease. European Journal of Pain, 9, 463–84.
Assendelft, W.J., Morton, S.C., Yu, E.I., Suttorp, M.J. and Shekelle, P.G. (2004). Spinal manipulative
therapy for low back pain. Cochrane Database of Systematic Reviews, 1, CD000447.
Baliki, M.N., Chialvo, D.R., Geha, P.Y., et al. (2006). Chronic pain and the emotional brain: specific brain
activity associated with spontaneous fluctuations of intensity of chronic back pain. The Journal of
Neuroscience, 26, 12165–73.
Baliki, M.N., Geha, P.Y., Jabakhanji., Harden, N., Schnitzer, T.J. and Apkarian, A.V. (2008). A preliminary
fMRI study of analgesic treatment in chronic back pain and knee osteoarthritis. Molecular pain, 4, 47.
Banic, B., Petersen-Felix, S., Andersen, O.K., et al. (2004). Evidence for spinal cord hypersensitivity in
chronic pain after whiplash injury and in fibromyalgia. Pain, 107, 7–15.
Battie, M.C., Videman, T., Levalahti, E., Gill, K., and Kaprio, J. (2007). Heritability of low back pain and
the role of disc degeneration. Pain, 131, 272–80.
Bergman, S., Herrstrom, P., Jacobsson, L.T., and Petersson, I.F. (2002). Chronic widespread pain: a three
year followup of pain distribution and risk factors. The Journal of Rheumatology, 29, 818–25.
Bondy, B., Spaeth, M., Offenbaecher, M., et al. (1999). The T102C polymorphism of the 5-HT2A-receptor
gene in fibromyalgia. Neurobiology of Disease, 6, 433–39.
Borge, A.I. and Nordhagen, R. (2000). Recurrent pain symptoms in children and parents. Acta Paediatrica,
89, 1479–83.
Borsook, D. and Becerra, L. (2003). Pain imaging: future applications to integrative clinical and basic
neurobiology. Advanced Drug Delivery Reviews, 55, 967–86.
Buchbinder, R. and Jolley, D. (2004). Population based intervention to change back pain beliefs: three year
follow up population survey. British Medical Journal, 28, 321.
Chatzitheodorou, D., Mavromoustakos, S. and Milioti, S. (2008). The effect of exercise on adrenocortical
responsiveness of patients with chronic low back pain, controlled for psychological strain. Clinical
Rehabilitation, 22, 319–28.
Clauw, D.J. and Chrousos, G.P. (1997). Chronic pain and fatigue syndromes: overlapping clinical
and neuroendocrine features and potential pathogenic mechanisms. Neuroimmunomodulation,
4, 134–53.
Cook, D.B., Lange, G., Ciccone, D.S., Liu, W.C., Steffner, J., and Natelson, B.H. (2004). Functional
imaging of pain in patients with primary fibromyalgia. The Journal of Rheumatology, 31, 364–78.
Croft, P.R., Macfarlane, G.J., Papageorgiou, A.C., Thomas, E., and Silman, A.J. (1998). Outcome of low
back pain in general practice: a prospective study. British Medical Journal, 316, 1356–59.
Diatchenko, L., Nackley, A.G., Slade, G.D., Fillingim, R.B., and Maixner, W. (2006). Idiopathic pain
disorders—pathways of vulnerability. Pain, 123, 226–30.
Diatchenko, L., Nackley, A.G., Tchivileva, I.E., Shabalina, S.A., and Maixner, W. (2007). Genetic
architecture of human pain perception. Trends in Genetics, 23, 605–13.
Dionne, C.E., Dunn, K.M., Croft, P.R., et al. (2008). A consensus approach toward the standardization of
back pain definitions for use in prevalence studies. Spine, 33, 95–103.
Edwards, P.W., Zeichner, A., Kuczmierczyk, A.R., and Boczkowski, J. (1985). Familial pain models: the
relationship between family history of pain and current pain experience. Pain, 21, 379–84.
Eisenach, J.C. (2004). Fishing for genes: practical ways to study genetic polymorphisms for pain.
Anesthesiology, 100, 1343–4.
Elliott, A.M., Smith, B.H., Penny, K.I., Smith, W.C., and Chambers, W.A. (1999). The epidemiology of
chronic pain in the community. Lancet, 354, 1248–52.
Elliott, A.M., Smith, B.H., Hannaford, P.C., Smith, W.C. and Chambers, W.A. (2002). The course of
chronic pain in the community: results of a 4-year follow-up study. Pain, 99, 299–307.
Fishbain, D.A., Fishbain, D., Lewis, J., et al. (2004). Genetic testing for enzymes of drug metabolism:
does it have clinical utility for pain medicine at the present time? A structured review. Pain Medicine,
5, 81–93.
Frost, H., Lamb, S.E., Doll, H.A., Carver, P.T., and Stewart-Brown, S. (2004). Randomised controlled trial
of physiotherapy compared with advice for low back pain. British Medical Journal, 329, 708.
Gaab, J., Baumann, S., Budnoik, A., Gmunder, H., Hottinger, N., and Ehlert, U. (2005). Reduced reactivity
and enhanced negative feedback sensitivity of the hypothalamus-pituitary-adrenal axis in chronic
whiplash-associated disorder. Pain, 119, 219–24.
Gracely, R.H., Petzke, F., Wolf, J.M., and Clauw, D.J. (2002). Functional magnetic resonance imaging
evidence of augmented pain processing in fibromyalgia. Arthritis and Rheumatism, 46, 1333–43.
Grachev, I.D., Frerickson, B.E., and Apkarian, A.V. (2000). Abnormal brain chemistry in chronic back
pain: an in vivo proton magnetic resonance spectroscopy study. Pain, 89, 7–18.
Grachev, I.D., Fredickson, B.E., and Apkarian, A.V. (2001). Dissociating anxiety from pain: mapping the
neuronal marker N-acetyl aspartate to perception distinguishes closely interrelated characteristics of
chronic pain. Molecular Psychiatry, 6, 256–8.
Grachev, I.D., Fredrickson, B.E., and Apkarian, A.V. (2002). Brain chemistry reflects dual states of pain
and anxiety in chronic low back pain. Journal of Neural Transmission, 109, 1309–34.
Grachev, I.D., Ramachandran, T.S., Thomas, P.S., Szeverenyi, N.M. and Fredrickson, B.E. (2003).
Association between dorsolateral prefrontal N-acetyl aspartate and depression in chronic back pain: an
in vivo proton magnetic resonance spectroscopy study. Journal of Neural Transmission, 110, 287–312.
Gran, J.T. (2003). The epidemiology of chronic generalized musculoskeletal pain. Best Practice & Research.
Clinical Rheumatology, 17, 547–61.
Griep, E.N., Boersma, J.W., Lentjes, E.G., Prins, A.P., Van Der Korst, J.K., and De Kloet, E.R. (1998).
Function of the hypothalamic-pituitary-adrenal axis in patients with fibromyalgia and low back pain.
The Journal of Rheumatology, 25, 1374–81.
Gursoy, S., Erdal, E., Herken, H., Madenci, E., Alaeshirli, B., and Erdal, N. (2003). Significance of catechol-
O-methyltransferase gene polymorphism in fibromyalgia syndrome. Rheumatology International, 23,
104–7.
Haddad, J.J., Saade, N.E., and Safieh-Garabedian, B. (2002). Cytokines and neuro-immune-endocrine
interactions: a role for the hypothalamic-pituitary-adrenal revolving axis. Journal of Neuroimmunology,
133, 1–19.
Hagen, K.B., Jamtvedt, G., Hilde, G., and Winnem, M.F. (2005). The updated cochrane review of bed rest
for low back pain and sciatica. Spine, 30, 542–6.
Harstall, C. and Ospina, M. (2003). How prevalent is chronic pain? Pain Clinical Updates, 11, 1–4.
Hartvigsen, J., Christensen, K., Frederiksen, H., and Petersen, H.C. (2004). Genetic and environmental
contributions to back pain in old age: a study of 2,108 Danish twins aged 70 and older. Spine, 29, 897–901.
Hasselhorn, H.M., Theorell, T., Vingard, E., and Musculoskeletal Intervention Center
(MUSIC)-NORRTALJE Study Group (2001). Endocrine and immunologic parameters indicative of
6-month prognosis after the onset of low back pain or neck/shoulder pain. Spine, 26, E24–9.
Hay, E.M., Mullis, R., Lewis, M., et al. (2005). Comparison of physical treatments versus a brief
pain-management programme for back pain in primary care: a randomised clinical trial in
physiotherapy practice. Lancet, 365, 2024–30.
Hay, E.M., Dunn, K.M., Hill, J.C., et al. (2008). A randomised clinical trial of subgrouping and targeted
treatment for low back pain compared with best current care. The STarT Back Trial Study Protocol.
BMC Musculoskeletal Disorders, 9, 58.
Hayden, J.A., Van Tulder, M.W., Malmivaara, A. and Koes, B.W. (2005). Exercise therapy for treatment of
non-specific low back pain. Cochrane Database of Systematic Reviews, 3(3), CD000335.
Henschke, N., Maher, C.G., Regshauge, K.M., et al. (2008). Prognosis in patients with recent onset low
back pain in Australian primary care: inception cohort study. British Medical Journal, 337, 171.
Herken, H., Erdal, E., Mutlu, N., et al. (2001). Possible association of temporomandibular joint pain and
dysfunction with a polymorphism in the serotonin transporter gene. American Journal of Orthodontics
and Dentofacial Orthopedics, 120, 308–13.
Heymans, M.W., Van Tulder, M.W., Esmail, R., Bombardier, C., and Koes, B.W. (2005). Back schools for
non-specific low-back pain: a systematic review within the framework for the Cochrane Collaboration
Back Review Group. Spine, 30, 2153–63.
Hotopf, M., Carr, S., Mayou, R., Wadsworth, M., and Wessely, S. (1998). Why do children have chronic
abdominal pain, and what happens to them when they grow up? Population based cohort study. British
Medical Journal, 316, 1196–200.
Jayson, M.I. (1997). Why does acute back pain become chronic? British Medical Journal, 314, 1639–40.
Jellema, P., Van Der Windt, D.A., Van Der Horst, H.E., Blankenstein, A.H., Bouter, L.M., and Stalman,
W.A. (2005a). Why is a treatment aimed at psychosocial factors not effective in patients with (sub)
acute low back pain? Pain, 118, 350–9.
Jellema, P., Van Der Windt, D.A., Van Der Horst, H.E., Twisk, J.W., Stalman, W.A., and Bouter, L.M.
(2005b). Should treatment of (sub)acute low back pain be aimed at psychosocial prognostic factors?
Cluster randomised clinical trial in general practice. British Medical Journal, 331, 84–8.
Jerjes, W.K., Cleare, A.J., Wessley, S., Wood, P.J. and Taylor, N.F. (2005). Diurnal patterns of salivary
cortisol and cortisone output in chronic fatigue syndrome. Journal of Affective Disorders, 87, 299–304.
Jerjes, W.K., Peters, T.J., Taylor, N.F., Wood, P.J., Wessley, S., and Cleare, A.J. (2006). Diurnal excretion
of urinary cortisol, cortisone, and cortisol metabolites in chronic fatigue syndrome. Journal of
Psychosomatic Research, 60, 145–53.
Jones, G., White, C., Sambrook, P., and Eisman, J. (1998). Allelic variation in the vitamin D receptor,
lifestyle factors and lumbar spinal degenerative disease. Annals of Rheumatic Diseases, 57, 94–9.
Jones, G.T., Johnson, R.E., Wiles, N.J., et al. (2006). Predicting persistent disabling low back pain in general
practice: a prospective cohort study. The British Journal of General Practice, 56, 334–41.
Jones, G.T., Silman, A.J., Power, C., and Macfarlane, G.J. (2007). Are common symptoms in childhood
associated with chronic widespread body pain in adulthood? Results from the 1958 British Birth
Cohort Study. Arthritis and Rheumatism, 56, 1669–75.
Kalichman, L. and Hunter, D.J. (2008). The genetics of intervertebral disc degeneration. Familial
predisposition and heritability estimation. Joint Bone Spine, 75, 383–7.
Kawaguchi, Y., Osada, R., Kanamori, M., et al. (1999). Association between an aggrecan gene
polymorphism and lumbar disc degeneration. Spine, 24, 2456–60.
Kawaguchi, Y., Kanamori, M., Ishihara, H., Ohmori, K., Matsui, H., and Kimura, T. (2002). The
association of lumbar disc disease with vitamin-D receptor gene polymorphism. The Journal of bone
and Joint Surgery, 84-A(11), 2022–8.
Koutantji, M., Pearce, S.A., and OakelyY, D.A. (1998). The relationship between gender and family history
of pain with current pain experience and awareness of pain in others. Pain, 77, 25–31.
Lamb, S.E., Lall, R., Hansen, Z. and the Back Skills Training Trial (BeST) Team (2007). Design
considerations in a clinical trial of a cognitive behavioural intervention for the management of low
back pain in primary care: Back Skills Training Trial. BMC Musculoskeletal Disorders, 8, 14.
Lambeek, L.C., Anema, J.R., Van Royen, B.J., et al. (2007). Multidisciplinary outpatient care program for
patients with chronic low back pain: design of a randomized controlled trial and cost-effectiveness
study. BMC Public Health, 7, 254.
Lau, R.R., Quadrel, M.J., and Hartman, K.A. (1990). Development and change of young adults’ preventive
health beliefs and behavior: influence from parents and peers. Journal of Health and Social Behavior,
31, 240–59.
Limer, K.L., Nicholl, B.I., Thomson, W., and McBeth, J. (2008). Exploring the genetic susceptibility of
chronic widespread pain: the tender points in genetic association studies. Rheumatology, 47, 572–7.
Linde, K., Witt, C.M., Streng, A., et al. (2007). The impact of patient expectations on outcomes in four
randomized controlled trials of acupuncture in patients with chronic pain. Pain, 128, 264–71.
Linton, S.J. (2005). Do psychological factors increase the risk for back pain in the general population in
both a cross-sectional and prospective analysis? European Journal of Pain, 9, 355–61.
Lis, A.M., Black, K.M., Korn, H., and Nordin, M. (2007). Association between sitting and occupational
LBP. European Spine Journal, 16, 283–98.
Macfarlane, G.J. (2006). Who will develop chronic pain and why? In: H. Flor, E. Kalso, and J.O. Dostrovsky
(eds) The Epidemiological Evidence Proceedings of the 11th World Congress on Pain. Seattle, WA:
International Association for the Study of Pain (IASP) Press.
Macfarlane, G.J., Jones, G.T., and Hannaford, P.C. (2006). Managing low back pain presenting to primary
care: where do we go from here? Pain, 122, 219–22.
Macfarlane, T.V., Blinkhorn, A., Worthington, H.V., Davies, R.M., and Macfarlane, G.J. (2002).
Sex hormonal factors and chronic widespread pain: a population study among women. Rheumatology,
41, 454–7.
MacGregor, A.J., Griffiths, G.O., Baker, J., and Spector, T.D. (1997). Determinants of pressure pain
threshold in adult twins: evidence that shared environmental influences predominate. Pain, 73, 253–7.
MacGregor, A.J., Andrew, T., Sambrook, P.N., and Spector, T.D. (2004). Structural, psychological, and
genetic influences on low back and neck pain: a study of adult female twins. Arthritis and Rheumatism,
51, 160–7.
MacMahon, B. and Pugh, T.F. (1970). Epidemiology: Principles and Methods. Boston, MA: Little Brown.
Magni, G., Marchetti, M., Moreschi, C., Merskey, H. and Luchini, S.R. (1993). Chronic musculoskeletal
pain and depressive symptoms in the National Health and Nutrition Examination. I. Epidemiologic
follow-up study. Pain, 53, 163–8.
Manchikanti, L. (2000). Epidemiology of low back pain. Pain physician, 3, 167–92.
Manek, N.J. and MacGregor, A.J. (2005). Epidemiology of back disorders: prevalence, risk factors, and
prognosis. Current Opinion in Rheumatology, 17, 134–40.
Matsui, H., Maeda, A., Tsuji, H., and Naruse, Y. (1997). Risk indicators of low back pain among workers in
Japan. Association of familial and physical factors with low back pain. Spine, 22, 1242–7.
Max, M.B. (2004). Assessing pain candidate gene studies. Pain, 109, 1–3.
Max, M.B. and Stewart, W.F. (2008). The molecular epidemiology of pain: a new discipline for drug
discovery. Nature Reviews. Drug Discovery, 7, 647–58.
McBeth, J., Macfarlane, G.J., Hunt, I.M., and Silman, A.J. (2001). Risk factors for persistent chronic
widespread pain: a community-based study. Rheumatology, 40, 95–101.
McBeth, J., Harkness, E.F., Silman, A.J., and Macfarlane, G.J. (2003). The role of workplace low-level
mechanical trauma, posture and environment in the onset of chronic widespread pain. Rheumatology,
42, 1486–94.
McBeth, J., Chui, Y.H., Silman, A.J., et al. (2005). Hypothalamic-pituitary-adrenal stress axis function and the
relationship with chronic widespread pain and its antecedents. Arthritis Research & Therapy, 7, R992–R1000.
McQuay, H.J., Derry, S., Moore, R.A., Poulain, P., and Legout, V. (2008). Enriched enrolment with
randomised withdrawal (EERW): Time for a new look at clinical trial design in chronic pain. Pain,
135, 217–20.
Melzack, R. and Casey, K. (1968). Sensory, motivational, and central control determinants of pain. In:
D.R. Kenshalo (ed.) The skin senses, 2nd edn., pp. 423–43. Springfield, IL: Charles C. Thomas.
Mogil, J.S., Wilson, S.G., Bon, K., et al. (1999). Heritability of nociception I: responses of 11 inbred mouse
strains on 12 measures of nociception. Pain, 80, 67–82.
Mogil, J.S., Wilson, S.G., Chesler, E.J., et al. (2003). The melanocortin-1 receptor gene mediates
female-specific mechanisms of analgesia in mice and humans. Proceedings of the National Academy of
Sciences of the United States of America, 100, 4867–72.
Mogil, J.S. (ed) (2004). The Genetics of Pain (Progress in Pain Research and Management). Seattle, WA:
IASP Press.
Morley, S., Eccleston, C., and Williams, A. (1999). Systematic review and meta-analysis of randomized
controlled trials of cognitive behaviour therapy and behaviour therapy for chronic pain in adults,
excluding headache. Pain, 80, 1–13.
Newton-John, T.R. and Geddes, J. (2008). The non-specific effects of group-based cognitive—behavioural
treatment of chronic pain. Chronic illness, 4, 199–208.
Omoigui, S. (2007). The biochemical origin of pain: the origin of all pain is inflammation and the
inflammatory response. Part 2 of 3 - inflammatory profile of pain syndromes. Medical hypotheses,
69, 1169–78.
Paassilta, P., Lohiniva, J., Goring, H.H., et al. (2001). Identification of a novel common genetic risk factor
for lumbar disk disease. Journal of the American Medical Association, 285, 1843–9.
Papageorgiou, A.C., Croft, P.R., Thomas, E., Ferry, S., Jayson, M.I., and Silman, A.J. (1996). Influence of
previous pain experience on the episode incidence of low back pain: results from the South Manchester
Back Pain Study. Pain, 66, 181–5.
Papageorgiou, A.C., Silman, A.J., and Macfarlane, G.J. (2002). Chronic widespread pain in the population:
a seven year follow up study. Annals of the Rheumatic Diseases, 61, 1071–4.
Pata, C., Erdal, E., Yazc, K., Camdeviren, H., Ozkaya, M., and Ulu, O. (2004). Association of the -1438 G/A
and 102 T/C polymorphism of the 5-Ht2A receptor gene with irritable bowel syndrome 5-Ht2A gene
polymorphism in irritable bowel syndrome. Journal of Clinical Gastroenterology, 38, 561–6.
Pincus, T., Burton A.K., Vogel, S., and Field, A.P. (2002). A systematic review of psychological factors as
predictors of chronicity/disability in prospective cohorts of low back pain. Spine, 27, E109–20.
Porta, M., Greenland, S., and Last, J.M. (eds). (2008). A Dictionary of Epidemiology, 5th edn. New York:
Oxford University Press.
Postacchini, F., Lami, R., and Pugliese, O. (1988). Familial predisposition to discogenic low-back pain. An
epidemiologic and immunogenetic study. Spine, 13, 1403–6.
Pountain, G., Hazelman, B., and Cawston, T.E. (1998). Circulating levels of IL-1beta, IL-6 and soluble IL-2
receptor in polymyalgia rheumatica and giant cell arteritis and rheumatoid arthritis. British Journal of
Rheumatology, 37, 797–8.
Rothman, K.J. and Greenland, S. (1998). Modern Epidemiology, 2nd edn. Oxford: Oxford University Press.
Royal College of General Practitioners (1999). Clinical Guidelines for the Management of Acute Low Back
Pain. London: Royal College of General Practitioners.
Russell, I.J. (1998). Advances in fibromyalgia: possible role for central neurochemicals. The American
Journal of the Medical Sciences, 315, 377–84.
Russell, I.J., Vaeroy, H., Javors, M. and Nyberg, F. (1992). Cerebrospinal fluid biogenic amine metabolites
in fibromyalgia/fibrositis syndrome and rheumatoid arthritis. Arthritis and Rheumatism, 35, 550–6.
Sambrook, P.N., Macgregor, A.J., and Spector, T.D. (1999). Genetic influences on cervical and lumbar disc
degeneration: a magnetic resonance imaging study in twins. Arthritis and Rheumatism, 42, 366–72.
Scuderi, G.J., Brusovanik, G.V., Anderson, D.G., et al. (2006). Cytokine assay of the epidural space lavage
in patients with lumbar intervertebral disk herniation and radiculopathy. Journal of Spinal Disorders &
Techniques, 19, 266–9.
Smith, A.K., White, P.D., Aslakson, E., Vollmer-Conna, U., and Rajeevan, M.S. (2006). Polymorphisms in
genes regulating the HPA axis associated with empirically delineated classes of unexplained chronic
fatigue. Pharmacogenomics, 7, 387–94.
Smith, B.H. (2001). Chronic pain: a challenge for primary care. British Journal of General Practice,
51, 524–6.
Smith, B.H., Chambers, W.A., and Smith, W.C. (1996). Chronic pain: time for epidemiology. Journal of the
Royal Society of Medicine, 89, 181–3.
Smith, B.H., Elliott, A.M. and Hannaford, P.C. (2004a). Is chronic pain a distinct diagnosis in primary
care? Evidence arising from the Royal College of General Practitioners’ Oral Contraception study.
Family practice, 21, 66–74.
Smith, B.H., Elliott, A.M., Hannaford, P.C., Chambers, W.A., and Smith, W.C. (2004b). Factors related to
the onset and persistence of chronic back pain in the community: results from a general population
follow-up study. Spine, 29, 1032–40.
Smith, B.H., Campbell, H., Blackwood, D., et al. (2006). Generation Scotland: the Scottish Family Health
Study; a new resource for researching genes and heritability. BMC Medical Genetics, 7, 74.
Smith, B.H., Macfarlane, G.J., and Torrance, N. (2007). Epidemiology of chronic pain, from the laboratory
to the bus stop: time to add understanding of biological mechanisms to the study of risk factors in
population-based research? Pain, 127, 5–10.
Solovieva, S., Kouhia, S., Leino-Arjas, P., et al. (2004a). Interleukin 1 polymorphisms and intervertebral
disc degeneration. Epidemiology, 15, 626–33.
Solovieva, S., Leino-Arjas, P., Saarela, J., Luoma, K., Raininko, R., and Riihimaki, H. (2004b). Possible
association of interleukin 1 gene locus polymorphisms with low back pain. Pain, 109, 8–19.
Sorensen, J., Graven-Nielsen, T., Henriksson, K.G., Bengtsson, M., and Arendt-Nielsen, L. (1998).
Hyperexcitability in fibromyalgia. The Journal of Rheumatology, 25, 152–5.
Staud, R. (2004). Fibromyalgia pain: do we know the source? Current Opinion in Rheumatology, 16, 157–63.
Szilagyi, A., Boor, K., Orosz, I., Szantai, E., et al. (2006). Contribution of serotonin transporter gene
polymorphisms to pediatric migraine. Headache, 46, 478–85.
Terre, L. and Ghiselli, W. (1997). A developmental perspective on family risk factors in somatization.
Journal of Psychosomatic Research, 42, 197–208.
UK BEAM Trial Team (2004). United Kingdom back pain exercise and manipulation (UK BEAM)
randomised trial: effectiveness of physical treatments for back pain in primary care. British Medical
Journal, 329, 1377.
Underwood, M.R., Morton, V., Farrin, A. and UK BEAM Trial Team (2007). Do baseline characteristics
predict response to treatment for low back pain? Secondary analysis of the UK BEAM dataset.
Vaeroy, H. (1989). Impaired neuroendocrine axis and FMS. The Journal of Rheumatology, 16, 1460–5.
Valdes, A.M., Hassett, G., Hart, D.J., and Spector, T.D. (2005). Radiographic progression of lumbar spine
disc degeneration is influenced by variation at inflammatory genes: a candidate SNP association study
in the Chingford cohort. Spine, 30, 2445–51.
Van Der Windt, D., Hay, E., Jellema, P., and Main, C. (2008). Psychosocial interventions for low back pain
in primary care: lessons learned from recent trials. Spine, 33, 81–9.
Van Tulder, M.W., Ostelo, R., Vlaeyen, J.W., Linton, S.J., Morley, S.J., and Assendelft, W.J. (2000).
Behavioral treatment for chronic low back pain: a systematic review within the framework of the
Cochrane Back Review Group. Spine, 25, 2688–99.
Verhaak, P.F., Kerssens, J.J., Dekker, J., Sorbi, M.J., and Bensing, J.M. (1998). Prevalence of chronic
benign pain disorder among adults: a review of the literature. Pain, 77, 231–9.
Videman, T., Leppavuori, J., Kaprio, J., et al. (1998). Intragenic polymorphisms of the vitamin D receptor
gene associated with intervertebral disc degeneration. Spine, 23, 2477–85.
Von Korff, M. (1999). Epidemiological methods In: I.K. Crombie, P.R. Croft, S.J. Linton, L. Leresche, and
M. Von Korff (eds.) Epidemiology of Pain, pp. 7–15. Seattle, WA: IASP Press.
Von Korff, M., Balderson, B.H., Saunders, K., et al. (2005). A trial of an activating intervention for chronic
back pain in primary care and physical therapy settings. Pain, 113, 323–30.
Wessely, S., Nimnuan, C., and Sharpe, M. (1999). Functional somatic syndromes: one or many? Lancet,
354, 936–9.
Williams, A.C., Richardson, P.H., Nicholas, M.K., et al. (1996). Inpatient vs. outpatient pain management:
results of a randomised controlled trial. Pain, 66, 13–22.
Wolf, C.R., Smith, G., and Smith, R.L. (2000). Science, medicine, and the future: Pharmacogenetics. British
Medical Journal, 320, 987–90.
Yu, L. (2004). Pharmacogentics: the OPRM (mu-opioid-receptor) gene. In: J.S. Mogil (ed.) The Genetics of
Pain (Progress in Pain research and Management), pp. 239–56. Seattle, WA: IASP Press.
Zubieta J.K., Heitzeg, M.M., Smith, Y.R., et al. (2003). COMT val158met genotype affects mu-opioid
neurotransmitter responses to a pain stressor. Science, 299, 1240–3.
Chapter 2
Defining Chronic Pain by Prognosis

Kate M. Dunn, Michael Von Korff, and Peter R. Croft
This book addresses the transition ‘from acute to chronic back pain’, but do we really know what
‘chronic’ means? The International Association for the Study of Pain (IASP) defines chronic pain
as pain lasting beyond the normal healing time, which is usually defined as 3 months or 6 months
(International Association for the Study of Pain 1986). Is this retrospective approach to defining
chronic pain sufficient, or are there better ways to characterize patients with chronic pain? In this
chapter we reconsider the duration-based, retrospective approach to defining chronic pain, con-
sider reasons why it might be inadequate, and develop an alternative approach to assessing and
defining chronic pain, based on its prognosis and multiple prognostic indicators grounded in the
biopsychosocial model of pain.
The prognostic approach to defining chronic pain developed in this chapter rests upon three
propositions that can be tested empirically. The first proposition is that chronic pain is better
characterized by a failure of pain and associated dysfunction to resolve than by progression to a
qualitatively distinct chronic pain state. That is, severe pain, pain-related activity limitations,
psychological distress, and other features used to characterize chronic pain patients are observa-
ble soon after pain onset. What typically differentiates ‘chronic pain’ from ‘acute pain’ is that
severe pain and associated dysfunction does not show meaningful improvement. While progres-
sive worsening of pain and pain dysfunction may be observed in some patients, this clinical
course is less common. The second proposition is that the seeds of chronic pain (prognostic indi-
cators) are observable early in the course of a pain condition, providing a basis for early identifica-
tion of patients at risk of an unfavourable outcome. The third proposition is that varied and
multi-faceted prognostic indicators can be summarized in a single measure—a prognostic risk
score. This prognostic risk score measures how likely it is that clinically significant pain
will continue to be present at a future point in time, providing a simple and clinically relevant
metric for quantifying chronic pain in terms of its future prognosis rather than its past duration
considered in isolation.
2.1 Duration-based definitions of chronic pain

In 1953, Bonica described chronic pain as pain persisting beyond normal healing time (Bonica
1953). The Classification of Chronic Pain of the IASP (International Association for the Study of
Pain 1986) defined chronic pain as pain that persists for at least 3 or 6 months for non-malignant
pain, corresponding to the time required for inflammation to subside or acute injuries to repair.
Similar definitions exist for specific pain conditions. Nachemson and Andersson (1982) defined
chronic back pain as back pain lasting 3 months or longer based on a study showing that 90% of
patients recovered within that time. The International Classification of Headache Disorders
defined chronic secondary headache disorders as headache persisting over a period longer than
3 months, and chronic primary headache disorders as attacks of headache occurring on most days
for over 3 months (International Headache Society 2004).
These definitions are each based on the assumption that duration or persistence of pain is the
critical defining feature of chronic pain. Duration-based definitions of chronic pain differentiate
persons with chronic pain empirically, as the passage of time sorts out those with chronic pain
from those with acute pain. However, as Loeser and Melzack pointed out (1999), ‘it is not the
duration of pain that distinguishes acute from chronic pain’. This statement reflects the implicit
understanding, shared by clinicians and patients alike, that the term ‘chronic pain’ carries greater
significance than a simple descriptive statement about how long pain has lasted. Defining chronic
pain by duration alone neglects other features of pain that have equal or greater prognostic value
(e.g. pain intensity, pain-related interference with activities, the diffuseness of pain, emotional
distress), and that have greater import for patient functioning and quality of life than pain
duration considered in isolation.
The conceptual underpinnings of duration-based definitions of chronic pain are problematic.
There are four significant limitations of defining chronic pain by duration alone:
◆ Multidimensionality—it is widely recognized that pain is a multidimensional phenomenon
(Waddell 1998; Loeser and Melzack 1999). Raspe et al. (2003) observed that ‘back pain is more
than pain in the back’. The traditional definition of chronic pain implies that acute and chronic
pain can be differentiated by pain duration alone, whereas physical and emotional suffering,
behavioural deactivation, and social role disability are important clinical features of chronic pain.
◆ Early identification—duration-based definitions do not provide a basis for identifying patients
at risk of a poor outcome early in the clinical course of a pain condition, when prevention
might be most effective. This is illustrated by Patient A (Box 2.1). According to a duration
Box 2.1 Examples of typical back pain patients

Patient A
This person presents to his primary care physician having had back pain for 1 month. His pain
is severe and he is clinically depressed. He reports pain in many other anatomical locations
as well. He is off work and has no plans to return to work, and is spending most of his time
resting and watching television. He has significant sleep disturbance.
Patient B
This person visits his doctor during an unusually severe flare-up of back pain. He has had back
pain almost every day for many years. The pain typically fluctuates during the day, but is not
too bothersome once he is up and about. He sometimes finds it difficult to make certain
movements due to his back pain, but he generally regards it as a minor nuisance that does not
substantially limit his activities. He does not experience significant psychological distress, and
his pain is limited to the lower back. The patient has had these flare-ups before, and they
always improve to his usual low level, typically within 3–5 days.
Patient C
This person comes to see her doctor because her back pain has gotten worse again. She has had
pain in the lower back for over a year, and this appears to be an exacerbation. She also says that
she is having headaches and cannot sleep because of the pain. Over the last 6 months, the
patient has taken time off work several times, which is worrying her, and she is no longer
getting much satisfaction out of her job. She is also having trouble doing the shopping as it
hurts her back to carry shopping bags.
DEFINING CHRONIC PAIN BY PROGNOSIS 23
based definition, this patient has acute back pain, and would not be classified as having chronic
back pain until his pain had lasted another 2 months. A primary care physician who focused
on pain duration to evaluate prognosis would manage this patient according to acute low back
pain guidelines, and might even advise the patient that he could expect to be ‘pain-free’ in a
matter of days or weeks. However, the other factors reported by this patient are established
indicators of significant psychosocial dysfunction linked to poor prognosis, and these might
not be detected, or might be regarded as understandable manifestations of severe (but time-
limited) back pain, that could be expected to resolve as the patient’s back pain improved with
the tincture of time.
◆ Timescale of neurophysiological and somatosensory mechanisms—as noted above, chronic pain
is traditionally distinguished from acute pain on a timescale measured in months. This does
not correspond to the timescale of neurophysiological changes hypothesized to contribute to
chronic pain.1 For example, long-term potentiation of nociceptive synaptic transmission can
be induced over a time scale measured in milliseconds to seconds (Klein et al., 2004). Wind-up,
or central pain sensitization, has been observed to take place over a timescale measured in
seconds to minutes (Staud et al. 2001). Burn sensitization occurs over a timescale of minutes
to hours (Woolf and McMahon 1985). And, inflammation has been observed to produce
central changes in pain pathways over a timescale measured in hours to days (Pitcher et al.
2007). Thus, neurophysiological changes that may contribute to induction of chronic pain can
occur early in the development of a pain condition, rather than unfolding gradually over a
period of months. At the same time, individual differences in susceptibility to induction of
chronic pain may be associated with premorbid somatosensory characteristics, such as soma-
tization or hypochondriasis, that are relatively stable over long periods of time (Barsky et al.
1998; De Gucht 2003; Fishbain et al. 2009). These more stable prognostic indicators may be
identifiable early in the course of a pain condition, or prior to onset. This raises the question
of the extent to which there is a transition from acute pain to a qualitatively distinct chronic
pain state that takes months to unfold or whether essential features of chronic pain are observ-
able early on. Duration-based definitions of chronic pain imply that there is a transition from
acute to chronic pain that takes months to unfold.
◆ Convergent validity—there is not a clear association between pain duration and the extent of
pain dysfunction. In addition, manifestations often referred to as identifying features of
‘chronic’ pain (e.g. severe pain, depression, pain-related activity limitations) are often present
in the acute phase of pain. If ‘chronic’ pain implies only that pain has lasted for 3 months or
more, the weak association of pain duration and pain dysfunction would be of limited import.
However, it is sometimes assumed that people with long-standing pain problems almost
always have greater dysfunction and require different treatment than persons with pain of
more recent origin. In fact, there is no clear cut-off for pain duration that alone is a strong
predictor of the severity of pain dysfunction (Von Korff et al. 1992; Dunn and Croft 2006).
Moreover, among people with long pain duration and significant pain dysfunction, some go
on to improve significantly even through they continue to experience pain. For such persons,
is their pain more chronic due to increased duration, or less chronic due to reduced pain dys-
function? In the population-at-large, there are many individuals with persistent or recurrent
pain who do not experience substantial activity limitations and suffering. For example,
1 We gratefully acknowledge Gregory Terman for pointing out the difference between the traditional times-
cale used to differentiate chronic from acute pain (3 to 6 months) and the much briefer time intervals over
which neuroplastic changes hypothesized to induce chronic pain have been observed.
although a substantial proportion of people with ‘chronic’ pain have a poor prognosis, up to
40% completely recover within a year (Costa et al. 2009). Among persons with comparably
long pain duration, those with less intense pain, lower levels of activity limitation, lower levels
of emotional distress, and less diffuse pain are more likely to recover (that is, their pain is less
likely to run a chronic course). This is illustrated by Patient B in Box 2.1, who has had long-
standing back pain, but whose pain is likely to improve to a manageable level as there are no
indicators of poor back pain prognosis other than pain of long duration.
In light of these conceptual difficulties with defining chronic pain by duration alone, there is a
need to consider approaches to defining chronic pain that address shortcomings of traditional,
duration-based definitions of chronic pain.
2.2 Conceptual bases for defining chronic pain2

Defining chronic pain by duration is based on the view that acute pain signals potential tissue
damage, whereas chronic pain results from central and peripheral sensitization in which pain is
sustained after nociceptive inputs have diminished (Bonica 1990). From this perspective, chronic
pain is a progressive condition characterized by changes in central and peripheral nerves facilitat-
ing transmission of pain signals at lower thresholds (e.g. allodynia and hyperalgesia) (Staud and
Spaeth 2008). Such changes in pain transmission may be accompanied by psychological changes
(e.g. onset of depressed mood, disturbed sleep), by behavioural changes (e.g. reduced activity
levels), and, in severe cases, by disruption of social role function (e.g. unemployment). The tran-
sition from acute to chronic pain is hypothesized to take place over a 3–6-month period in which
a qualitatively different chronic pain state emerges.
A contrasting perspective is that chronic pain reflects deficiencies in endogenous pain inhibi-
tion systems, individual differences in somatosensory processing, and differences in abilities to
cope with pain that are observable soon after pain onset or even prior to pain onset. Edwards has
hypothesized that individual differences in pain sensitivity and pain inhibition, differences which
reflect variability in central nervous system pain processing, influence chronic pain risk across
anatomic sites (Edwards 2005). Individuals with inadequate endogenous pain inhibition are at
increased risk that pain conditions will not spontaneously resolve. Individual differences in
somatosensory processing, indicated by diffuse somatic symptoms and/or hypochondriasis, may
also be related to unfavourable prognostic risk (Celestin et al. 2009; Fishbain et al. 2009). The
biopsychosocial perspective holds that psychological and behavioural domains are also integral
determinants of risks of pain dysfunction being sustained over time. Catastrophizing, emotional
distress, and helplessness reflect deficiencies in abilities to adapt to pain (Keefe et al. 2004).
Individuals with these characteristics may be less able to restore normal functioning. This per-
spective focuses on factors present early in a pain episode (and prior to onset) that differentiate
persons likely to have favourable pain outcomes from those less likely to recover. Chronic pain
represents a failure of homeostatic systems (neurophysiological, psychological, behavioural) to
control pain expression and restore normal functioning, rather than as a transition to a qualita-
tively different chronic pain state. Manifestations of chronic pain (e.g. severe pain, depression,
activity limitations) are present early on, while chronic pain is usually characterized by a failure
2 The authors are indebted to Judith Turner, Gregory Terman, Mark Sullivan, Ruth Landau, Alex Cahana,
and Gary Franklin for discussions that contributed to ideas discussed in the section of this chapter on
‘Conceptual bases for defining chronic pain’.
of these manifestations to resolve, rather than progressive worsening over time being a typical
clinical course. However, this perspective leaves room for a cumulative effect of the experience of
pain over time to interact with the psychological and behavioural characteristics mentioned
above. Thus, pain persistence (or duration) is a significant prognostic indicator for two reasons:
(1) pain persistence is an indicator of deficiencies in pain inhibitory systems; and (2) pain persist-
ence also reflects increased opportunity for interaction between the experience of pain and
psychological and behavioural processes which may influence the subsequent course of pain and
associated dysfunction.
A critical question addressed by this chapter is the extent to which outcomes can be predicted
early in an episode of a painful condition. If chronic pain is predominantly a progressive condi-
tion, marked by a transition from acute to chronic pain, then it may be necessary to rely on the
passage of time to empirically sort out patients who develop chronic pain, which would justify
relying on pain duration as the critical defining feature of chronic pain. However, if indicators of
deficient endogenous pain inhibition, somatosensory amplification and inadequate pain coping
can be identified in the early stages of pain episodes, then it may be possible to identify patients at
high risk of unfavourable pain outcomes for early intervention. This chapter considers the extent
to which patients at high risk of clinically significant persistent pain, and associated dysfunction,
can be empirically identified without relying primarily on the passage of time to distinguish those
for whom clinically significant pain does not resolve spontaneously. The prognostic variables
considered in this chapter include self-report measures of pain (intensity, persistence, diffuse-
ness), depression, and pain-related dysfunction. However, the prognostic approach developed in
this chapter could, and should, be extended to incorporate psychophysical measures of deficien-
cies in pain modulation such as tests of diffuse noxious inhibitory control (DNIC) and mechani-
cal temporal summation (Granot et al. 2006, 2008). DNIC is a measure of endogenous pain
inhibition capacity, specifically, the degree to which exposure to an ongoing painful ‘condition-
ing’ stimulus inhibits the pain caused by a brief, repeated painful ‘test’ stimulus. Mechanical
temporal summation is an indicator of the excitatory mechanism of pain processing. Other
measures of somatosensory processing (e.g. somatization, hypochondriasis) and pain coping (e.g.
catastrophizing, fear-avoidance) should be considered as prognostic indicators as well.
2.3 Duration-based and prognostic approaches

Duration-based definitions of chronic pain have been commonly used for research purposes. For
example, traditionally designed prognostic studies rely on an inception cohort (of acute patients),
determined by episode duration (Henschke et al. 2008) or healthcare use (Croft et al. 1998).
Similarly, patients recruited to randomized controlled trials of chronic pain are often identified
based on the duration of their symptoms, e.g. low back pain for at least 6 months (Spinhoven
et al. 2004) or a minimum of 3 months (Katz et al. 2003). However, it is often unclear exactly how
this duration is determined, for example does the pain have to be present every day, or most days?
It is rare that the exact question used to define duration is presented with information supporting
its reliability and validity. Research comparing different questions used to determine back pain
episode duration has shown poor reliability between different definitions. For example, 38% of
people who said their back pain had been present for less than 3 months also reported they had
not had a whole month without back pain for over a year (Dunn et al. 2006a). Such difficulties
likely result from the nature of pain, which commonly runs a recurrent or episodic course (Von
Korff et al. 1993a). A definition of chronic pain based purely on duration implies a linear course
of pain (Cedraschi et al. 1998), making it difficult to accurately classify persons with recurrent
pain as either acute or chronic. In a clinical consultation, it may be possible to clarify the duration
of pain by requesting further information from the patient, but in a research setting where data
collection is standardized, this is a difficult task.
Researchers have recognized these difficulties, and proposed alternate definitions of chronic
pain. Definitions can take the episodic nature of some pain conditions into account, e.g. defining
chronic back pain as pain present on at least half the days in a 12-month or a 6-month period
(Von Korff 1994). Consensus definitions about how an episode should be defined more specifi-
cally have been proposed. De Vet et al. (2002) defined episodes of low back pain, care for low back
pain, and work absence due to low back pain, summarized in Box 2.2. The definition of low back
pain episodes has been examined among primary care back pain patients and found to have good
reliability and validity (Dunn et al. 2006a). However, there are still many problems in defining
duration. In a recent Delphi study, back pain researchers had substantial difficulty in reaching
consensus on how to operationalize back pain duration (Dionne et al. 2008). In the end, the
participants agreed to recommend using time since the last pain-free month to assess duration,
but with an additional category to identify patients with less than 3 months duration.
Apart from conceptual difficulties, the empirical bases for the cutpoints used to define chronic-
ity are not well supported. The justification that 90% of patients recover within 3 months
(Nachemson and Andersson 1982) has since been refuted. Hestbaek et al. carried out a systematic
review including 36 articles, and reported that an average of 62% of patients consulting with low
back pain still experienced pain after 12 months (Hestbaek et al. 2003). Furthermore, research
comparing primary care back pain patients with different episode durations did not support
a cut-off at 3 months’ duration, and offered only limited support for a cut-off at 6 months’
duration (Dunn and Croft 2006).
Prior episodes of pain have been found to be one of the main predictors of outcome
(Papageorgiou et al. 1996; Hestbaek et al. 2003). This suggests that defining chronicity purely on
the basis of the duration of the current episode, and ignoring prior history, might not be the most
appropriate approach. So-called ‘inception cohorts’ may not actually identify people at a
Box 2.2 Definitions of low back pain episodes

Episode of low back pain
Definition: a period of pain in the lower back lasting over 24 hours, preceded and followed by
1 month or more without low back pain (de Vet et al. 2002).
Question for current pain sufferers: ‘How long is it since you had a whole month without any
back pain?’ (Dunn and Croft 2006).
Question for general population samples: ‘If you had low back pain in the past 4 weeks, how
long was it since you had a whole month without any low back pain?’ (Dionne et al. 2008).
Episode of care for low back pain

Definition: a consultation or a series of consultations for low back pain, preceded and followed
by at least 3 months without consultation for low back pain (de Vet et al. 2002).
Episode of work absence due to low back pain

Definition: a period of work absence due to low back pain, preceded and followed by a period
of at least 1 day at work in the normal job (de Vet et al. 2002).
common point in the course of their condition, if measured by duration of the current episode,
when prior back pain episodes are taken into account.
2.4 Chronic pain as a multidimensional phenomenon

Risk factors for chronic pain come from a range of domains. Biological mechanisms can include
factors such as genetics (Hartvigsen et al. 2004; Hestbaek et al. 2004) or abnormalities in the
hypothalamic–pituitary–adrenal (HPA) stress-response system (McBeth et al. 2007), and biome-
chanical factors may incorporate components such as muscle strength (Smeets and Wittink 2007)
or central sensitization (Werneke and Hart 2001). Psychological factors can involve factors such
as distress or depressive mood (Pincus et al. 2002). Social factors may include job stress (Turner
et al. 2008) or social isolation (Steenstra et al. 2005). These factors are all known prognostic indi-
cators, but also characterize people with chronic pain. For example, 20% of people with chronic
pain also have major depression (Currie and Wang 2004), and chronic pain patients are much
more likely to have fear avoidance beliefs than acute pain patients (Grotle et al. 2004). Studies
including factors from several domains have confirmed the multidimensional nature of chronic
pain for a range of conditions including hand and wrist problems, back pain, shoulder pain, and
neck pain (Smith et al. 2004; Carragee et al. 2005; Kamper et al. 2008; Reilingh et al. 2008; Spies-
Dorgelo et al. 2008).
The recommendations for management of chronic pain also highlight the multidimensional
nature of the problem. For example, the European guidelines for the management of chronic
non-specific low back pain recommend assessment of domains including work-related factors,
psychological distress and depressive mood, and point out that no single intervention is likely to
be effective due to the multidimensional nature of chronic low back pain (Airaksinen et al. 2006;
see also Chapter 24). The guidelines for persistent pain from the American Geriatrics Society
recommend assessment of pain characteristics and impairments in physical and social function,
as well as factors such as attitudes and beliefs about pain (American Geriatrics Society Panel on
Persistent Pain in Older Persons 2002). The basis for such assessment stems from systems such as
Turk and Rudy’s Multiaxial Assessment of Pain, which integrated physical, psychosocial, and
behavioural information (Turk and Rudy 1987). The assessment of psychological factors has been
significantly advanced by the ‘yellow flags’ approach to assessment of back pain (ACC and the
National Health Committee 1997; Kendall et al. 2009; Nicholas et al. 2011), complementing the
‘red flag’ indicators of serious underlying conditions (Croft 1999; Henschke et al. 2009), and ‘blue
flags’ addressing individual-level occupational factors (Shaw et al. 2009). These ‘flags’ are now
commonly recommended for use in conditions such as low back pain (Samanta et al. 2003; Main
and Spanswick 2003; see also Chapter 13). This approach to back pain assessment is consistent
with the prognostic approach to defining chronic pain developed in this chapter.
Despite the acceptance of prognostic factors for chronic pain from multiple domains, and the
integration of the multidimensional approach into back pain assessment and treatment guide-
lines, the definition of chronic pain by duration is widely employed. For example, ‘chronic’ pain
patients are still identified for randomized controlled trials (e.g. Smeets et al. 2006; Critchley et al.
2007) and cohort studies (e.g. Keeley et al. 2008; Reilingh et al. 2008) using duration-based defini-
tions of chronicity. As stated by Cedraschi et al. over 10 years ago, there is a ‘necessity to include
not only clinical findings but also elements in relation to a biopsychosocial model of low back
pain in the definition of chronicity’ (Cedraschi et al. 1998). Limiting ourselves to studying chronic
pain in terms of duration restricts the investigation of genetic, neurophysiological and biome-
chanical processes, and psychosocial mechanisms that may represent important influences on the
development and course of chronic pain.
2.5 Chronic pain—static or dynamic?

The term ‘chronic’ can be interpreted as meaning unlikely to change (Von Korff and Dunn 2008),
but chronic pain is not a static phenomenon. Descriptions of the course of low back pain, for
example, describe pain episodes with exacerbations and pain-free intervals (Deyo 1993). Chronic
pain is most commonly characterized by recurrences or repeated episodes (Von Korff et al.
1993a). Population studies appear to give a slightly more stable picture of chronic pain, with a
study in Scotland showing that almost 80% of people with chronic pain still had reported pain
4 years later (Elliott et al. 2002), but this apparent stability in the presence of pain masks large
variability in pain severity. Recent in-depth analysis of pain trajectories over time has found
highly variable patterns of pain over time. Among primary care back pain consulters, groups
could be characterized as having recovering, persistent mild, fluctuating or severe chronic pain
trajectories (Dunn et al. 2006b).
Labelling patients as having ‘chronic pain’ has negative connotations in clinical practice, imply-
ing a problem patient (Cedraschi et al. 1999), or someone with long-standing psychiatric prob-
lems for whom treatment is likely to be ineffective (American Geriatrics Society Panel on Persistent
Pain in Older Persons 2002). From a patient’s perspective, defining chronic pain based on the
duration of symptoms only gives information on their pain experience up to that point, and gives
no indication of likely pain outcomes (Richardson et al. 2006). This lack of knowledge about
the future course of pain compounds the overall uncertainty of many pain conditions, prevents
planning, and may be a barrier to accepting and coping with pain (Richardson et al. 2006).
Interestingly, research from clinical practice has shown that clinicians (rheumatologists and
chiropractors) actually do not simply use pain duration to define chronicity. Rather, they incor-
porate a broader psychosocial context including symptoms and signs, psychological difficulties,
and working conditions (Cedraschi et al. 1999). And individuals with chronic pain themselves
may be expressing different things when they are reporting chronicity, for example they may take
the wider experience of their symptoms or the severity of the effects of pain on their daily lives
into account, rather than simply referring to the duration of their pain.
Despite problems with purely duration-based approaches, episode duration as reported and
perceived by patients does have some prognostic value and should not be discarded. Patients with
different episode durations have different characteristics and different prognoses, even after
adjustment for other factors (Bot et al. 2005; Dunn and Croft 2006; Demmelmaier et al. 2008).
Episode duration may also influence treatment outcomes, for example back pain patients in a
disability rehabilitation programme with 3–6 months’ duration of sick leave did better than those
with over 6 months’ duration (Sullivan et al. 2008). Classification systems, such as that produced
by the Quebec Task Force (Spitzer et al. 1987), do include duration alongside other prognostic
indicators.
Alternative approaches to defining chronic pain that are consistent with the prognostic
approach developed in this chapter have been proposed and evaluated. As mentioned earlier, the
Multiaxial Assessment of Pain (Turk and Rudy’s Dysfunctional Chronic Pain taxonomy) assesses
chronic pain status in terms of dimensions including activity interference, emotional distress,
pain intensity, and perceived support (Turk and Rudy 1987). The Chronic Pain Grade (Von Korff
et al. 1992) is a widely used method of classifying dysfunctional chronic pain. It was developed on
samples of patients with headache, back pain, and temporomandibular pain, and has since been
used in other conditions such as neck pain (Côté et al. 1998), hand pain (Dziedzic et al. 2007), and
hip pain (Dasch et al. 2008), and in population-based samples (Cassidy et al. 1998; Elliott et al.
1999; Thomas et al. 2004). It incorporates information on pain intensity and pain interference to
classify individuals into one of five severity grades and persistent versus non-persistent pain.
While it considers pain intensity and pain-related interference with activities, it is not multi-
dimensional, as factors now recognized to be important, such as yellow flag psychosocial indica-
tors, are not included.
Overall, chronic pain is characterized by its uncertainty (Richardson et al. 2006), and should
not be seen as an inherently fixed or stable trait, but as being mutable over time (Von Korff and
Dunn 2008). In the following section, we describe an approach to integrating multidimensional
prognostic variables relevant to chronic pain, using this information to predict probabilities of
unfavourable pain outcomes.
2.6 Defining chronic pain by outcome probability

A prognostic approach to defining chronic pain combines information on past and current pain
status, with information on other prognostic variables, into a definition based on outcome prob-
abilities. This approach was initially developed by Von Korff and Miglioretti using data from a
cohort of primary care back pain consulters in Washington State, USA (Von Korff and Miglioretti
2005, 2006), and was subsequently replicated across several pain conditions in independent
studies (Dunn et al. 2008; Thomas et al. 2008; Von Korff and Dunn 2008).
2.7 Development of the approach

The definition of chronic pain used in the development of the approach was clinically significant
pain likely to be present 1 or more years in the future. This outcome (clinically significant pain)
was operationalized using Chronic Pain Grades II to IV (Von Korff et al. 1992) at the follow-up
time points. This identifies people with moderate to severe pain intensity and mild to severe dys-
function 1, 2, or 5 years after the baseline evaluation, which typically occurred about 2–4 weeks
after an index pain visit in a primary care setting.
In the initial work, latent transition regression analysis was used to identify categories of latent
pain severity: no pain, mild pain, moderate pain/limitation, and severe limiting pain. The observed
probabilities of severe pain at the subsequent observation demonstrated the utility of pain sever-
ity measures in predicting risk of future clinically significant pain, as people with severe pain
tended to continue to have severe pain, and people with mild or moderate pain were very unlikely
to develop severe pain. This observation suggested that chronicity could be measured by the
probability of clinically significant pain continuing at a future point in time.
The next step was to consider prognostic variables other than pain severity, to further differen-
tiate people with lower or higher risk of future clinically significant pain. The variables considered
were: depression, measured using the SCL-90-R (Derogatis et al. 1974), the number of days with
pain in the previous 6 months, and the number of other pain sites (headache, abdominal pain,
chest pain, and facial pain). These variables were then combined with ratings of pain intensity and
interference with activities to produce a risk score. Table 2.1 shows the scoring rules used to
calculate the risk score for each patient. The items were scored using a limited number of catego-
ries established by examining distributional properties of individual items, with 0 scores assigned
to low values. It was subsequently assessed whether the predictive accuracy of scoring could be
improved by optimizing the scoring weights for each of the items. It was found that predictive
validity was not materially improved by using regression-based scoring weights.
The ability of the risk score to predict chronic pain at follow-up was then assessed using a
smoothed plot of the baseline risk score against the probability of clinically significant pain 1 year
after the index consultation. Probable chronic pain was defined by an 80% or greater probability
of future clinically significant pain, possible chronic pain as a 50% or greater probability, and
Table 2.1 Scoring rules for estimating prognostic risk score
Item Item value Risk score value

Average pain intensity 0–3 0
4–6 1
7–10 2
Worst pain intensity 0–4 0
5–7 1
8–10 2
Current pain intensity 0–2 0
3–4 1
5–10 2
Interference with usual activities 0–2 0
3–4 1
5–10 2
Interference with work/household activities 0–2 0
3–4 1
5–10 2
Interference with family/social activities 0–2 0
3–4 1
5–10 2
Days of activity limitation due to pain in prior 0–2 0
3 months
3–6 1
7–15 2
16–24 3
25–90 4
SCL-90-R† Depression score <0.50 0
0.50–<1.0 1
1.0–<1.5 2
1.5–<2.0 3
2.0–4.0 4
Number of other pain sites 0 0
1 1
2 2
3 3
4 4
Table 2.1 (continued) Scoring rules for estimating prognostic risk score (Continued)
Item Item value Risk score value

Number of days with index pain in prior 6 months 0–30 0
31–89 1
90–120 2
121–160 3
161–180 4
Total risk score 0–28
* Risk score values are summed across items.
‡ Measured on 0–10 numerical rating scale.
† Symptom Checklist-90-R (Derogatis et al. 1974).
intermediate risk as a 20% or greater probability of future clinically significant pain, with low risk
below 20%. Cut-points on the risk score were then determined from the probability plot to define
chronic pain status in terms of the probability of having clinically significant pain at a future time
point. Low risk was determined as risk scores of 0–7 (31% of the sample), intermediate risk from
8–15 (43%), possible chronic pain as a score of 16–21 (20%) and probable chronic pain as a risk
score of 22 or more (6% of the sample). These categories strongly predicted clinically significant
pain at 1 year, as would be expected (see Table 2.2), but they also strongly predicted clinically
significant pain at the 2-year and 5-year follow-ups.
These findings provided initial support for the predictive validity of a prognostic approach to
defining chronic pain, and supported a multivariable approach to defining chronic pain in terms
of outcome probabilities.
2.8 Generalizability of the prognostic approach to other samples

Since the initial development of the prognostic approach, it has been tested in a number of other
situations. Dunn et al. (2008) investigated the generalizability of the prognostic approach among
another low back pain sample, this time in UK primary care consulters. The researchers used the
same definitions of chronic pain as the original study, and had the same pain severity variables.
They also measured the same prognostic domains, although the measurement instruments used
Table 2.2 Probability of clinically significant pain at follow-up (percent of each risk score group with
chronic pain grade II–IV at follow-up)
Baseline risk score group

Sample Low risk Intermediate Possible chronic Probable chronic
(0–7) risk (8–15) pain (16–21) pain (22+)
US LBP 1 year1 10.9 32.1 58.7 82.1
UK LBP 1 year2 22.9 29.9 51.1 90.3
UK knee 18 months3 20.3 51.1 84.3 88.6
LBP, low back pain.
1 Von Korff and Miglioretti (2005).
2 Dunn et al. (2008).
3 Thomas et al. (2008).
to assess these domains were different, e.g. the Hospital Anxiety and Depression Scale (Zigmond
and Snaith 1983) was used to measure depression, rather than the SCL-90-R (Derogatis et al.
1974), and time since the last pain-free month (Dunn and Croft 2006) was used to measure pain
duration, rather than days with pain in the prior 6 months. The results showed that the prognos-
tic approach was applicable to UK primary care patients. For example, over 90% of people defined
with probable chronic back pain actually had clinically significant pain a year later (see Table 2.2).
These results also supported the feasibility of substituting different measures of the prognostic
indicators when the originally tested prognostic indicators are not available, suggesting that these
results are robust across alternative measures of the same prognostic indicator.
Von Korff and Dunn (2008) investigated the use of the prognostic approach in another pri-
mary care low back pain sample in the USA. Again, the results showed predictive validity of the
approach for predicting chronic pain. They also showed that the prognostic risk scores signifi-
cantly predicted unemployment and long-term opioid use, as well as the pain itself, with ‘proba-
ble’ chronic back pain sufferers having 15 times the odds of being unemployed, and three times
the odds of being on long-term opioid therapy compared to people in the low-risk score group.
The prognostic approach has also been evaluated in conditions other than back pain. In sam-
ples of primary care patients with headache and orofacial pain, it was shown to have good predic-
tive validity with similar cut-points to those initially proposed (Von Korff and Dunn 2008).
Thomas et al. (2008) have explored the use of the prognostic Risk Score in a general population
sample of older adults with knee pain in the UK. They showed that the approach had good prog-
nostic ability in this population as well. For example, 89% of people defined with probable
chronic knee pain at baseline had clinically significant pain a year later (see Table 2.2). The find-
ings were consistent when considered across different time-frames. This sample was population-
based, and therefore contained a mixture of people seeking healthcare for their knee pain, and
those who had not sought care. This work suggested that risk score cut-points may need to be
modified for use in general population samples, with lower cut-points being employed in general
population samples than in primary care samples.
These studies have shown that using a multifactorial prognostic approach to defining chronic
pain is applicable to individuals with back pain, headache, orofacial pain, and knee pain, in both
US and UK settings, and in primary care and population settings. While the approach appears to
be generalizable, the absolute risk score cut-points did differ between primary care and general
population samples when the likelihood of clinically significant pain at follow-up was considered.
Thus, while the general approach is reproducible and useful, it may be that cut-points (e.g. for
probable or possible chronic pain) need to be determined for particular situations. Specifically,
primary care samples assessed in proximity to an initial visit are assessed in a period of pain exac-
erbation, whereas general population samples are assessed at an arbitrary point in time in the
course of their pain condition. For this reason, lower cut-points for possible and probable chronic
pain may be appropriate in a general population sample than in patients seeking treatment for a
pain condition. Additional research on this issue is needed before a prognostic risk score is used
to estimate the prevalence of chronic pain in a general population sample. With that caveat,
research to date suggests that the risk score approach not only predicts clinically significant pain,
but also behavioural outcomes such as unemployment and medication use, supporting its utility
for predicting a range of outcomes that are associated with chronic pain.
Application to the general population is useful for estimating the prevalence of chronic pain in
the population-at-large. Applying the original cut-points to a previous general population study
(Von Korff et al. 1988, 1993b), the prevalence of probable chronic back pain, headache, or facial
pain was approximately 1%. An additional 8% of people had possible chronic pain using this
definition. When risk score cut-points defined for general population samples were employed
(18 for probable chronic pain and 12 for possible chronic pain), the prevalence of prognostically
defined chronic pain was 5% for probable chronic pain and 19% for possible chronic pain (or
24% of the population having possible or probable chronic pain considering back pain, headache
and orofacial pain in combination).
2.9 Comparison of prognostic and duration approaches

One important part of the assessment of the prognostic approach to defining chronic pain was to
compare it to more traditional duration-based approaches. In the UK back pain sample, classify-
ing the sample by duration showed that 52% of people with over 3 months of pain had a poor
outcome, whereas classification using the risk score approach indicated that 67% of people with
possible or probable chronic pain had a poor outcome (Dunn et al. 2008). Extending this to the
US back pain, headache, and orofacial pain samples, the risk score was consistently a better pre-
dictor of future clinically significant pain than pain days in the previous year (Von Korff
and Dunn 2008). In the population-based knee pain sample, Risk score again proved to have
improved prognostic ability when compared to pain duration alone, with the area under the
receiver-operating characteristic curve being 0.82 (95% confidence interval [CI]: 0.78, 0.85) for
the prognostic approach compared to 0.56 (CI: 0.51, 0.61) for pain duration alone (Thomas et al.
2008), where area under of the curve of 0.50 represents no better than chance prediction.
The evidence shows that the prognostic approach to defining chronic pain shows improved
predictive abilities when compared to using measures of pain duration alone, but it has a number
of conceptual advantages as well. Traditional duration-based approaches group people into acute
and chronic categories. Research into episode duration has shown that the reality is more of a
continuum, with little support for the traditional cut-points (Dunn and Croft 2006). Evaluation
of the risk score approach has shown that it provides a continuum, where increasing scores indi-
cate increasing risk of future clinically significant pain. The use of cut-points can provide a useful
categorization of these scores, and the labels of probable and possible chronic pain are appropri-
ate for a condition where uncertainty about clinical course is inherent. Labelling patients as
‘chronic’ without indicating the potential for change can have potentially stigmatizing effects. Of
greatest importance, the risk score integrates information on pain severity, pain duration, and
other prognostic indicators (e.g. depression, number of pain sites), summarizing the results of a
multidimensional assessment of chronic pain status, consistent with the biopsychosocial model,
in terms of a single risk score that quantifies long-term outcome probabilities.
2.10 Clinical implications

This chapter challenges the notion that acute and chronic pain, defined by the length of time the
pain has been present, should be a predominant basis for prognostic classification of pain patients.
For all patients, whether pain duration is brief or long, more accurate prediction of likely out-
come than that based on pain duration alone is possible. Persons with recent onset pain who have
multiple risk indicators may have a less favourable prognosis than persons with long-lasting pain
who have few other unfavourable prognostic indicators. With a more accurate prognostic evalu-
ation, Patient A (see Box 2.1) would be found to be at high risk of chronic dysfunction and steps
could be taken to manage his pain, increase his activity levels, and get the patient back to work as
soon as possible. Similarly, people who have had their pain for a long time may still have a high
likelihood of a favourable outcome if they have few unfavourable prognostic characteristics
(see Patient B in Box 2.1). With an accurate prognostic evaluation, the lack of risk factors would
be clarified, and the clinician could recognize that the patient is likely to recover to his normal
manageable pain levels soon. The patient could be reassured to expect improvement, and offered
short-term palliative remedies if desired by the patient. If Patient B’s physician focused on pain
duration alone, the patient might be offered more aggressive or longer-term medical treatments
than warranted. The prognostic approach provides a useful basis for assessing patients, while
acknowledging the inherent uncertainty of pain and its variability over time.
For clinicians, using a prognostic approach to defining chronic pain could help to identify tar-
gets for intervention early in the clinical course of a pain condition, before social role dysfunction
has become entrenched. The prognostic approach focuses attention on what can be done to
improve the likelihood of a favourable outcome rather than labelling patients as intractable and
unlikely to change. For patients with a short-term problem and no other risk factors, the approach
may give a clinician more confidence in offering reassurance and short-term pain relief. The
prognostic approach can also provide information to facilitate better management of back pain
patients with long term problems. For Patient C (Box 2.1), a clinician using a duration-based
definition of chronic might simply tell the person that she is not likely to get better. But using a
prognostic approach, prognostic indicators might provide a basis for discussion of how to increase
the likelihood of a more favourable outcome, by increasing activity levels in work and family life.
This multidimensional approach provides a basis for communicating with patients about what
can be done to reduce risks of an unfavourable pain outcome rather than an exclusive focus on
medical treatments to ‘cure’ the underlying condition.
The prognostic approach is consistent with current pain management guidelines, which take a
biopsychosocial approach and encourage assessment of a wide range of factors. Using a probabi-
listic definition incorporates the assumption that the future course of symptoms and prognosis is
inherently uncertain, while at the same time identifying possible targets for intervention. This is
in contrast to labelling patients as ‘chronic’, which provides little information to the patient or
clinician, and can have connotations of an intractable and enduring problem about which little
can be done. Of course, a probabilistic statement about pain prognosis could also be used to label
patients as unlikely to improve, or hopeless, but that is contrary to our intent. The term chronic
pain is qualified to reflect a possible or probable outcome, not a certainty. And, each patient’s risk
score is quantified in terms of prognostic factors that can be modified even if pain cannot be
eliminated. For these reasons, we believe a prognostic approach to chronic pain is less likely to be
used to label and inadvertently stigmatize patients, and more likely to be used to focus attention
on a broader set of prognostic factors that can be influenced to improve patient outcomes than
pain alone.
Research on a prognostic approach to defining chronic pain has used a range of measurement
instruments to assess prognostic factors. It is entirely appropriate to substitute different measures
of pain intensity, interference, number of pain sites, pain duration, and psychological distress for
those used in prior research. In fact, results have been generally comparable when different meas-
ures have been employed. Given the time and cost constraints usually present in clinical practice,
briefer instruments may provide the necessary information without taking up too much time.
However, the use of risk scores is increasingly feasible in settings that use electronic medical
records, and various risk scores are being evaluated for clinical use in many different areas of
clinical management of chronic disease and prevention (Ferrer et al. 2005). There is a need for
additional research which tests the prognostic ability of alternative self-report measures for pre-
dicting long-term pain outcomes to improve the predictive power of this approach. It would also
be entirely consistent with a prognostic approach to defining chronic pain to employ psycho-
physical measures (e.g. diffuse noxious inhibitory control) or genetic markers, if these variables
were found to improve prediction of long-term pain outcomes. Research assessing the prognostic
value of measures of pain coping and somatosensory functioning could also be evaluated to assess
their contribution of prediction of future clinically significant pain. Such research could not only
enhance the accuracy of a prognostic classification of chronic pain, it could also shed light on the
nature and underlying mechanisms of chronic pain.
To some, viewing a patient in a broader context than the duration of their pain alone may seem
obvious. Many clinicians intuitively incorporate indicators of prognosis into their assessment
and management of back pain patients already. The adoption of a more structured approach to
defining chronic pain by multiple prognostic indicators could result in these approaches being
used more consistently across settings and healthcare professionals, providing an evidence-
based foundation for patient care. Thus, a prognostic approach to defining chronic pain has the
potential to provide a theory-based and evidence-based approach to classification of diverse
chronic pain conditions for epidemiological, clinical, and basic research. It could also provide a
useful classification for clinical practice, but further research is needed to examine applicability to
clinical care.
2.11 Conclusions
In this chapter, traditional duration-based approaches to defining chronic pain have been
contrasted with multifactorial models of pain based in the biopsychosocial model. A new approach
to defining chronic pain based on outcome probabilities that are a function of multiple prognos-
tic indicators is proposed. We believe that this approach offers improved predictive validity, as
well as better conceptual links to the biopsychosocial model, than approaches based on pain
duration alone. It is consistent with current guidelines for managing chronic pain, and is likely to
provide more useful information to patients and clinicians.
As an alternative to the IASP definition of chronic pain, we propose that chronic pain be
defined by the risk that clinically significant pain and associated dysfunction will be present at a
future time point, where the likelihood of future pain and dysfunction is predicted by multiple
prognostic factors. The prognostic indicators currently investigated for this purpose include pain
severity, pain duration, number of anatomical pain sites, the severity of pain-related activity
limitations, and psychological distress. However, these should not be considered as a restrictive or
exhaustive list. Other prognostic indicators should be evaluated as they are identified or as the
purpose fits, including psychological, psychophysical, and genetic variables. Such factors should
be investigated and incorporated while keeping in mind that any assessment has to be acceptable
and practical in the setting where it will be applied. The current approach, and any future exten-
sions of the approach, will require testing in clinical practice to establish validity and utility out-
side the research setting. This definition of chronic pain based on risk of future clinically significant
pain defines chronic pain status in probabilistic terms (possible and probable chronic pain), indi-
cating to patients and providers alike that change in pain status over time, both improvement and
deterioration, is common. This may help patients and clinicians to view chronic pain as a dynamic
process, with potential for change, rather than a label applied to patients deemed unlikely
to improve or hopeless. The fact that outcomes are a function of multiple factors calls attention
to the possibility that chronic pain outcomes can be improved in ways other than reducing
pain alone.
The proposed prognostic approach should not be seen as representing a linear progression over
time, although changes in any of the variables included in the risk score may indicate a transition
between pain status categories. For example, an increase in the level of depression reported could
move someone from having possible to probable chronic pain. Equally, a reduction in pain inten-
sity could move someone from probable to possible chronic pain. Changes in the number of
anatomical sites with pain or pain interference could also trigger shifts between categories of
pain status. There is no underlying implication in the prognostic approach that pain sufferers
progress from low risk through to probable chronic pain over time. This contrasts with a purely
duration-based approach which, by its definition, implies deterioration in status or worsening
prognosis over time, but is not based on strong empirical evidence and does not give further
information on what should be done to improve outcomes other than eliminating pain.
Using a definition of chronic pain based on pain duration alone seems inappropriate when
biomedical, psychological and social factors are all accepted contributors to the experience,
assessment, mechanisms and management of chronic pain. A definition based on prognosis,
quantified in terms of outcome probabilities, encompasses both the complexity and the variability
of pain, and appears to be more consistent with the biopsychosocial model of pain.
References
ACC and the National Health Committee. (1997). New Zealand Acute Low Back Pain Guide. Wellington:
Accident Rehabilitation & Compensation Insurance Corporation of New Zealand and the National
Health Committee.
Airaksinen, O., Brox, J.I., Cedraschi, C., et al. (2006). European guidelines for the management of chronic
nonspecific low back pain. Eur Spine J, 15, S192–S300.
American Geriatrics Society Panel on Persistent Pain in Older Persons. (2002). The management of
persistent pain in older persons. J Am Geriatr Soc, 50, S205–24.
Barsky, A.J., Fama, J.M., Bailey, E.D., Ahern, D.K. (1998). A prospective 4- to 5-year study of DSM-III-R
hypochondriasis. Arch Gen Psychiatry, 55, 737–44.
Bonica, J.J. (1953). The Management of Pain. Philadelphia, PA: Lea & Febiger.
Bonica, J.J. (1990). General considerations of chronic pain. In Bonica J.J. (ed.) The Management of Pain,
pp. 159–79. Philadelphia, PA: Lea and Febiger.
Bot, S.D., van der Waal, J.M., Terwee, C.B., et al. (2005). Predictors of outcome in neck and shoulder
symptoms: a cohort study in general practice. Spine, 30, E459–70.
Carragee, E.J., Alamin, T.F., Miller, J.L., Carragee, J.M. (2005). Discographic, MRI and psychosocial
determinants of low back pain disability and remission: a prospective study in subjects with benign
persistent back pain. Spine J, 5, 24–35.
Cassidy, J.D., Carroll, L.J., Côté, P. (1998). The Saskatchewan health and back pain survey. The prevalence
of low back pain and related disability in Saskatchewan adults. Spine, 23, 1860–6.
Cedraschi, C., Nordin, M., Nachemson, A.L., Vischer, T.L. (1998). Health care providers should use a
common language in relation to low back pain patients. Baillière’s Clin Rheumatol, 12, 1–15.
Cedraschi, C., Robert, J., Goerg, D., Perrin, E., Fischer, W., Vischer, T.L. (1999). Is chronic non-specific low
back pain chronic? Definitions of a problem and problems of a definition. Br J Gen Pract, 49, 358–62.
Celestin, J., Edwards, R.R., Jamison, R.N. (2009). Pretreatment psychosocial variables as predictors of
outcomes following lumbar surgery and spinal cord stimulation: a systematic review and literature
synthesis. Pain Med, 10, 639–53.
Costa, L.C., Maher, C.G., McAuley, J.H., et al. (2009). Prognosis for patients with chronic low back pain:
inception cohort study. BMJ, 339, b3829.
Côté, P., Cassidy, J.D., Carroll, L. (1998). The Saskatchewan Health and Back Pain Survey. The prevalence
of neck pain and related disability in Saskatchewan adults. Spine, 23, 1689–98.
Critchley, D.J., Ratcliffe, J., Noonan, S., Jones, R.H., Hurley, M.V. (2007). Effectiveness and
cost-effectiveness of three types of physiotherapy used to reduce chronic low back pain disability: a
pragmatic randomized trial with economic evaluation. Spine, 32, 1474–81.
Croft, P.R. (1999). Diagnosing regional pain: the view from primary care. Baillière’s Best Pract Res Clin
Rheumatol, 13, 231–42.
Croft, P.R., Macfarlane, G.J., Papageorgiou, A.C., Thomas, E., Silman, A.J. (1998). Outcome of low back
pain in general practice: a prospective study. BMJ, 316, 1356–9.
Currie, S.R. and Wang, J. (2004). Chronic back pain and major depression in the general Canadian
population. Pain, 107, 54–60.
Dasch, B., Endres, H.G., Maier, C., et al. (2008). Fracture-related hip pain in elderly patients with proximal
femoral fracture after discharge from stationary treatment. Eur J Pain, 12, 149–56.
De Gucht, V. (2003). Stability of neuroticism and alexithymia in somatization. Compr Psychiatry, 44, 466–71.
de Vet, H.C., Heymans, M.W., Dunn, K.M., et al. (2002). Episodes of low back pain: a proposal for
uniform definitions to be used in research. Spine, 27, 2409–16.
Demmelmaier, I., Lindberg, P., Asenlof, P., Denison, E. (2008). The associations between pain intensity,
psychosocial variables, and pain duration/recurrence in a large sample of persons with nonspecific
spinal pain. Clin J Pain, 24, 611–19.
Derogatis, L.R., Lipman, R.S., Rickels, K., Uhlenhuth, E.H., Covi, L. (1974). The Hopkins Symptom
Checklist (HSCL): a self-report symptom inventory. Behav Sci, 19, 1–15.
Deyo, R.A. (1993). Practice variations, treatment fads, rising disability. Do we need a new clinical research
paradigm? Spine, 18, 2153–62.
Dionne, C.E., Dunn, K.M., Croft, P.R., et al. (2008). A consensus approach toward the standardization of
back pain definitions for use in prevalence studies. Spine, 33, 95–103.
Dunn, K.M., and Croft, P.R. (2006). The importance of symptom duration in determining prognosis. Pain,
121, 126–32.
Dunn, K.M., de Vet, H.C., Hooper, H., Ong, B.N., Croft, P.R. (2006a). Measurement of back pain episode
inception in questionnaires: a study combining quantitative and qualitative methods. J Musculoskeletal
Pain, 14, 29–37.
Dunn, K.M., Jordan, K., Croft, P.R. (2006b). Characterising the course of low back pain: a latent class
analysis. Am J Epidemiol, 163, 754–61.
Dunn, K.M., Croft, P.R., Main, C.J., Von Korff, M. (2008). A prognostic approach to defining chronic
pain: replication in a UK primary care low back pain population. Pain, 135, 48–54.
Dziedzic, K., Thomas, E., Hill, S., Wilkie, R., Peat, G., Croft, P.R. (2007). The impact of musculoskeletal
hand problems in older adults: findings from the North Staffordshire Osteoarthritis Project
(NorStOP). Rheumatology (Oxford), 46, 963–7.
Edwards, R.R. (2005). Individual differences in endogenous pain modulation as a risk factor for chronic
pain. Neurology, 65, 437–43.
Elliott, A.M., Smith, B.H., Penny, K.I., Smith, W.C., Chambers, W.A. (1999). The epidemiology of chronic
pain in the community. Lancet, 354, 1248–52.
Elliott, A.M., Smith, B.H., Hannaford, P.C., Smith, W.C., Chambers, W.A. (2002). Assessing change in
chronic pain severity: the chronic pain grade compared with retrospective perceptions. Br J Gen Pract,
52, 269–74.
Ferrer, J., Neyro, J.L., Estevez, A. (2005). Identification of risk factors for prevention and early diagnosis of
a-symptomatic post-menopausal women. Maturitas, 52(Suppl 1), S7–22.
Fishbain, D.A., Lewis, J.E., Gao, J., Cole, B., Steele, R.R. (2009). Is chronic pain associated with
somatization/hypochondriasis? An evidence-based structured review. Pain Pract, 9, 449–67.
Granot, M., Granovsky, Y., Sprecher, E., Nir, R.R., Yarnitsky, D. (2006). Contact heat-evoked temporal
summation: tonic versus repetitive-phasic stimulation. Pain, 122, 295–305.
Granot, M., Weissman-Fogel, I., Crispel, Y., et al. (2008). Determinants of endogenous analgesia
magnitude in a diffuse noxious inhibitory control (DNIC) paradigm: do conditioning stimulus
painfulness, gender and personality variables matter? Pain, 136, 142–9.
Grotle, M., Vøllestad, N.K., Veierød, M.B., Brox, J.I. (2004). Fear-avoidance beliefs and distress in relation
to disability in acute and chronic low back pain. Pain, 112, 343–52.
Hartvigsen, J., Christensen, K., Frederiksen, H., Pedersen, H.C. (2004). Genetic and environmental
contributions to back pain in old age: a study of 2,108 Danish twins aged 70 and older. Spine, 29, 897–901.
Henschke, N., Maher, C.G., Refshauge, K.M., et al. (2008). Prognosis in patients with recent onset low back
pain in Australian primary care: inception cohort study. BMJ, 337, a171.
Henschke, N., Maher, C.G., Refshauge, K.M., et al. (2009). Prevalence of and screening for serious spinal
pathology in patients presenting to primary care settings with acute low back pain. Arthritis Rheum, 60,
3072–80.
Hestbaek, L., Leboeuf-Yde, C., Manniche, C. (2003). Low back pain: what is the long-term course?
A review of studies of general patient populations. Eur Spine J, 12, 149–65.
Hestbaek, L., Iachine, I.A., Leboeuf-Yde, C., Kyvik, K.O., Manniche, C. (2004). Heredity of low back pain
in a young population: a classical twin study. Twin Res, 7, 16–26.
International Association for the Study of Pain Subcommittee onTaxonomy (1986). Classification of
chronic pain. Descriptions of chronic pain syndromes and definitions of pain terms. Pain Suppl,
3, S1–226.
International Headache Society H.C.S. (2004). Definition of terms. Cephalalgia, 24, 150–51.
Kamper, S.J., Rebbeck, T.J., Maher, C.G., McAuley, J.H., Sterling, M. (2008). Course and prognostic
factors of whiplash: A systematic review and meta-analysis. Pain, 138, 617–29.
Katz, N.P., Ju, W.D., Krupa, D.A., et al. (2003). Efficacy and safety of rofecoxib in patients with chronic
low back pain: results from two 4-week, randomized, placebo-controlled, parallel-group, double-blind
trials. Spine, 28, 851–59.
Keefe, F.J., Rumble, M.E., Scipio, C.D., Giordano, L.A., Perri, L.M. (2004). Psychological aspects of
persistent pain: current state of the science. J Pain, 5, 195–211.
Keeley, P., Creed, F., Tomenson, B., Todd, C., Borglin, G., Dickens, C. (2008). Psychosocial predictors of
health-related quality of life and health service utilisation in people with chronic low back pain. Pain,
135, 142–50.
Kendall, N.A.S., Burton, A.K., Main, C.J., Watson, P.J. (2009). Tackling musculoskeletal problems, a guide
for the clinic and workplace: Identifying obstacles using the psychosocial flags framework. London:
The Stationery Office.
Klein, T., Magerl, W., Hopf, H-C., Sandkühler, J., Treede, R-D. (2004). Perceptual Correlates of
Nociceptive Long-Term Potentiation and Long-Term Depression in Humans. J Neurosci, 24, 964–71.
Loeser, J.D. and Melzack, R. (1999). Pain: an overview. Lancet, 353, 1607–9.
Main, C.J. and Spanswick, C. (2003). Pain management: an interdisciplinary approach. Edinburgh:
Churchill Livingstone.
McBeth, J., Silman, A.J., Gupta, A., et al. (2007). Moderation of psychosocial risk factors through
dysfunction of the hypothalamic-pituitary-adrenal stress axis in the onset of chronic widespread
musculoskeletal pain: findings of a population-based prospective cohort study. Arthritis Rheum, 56,
360–71.
Nachemson, A.L. and Andersson, G.B. (1982). Classification of low-back pain. Scand J Work Environ
Health, 8, 134–6.
Nicholas, M.K., Linton, S.J., Watson, P.J., Main, C.J., “Decade of the Flag” Working Group (2011). Early
identification and management of psychological risk factors (“yellow flags”) in patients with low back
pain: a reappraisal. Phys Therapy, 91, 737–53.
Papageorgiou, A.C., Croft, P.R., Thomas, E., Ferry, S., Jayson, M.I., Silman, A.J. (1996). Influence of
previous pain experience on the episode incidence of low back pain: results from the South Manchester
Back Pain Study. Pain, 66, 181–5.
Pincus, T., Burton, A.K., Vogel, S., Field, A.P. (2002). A systematic review of psychological factors as
Pitcher, M.H., Ribeiro-da-Silva, A., Coderre, T.J. (2007). Effects of inflammation on the ultrastructural
localization of spinal cord dorsal horn group I metabotropic glutamate receptors. J Comp Neurol, 505,
412–23.
Raspe, H., Huppe, A., Matthis, C. (2003). Theories and models of chronicity: on the way to a broader
definition of chronic back pain. Schmerz, 17, 359–66.
Reilingh, M.L., Kuijpers, T., Tanja-Harfterkamp, A.M., van der Windt, D.A. (2008). Course and prognosis
of shoulder symptoms in general practice. Rheumatology, 47, 724–30.
Richardson, J.C., Ong, B.N., Sim, J. (2006). Remaking the future: contemplating a life with chronic
widespread pain. Chronic Illness, 2, 209–18.
Samanta, J., Kendall, J., Samanta, A. (2003). 10-minute consultation. Chronic low back pain. BMJ, 326, 535.
Shaw, W.S., van der Windt, D.A., Main, C.J., Loisel, P., Linton S.J. (2009). Early patient screening and
intervention to address individual-level occupational factors (‘blue flags’) in back disability. J Occup
Rehabil, 19, 64–80.
Smeets, R.J. and Wittink, H. (2007). The deconditioning paradigm for chronic low back pain unmasked?
Pain, 130, 201–202.
Smeets, R.J., Vlaeyen, J.W., Hidding, A., et al. (2006). Active rehabilitation for chronic low back pain:
Cognitive-behavioral, physical, or both? First direct post-treatment results from a randomized
controlled trial [ISRCTN22714229]. BMC Musculoskelet Disord, 7, 5.
Smith, B.H., Elliott, A.M., Hannaford, P.C., Chambers, W.A., Smith, W.C. (2004). Factors Related to the
Onset and Persistence of Chronic Back Pain in the Community: Results From a General Population
Follow-up Study. Spine, 29, 1032–40.
Spies-Dorgelo, M.N., van der Windt, D.A., Prins, A.P., Dziedzic, K.S., van der Horst, H.E. (2008). Clinical
course and prognosis of hand and wrist problems in primary care. Arthritis Rheum, 59, 1349–57.
Spinhoven, P., Ter Kuile, M., Kole-Snijders, A.M., Hutten, M.M., Den Ouden, D.J., Vlaeyen, J.W. (2004).
Catastrophizing and internal pain control as mediators of outcome in the multidisciplinary treatment
of chronic low back pain. Eur J Pain, 8, 211–19.
Spitzer, W.O., LeBlanc, F.E., Dupuis, M., et al. (1987). Scientific approach to the assessment and
management of activity-related spinal disorders: A monograph for clinicians. Report of the Quebec
Task Force on Spinal Disorders. Spine 12(7 Supp ), S1–S59.
Staud, R. and Spaeth, M. (2008). Psychophysical and neurochemical abnormalities of pain processing in
fibromyalgia. CNS Spectr, 13, 12–17.
Staud, R., Vierck, C.J., Cannon, R.L., Mauderli, A.P., Price, D.D. (2001). Abnormal sensitization and
temporal summation of second pain (wind-up) in patients with fibromyalgia syndrome. Pain, 91, 165–75.
Steenstra, I.A., Verbeek, J.H., Heymans, M.W., Bongers, P.M. (2005). Prognostic factors for duration of
sick leave in patients sick listed with acute low back pain: a systematic review of the literature.
Occupational and Environmental Medicine, 62, 851–60.
Sullivan, M.J., Adams, H., Tripp, D., Stanish, W.D. (2008). Stage of chronicity and treatment response in
patients with musculoskeletal injuries and concurrent symptoms of depression. Pain, 135, 151–9.
Thomas, E., Peat, G., Harris, L., Wilkie, R., Croft, P.R. (2004). The prevalence of pain and pain interference
in a general population of older adults: cross-sectional findings from the North Staffordshire
Osteoarthritis Project (NorStOP). Pain, 110, 361–68.
Thomas, E., Dunn, K.M., Mallen, C.D., Peat, G. (2008). A prognostic approach to chronic pain: application
to knee pain in older adults. Pain, 139, 389–97.
Turk, D.C. and Rudy, T.E. (1987). Towards a comprehensive assessment of chronic pain patients. Behav
Res Ther, 25, 237–49.
Turner, J.A., Franklin, G., Fulton-Kehoe, D., et al. (2008). ISSLS prize winner: Early predictors of
chronic work disability: a prospective, population-based study of workers with back injuries. Spine,
33, 2809–18.
Von Korff, M. (1994). Studying the natural history of back pain. Spine, 19, 2041S–2046S.
Von Korff, M. and Dunn, K.M. (2008). Chronic pain reconsidered. Pain, 138, 267–76.
Von Korff, M. and Miglioretti, D.L. (2005). A prognostic approach to defining chronic pain. Pain,
117, 304–13.
Von Korff, M. and Miglioretti, D.L. (2006). A prospective approach to defining chronic pain. In Flor H.,
Kalso E., Dostrovsky J.O. (eds.) Proceedings of the 11th World Congress on Pain. pp. 761–9. Seattle, WA:
IASP Press.
Von Korff, M., Dworkin, S.F., Le Resche, L., Kruger, A. (1988). An epidemiologic comparison of pain
complaints. Pain, 32, 173–83.
Von Korff, M., Ormel, J., Keefe, F.J., Dworkin, S.F. (1992). Grading the severity of chronic pain. Pain,
50, 133–49.
Von Korff, M., Deyo, R.A., Cherkin, D.C., Barlow, W. (1993a). Back pain in primary care. Outcomes at
1 year. Spine, 18, 855–62.
Von Korff, M., Le Resche, L., Dworkin, S.F. (1993b). First onset of common pain symptoms: a prospective
study of depression as a risk factor. Pain, 55, 251–8.
Waddell, G. (1998). The Back Pain Revolution. Edinburgh: Churchill Livingstone.
Werneke, M. and Hart, D.L. (2001). Centralization phenomenon as a prognostic factor for chronic low
back pain and disability. Spine, 26, 758–64.
Woolf, C.J. and McMahon, S.B. (1985). Injury-induced plasticity of the flexor reflex in chronic decerebrate
rats. Neuroscience, 16, 395–404.
Zigmond, A.S. and Snaith, R.P. (1983). The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand,
67, 361–70.
Part 2
Risk Factors of Chronic

Back Pain and Disability:
Biological Mechanisms
Chapter 3
Genetic Factors Modulating Chronic

Back Pain
Julia Metzner and Irmgard Tegeder
3.1 Introduction
The manifestation of back pain is contributed by structural, psychosocial, and occupational
influences (Hartvigsen et al. 2004). Biochemical and inflammatory factors modulate the transi-
tion of acute towards chronic pain and genetic factors may impact on any of these factors.
Research has been mainly focused on genes that determine bone and cartilage structure and
are accompanied by morphological signs in magnetic resonance imaging (MRI). Genetic associa-
tions were found for disc height narrowing, disc herniation, and different definitions of back
pain, such as duration of the worst back pain episode and hospitalization for back problems
(Battie et al. 2007). The heritability estimates for these back pain variables ranged from 30–45%
(Battie et al. 2007).
However, only a minority of the genetic influences was caused by genes affecting disc degen-
eration suggesting that genes involved in pain perception, signalling, and psychological process-
ing (Foulkes and Wood 2008), and genetic variants of immune genes (Solovieva et al. 2004b)
contribute to the proportion of heritability of chronic back pain. The genetic variability in pain
signalling pathways contributes to the variance in pain sensitivity and the individual response to
treatment strategies. In this chapter we will summarize genetic factors that specifically modify
intervertebral disc stability and pain signalling that may independently impact on the risk of
developing chronic back pain.
3.2 Polymorphisms associated with subtle modulations of

pain sensitivity
MRI signs of bone and cartilage structure are weakly predictive for chronic back pain suggesting
that interindividual differences in pain sensation and processing contribute to the individual risk
of chronic back pain. Genetic variants in genes modifying pain sensation and signalling are likely
to explain a certain proportion of the high heritability estimates for chronic back pain. Only
recently it has been increasingly recognized that some frequent genetic polymorphisms are asso-
ciated with modulations of pain sensitivity or the development of hyperalgesia. These frequent
polymorphisms are not the cause of serious diseases but may change the risk for chronic pain
including chronic back pain. Associations between candidate genes and chronic back pain with
lumbar root pain were evaluated in patients following lumbar disc surgery (Tegeder et al. 2006).
The patients suffered from serious long-lasting radicular sciatic pain before surgery. In the first
2 years after surgery, pain scores for frequency and intensity at rest and during walking were
recorded. The statistical analysis revealed that patients with a defined set of single nucleotide
polymorphisms (SNPs) in the gene of the GTP cyclohydrolase 1 (GCH1) had consistently less
pain than non-carriers and a significantly better outcome after surgery (Tegeder et al. 2006). The
polymorphisms in GCH1 that were associated with reduced pain constitute a certain haplotype,
referred to as ‘pain-protective’ haplotype of GCH1, with a frequency of about 16% in the Caucasian
population. The GTP cyclohydrolase is the rate-limiting enzyme in the synthesis of the enzyme
cofactor, tetrahydrobiopterin which is essential for the production of biogenic amines (Blau
and Niederwieser 1985) and nitric oxide (Gross and Levi 1992). It has been confirmed in further
studies that the ‘pain-protective’ haplotype is associated with a reduction of the sensitivity to
mechanical and heat stimulation and particularly the development of hyperalgesia (Tegeder et al.
2008). Functionally, this haplotype prevents the up-regulation of GCH1 and overproduction of
tetrahydrobiopterin that normally occurs upon pro-inflammatory stimulation or peripheral
nerve injury (Tegeder et al. 2006, 2008).
Other genetic polymorphisms have been associated with reduced pain sensitivity (Oertel and
Lötsch 2008) but most studies did not specifically address chronic back pain. However, it appears
reasonable to hypothesize that polymorphisms that modulate other types of chronic pain may
also impact on the liability to develop chronic back pain. Such pain-modulating gene variants
were found for catechol-O-methyltransferase (COMT), transient receptor potential channel A1
(TRPA1), melanocortin-1 receptor (MC1R), fatty acid amide hydrolase (FAAH), and the μ-opioid
receptor (OPRM1). COMT modulates specific aspects of human pain perception and the risk for
developing complex pain conditions including migraine (Diatchenko et al. 2006). COMT metab-
olizes catecholamines and modifies thereby the transmission of dopaminergic, adrenergic and
noradrenergic pathways in the brain. A single nucleotide exchange in the coding region of COMT
leads to an amino acid substitution of valine to methionine at position 158 (V158M), with
impaired translation of COMT-mRNA, reduction of its enzyme activity and reduced thermosta-
bility (Diatchenko et al. 2006). The V158M genotype was primarily associated with the rate of
temporal summation of heat pain (Diatchenko et al. 2006). Other SNPs of COMT exert a greater
influence on baseline nociceptive sensitivity and are inversely correlated with enzyme activity
and the risk of developing the myogenous temporomandibular joint disorder (TMD) (Slade
et al. 2007). TMD is a common pain syndrome with a prevalence of about 10% and 3:1 female to
male ratio with high heritability estimates (Oakley and Vieira 2008; Stohler 2006) and is often
associated with chronic back pain (Oakley and Vieira 2008; Stohler 2006).
Variants of the melanocortin 1 receptor gene (MC1R) were reported to reduce pain sensitivity
(Mogil et al. 2005) although this finding was not reproduced in another human cohort (Liem
et al. 2005). Nevertheless, in humans with red-hair and fair skin, which is the visible phenotype
associated with non-function variants in the MC1R gene, the analgesic efficacy of μ-opioid
agonists was increased in both women and men, whereas κ -agonists reduced pain only in
females (Mogil et al. 2005). Interestingly, female but not male carriers of a genetic variant of the
cold receptor, TRPA1 had increased sensitivity to cold-induced pain compared to carriers of the
wild-type allele (Kim et al. 2006a).
The sensitivity to pain and to opioid analgesia and side effects is known to be modulated by
variants in opioid receptors (Lötsch and Geisslinger 2005). The μ-opioid receptor, encoded by
the OPRM1 gene (OPRM1), is the primary site of action of endogenous and of the most potent
exogenous common clinical opioid analgesics such as morphine, fentanyl, or methadone. A large
number of polymorphisms have been identified in the promoter region, in exons and introns
of OPRM1 (Ikeda et al. 2005; Lötsch and Geisslinger 2005). So far, most information has been
accumulated about the OPRM1 118A>G SNP in exon 1 leading to an amino acid substitution
from asparagine to aspartate at position 40 of the protein thereby deleting a putative extracellu-
lar glycosylation site. This SNP is highly prevalent in the population with an allele frequency
of approximately 16% in Caucasians. At the molecular level, it is thought to either decrease
GENETIC FACTORS MODULATING CHRONIC BACK PAIN 45
μ-opioid receptor expression (Zhang et al. 2005) or, in a brain-region dependent manner,
agonist-stimulated receptor signalling (Oertel et al. 2008). With respect to pain sensitivity, carri-
ers of this variant displayed higher pain thresholds (Fillingim et al. 2005) or lower cortical
responses to experimental pain stimuli (Lötsch et al. 2006b). With respect to pain treatment
with exogenous opioids, the 118A>G SNP significantly reduced the potency of morphine-6-
glucuronide (Lötsch et al. 2002; Skarke et al. 2003), morphine (Skarke et al. 2003), or levometha-
done (Lötsch et al. 2006a) in experimental studies in healthy volunteers that evaluated pupil
constrictory effects of the opioids (Lötsch et al. 2002). The OPRM1 118A>G polymorphism
reduced both analgesic (Oertel and Lötsch 2008; Oertel et al. 2006) and respiratory depressive
(Oertel et al. 2006) effects of alfentanil suggesting that the efficacy of various μ-opioid receptor
agonists is affected by this variant. In the clinical setting, carriers of the OPRM1 118A>G poly-
morphism required significantly higher doses of morphine in the early postoperative period
following knee surgery (Chou et al. 2006b), abdominal hysterectomy (Chou et al. 2006a), or other
major abdominal surgical interventions (Hayashida et al. 2008). Cancer patients homozygous for
the 118 G allele also needed higher morphine doses to achieve pain control (Klepstad et al. 2004),
further modulated when these patients had the wild type COMT Val/Val genotype (Reyes-Gibby
et al. 2007).
Endogenous cannabinoids, cannabis, and its congeners reduce pain and modify emotional
components of pain through agonistic action at peripheral and central cannabinoid-1 CB1 recep-
tors (Agarwal et al. 2007; Manning et al. 2001). Cannabinoids additionally modify the immune
response through cannabinoid CB2 receptors. Tetrahydrocannabinol, one of the constituents of
marihuana has the potential to reduce serious neuropathic pain in patients with multiple sclerosis
or other neuropathic pain syndromes (Svendsen et al. 2004). However, polymorphisms in the
CB1 gene CNR1 have not been associated with specific pain phenotypes, but were associated with
obesity (Aberle et al. 2008; Russo et al. 2007), schizophrenia or efficacy of neuroleptic treatment
(Hamdani et al. 2008; Ujike et al. 2002) and drug and alcohol dependence (Agrawal et al. 2008;
Zhang et al. 2004 ; Zuo et al. 2007 ). Polymorphisms in the CB2 gene CNR2 play a role in
osteoporosis (Karsak et al. 2005) and thereby possibly chronic back pain that is often caused by
osteoporosis of the spine with and without vertebral fractures. Genetic variances in CNR2 might
also modulate the susceptibility to autoimmune disorders (Sipe et al. 2005) such as rheumatic
diseases. The endocannabinoid anandamide is rapidly metabolized by fatty acid amide hydrolase
(FAAH) (Thomas et al. 1997) and FAAH inhibitors reduce pain due to a prolonged half life of
anandamide (Lichtman et al. 2004). Polymorphisms in the FAAH gene (Doehring et al. 2007)
result in decreased FAAH enzyme catalytic activity (Kim et al. 2006a) and slightly reduced sensi-
tivity to cold pain in experimental settings in healthy volunteers (Kim et al. 2006a). The potential
modulation of the susceptibility to chronic back pain has not been addressed. However, consider-
ing the functions of endocannabinoids on bone formation and density and pain signalling (Ofek
et al. 2006; Tam et al. 2008) it is likely that alterations in their metabolism affect the susceptibility
to chronic back pain.
3.3 Genetic polymorphisms associated with intervertebral

disc disease
Intervertebral disc disease (IDD) is characterized by disc degeneration and herniation and is often
associated with low back pain and lumbar radicular pain due to nerve root compression or
inflammation. Sensory neurons and afferent fibres from multiple spinal cord levels innervate
intervertebral discs (Zhang et al. 2006) explaining the often widespread back pain. An increased
risk of low back pain was found in relation to all signs of disc degeneration (Luoma et al. 2000;
Sakai et al. 2007). However, structural aspects alone are of limited predictive value for chronic
back pain.
Collagen is a major component of the extracellular matrix and regulates cartilage fibril forma-
tion within intervertebral discs. It is a heterotrimeric protein consisting of three α-chains.
Polymorphisms in the genes coding for the α2 and α3 chains of collagen IX, COL9A2, and
COL9A3 were associated with alterations in the mechanical properties of human intervertebral
discs and contribute to the susceptibility for lumbar disc herniation and back pain (Annunen
et al. 1999; Paassilta et al. 2001; Seki et al. 2006). Several IDD associated COL9A2 alleles were
identified. The so called Trp2 allele representing a Gln326Trp amino acid exchange in the α2
chain was associated with premature disc degeneration and back pain in Finnish, Japanese and
Chinese populations (Annunen et al. 1999; Higashino et al. 2007; Jim et al. 2005). The Trp2
variant however was not detected in Germans (Knoeringer et al. 2008) but a high relapse rate
of lumbar disc disease after surgery was detected in carriers of the Gln326Arg variant of
COL9A2 (Knoeringer et al. 2008). A Trp3 variant of the α3 chain was also associated with an
increased risk of disc degeneration (Paassilta et al. 2001). Presumably, the collagen IX variants
reduce the stability of the intervertebral cartilage and thereby increase the risk of disc degenera-
tion and herniation.
Type X1 collagen in the annulus fibrosus and nucleus pulposus also plays a major role in the
stability of the intervertebral disc. It is also composed of three α-chains, α1(XI), α2(XI), and
α3(II), which are encoded by COL11A1, COL11A2, and COL2A1, respectively. Type XI collagen
regulates the diameter of cartilage collagen fibrils. Genetic variants in COL11A1 were associated
with an increased risk of lumbar disc herniation in a Japanese cohort (Mio et al. 2007). A single
nucleotide polymorphism in the coding region of COL11A1 showed the strongest association
with disc herniation. The stability of the transcript of the disease-associated allele was reduced
and the expression level inversely correlated with the severity of disc degeneration (Mio et al.
2007). In a Finnish population, carriers of a deletion splice site variant of COL11A2 encoding the
α2(XI) chain had a lower risk for degenerative spinal stenosis with radicular pain than carriers of
the more abundant ‘wild-type’ allele (Noponen-Hietala et al. 2003).
Carriers of the so called Sp1 promoter variant of type I collagen (COL1A1) were at risk to
develop intervertebral disc degeneration (Pluijm et al. 2004), reduced bone mineral density and
osteoporotic bone fractures particularly in postmenopausal women (Ashford et al. 2001 ;
MacDonald et al. 2001; Pluijm et al. 2004). The Sp1 variant interferes with the binding of the
transcription factor Sp1 resulting in reduced expression of COL1A1 (Grant et al. 1996 ).
Osteoporosis represents a major cause of chronic low back pain (Schneider et al. 2007) and poly-
morphisms of various other genes such as transforming growth factor beta (Dick et al. 2003;
Langdahl et al. 2003), oestrogen receptor (Kung et al. 2006), the vitamin D receptor (Hustmyer
et al. 1994; Morrison et al. 1994) and the low density lipoprotein genes LRP5 and LRP6 (Richards
et al. 2008; van Meurs et al. 2008) contribute to the pathogenesis of osteoporosis and may there-
fore contribute to the development of back pain.
Intervertebral disc disease was also associated with ‘variable number of tandem repeats’ (VNTR)
in the coding region of the human aggrecan gene (AGAN) (Solovieva et al. 2007). Thirteen differ-
ent alleles have been identified, with repeat numbers ranging from 13 to 33. The polymorphism
results in individuals with differing length aggrecan core proteins, bearing different numbers of
potential attachment sites for chondroitin sulfate. Magnetic resonance imaging in healthy volun-
teers revealed the association of ACAN polymorphisms with MRI-signs of intervertebral disc
disease (Solovieva et al. 2007). Particularly, multilevel and severe disc degeneration was present in
the patients with shorter VNTR length of the aggrecan gene whereas a higher number of repeats
(>25) had protective effects (Kawaguchi et al. 1999).
Variants of the gene coding for the cartilage intermediate layer protein (CILP) (Kalichman and
Hunter 2008; Seki et al. 2005) were recently reported to be associated with intervertebral disc
degeneration (Seki et al. 2005). CILP was expressed abundantly in intervertebral discs, and its
expression increased during progression of disc degeneration. CILP colocalized with transform-
ing growth factor TGF-β1 and directly inhibited TGF-1 mediated induction of cartilage matrix
genes and TGF-β 1 mediated inhibition of metalloproteinase transcription. The aberrantly
increased inhibitory effects of CILP attributed to the variant allele probably perturb the balance
of TGF-β control over chondrocyte metabolism and intervertebral disc tissue maintenance.
A single adenine insertion/deletion polymorphism (6A/5A) in the metalloproteinase-3 (MMP3)
promoter region was associated with MRI so-called modic changes in endplates of lumbar verte-
bral bodies in Finnish male workers, particularly if carriers of the MMP3 variant had additional
polymorphisms in interleukin-1 locus genes (Karppinen et al. 2008). In a cohort of Japanese
middle-aged women MMP3 genotypes with the shorter allele (5 adenine nucleotides) were associ-
ated with degenerative changes in lumbar intervertebral discs (Takahashi et al. 2001). The 5A
promoter has higher activity compared to that of the 6A allele. Thus, subjects without the variant
5A allele would be predicted to have higher expression and activity of MMP3. As MMP3 plays a
major role in the neuroimmune responses following nerve injury and contributes to the develop-
ment of neuropathic pain in rodents (Kawasaki et al. 2008) this variant may additionally increase
the susceptibility to radicular sciatic pain in degenerative disc disease.
In addition to MMP3, a single missense polymorphism in metallaoproteinase-9 (MMP9) was
strongly associated with lumbar disc herniation in two independent Japanese cohorts (Hirose
et al. 2008). This polymorphism showed a combinatorial effect with a non-coding variant of the
thrombospondin-2 gene. The THBS2-SNP caused differences on exon 11 skipping rates, with
decreased thrombospondin-2 interaction with MMP2 and MMP9 (Hirose et al. 2008). MMP9
also plays a major role in the manifestation of neuropathic pain following nerve injury (Kawasaki
et al. 2008) suggesting that variants in thrombospondin and MMP9 may also increase the risk of
pain hypersensitivity in carriers of these variants.
Carriers of some variants of the vitamin D receptor gene may also be at a higher risk for interver-
tebral disc disease and osteoporosis. A common polymorphism in exon 2 of the vitamin D recep-
tor gene (VDR) introduces a new translation start site and produces a protein that differs in
length by three amino acids. Some studies suggest that the longer variant is correlated with lower
bone mineral density in some populations (Morrison et al. 1994) and lumbar disc degeneration
(Videman et al. 1998).
As extracellular matrix protein polymorphisms increase the susceptibility to degeneration of
discs and/or bone of the spine they may increase the risk for chronic back pain. However, these
genetic variants presumably do not directly affect pain sensation or signalling or adaptations of
peripheral and central pain circuits, with exception of the metalloproteinase polymorphisms. The
risk conferred by genetic variants of ECM-genes is further contributed by the risk conferred by
inflammatory genes.
3.4 Polymorphisms in pro-inflammatory genes

Inflammation plays a major role in the development of intervertebral disc disease (Battie et al.
1995). The extent and resolution of inflammation is modified by genetic variants in cytokine
genes. Particularly, polymorphisms in the interleukin-1 gene (IL-1) locus contribute to the devel-
opment of low back pain. Variants in the genes of IL-1α, IL-1β and the IL-1 receptor antagonist
(IL1RN) modify bone mineral density (Kim et al. 2006b) and promote intervertebral disc disease
with low back pain (Solovieva et al. 2004a; Solovieva et al. 2006). Herniated discs produce several
inflammatory mediators such as IL-1, IL-6 and tumour necrosis factor alpha (TNFα) which
maintain the inflammatory process and sensitize nociceptors that innervate the affected discs or
the surrounding tissue. In a Finnish study in middle-aged men carriers of the IL-1 receptor
antagonist G1821>A allele had an increased risk of low back pain (Solovieva et al. 2004b). In
addition, carriers of this allele in combination with variants of IL-1α or IL-1β had a higher risk to
develop low back pain than non-carriers, and reported more days with pain and higher intensities
of low back pain (Solovieva et al. 2004b). The results suggest that IL-1 gene locus polymorphisms
promote or prolong low back pain, supported by a recent study that revealed MRI changes in
endplates of lumbar vertebral bodies in carriers of combined polymorphisms in IL1a and the
5-adenine promoter polymorphism of metalloproteinase-3 (Karppinen et al. 2008). In addition
to IL-1, polymorphisms in the IL-6 gene were associated with intervertebral disc disease in
patients with discogenic lumbar radicular pain (Noponen-Hietala et al. 2005). Several other pro-
inflammatory gene polymorphisms including variants of IL-2, TNF-α, IL-4 and INFγ were
analysed but had almost identical allele frequencies in low back pain patients and controls.
3.5 Polymorphisms contributing to chronic widespread

musculoskeletal pain
Chronic back pain is often associated with widespread pain in complex musculoskeletal pain
syndromes such as fibromyalgia (Arnold et al. 2008 ), temporomandibular joint disorder
(Wiesinger et al. 2007), and chronic fatigue syndrome (Meeus et al. 2007). Fibromyalgia is a gen-
eralized widespread chronic pain disorder characterized by diffuse muscle pain throughout the
body, most often including chronic back and neck pain, muscle weakness, fatigue, increased
negative mood, sleep disturbance (Arnold et al. 2008) and comorbidity with anxiety and depres-
sion (Thieme et al. 2004). The precise role of genetic factors in the aetiopathology is still unclear
but polymorphisms in genes of the serotoninergic and catecholaminergic systems have been asso-
ciated with an increased risk of fibromyalgia and chronic fatigue syndrome. COMT polymor-
phisms may contribute to the pathogenesis because approximately 74% of fibromyalgia patients
had low or intermediate COMT activity resulting in low catecholamine degradation (Gursoy et al.
2003; Tander et al. 2008).
Polyarthropathic diseases such as ankylosing spondylitis, rheumatoid arthritis, psoriatic arthri-
tis, or systemic lupus erythematosus cause inflammatory back pain due to arthritis of the small
intervertebral or sacroiliacal joints and myositis. Polymorphisms in the major histocompatibility
complex (MHC) molecules contribute to the development of these diseases (Brown et al. 1998a;
Brown et al. 1998b). However, several other immune genes such as immunoglobulin receptors,
TNF receptor, transcription factors such as Stat4, various cytokines and chemokines were shown
to be associated with the risk to develop these chronic inflammatory autoimmune diseases and to
modify the response to pharmacologic treatments with e.g. anti-TNF monoclonal antibodies or
immunosuppressive agents.
3.6 Polymorphisms causing complex syndromes with a loss of

pain perception
Several hereditary maladies with complete loss of pain sensitivity have been genetically defined.
Although the molecular mechanisms differ among them, all syndromes are characterized by an
interruption of transmission or processing at key points of the nociceptive system (for full details,
see the ‘Online Mendelian Inheritance in Man’ database (OMIM): http://www.ncbi.nlm.nih.gov/
sites/entrez?db=omim) (Oertel and Lötsch 2008). This includes (1) the channelopathy associated
insensitivity to pain, which is based on loss-of-function mutations of the alpha-subunit of the

voltage-gated sodium channel, Na(v)1.7 (Cox et al. 2006; Goldberg et al. 2007), (2) the Hereditary
sensory and autonomic neuropathy type I (HSAN-I), caused by mutations in the serine palmi-
toyltransferase, long chain base subunit 1 gene (Bejaoui et al. 2001; Dawkins et al. 2001; Verhoeven
et al. 2006), (3) the HSAN-II, based on mutations in the hereditary sensory neuropathy, type II
(Lafreniere et al. 2004; Riviere et al. 2004; Cho et al. 2006; Roddier et al. 2005), (4) the HSAN-III
due to mutations in the inhibitor of kappa light polypeptide gene enhancer in B-cells, kinase
complex-associated protein gene (Slaugenhaupt and Gusella 2002; Anderson et al. 2001; Leyne
et al. 2003), (5) the HSAN-IV, also called congenital insensitivity to pain with anhidrosis (CIPA)
and based on mutations in the neurotrophic tyrosine kinase, receptor, type 1 gene (Indo et al.
1996; Miura et al. 2000), and (6) the HSAN-V caused by mutations in the nerve growth factor,
beta polypeptide gene (Einarsdottir et al. 2004). All these syndromes are very rare affecting a few
families. Therefore, their specific association with back pain has not been shown but it is reason-
able to assume that patients with these syndromes will not develop back pain.
3.7 Summary
Several genetic factors contribute to the risk for chronic back pain and widespread pain syndromes.
The experience of pain results from a complex interaction between several genetic variants
involved in different steps of neuronal processing of nociceptive information with additional
contribution of other genetic, structural, environmental and psychosocial factors. The investiga-
tion of interactions between genetic variants that modify pain signalling and the variants affecting
the bone and intervertebral cartilage and the identification of the molecular consequences
of functional variants are further required to understand the genetics of pain and specially to
identify genetic predictors of chronic back pain.
References
Aberle, J., Flitsch, J., Beck, N.A., et al. (2008). Genetic variation may influence obesity only under
conditions of diet: analysis of three candidate genes. Mol Genet Metab 95, 188–91.
Agarwal, N., Pacher, P., Tegeder, I., et al. (2007). Cannabinoids mediate analgesia largely via peripheral
type 1 cannabinoid receptors in nociceptors. Nat Neurosci 10, 870–9.
Agrawal, A., Wetherill, L., Dick, D.M., et al. (2008). Evidence for association between polymorphisms in
the cannabinoid receptor 1 (CNR1) gene and cannabis dependence. Am J Med Genet B Neuropsychiatr
Genet 150B(5), 736–40.
Anderson, S.L., Coli, R., Daly, I.W., et al. (2001). Familial dysautonomia is caused by mutations of the
IKAP gene. Am J Hum Genet 68, 753–8.
Annunen, S., Paassilta, P., Lohiniva, J., et al. (1999). An allele of COL9A2 associated with intervertebral
disc disease. Science 285, 409–12.
Arnold, B.S., Alpers, G.W., Suss, H., et al. (2008). Affective pain modulation in fibromyalgia, somatoform
pain disorder, back pain, and healthy controls. Eur J Pain 12, 329–38.
Ashford, R.U., Luchetti, M., McCloskey, E.V., et al. (2001). Studies of bone density, quantitative
ultrasound, and vertebral fractures in relation to collagen type I alpha 1 alleles in elderly women. Calcif
Tissue Int 68, 348–51.
Battie, M.C., Videman, T., Gibbons, L.E., Fisher, L.D., Manninen, H., and Gill, K. (1995). 1995 Volvo
Award in clinical sciences. Determinants of lumbar disc degeneration. A study relating lifetime
exposures and magnetic resonance imaging findings in identical twins. Spine 20, 2601–12.
Battie, M.C., Videman, T., Levalahti, E., Gill, K., and Kaprio, J. (2007). Heritability of low back pain and
the role of disc degeneration. Pain 131, 272–80.
Bejaoui, K., Wu, C., Scheffler, M.D., et al. (2001). SPTLC1 is mutated in hereditary sensory neuropathy,
type 1. Nat Genet 27, 261–2.
Blau, N., and Niederwieser, A. (1985). GTP-cyclohydrolases: a review. J Clin Chem Clin Biochem 23,
169–76.
Brown, M.A., Kennedy, L.G., Darke, C., et al. (1998a). The effect of HLA-DR genes on susceptibility to and
severity of ankylosing spondylitis. Arthritis Rheum 41, 460–5.
Brown, M.A., Pile, K.D., Kennedy, L.G., et al. (1998b). A genome-wide screen for susceptibility loci in
ankylosing spondylitis. Arthritis Rheum 41, 588–95.
Cho, H.J., Kim, B.J., Suh, Y.L., An, J.Y., and Ki, C.S. (2006). Novel mutation in the HSN2 gene in a Korean
patient with hereditary sensory and autonomic neuropathy type 2. J Hum Genet 51, 905–8.
Chou, W.Y., Wang, C.H., Liu, P.H., Liu, C.C., Tseng, C.C., and Jawan, B. (2006a). Human opioid receptor
A118G polymorphism affects intravenous patient-controlled analgesia morphine consumption after
total abdominal hysterectomy. Anesthesiology 105, 334–7.
Chou, W.Y., Yang, L.C., Lu, H.F., et al. (2006b). Association of mu-opioid receptor gene polymorphism
(A118G) with variations in morphine consumption for analgesia after total knee arthroplasty. Acta
Anaesthesiol Scand 50, 787–92.
Cox, J.J., Reimann, F., Nicholas, A.K., et al. (2006). An SCN9A channelopathy causes congenital inability
to experience pain. Nature 444, 894–8.
Dawkins, J.L., Hulme, D.J., Brahmbhatt, S.B., Auer-Grumbach, M., and Nicholson, G.A. (2001).
Mutations in SPTLC1, encoding serine palmitoyltransferase, long chain base subunit-1, cause
hereditary sensory neuropathy type I. Nat Genet 27, 309–12.
Diatchenko, L., Nackley, A.G., Slade, G.D., et al. (2006). Catechol-O-methyltransferase gene
polymorphisms are associated with multiple pain-evoking stimuli. Pain 125, 216–24.
Dick, I.M., Devine, A., Li, S., Dhaliwal, S.S., and Prince, R.L. (2003). The T869C TGF beta polymorphism
is associated with fracture, bone mineral density, and calcaneal quantitative ultrasound in elderly
women. Bone 33, 335–41.
Doehring, A., Geisslinger, G., and Lötsch, J. (2007). Rapid screening for potentially relevant
polymorphisms in the human fatty acid amide hydrolase gene using Pyrosequencing. Prostaglandins
Other Lipid Mediat 84, 128–37.
Einarsdottir, E., Carlsson, A., Minde, J., et al. (2004). A mutation in the nerve growth factor beta gene
(NGFB) causes loss of pain perception. Hum Mol Genet 13, 799–805.
Fillingim, R.B., Kaplan, L., Staud, R., et al. (2005). The A118G single nucleotide polymorphism of the
mu-opioid receptor gene (OPRM1) is associated with pressure pain sensitivity in humans. J Pain
6, 159–67.
Foulkes, T., and Wood, J.N. (2008). Pain genes. PLoS Genet 4, e1000086.
Goldberg, Y.P., MacFarlane, J., MacDonald, M.L., et al. (2007). Loss-of-function mutations in the Nav1.7
gene underlie congenital indifference to pain in multiple human populations. Clin Genet 71, 311–19.
Grant, S.F., Reid, D.M., Blake, G., Herd, R., Fogelman, I., and Ralston, S.H. (1996). Reduced bone density
and osteoporosis associated with a polymorphic Sp1 binding site in the collagen type I alpha 1 gene.
Nat Genet 14, 203–5.
Gross, S.S., and Levi, R. (1992). Tetrahydrobiopterin synthesis. An absolute requirement for
cytokine-induced nitric oxide generation by vascular smooth muscle. J Biol Chem 267, 25722–9.
Gursoy, S., Erdal, E., Herken, H., Madenci, E., Alasehirli, B., and Erdal, N. (2003). Significance of
catechol-O-methyltransferase gene polymorphism in fibromyalgia syndrome. Rheumatol Int 23, 104–7.
Hamdani, N., Tabeze, J.P., Ramoz, N., et al. (2008). The CNR1 gene as a pharmacogenetic factor for
antipsychotics rather than a susceptibility gene for schizophrenia. Eur Neuropsychopharmacol 18, 34–40.
Hartvigsen, J., Christensen, K., Frederiksen, H., and Pedersen, H.C. (2004). Genetic and environmental
contributions to back pain in old age: a study of 2,108 danish twins aged 70 and older. Spine 29,
897–901; discussion 902.
Hayashida, M., Nagashima, M., Satoh, Y., et al. (2008). Analgesic requirements after major abdominal
surgery are associated with OPRM1 gene polymorphism genotype and haplotype. Pharmacogenomics
9, 1605–16.
Higashino, K., Matsui, Y., Yagi, S., et al. (2007). The alpha2 type IX collagen tryptophan polymorphism is
associated with the severity of disc degeneration in younger patients with herniated nucleus pulposus
of the lumbar spine. Int Orthop 31, 107–11.
Hirose, Y., Chiba, K., Karasugi, T., et al. (2008). A functional polymorphism in THBS2 that affects
alternative splicing and MMP binding is associated with lumbar-disc herniation. Am J Hum Genet
82, 1122–9.
Hustmyer, F.G., Peacock, M., Hui, S., Johnston, C.C., and Christian, J. (1994). Bone mineral density in
relation to polymorphism at the vitamin D receptor gene locus. J Clin Invest 94, 2130–4.
Ikeda, K., Ide, S., Han, W., Hayashida, M., Uhl, G.R., and Sora, I. (2005). How individual sensitivity to
opiates can be predicted by gene analyses. Trends Pharmacol Sci 26, 311–17.
Indo, Y., Tsuruta, M., Hayashida, Y., et al. (1996). Mutations in the TRKA/NGF receptor gene in patients
with congenital insensitivity to pain with anhidrosis. Nat Genet 13, 485–88.
Jim, J.J., Noponen-Hietala, N., Cheung, K.M., et al. (2005). The TRP2 allele of COL9A2 is an
age-dependent risk factor for the development and severity of intervertebral disc degeneration. Spine
30, 2735–42.
Kalichman, L., and Hunter, D.J. (2008). The genetics of intervertebral disc degeneration. Associated genes.
Joint Bone Spine 75, 388–96.
Karppinen, J., Daavittila, I., Solovieva, S., et al. (2008). Genetic factors are associated with modic changes
in endplates of lumbar vertebral bodies. Spine 33, 1236–41.
Karsak, M., Cohen-Solal, M., Freudenberg, J., et al. (2005). Cannabinoid receptor type 2 gene is associated
with human osteoporosis. Hum Mol Genet 14, 3389–96.
Kawaguchi, Y., Osada, R., Kanamori, M., et al. (1999). Association between an aggrecan gene
polymorphism and lumbar disc degeneration. Spine 24, 2456–60.
Kawasaki, Y., Xu, Z.Z., Wang, X., et al. (2008). Distinct roles of matrix metalloproteases in the early- and
late-phase development of neuropathic pain. Nat Med 14, 331–6.
Kim, H., Mittal, D.P., Iadarola, M.J., and Dionne, R.A. (2006a). Genetic predictors for acute experimental
cold and heat pain sensitivity in humans. J Med Genet 43, e40.
Kim, J.G., Lim, K.S., Ku, S.Y., Kim, S.H., Choi, Y.M., and Moon, S.Y. (2006b). Relations between
interleukin-1, its receptor antagonist gene polymorphism, and bone mineral density in
postmenopausal Korean women. J Bone Miner Metab 24, 53–7.
Klepstad, P., Rakvag, T.T., Kaasa, S., et al. (2004). The 118 A > G polymorphism in the human mu-opioid
receptor gene may increase morphine requirements in patients with pain caused by malignant disease.
Acta Anaesthesiol Scand 48, 1232–9.
Knoeringer, M., Reinke, A., Trappe, A.E., and Schlegel, J. (2008). Absence of the mutated Trp2 allele but a
common polymorphism of the COL9A2 collagen gene is associated with early recurrence after lumbar
discectomy in a German population. Eur Spine J 17, 463–67.
Kung, A.W., Lai, B.M., Ng, M.Y., Chan, V., and Sham, P.C. (2006). T-1213C polymorphism of estrogen
receptor beta is associated with low bone mineral density and osteoporotic fractures. Bone 39,
1097–106.
Lafreniere, R.G., MacDonald, M.L., Dube, M.P., et al. (2004). Identification of a novel gene (HSN2)
causing hereditary sensory and autonomic neuropathy type II through the Study of Canadian Genetic
Isolates. Am J Hum Genet 74, 1064–73.
Langdahl, B.L., Carstens, M., Stenkjaer, L., and Eriksen, E.F. (2003). Polymorphisms in the transforming
growth factor beta 1 gene and osteoporosis. Bone 32, 297–310.
Leyne, M., Mull, J., Gill, S.P., et al. (2003). Identification of the first non-Jewish mutation in familial
Dysautonomia. Am J Med Genet A 118A, 305–8.
Lichtman, A.H., Leung, D., Shelton, C.C., et al. (2004). Reversible inhibitors of fatty acid amide hydrolase
that promote analgesia: evidence for an unprecedented combination of potency and selectivity.
J Pharmacol Exp Ther 311, 441–8.
Liem, E.B., Joiner, T.V., Tsueda, K., and Sessler, D.I. (2005). Increased sensitivity to thermal pain and
reduced subcutaneous lidocaine efficacy in redheads. Anesthesiology 102, 509–14.
Lötsch, J., and Geisslinger, G. (2005). Are mu-opioid receptor polymorphisms important for clinical opioid
therapy? Trends Mol Med 11, 82–89.
Lötsch, J., Skarke, C., Grosch, S., Darimont, J., Schmidt, H., and Geisslinger, G. (2002). The
polymorphism A118G of the human mu-opioid receptor gene decreases the pupil constrictory
effect of morphine-6-glucuronide but not that of morphine. Pharmacogenetics 12, 3–9.
Lötsch, J., Skarke, C., Wieting, J., et al. (2006a). Modulation of the central nervous effects of
levomethadone by genetic polymorphisms potentially affecting its metabolism, distribution, and drug
action. Clin Pharmacol Ther 79, 72–89.
Lötsch, J., Stuck, B., and Hummel, T. (2006b). The human mu-opioid receptor gene polymorphism
118A > G decreases cortical activation in response to specific nociceptive stimulation. Behav Neurosci
120, 1218–24.
Luoma, K., Riihimaki, H., Luukkonen, R., Raininko, R., Viikari-Juntura, E., and Lamminen, A. (2000).
Low back pain in relation to lumbar disc degeneration. Spine 25, 487–92.
MacDonald, H.M., McGuigan, F.A., New, S.A., et al. (2001). COL1A1 Sp1 polymorphism predicts
perimenopausal and early postmenopausal spinal bone loss. J Bone Miner Res 16, 1634–41.
Manning, B.H., Merin, N.M., Meng, I.D., and Amaral, D.G. (2001). Reduction in opioid- and
cannabinoid-induced antinociception in rhesus monkeys after bilateral lesions of the amygdaloid
complex. J Neurosci 21, 8238–46.
Meeus, M., Nijs, J., and Meirleir, K.D. (2007). Chronic musculoskeletal pain in patients with the chronic
fatigue syndrome: a systematic review. Eur J Pain 11, 377–86.
Mio, F., Chiba, K., Hirose, Y., et al. (2007). A functional polymorphism in COL11A1, which encodes the
alpha 1 chain of type XI collagen, is associated with susceptibility to lumbar disc herniation. Am J Hum
Genet 81, 1271–7.
Miura, Y., Mardy, S., Awaya, Y., et al. (2000). Mutation and polymorphism analysis of the TRKA (NTRK1)
gene encoding a high-affinity receptor for nerve growth factor in congenital insensitivity to pain with
anhidrosis (CIPA) families. Hum Genet 106, 116–24.
Mogil, J.S., Ritchie, J., Smith, S.B., et al. (2005). Melanocortin-1 receptor gene variants affect pain and
mu-opioid analgesia in mice and humans. J Med Genet 42, 583–7.
Morrison, N.A., Qi, J.C., Tokita, A., et al. (1994). Prediction of bone density from vitamin D receptor
alleles. Nature 367, 284–7.
Noponen-Hietala, N., Kyllonen, E., Mannikko, M., et al. (2003). Sequence variations in the collagen IX and
XI genes are associated with degenerative lumbar spinal stenosis. Ann Rheum Dis 62, 1208–14.
Noponen-Hietala, N., Virtanen, I., Karttunen, R., et al. (2005). Genetic variations in IL6 associate with
intervertebral disc disease characterized by sciatica. Pain 114, 186–94.
Oakley, M., and Vieira, A.R. (2008). The many faces of the genetics contribution to temporomandibular
joint disorder. Orthod Craniofac Res 11, 125–35.
Oertel, B., and Lötsch, J. (2008). Genetic mutations that prevent pain: implications for future pain
medication. Pharmacogenomics 9, 179–94.
Oertel, B.G., Preibisch, C., Wallenhorst, T., et al. (2008). Differential opioid action on sensory and affective
cerebral pain processing. Clin Pharmacol Ther 83, 577–88.
Oertel, B.G., Schmidt, R., Schneider, A., Geisslinger, G., and Lötsch, J. (2006). The mu-opioid receptor
gene polymorphism 118A>G depletes alfentanil-induced analgesia and protects against respiratory
depression in homozygous carriers. Pharmacogenet Genomics 16, 625–36.
Ofek, O., Karsak, M., Leclerc, N., et al. (2006). Peripheral cannabinoid receptor, CB2, regulates bone mass.
Proc Natl Acad Sci U S A 103, 696–701.
Paassilta, P., Lohiniva, J., Goring, H.H., et al. (2001). Identification of a novel common genetic risk factor
for lumbar disk disease. Jama 285, 1843–9.
Pluijm, S.M., van Essen, H.W., Bravenboer, N., et al. (2004). Collagen type I alpha1 Sp1 polymorphism,
osteoporosis, and intervertebral disc degeneration in older men and women. Ann Rheum Dis 63, 71–7.
Reyes-Gibby, C.C., Shete, S., Rakvag, T., et al. (2007). Exploring joint effects of genes and the clinical
efficacy of morphine for cancer pain: OPRM1 and COMT gene. Pain 130, 25–30.
Richards, J.B., Rivadeneira, F., Inouye, M., et al. (2008). Bone mineral density, osteoporosis, and
osteoporotic fractures: a genome-wide association study. Lancet 371, 1505–12.
Riviere, J.B., Verlaan, D.J., Shekarabi, M., et al. (2004). A mutation in the HSN2 gene causes sensory
neuropathy type II in a Lebanese family. Ann Neurol 56, 572–5.
Roddier, K., Thomas, T., Marleau, G., et al. (2005). Two mutations in the HSN2 gene explain the high
prevalence of HSAN2 in French Canadians. Neurology 64, 1762–7.
Russo, P., Strazzullo, P., Cappuccio, F.P., et al. (2007). Genetic variations at the endocannabinoid type 1
receptor gene (CNR1) are associated with obesity phenotypes in men. J Clin Endocrinol Metab 92,
2382–6.
Sakai, Y., Matsuyama, Y., Hasegawa, Y., et al. (2007). Association of gene polymorphisms with
intervertebral disc degeneration and vertebral osteophyte formation. Spine 32, 1279–86.
Schneider, S., Mohnen, S.M., Schiltenwolf, M., and Rau, C. (2007). Comorbidity of low back pain:
representative outcomes of a national health study in the Federal Republic of Germany. Eur J Pain
11, 387–97.
Seki, S., Kawaguchi, Y., Chiba, K., et al. (2005). A functional SNP in CILP, encoding cartilage intermediate
layer protein, is associated with susceptibility to lumbar disc disease. Nat Genet 37, 607–12.
Seki, S., Kawaguchi, Y., Mori, M., et al. (2006). Association study of COL9A2 with lumbar disc disease in
the Japanese population. J Hum Genet 51, 1063–7.
Sipe, J.C., Arbour, N., Gerber, A., and Beutler, E. (2005). Reduced endocannabinoid immune modulation
by a common cannabinoid 2 (CB2) receptor gene polymorphism: possible risk for autoimmune
disorders. J Leukoc Biol 78, 231–38.
Skarke, C., Darimont, J., Schmidt, H., Geisslinger, G., and Lötsch, J. (2003). Analgesic effects of morphine
and morphine-6-glucuronide in a transcutaneous electrical pain model in healthy volunteers.
Clin Pharmacol Ther 73, 107–21.
Slade, G.D., Diatchenko, L., Bhalang, K., et al. (2007). Influence of psychological factors on risk of
temporomandibular disorders. J Dent Res 86, 1120–25.
Slaugenhaupt, S.A., and Gusella, J.F. (2002). Familial dysautonomia. Curr Opin Genet Dev 12, 307–11.
Solovieva, S., Kouhia, S., Leino-Arjas, P., et al. (2004a). Interleukin 1 polymorphisms and intervertebral
disc degeneration. Epidemiology 15, 626–33.
Solovieva, S., Leino-Arjas, P., Saarela, J., Luoma, K., Raininko, R., and Riihimaki, H. (2004b). Possible
association of interleukin 1 gene locus polymorphisms with low back pain. Pain 109, 8–19.
Solovieva, S., Lohiniva, J., Leino-Arjas, P., et al. (2006). Intervertebral disc degeneration in relation to the
COL9A3 and the IL-1ss gene polymorphisms. Eur Spine J 15, 613–19.
Solovieva, S., Noponen, N., Mannikko, M., et al. (2007). Association between the aggrecan gene variable
number of tandem repeats polymorphism and intervertebral disc degeneration. Spine 32, 1700–5.
Stohler, C.S. (2006). TMJD 3: a genetic vulnerability disorder with strong CNS involvement. J Evid Based
Dent Pract 6, 53–57.
Svendsen, K.B., Jensen, T.S., and Bach, F.W. (2004). Does the cannabinoid dronabinol reduce central pain
in multiple sclerosis? Randomised double blind placebo controlled crossover trial. Bmj 329, 253.
Takahashi, M., Haro, H., Wakabayashi, Y., Kawa-uchi, T., Komori, H., and Shinomiya, K. (2001). The
association of degeneration of the intervertebral disc with 5a/6a polymorphism in the promoter of the
human matrix metalloproteinase-3 gene. J Bone Joint Surg Br 83, 491–95.
Tam, J., Trembovler, V., Di Marzo, V., et al. (2008). The cannabinoid CB1 receptor regulates bone
formation by modulating adrenergic signaling. Faseb J 22, 285–94.
Tander, B., Gunes, S., Boke, O., et al. (2008). Polymorphisms of the serotonin-2A receptor and catechol-
O-methyltransferase genes: a study on fibromyalgia susceptibility. Rheumatol Int 28, 685–91.
Tegeder, I., Adolph, J., Schmidt, H., Woolf, C.J., Geisslinger, G., and Lötsch, J. (2008). Reduced
hyperalgesia in homozygous carriers of a GTP cyclohydrolase 1 haplotype. Eur J Pain 12, 1069–77.
Tegeder, I., Costigan, M., Griffin, R.S., et al. (2006). GTP cyclohydrolase and tetrahydrobiopterin regulate
pain sensitivity and persistence. Nat Med 12, 1269–77.
Thieme, K., Turk, D.C., and Flor, H. (2004). Comorbid depression and anxiety in fibromyalgia syndrome:
relationship to somatic and psychosocial variables. Psychosom Med 66, 837–44.
Thomas, E.A., Cravatt, B.F., Danielson, P.E., Gilula, N.B., and Sutcliffe, J.G. (1997). Fatty acid amide
hydrolase, the degradative enzyme for anandamide and oleamide, has selective distribution in neurons
within the rat central nervous system. J Neurosci Res 50, 1047–52.
Ujike, H., Takaki, M., Nakata, K., et al. (2002). CNR1, central cannabinoid receptor gene, associated with
susceptibility to hebephrenic schizophrenia. Mol Psychiatry 7, 515–18.
van Meurs, J.B., Trikalinos, T.A., Ralston, S.H., et al. (2008). Large-scale analysis of association between
LRP5 and LRP6 variants and osteoporosis. Jama 299, 1277–90.
Verhoeven, K., Timmerman, V., Mauko, B., Pieber, T.R., De Jonghe, P., and Auer-Grumbach, M. (2006).
Recent advances in hereditary sensory and autonomic neuropathies. Curr Opin Neurol 19, 474–80.
Videman, T., Leppavuori, J., Kaprio, J., et al. (1998). Intragenic polymorphisms of the vitamin D receptor
gene associated with intervertebral disc degeneration. Spine 23, 2477–85.
Wiesinger, B., Malker, H., Englund, E., and Wanman, A. (2007). Back pain in relation to musculoskeletal
disorders in the jaw-face: a matched case-control study. Pain 131, 311–19.
Zhang, P.W., Ishiguro, H., Ohtsuki, T., et al. (2004). Human cannabinoid receptor 1: 5’ exons,
candidate regulatory regions, polymorphisms, haplotypes and association with polysubstance abuse.
Mol Psychiatry 9, 916–31.
Zhang, Y., Kerns, J.M., Anderson, D.G., et al. (2006). Sensory neurons and fibers from multiple spinal cord
levels innervate the rabbit lumbar disc. Am J Phys Med Rehabil 85, 865–71.
Zhang, Y., Wang, D., Johnson, A.D., Papp, A.C., and Sadee, W. (2005). Allelic expression imbalance of
human mu opioid receptor (OPRM1) caused by variant A118G. J Biol Chem 280, 32618–24.
Zuo, L., Kranzler, H.R., Luo, X., Covault, J., and Gelernter, J. (2007). CNR1 variation modulates risk for
drug and alcohol dependence. Biol Psychiatry 62, 616–26.
Chapter 4
Peripheral and Central Sensitization

as Risk Factors of Low Back Pain
Hermann O. Handwerker
4.1 Introduction
Acute low back pain is one of the most common complaints of adult humans, probably due to
degenerative and/or use-dependent damages of the vertebral column and surrounding structures.
When low back pain becomes chronic, it is a disabling disease. Chronic low back pain (CLBP),
probably the most common chronic pain disease, is very difficult to treat (see Chapter 25).
Though pathological processes at and around the vertebral column are the source of the pain,
there is surprisingly little correlation between the magnitude e.g. of degenerative alterations of the
spine and the pain intensity.
Chronic pain states are generally due to a complex interplay between peripheral and central
nociceptive mechanisms. In CLBP this interplay is particularly complex, since it includes not only
sensitized nociceptors from vertebrae, muscles, tendons, and small joints, but very often also
has neuropathic components from damaged vertebral nerve roots. Sensitized nociceptors, or
damaged axons, may bombard the central nervous system and this may lead to a disordered
matrix of central nociceptive neuronal populations. In this respect, marked differences exist
between muscle and skin nociceptors. Muscle pain has a much stronger tendency to be referred
and to become chronic and this tendency is particularly strong in low back muscles (Taguchi et al.
1985, 1986). In addition, nocifensive reflexes may induce inadequate contractions of the spinal
musculature which again create a changed micro-environment of the nociceptor terminals and
hence lead to further nociceptor sensitization.
This short review of the pathophysiology of CLBP concentrates on three sources of chronic
pain:
1) Nociceptor sensitization and damage.
2) Altered processing at central synapses.
3) Changes of the reactivity and structure of the brain in chronic pain states.
Indirect influences, e.g. the role of hormonal processes are covered by another chapter of this
book (see Chapter 5) and only marginally implemented in this review. Details about the interac-
tion of muscle activation and CLBP are discussed in two other chapters of this book (see Chapters
8 and 9).
4.2 Nociceptor functions

The term ‘nociceptor’ denominates a large subpopulation of primary afferent neurons. Each
nociceptor consists of a cell body in a dorsal root ganglion (DRG), an axon running in a nerve
which terminally diverts in branches and sprouts into the target tissue. The nociceptor terminals
in the periphery are equipped with a mosaic of membrane proteins functioning as receptors
and membrane channels. The central branches of the nociceptor axons project into the spinal
cord dorsal horn. A review of nociceptor functions has been recently published by Woolf
(2007). The author of this chapter has published an audio-seminar on nociceptor functions
(Handwerker 2009).
It has long been known that nociceptors constitute a heterogeneous population of primary
afferent neurons with respect to functions of their terminals, conduction velocity of axons, and
central projections. According to their neurogenesis two major classes can be distinguished: a
peptidergic population defined by the presence of calcitonin-gene related peptide (CGRP) and
often also substance P. These neurons express the membrane receptor trkA (a receptor tyrosine
kinase of the Trk family) and are dependent on the nerve growth factor (NGF). By contrast,
non-peptidergic nociceptors express Ret, the receptor for the glial cell line-derived neuro-
trophic factor. These neurons are histochemically characterized by binding isolectin B4 (IB4).
However, the two cell populations are not entirely exclusive (Snider and McMahon 1998; Gold
and Gebhart 2010).
Part of the peptidergic primary afferents release CGRP (and in some species also substance P)
from their peripheral terminals and these neuropeptides may have a function in safeguarding the
integrity of the innervated tissues (Kruger 1988). In the course of sensitization this release of
afferent transmitters can be markedly increased leading to neurogenic inflammation (Richardson
and Vasko 2002). Apart of the important distinction in peptidergic and non-peptidergic units
various classes of nociceptors have been distinguished according to the function of their periph-
eral terminals (Handwerker 2010). The most common type has been named ‘polymodal nocicep-
tor’ some decades ago (Bessou and Perl 1969). These units are characterized by unmyelinated
axons and sensitivity to heating and strong mechanical stimulation. Though this characterization
was originally made for cutaneous nociceptors, the term has been transferred to nociceptor
populations in other tissues. Nociceptors with thin myelinated axons may have similar proper-
ties, or serve exclusively as high threshold mechanoreceptors (Adriaensen et al. 1983). ‘Polymodal
nociceptors’ can probably be found in the peptidergic and in the non-peptidergic neuronal popu-
lation. It is not always clear which units belong to this group in different tissues, species and
preparations (Handwerker 2009).
During the last decade another group of nociceptors emerged which are insensitive to mechan-
ical stimulation—as long as the tissue surrounding their terminals is intact. Afferents of this kind
were first described in the knee joint of the cat. They become activated, however, in the time course
of the development of joint inflammation. (Schaible and Schmidt 1983). The authors called these
fibres ‘sleeping nociceptors’. From these early findings it became immediately clear that this group
of nociceptors must play a salient role in the development of inflammatory pain and hyperalgesia.
Later on similar types of nociceptors responding to electrical stimulation of their axons, but not
to mechanical stimulation of their terminals were found in the monkey skin. They were termed
MIAs (mechanically insensitive afferents) (Meyer et al. 1991). Finally, mechanically insensitive
nociceptors were encountered in microneurography experiments in human skin nerves and
named CMi (for mechanically insensitive C-fibres) (Schmidt et al. 1995). This latter group of
units is peptidergic and responsible for the axon reflex flare provoked by the CGRP release from
the terminals (Schmelz et al. 2000). CMi units are characterized by very slow conduction velocities
of their axons and very large receptive fields indicating extensive arborization of their terminals in
the skin. It has been shown that CMi units become mechanically responsive, when their receptive
field is treated with irritants, such as capsaicin (ligand of the TRPV1-receptor, see below), or
mustard oil (ligand of the TRPA1-receptor). CMi units play a decisive role in the development of
mechanical hyperalgesia of inflamed tissues (Torebjörk et al. 1996; Schmidt et al. 2000).
PERIPHERAL AND CENTRAL SENSITIZATION AS RISK FACTORS OF LOW BACK PAIN 57
Unfortunately detailed analyses of the nociceptive afferents from the vertebral column and
surrounding tissues are not available and their properties have to be deduced from comparisons
with those of larger muscles and joints, in some cases even from skin (see below).
4.2.1Transduction and transformation processes in

nociceptor terminals.
It has long been known that plasticity of the nociceptor function and nociceptor sensitization is a
major source of hyperalgesia (which has been termed ‘primary’ hyperalgesia when due to sensi-
tized nociceptor terminals). Two steps are involved in the excitation of nociceptor (and other)
receptor terminals: at first, in a transduction process stimuli generate depolarizations of the
terminals (generator potentials). In a second step these local depolarizations are transformed into
sequences of action potentials. In nociceptors these two processes are not easy to distinguish,
since they occur in parallel at the same sites of the terminal membrane (Carr et al. 2009). Several
types of trans-membrane proteins play a role in the transduction: ion channels, G-protein coupled
receptors (GPCRs), receptors for cytokines and for neurotrophins. The different composition of
these protein populations are a major source of the diversity of nociceptors. Several comprehen-
sive reviews on transduction processes in nociceptors have been published (Woolf and Ma
2007; Gold and Gebhart 2010), and only a non-comprehensive summary of the processes most
important for CLBP can be given in this chapter.
◆ Membrane channels are the prototypic transducing molecules. Transient receptor potential
(TRP) channels seem to play a key role in nociception, most importantly TRPV1 and TRPA1.
The former is gated by heating, low pH and by an exogenous irritant, capsaicin. Endogenous
gating substances have been hypothesized (e.g. anandamide). TRPA1 is gated by various
chemical irritants. Other channels found in nociceptors are gated by ATP, belonging to the
P2X channel family. In addition channels that influence passive membrane properties, the
two-pore K+ channels TREK 1 (Patel and Honore 2001) and TRAAK (Maingret et al. 1999)
contribute to the terminal membrane potential. Though TREK and TRAAK channels are
gated by mechanical forces (Chemin et al. 2007), they probably do not contribute to the
mechanical activation of nociceptors, and neither does one of the above channels. This is an
important gap in our knowledge in particular in the context of low back pain which is in part
initiated by mechanical forces. Important in muscle nociception and the development of
hyperalgesia are channels gated by the low PH of inflamed tissue, the ASIC channels and
TRPV1. Also purinergic membrane receptors (P2X3) gated by ATP seem to play a greater role
than in skin nociceptors (Sluka et al. 2001; Ellrich and Makowska 2007; Heinricher et al. 2009;
Walder et al. 2010).
◆ G protein-coupled receptors do not directly change the membrane potential, but are impor-
tant for the sensitization of nociceptors. Many of them bind to inflammatory substances,
e.g. EP-receptors for prostaglandines, BK-receptors for bradykinin, but also receptors binding
to histamine (H1), serotonin (5-HT2), and others. Upon activation by their ligands these
receptors activate second messenger cascades. Some of them, e.g. the EP-receptors are charac-
terized by Gs proteins coupled to the adenylate cyclase–cAMP–protein kinase A (PKA) path-
way. Others, e.g. bradykinin receptors, are linked to Gq proteins controlling the pathway
phospholipase–IP3–protein kinase C (PKC). In both cases protein kinases, PKA or PKC are at
the end of the cascade and phosphorylate membrane channel proteins, e.g. TRPV1 and Nav
channels. The resulting sensitization processes are discussed below.
◆ Neurotrophin receptors are important for the development and maintenance of nocicep-
tor functions. As mentioned earlier, NGF binding to the TrkA-receptor is mandatory for
peptidergic nociceptors. A genetic defect of TrkA leads to the long known (rare) congenital
insensitivity to pain with anhidrosis (CIPA syndrome) (Indo et al. 1996). However, NGF is also
known to sensitize nociceptors and to produce hypersensitivity (Lewin et al. 1994).
◆ Voltage gated sodium channels are dedicated to the process of transformation, i.e. the
generation of action potentials from the generator potential. It has been mentioned that in
nociceptors the processes of transduction and transformation take place in parallel in the
peripheral axon terminal. Subtypes of voltage dependent Na+ receptors have been described
which are specifically expressed in small DRG-neurons of slowly conducting primary affer-
ents: NaV 1.7, NaV 1.8 and NaV 1.9. NaV 1.9 and 1.8 are resistant to the specific antagonist
tetrodotoxin, NaV 1.7 is sensitive to this toxin. Under normal conditions NaV 1.8 is responsi-
ble for the current generating action potentials. However, NaV 1.7 enhances depolarizing
input leading to action potentials. Mutations preventing the expression of NaV 1.7 in man
lead to congenital insensitivity to pain (Goldberg et al. 2007). People carrying a mutation
leading to a gain of function of NaV 1.7 develop chronic pain diseases, in particular eryth-
romelalgia (Dib-Hajj et al. 2008; Ahn et al. 2010). Polymorphisms of the NaV 1.7 gene
may lead to an overexcitability of DRG neurons and hence to altered pain perceptions prob-
ably also in muscle pain (Estacion et al. 2009; Reimann et al. 2010). Phosphorylation of
NaV 1.8 by protein kinases (PKC or PKA) caused by activation of G-protein coupled receptors
of inflammatory mediators, also results in an increase in nociceptor excitability (Amir
et al. 2006).
Sensitization of nociceptor terminals and the initiation of

4.2.2
pathological pain and hyperalgesia
Inflammatory mediators binding to G-protein coupled receptors and initiating second messenger
cascades with activation of protein kinases are well known processes of nociceptor sensitization.
PKA and PKC phosphorylate transduction channels, e.g. TRPV1, and also spike initiating NaV-
channels. In both cases activation thresholds of nociceptors are lowered and suprathreshold
stimuli lead to more frequent nerve impulses. In case of acute events, e.g. a short-acting trauma,
this process will be reversed after hours or days. Additional mechanisms are required for nocicep-
tor sensitization to become chronic. One such mechanism seems to be the activation of the epsi-
lon isoform of protein kinase C (PKCepsilon) by a recently discovered second messenger pathway
leading from cAMP through the cAMP-activated guanine exchange factor EPAC to PKCepsilon.
This pathway is restricted to non-peptidergic (IB4 postitive) nociceptors. In animal experiments
sustained mechanical hyperalgesia was induced through this pathway (Hucho et al. 2000).
Agents leading to more chronic changes of nociceptor sensitivity are neurotrophins, mentioned
above; e.g. NGF is increased in inflamed tissue and has been shown to induce long lasting sensiti-
zation of peptidergic nociceptors (Hefti et al. 2006; Rukwied et al. 2010). After binding to the trkA
receptor NGF is incorporated by nociceptive nerve endings and transported to the cell bodies
where it alters functions of the genome which contribute to long lasting hyperexcitability
(Hoheisel et al. 2005; Amir et al. 2006; Murase et al. 2010).
Cytokines, which are known to play a major role in the development of arthritis, are also impor-
tant for chronic inflammatory pain states of ligaments, small joints, muscle, and tendons of the
vertebral column. Their role in nociceptive processing has been addressed in a recent review
(Schaible et al. 2010). Tumour necrosis factor alpha (TNF-α), interleukin 1 and 6 bind to
membrane receptors of nociceptors (TNF-R1,2; IL1-R, IL6-R) and mediate long lasting sensitiza-
tion in the course of inflammation. Their effects include enhancement of prostaglandine effects
and upregulation of TRP- and NaV-channels. Increased levels of TNF-α and other cytokines have
been detected in herniated disc tissue compared with normal non-degenerated disc tissue
(Yamashita et al. 2008; Lee et al. 2009). TNF-α applied to nerve roots produces the pathologic
changes observed with nerve root exposure to nucleus pulposus extracts (Igarashi et al. 2000).
Furthermore, in an animal model of radiculopathy induced by intervertebral disc material,
inhibitors of TNF-α had a moderate effect on the radicular pathology (Norimoto et al. 2008).
There is ample evidence that neurotrophins and cytokines induce not only acute effects, like
gating of membrane receptors, but also profound changes in the molecular functions of the noci-
ceptor neurons, such as the expression of membrane receptors. In animal models these effects
were particularly prominent in neuropathy models.
4.2.3 Neuropathic components of CLBP

CLBP is characterized in many cases by a combination of inflammatory and neuropathic mecha-
nisms (Baron 2009) (see also Chapter 25). This may be a salient reason for the difficulties of find-
ing effective pharmaco-therapies for this kind of chronic pain (see Chapter 25). Neuropathic
alterations can be due to mechanical forces, i.e. chronic pressure on nerve roots, but chemical
mediators are probably more important, e.g. by the action of the material from disc hernia on the
conductile axons. However, also in the absence of a ruptured nucleus pulposus inflammatory
cytokines, in particular TNFα released from macrophages, can induce ectopic spike activity from
injured and also from uninjured nerve fibres (Sommer 2003; Sorkin et al. 1998). Interestingly,
macrophages may accumulate in DRG even when the nerve lesion is located at a distance
(Lu et al. 1993).
While chronic inflammatory pain is due to alterations of the transduction and transformation
process at the nociceptor terminals, neuropathic changes occur along the conductile membrane
of the axons. Several animal models of neuropathic pain have been developed and the results have
been described in recent comprehensive reviews (Decosterd and Woolf 2000; Woolf and Salter
2000; Decosterd and Berta 2009; Ossipov and Porreca 2009). In the context of this chapter models
of mononeuropathies are most important including loose ligature of a nerve (Bennett and Xie
1988) and partial nerve transection (Decosterd and Woolf 2000).
A large number of cellular changes are the consequence of sustained alterations of the conduc-
tile axonal membrane. Partial nerve injury leads to increased levels of messenger RNA for mem-
brane receptors and channels. In particular, sodium channels NaV 1.8 and 1.9 are accumulated at
the site of a nerve lesion, and in addition an embryonic channel NaV 1.3 can be expressed (Wood
et al. 2004; Omana-Zapata et al. 2005. This can lead to ectopic action potentials. Not only at the
lesion site, but also in the cell bodies in the DRG channel expressions are changed. Even more
important, altered channel expressions are also found in unaffected neurons. After traumatic
nerve injury, redistribution of NaV 1.8 to the axons of uninjured neighbouring nociceptors plays
a major role in mechanical hypersensitivity (Amir et al. 2006). The expression of TRPV1 channels
at the nerve terminals distal to the lesion site is downregulated after partial nerve lesion but novel
expression of TRPV1 in uninjured neighbouring axons has been observed even in myelinated
afferents and their ganglion cells (Hudson et al. 2001; Ma et al. 2005). Also receptors which are
normally not expressed in primary afferent nociceptive neurons may be produced in spared
axons and cell bodies. After a traumatic lesion of a nerve or root, for example, cold sensing
TRPM8 (transient receptor potential cation channel, subfamily M, member 8) channels (Obata
et al 2005) and alpha adrenoreceptors (Sato and Perl 1991) are expressed in the terminal mem-
brane. There is evidence that the impaired capacity of lesioned axons to take neurotrophins up
(e.g. NGF) which are produced in the tissue surrounding the nociceptor terminals leads to ‘over-
feeding’ of spared axons. Since these neurotrophins are taken up by their endings and transported
to the cell bodies in the DRG, they can alter the genetic programming in the nucleus leading to
the production of increased amounts of mRNA and even altered mRNA fostering the production
of signal proteins and their transport to the nerve terminals (Wu et al. 2001).
4.3 Altered processing in the central nervous system

4.3.1 Synaptic plasticity in the spinal cord dorsal horn
The changes in nociceptor functions following inflammation and axonal damage have profound
consequences for pain processing in the central nervous system. Most information is available
from animal studies on the changes in the spinal dorsal horn, the first station of central pain
processing.
The secondary neurons of the nociceptive pathway are located in the superficial laminae of the
dorsal horn. The transmission neurons in lamina I receive more or less specific input from
nociceptors (nociception specific neurons), whereas larger multimodal projection neurons are
found in lamina V. Both neuronal populations project to higher centres through the classical
anterolateral spinothalamic tract and also through multisynaptic projections, relayed through
brainstem nuclei.
It has been shown that low-frequency input from muscle nociceptors or even subthreshold
synaptic potentials evoked by muscle nociceptors are sufficient to induce a long-lasting hyperex-
citability in these neurons (Hoheisel et al. 2007). Important signs of central sensitization are
increased neuronal activity to noxious stimuli, expansion of the receptive fields, and spread of
excitation to other spinal segments (a phenomenon closely linked with expansion of the periph-
eral input). If central sensitization is established, innocuous tactile stimuli can become capable of
exciting nociceptive projection neurons leading to ‘touch-evoked’ hyperalgesia (Tal et al. 1994).
Recent reviews provide comprehensive overviews of the plastic changes in nociceptive trans-
mission neurons (Woolf and Salter 2000; Kuner 2010), a short synopsis of the most important
mechanisms is given here.
Inflammation or nerve injury causes long-term potentiation in lamina I nociception specific
neurons which express the neurokinin 1 receptor for substance P. This synaptic augmentation by
a co-transmitter involves activation of T-type voltage gated calcium channels (Ikeda et al. 2003,
2006). Whereas substance P (together with CGRP and neurokinin A) acts as co-transmitter at
spinal nociceptive synapses, the main transmitter is glutamate, acting at AMPA, NMDA, and
metabotropic receptors. When the postsynaptic neurons are depolarized, a magnesium block of
the NMDA-receptors is removed and a more profound depolarization takes place. This leads to
short-term enhancement of synaptic transmission which may be the starting point of many
longer lasting plastic changes. Whereas NMDA-R always permit Ca++ entry upon activation,
AMPA-R influence the influx of Ca++ in a more subtle way, by expression and inclusion of subu-
nits which cause greater Ca++ permeability (Burnashev et al. 1992). It has been shown that this
mechanism is crucial for inflammation induced plasticity of spinal transmission (Hartmann et al.
2004; Luo et al. 2008).
A key effect of synaptic plasticity is the increase of intracellular Ca++ level which leads to the
activation of several calcium-dependent kinases, e.g. CamKII alpha, cyclooxygenase-2 and
NO-synthase. Their products, prostaglandin E2 and NO, have been shown to facilitate nocicep-
tive transmission. Another important source of central sensitization is the mitogen-activated
protein kinase (MAPK) system, ERK1 and 2 (Ji and Woolf 2001; Kawasaki et al 2004). ERK 1 and
2 can phosphorylate ion channels and thus induce a short term potentiation, but they play also a
role in long term sensitization by acting as synapse-to-nucleus communicators changing gene
transcription (Garry et al. 2003a, 2003b).
4.3.2 Contribution of glial cells to neuropathic pain

In recent years increasing evidence has emerged that not only neuronal elements, but also glial
cells play an important role in the initiation and maintenance of neuropathic pain. In experimen-
tal pain states in animals, astrocytes and microglia are activated by neuronal transmitters includ-
ing substance P and glutamate (Wieseler-Frank et al. 2005 ). In turn, glial cells release
proinflammatory cytokines and other sensitizing compounds that promote long-lasting spinal
hyperexcitability and therefore are key factors for the transition from acute to chronic pain
(Chacur et al. 2009). For example, brain-derived neurotrophic factor (BDNF) released from acti-
vated glial cells fosters the reduction of the potassium chloride exporter (KCC2) in lamina I
neurons. A consequence is an alteration of chloride homeostasis in these cells leading to depolari-
zation. This shift inverts the polarity of currents activated by GABA which become excitatory and
thus enhance the sensitization of projecting neurons (Coull et al. 2003, 2005).
4.3.3 Central disinhibition

The output of the dorsal horn to brain centres depends on a balance of inhibitory and excitatory
mechanisms. Hyperalgesia is not only due to hypersensitivity of transmission neurons, but also to
a diminished control by inhibitory neurons. Locally, these are mainly GABA-ergic and glycinergic
interneurons, apart from the more nociception-specific enkephalinergic and dynorphinergic
interneurons of the endogenous opioid system.
It has been proposed that loss of GABAergic interneurons in the spinal dorsal horn after nerve
injury causes an imbalance between excitation and inhibition. Peripheral nerve injury may lead
to selective apoptosis of GABAergic interneurons in the dorsal horn (Moore et al. 2002; Scholz
et al. 2005). For the understanding of the effect of cyclo-oxygenase inhibitors it is important to
realise that PGE2 induces protein kinase A-dependent phosphorylation of glycin receptors
causing inhibition of these inhibitory interneurons and hence facilitation of nociceptive trans-
mission (Harvey et al. 2004). This important finding has the implication that the effect of
COX-inhibiting drugs is not confined to the primary afferents. These drugs have also central
actions (see Chapter 25).
Dorsal horn neurons are under a powerful descending control from facilitatory and inhibitory
supraspinal brainstem centres. The periaqueductal grey (PAG) and the nucleus raphe magnus
and adjacent structures of the rostral ventromedial medulla (RVM) project to the spinal cord
dorsal horn. This descending control pathway contains serotonergic neurons and also involves
noradrenergic neurons from the locus coeruleus and other noradrenergic brainstem cell groups.
Considerable evidence has recently emerged showing modulation of this system in persistent pain
conditions due to inflammation or neuropathy. It seems that persistent nociception simultane-
ously triggers descending facilitation and inhibition from the RVM (Porreca et al. 2002; Heinricher
et al. 2009). In models of inflammation, descending inhibition may predominate over facilitation
in pain circuits with input from the inflamed tissue, while descending facilitation predominates
over inhibition in pain circuits with input from neighbouring tissues, and thus facilitates second-
ary hyperalgesia (Vanegas and Schaible 2004). Although both descending facilitation and inhibi-
tion seem to stem from the RVM, they are probably due to different neuronal pools. According
to one hypothesis ‘on’ and ‘off’ neurons in this region, i.e. neurons reacting to switching on and
off nociceptive input, respectively, are responsible for the balance between facilitation and inhibi-
tion. This modulation may be one of the important target mechanisms for the endogenous opioid
antinociception (Fields 2000 ). Because in all models of inflammatory and neuropathic
pain descending facilitation and inhibition are triggered simultaneously, it will be important to
elucidate why inhibition predominates in some conditions and facilitation in others.
Tipping the balance between descending inhibition and facilitation towards inhibition of
chronic pain states is the final goal of centrally-acting pain therapies via serotonin and noradren-
aline re-uptake inhibitors, e.g. tricyclic antidepressants (see Chapter 25).
4.4 Supraspinal mechanisms of CLBP

4.4.1 Functional plasticity
When turning to the supraspinal mechanisms of chronification of LBP, the available evidence
on the level of individual neurons and synapses becomes scarce, though some synaptological
studies exist, e.g. involving thalamic neurons (Cheong et al. 2008) and neurons in the anterior
cingulated gyrus which is regarded as part of the ‘medial’, affective system (Cao et al. 2009: Jeon
et al. 2010).
Most relevant studies are performed in humans suffering from LBP using functional cerebral
imaging, mainly functional magnetic resonance imaging (fMRI). These methods have provided
an invaluable insight into the cerebral processing of acute and chronic pain. However, one has to
keep in mind that they do not allow a direct analysis at the cellular and synaptic level. In the time
and in the space domain the resolution is by several orders of magnitude too low for synaptic
studies, being typically more than 1mm3 and more than 1s in fMRI brain scans. Alternatively,
encephalographic (EEG) and magnetoencephalographic (MEG) mappings have been performed
which reflect the neuronal mass activity of neuron pools at a high time resolution, but with a
limited spatial resolution. These EEG and MEG recordings have also the disadvantage of reflect-
ing only cortical processing.
A thorough discussion of altered cerebral responsiveness in CLBP can be found in this book in
Chapter 6. Only a short summary will be given in this chapter for the purpose of discussing
pathophysiological implications.
There are numerous studies on pain-related cerebral activation patterns. The majority dealt
with the responses of healthy volunteers to experimentally-induced pain. However, some studies
tried to compare cerebral processing of CLBP patients and healthy subjects. In an early study,
MEG recordings were used (Flor et al. 1997). The experiment employed electrical stimulation of
the skin and revealed significant brain activation differences between CLBP patients and healthy
controls. Stimulation at the affected back skin lead to a significantly higher potential after about
80ms (N80) in CLBP patients. Since the signal strength and the duration of the chronic pain were
positively correlated, the authors reasoned that pain chronification leads to increased and also to
more widely spread cortical responses. These results have been confirmed and extended by this
group in a newer EEG study employing also intramuscular electrical stimuli (Diers et al. 2007).
According to these studies CLBP leads to an expansion of the cortical zone excited by a noxious
stimulus in the affected region and to stronger responses, regardless whether the affected tissue or
the overlying skin is stimulated.
Later on, fMRI became the predominant tool for those studies. With this method changes in
the local blood oxygenation are assessed (BOLD-effect) as an indication of the activation of
groups of neurons. The time resolution is 1000-fold lower than in EEG/MEG, but more tonic
stimuli can be used and sustained responses can be assessed. Subcortical areas became also acces-
sible with this method. In one such study CLBP patients were compared with controls, and with
other painful ailments. Tonic pressure pain was induced in a remote body site (Giesecke et al.
2004), in another study the patients were asked to rate their ongoing back pain in the course of an
fMRI session (Baliki et al. 2006). In both studies cortical brain regions were activated which have
been often associated with pain perceptions: the somatosensory cortex S1 and S2, the inferior
parietal lobe, the insula, and the anterior cingulated gyrus. These regions are often called the
‘pain matrix’ (Melzack 1999). Prior to the advent of functional brain imaging these cortical
regions had been related to pain processing, and in particular to different dimensions of pain
perception: a sensory-discriminative dimension which is supposed to be processed by a ‘lateral’
projection system mainly projecting to the somatosensory projection areas, an affective-
emotional dimension which is supposed to be represented in medial and limbic structures, and
an evaluative dimension related to frontal cortical fields (Melzack and Casey 1968; Treede et al.
1999). However, more recent fMRI studies raised some doubt, if there is really a ‘pain matrix’
reflected in the BOLD responses of the above structures. If in normal subjects non-painful tactile
stimuli were applied, very similar BOLD activation patterns emerged in the cerebral cortex
(Schoedel et al. 2008). In a later study it was shown that the fMRI responses triggered by nocic-
eptive stimuli can be largely explained by a combination of multimodal neural activities (i.e.
activities elicited by all stimuli regardless of sensory modality) and somatosensory-specific
but not nociceptive-specific neural activities (i.e. activities elicited by both nociceptive and non-
nociceptive somatosensory stimuli) (Mouraux et al. 2011). These authors concluded that the
BOLD responses in the so-called pain matrix reflect saliency of a stimulus rather than painfulness
per se (Legrain et al. 2011). These results do not necessarily imply that the activations in the dif-
ferent elements of the so-called pain matrix have nothing to do with pain processing. Certainly
pain, in particular LBP, has a greater saliency for the brain of a chronic pain patient than for a
healthy subject, and this greater saliency is part of the pain chronicity. Further, one has to keep in
mind that the spatial resolution of present fMRI devices is far too low to allow conclusions about
small neuronal circuits in the activated brain regions and their connections.
4.4.2 Structural plasticity

Two lines of evidence indicate that structural changes also occur in cortical regions as a conse-
quence of CLBP. Flor et al. (Flor et al. 1997; Flor 2002) observed a reorganization of the soma-
totopy in the primary sensory cortex comparable to that observed in amputees. This cortical
re-organization was reversible—at least in part—under successful therapy. Later on, several
morphometric MRT studies have been performed which showed consistently that the cortical
grey matter was shrinking in the course of the development of chronic pain. This topic is exten-
sively discussed in another chapter of this book (see Chapter 7). It is still unclear which mecha-
nisms are leading to these cortical changes. As far as they are reversible they are possibly not due
to the loss of neurons. A loss of dendrites and thus of functional synapses is an intriguing possibility.
Future cellular studies will certainly help to fully comprehend this striking phenomenon.
4.5 Conclusions
CLBP apparently starts with a trauma in the structures in and around the vertebral column. The
extension of this trauma is not the salient factor for the development of a chronic pain disease, but
a chronification process which is still not fully understood, in spite of the plethora of contributing
factors recently discovered which are outlined in this chapter.
Certainly, the special structure of our vertebral column—the continuous stress on it by
our upright locomotion combined with a predominantly sedentary way of life in modern times—
are disposing factors. However, it is not fully understood why some people get CLBP and others
not. Many factors may be genetic and/or psychosocial and will be explained in other chapters of
this book.
On a pathophysiological level, the interactions between cerebral cortex, brainstem, spinal cord
and the primary afferent system are only incompletely understood. For many questions we have
not yet developed a sufficiently powerful scientific repertoire.
References
Adriaensen H, Gybels J, Handwerker HO, Van Hees J. Response properties of thin myelinated (A-delta)
fibers in human skin nerves. J Neurophysiol 1983; 49:111–22.
Ahn HS, Dib-Hajj SD, Cox JJ, et al. A new Nav1.7 sodium channel mutation I234T in a child with severe
pain. Eur J Pain 2010; 14:944–50.
Amir R, Argoff CE, Bennett GJ, et al. The role of sodium channels in chronic inflammatory and
neuropathic pain. J Pain 2006; 7:S1–29.
Baliki MN, Chialvo DR, Geha PY, Levy RM, Harden RN, Parrish TB, Apkarian AV. Chronic pain and the
emotional brain: specific brain activity associated with spontaneous fluctuations of intensity of chronic
back pain. J Neurosci 2006; 26:12165–73.
Baron R. Neuropathic Pain: Clinical. In: Basbaum AI, Bushnell CM (eds) Science of Pain. Amsterdam:
Elsevier, 2009. pp. 865–900.
Bennett GJ, Xie YK. A peripheral mononeuropathy in rat that produces disorders of pain sensation like
those seen in man. Pain 1988; 33:87–107.
Bessou P, Perl ER. Responses of cutaneous sensory units with unmyelinated fibers to noxious stimuli.
J Neurophysiol 1969; 32:1025–43.
Burnashev N, Monyer H, Seeburg PH, Sakmann B. Divalent ion permeability of AMPA receptor channels
is dominated by the edited form of a single subunit. Neuron 1992; 8:189–98.
Cao H, Gao YJ, Ren WH, et al. Activation of extracellular signal-regulated kinase in the anterior cingulate
cortex contributes to the induction and expression of affective pain. J Neurosci 2009; 29:3307–21.
Carr RW, Pianova S, McKemy DD, Brock JA. Action potential initiation in the peripheral terminals of
cold-sensitive neurones innervating the guinea-pig cornea. J Physiol 2009; 587:1249–64.
Chacur M, Lambertz D, Hoheisel U, Mense S. Role of spinal microglia in myositis-induced central
sensitisation: an immunohistochemical and behavioural study in rats. Eur J Pain 2009; 13:915–23.
Chemin J, Patel AJ, Duprat F, Sachs F, Lazdunski M, Honore E. Up- and down-regulation of the
mechano-gated K(2P) channel TREK-1 by PIP (2) and other membrane phospholipids. Pflugers Arch
2007; 455:97–103.
Cheong E, Lee S, Choi BJ, Sun M, Lee CJ, Shin HS. Tuning thalamic firing modes via simultaneous
modulation of T- and L-type Ca2+ channels controls pain sensory gating in the thalamus. J Neurosci
2008; 28:13331–40.
Coull JA, Beggs S, Boudreau D, et al. BDNF from microglia causes the shift in neuronal anion gradient
underlying neuropathic pain. Nature 2005; 438:1017–21.
Coull JA, Boudreau D, Bachand K, et al. Trans-synaptic shift in anion gradient in spinal lamina I neurons
as a mechanism of neuropathic pain. Nature 2003; 424:938–42.
Decosterd I, Berta T. Animal Models and Neuropathic Pain. In: Basbaum A.I., Bushnell C.M., editors.
Science of Pain. Amsterdam: Elsevier, 2009. pp. 857–64.
Decosterd I, Woolf CJ. Spared nerve injury: an animal model of persistent peripheral neuropathic pain.
Pain 2000; 87:149–58.
Dib-Hajj SD, Yang Y, Waxman SG. Genetics and molecular pathophysiology of Na(v)1.7-related pain
syndromes. Adv Genet 2008; 63:85–110.
Diers M, Koeppe C, Diesch E, et al. Central processing of acute muscle pain in chronic low back pain
patients: an EEG mapping study. J Clin Neurophysiol 2007; 24:76–83.
Ellrich J, Makowska A. Nerve growth factor and ATP excite different neck muscle nociceptors in
anaesthetized mice. Cephalalgia 2007; 27:1226–35.
Estacion M, Harty TP, Choi JS, Tyrrell L, Dib-Hajj SD, Waxman SG. A sodium channel gene SCN9A
polymorphism that increases nociceptor excitability. Ann Neurol 2009; 66:862–6.
Fields HL. Pain modulation: expectation, opioid analgesia and virtual pain. Prog Brain Res 2000;
122:245–53
Flor H. The modification of cortical reorganization and chronic pain by sensory feedback. Appl
Psychophysiol Biofeedback 2002; 27:215–27.
Flor H, Braun C, Elbert T, Birbaumer N. Extensive reorganization of primary somatosensory cortex in

chronic back pain patients. Neurosci Lett 1997; 224:5–8.
Garry EM, Moss A, Delaney A, et al. Neuropathic sensitization of behavioral reflexes and spinal NMDA
receptor/CaM kinase II interactions are disrupted in PSD-95 mutant mice. Curr Biol 2003a; 13:321–8.
Garry EM, Moss A, Rosie R, Delaney A, Mitchell R, Fleetwood-Walker SM. Specific involvement in
neuropathic pain of AMPA receptors and adapter proteins for the GluR2 subunit. Mol Cell Neurosci
2003b; 24:10–22.
Giesecke T, Gracely RH, Grant MA, et al. Evidence of augmented central pain processing in idiopathic
chronic low back pain. Arthritis Rheum 2004; 50:613–23.
Gold MS, Gebhart GF. Nociceptor sensitization in pain pathogenesis. Nat Med 2010; 16:1248–57.
Goldberg YP, MacFarlane J, MacDonald ML, et al. Loss-of-function mutations in the Nav1.7 gene underlie
congenital indifference to pain in multiple human populations. Clin Genet 2007; 71:311–19.
Handwerker HO. What is a polymodal nociceptor? J Pain 2008; 9:309–10.
Handwerker HO. Primary afferent nociceptors. In: Basbaum Aed (ed) The Biomedical & Life Sciences
Collection, Henry Stewart Talks Henry Stewart Talks, Ltd, London, 2009.
Handwerker HO. Classification of nociceptors - To what purpose? Pain 2010; 148(3):355–6.
Hartmann B, Ahmadi S, Heppenstall PA, et al. The AMPA receptor subunits GluR-A and GluR-B
reciprocally modulate spinal synaptic plasticity and inflammatory pain. Neuron 2004; 44:637–50.
Harvey RJ, Depner UB, Wassle H, et al. GlyR alpha3: an essential target for spinal PGE2-mediated
inflammatory pain sensitization. Science 2004; 304:884–7.
Hefti FF, Rosenthal A, Walicke PA, et al. Novel class of pain drugs based on antagonism of NGF. Trends
Pharmacol Sci 2006; 27:85–91.
Heinricher MM, Tavares I, Leith JL, Lumb BM. Descending control of nociception: Specificity,
recruitment and plasticity. Brain Res Rev 2009; 60:214–25.
Hoheisel U, Reinohl J, Unger T, Mense S. Acidic pH and capsaicin activate mechanosensitive group IV
muscle receptors in the rat. Pain 2004; 110:149–57.
Hoheisel U, Unger T, Mense S. Excitatory and modulatory effects of inflammatory cytokines and
neurotrophins on mechanosensitive group IV muscle afferents in the rat. Pain 2005; 114:168–76.
Hoheisel U, Unger T, Mense S. Sensitization of rat dorsal horn neurons by NGF-induced subthreshold
potentials and low-frequency activation. A study employing intracellular recordings in vivo. Brain Res
2007; 1169:34–43.
Hucho TB, Dina OA, Levine JD. Epac mediates a cAMP-to-PKC signaling in inflammatory pain: an
isolectin B4(+) neuron-specific mechanism. J Neurosci 2005; 25:6119–26.
Hudson LJ, Bevan S, Wotherspoon G, Gentry C, Fox A, Winter J. VR1 protein expression increases in
undamaged DRG neurons after partial nerve injury. Eur J Neurosci 2001; 13:2105–14.
Igarashi T, Kikuchi S, Shubayev V, Myers RR. (2000). Volvo Award winner in basic science studies:
Exogenous tumor necrosis factor-alpha mimics nucleus pulposus-induced neuropathology. Molecular,
histologic, and behavioral comparisons in rats. Spine 25:2975–80.
Ikeda H, Heinke B, Ruscheweyh R, Sandkuhler J. Synaptic plasticity in spinal lamina I projection neurons
that mediate hyperalgesia. Science 2003; 299:1237–40.
Ikeda H, Stark J, Fischer H, et al. Synaptic amplifier of inflammatory pain in the spinal dorsal horn. Science
2006; 312:1659–62.
Indo Y, Tsuruta M, Hayashida Y, et al. Mutations in the TRKA/NGF receptor gene in patients with
congenital insensitivity to pain with anhidrosis. Nat Genet 1996; 13:485–8.
Jeon D, Kim S, Chetana M, et al. Observational fear learning involves affective pain system and Cav1.2
Ca2+ channels in ACC. Nat Neurosci 2010; 13:482–8.
Ji RR, Woolf CJ. Neuronal plasticity and signal transduction in nociceptive neurons: implications for the
initiation and maintenance of pathological pain. Neurobiol Dis 2001; 8:1–10.
Kawasaki Y, Kohno T, Zhuang ZY, et al. Ionotropic and metabotropic receptors, protein kinase A, protein
kinase C, and Src contribute to C-fiber-induced ERK activation and cAMP response element-binding
protein phosphorylation in dorsal horn neurons, leading to central sensitization. J Neurosci 2004; 24:8310–21.
Kruger L. Morphological features of thin sensory afferent fibers: a new interpretation of ‘nociceptor’
function. Prog Brain Res 1988; 74:253–7.
Kuner R. Central mechanisms of pathological pain. Nat Med 2010; 16:1258–66.
Lee S, Moon CS, Sul D, et al. Comparison of growth factor and cytokine expression in patients with
degenerated disc disease and herniated nucleus pulposus. Clin Biochem 2009; 42:1504–11.
Legrain V, Iannetti GD, Plaghki L, Mouraux A. The pain matrix reloaded A salience detection system for
the body. Prog Neurobiol 2011; 93(1):111–24.
Lewin GR, Rueff A, Mendell LM. Peripheral and central mechanisms of NGF-induced hyperalgesia. Eur
J Neurosci 1994; 6:1903–12.
Lu X, Richardson PM. Responses of macrophages in rat dorsal root ganglia following peripheral nerve
injury. J Neurocytol 1993; 22:334–41.
Luo C, Seeburg PH, Sprengel R, Kuner R. Activity-dependent potentiation of calcium signals in spinal
sensory networks in inflammatory pain states. Pain 2008; 140:358–67.
Ma W, Zhang Y, Bantel C, Eisenach JC. Medium and large injured dorsal root ganglion cells increase
TRPV-1, accompanied by increased alpha2C-adrenoceptor co-expression and functional inhibition by
clonidine. Pain 2005; 113:386–94.
Maingret F, Fosset M, Lesage F, Lazdunski M, Honore E. TRAAK is a mammalian neuronal
mechano-gated K+ channel. J Biol Chem 1999; 274:1381–7.
Melzack R. From the gate to the neuromatrix. Pain 1999; Suppl 6, S121–6.
Melzack R, Casey KL. Sensory, motivational, and central control determinants of pain. In: Kenshalo DR
(ed) The skin senses. Springfield, IL: Thomas, 1968. pp. 423–43.
Meyer RA, Davis KD, Cohen RH, Treede RD, Campbell JN. Mechanically insensitive afferents (MIAs) in
cutaneous nerves of monkey. Brain Res 1991; 561:252–61.
Moore KA, Kohno T, Karchewski LA, Scholz J, Baba H, Woolf CJ. Partial peripheral nerve injury promotes
a selective loss of GABAergic inhibition in the superficial dorsal horn of the spinal cord. J Neurosci
2002; 22:6724–31.
Mouraux A, Diukova A, Lee MC, Wise RG, Iannetti GD. A multisensory investigation of the functional
significance of the ‘pain matrix’. Neuroimage 2011; 54(3):2237–49.
Murase S, Terazawa E, Queme F, et al. Bradykinin and nerve growth factor play pivotal roles in muscular
mechanical hyperalgesia after exercise (delayed-onset muscle soreness). J Neurosci 2010; 30:3752–61.
Norimoto M, Ohtori S, Yamashita M, et al. Direct application of the TNF-alpha inhibitor, etanercept, does
not affect CGRP expression and phenotypic change of DRG neurons following application of nucleus
pulposus onto injured sciatic nerves in rats. Spine 2008; 33:2403–8.
Obata K, Katsura H, Mizushima T, et al. TRPA1 induced in sensory neurons contributes to cold
hyperalgesia after inflammation and nerve injury. J Clin Invest 2005; 115:2393–401.
Omana-Zapata I, Khabbaz MA, Hunter JC, Clarke DE, Bley KR. Tetrodotoxin inhibits neuropathic ectopic
activity in neuromas, dorsal root ganglia and dorsal horn neurons. Pain 1997; 72:41–9.
Ossipov MH, Porreca F. Neuropathic Pain:Basic Mechanisms (animal). In: Basbaum A.I., Bushnell C.M.
(eds) Science of Pain. Amsterdam: Elsevier, 2009. pp. 833–55.
Patel AJ, Honore E. Properties and modulation of mammalian 2P domain K+ channels. Trends Neurosci
2001; 24:339–46.
Porreca F, Ossipov MH, Gebhart GF. Chronic pain and medullary descending facilitation. Trends Neurosci
2002; 25:319–25.
Reimann F, Cox JJ, Belfer I, et al. Pain perception is altered by a nucleotide polymorphism in SCN9A. Proc
Natl Acad Sci U S A 2010; 107:5148–53.
Richardson JD, Vasko MR. Cellular mechanisms of neurogenic inflammation. J Pharmacol Exp Ther 2002;
302:839–45.
Rukwied R, Mayer A, Kluschina O, Obreja O, Schley M, Schmelz M. NGF induces non-inflammatory
localized and lasting mechanical and thermal hypersensitivity in human skin. Pain 2010; 148:407–13.
Sato J, Perl ER. Adrenergic excitation of cutaneous pain receptors induced by peripheral nerve injury.
Science 1991; 251:1608–10.
Schaible HG, von Banchet GS, Boettger MK, et al. The role of proinflammatory cytokines in the generation
and maintenance of joint pain. Ann N Y Acad Sci 2010; 1193:60–9.
Schaible H-G, Schmidt RF. Responses of fine medial articular nerve afferents to passive movements of knee
joints. J Neurophysiol 1983; 49:1118–26.
Schmelz M, Michael K, Weidner C, Schmidt R, Torebjork HE, Handwerker HO. Which nerve fibers
mediate the axon reflex flare in human skin? Neuroreport 2000; 11:645–8.
Schmidt R, Schmelz M, Forster C, Ringkamp M, Torebjork E, Handwerker H. Novel classes of responsive
and unresponsive C nociceptors in human skin. J Neurosci 1995; 15:333–41.
Schmidt R, Schmelz M, Torebjork HE, Handwerker HO. Mechano-insensitive nociceptors encode pain
evoked by tonic pressure to human skin. Neuroscience 2000; 98:793–800.
Schoedel AL, Zimmermann K, Handwerker HO, Forster C. The influence of simultaneous ratings on
cortical BOLD effects during painful and non-painful stimulation. Pain 2008; 135:131–41.
Scholz J, Broom DC, Youn DH, et al. Blocking caspase activity prevents transsynaptic neuronal apoptosis
and the loss of inhibition in lamina II of the dorsal horn after peripheral nerve injury. J Neurosci 2005;
25:7317–23.
Sluka KA, Kalra A, Moore SA. Unilateral intramuscular injections of acidic saline produce a bilateral,
long-lasting hyperalgesia. Muscle Nerve 2001; 24:37–46.
Snider WD, McMahon SB. Tackling pain at the source: new ideas about nociceptors. Neuron 1998; 20:629–32.
Sommer C. Painful neuropathies. Curr Opin Neurol 2003; 16:623–8.
Sorkin LS, Puig S, Jones DL. Spinal bicuculline produces hypersensitivity of dorsal horn neurons: effects of
excitatory amino acid antagonists. Pain 1998; 77:181–90.
Taguchi T, Hoheisel U, Mense S. Dorsal horn neurons having input from low back structures in rats. Pain
2008; 138:119–29.
Taguchi T, John V, Hoheisel U, Mense S. Neuroanatomical pathway of nociception originating in a low
back muscle (multifidus) in the rat. Neurosci Lett 2007; 427:22–7.
Tal M, Bennett GJ. Extra-territorial pain in rats with a peripheral mononeuropathy: mechano-hyperalgesia
and mechano-allodynia in the territory of an uninjured nerve. Pain 1994; 57:375–82.
Torebjörk HE, Schmelz M, Handwerker HO. Functional properties of human cutaneous nociceptors and
their role in pain and hyperalgesia. In: Belmonte C, Cervero F (eds) Neurobiology of Nociceptors.
Oxford: Oxford University Press, 1996. pp. 349–69.
Treede RD, Kenshalo DR, Gracely RH, Jones AK. The cortical representation of pain. Pain 1999; 79:105–11.
Vanegas H, Schaible HG. Descending control of persistent pain: inhibitory or facilitatory? Brain Res Brain
Res Rev 2004; 46:295–309.
Walder RY, Rasmussen LA, Rainier JD, Light AR, Wemmie JA, Sluka KA. ASIC1 and ASIC3 play different
roles in the development of Hyperalgesia after inflammatory muscle injury. J Pain 2010; 11:210–18.
Wieseler-Frank J, Maier SF, Watkins LR. Immune-to-brain communication dynamically modulates pain:
physiological and pathological consequences. Brain Behav Immun 2005; 19:104–11.
Wood JN, Boorman JP, Okuse K, Baker MD. Voltage-gated sodium channels and pain pathways.
J Neurobiol 2004; 61:55–71.
Woolf CJ, Ma Q. Nociceptors—noxious stimulus detectors. Neuron 2007; 55:353–64.
Woolf CJ, Salter MW. Neuronal plasticity: increasing the gain in pain. Science 2000; 288:1765–69.
Wu G, Ringkamp M, Hartke TV, et al. Early onset of spontaneous activity in uninjured C-fiber nociceptors
after injury to neighboring nerve fibers. J Neurosci 2001; 21:RC140.
Yamashita M, Ohtori S, Koshi T, et al. Tumor necrosis factor-alpha in the nucleus pulposus mediates
radicular pain, but not increase of inflammatory peptide, associated with nerve damage in mice. Spine
2008; 33:1836–42.
Chapter 5
Dysfunction of the
Hypothalamic–Pituitary–Adrenal Axis
and Associated Stress Axes in the
Development of Chronic Low Back Pain
John McBeth and Andrea Power
5.1 Introduction
As discussed in Parts 2 to 4 of this volume, low back pain (LBP) is a complex disorder whose
aetiology is associated with a host of factors across multiple domains that include biological, psy-
chological, and individual- and higher-order social factors. By definition chronic LBP is LBP that
persists beyond the normal healing time required for any insult or injury that may have preceded
the onset of symptoms (IASP 2002). That is, for most cases, LBP in the absence of obvious pathol-
ogy. The challenge then is to identify the factors that are associated with symptom perpetuation,
or chronification.
It is clear that to determine the factors that are associated with the transition from acute to
chronic LBP, we cannot limit our perspective. Focusing on the role of the biological determinants
of chronicity in isolation from psychological and social factors would be meaningless. The inter-
relationships between risk factors for chronicity can be usefully conceptualized through a tradi-
tional biopsychosocial model. However in an excellent theoretical paper (Deary et al. 2007)
the biopsychosocial model has been expanded to develop a model of symptom perpetuation that
is autopoietic (a term borrowed from systems theory and cell biology that literally means ‘self-
creating’) (see Figure 5.1). The core tenet of that model is a multifactorial autopoietic cycle that
underlies the perpetuation of symptoms. The strength of this conceptualization is that while
across groups of individuals there may be similarities in the factors associated with the chronifica-
tion of LBP, the model allows for interactions between these factors that may be unique to sub-
groups or even the individual experience and addresses the important issue of phenotypic
heterogeneity. Expanding beyond the original aim of the Deary et al. (2007) paper (to explore
the onset and perpetuation of ‘medically unexplained’ symptoms such as chronic fatigue and
irritable bowel syndrome) this model can usefully focus our thoughts on the role of multiple fac-
tors in the perpetuation of LBP. The reader is encouraged to place the findings reported below in
the context of the model outlined in Figure 5.1.
Stress response systems are notoriously sensitive to multiple factors: prior experience and life
stress (Faravelli et al. 2010), psychological distress, cognitive and behavioural factors (Geiss et al.
2005), and body composition (Vicennati et al. 2009) to name a few. In this chapter we will outline
the putative role of innate stress response systems in the perpetuation of pain. The main focus will
be the primary stress response system in humans the hypothalamic–pituitary–adrenal (HPA) axis.
We will also briefly explore the relationship with other related systems; the hypothalamic–
pituitary–gonadal (HPG) and hypothalamic–pituitary–growth hormone (HPGH) axes.
Predisposing and precipitating factors
Life events
Early adversity Threat sensitivity
stressors
Physiological sensitization
Distress intolerance
Chronic low
back pain
Cognitive processes
Attention Physiology Behaviours Social factors
Attribution Arousal Avoidance Medical uncertainty
Rumination Hypocortisolism Illness response Secondary gain
Catastrophizing
Perpetuating factors
Fig. 5.1 Autopoietic cycle of low back pain maintenance. This model adapted from Deary et al.
(2007) proposes that low back pain is a multifactorial disorder and that pain chronification is
associated with a number of predisposing and precipitating factors (early experiences, threat
sensitivity, life events) that interact with perpetuating factors (cognitive processes, physiology,
behaviours, social) in an autopoietic (self-perpetuating) cycle. This model explains why pain may
persist in the absence of physical pathology.
We will argue that these stress response systems may bring about the transition from acute to
chronic LBP.
However, while the available data is suggestive, these relationships have not been adequately
tested in well-conducted longitudinal epidemiological studies. Data from other disorders that are
strongly associated with stress and that have pain as the cardinal symptom will occasionally be
included in the discussion where data is not directly available on LBP. We conclude with a research
recommendation that outlines a testable hypothesis to establish the true relationship between
functioning of stress response systems and the chronification of LBP.
5.2 Allostatic load: the molecular switch from acute to

chronic pain?
Allostasis enables the system to maintain stability through adaptation by the activation of differ-
ent physiological regulatory pathways. These adjustments are homeostatic mechanisms vital for
survival; however persistent adaptation of the system to chronic stressors is reported to have a
detrimental effect (Chapman et al. 2008). The point at which the system is unable to successfully
adapt to this enduring stress is known as the ‘allostatic load’. At this point, activity within homeo-
static systems appears to be permanently altered, with disruptions to signalling and hence
communication between key pathways. Animal models displayed significant alterations within
the limbic system following 21 days of stress. Here structural alterations to the hippocampus and
amygdale were observed, as well as changes in behaviours such as anxiety and aggression (McEwen
2003; Pham et al. 2003).
HPA AXIS AND BACK PAIN 71
In typical situations, activation of the stress pathways discussed below enables physiological
adaptation to short term stressors. However, chronic exposure to emotional and physical stress
results in the alteration and adaptation of the HPA axis and associated axes to the environment.
Although there are sizeable gaps in our knowledge regarding the molecular transition associated
with chronic stress states, HPA dysfunctions are usually associated with significant alterations to
cortisol regulation. This may eventually manifest as either hypocortisolism or hypercortisolism.
For the purpose of this review we will focus on hypocortisolism and examine abnormal commu-
nication between the HPA and its surrounding pathways which may influence the transition from
acute to chronic pain.
5.3 Stress and the HPA axis

5.3.1 Normal HPA function
The HPA axis, the central pathway for stress-adaptation, comprises three brain sites: the hypoth-
alamus, the anterior pituitary, and the adrenal cortex (Papadimitriou & Priftis 2009). The
upstream component of this pathway, corticotrophin releasing hormone (CRH), is expressed by
cells within the paraventricular nucleus (PVN) of the hypothalamus (see Figure 5.2). It is then
secreted into the hypophyseal portal system, before being carried to the anterior pituitary, where
it stimulates the release of proopiomelanocortin (POMC). POMC is a large molecule which dif-
ferentially cleaved, produces peptides with a variety of biological functions, such as endogenous
opioid, ß-endorphin and ß-melanocyte stimulating hormone (ß-MSH). However, for the pur-
pose of this section, we will concentrate on the cleavage product and stress axis regulator, adreno-
corticotropin hormone (ACTH). ACTH subsequently triggers the release of the key stress
hormone, cortisol, from the zona reticularis of the adrenal cortex.
On release into the blood system, approximately 90% of cortisol in circulation binds to special-
ized plasma proteins. The remaining unbound or ‘free’ cortisol is biologically active. This compo-
nent, is not only responsible for driving the following stress response, but in sufficient
concentrations, suppresses HPA function at both hypothalamic and pituitary levels, and regulates
a number of further physiological actions, such as metabolism, immune response and memory
consolidation (Harbuz 2002a, 2002b; Henckens et al. 2009; Mussig et al. 2010). Also, by control-
ling glucose metabolism, cortisol modulates glucose levels within the blood, providing energy
substrates for key organs such as the brain and heart. Cortisol is also responsible for the suppres-
sion of immune function. It impedes the release of arachidonic acid, a key precursor for numer-
ous inflammatory mediators, preventing the differentiation of mast cells, and the production of
nitric oxide, interleukins and gamma interferon (Harbuz 2002b). Furthermore, cortisol exerts its
effect on the CNS by directly modulating neuronal electrical activity via type 1 and 2 glucocorti-
coid receptors, which are widely expressed in the limbic system and hippocampus. Indeed cortisol
has demonstrated the ability to reduce hippocampal volume as well as memory (Kumsta et al.
2010; Muller et al. 2002; Patel et al. 2008).
Although CRH, ACTH and cortisol are secreted in small intermittent pulses throughout the
day, plasma levels of ACTH and cortisol characteristically peak during early morning, initiated by
the release of CRH several hours before waking. Subsequently, plasma ACTH and cortisol levels
are at nadir round the time of sleep (Lightman et al. 2000).
5.3.2 The impact of acute stress

On receiving stress-associated signals from brain areas, such as the limbic system, cells of the PVN
synthesize and release CRH, resulting in cortisol release into the bloodstream (Figure 5.3a).
HYP
CRH GnRH
SS GHRH
PT
GnIH
POMC
GH LH,
α-MSH
Ghrelin β-End FSH
CNS
LIVER function ACTH
Analgesia
5-HT
IGF-1 AC
Ovaries Oestrogen &
Bone & muscle ↑Gluconeogenesis
testosterone
growth
Testes
IL-6
TNF-α
Cortisol
HPGH axis ↑Plasma
HPG axis
Glucose
Cytokines
Stress
adaptation
HPA axis
Fig. 5.2 (Also see Colour Plate 1) The three hypothalamic axes. In humans, the primary stress
response system, the hypothalamic pituitary adrenal (HPA) axis (grey), interacts with two other
related neuroendocrine pathways, the hypothalamic pituitary growth hormone (HPGH, green) and
hypothalamic pituitary gonodal (HPG, blue) axes. Under normal, stress-free conditions, these three
hypothalamic pathways function independently and maintain homeostasis through positive (arrows)
and negative (bars) regulation. Each axis is driven by hormonal secretion from the pituitary gland
(PT). Corticotropin releasing hormone (CRH) released from the hypothalamus (HYP) initiates HPA
activity, by stimulating pro-opiomelanocortin (POMC) production from the PT POMC is sequentially
cleaved into a number of active components such as adrenocorticotropin hormone (ACTH)
β-endorphin (β-E) and α-melanocyte-stimulating hormone (α-MSH). ACTH is responsible for
activating the release of downstream element, cortisol, from the adrenal cortex (AC). Cortisol then
suppresses HPA function at various levels as well as inhibiting the production of immune system
components, such as interleukin-6 (IL-6) and tumour necrosis factor alpha (TNF-α). In comparison,
HPGH activity is instigated by the release of gonadotrophin releasing hormone (GnRH) and the
subsequent secretion of growth hormone (GH) by the PT Circulating GH then induces the liver to
secrete insulin-like growth factor 1 (IGF-1) resulting in bone and muscle growth and
gluconeogenesis. HPG activity is prompted by the hypothalamic release of gonadotrophin releasing
hormone (GnRH) which stimulates luteinizing hormone (LH) and follicle stimulating hormone (FSH)
secretion from the PT This results in the downstream release of oestrogen and testosterone and
sequential inhibition of GnRH. Further, control of the HPG pathway is elicited by gonadotrophin
inhibitory hormone (GnIH) which inhibits GnRH, LH and FSH secretion.
Under stressful conditions, cortisol re-establishes physiological equilibrium within the organism,
by halting the stress response, and consequently initiating the recovery process and coping mech-
anisms (Olff et al. 1995; Stein-Behrens & Sapolsky 1992; Ursin & Olff 1993). Cortisol impedes
further stress response by inhibiting the HPA axis at several levels. Within minutes of cortisol
release, CRH and ACTH releasing cells of the PVN and anterior pituitary are suppressed (Steckler
et al. 1999). Then, approximately 60–90 minutes later, cortisol inhibits the hippocampal cells
responsible for HPA activation (ffrench-Mullen 1995; Young 1995; Young & Vazquez 1996). It
therefore makes sense that during acute stress, HPA hyperactivity has been recorded with reported
increases in circulating ACTH, urinary cortisol and CSF-CRH (Reul et al. 2000; Steckler et al.
1999; Van Praag 2004).
However slight elevations in basal cortisol levels are sufficient to induce a metabolic response,
immunosuppressive effect or alter blood pressure regulation—all of which may be detrimental to
the system. In fact, HPA hyperactivity and hypercortisolism has been associated with hyperten-
sion, abdominal obesity, diabetes II and osteoporosis (Chiodini et al. 2008; Kudielka et al. 2004;
Morelli et al. 2010).
5.3.3 The impact of chronic stress

It is difficult to ascertain the juncture at which HPA function is permanently altered by stressors.
Persistent exposure to stress down-regulates critical components of the HPA axis, resulting in
(a)
HYP
↑CRH ↓GnRH
↑SS GHRH
PT
GnIH
↑POMC
↓GH ↓LH,
α-MSH
↑Ghrelin β-End FSH
CNS
LIVER function ↑ACTH
Analgesia
5-HT
↓IGF-1 AC
Ovaries ↓Oestrogen &
testosterone
↓Bone & muscle ↓Gluconeogenesis Testes
growth IL-6
TNF-α
↑Cortisol
HPGH axis
↑Plasma
Glucose Cytokines
Stress
adaptation
Fig. 5.3 (Also see Colour Plate 2) The HPA dampens HPGH and HPG activity during acute stress (a),
while chronic stress leads to HPA hypoactivity, excessive cytokine production and further
down-regulation of the HPGH and HPG axes (b). HYP, hypothalamus; CRH, corticotropin releasing
hormone; ACTH, adrenocorticotropin hormone; AC, adrenal cortex; POMC, pro-opiomelanocortin;
β-End, β-endorphin; α-MSH, α-melanocyte-stimulating hormones; SS, somatostatin; GnRH,
gonadotrophin releasing hormone; LH, luteinizing hormone; FSH, follicle stimulating hormone;
GnIH, gonadotrophin inhibitory hormone.
(a) Acute stress up-regulates the production of components within the hypothalamic–pituitary–
adrenal (HPA) axis, which then down-regulate both hypothalamic pituitary growth hormone (HPGH)
and hypothalamic–pituitary–gonodal (HPG) activity. Cortisol suppresses growth hormone (GH)
activity both directly and indirectly (via activation of GH-inhibiting components somatostation (SS)
and ghrelin). Cortisol also prevents secretion of gonadotrophin releasing hormone (GnRH),
luteinizing hormone (LH) and follicle stimulating hormone (FSH), as well as suppressing the release
of immune elements, interleukin-6 (IL-6) and tumour necrosis factor alpha (TNF-α, red bars).
(b)
HYP
↓CRH ↓GnRH
↑SS GHRH
X
PT
GnIH
X ↓POMC
X
↓GH XX ↓LH,
↑Ghrelin
α-MSH
β-End X FSH
LIVER
CNS
function
↓ACTH
5-HT
Analgesia
X
↓IGF-1 AC Sympathetic
mediated
pain &
Ovaries ↓Oestrogen &
fatigue testosterone
↓Bone & muscle ↓Gluconeogenesis
Testes
growth X ↑SP
↓Cortisol 5HT
HPGH axis ↓Plasma X HPG axis
Glucose X ↑Cytokines
Stress ie. IL-6,
TNF-α
adaptation
HPA axis
Fig. 5.3 (Continued) (b) HPA activity is severely compromised during chronic stress, resulting in
HPA hypoactivity and pituitary gland (PT) desensitization. Reduced cortisol levels fail to regulate
cytokine-suppression resulting in the subsequent release of cytokines such as IL-6 and TNF-α.
Indeed, these cytokines are responsible for suppressing a variety of components within the HPGH
and HPG axes. Cytokines inhibit serotonin (5-HT) production within the HPGH, leading indirectly to
reductions in GH, as well as suppression of HPG activity at various levels (red bars). In addition,
5-HT suppression increases CWP-related symptoms, such as sympathetic mediated pain and fatigue,
via elevated substance P (SP) levels. Arrows and bars indicate the positive and negative regulation,
respectively.
ACTH deficiency and prevents the subsequent secretion of cortisol from the adrenal gland
(Figure 5.3b). Hypocortisolism is a deficiency associated with pain disorders, fatigue and enhanced
stress sensitivity (Fries et al. 2005; Meeus et al. 2008; Neeck & Crofford 2000). Indeed, patients
with fibromyalgia (FM), a disorder characterized by chronic widespread body pain (Wolfe
1990) have reduced levels of urinary and plasma cortisol, cortisol feedback resistance and hypo-
responsive pituitary function (Griep et al. 1998b), as well as abnormal immune response. As
cortisol is responsible for downstream immune-suppression, hypocortisolism results in atypical
elevations in cytokines IL-8, TNF-α, IL-1 and IL-6 (Maier & Watkins 1998). Indeed the excessive
levels of cytokines may contribute to pain phenotypes including LBP. Impairments within this
axis ultimately impact on communications between the parallel HPG and HPGH pathways.
5.4 Stress and HPG axis function

5.4.1 Normal function
Cells containing gonadotropin-releasing hormone (GnRH) project from the olfactory
bulb via the hypothalamus and preoptic area to the medial basal hypothalamus (MBH) and
median eminence (ME). Ensuing GnRH secretion (Figure 5.2) from the ME stimulates the release
of luteinizing hormone (LH) and follicle stimulating hormone (FSH) from gonadotropic cells of
the anterior pituitary into the portal system. Circulating LH and FSH bind to receptors of the
ovary and testis to up-regulate sex steroid production and gametogenesis. In the testes, LH
induces testosterone production from the Leydig cells, FSH induces testicular growth and expres-
sion of androgen-binding protein by Sertoli cells, while a combination of these hormones ensures
spermatozoa maturation. In the ovaries, LH drives oestrogen and progesterone production and
ovulation, FSH regulates the development of the ovarian follicle, and combined LH and FSH
control the follicular secretion of oestrogen. LH and FSH then, by directly targeting the pituitary,
act as potent negative feedback regulators of the HPG axis exerting homeostatic control
(Vadakkadath & Atwood 2005).

Gonadotropin inhibitory hormone (GnIH) opposes the action of GnRH, by inhibiting the pro-
duction of LH and FSH and it has been implicated in stress-mediated reproductive dysfunction
(Figure 5.3a). In rodents, GnIH expression occurs within cells of the dorsomedial nucleus of the
hypothalamus (DMH). These cells then project to the ME and to the POA, the aforementioned
area of GnRH neurons secretion (Kriegsfeld et al. 2006). Receptors homologous to GnIH recep-
tors are expressed in the hypothalamus, pituitary, and testes of other species (Bentley et al. 2008;
Yin et al. 2005). In vivo, up-regulation of this receptor or the GnIH ligand, suppress HPG func-
tion, LH secretion and sexual activity (Bentley et al. 2006; Calisi et al. 2008; Murakami et al. 2008;
Osugi et al. 2004). This perhaps suggests that GnIH acts as a mediator of stressors on mammalian
reproduction. Furthermore, GnIH-expressing cells also express stress hormone receptors and as
an adrenalectomy prevents increased GnIH expression following stress, it may be that the HPG
axis is regulated by stress-induced adrenal hormone release.
The HPA and HPG axes appear to have a bidirectional regulatory relationship. HPG activity is
controlled by components of the HPA, and visa-versa. Firstly, both CRH and glucocorticoids
are shown to suppress HPG function by altering GnRH production. Whilst CRH inhibits the
expression of GnRH from neurons of the arcuate nucleus, glucocorticoids appear to regulate
HPG activity on multiple levels. Hence, glucocorticoids not only suppress GnRH production,
pituitary secretion of LH and steroidogenesis within the gonads, they also desensitize target
tissues to circulating sex-steroids (Rabin et al. 1990; Rivier et al. 1996). In comparison, several
components of the HPG axis reportedly activate the HPA. For instance, oestrogen response
elements have been identified in the CRH gene promoter region (Vamvakopoulos & Chrousos
1993). Therefore the HPA has an inhibitory effect on the HPG, whist the HPG appears to stimu-
late HPA activity (Chrousos & Harris 1998; Mastorakos et al. 2006).

The HPG relies on daily signals in CRH and cortisol expression to maintain homeostatic regula-
tion. If, as discussed, chronic stress results in a HPA hypo-function, these daily inhibitory signals
fail to adequately control HPG function, the HPG may itself become hypoactive (Figure 5.3b).
Availability of sex hormones has been associated with fluctuations in pain. Rheumatoid arthritis
tends to improve during pregnancy, during oestrogen replacement therapy and during treatment
with oestrogen-containing oral contraceptives (Van Vollenhoven & McGuire 1994), temporo-
mandibular (LeResche et al. 2003) and experimental (Teepker et al. 2010) pain fluctuate over the
course of the menstrual cycle, and testosterone has been associated with the amelioration of pain in
clinic patients with FM (Dessein et al. 1999). However, the HPG is influenced by a host of factors.
Body weight is an independent regulator of the HPG axis activity. Leptin, an adipose tissue
derived hormone (Bray 1997; Mitchell et al. 2005), primarily modulates appetite and energy
expenditure within hypothalamic nuclei. It also regulates a variety of functions within the HPG
axis, such as the onset of puberty, reproductive capacity and facilitating implantation and
pregnancy (Lado-Abeal & Norman 2002; Margetic et al. 2002) This is because adipose tissue
metabolizes sex-steroids and therefore significantly alters functionality of the reproductive axis by
reducing the levels of circulating sex-steroids (Haffner 2000; Tchernof et al. 2000). Indeed, obes-
ity constitutes as a low-grade inflammatory state (Yudkin 2007) as increased adipose tissue causes
enhanced secretion of pro-inflammatory hormones and cytokines, such as TNF-α and IL-6, into
the circulation. This imposes systemic chronic inflammatory stress on the body (Tsigos et al.
1999) which is further aggravated by severe reductions in cortisol-mediated and subsequent
immune activity.
A further interesting observation is the role of cytokines on HPG function. Pro-inflammatory
cytokines, such as TNFα, stimulate HPA axis function via CRH activation, whilst at the same
time reducing HPG activity, by preventing GnRH secretion and sex-steroid production (Tsigos
et al. 1999). Therefore, prolonged suppression of gonadal function, as a result of abnormal HPA
activity, can result in reduced LH and testosterone concentrations in males and menstrual disor-
ders and amenorrhea in females (Meczekalski et al. 2008; Tsutsumi & Webster 2009).
5.5 Stress and HPGH axis function

5.5.1 Normal function
Growth hormone (GH) is secreted by the anterior pituitary in a pulsatile manner and regulated
by the opposing actions of hypothalamic GH releasing hormone (GHRH), and somatostatin
(SS)—GHRH stimulates while SS inhibits GH release (Figure 5.2). The secretion of GH into the
bloodstream stimulates IGF-I production in many organs, and subsequent elevations in circulat-
ing IGF-I. IGF-1 then feeds back to inhibit the axis hence preventing further secretion of GH
by the pituitary. Experimentally, administration of exogenous IGF-I lowers GH secretion, and
conversely, IGF-I knock-out mice show significant increases in GH secretion (Yakar et al. 2001).
In addition, this pathway is modulated by the regulatory control exerted over SS by key
components of this and other pathways. For instance, serotonin (5-HT) diminishes levels of
circulating SS while GH, insulin-like growth factor-1 (IGF-1) and CRH individually upregulate
SS. Ghrelin, a novel gastric hormone important in appetite regulation has a bi-directional rela-
tionship with SS. Whilst ghrelin induces SS expression, SS suppresses ghrelin production, and
therefore ghrelin-induced activation of components such as CRH, ACTH, cortisol and vaso-
pressin (Unniappan 2010).
Independently of these control mechanisms, concentrations of circulating GH express diur-
nal rhythms, peaking during the late night/early morning and at nadir midday (Uchiyama
et al. 1998). In bone IGF-I acts on the epiphysial plate, leading to longitudinal bone growth
(Kemp 2009; Báez-Saldaña et al. 2009). As illustrated, GH also has direct effects on many organs,
including kidney and cartilage, which can be independent of IGF-I action (Gershberg, 1960;
Olney, 2009).

Interactions between the HPGH and HPA are well documented and occur at both the hypothalamic
and pituitary levels (Figure 5.3a). Peaks in CRH or cortisol result in elevation of GHRH, SS and
ghrelin and conversely, reduced GH secretion (Maclean & Jackson 1988; Rigamonti et al. 2002).
Additionally, studies have shown that prolonged GH secretion could have a variety of physiolog-
ical effects such as muscle weakness (Mukherjee et al. 2004).

Chronic activation of the HPA axis impedes HPGH activity (Figure 5.3b). In fact, HPA hypofunc-
tion due to prolonged stress has been linked to the stress-associated suppression of GH secretion
and consequent desensitization of the pituitary gland. In clinical trials fibromyalgia (FM)—
a chronic pain disorder characterized by chronic widespread pain that includes axial (often low
back) pain (Wolfe et al. 1990)—patients have abnormal HPGH function and basal levels of both
GH and IGF-1, while further studies have implicated GH deficiency with depression and other
chronic illnesses (Cuatrecasas et al. 2007; Griep et al. 1993).
5.6 Does stress system functioning underpin the transition

from acute to chronic LBP?
While there is robust evidence that links physical, psychological and psychosocial stressors with
the chronification of LBP, there is a paucity of data that has examined whether dysfunction of
stress response systems underpin these relationships. There are no methodologically robust stud-
ies available that have directly tested that hypothesis. The data available and discussed below is
‘strongly suggestive’, having directly assessed the relationships between markers of stress system
functioning and chronic LBP or having assessed the relationship between factors known to be
associated with chronicity of acute LBP and stress system functioning.
5.7 Stress system functioning in chronic LBP

Dysfunction of the HPA axis has been associated with chronic LBP in a number of cross-sectional
studies. Compared to healthy pain free individuals those with chronic LBP had lower 24-hour
urinary free cortisol (Lentjes et al. 1997), an exaggerated ACTH response to CRH challenge
(Griep et al. 1998b) although subsequent cortisol release was identical in those with and without
pain (Griep et al. 1998b). While the number of glucocorticoid receptors in circulating mono-
nuclear cells was the same, the binding affinity was lower in those with chronic LBP (Lentjes et al.
1997). These data suggest mild hypocortisolaemia with attenuated feedback resistance similar,
although less pronounced, to that observed in patients with FM (Crofford et al. 1994; Griep et al.
1998a), a chronic pain disorder characterized by chronic widespread pain that includes axial
(often low back) pain (Wolfe et al. 1990). LBP is associated with the onset of chronic widespread
pain and FM (Forseth et al. 1999) and the reported differences in HPA axis functioning may be
explained by quantitative differences in pain extent between the two groups. In addition FM is
associated with higher rates of psychological distress and other factors (Staud 2009) that could
influence HPA function.
Secretion of proinflammatory cytokines including TNF-α (Aoki et al. 2002) and IL-6 (Geiss
et al. 2005) have been related to sciatic pain and chronic LBP (Wang et al. 2008), and IL-6 to post-
operative pain in patients undergoing surgery for disc herniation (Geiss et al. 2005). Other studies
have indirectly suggested an association. Analysis of data from the Third National Health and
Nutrition Examination Survey revealed that among 6814 adults aged 20–39 years 1452 (21.3%,
weighted for non-response) reported a history of allergic conditions (asthma, hay fever, and aller-
gic reactions) that are associated with increased secretion of pro-inflammatory cytokines includ-
ing IL-1β and altered functioning of the HPA axis (Hurwitz & Morgenstern 1999). Exposure to an
allergic reaction was associated with a 53% increase in the odds of reporting LBP (independent of
age, sex, race, marital status and employment status) and a three fold increase in the odds of
reporting LBP and depression (OR=3.2; 95% CI: 1.4, 7.3).
Women are more likely to report chronic pain when compared to men suggesting a role of
female sex hormones in the onset and persistence of symptoms or conversely a protective role of
male sex hormones. However the data exploring the role of sex hormones in chronic LBP is
unclear. In a prospective survey of women first surveyed when 14 years old ‘menstruation’ and
‘pregnancy’ were independent significant predictors of LBP at 38 years old (Harreby et al. 1996).
In a population based study of 11,428 women, hormonal and reproductive factors associated with
increased oestrogen levels (prolonged menstrual cycle, past—though not current—pregnancy,
younger maternal age at first birth, use of oral contraceptives, and use of oestrogens during
menopause) were associated with the presence of chronic LBP (Wijnhoven et al. 2006). This is in
direct contrast to the hypothesized stress-related reduced availability of oestrogens discussed
above. However, in that study hysterectomy was also associated with LBP. Age related declines in
oestrogen levels are associated with osteoporosis and osteoporotic fracture which in turn are
associated with LBP (Nevitt et al. 1998) with the strength of the association increasing with the
number and severity of fractures (O’Neill et al. 2004).
Compared to healthy controls subjects with chronic LBP have poorer overall sleep, increased
insomnia symptoms, and less efficient sleep (O’Donoghue et al. 2009) that will impact on pro-
duction of GH. Serum IGF-I, IGFBP-3, and IGFBP-5 were significantly lower in women with
osteoporotic fractures (Yamaguchi et al. 2006). Markers of HPGH axis function have been associ-
ated with widespread pain disorders including FM (Bennett 1998; Jones et al. 2007) although the
relationship may be explained by higher levels of obesity (McBeth et al. 2009) and the association
with LBP remains unclear. Similarly there is a paucity of data examining the relationship with
androgen levels although among 119 newspaper employees who were classified according to the
presence or absence of neck, shoulder and back pain those with pain had lower levels of serum
DHEA-S and testosterone (Schell et al. 2008). After adjustment for age and gender the latter two
relationships were no longer significant.
Of course these studies have been conducted on subjects with established LBP and although
they suggest a relationship they are unable to establish the temporal relationship between stress
mediators and chronic pain.
5.8 The transition from acute to chronic symptoms

Garofalo and colleagues (Garofalo et al. 2006) recruited 37 patients with acute (<6 months’ dura-
tion) LBP. Using an algorithm developed by the authors (Gatchel et al. 1995) subjects were
classified into ‘low’ and ‘high’ risk of developing chronic LBP. All subjects collected two saliva
samples (one on wakening and the second 20 minutes later) every day for the first 2 weeks after
recruitment. Two measures of cortisol levels were reported: mean cortisol secretions over the first
2 weeks and the difference between wakening and the second sample 20 minutes later. The data
suggested that compared to low risk subjects those at high risk had significantly lower mean
cortisol concentrations1 over the first 2 weeks (mean (SD): 0.392 (0.251) and 0.367 (0.131)
respectively). These differences were apparent in women but not men. Subjects at high risk also
had greater variability in the wakening response when compared to low risk subjects with the
former group tending to have a significantly greater secretion 20 minutes after wakening (0.410
(0.209) versus 0.305 (0.214) respectively). The difference in variability in cortisol secretion was
1 No values for cortisol concentrations are given in the original manuscript

observed in both men and women although the degree of variation was significantly greater in
women. The authors concluded that ‘patients at high risk for chronic pain were found to exhibit
differences in both quantity of cortisol concentrations and variability, relative to low-risk patients’
(Garofalo et al. 2006, p. 173).
At this point it would be prudent to further examine the authors’ classification system. The
classification algorithm was developed on data collected from 324 patients recruited via one occu-
pational health clinic and two orthopaedic clinics (Gatchel et al. 1995). All patients had presented
with pain in the lumber spine that had been present for less than 6 weeks (as compared to the six
month period in the study described above). At baseline subjects completed a number of assess-
ments: a demographic and medical history form, assessment of psychopathology using the
Structured Clinical Interview for DSM-III-R (Spitzer & Williams 1988), and the Million Visual
Analog Scale (Million et al. 1982) an occupational based assessment of pain and associated disa-
bility that includes 15 items and has a total score of between 0 and 150 with higher scores indicat-
ing more pain and associated disability. All subjects were followed up 6 months after initial
assessment and classified into one of two groups: not disabled (currently working or vocational
training or school/retraining, n=274) and disabled (currently disabled and not working because
of the original back injury, n=36). Perhaps unsurprisingly the baseline factors which best
discriminated between these two groups were higher levels of baseline pain and disability and the
presence of an Axis II personality disorder.
This algorithm identifies subjects with LBP who are distressed and who report considerable
pain-related disability. It is perhaps not surprising then that those identified as being at high risk
of chronic symptoms were those with derangements of the HPA axis. The study did not follow
subjects up. It would have been interesting to know, for example, whether baseline levels of
cortisol predicted the chronicity of symptoms, say 6 months later. In a second study from the
same group (Garofalo et al. 2007) among subjects with acute (<6 months’ duration) LBP, lower
cortisol levels were significantly associated with higher reported pain severity. Whether these
results simply reflected greater associated distress was unclear.
A more recent pilot study sought to examine the relationship between pain-related coping
strategies and functioning of the HPA axis (Sudhaus et al. 2009) and to compare the relationships
in subjects with acute to those with chronic LBP. A total of 19 patients with acute and 24 with
chronic non-specific LBP with and without distal radiation were recruited. All subjects collected
five saliva samples 0, 15, 30, 45, and 60 minutes after waking up in the morning over a 2-day
period. The cortisol awakening response was estimated as: (1) the mean value over the 2-day
period for each data collection point and (2) the ‘area under the curve with respect to ground’
(AUCG), an estimation of the total cortisol output which is calculated from repeat measure data
with identical time intervals between measurements (Pruessner et al. 2003). Subjects completed
the Kiel Pain Inventory (KPI) (Hasenbring et al. 1994) from which three categories of pain-
related variables were extracted: ‘endurance coping’ (EC) comprised measures of behavioural
endurance, daily active coping strategies, thought suppression; ‘fear avoidance coping’ (FAC)
included measures of avoidance of social activities, avoidance of physical activities, and helpless-
ness/hopelessness; and ‘nonverbal pain behaviour’ (NPB). Subjects also completed measures of
depression (Beck Depression Inventory (Beck et al. 1961)) and fatigue (General Fatigue Scale of
the Multidimensional Fatigue Inventory (Smets et al. 1995)).
For both acute and chronic LBP, mean cortisol levels followed a typical wakening response
increasing from wakening through the first 30 minutes and decreasing thereafter (45 and 60 minutes
after wakening). However, there were no significant differences between the groups at any time
point nor in the total concentration of cortisol over the study period (AUCG (μg/dl): acute LBP
3.24 ± 0.85 and chronic LBP 3.41 ± 1.55). Sex, body mass index, smoking or pain intensity did not
influence the results. However distinct patterns of association between psychosocial factors and
AUCG cortisol concentrations were observed. Among subjects with acute LBP positive EC strate-
gies (behavioural endurance and daily active coping strategies) were associated with reduced
AUCG cortisol concentrations. Among those with chronic LBP ‘maladaptive’ FAC strategies
(avoidance of social activities, avoidance of physical activity, and helplessness/hopelessness) were
associated with reduce AUCG cortisol concentrations. Importantly, within the chronic LBP group
when compared to those subjects with low levels those with high levels of NPB, depression and
fatigue had significantly less variation in the wakening cortisol response indicating a blunted or
hypofunctioning HPA axis. There were no similar observations among those with acute LBP.
The Swedish Norrtälje Musculoskeletal Intervention Centre study was set up to investigate the
role of workplace risk factors in the onset and maintenance of musculoskeletal disorders includ-
ing LBP (Theorell et al. 2000). A subgroup of subjects who had presented to a ‘caregiver’ (tradi-
tional and alternative) with an episode of acute LBP that had been preceded by 6 months without
consultation for LBP were identified (Theorell et al. 2000). 26 women and 19 men (with 85 and
72 respectively gender- and age-matched control subjects identified through a population regis-
ter) provided two blood samples at 08.00 and 12.00 hours and completed a battery of psycho-
social measures: pain and associated disability, decision latitude, job demand, job strain, job
satisfaction, and social support. While both male and female cases and controls showed a typical
pattern of decreasing cortisol levels from wakening to midday, female cases had a significantly
reduced change in cortisol levels when compared to female controls (proportion change in serum
cortisol: 7.8% vs. −22.7). Similarly among women, cases had significantly higher midday serum
IL-6 levels when compared to controls (pg/ml (95% confidence intervals): 4.4 (3.6–5.4) versus
3.1 (2.6–3.6)). Among men there were no differences between cases and controls on either
biomarker although low decision latitude was associated with high IL-6 and higher levels of social
support were associated with higher morning cortisol levels. In a follow up study serum IL-6 did
not predict persistent pain or subsequent disability at 3 or 6 months post baseline in women.
However, among men high IL-6 was associated with persistent pain although this relationship
was explained by differences in pain intensity and disability (Hasselhorn et al. 2001)2. This is an
interesting observation and suggests that the factors driving the transition from acute to chronic
LBP are psychosocial independent of the underlying stress axis function.
Together these findings suggest qualitative differences between acute and chronic states in rela-
tion to stress axis function. However markers of stress axis dysfunction appear to reflect psycho-
social factors and those relationships differ between men and women: lower CSF cortisol levels
were associated with more intense or disabling pain in men and with higher levels of anxiety and
somatization in men and women (Alaranta et al. 1983), while CSF concentrations of prolactin
correlated positively with levels of anxiety and depression (Alaranta et al. 1983) although much of
the variation in prolactin levels was explained by sex (Hyyppa et al. 1985). The rate of post-
dexamethasone non-suppression in individuals with major depressive disorder was independent
of the presence of chronic LBP (France et al. 1987). In a group of 80 patients attending an inpa-
tient pain management programme who had reported daily back pain for 6 months or more 35
(43.8%) satisfied DSM-III criteria for major depression (France & Krishnan 1985). On adminis-
tration of 1mg dexamethasone 14 (40%) individuals with LBP and major depression failed to
suppress (serum cortisol >5μg/dl) while none of the group with LBP in the absence of major
2 No results for the relationship between cortisol levels and subsequent pain and disability were pre-
sented although the original paper shows that high IL-6 was associated with low cortisol in this group of
subjects.
depression failed to suppress. Among those with LBP and major depression 21 cases were deemed
to have an organic cause (physical examination positive for radiculopathy, abnormal radiograph,
or positive electromyography) and no organic cause could be found for 11 cases. The rate of post-
dexamethasone non-suppression among the latter group was significantly higher (n=7, 63.6%)
when compared to those with an organic pathology (n=7, 29.2%). This observation could be
explained by higher rates of maladaptive coping strategies in those with non-specific back pain of
unknown origin.
Maladaptive coping strategies have been shown to be associated with hypofunctioning of the
HPA axis (Abelson et al. 2008, 2010) and potentially with other associated systems such as the
HPG axis (Kajantie & Phillips 2006) and proinflammatory cytokines. Patients with lumbar disc
herniation and low diurnal cortisol variability had lower levels of physical functioning, lower
perceived possibilities of influencing their pain and higher levels of pain related catastrophizing
when compared to those with higher diurnal cortisol variability (Johansson et al. 2008 ).
Experimental studies have confirmed the relationship between catastrophizing and stress system
functioning. In subjects free of chronic pain the administration of acute painful stimuli (heat,
cold and pressure) was associated with post-test IL-6 but not cortisol levels in those with high
levels of situation specific catastrophizing (i.e. catastrophic thoughts associated with the admin-
istered stimuli) (Edwards et al. 2008). Patients with temporomandibular joint disorder under-
went a series of tests to establish response to pressure and thermal stimuli (Quartana et al. 2010).
Those with high pain catastrophizing scores had a blunted decline in salivary cortisol levels
20 minutes post-test. Interestingly, a similar relationship was observed in a group of individuals
free of pain. These observations in pain free individuals suggest that among those with high levels
of catastrophizing exposure to acute painful stimuli may lead to chronic pain via altered function-
ing of the HPA axis or an increase in pro-inflammatory cytokines. There are a number of putative
mechanisms of action including reduced prefrontal cortical modulation of pain signals among
those with catastrophic cognitions that may inhibit disengaging and suppressing pain signals
(Seminowicz & Davis 2006) or an increased central sensitizaton and sensitivity to stimuli via the
action of proinflammatory cytokines in the long-term activation of spinal cord glia and dorsal
horn neurons (Diers et al. 2007; Gur & Oktayoglu 2008).
5.9 Causal mechanisms

There are numerous methodological problems with the studies described above, many of which
are acknowledged by the authors, including small sample sizes and the challenge of between and
within individual variation in stress biomarkers. But perhaps the most fundamental problem has
been the inability of these studies to robustly assess causality. The main question remains unan-
swered: among individuals with acute LBP, does functioning of the HPA and associated axes
predict who will go on to develop chronic symptoms? When considering the relationship between
a risk factor (stress system function) and outcome (the chronification of acute LBP) we should be
mindful that while these factors may be associated with one another an observed association does
not necessarily imply causality. In Figure 5.4, a number of possible explanations for the observed
associations are outlined: in (a) acute LBP could be associated with chronic LBP directly (dashed
line) or via dysfunction of stress response systems (solid lines). Stress system derangement may be
pre-existing or may be associated with the high levels of stress and other factors associated with
acute LBP that are known to influence chronicity. In (b) both acute LBP and chronic LBP are a
consequence of stress system derangement but are independent of each other and not causally
related. In (c) acute LBP is causally associated with chronic LBP and stress system derangement is
a consequence of chronic LBP. While the true nature of the role of stress system functioning in
(a) Stress system

derangement
ALBP CLBP
ALBP
(b) Stress system
derangement
CLBP
(c) Stress system

ALBP CLBP
derangement
Fig. 5.4 Pathways of association. In (a) ALBP could be associated with CLBP directly (dashed line)
or via dysfunction of stress response systems (solid lines). Stress system derangement may be
pre-existing or may be associated with the high levels of stress and other factors associated with
ALBP that are known to influence chronicity. In (b) both ALBP and CLBP are a consequence of
stress system derangement but are independent of each other and not causally related. In (c) ALBP
is causally associated with CLBP and stress system derangement is a consequence of CLBP. ALBP,
acute low back pain; CLBP, chronic low back pain.
the transition from acute to chronic LBP awaits rigorous investigation, a small number of pro-
spective studies have investigated the role of altered functioning of the HPA and other axes in the
onset and/or perpetuation of more general musculoskeletal pain disorders.
In a two-year prospective study (Wrosch et al. 2008) 184 individuals were asked to report their
experience of 12 physical symptoms that included LBP over a 2-day period and to collect five saliva
samples (30 minutes after awakening, 14.00 hrs, 16.00 hrs, and before bedtime) on three consecu-
tive days. Baseline cortisol levels predicted an increase in physical symptoms at follow-up but only
among those with high levels of negative affect and low sleep efficiency at baseline. Among par-
ticipants with neck, shoulder or back pain lower levels of serum GH, but not IGF-1, was associated
with a worsening of pain 12 months later (Schell et al. 2008). There were no associations with
cortisol, DHEA-S, oestradiol, testosterone, or cytokine levels. In individuals free of widespread
pain but who were psychologically distressed markers of HPA axis dysfunction predicted the onset
of symptoms 12 months later (McBeth et al. 2007). This relationship was not explained by psycho-
logical factors although there was no measure of catastrophizing included in the study. Jennifer
Glass has elegantly demonstrated the predictive ability of brief exercise cessation in relation to pain
onset among healthy people but only among those with a hypofunctioning HPA axis (Glass et al.
2004). Among women who were followed up over a nine year period fluctuations in levels of
oestrodial were associated with the development of aches and pains (Freeman et al. 2007).
5.10 Conclusions
It is unclear whether dysfunction of stress response systems is associated with the transition from
acute to chronic LBP. To answer that question a group of individuals with acute LBP need to be
identified. Within that group measures of HPA (and other) axis function and other factors
known to influence stress system function should be rigorously assessed. All subjects should be
followed-up over a reasonable period of time (say, 12 months) to identify those who do and those
who do not develop chronic LBP. Only then will the role, if any, of HPA axis function in the
chronification of acute LBP become clear. However, the aetiology of LBP is multifactorial.
Bringing together multiple levels of causality in a single study is challenging: there is a pay-off
between study size and detail in exposure and phenotype measurement. Using novel methodolo-
gies (e.g. case-crossover designs, two-stage sampling procedures) and novel statistical methods
(e.g. multilevel modelling) will enhance our investigations. We have begun to explain the transi-
tion from acute to chronic LBP and although we still need to identify other slices of the ‘causal
pie’ the usefulness of multilevel models in informing interventions across all levels of causality are
apparent. In intervention studies of chronic LBP the message is clear; one size does not fit all. The
hurdle to developing effective interventions will not be in our conceptualization of the problem
to be addressed or how to address it, but in the practicalities of doing so.
Acknowledgements
John McBeth would like to thank Sarah Parsons for her invaluable help with literature searching,
retrieval of manuscripts and, together with Mary Ingram, help with references.
References
Abelson, J. L., Khan, S., Liberzon, I., Erickson, T. M., & Young, E. A. (2008). Effects of perceived control
and cognitive coping on endocrine stress responses to pharmacological activation. Biol Psychiatry,
64(8), 701–7.
Abelson, J. L., Khan, S., Young, E. A., & Liberzon, I. (2010). Cognitive modulation of endocrine responses
to CRH stimulation in healthy subjects. Psychoneuroendocrinology, 35(3), 451–9.
Alaranta, H., Hu Báez-Saldaña rme, M., Lahtela, K., & Hyyppa, M. T. (1983). Prolactin and cortisol in
cerebrospinal fluid: Sex-related associations with clinical and psychological characteristics of patients
with low back pain. Psychoneuroendocrinology, 8(3), 333–41.
Aoki, Y., Rydevik, B., Kikuchi, S., & Olmarker, K. (2002). Local application of disc-related cytokines on
spinal nerve roots. Spin, 27(15), 1614–17.
Báez-Saldaña, A., Camacho-Arroyo, I., Espinosa-Aguirre, J.J., et al. (2009). Biotin deficiency and biotin
excess: effects on the female reproductive system. Steroids, 74(10–11), 863–9.
Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for measuring
depression. Arch Gen Psychiatry, 4, 561–71.
Bennett, R. M. (1998). Disordered growth hormone secretion in fibromyalgia: a review of recent findings
and a hypothesized etiology. Zeitschrift fur Rheumatologie, 57, 72–6.
Bentley, G. E., Kriegsfeld, L. J., Osugi, T., et al. (2006). Interactions of gonadotropin-releasing hormone
(GnRH) and gonadotropin-inhibitory hormone (GnIH) in birds and mammals. J Exp Zool A Comp
Exp Biol, 305(9), 807–14.
Bentley, G. E., Ubuka, T., McGuire, N. L., et al. (2008). Gonadotropin-inhibitory hormone and its receptor
in the avian reproductive system. Gen Comp Endocrinol, 156(1), 34–43.
Bray, G. A. (1997). Obesity and reproduction. Hum Reprod, 12(Suppl 1), 26–32.
Calisi, R. M., Rizzo, N. O., & Bentley, G. E. (2008). Seasonal differences in hypothalamic EGR-1 and GnIH
expression following capture-handling stress in house sparrows (Passer domesticus). Gen Comp
Endocrinol, 157(3), 283–7.
Chapman, C. R., Tuckett, R. P., & Song, C. W. (2008). Pain and stress in a systems perspective: reciprocal
neural, endocrine, and immune interactions. J Pain, 9(2), 122–45.
Chiodini, I., Francucci, C. M., & Scillitani, A. (2008). Densitometry in glucocorticoid-induced
osteoporosis. J Endocrinol Invest, 31(7 Suppl), 33–7.
Chrousos, G. P. & Harris, A. G. (1998). Hypothalamic-pituitary-adrenal axis suppression and inhaled
corticosteroid therapy. 1. General principles. Neuroimmunomodulation, 5(6), 277–87.
Crofford, L. J., Pillemer, S. R., & Kalogeras, K. T. (1994). Hypothalamic-pituitary-adrenal axis Pertubations
in Patients with Fibromyalgia. Arthritis Rheum, 37, 1583–92.
Cuatrecasas, G., Riudavets, C., Guell, M. A., & Nadal, A. (2007). Growth hormone as concomitant
treatment in severe fibromyalgia associated with low IGF-1 serum levels. A pilot study. BMC
Musculoskelet Disord, 8, p. 119.
Deary, V., Chalder, T., & Sharpe, M. (2007). The cognitive behavioural model of medically unexplained
symptoms: a theoretical and empirical review. Clin Psychol Rev, 27(7), 781–97.
Dessein, P. H., Shipton, E. A., Joffe, B. I., Hadebe, D. P., Stanwix, A. E., & Van der Merwe, B. A. (1999).
Hyposecretion of adrenal androgens and the relation of serum adrenal steroids, serotonin and insulin-
like growth factor-1 to clinical features in women with fibromyalgia. Pain, 83(2), 313–319.
Diers, M., Koeppe, C., Diesch, E., et al. (2007). Central processing of acute muscle pain in chronic low back
pain patients: an EEG mapping study. J Clin Neurophysiol, 24(1), 76–83.
Edwards, R. R., Kronfli, T., Haythornthwaite, J. A., Smith, M. T., McGuire, L., & Page, G. G. (2008).
Association of catastrophizing with interleukin-6 responses to acute pain. Pain, 140(1), 135–44.
Faravelli, C., Amedei, S. G., Rotella, F., et al. (2010). Childhood traumata, Dexamethasone Suppression
Test and psychiatric symptoms: a trans-diagnostic approach. Psychol Med, 40, 1–12.
ffrench-Mullen, J. M. (1995). Cortisol inhibition of calcium currents in guinea pig hippocampal CA1
neurons via G-protein-coupled activation of protein kinase C. J Neurosci, 15(1 Pt 2), 903–11.
Forseth, K. O., Husby, G., Gran, J. T., & Forre, O. (1999). Prognostic factors for the development of
fibromyalgia in women with self-reported musculoskeletal pain. A prospective study. Journal of
Rheumatology, 26(11), 2458–67.
France, R. D. & Krishnan, K. R. (1985). The dexamethasone suppression test as a biologic marker of
depression in chronic pain. Pain, 21(1), 49–55.
France, R. D., Krishnan, K. R., Trainor, M., & Pelton, S. (1987). Chronic pain and depression. IV. DST as a
discriminator between chronic pain and depression. Pain, 28(1), 39–44.
Freeman, E. W., Sammel, M. D., Lin, H., et al. (2007). Symptoms associated with menopausal transition
and reproductive hormones in midlife women. Obstet Gynecol, 110(2 Pt 1), 230–40.
Fries, E., Hesse, J., Hellhammer, J., & Hellhammer, D. H. (2005). A new view on hypocortisolism.
Psychoneuroendocrinology, 30(10), 1010–16.
Garofalo, J. P., Robinson, R. C., & Gatchel, R. J. (2006). Hypothalamic-pituitary-adrenocortical axis
dysregulation in acute temporomandibular disorder and low back pain: A marker for chronicity?
J Applied Biobehav Res, 11(3–4), 166–78.
Garofalo, J. P., Robinson, R. C., Gatchel, R. J., & Wang, Z. (2007). A pain severity-hypothalamic-
pituitary-adrenocortical axis interaction: The effects on pain pathways. Applied Biobehav Res,, 12(1),
35–42.
Gatchel, R. J., Polatin, P. B., & Kinney, R. K. (1995). Predicting outcome of chronic back pain using clinical
predictors of psychopathology: A prospective analysis. Health Psychol, 14, 415–20.
Geiss, A., Rohleder, N., Kirschbaum, C., Steinbach, K., Bauer, H. W., & Anton, F. (2005). Predicting the
failure of disc surgery by a hypofunctional HPA axis: evidence from a prospective study on patients
undergoing disc surgery. Pain, 114(1–2), 104–17.
Gershberg, H. (1960). Metabolic and renotropic effects of human growth hormone in disease. J Clin
Endocrinol Metab, 20, 1107–19.
Glass, J. M., Lyden, A. K., Petzke, F., et al. (2004). The effect of brief exercise cessation on pain, fatigue,
and mood symptom development in healthy, fit individuals. Journal of Psychosomatic Research, 57(4),
391–8.
Griep, E. N., Boersma, J. W., & de Kloet, E. R. (1993). Altered reactivity of the hypothalamic-pituitary-
adrenal axis in the primary fibromyalgia syndrome. Journal of Rheumatology, 20(3), 469–74.
Griep, E. N., Boersma, J. W., Lentjes, E. G., Prins, A. P., Van der Korst, J. K., & de Kloet, E. R. (1998a).
Function of the hypothalamic-pituitary-adrenal axis in patients with fibromyalgia and low back pain.
Journal of Rheumatology, 25(7), 1374–81.
Griep, E. N., Boersma, J. W., Lentjes, E. G. W. M., Prins, A. P. A., Van der Korst, J. K., & de Kloet, E. R.
(1998b). Function of the hypothalamic-pituitary-Adrenal Axis in patients with fibromyalgia and low
back pain. J Rheumatol, 25, 1374–81.
Gur, A. & Oktayoglu, P. (2008). Status of immune mediators in fibromyalgia. Curr Pain Headache Rep,
12(3), 175–81.
Haffner, S. M. (2000). Sex hormones, obesity, fat distribution, type 2 diabetes and insulin resistance:
epidemiological and clinical correlation. Int J Obes Relat Metab Disord, 24(Suppl 2), S56–58.
Harbuz, M. (2002a). Neuroendocrine function and chronic inflammatory stress. Exp Physiol, 87(5),
519–25.
Harbuz, M. (2002b). Neuroendocrinology of autoimmunity. Int Rev Neurobiol, 52, 133–61.
Harreby, M., Kjer, J., Hesselsoe, G., & Neergaard, K. (1996). Epidemiological aspects and risk factors for
low back pain in 38-year- old men and women: a 25-year prospective cohort study of 640 school
children. European Spine Journal, 5(5), 312–18.
Hasenbring, M., Marienfeld, G., Kuhlendahl, D., & Soyka, D. (1994). Risk factors of chronicity in lumbar
disc patients. A prospective investigation of biologic, psychologic, and social predictors of therapy
outcome. Spine, 19(24), 2759–65.
Hasselhorn, H. M., Theorell, T., & Vingard, E. (2001). Endocrine and immunologic parameters indicative
of 6-month prognosis after the onset of low back pain or neck/shoulder pain. Spine, 26(3, E24–29.
Henckens, M. J., Hermans, E. J., Pu, Z., Joels, M., & Fernandez, G. (2009). Stressed memories: how acute
stress affects memory formation in humans. J Neurosci, 29(32), 10111–19.
Hurwitz, E. L. & Morgenstern, H. (1999). Cross-sectional associations of asthma, hay fever, and other
allergies with major depression and low-back pain among adults aged 20–39 years in the United States.
American Journal of Epidemiology, 150(10), 1107–16.
Hyyppa, M. T., Alaranta, H., Hurme, M., & Lahtela, K. (1985). Prolactin and cortisol responses to the
experience of low back pain. Pain, 23(3), 231–42.
IASP (2002). Epidemiology of Pain a report of the task force on epidemiology of the International Association of
Pain. Seattle, WA: IASP Press.
Johansson, A. C., Gunnarsson, L. G., Linton, S. J., Bergkvist, L., Stridsbergf, M., Nilsson, O., &
Cornefjord, M. (2008). Pain, disability and coping reflected in the diurnal cortisol variability in
patients scheduled for lumbar disc surgery. European Journal of Pain, 12(5), 633–40.
Jones, K. D., Deodhar, P., Lorentzen, A., Bennett, R. M., & Deodhar, A. A. (2007). Growth hormone
perturbations in fibromyalgia: a review. Seminars in Arthritis and Rheumatism, 36(6), 357–79.
Kajantie, E. & Phillips, D. I. (2006). The effects of sex and hormonal status on the physiological response to
acute psychosocial stress. Psychoneuroendocrinology, 31(2), 151–78.
Kemp, S. F. (2009). Insulin-like growth factor-I deficiency in children with growth hormone insensitivity:
current and future treatment options. BioDrugs, 23(3), 155–63.
Kriegsfeld, L. J., Mei, D. F., Bentley, G. E., et al. (2006). Identification and characterization of a
gonadotropin-inhibitory system in the brains of mammals. Proc Natl Acad Sci U S A, 103(7), 2410–15.
Kudielka, B. M., Schommer, N. C., Hellhammer, D. H., & Kirschbaum, C. (2004). Acute HPA axis
responses, heart rate, and mood changes to psychosocial stress (TSST) in humans at different times
of day. Psychoneuroendocrinology, 29(8), 983–92.
Kumsta, R., Entringer, S., Koper, J. W., van Rossum, E. F., Hellhammer, D. H., & Wust, S. (2010).
Working memory performance is associated with common glucocorticoid receptor gene
polymorphisms. Neuropsychobiology, 61(1), 49–56.
Lado-Abeal, J. & Norman, R. L. (2002). Leptin and reproductive function in males. Semin Reprod Med,
20(2), 145–51.
Lentjes, E. G. W. M., Griep, E. N., Boersma, J. W., Romijn, F. P. T. H., & de Kloet, E. R. (1997).
Glucocorticoid receptors, fibromyalgia and low back pain. Psychoneuroendocrinology, 22(8), 603–14.
LeResche, L., Mancl, L., Sherman, J. J., Gandara, B., & Dworkin, S. F. (2003). Changes in
temporomandibular pain and other symptoms across the menstrual cycle. Pain, 106(3), 253–61.
Lightman, S. L., Windle, R. J., Julian, M. D., et al. (2000). Significance of pulsatility in the HPA axis.
Novartis Found Symp, 227, 244–57.
Maclean, D. B. & Jackson, I. M. (1988). Molecular biology and regulation of the hypothalamic hormones.
Baillieres Clin Endocrinol Metab, 2(4), 835–68.
Maier, S. F. & Watkins, L. R. (1998). Cytokines for psychologists: implications of bidirectional immune-
to-brain communication for understanding behavior, mood, and cognition. Psychol Rev, 105(1),
83–107.
Margetic, S., Gazzola, C., Pegg, G. G., & Hill, R. A. (2002). Leptin: a review of its peripheral actions and
interactions. Int J Obes Relat Metab Disord, 26(11), 1407–33.
Mastorakos, G., Pavlatou, M. G., & Mizamtsidi, M. (2006). The hypothalamic-pituitary-adrenal and the
hypothalamic- pituitary-gonadal axes interplay. Pediatr Endocrinol Rev, 3(Suppl 1), 172–81.
McBeth, J., Silman, A. J., Gupta, A., et al. (2007). Moderation of psychosocial risk factors through
dysfunction of the hypothalamic-pituitary-adrenal stress axis in the onset of chronic widespread
musculoskeletal pain - Findings of a population-based prospective cohort study. Arthritis Rheum,
56(1), 360–71.
McBeth, J., Symmons, D. P., Silman, A. J., et al. (2009). Musculoskeletal pain is associated with a long-term
increased risk of cancer and cardiovascular-related mortality. Rheumatology, 48(1), 74–77.
McEwen, B. S. (2003). Mood disorders and allostatic load. Biological Psychiatry, 54(3), 200–207.
Meczekalski, B., Podfigurna-Stopa, A., Warenik-Szymankiewicz, A., & Genazzani, A. R. (2008). Functional
hypothalamic amenorrhea: current view on neuroendocrine aberrations. Gynecol Endocrinol, 24(1), 4–11.
Meeus, M., Nijs, J., Van de, W. N., Toeback, L., & Truijen, S. (2008). Diffuse noxious inhibitory control is
delayed in chronic fatigue syndrome: an experimental study. Pain, 139(2), 439–48.
Million, R., Hall, W., Haavik Dilsen, K., Baker, R., & Jayson, M. I. V. (1982). 1981 Volvo Award in Clinical
Science - Assessment of the progress of the Back-pain patient. Spine, 7, 204–12.
Mitchell, M., Armstrong, D. T., Robker, R. L., & Norman, R. J. (2005). Adipokines: implications for female
fertility and obesity. Reproduction, 130(5), 583–97.
Morelli, V., Masserini, B., Salcuni, A. S., et al. (2010). Subclinical hypercortisolism: correlation between
biochemical diagnostic criteria and clinical aspects. Clin Endocrinol (Oxf).
Mukherjee, A., Murray, R. D., & Shalet, S. M. (2004). Impact of growth hormone status on body
composition and the skeleton. Horm Res, 62(Suppl 3), 35–41.
Muller, M., Holsboer, F., & Keck, M. E. (2002). Genetic modification of corticosteroid receptor signalling:
novel insights into pathophysiology and treatment strategies of human affective disorders.
Neuropeptides, 36(2–3), 117–31.
Murakami, M., Matsuzaki, T., Iwasa, T., Yasui, T., Irahara, M., Osugi, T., & Tsutsui, K. (2008).
Hypophysiotropic role of RFamide-related peptide-3 in the inhibition of LH secretion in female rats.
J Endocrinol, 199(1), 105–12.
Mussig, K., Remer, T., & Maser-Gluth, C. (2010). Brief review: Glucocorticoid excretion in obesity.
J Steroid Biochem Mol Biol, 121(3–5), 589–93.
Neeck, G. & Crofford, L. J. (2000). Neuroendocrine perturbations in fibromyalgia and chronic fatigue
syndrome. Rheum Dis Clinics North Am, 26(4), 989–1002.
Nevitt, M. C., Ettinger, B., Black, D. M., et al. (1998). The association of radiographically detected
vertebral fractures with back pain and function: A prospective study. Annals of Internal Medicine, 128,
793–800.
O’Donoghue, G. M., Fox, N., Heneghan, C., & Hurley, D. A. (2009). Objective and subjective assessment
of sleep in chronic low back pain patients compared with healthy age and gender matched controls:
a pilot study. BMC Musculoskelet Disord, 10, 122.
O’Neill, T. W., Cockerill, W., Matthis, C., et al. (2004). Back pain, disability, and radiographic vertebral
fracture in European women: a prospective study. Osteoporosis International, 15(9), 760–5.
Olff, M., Brosschot, J. F., Godaert, G., et al. (1995). Modulatory effects of defense and coping on
stress-induced changes in endocrine and immune parameters. Int J Behav Med, 2(2), 85–103.
Olney, R. C. (2009). Mechanisms of impaired growth: effect of steroids on bone and cartilage. Horm Res,
72(S1), 30–5.
Osugi, T., Ukena, K., Bentley, G. E., et al. (2004). Gonadotropin-inhibitory hormone in Gambel’s
white-crowned sparrow (Zonotrichia leucophrys gambelii): cDNA identification, transcript
localization and functional effects in laboratory and field experiments. J Endocrinol, 182(1), 33–42.
Papadimitriou, A. & Priftis, K. N. (2009). Regulation of the hypothalamic-pituitary-adrenal axis.
Neuroimmunomodulation, 16(5), 265–71.
Patel, P. D., Katz, M., Karssen, A. M., & Lyons, D. M. (2008). Stress-induced changes in corticosteroid
receptor expression in primate hippocampus and prefrontal cortex. Psychoneuroendocrinology, 33(3),
360–7.
Pham, K., Nacher, J., Hof, P. R., & McEwen, B. S. (2003). Repeated restraint stress suppresses neurogenesis
and induces biphasic PSA-NCAM expression in the adult rat dentate gyrus. Eur J Neurosci, 17(4), 879–86.
Pruessner, J. C., Kirschbaum, C., Meinlschmid, G., & Hellhammer, D. H. (2003). Two formulas for
computation of the area under the curve represent measures of total hormone concentration versus
time-dependent change. Psychoneuroendocrinology, 28(7), 916–31.
Quartana, P. J., Buenaver, L. F., Edwards, R. R., Klick, B., Haythornthwaite, J. A., & Smith, M. T. (2010).
Pain catastrophizing and salivary cortisol responses to laboratory pain testing in temporomandibular
disorder and healthy participants. J Pain, 11(2), 186–94.
Rabin, D. S., Johnson, E. O., Brandon, D. D., Liapi, C., & Chrousos, G. P. (1990). Glucocorticoids inhibit
estradiol-mediated uterine growth: possible role of the uterine estradiol receptor. Biol Reprod, 42(1),
74–80.
Reul, J. M., Bilang-Bleuel, A., Droste, S., Linthorst, A. C., Holsboer, F., & Gesing, A. (2000). New mode of
hypothalamic-pituitary-adrenocortical axis regulation: significance for stress-related disorders.
Zeitschrift fur Rheumatologie, 59(Suppl 2), II/22–II/25.
Rigamonti, A. E., Bonomo, S. M., Cella, S. G., & Muller, E. E. (2002). GH and cortisol rebound rise during
and following a somatostatin infusion: studies in dogs with the use of a GH-releasing peptide.
J Endocrinol, 174(3), 387–94.
Rivier, J., Jiang, G. C., Lahrichi, S. L., et al. (1996). Dose relationship between GnRH antagonists and
pituitary suppression. Hum Reprod, 11(Suppl 3), 133–47.
Schell, E., Theorell, T., Hasson, D., Arnetz, B., & Saraste, H. (2008). Stress biomarkers’ associations to pain
in the neck, shoulder and back in healthy media workers: 12-month prospective follow-up. Eur Spine J,
17(3), 393–405.
Seminowicz, D. A. & Davis, K. D. (2006). Cortical responses to pain in healthy individuals depends on pain
catastrophizing. Pain, 120(3), 297–306.
Smets, E. M. A., Garssen, B., Bonke, B., & de Haes, J. C. (1995). The Multidimensional Fatigue Inventory
(MFI) psychometric qualities of an instrument to assess fatigue. J Psychosomat Res, 39(3), 315–25.
Spitzer, R. L. & Williams, J. B. (1988). Revised diagnostic criteria and a new structured interview for
diagnosing anxiety disorders. J Psychiatric Res, 22(Suppl 1), 55–85.
Staud, R. (2009). Chronic widespread pain and fibromyalgia: two sides of the same coin?. Curr Rheumatol
Rep, 11(6), 433–36.
Steckler, T., Holsboer, F., & Reul, J. M. (1999). Glucocorticoids and depression. Baillieres Best Pract Res
Clin Endocrinol Metab, 13(4), 597–614.
Stein-Behrens, B. A. & Sapolsky, R. M. (1992). Stress, glucocorticoids, and aging. Aging, 4(3), 197–210.
Sudhaus, S., Fricke, B., Stachon, A., Schneider, S., Klein, H., von During, M., & Hasenbring, M. (2009).
Salivary cortisol and psychological mechanisms in patients with acute versus chronic low back pain.
Psychoneuroendocrinology, 34(4), 513–22.
Tchernof, A., Poehlman, E. T., & Despres, J. P. (2000). Body fat distribution, the menopause transition,
and hormone replacement therapy. Diabetes Metab, 26(1), 12–20.
Teepker, M., Peters, M., Vedder, H., Schepelmann, K., & Lautenbacher, S. (2010). Menstrual Variation in
Experimental Pain: Correlation with Gonadal Hormones. Neuropsychobiology, 61(3), 131–40.
Theorell, T., Hasselhorn, H. M., Vingard, E., Andersson, B., & MUSIC-Norrtalje Study Group (2000).
Interleukin 6 and cortisol in acute musculoskeletal disorders: Results from a case-referent study in
Sweden. Stress Medicine, 16(1), 27–35.
Tsigos, C., Papanicolaou, D. A., Kyrou, I., Raptis, S. A., & Chrousos, G. P. (1999). Dose-dependent effects
of recombinant human interleukin-6 on the pituitary-testicular axis. J Interferon Cytokine Res, 19(11),
1271–76.
Tsutsumi, R. & Webster, N. J. (2009). GnRH pulsatility, the pituitary response and reproductive
dysfunction. Endocr J, 56(6), 729–37.
Uchiyama, M., Ishibashi, K., Enomoto, T., et al. (1998). Twenty-four hour profiles of four hormones
under constant routine. Psychiatry Clin Neurosci, 52(2), 241–43.
Unniappan, S. (2010). Ghrelin: An emerging player in the regulation of reproduction in non-mammalian
vertebrates. Gen Comp Endocrinol, 167(3), 340–3.
Ursin, H. & Olff, M. (1993). Psychobiology of coping and defence strategies. Neuropsychobiology, 28(1–2),
66–71.
Vadakkadath, M. S. & Atwood, C. S. (2005). The role of hypothalamic-pituitary-gonadal hormones in the
normal structure and functioning of the brain. Cell Mol Life Sci, 62(3), 257–70.
Vamvakopoulos, N. C. & Chrousos, G. P. (1993). Evidence of direct estrogenic regulation of human
corticotropin-releasing hormone gene expression. Potential implications for the sexual dimophism of
the stress response and immune/inflammatory reaction. J Clin Invest, 92(4), 1896–1902.
Van Praag, H. M. (2004). The cognitive paradox in posttraumatic stress disorder: a hypothesis. Prog
Neuropsychopharmacol Biol Psychiatry, 28(6), 923–35.
Van Vollenhoven, R. F. & McGuire, J. L. (1994). Estrogen, progesterone, and testosterone: can they be used
to treat autoimmune diseases? Cleve Clin J Med, 61(4), 276–84.
Vicennati, V., Pasqui, F., Cavazza, C., Pagotto, U., & Pasquali, R. (2009). Stress-related development of
obesity and cortisol in women. Obesity, 17(9), 1678–83.
Wang, H., Schiltenwolf, M., & Buchner, M. (2008). The role of TNF-alpha in patients with chronic low
back pain-a prospective comparative longitudinal study. Clin J Pain, 24(3), 273–8.
Wijnhoven, H. A., de Vet, H. C., Smit, H. A., & Picavet, H. S. (2006). Hormonal and reproductive factors
are associated with chronic low back pain and chronic upper extremity pain in women—the MORGEN
study. Spine, 31(13), 1496–502.
Wolfe, F. (1990). Fibromyalgia. Rheum Dis Clinics North Am, 16, 681–98.
Wolfe, F., Smythe, H. A., Yunus, M. B., et al. (1990). The American College of Rheumatology 1990 Criteria
for the Classification of Fibromyalgia. Report of the Multicenter Criteria Committee. Arthritis and
Rheumatism, 33(2), 160–72.
Wrosch, C., Miller, G. E., Lupien, S., & Pruessner, J. C. (2008). Diurnal cortisol secretion and 2-year
changes in older adults’ physical symptoms: the moderating roles of negative affect and sleep. Health
Psychology, 27(6), 685–93.
Yakar, S., Liu, J. L., Fernandez, A. M., et al. (2001). Liver-specific igf-1 gene deletion leads to muscle insulin
insensitivity. Diabetes, 50(5), 1110–18.
Yamaguchi, T., Kanatani, M., Yamauchi, M., et al. (2006). Serum levels of insulin-like growth factor (IGF);
IGF-binding proteins-3, -4, and -5; and their relationships to bone mineral density and the risk of
vertebral fractures in postmenopausal women. Calcified Tissue International, 78(1), 18–24.
Yin, H., Ukena, K., Ubuka, T., & Tsutsui, K. (2005). A novel G protein-coupled receptor for gonadotropin-
inhibitory hormone in the Japanese quail (Coturnix japonica): identification, expression and binding
activity. J Endocrinol, 184(1), 257–66.
Young, E. A. (1995). Normal glucocorticoid fast feedback following chronic 50% corticosterone pellet
treatment. Psychoneuroendocrinology, 20(7), 771–84.
Young, E. A. & Vazquez, D. (1996). Hypercortisolemia, hippocampal glucocorticoid receptors, and fast
feedback. Mol Psychiatry, 1(2), 149–59.
Yudkin, J. S. (2007). Inflammation, obesity, and the metabolic syndrome. Horm Metab Res, 39(10), 707–9.
Chapter 6
Central Imaging of Pain and the

Process of Chronicity
Sandra Kamping and Herta Flor
6.1 Introduction
Neuroimaging methods were introduced to clinical and experimental pain research only a few
decades ago but have had a major impact. In this chapter we will focus on neuroimaging in
humans and give a short overview of brain areas involved in the perception and evaluation of
pain. We will then discuss functional and structural changes found in patients with chronic back
pain (CBP) and will specifically emphasize the role of emotional learning and memory as well
as cognitive factors in the chronicity process. This chapter will consider methods ranging from
electroencephalography (EEG) and magnetoencephalography (MEG) to functional and struc-
tural magnetic resonance imaging (MRI), including magnetic resonance spectroscopy and posi-
tron emission tomography (PET). We do not aim to provide an exhaustive review but will
highlight key findings and discuss the impact they had on our understanding of CBP and factors
that lead to chronicity.
6.2 The pain matrix: a short introduction

Brain regions involved in the processing of acute pain have been identified by applying nocicep-
tive stimuli (e.g. heat, electric stimulation, pressure) mostly to the skin of a person. Several key
brain regions have been identified, among them the primary (SI) and secondary (SII) somatosen-
sory cortex, the insula, the anterior cingulate cortex (ACC), the prefrontal cortex (PFC), as well
as the thalamus (for reviews see Apkarian et al. 2005; Tracey & Mantyh 2007; see also Figure 6.1).
Depending on the task of the study or the particular the set of circumstances for the individual,
other brain regions such as temporal and parietal areas, the basal ganglia, the cerebellum, the
amygdala, and hippocampus are also active.
Traditionally pain has been separated into three components: (1) sensory-discriminative,
(2) motivational-affective, and (3) cognitive-evaluative (Melzack & Casey 1968). Activation in SI
has been associated with the sensory-discriminative aspect of pain (Bushnell et al. 1999), and
most imaging studies showed a robust activation of SI, depending on the type of stimulation and
the imaging procedure (e.g. EEG or functional magnetic resonance imaging (fMRI)) that was
used. Unfortunately, there are almost as many studies that show activation in SI as there are stud-
ies that show no activation (Bingel et al. 2003; Bornhovd et al. 2002; Coghill et al. 1999; Maihofner
et al. 2002). It has been suggested that the crucial variable for activation in SI is the amount of
spatial and temporal summation (Peyron, Laurent, & Garcia-Larrea 2000).
The secondary somatosensory cortex (SII) plays a role in the sensory analysis of pain, seems to
encode the ‘painfulness’ of the stimulation (e.g. Ringler et al. 2003), is involved in the temporal
encoding of the pain experience (Chen et al. 2002) and also contributes to tactile and pain-related
M1
S1
(a) (b) 1 2 3
SMA
PPC
2.
ACC PCC
Insula S2
PF
BG Thalamus
Amyg HT
PAG
PB
1. 3.
Fig. 6.1 (Also see Colour Plate 3) Cortical and subcortical regions involved in pain perception. The
locations of the regions are superimposed onto an example MRI. The areas primarily involved in
pain perception are the primary somatosensory cortex (S1, red), the secondary somatosensory
cortex (S2, orange), the anterior cingulate cortex (ACC, green), the insula (light blue), the thalamus
(yellow), as well as the basal ganglia (BG, pink), the prefrontal cortex (PF; purple) and the primary
motor cortex (M1, blue). Other regions are the supplementary motor area (SMA), the posterior
cingulate cortex (PCC), the amygdale (AMYG), the parabrachial nuclei (PB) and the periaqueductal
grey (PAG). Reprinted from Progress in Neurobiology, 87(2), A. V. Apkarian, M. N. Baliki, and
P. Y. Geha, Towards a theory of chronic pain, pp. 81–97, Copyright (2009), with permission
from Elsevier.
memory (Gallace & Spence 2009; Harris, Harris, & Diamond 2001). It seems to be activated via a
direct pathway from the thalamus (Ploner, Freund, & Schnitzler 1999) and receives projections
via SI. Activation in this region is more frequently and robustly observed than in SI (Ferretti et al.
2003, Peyron, Laurent, & Garcia-Larrea 2000; Treede et al. 2000).
One area that has been consistently activated in imaging studies of pain is the insula. It seems
to have an important role in nociceptive processing and can be divided into a posterior and an
anterior part. The posterior insula is activated, predominantly contralateral, in response to noci-
ceptive stimuli (Bingel et al. 2003; Brooks et al. 2002) and also encodes temperature (Craig et al.
2000). Activation of the anterior insula is also not specific to pain. It has been described in
response to tactile (Baron et al. 1999; Iadarola et al. 1998), vibratory (Coghill et al. 1994), and
olfacto-gustatory (Faurion et al. 1999; Small et al. 2004) stimulation. Numerous studies show the
involvement of the insula in processes such as memory functions, emotional responses, visceral
sensory and motor integration, and self-awareness as well as the processing of painful stimuli.
It has also been discussed as the site of the brain that contains the anatomical substrate for
self-awareness (Craig 2002, 2009).
The ACC encodes, among others, the affective component of pain (Craig et al. 1996; Rainville
et al. 1997). The ACC has been thoroughly investigated and a midcingulate caudal ‘cognitive’ part
and a ventral perigenual cingulate ‘emotional’ part have been identified. The midcingulate region
is activated in all pain studies and most likely related to attentional factors (Davis et al. 1997;
Kwan et al. 2000). Activation of the rostral part of the ACC (rACC) seems to be associated with
anticipation of pain (Ploghaus et al. 1999). Activations of the perigenual ACC (pACC) and the
ventral pACC have been related to pain intensity (Büchel et al. 2002). The rACC seems to play an
CENTRAL IMAGING OF PAIN AND THE PROCESS OF CHRONICITY 91
important role in placebo analgesia. Activity of the rACC during placebo analgesia covaries with
subcortical activity in the periaqueductal grey (PAG) and the amygdala and seems to be the
cognitive link of the endogenous pain control system (Bingel et al. 2006; Bingel, Schoell, & Buchel
2007). Activations of the anterior ACC are more unspecific and are most likely related to working
memory and attentional processes. From the activation of the ACC during various task and
stimulations it seems likely that the ACC is a multi-integrative structure: single ACC regions may
be involved in several functional networks, and their processing capacities with respect to a given
function are probably modulated by other concomitant processes (Apkarian et al. 2005; Lorenz &
Casey 2005).
6.3 Brain processes in chronic back pain

Using painful electric stimulation of the skin at the back or at the finger of CBP patients during a
MEG recording, Flor et al. (1997) showed significant brain activation differences between CBP
patients and healthy controls. Stimulation at the affected back but not at the finger led to a
significantly higher magnetic field in the early time window (<100 ms) for CBP compared to
controls. The signal strength in this time window and the duration of pain were positively corre-
lated, suggesting increased cortical responsivity with increasing chronicity. The source of this
early activity originated from SI and the localization of the fingers did not differ between patients
and controls. However, the localization of the back was more inferior and medial in the patients.
This indicates a shift and expansion toward the cortical representation of the leg (see Figure 6.2)
suggesting that chronic pain leads to an expansion of the cortical representation zone related to
nociceptive input, and that the amount of cortical reorganization increases with pain duration.
This is similar to the expansions of cortical representations that have been documented to occur
with other types of behaviourally relevant stimulation, but different from the changes that occur
with deafferentation where the changes correlate with pain intensity rather than chronicity.
Flor et al (2004) used painful and non-painful electric stimulation to the finger in CBP patients,
patients with headache and controls. Only the CBP patients displayed significantly lower pain
thresholds and tolerance as well as a reduced habituation, i.e. sensitization. Although the stimula-
tion intensity for the three groups was different (i.e. lower in the CBP group), the CBP patients
showed equally high levels of evoked brain responses as assessed by multichannel EEG recordings
suggesting also central sensitization. A later study by Diers et al. (2007) expanded these results to
intramuscular recordings and observed both higher pain ratings (perceptual sensitization) over
the course of the stimulation in the CBP patients and also enhanced central nervous system
processing as evidenced by an enhanced N80 component (see Figure 6.3).
Enhanced perceptual sensitization to tonic rather than phasic stimulation was found in CBP
patients by Kleinböhl et al. (1999). They used extended heat pain and a behavioral response as
indicator of sensitization.
Giesecke et al. (2004) compared chronic low back pain patients, patients with fibromyalgia and
healthy controls. During the fMRI measurements they administered two kinds of painful stimula-
tion to the thumbnails of the participants: (1) stimuli of equal pressure and (2) equal subjective
pain intensity. The pressure required to produce moderate pain was significantly higher for both
patient groups than for the control participants, indicating hyperalgesia. Equal amounts of pres-
sure applied to the thumb revealed activation in five common brain regions for the two patients
groups: contralateral SI, bilateral SII, the inferior parietal lobule and the cerebellum. In healthy
controls this stimulation resulted only in activation of the contralateral SII. These results can be
interpreted as an increased central pain processing in CBP even when the painful stimulus is at a
site distant to the region involved in clinical pain. When subjects received a stimulus that resulted
Fig. 6.2 (Also see Colour Plate 4) Localization of the representation of the digits and the back in
primary somatosensory cortex in back pain patients and healthy controls. Stimulation was on the
left side of the body, the representations are on the hemisphere contralateral to the stimulation
side. Please note the shift of the back representation of the back pain patients into a more medial
position (i.e. towards the leg representation). The shift amounted to about 2–3 cm. Reprinted from
Neuroscience Letters, 224(1), H. Flor, C. Braun, T. Elbert, and N. Birbaumer, Extensive
reorganization of primary somatosensory cortex in chronic back pain patients, pp. 5–8,
Copyright (1997), with permission from Elsevier.
in the same amount of pain, all three groups showed activation in the central components of the
pain matrix (SI, SII, insula, ACC, inferior parietal lobe), with the magnitude of activation being
greater for the two patient groups.
Baliki et al. (2006) examined variations in habitual pain rather than applying acute pain stimuli
in patients with CBP. During the fMRI measurements patients were asked to continuously rate
their habitual pain level on a scale from 0–10. An increase of spontaneous pain in CBP patients
activated regions also seen in acute pain (i.e. the anterior and posterior insula, SII, the mid-
cingulate cortex, SI and the cerebellum). However, sustained high pain additionally engaged
brain areas involved in emotion, cognition, and motivation such as the medial prefrontal cortex,
rostral anterior cingulate cortex, posterior thalami, ventral striatum, and extended amygdala.
Insular activity was also present and correlated with pain duration leading the authors to con-
clude that it may reflect the chronicity of CBP. In contrast, activation of the medial prefrontal
cortex correlated with pain intensity ratings.
In a later study Baliki et al. (2008) investigated two groups of chronic pain patients: a group with
CBP and a group with osteoarthritis (OA) of the knee. Both groups were placed in the fMRI scan-
ner and rated the intensity of their pain with a finger-span logging device. The CBP patients rated
their current pain and its fluctuations and the OA patients rated their pain in response to pressure
applied to their knee. During the pain rating task CBP patients reported fluctuations of their
spontaneous pain intensity in the absence of any overt experimental stimulus. Activation related
(a) (b)
10
Perceived intensity of stimulation (NRS 0–10) of the arm

μV Begin of stimulation
9
End of stimulation
Evoked potentials to intramuscular 8
–2.0 stimulation of the arm Chronic low back Healthy controls
7 pain patients
–1.5
6
–1.0
5
–0.5
4
0
3
0.5
2
1.0 ms 1
–50 50 150 250 350 0
Fig. 6.3 Central processing and behavioral correlates of acute muscle pain in chronic low back pain
patients. Part (a) of the figure shows the pain-related somatosensory evoked potential wave forms
for intramuscular stimulation of the left lower arm. The solid line represents the activation in the
CBP patients and the dotted line the healthy controls. The analysis revealed significantly higher
amplitude of the N80 for the pain patients. Part (b) of the figure displays the differences in
perceived intensity of the stimulation of the left lower arm at the beginning and at the end of the
stimulation period. Both groups sensitized during the course of the experiment, with the CBP
displaying higher sensitization than the healthy controls. Means and standard errors are displayed.
Reprinted from Journal of Clinical Neurophysiology, 24(1), M. Diers, C. Koeppe, E. Diesch, et al.
Central processing of acute muscle pain in chronic low back pain patients: An EEG mapping study.
Copyright (2007), with permission from Wolters Kluwer Health.
to these spontaneous fluctuations seemed to be mainly related to emotion/reward mediating

areas (i.e. the medial prefrontal cortex and the nucleus accumbens). Activations due to pressure
pain in the OA group were found in SII, the insula, the supplementary motor area, the ACC, the
medial frontal gyrus, the thalamus, the right putamen and the left amygdala. This activation pat-
tern was similar to that found in healthy participants in response to acute pain (i.e. activation of
the pain matrix). In a pre-post design the authors also investigated the activation of brain regions
before and after a 2-week treatment with a lidocaine patch applied to the painful body part. On a
behavioral level, CBP patients experienced a robust decrease in pain intensity in response to the
two week treatment. Before treatment, CBP patients showed increased brain activity mainly in
the frontal cortex (including the medial prefrontal cortex, the rostral anterior cingulate cortex,
bilateral superior frontal gyrus, and the nucleus accumbens). Increased brain activity was also
found in the inferior temporal gyrus, and the left posterior parietal cortex. After treatment the
authors reported only one significant cluster of activation that showed increased activity for
spontaneous pain, the left motor area (MT). The medial prefrontal cortex was the brain area the
correlated best with the treatment effect. The authors conclude that the spontaneous pain occur-
ring in CBP is primarily of an emotional nature and assume that lidocaine treatment decreases
mainly emotional pain aspects.
Using PET and noxious heat stimulation Derbyshire et al. (2002) examined 16 CBP patients
and 16 matched controls. They were interested in the correlation of pain scores and brain activa-
tion patterns. Significant bilateral activation was present in the cerebellum, the midbrain (includ-
ing the PAG region), the thalamus, the lentiform nucleus, the insula and the midcingulate cortex
for both healthy controls and CBP patients. They reported no significant differences between the
two groups and concluded that there is no abnormal nociceptive processing in patients with low
back pain. Although the authors reported a significant difference in anxiety between the patients
and the controls, with the patients being more anxious than the controls, they did not include this
difference in their analysis. They also reported a trend towards a difference in depression scores
between the two groups, again with the patients being more depressed than the controls and
discuss the possible implications of depression on their results.
Thunberg et al. (2005) also used PET in healthy volunteers and injected hypertonic saline into
the right m. erector spinae at the level of L3 in an attempt to mimic the pain of low back pain
patients. The authors distinguished between two phases: an early acute phase (4min after the start
of the infusion), and a late tonic phase (21min after the start of the infusion). They assumed that
the latter should be a more realistic model of the chronic muscle pain found in patients. The cer-
ebral response differed in the acute and tonic phase and longer lasting tonic muscle pain resulted
in decreases in three clusters (left insula, right insula and right cingulate cortex). Increases in
rCBF were only found bilaterally in the occipital cortex. The authors interpret the decrease in
subjective levels between the early phase and the late phase as a habituation effect and thus also
attribute the changes in brain activity to this habituation. When the late phase was compared with
a baseline condition, regions that are implicated in the processing of all three components of pain
(sensory-discriminative, motivational-affective, and cognitive-evaluative) were active. These
included an increase of activation in the contralateral medial PFC as well as a decrease of activa-
tion in the insula, the ACC and the dorsolateral prefrontal cortex of the ipsilateral hemisphere.
The authors conclude that this dysfunction of the network may contribute to the development of
chronic pain.
Almost all studies on brain imaging revealed that certain brain regions show a deactivation
during task performance. This fact has led to the proposal of a ‘default mode network’ (DMN) of
brain function (Raichle et al. 2001). Baliki et al. (2008) suggested that long-term pain alters the
functional connectivity of the cortical regions of the DMN, which suggests that chronic pain has
a widespread impact on overall brain function and not only on the pain matrix. They studied CBP
patients and healthy controls during a simple visual attention task while the blood-oxygen-level-
dependent (BOLD) response was measured. Task performance was similar in both groups, but
the CBP patients showed a reduced deactivation in the medial prefrontal cortex, amygdalae and
PCC. The disruption of the DMN was interpreted as cause of many of the cognitive and behavio-
ral impairments that accompany chronic pain (i.e. depression, anxiety, sleep disturbances, or
decision-making abnormalities).
6.4 Structural and biochemical changes in the brains of

patients with CBP
So far we have only discussed functional changes in patients with CBP. Prolonged and sustained
functional changes can, however, also lead to structural changes of the brain (see also Chapter 7).
Apkarian et al. (2004) examined the grey matter volume of patients with CBP and healthy using
voxel-based morphometry (VBM) as well as alternative methods of structural analyses. With
VBM they found a 5.4% decrease in whole-brain neocortical grey matter volume in CBP patients
compared to healthy controls. With multiple linear regression they showed that after correcting
for age and sex, pain duration was a significant predictor for whole-brain grey matter volume in
CBP. A separate analysis of the thalamus showed a significant decrease in grey matter density in
the CBP group for the right anterior thalamus. Additionally, the dorsolateral prefrontal cortex
was the main region showing local decreases in grey matter density in CBP patients compared
to controls. The magnitude of decrease in grey matter volume (between 5–11%) is equivalent to
the grey matter volume lost in 10–20 years of normal aging.
Schmidt-Wilcke et al. (2006) also used VBM and demonstrated a reduction of the grey matter
of the primary somatosensory cortex and brainstem in individuals with CBP. These decreases
were, in contrast, significantly correlated with the unpleasantness and intensity of pain. Increases
in grey matter were observed bilaterally in the striatum and the left thalamus. Older adults with
CBP showed structural brain changes in the middle corpus callosum, the grey matter of the pos-
terior parietal cortex as well as impaired attention and mental flexibility (Buckalew et al. 2008).
Investigating a slightly different group Valet et al. (2009) studied patients with the DSM-IV
diagnosis of pain disorder with VBM. The predominant clinical pain in these patients was either
located in the head, neck and shoulder region; the lower back region; the pelvic region; or the
lower limbs. Some persons reported more than one predominant pain region. The authors found
significant grey matter decreases in structures involved in pain processing (i.e. insula, PPC, OFC,
vmPFC), confirming studies of patients with fibromyalgia, chronic tension-type headache and
CBP, that also showed these decreases (Schmidt-Wilcke et al. 2005, 2006, 2007).
Magnetic resonance spectroscopy (MRS) monitors biochemical changes in specific regions of
the brain. Early studies on patients showed that chronic low back pain is associated with altered
brain chemistry (Grachev, Fredrickson, & Apkarian 2000). The authors demonstrated a reduc-
tion of two molecules (N-acetyl aspartate (NAA) and glucose) in the dorsolateral prefrontal cor-
tex specifically in CBP patients. A later study by the same authors, however, showed that there was
a stronger association of reduced NAA with depression rather than with pain (Grachev et al.
2003). This nicely illustrates that MRS data have to be interpreted carefully since biochemical
changes in the brain that may be due to other factors than pain such as depression or medication.
Siddall et al. (2006) used a different method of spectral analysis in a study that compared indi-
viduals with chronic low back pain and persons without pain. They obtained MR spectra from
several brain regions (the ACC, the thalamus and the prefrontal cortex) and analyzed and com-
pared them using a pattern recognition method. With this method they were able to discriminate
individuals with low back pain from controls with an accuracy of 97% using the spectrum from
the PFC, 99% using the spectrum of the thalamus and 100% using the spectrum of the ACC based
only on the regional chemical brain differences. Unfortunately without proper longitudinal stud-
ies, the causes and effects for the structural and biochemical changes in the brains of chronic pain
patients cannot be untangled. It could be that the structural changes predate the onset of pain and
might even be a vulnerability factor (see Chapter 7 for further information).
6.5 The role of learning and memory

Earlier we have discussed that the brain, and specifically the primary sensory cortex reorganizes
in chronic pain (Flor et al. 1997). This reorganization and expansion can be viewed as an implicit
learning and memory process where the experience of pain leaves a long-term alteration in the
brain, which influences the subsequent processing of painful and non-painful somatosensory
stimuli. This non-associative sensitization process may be complemented by operant and respond-
ent conditioning and other associative learning processes and thus a variety of pain associated
stimuli and responses may become part of a larger pain-related network that may be maladaptive
in the process of chronicity (Apkarian, Baliki, & Geha 2009; Flor 2009; Flor & Diers 2007).
In a series of studies Pincus et al. (1993, 1995, 1996) reported that chronic pain patients not
only selectively perceive pain-related words, they also make more pain-related associations to
ambiguous words and are more likely to retrieve pain-associated words from memory. In an EEG
study, Flor et al. (1997) examined whether CBP patients detect pain- and body-related words
earlier, allocate more attention to them, and process them more deeply than neutral words.
Additionally they examined peripheral responses (heart rate, muscle tension and skin conduct-
ance level). Although the CBP group did not recognize the pain- and body-related words any
better or earlier than the healthy controls, they showed selectively enhanced N100 and N200
components in response to pain-related words. This increase in early reactivity was accompanied
by enhanced skin conductance levels. Overall, the data confirm the assumption of altered implicit
pain memories and selective attention to pain-related materials in chronic pain patients.
Similar results were found for pre-CBP patients by Knost et al. (1997). In contrast to the CBP
patients, these individuals reported episodic upper back pain for less than 6 months. Again word
recognition performance was not different between the two groups; but a significant enhance-
ment of the N100 component for pain-related words was found in the prechronic pain group.
These results suggest that a differential cortical processing of pain-related material may be a
vulnerability marker. Montoya et al. (1996) showed that these central differences in word process-
ing may be related to emotional learning. When certain words were associated with painful shock
in healthy humans, these subjects showed larger N100 amplitudes and a more negative-going
slow wave 400–800 ms after word presentation to shock-related compared to non-shock words.
The authors interpreted their results as an activation of cell assemblies representing the memory
of the learned word-shock contingency. Furthermore, the increased N100 amplitude can be inter-
preted as an increased attentive facilitation to aversive pain-related information as a consequence
of conditioning.
Another study that examined the role of classical conditioning on muscular responses and EEG
measures in chronic pain patients was conducted by Schneider et al. (2004). They used aversive
slides as conditioned stimuli, which were followed by intracutaneous electric shocks to the left
index finger. The CBP patients already showed an increased muscular response in the precondi-
tioning phase, selective to the area where the shock was later administered. This can be inter-
preted as an anticipatory muscular response to predicted painful stimuli in CBP patients. In the
acquisition phase only the CBP patients showed the expected classical conditioning effect for both
the shock site and the right trapezius (i.e. the site of their pain). The results of the extinction phase
suggest that the patients generalized the conditioned response (elevated muscular tension) to
other stimuli (in this case to the CS). In addition the EEG measures indicated a dysfunctional
brain response to the anticipation of the painful stimulus as well as in consequence to it. These
results suggest that classical conditioning processes may be important in the development and
maintenance of chronic back pain.
Diesch and Flor (2007) examined if non-painful tactile stimuli could serve as conditioned
stimuli and could be connected to painful stimuli via classical conditioning in healthy controls.
They used innocuous electrical stimuli to one finger as conditioned stimulus and painful electric
shock to the lower back as the unconditioned stimulus. They found that muscle tension levels
were easily conditioned and that the representation of the finger involved in the conditioning
procedure in somatosensory cortex expanded and shifted towards the back representation. These
results support the hypothesis that classical conditioning may also be present in the somato-
sensory system and may evoke pain sensations to normally not painful stimuli. Priming is another
form of implicit memory where, for example, repetitions of previously experienced stimuli lead
to altered behavior. Dillmann et al. (2000) investigated the effect of different semantic primes on
the processing of painful stimuli. Three different categories were used as prime stimuli: somato-
sensory pain-related words, affective pain-related words and neutral adjectives. While viewing
the primes subjects received a painful laser stimulus to their left hand. Painful stimuli applied
while the subjects viewed pain-related words (affective and somatosensory words) resulted
in larger evoked amplitudes 370ms after stimulus application when compared to neutral words.
The authors interpret this result as a preactivation of pain memory due to the processing of pain-
related words.
Whereas respondent conditioning focuses on the development of stimulus associations, oper-
ant conditioning deals with the association of responses and reinforcing stimuli. Fordyce (1976)
was the first to describe the important role of operant conditioning in chronic pain. He suggested
that both positive and negative reinforcement of pain behaviors as well as a lack of reinforcement
for well behaviors contribute to chronicity. Although this model had a significant impact on the
treatment of chronic pain, there is little experimental evidence on the role of operant factors in
chronicity and the underlying brain mechanisms. Flor et al. (2002) examined the role of operant
conditioning during acute painful stimulation in CBP patients and healthy controls. EEG, heart
rate, skin conductance and muscle tension were concurrently assessed. Half of the chronic pain
patients and healthy controls were reinforced for decreased pain reports and the other half were
reinforced for increased pain reports. While both groups showed similar learning rates, the CBP
group displayed slower extinction of both the verbal and the cortical (N150) pain response. There
was no significant correlation between resistance to extinction and pain duration. The authors
interpreted this result as evidence that the prolonged operant conditioning effects might precede
the onset of chronic pain and might thus be a predisposing factor.
From an operant learning perspective, tensing painful muscles can also be viewed as a learned
pain behavior that may be maintained by its consequences (reduced pain impact). Chronic pain
patients might have learned to increase their muscle tension in anticipation to painful stimula-
tion, in an attempt to decrease the painfulness of the episode. While this coping behavior might
temporarily alleviate pain, it may, however, over time lead to enhanced pain perception as it
stimulates and sensitizes nociceptors. Knost et al. (1999) investigated CBP patients, people at
high risk for CBP, and healthy controls who received painful electric stimuli to the forearm or
lower back, while they had to produce high or low muscle tension levels. The CBP showed sig-
nificantly elevated and the high risk group a trend to higher N150 and N150/P260 amplitudes
when the low muscle tension condition was compared with then high muscle tension condition
(see Figure 6.4). These results suggest that operantly conditioned increased muscle tension levels
and accompanying cortical network changes contribute to pain chronicity.
Hölzl et al. (2005) devised an operant conditioning design that relies on implicit, non-conscious
changes in pain-related reinforcement, which might be very relevant for pain development. They
employed a sequence of interconnected trials with tonic heat stimuli applied to the thenar of the
dominant hand. Stimulus temperature increased or decreased from trial to trial depending on the
participant’s adjustment of the heat stimuli in the previous trial and the responses of the subjects
were reinforced temperature changes without the subject’s knowledge. They showed that increases
and decreases in pain sensitivity could be learned implicitly without the participants’ awareness.
6.6 The role of cognitive factors

The role of cognitive factors such as helplessness, lack of coping skills and catastrophizing in the
chronicity process has been emphasized for a long time (cf. Turk, Meichenbaum, & Genest 1983).
In addition, the fear-avoidance model of chronic back pain by Vlaeyen & Linton (2000) places an
emphasis on the mis-interpretation and negative evaluation of acute pain. When pain is (mis-)
interpreted and negatively evaluated it can lead to the development of a vicious circle of fear of
pain, pain anxiety, avoidance, and depression. Pain catastrophizing has been extensively exam-
ined in chronic pain patients and has been consistently associated with disability and enhanced
pain behaviours (Peters, Vlaeyen, & Weber 2005; Sullivan, Lynch, & Clark 2005; Turner, Mancl,
& Aaron 2004).
–6 Healthy controls
Subchronic group
Chronic back pain
–5
–4
μ Volt
–3
–2
–1
0
Low tension High tension
Fig. 6.4 Mean N150 amplitudes averaged across all electrodes for the painful stimulation (averaged
across back and arm stimulation) are displayed. Healthy controls showed no difference in EEG
amplitudes between the high and low tension condition, whereas CBP patients displayed a
significant difference, with higher amplitudes during the low tension condition. The results of the
subchronic group showed a trend towards significance. Reproduced from Psychophysiology 36(6),
J. C. Ziegler, A. Benraïss, and M. Besson, From print to meaning: An electrophysiological
investigation of the role of phonology in accessing word meaning, pp. 775–85, © 1999 with
permission from John Wiley and Sons.
Based on the findings associated with chronicity Lloyd et al. (2008) studied the brain correlates
of painful stimulation in patients with high and low levels of clinical indicators of continued
chronicity, the so-called non-organic signs and symptoms as first described by Waddell et al.
(1980). High scores of Waddell signs (WS) are associated with poor physical performance,
increased catastrophizing and poor treatment outcome whereas patients with low scores do not
differ from healthy controls. The authors delivered tactile (intense and unpleasant but not pain-
ful) stimuli to the back of the participants with high (WS-H) or low Waddell (WS-L) signs. WS-H
patients, who have high catastrophizing and disability scores, did not differ in brain activation
patters from the healthy controls. In contrast, WS-L patients showed significantly more activation
in the left superior parietal lobe and the left extrastriate cortex than the controls. This result is
interesting since healthy controls and WS-L patients did not differ in their behavior, but showed
differential brain activation in response to pain. When the two patient groups were directly com-
pared, the WS-L patients showed more activation in the posterior cingulate cortex and the extras-
triate cortex. The authors note that these regions, which are associated with affective-cognitive
processing, might reflect better coping (WS-L group). There were no areas that were more active
in the WS-H compared to the WS-L group. A correlation analysis between catastrophizing and
the BOLD response was significant in the inferior parietal cortex and the superior parietal lobe.
In accordance with prior findings on chronic back pain patients, Lloyd et al. (2008) found a
moderate medial shift of activation (4mm) in primary somatosensory cortex after tactile stimula-
tion of the back for the WS-H compared to the WS-L group. However, patients were allowed to
continue their stable medication, including opioids, antiepileptics, and antidepressants. Although
medication did not differ between the two patient groups, none of the healthy controls took anal-
gesic medication. Since these drugs directly influence the brain chemistry, some of the differences
between the groups examined might be related to this difference. Another variable that needs to
be considered is the significant age difference between the healthy controls and the patient groups
since age is known to have an influence on brain activation (Abe et al. 2002; Hock et al. 1995;
Tumeh et al. 2007).
Gracely et al. (2004) examined the association of catastrophizing independent of depressive
symptoms and brain responses to pain in fibromyalgia patients. Activation of the ipsilateral SII
was more than twice as large in the persons high on catastrophizing compared to those with a low
score. In addition, patients high on catastrophizing scores displayed an additional activation in
the contralateral rostral ACC and bilateral lentiform nuclei that was not present in the low
catastrophizers. More research is needed to determine the brain mechanisms that underlie
catastrophizing in chronic pain.
The effects of attention on pain have been known for a long time and have been investigated in
many studies in healthy humans. For example, Miltner et al. (1989) examined effects of attention
on pain perception using multichannel EEG. They delivered painful and non-painful stimuli
intracutaneously to the tip of the finger while recording somatosensory event-related potentials.
In one condition subjects had to attend to the painful stimuli by silently counting them and in the
other condition they were told to ignore the stimuli and additionally solve a difficult puzzle. Their
results show that the degree of attention a subject pays to a painful stimulus modulates the ampli-
tudes of the later components of the EEG (P200 and P300), which were larger for attended than
for unattended stimuli. For example, Bantick et al. (2002) assessed how attention to pain versus
distraction from pain (by a cognitively demanding Stroop task) affects central processing of
healthy subjects using fMRI. Cognitive distraction reduced activation in a number of areas, e.g.
the thalamus, the insula, and the cognitive division of the ACC. Increased activation was found in
the affective division of the ACC and orbitofrontal regions related to effects of both pain and
distraction. Tracey et al. (2002) showed that in healthy subjects the periaqueductal grey—an area
important for opioid analgesia—is also involved in attentional control and thus cognitive
functions related to pain. Similar studies in chronic pain patients and those at risk for chronicity
are needed.
6.7 Imaging and therapeutic interventions

Research about effective treatment strategies for CBP patients has shown that many different
behavioural and somatic treatments are effective in interdisciplinary settings (Cherkin et al. 2003;
Scascighini et al. 2008). Unfortunately, for CBP patients there are not many studies that investi-
gated the brain changes that occur as an effect of pain reduction. deCharms et al. (2005) used
real-time fMRI to teach participants to deliberately control the activation in the rACC, a region
involved in pain perception and modulation. During the control condition of the experiment
subjects saw a scrolling line chart of the activation of their rACC region next to a visual image of
a smaller or larger fire. Participants where then instructed to increase or decrease the activation.
After sufficient time for training, subjects received painful heat stimuli to their left palm. During
the early trials participants did not rate the pain stimuli any different in the two conditions
‘increase’ and ‘decrease’, however at the end of the training stimuli received during the ‘increase’
condition were rated as significantly more painful than those during the ‘decrease’ condition.
Using a regression analysis de Charmes et al. (2005) showed that changes in activity in the
rACC and changes in pain perception were significantly correlated. Additionally chronic pain
patients were examined with the same setup, but without nociceptive stimulation. The authors
also reported a significant decrease of the patients’ overall pain level after the one day of training.
As they pointed out it is necessary for participants to learn control over the activity of a specific
brain area, since participants taught to control the activity of the PCC did not show the same
decrease in pain perception. Participants presented with a pre-recorded video of the real-
time information of an earlier group did also not display any effect. Direct feedback thus seems to
be important.
Crawford et al. (1998) used hypnotic instructions during a cold-pressor test training and subse-
quently examined somatosensory event-related potentials during noxious electrical stimulation.
In the first part of the experiment subjects received a hypnotic analgesia instruction (cold pressor
pain training) and in the second part they received painful electrical stimuli during two conditions
(hypnotic analgesia and attention to the painful stimuli). The participants showed an increased
N140 potential in the anterior frontal region during the hypnotic analgesia condition and a pre-
stimulus positive-going contingent cortical potential at the same site. A contingent positive varia-
tion seems to be a reflection of the participant’s effort to inhibit irrelevant movement and is a
variation of the contingent negative variation or readiness potential. Additionally, reduced ampli-
tudes of the P200 and P300 components were found and interpreted as a decreased spatiotempo-
ral perception of the painful stimuli. On a behavioural level, the hypnotic analgesia instruction led
to a significantly reduced perception of pain and distress. In chronic pain patients a successful
transfer of the hypnotic analgesia suggestion into daily life was found. After three experimental
sessions the patients reported reduced pain levels, increased well-being and increased quality of
sleep. There are no studies available that examined the effects of cognitive-behavioural pain treat-
ment or preventive interventions on brain function in patients with CBP or those at risk.
6.8 Conclusions and outlook

We have provided a brief overview over functional and structural changes in the brains of patients
with CBP or at high risk for chronicity. We delineated the brain areas involved in the processing
of pain and pointed out differences between acute and chronic pain. The examination of
the structural and functional neuronal correlates of chronicity is an important task for future
research. Longitudinal studies that elucidate the role of emotional learning, and cognitive factors,
and their brain correlates could shed light on the causality of the factors involved but are currently
lacking.
Acknowledgements
This work was supported by the Bundesministerium für Bildung und Forschung (grant #
01GW0531) and the Award for Basic Research of the State of Baden-Württemberg to HF.
References
Abe, O, Aoki, S, Hayashi, N, et al. (2002). Normal aging in the central nervous system: quantitative MR
diffusion-tensor analysis. Neurobiology of Aging, 23, 433–41.
Apkarian, AV, Baliki, MN & Geha, PY (2009). Towards a theory of chronic pain. Progress in Neurobiology,
87, 81–97.
Apkarian, AV, Bushnell, MC, Treede, RD & Zubieta, JK (2005). Human brain mechanisms of pain
perception and regulation in health and disease. European Journal of Pain, 9, 463–84.
Apkarian, AV, Sosa, Y, Sonty, S, et al. (2004). Chronic back pain is associated with decreased prefrontal
and thalamic gray matter density. Journal of Neuroscience, 24, 10410–5.
Baliki, MN, Chialvo, DR, Geha, PY, et al. (2006). Chronic pain and the emotional brain: specific brain
activity associated with spontaneous fluctuations of intensity of chronic back pain. Journal of
Baliki, MN, Geha, PY, Apkarian, AV & Chialvo, DR (2008). Beyond feeling: chronic pain hurts the brain,
disrupting the default-mode network dynamics. Journal of Neuroscience, 28, 1398–403.
Baliki, MN, Geha, PY, Jabakhanji, R, Harden, N, Schnitzer, TJ & Apkarian, AV (2008). A preliminary
fMRI study of analgesic treatment in chronic back pain and knee osteoarthritis. Molecular Pain, 4, 47.
Bantick, SJ, Wise, RG, Ploghaus, A, Clare, S, Smith, SM & Tracey, I (2002). Imaging how attention
modulates pain in humans using functional MRI. Brain, 125, 310–19.
Baron, R, Baron, Y, Disbrow, E & Roberts, TP (1999). Brain processing of capsaicin-induced secondary
hyperalgesia: a functional MRI study. Neurology, 53, 548–57.
Bingel, U, Lorenz, J, Schoell, E, Weiller, C & Buchel, C (2006). Mechanisms of placebo analgesia: rACC
recruitment of a subcortical antinociceptive network. Pain, 120, 8–15.
Bingel, U, Quante, M, Knab, R, Bromm, B, Weiller, C & Buchel, C (2003). Single trial fMRI reveals
significant contralateral bias in responses to laser pain within thalamus and somatosensory cortices.
Neuroimage, 18, 740–8.
Bingel, U, Schoell, E & Buchel, C (2007). Imaging pain modulation in health and disease. Current Opinions
in Neurology, 20, 424–31.
Bornhovd, K, Quante, M, Glauche, V, Bromm, B, Weiller, C & Buchel, C (2002). Painful stimuli evoke
different stimulus-response functions in the amygdala, prefrontal, insula and somatosensory cortex: a
single-trial fMRI study. Brain, 125, 1326–36.
Brooks, JC, Nurmikko, TJ, Bimson, WE, Singh, KD & Roberts, N (2002). fMRI of thermal pain: effects of
stimulus laterality and attention. Neuroimage, 15, 293–301.
Buchel, C, Bornhovd, K, Quante, M, Glauche, V, Bromm, B & Weiller, C (2002). Dissociable neural
responses related to pain intensity, stimulus intensity, and stimulus awareness within the anterior
cingulate cortex: a parametric single-trial laser functional magnetic resonance imaging study. Journal of
Buckalew, N, Haut, MW, Morrow, L & Weiner, D (2008). Chronic pain is associated with brain volume
loss in older adults: preliminary evidence. Pain Medicine, 9, 240–8.
Bushnell, MC, Duncan, GH, Hofbauer, RK, Ha, B, Chen, JI & Carrier, B (1999). Pain perception: is
there a role for primary somatosensory cortex? Proceedings of the National Academy of Sciences,
96, 7705–9.
Chen, JI, Ha, B, Bushnell, MC, Pike, B & Duncan, GH (2002). Differentiating noxious- and innocuous-
related activation of human somatosensory cortices using temporal analysis of fMRI. Journal of
Neurophysiology, 88, 464–74.
Cherkin, DC, Sherman, KJ, Deyo, RA & Shekelle, PG (2003). A review of the evidence for the effectiveness,
safety, and cost of acupuncture, massage therapy, and spinal manipulation for back pain. Annals of
Internal Medicine, 138, 898–906.
Coghill, RC, Sang, CN, Maisog, JM & Iadarola, MJ (1999). Pain intensity processing within the human
brain: a bilateral, distributed mechanism. Journal of Neurophysiology, 82, 1934–43.
Coghill, RC, Talbot, JD, Evans, AC, et al. (1994). Distributed processing of pain and vibration by the
human brain. Journal of Neuroscience, 14, 4095–108.
Craig, AD (2002). How do you feel? Interoception: the sense of the physiological condition of the body.
Nature Reviews Neuroscience, 3, 655–66.
Craig, AD (2009). How do you feel—now? The anterior insula and human awareness. Nature Reviews
Craig, AD, Chen, K, Bandy, D & Reiman, EM (2000). Thermosensory activation of insular cortex. Nature
Craig, AD, Reiman, EM, Evans, A & Bushnell, MC (1996). Functional imaging of an illusion of pain.
Nature, 384, 258–60.
Crawford, HJ, Knebel, T, Kaplan, L, et al. (1998). Hypnotic analgesia: 1. Somatosensory event-related
potential changes to noxious stimuli and 2. Transfer learning to reduce chronic low back pain.
International Journal of Clinical and Experimental Hypnosis, 46, 92–132.
Davis, KD, Taylor, SJ, Crawley, AP, Wood, ML & Mikulis, DJ (1997). Functional MRI of pain- and
attention-related activations in the human cingulate cortex. Journal of Neurophysiology, 77, 3370–80.
deCharms, RC, Maeda, F, Glover, GH, et al. (2005). Control over brain activation and pain learned by
using real-time functional MRI. Proceedings of the National Academy of Sciences, 102, 18626–31.
Derbyshire, SW, Jones, AK, Creed, F, et al. (2002). Cerebral responses to noxious thermal stimulation in
chronic low back pain patients and normal controls. Neuroimage, 16, 158–68.
Diers, M, Koeppe, C, Diesch, E, et al. (2007). Central processing of acute muscle pain in chronic low back
pain patients: an EEG mapping study. Journal of Clinical Neurophysiology, 24, 76–83.
Diesch, E & Flor, H (2007). Alteration in the response properties of primary somatosensory cortex related
to differential aversive Pavlovian conditioning. Pain, 131, 171–80.
Dillmann, J, Miltner, WH & Weiss, T (2000). The influence of semantic priming on event-related
potentials to painful laser-heat stimuli in humans. Neuroscience Letters, 284, 53–6.
Faurion, A, Cerf, B, Van De Moortele, PF, Lobel, E, Mac Leod, P & Le Bihan, D (1999). Human taste
cortical areas studied with functional magnetic resonance imaging: evidence of functional lateralization
related to handedness. Neuroscience Letters, 277, 189–92.
Ferretti, A, Babiloni, C, Gratta, CD, et al. (2003). Functional topography of the secondary somatosensory
cortex for nonpainful and painful stimuli: an fMRI study. Neuroimage, 20, 1625–38.
Flor, H (2009). Extinction of Pain Memories: Importance for the Treatment of Chronic Pain. in J Castro-
Lopes (ed.), Current Topics in Pain: 12th World Congress on Pain, IASP Press, Seattle, 221–44.
Flor, H, Braun, C, Elbert, T & Birbaumer, N (1997). Extensive reorganization of primary somatosensory
cortex in chronic back pain patients. Neuroscience Letters, 224, 5–8.
Flor, H & Diers, M (2007). Limitations of pharmacotherapy: behavioral approaches to chronic pain.
Handbook of Experimental Pharmacology, 177, 415–27.
Flor, H, Diers, M & Birbaumer, N (2004). Peripheral and electrocortical responses to painful and
non-painful stimulation in chronic pain patients, tension headache patients and healthy controls.
Neuroscience Letters, 361, 147–50.
Flor, H, Knost, B & Birbaumer, N (1997). Processing of pain- and body-related verbal material in chronic
pain patients: central and peripheral correlates. Pain, 73, 413–21.
Flor, H, Knost, B & Birbaumer, N (2002). The role of operant conditioning in chronic pain: an
experimental investigation. Pain, 95, 111–8.
Fordyce, WE (1976) Behavioral concepts in chronic pain and illness. in PO Davidson (ed.), The behavioral
treatment of anxiety, depression and pain, Bruner & Mazel, New York, 147–88.
Gallace, A & Spence, C (2009). The cognitive and neural correlates of tactile memory. Psychological Bulletin,
135, 380–406.
Giesecke, T, Gracely, RH, Grant, MA, et al. (2004). Evidence of augmented central pain processing in
idiopathic chronic low back pain. Arthritis and Rheumatism, 50, 613–23.
Gracely, RH, Geisser, ME, Giesecke, T, et al. (2004). Pain catastrophizing and neural responses to pain
among persons with fibromyalgia. Brain, 127, 835–43.
Grachev, ID, Fredrickson, BE & Apkarian, AV (2000). Abnormal brain chemistry in chronic back pain: an
in vivo proton magnetic resonance spectroscopy study. Pain, 89, 7–18.
Grachev, ID, Ramachandran, TS, Thomas, PS, Szeverenyi, NM & Fredrickson, BE (2003). Association
between dorsolateral prefrontal N-acetyl aspartate and depression in chronic back pain: an in vivo
proton magnetic resonance spectroscopy study. Journal of Neural Transmission, 110, 287–312.
Harris, JA, Harris, IM & Diamond, ME (2001). The topography of tactile working memory. Journal of
Hock, C, Muller-Spahn, F, Schuh-Hofer, S, Hofmann, M, Dirnagl, U & Villringer, A (1995). Age
dependency of changes in cerebral hemoglobin oxygenation during brain activation: a near-infrared
spectroscopy study. Journal of Cerebral Blood Flow and Metabolism, 15, 1103–8.
Iadarola, MJ, Berman, KF, Zeffiro, TA, et al. (1998). Neural activation during acute capsaicin-evoked pain
and allodynia assessed with PET. Brain, 121, 931–47.
Kleinbohl, D, Holzl, R, Moltner, A, Rommel, C, Weber, C & Osswald, PM (1999). Psychophysical
measures of sensitization to tonic heat discriminate chronic pain patients. Pain, 81, 35–43.
Knost, B, Flor, H, Birbaumer, N & Schugens, MM (1999). Learned maintenance of pain: muscle tension
reduces central nervous system processing of painful stimulation in chronic and subchronic pain
patients. Psychophysiology, 36, 755–64.
Knost, B, Flor, H, Braun, C & Birbaumer, N (1997). Cerebral processing of words and the development of
chronic pain. Psychophysiology, 34, 474–81.
Kwan, CL, Crawley, AP, Mikulis, DJ & Davis, KD (2000). An fMRI study of the anterior cingulate cortex
and surrounding medial wall activations evoked by noxious cutaneous heat and cold stimuli. Pain,
85, 359–74.
Lloyd, D, Findlay, G, Roberts, N & Nurmikko, T (2008). Differences in low back pain behavior are
reflected in the cerebral response to tactile stimulation of the lower back. Spine, 33, 1372–7.
Lorenz, J & Casey, KL (2005). Imaging of acute versus pathological pain in humans. European Journal of
Pain, 9, 163–5.
Maihofner, C, Kaltenhauser, M, Neundorfer, B & Lang, E (2002). Temporo-spatial analysis of cortical
activation by phasic innocuous and noxious cold stimuli—a magnetoencephalographic study. Pain,
100, 281–90.
Melzack, R & Casey, KL (1968). Sensory, motivational, and central control determinants of pain. in
DR Kenshalo (ed.), The skin senses, Charles C. Thomas, Springfield (IL), 423–39.
Miltner, W, Johnson, R, Jr., Braun, C & Larbig, W (1989). Somatosensory event-related potentials to
painful and non-painful stimuli: effects of attention. Pain, 38, 303–12.
Montoya, P, Larbig, W, Pulvermuller, F, Flor, H & Birbaumer, N (1996). Cortical correlates of semantic
classical conditioning. Psychophysiology, 33, 644–9.
Peters, ML, Vlaeyen, JW & Weber, WE (2005). The joint contribution of physical pathology, pain-related
fear and catastrophizing to chronic back pain disability. Pain, 113, 45–50.
Peyron, R, Laurent, B & Garcia-Larrea, L (2000). Functional imaging of brain responses to pain. A review
and meta-analysis (2000). Clinical Neurophysiology, 30, 263–88.
Pincus, T, Pearce, S, McClelland, A & Isenberg, D (1995). Endorsement and memory bias of
self-referential pain stimuli in depressed pain patients. British Journal of Clinical Psychology,
34 (Pt 2), 267–77.
Pincus, T, Pearce, S, McClelland, A & Turner-Stokes, L (1993). Self-referential selective memory in pain
patients. British Journal of Clinical Psychology, 32 (Pt 3), 365–74.
Pincus, T, Pearce, S & Perrott, A (1996). Pain patients’ bias in the interpretation of ambiguous
homophones. British Journal of Medical Psychology, 69 (Pt 3), 259–66.
Ploghaus, A, Tracey, I, Gati, JS, et al. (1999). Dissociating pain from its anticipation in the human brain.
Science, 284, 1979–81.
Ploner, M, Freund, HJ & Schnitzler, A (1999). Pain affect without pain sensation in a patient with a
postcentral lesion. Pain, 81, 211–4.
Raichle, ME, MacLeod, AM, Snyder, AZ, Powers, WJ, Gusnard, DA & Shulman, GL (2001). A default
mode of brain function. Proceedings of the National Academy of Sciences, 98, 676–82.
Rainville, P, Duncan, GH, Price, DD, Carrier, B & Bushnell, MC (1997). Pain affect encoded in human
anterior cingulate but not somatosensory cortex. Science, 277, 968–71.
Ringler, R, Greiner, M, Kohlloeffel, L, Handwerker, HO & Forster, C (2003). BOLD effects in different
areas of the cerebral cortex during painful mechanical stimulation. Pain, 105, 445–53.
Scascighini, L, Toma, V, Dober-Spielmann, S & Sprott, H (2008). Multidisciplinary treatment for chronic
pain: a systematic review of interventions and outcomes. Rheumatology (Oxford), 47, 670–8.
Schmidt-Wilcke, T, Leinisch, E, Ganssbauer, S, et al. (2006). Affective components and intensity of pain
correlate with structural differences in gray matter in chronic back pain patients. Pain, 125, 89–97.
Schmidt-Wilcke, T, Leinisch, E, Straube, A, et al. (2005). Gray matter decrease in patients with chronic
tension type headache. Neurology, 65, 1483–6.
Schmidt-Wilcke, T, Luerding, R, Weigand, T, et al. (2007). Striatal grey matter increase in patients
suffering from fibromyalgia—a voxel-based morphometry study. Pain, 132 (Suppl 1), S109–16.
Schneider, C, Palomba, D & Flor, H (2004). Pavlovian conditioning of muscular responses in chronic pain
patients: central and peripheral correlates. Pain, 112, 239–47.
Siddall, PJ, Stanwell, P, Woodhouse, A, et al. (2006). Magnetic resonance spectroscopy detects biochemical
changes in the brain associated with chronic low back pain: a preliminary report. Anesthesia &
Analgesia, 102, 1164–8.
Small, DM, Voss, J, Mak, YE, Simmons, KB, Parrish, T & Gitelman, D (2004). Experience-dependent
neural integration of taste and smell in the human brain. Journal of Neurophysiology, 92, 1892–903.
Sullivan, MJ, Lynch, ME & Clark, AJ (2005). Dimensions of catastrophic thinking associated with pain
experience and disability in patients with neuropathic pain conditions. Pain, 113, 310–5.
Thunberg, J, Lyskov, E, Korotkov, A, et al. (2005). Brain processing of tonic muscle pain induced by
infusion of hypertonic saline. European Journal of Pain, 9, 185–94.
Tracey, I & Mantyh, PW (2007). The cerebral signature for pain perception and its modulation. Neuron,
55, 377–91.
Tracey, I, Ploghaus, A, Gati, JS, et al. (2002). Imaging attentional modulation of pain in the periaqueductal
gray in humans. Journal of Neuroscience, 22, 2748–52.
Treede, RD, Apkarian, AV, Bromm, B, Greenspan, JD & Lenz, FA (2000). Cortical representation of pain:
functional characterization of nociceptive areas near the lateral sulcus. Pain, 87, 113–9.
Tumeh, PC, Alavi, A, Houseni, M, et al. (2007). Structural and functional imaging correlates for age-
related changes in the brain. Seminars in Nuclear Medicine, 37, 69–87.
Turk, DC, Meichenbaum, D & Genest, M (1983). Pain and behavioral medicine: A cognitive-behavioral
perspective, Guilford Press, New York.
Turner, JA, Mancl, L & Aaron, LA (2004). Pain-related catastrophizing: a daily process study. Pain,
110, 103–11.
Valet, M, Gundel, H, Sprenger, T, et al. (2009). Patients with pain disorder show gray-matter loss in
pain-processing structures: a voxel-based morphometric study. Psychosomatic Medicine, 71, 49–56.
Vlaeyen, JW & Linton, SJ (2000). Fear-avoidance and its consequences in chronic musculoskeletal pain:
a state of the art. Pain, 85, 317–32.
Waddell, G, McCulloch, JA, Kummel, E & Venner, RM (1980). Nonorganic physical signs in low-back
pain. Spine, 5, 117–25.
Chapter 7
Structural Brain Changes in Patients

with Chronic Back Pain
Arne May and A. Vania Apkarian
7.1 Introduction
In the past decade of pain research, a network of pain transmitting areas within the CNS has been
established, based on both animal studies (Wall and Melzack 2006) and findings from functional
imaging studies in humans (Peyron et al. 2000). Consequently, the neurobiology of pain is
increasingly understood as an integration of activity in distinct neuronal structures. Evidence of
altered local brain chemistry (Grachev et al. 2000) and functional reorganization in chronic back
pain (CBP) patients (Flor et al. 1997) supports the idea that chronic pain could be understood not
only as an altered functional state, but also as a consequence of central plasticity (Flor 2003).
Recent neurobiological findings suggest cortical reorganization on a functional level (Grusser
et al. 2004). For example, amputation of a limb is very often accompanied by phantom pain. In
these patients, the deafferentation leads to cortical reorganization where the representational
fields of adjacent areas move into the into the representation zone of the deafferented limb (Pons
et al. 1991; Flor et al 2006). This ‘functional reorganization’ was not only detected in patients suf-
fering from phantom limb pain (Flor et al. 1995), but also in CBP patients (Flor et al. 1997).
Regarding CBP, increased cortical activity and a shift of the cortical representation of the back,
which was interpreted as an expansion of the back’s representation into the neighbouring foot
and leg area (Flor et al. 1997), was found. In patients with chronic regional pain syndrome (CRPS
Type I), a shrinkage of the representational field of the affected arm was found and the extent of
shrinkage correlated highly with the intensity of pain and the magnitude of mechanical hyperal-
gesia (Maihofner et al. 2003; Pleger et al. 2004). It is noteworthy, that the functional changes in
CRPS (Maihofnere et al. 2004) and in phantom pain (Flor et al. 2001) were dynamic, (i.e. cortical
reorganization reversed coincident with clinical improvement).
Currently, chronic pain states are attributed to abnormal nociceptive/antinociceptive function
on different levels of the neuroaxis (Wall and Melzack 2006) with a normal brain structure.
However, any significant challenge that requires a specific function, including learning a specific
task, has the potential to alter brain structure (May et al. 2007). Given that the initiation of
chronification of pain involves nociceptive input, one would expect that neuroplasticity would
probably occur in modulatory areas of nociception—namely, the antinociceptive system.
An alternative view of the cumulative human brain imaging studies regarding pain is that we
have firmly established that various brain regions are activated in a reproducible pattern for acute
and experimental pain conditions. Yet, brain activity for chronic pain is distinct from acute pain
(Apkarian et al. 2005). Moreover, the ‘pain matrix’ assumed to underlie pain in general in fact
corresponds to spinothalamic inputs to the cortex and as such identifies brain areas involved in
acute nociception. Even for acute pain, the involvement of primary somatosensory cortex
(S1) remains unresolved. Multiple evoked potential studies assert its role while functional brain
imaging, primarily functional magnetic resonance imaging (fMRI), studies often fail to find
activity in appropriate parts of S1, however, with some exception (Bingel 2004). Animal studies
have shown that the S1 body map is dynamic and reorganizes with body part use or neglect, as
well as with peripheral nerve injuries. Recently, this was also shown for humans (Flor et al. 1995,
1997, 2002). Thus, changes in somatotopy in S1, which have been described in various chronic
pain conditions and which were determined mainly for non-painful tactile stimuli, may be
a consequence of behavioural modifications of body use due to coping strategies necessary for
living with chronic pain.
There is now accumulating evidence that brain activity for acute and chronic pain (especially
for fluctuations of spontaneous pain) are distinct from each other, and that brain activity for
various clinical chronic pain conditions are also distinct from each other (Apkarian et al. 2009).
For example, in CBP patients, activity in the medial prefrontal cortex (mPFC) identifies the
intensity of patients’ pain with a predictive power larger than 80% (Baliki et al. 2006). When
in the same patients brain activity for acute thermal pain is examined, applied on the back at a
location closest to the back pain, the mPFC activity show no related activity, and instead parts of
the insula code the perceived intensity at a comparable predictive level as the mPFC for spontane-
ous pain. Moreover, the brain activations for acute thermal pain in CBP engages the same brain
regions as seen in healthy subjects, and contrasting these activations between the two groups
shows no brain region to be more active or less active in either group, thus demonstrating that
acute thermal pain brain activity does not differ between CBP and normals, and spontaneous
pain of CBP activates distinct pattern.
Brain activity for spontaneous pain was also reported for patients with chronic postherpetic
neuralgia (PHN) (Geha et al. 2007), a prototypical neuropathic chronic pain condition. The PHN
patients were studied in the same lab as CBP patients, using the same methods and similar analyses,
thus the result can be directly compared. In PHN spontaneous pain activated a more elaborate set
of brain regions, but most importantly parts of the basal ganglia and amygdala, and these regions
best reflected the intensity as well as multiple questionnaire based properties of PHN pain. It
therefore seems that brain activity for spontaneous pain engages distinct brain regions in different
chronic pain conditions. We should add, however, that the mPFC is a brain region tightly con-
nected to basal ganglia and amygdala, and in the CBP patients’ study these structures were also
activated but at a lower statistical level.
The PHN patients’ study also examined the effects of a therapy on brain activity, and the same
therapy effects were also recently reported in CBP patients for their spontaneous pain (Baliki et al.
2008). Both studies examined the effects of 5% lidocaine patch application on the painful body
part (a sodium channel blocker that should only be effecting neural transmission for the tissue
just underneath the patch), and correlated the change in pain perception with that of brain activ-
ity determined for spontaneous pain. In both studies decreased pain perception was accompanied
with decreased brain activity, and in each group the brain regions involved were distinct and
corresponded to those best reflecting spontaneous pain, that is, lidocaine therapy decreased
activity in mPFC in CBP patients while in PHN patients it decreased activity in basal ganglia
and amygdala. These results and similar accumulating evidence (Apkarian et al. 2009) question
the utility of the notion of the ‘pain matrix’ and imply that chronic pain conditions need to be
investigated as distinct entities with unique brain properties.
7.2 Structural brain imaging

In the past, studies of brain morphology completely depended on autopsy material. Fortunately,
chronic pain is not a life limiting factor, and given that the common assumption of chronic pain
STRUCTURAL BRAIN CHANGES IN PATIENTS WITH CHRONIC BACK PAIN 107
has been that it is simply pain sustained over a long time, there was no urge to examine the
anatomy of autopsies of patients suffering from chronic pain. This situation changed with the
advent of modern in vivo imaging methods, in particular magnetic resonance (MR) imaging. All
quantitative MR-based methods are subsumed under the heading of MR morphometry of the
brain and are based on the idea of using a common coordinate system or atlas. Normally, three-
dimensional, high-resolution, T1-weighted MR images acquired with conventional 1.5 T MR
scanners and 1 mm3 voxels provide sufficient detail and contrast (Ashburner et al. 2003) whereas
scanners with higher field strength (3 Tesla, 7 Tesla) allow for higher spatial resolution. Images
from several subjects can be grouped together and analysed by mapping them onto a standardized
coordinate space. Voxel-based morphometry (VBM) is the most commonly applied technique. It
is relatively simple to use, has moderate demands on computational resources and is available in
common software packages like FSL (FSL;, Smith et al. 2004) or SPM (Friston et al. 1999).
As a non-invasive procedure, MR morphometry is the ideal tool for finding the morphological
substrates of diseases, deepening our understanding of the relationship between brain structure
and function, and even to monitor therapeutic interventions. It has to be said that most studies
focus on cohort or cross-sectional studies, leading to the question of cause and consequence.
Important contributions to the exact causes of structural changes will come from studies that
look at the time parameters of these changes and include independent factors (i.e. electrophysiol-
ogy or genetics). Moreover, the exact cause of lesion- and training-related morphological
changes in the adult brain is still not known. A decrease in the brain grey matter, such as has been
described in chronic pain patients, does not necessarily mean neuronal destruction. Potential
correlates of the observed morphometric changes include a simple change in cell size, shrinkage
or atrophy of neurons or glia, as well as changes in the intra-cortical axonal architecture (synaptic
loss) (May and Glaser 2006).
The analysis of the cortical surface complements voxel-based methods. For these approaches,
the cortical surface is extracted from brain scans and further computations are applied to then
calculate parameters such as cortical thickness (Fjell et al. 2006) or complexity (Thompson et al.
2005). More recently, this has also allowed the local three-dimensional computation of the gyri-
fication index as a measure of the degree of folding of a given cortical area (Luders et al. 2006). An
advantage over VBM and DBM is the improved reduction of intersubject variability in cortical
folding patterns (Argall et al. 2006). However, the extraction of the cortical surface imposes high
demands on computational resources and crucially depends on the quality of surface extraction,
which sometimes requires additional manual correction or interaction.
A disadvantage of all of the methods described is that they are insensitive to changes in white
matter, where diffusion tensor imaging (DTI) is the method of choice. DTI takes advantage of
water diffusion along axons and allow the study of the integrity and structural connectivity of the
white matter (Beaulieu 2002). By fitting a diffusion tensor model to DTI measurements at each
voxel, one can measure fractional anisotropy (FA) and its three principle diffusivities (eigenvalues
l1, l2, and l3), all of which characterize the microenvironment of the white matter tissue (Beaulieu
2002; Buchel et al. 2004). With the development of novel computational techniques during the
last few years and increasing image resolution, the era of MRI-based morphometry has begun to
expand. Recent findings and further methodological developments should lead to scientific
breakthroughs that will change how we think of the brain (May 2011).
7.3 Structural imaging in chronic back pain

The mechanisms of chronification are the subject of intense research and debate and several
studies have investigated the possible functional changes in back pain (Grachev et al. 2000;
Apkarian et al. 2001, 2004a). As sophisticated structural brain imaging methods became available,
the question emerged whether the brain of patients experiencing chronic back pain (CBP) may be
different from healthy controls. The pioneering work by Apkarian et al. used VBM methods in
17 patients with CBP and demonstrated a significant but spatially restricted brain atrophy in these
patients. The authors were again able to show that these morphological changes are correlated
with the clinical parameters of the condition (duration of pain) and suggested that the patho-
physiology of chronic pain includes thalamo-cortical processes ( Apkarian et al. 2004b ).
Specifically, this study found a decrease in grey matter in the dorsolateral prefrontal cortices
(DLPFC) (a region outside of the spinothalamic projections to the cortex) bilaterally and a
decrease in grey matter in the right thalamus (implicating involvement of spinothalamic projec-
tions). The authors suggested that neurodegeneration—rather than tissue shrinkage—without a
substantial impact on neuronal properties may be the cause of this finding, based on an earlier
observation that N-acetyl-aspartate concentrations are decreased in DLPFC in back pain (Grachev
et al. 2000).
The Apkarian et al. study was in part replicated by Schmidt-Wilcke et al. (Schmidt-Wilcke et al.
2006), who investigated 18 patients with matched healthy controls and also found a decrease in
grey matter in the DLPFC. However, this study also found an increase in thalamic grey matter and
an additional decrease in the dorsolateral pons and the somatosensory cortex. Interestingly, the
correlation analyses suggested that the grey matter decrease in the brainstem does not correlate
with pain duration, but rather with pain intensity and pain unpleasantness experienced at
the time of scanning, thus possibly accounting for the degree of impaired antinociception at that
time. The differences between the studies by Apkarian et al. and Schmidt-Wilcke et al. are
not easy to interpret. One explanation is the relatively small sample sizes used in these studies, so
that negative findings have little meaning. However, Schmidt-Wilcke et al. only included patients
without radiating pain, including radiculopathy, whereas Apkarian studied a mixture of patients
with and without neurological manifestations as well as patients with pain outside of his region
(for example, in the upper back). Moreover, in the Apkarian study the clinical properties of back
pain could be related to the extent of grey matter decrease in DLPFC: the amount of this regional
decrease was significantly larger for back pain with radiculopathy in contrast to back pain without
radiculopathy. In a subsequent study where brain activity for spontaneous pain was studied in
back pain patients and mPFC activity identified, the authors demonstrated that activity in DLPFC
was negatively correlated with that of mPFC (Baliki et al. 2006), leading to the suggestion that the
anatomical changes in DLPFC may underlie the increased activity in mPFC and thus lead to a
more emotional perception of the pain of back pain. In summary, chronic back pain is accompa-
nied by specific morphological alterations in structures known to play a crucial role in either
antinociception or in the control of complex responses to the pain specifically in emotional control
for back pain, which may correlate to the intensity and unpleasantness of pain (May 2011).
7.4 Are the brain changes in chronic back pain specific?

The morphometric data in CBP patients have to be seen in relation to nearly a dozen morpho-
metric studies done in different chronic pain diseases (May 2008). All studies showed local
morphologic alterations of the brain in areas ascribable to the transmission of pain in patients
with phantom pain (Draganski et al. 2006), irritable bowl syndrome (Davis 2000), fibromyalgia
(Kuchinad et al. 2007; Schmidt-Wilcke et al. 2007), tension type headache (Schmidt-Wilcke et al.
2005), migraine (Granziera et al. 2006; Rocca et al. 2006; Kin et al. 2008; Schmidt-Wilcke et al.
2008; Valfre et al. 2008), chronic vulvar pain (Schweinhardt et al. 2008), chronic complex regional
pain syndrome (CRPS) (Geha et al. 2008), and chronic pain following thoracic spinal damage
(Wrigley et al 2008). These alterations were different for each pain syndrome, but most studies
showed a decrease in grey matter in the cingulate cortex, the parts of the prefrontal cortex and
the anterior insula (May 2008). These regions are suggested to function as multi-integrative
structures during the experience and the anticipation of pain.
As neither pain related inactivity (Draganski et al. 2006), nor pain medication (Apkarian et al.
2004b; Schmidt-Wilcke et al. 2005; Kuchinad et al. 2007) explain these findings, the in vivo
demonstration of a loss of brain grey matter in patients experiencing chronic pain compared to
age- and sex-matched healthy controls could represent the heavily discussed neuroanatomical
substrate for pain memory. These alterations were different for the different pain syndromes, but,
in terms of functional systems, overlapped to an astounding extent. It is, however, crucial to stress
that this phenomenon, which coincides with the chronicity of pain does not act in isolation.
It influences and is influenced by nociceptive and spinal events. The plasticity and dynamic inter-
action among nociceptors (Woolf and Salter 2000), neurotransmitters (Koltzenburg 1999), glial,
neuronal and endothelial cells (Gordh et al. 2006), receptive fields (Suzuki et al. 2000) and
the immune system (Marchand et al. 2005) as well as higher cognitive functions (Flor et al.
2002; Gureje 2007) needs to be considered. They comprise a complex collection of many discrete,
but interacting, systems (Melzack et al. 2001). However, although the complexity of pain percep-
tion involves many levels of the neuraxis, cortical plasticity is certainly not restricted to functional
changes.
The data indicating that the cingulate cortex is affected in chronic pain raises some speculation
on the neuronal pathways of chronic pain. The anterior cingulate cortex and the anterior insulae
function as multi-integrative structures during the experience and the anticipation of pain
(Peyron et al. 2000; Porro et al. 2002). The anterior cingulate cortex, including the perigenual
part, is of particular interest, since it plays a deterministic role in pain modulation and analgesia.
The analgesic/modulating effect is mediated through the interaction with other structures, such
as the orbitofrontal cortex, the amygdala and the PAG (Zhang et al. 1997). All of these structures
also showed alteration in some of the studies mentioned here. The collective evidence for the
rACC as a crucial site for endogenous pain control has been observed in the context of pain
modulation by attention, anticipation of pain and placebo analgesia (for a review see (Kupers
et al. 2005) and also studies of neurostimulation for pain relief (Davis 2006)). A recent study
by Apkarian and colleagues raise novel questions as to the role of ACC and of insulae in acute
nociception (Baliki et al 2009). The temporal properties of ACC activity for acute thermal pain
seem to closely correspond to that for the amygdalae, with both activations preceding the peak
thermal stimulus, which is consistent with their role in pain anticipation rather than encoding
its properties.
The insular activity, on the other hand, was segregated into two functionally distinct portions
one activated early and closely related to the stimulus properties (encoding nociceptive input)
while the other was activated later, encoded perceived subjective pain, and was an area that
equally well encoded the size of bars presented visually (representing magnitude estimation).
Thus, depending on the specific part of the insula showing decreased grey matter one would reach
very different underlying reasons and implications regarding processing pain and other sensory
or cognitive events. Moreover, given the distinct brain activity patterns associated with various
clinical pain conditions, the interaction between brain morphology, morphological reorganiza-
tion and brain activity needs to be studied far more intensively.
The decrease of grey matter in brain regions which are highly associated with pain suppression
could certainly lead to dysfunction in effective antinociception. Abnormal modulation of
brain nociceptive systems, at first transient, but becoming permanent with continuing illness,
could in part explain the shift from acute to chronic pain. Very recently, and using a different
methodological approach, Davis et al. were able to show that patients suffering from irritable
bowel syndrome (IBS) show a significant cortical thinning of the ACC, a finding which has strik-
ing similarity to the findings discussed above (Davis et al. 2008).
7.5 Relating brain grey matter changes to white

matter connectivity
Grey matter morphological changes in back pain (Apkarian et al. 2004) and in CRPS (Geha et al.
2008) were studied in the same lab using similar imaging and data analysis techniques. The results
indicate that the two clinical conditions affect non-overlapping brain regions, and yet in both the
changes are correlated with duration and/or intensity of the pain. In the CRPS patients, DTI
analyses were used to examine relationship between grey matter decreased density and white
matter connectivity. The study generally indicates that brain regions where grey matter is reduced
is also accompanied with a general decrease in white matter connectivity, although in some cases
this was also accompanied with target specific increased connectivity as well. This is the first study
linking grey matter changes to white matter properties and the results are consistent with the
general idea of loss of neurons, but also suggest that it is a result of competitive reorganization of
connectivity across brain regions. More importantly this study demonstrates the power of com-
bining various brain anatomical imaging techniques to begin to unravel the processes underlying
brain reorganization with chronic pain.
7.6 Brain changes in chronic pain: cause or consequence?

As chronic pain patients may have a common ‘brain signature’ in areas known to be involved in
pain control, the question arises whether the central reorganization processes in chronic pain
syndromes could involve a ‘degeneration’ of specific brain areas. This question is not redundant,
as a degenerative process is irreversible. Although many VBM studies demonstrate changes in
grey matter, the neurobiological basis of these structural alterations on a microscopic level are not
well defined (May and Gaser 2006). VBM detects changes in grey matter concentration per voxel
as well as changes in the classification of individual voxels (e.g. from white to grey matter (Good
et al. 2001) and probably a combination of both). In general, a decrease in grey matter could be
due to a simple decrease in cell size, neural or glial cell apoptosis, a decrease in spine density or
even changes in blood flow or interstitial fluid. Unfortunately, all available studies compared
cohorts of patients and therefore no statement regarding dynamic changes can be made.
It is not understood why only a relatively small proportion of humans develop a chronic pain
syndrome, considering that pain is a universal experience. As the adult human brain may change
its structure in response to environmental demands (Flor et al. 2001; Draganski et al. 2004), the
question arises whether in some humans a (possibly genetically) structural difference in central
pain transmitting systems may act as a diathesis for chronic pain. What argues against the assump-
tion that the people with chronic pain have a priori a different brain structure and thus
are susceptible to chronicity is the fact, that people experiencing phantom pain and chronic
pain following spinal cord damage show some similar brain changes as, for example, people with
CBP. Moreover, a recent study showed in patients with chronic hip pain that these changes are
in part reversible (Rodriguez-Raecke et al. 2009). On the other hand the opposite argument has
just as convincing data, namely, brain activity for spontaneous pain seems distinct for diverse
clinical chronic pain conditions and brain morphometric changes, especially between CRPS and
back pain.
A recent study aimed to investigate how repeated painful stimulation over several days is proc-
essed, perceived and modulated in the healthy human brain and whether repeated painful stimu-
lation may lead to structural changes of the brain. A standardized nociceptive heat paradigm for
8 consecutive days is reflected in increased grey matter in brain areas ascribable to the transmis-
sion of sensory input and pain perception (Teutsch et al. 2008). This data is in line with most
morphometric studies investigating structural brain plasticity as a result of exercise and learning
(Flor et al. 2001; Gaser et al. 2003; Draganski et al. 2004; Boyke et al. 2008). Moreover, these find-
ings of structural changes follow the previously described functional pattern (Bingel et al. 2007)
precisely (i.e. a significant change during the protocol that reverses to pre-stimulation levels at the
fourth time-point, namely, after 1 year (Bingel et al. 2008).
However, it is an intriguing fact that chronic pain patients have constant pain, but seem not to
develop an increase in grey matter in somatosensory areas (in fact the bulk of the data points to
decreased grey matter in areas outside of the somatosensory regions), although several studies
showed that exercise is accompanied by an increase of grey matter in the regions which are
specific for the respective task (for review see (Friston et al. 1996)).
One explanation for this lack of grey matter increase in chronic pain patients is that they do
not have a significant noxious input to explain the pain experience (any more, although there is
no real convincing data for this notion; for spontaneous pain of back pain a transient afferent
nociceptive input has been described (Baliki et al. 2006)). In that case the experience of constant
pain would be mostly driven by the brain itself and the afferent (peripheral noxious) input is no
longer needed for this experience. Another possibility is, that a given task-specific exercise will
only increase grey matter in corresponding brain areas until the task is learned and that this
change recedes once the task is learned sufficiently.
There is no conclusive data regarding the cause or the consequence of the different cortical and
subcortical morphological changes, although the correlation of pain duration and degree of grey
matter decrease in most studies suggests that the morphological changes are at least in part
secondary to constant pain. This notion is also consistent with the evidence of the same brain
region showing decreased metabolite concentration as well as decreased white matter connectiv-
ity for the same regions that also show decreased grey matter density. Still, thorough longitudinal
studies need to address whether the morphological changes reverse when the disproportionate
amount of nociceptive stimulation (i.e. following sufficient pain treatment) stops. The idea of
structural maladaptive plasticity however, serves certainly well as a good model for structural
cortical/subcortical reorganization following chronic input of nociceptive information.
References
Apkarian AV, Krauss BR, Fredrickson BE, Szeverenyi NM. Imaging the pain of low back pain: functional
magnetic resonance imaging in combination with monitoring subjective pain perception allows the
study of clinical pain states. Neurosci Lett 2001; 299(1–2):57–60.
Apkarian AV, Sosa Y, Krauss BR, et al. Chronic pain patients are impaired on an emotional decision-
making task. Pain 2004; 108(1–2):129–36.
Apkarian AV, Sosa Y, Sonty S, et al. Chronic back pain is associated with decreased prefrontal and thalamic
gray matter density. J Neurosci 2004; 24(46):10410–5.
Apkarian AV, Bushnell MC, Treede RD, Zubieta JK. Human brain mechanisms of pain perception and
regulation in health and disease. Eur J Pain 2005; 9(4):463–84.
Argall BD, Saad ZS, Beauchamp MS. Simplified intersubject averaging on the cortical surface using SUMA.
Hum Brain Mapp 2006; 27(1):14–27.
Apkarian AV, Baliki MN, Geha PY. Towards a theory of chronic pain. Prog Neurobiol 2009; 87(2):81–97.
Ashburner J, Csernansky JG, Davatzikos C, Fox NC, Frisoni GB, Thompson PM. Computer-assisted
imaging to assess brain structure in healthy and diseased brains. Lancet Neurol 2003; 2(2):79–88.
Baliki MN, Chialvo DR, Geha PY, et al. Chronic pain and the emotional brain: specific brain activity
associated with spontaneous fluctuations of intensity of chronic back pain. J Neurosci 2006; 26(47):
12165–73.
Baliki MN, Geha PY, Jabakhanji R, Harden N, Schnitzer TJ, Apkarian AV. A preliminary fMRI study of
analgesic treatment in chronic back pain and knee osteoarthritis. Mol Pain 2008; 4:47.
Baliki MN, Geha PY, Apkarian AV. Parsing pain perception between nociceptive representation and
magnitude estimation. J Neurophysiol 2009; 101(2):875–87.
Beaulieu C. The basis of anisotropic water diffusion in the nervous system - a technical review. NMR
Biomed 2002; 15:435–55.
Bingel U, Lorenz J, Glauche V, et al. Somatotopic organization of human somatosensory cortices for pain:
a single trial fMRI study. Neuroimage 2004; 23(1):224–32.
Bingel U, Herken W, Teutsch S, May A. Habituation to painful stimulation involves the antinociceptive
system—a 1-year follow-up of 10 participants. Pain 2008; 140(2):393–4.
Bingel U, Schoell E, Herken W, Buchel C, May A. Habituation to painful stimulation involves the
antinociceptive system. Pain 2007; 131(1–2):21–30.
Boyke J, Driemeyer J, Gaser C, Büchel C, May A. Training induced brain structure changes in the Elderly.
J Neuroscience 2008; 28:7031–35.
Buchel C, Raedler T, Sommer M, Sach M, Weiller C, Koch MA. White matter asymmetry in the human
brain: a diffusion tensor MRI study. Cereb Cortex 2004; 14(9):945–51.
Davis KD. The neural circuitry of pain as explored with functional MRI. Neurol Res 2000; 22(3):313–7.
Davis KD, Pope G, Chen J, Kwan CL, Crawley AP, Diamant NE. Cortical thinning in IBS: implications for
homeostatic, attention, and pain processing. Neurology 2008; 70(2):153–4.
Draganski B, Gaser C, Busch V, Schuierer G, Bogdahn U, May A. Neuroplasticity: Changes in grey matter
induced by training. Nature 2004; 427(6972):311–12.
Draganski B, Moser T, Lummel N, et al. Decrease of thalamic gray matter following limb amputation.
Neuroimage 2006; 31(3):951–7.
Fjell AM, Walhovd KB, Reinvang I, et al. Selective increase of cortical thickness in high-performing
elderly—structural indices of optimal cognitive aging. Neuroimage 2006; 29(3):984–94.
Flor H. The modification of cortical reorganization and chronic pain by sensory feedback. Appl
Psychophysiol Biofeedback 2002; 27(3):215–27.
Flor H. Cortical reorganisation and chronic pain: implications for rehabilitation. J Rehabil Med
2003(41 Suppl):66–72.
Flor H, Elbert T, Knecht S, et al. Phantom-limb pain as a perceptual correlate of cortical reorganization
following arm amputation. Nature 1995; 375(6531):482–4.
Flor H, Braun C, Elbert T, Birbaumer N. Extensive reorganization of primary somatosensory cortex in
chronic back pain patients. Neurosci Lett 1997; 224(1):5–8.
Flor H, Denke C, Schaefer M, Grusser S. Effect of sensory discrimination training on cortical
reorganisation and phantom limb pain. Lancet 2001; 357(9270):1763–4.
Flor H, Knost B, Birbaumer N. The role of operant conditioning in chronic pain: an experimental
investigation. Pain 2002; 95(1–2):111–18.
Flor H, Nikolajsen L, Staehelin Jensen T. Phantom limb pain: a case of maladaptive CNS plasticity?
Nat Rev Neurosci 2006; 7(11):873–81.
Friston KJ, Holmes AP, Ashburner J, Poline J-B. SPM99. World Wide Web 1999; http://www.fil.ion.ucl.
ac.uk/spm
FSL: http://www.fmrib.ox.ac.uk/fsl/
Gaser C, Schlaug G. Brain structures differ between musicians and non-musicians. J Neurosci 2003;
23(27):9240–5.
Geha PY, Baliki MN, Chialvo DR, Harden RN, Paice JA, Apkarian AV. Brain activity for spontaneous pain
of postherpetic neuralgia and its modulation by lidocaine patch therapy. Pain 2007; 128(1–2):88–100.
Geha PY, Baliki MN, Harden RN, Bauer WR, Parrish TB, Apkarian AV. The brain in chronic CRPS pain:
abnormal gray-white matter interactions in emotional and autonomic regions. Neuron 2008;
60(4):570–81.
Good CD, Johnsrude IS, Ashburner J, Henson RN, Friston KJ, Frackowiak RS. A voxel-based
morphometric study of ageing in 465 normal adult human brains. Neuroimage 2001; 14(1 Pt 1):21–36.
Gordh T, Chu H, Sharma HS. Spinal nerve lesion alters blood-spinal cord barrier function and activates
astrocytes in the rat. Pain 2006; 124(1–2):211–21.
Grachev ID, Frederickson BE, Apkarian AV. Abnormal brain chemistry in chronic back pain; an in vivi
proton magnetic resonance spectroscopy study. Pain 2000; 89:7–18.
Granziera C, Dasilva AF, Snyder J, Tuch DS, Hadjikhani N. Anatomical alterations of the visual motion
processing network in migraine with and without aura. PLoS Med 2006; 3(10):e402.
Grusser SM, Muhlnickel W, Schaefer M, Villringer K, Christmann C, Koeppe C, Flor H. Remote
activation of referred phantom sensation and cortical reorganization in human upper extremity
amputees. Exp Brain Res 2004; 154(1):97–102.
Gureje O. Psychiatric aspects of pain. Curr Opin Psychiatry 2007; 20(1):42–6.
Kim JH, Suh SI, Seol HY, et al. Regional grey matter changes in patients with migraine: a voxel-based
morphometry study. Cephalalgia 2008; 28(6):598–604.
Koltzenburg M. The changing sensitivity in the life of the nociceptor. Pain 1999; Suppl 6:S93–102.
Kuchinad A, Schweinhardt P, Seminowicz DA, Wood PB, Chizh BA, Bushnell MC. Accelerated brain gray
matter loss in fibromyalgia patients: premature aging of the brain? J Neurosci 2007; 27(15):4004–7.
Kupers R, Faymonville ME, Laureys S. The cognitive modulation of pain: hypnosis- and placebo-induced
analgesia. Prog Brain Res 2005; 150:251–69.
Luders E, Thompson PM, Narr KL, Toga AW, Jancke L, Gaser C. A curvature-based approach to estimate
local gyrification on the cortical surface. Neuroimage 2006; 29(4):1224–30.
Maihofner C, Handwerker HO, Neundorfer B, Birklein F. Patterns of cortical reorganization in complex
regional pain syndrome. Neurology 2003; 61(12):1707–15.
Maihofner C, Handwerker HO, Neundorfer B, Birklein F. Cortical reorganization during recovery from
complex regional pain syndrome. Neurology 2004; 63(4):693–701.
Marchand F, Perretti M, McMahon SB. Role of the immune system in chronic pain. Nat Rev Neurosci
2005; 6(7):521–32.
May A. Chronic pain may change the structure of the brain. Pain 2008; 137(1):7–15.
May A. Structural Brain imaging: A window into chronic pain. The Neuroscientist 2011; 17:209–20.
May A, Gaser C. Magnetic resonance-based morphometry: a window into structural plasticity of the brain.
Curr Opin Neurol 2006; 19(4):407–11.
May A, Hajak G, Ganssbauer S, et al. Structural brain alterations following 5 days of intervention: dynamic
aspects of neuroplasticity. Cereb Cortex 2007; 17(1):205–10.
Melzack R, Coderre TJ, Katz J, Vaccarino AL. Central neuroplasticity and pathological pain. Ann N Y Acad
Sci 2001; 933:157–74.
Peyron R, Laurent B, Garcia-Larrea L. Functional imaging of brain responses to pain. A review and
meta-analysis (2000). Neurophysiol Clin 2000; 30(5):263–88.
Pleger B, Tegenthoff M, Schwenkreis P, et al. Mean sustained pain levels are linked to hemispherical
side-to-side differences of primary somatosensory cortex in the complex regional pain syndrome I.
Exp Brain Res 2004; 155:115–19.
Pons TP, Garraghty PE, Ommaya AK, Kaas JH, Taub E, Mishkin M. Massive cortical reorganization after
sensory deafferentation in adult macaques. Science 1991; 252(5014):1857–60.
Porro CA, Baraldi P, Pagnoni G, et al. Does anticipation of pain affect cortical nociceptive systems?
J Neurosci 2002; 22(8):3206–14.
Rocca MA, Ceccarelli A, Falini A, et al. Brain gray matter changes in migraine patients with T2-visible
lesions: a 3-T MRI study. Stroke 2006; 37(7):1765–70.
Rodriguez-Raecke R, Niemeier A, Ihle K, Ruether W, May A. Brain gray matter decrease in chronic pain is
the consequence and not the cause of pain. J Neurosci 2009; 29(44):13746–50.
Schmidt-Wilcke T, Leinisch E, Straube A, et al. Gray matter decrease in patients with chronic tension type
headache. Neurology 2005; 65(9):1483–6.
Schmidt-Wilcke T, Leinisch E, Ganssbauer S, et al. Affective components and intensity of pain correlate
with structural differences in gray matter in chronic back pain patients. Pain 2006; 125(1–2):89–97.
Schmidt-Wilcke T, Luerding R, Weigand T, et al. Striatal grey matter increase in patients suffering from
fibromyalgia - A voxel-based morphometry study. Pain 2007; 132(Suppl 1):S109–16.
Schmidt-Wilcke T, Ganssbauer S, Neuner T, Bogdahn U, May A. Subtle grey matter changes between
migraine patients and healthy controls. Cephalalgia 2008; 28(1):1–4.
Schweinhardt P, Kuchinad A, Pukall CF, Bushnell MC. Increased gray matter density in young women
with chronic vulvar pain. Pain 2008; 140(3):411–19.
Smith SM, Jenkinson M, Woolrich MW, et al. Advances in functional and structural MR image analysis
and implementation as FSL. Neuroimage 2004; 23 Suppl 1:S208–19.
Suzuki R, Kontinen VK, Matthews E, Williams E, Dickenson AH. Enlargement of the receptive field size to
low intensity mechanical stimulation in the rat spinal nerve ligation model of neuropathy. Exp Neurol
2000; 163(2):408–13.
Teutsch S, Herken W, Bingel U, Schoell E, May A. Changes in brain gray matter due to repetitive painful
stimulation. Neuroimage 2008; 42(2):845–9.
Thompson PM, Lee AD, Dutton RA, et al. Abnormal cortical complexity and thickness profiles mapped in
Williams syndrome. J Neurosci 2005; 25(16):4146–58.
Valfre W, Rainero I, Bergui M, Pinessi L. Voxel-based morphometry reveals gray matter abnormalities in
migraine. Headache 2008; 48(1):109–17.
Wall P, Melzack R. Textbook of pain. Edinburgh, Churchill Livingstone, 2006.
Woolf CJ, Salter MW. Neuronal plasticity: increasing the gain in pain. Science 2000; 288(5472):1765–9.
Wrigley PJ, Gustin SM, Macey PM, et al. Anatomical changes in human motor cortex and motor pathways
following complete thoracic spinal cord injury. Cereb Cortex 2008; 19(1):224–32.
Zhang YQ, Tang JS, Yuan B, Jia H. Inhibitory effects of electrically evoked activation of ventrolateral
orbital cortex on the tail-flick reflex are mediated by periaqueductal gray in rats. Pain 1997;
72(1–2):127–35.
Chapter 8
The Psychophysiology of Chronic

Back Pain Patients
Kati Thieme and Richard H. Gracely
8.1 Introduction
Psychophysiological methods investigate peripheral and central physiological responses pro-
voked by stressors, cognitions, emotions, and behaviours. These methods involve physiology,
functional anatomy, and neuroscience as well as general and physiological psychology. Studies of
medical conditions, including cardiology, psychiatry, sleep, and chronic pain, use psychophysio-
logical methods to determine mechanisms mediating diseases and treatments. In contrast to
simple physiological theories, pain can be considered to be a modulated complex of physiological
and psychological processes (Birbaumer and Schmidt 1996). These processes are modulated by
factors such as sensitization and both classical and operant conditioning. While the psychophysi-
ological experiments of the past focused on one selected physiological system such as muscle
tension, present studies record a complex of interrelated physiological systems. The goal of
the multimethod assessment approach is to determine individual-specific stress reactivity and
symptom-specific stress responses that may initiate and maintain pathological processes.
Psychophysiological pain treatment includes biofeedback (e.g. Guzman et al. 2007 ; Ostelo
et al. 2005), deep brain stimulation (IBS; Coffey et al. 2001), and brain–computer interface (BCI;
Birbaumer and Flor 1999; Birbaumer et al. 1999a; DeCharms et al. 2005). These are usually
combined with behavioural-oriented psychological treatments such as operant and cognitive-
behavioural therapies (e.g. Glombiewski et al. 2010; Newton-Jon 1995). Common modalities for
biofeedback include surface electromyography, respiration rate and depth, skin surface tempera-
ture, cardiovascular reactivity, and electrodermal response. Clinical biofeedback therapy broadly
involves the direct feedback learning model as a therapeutic stress-management biofeedback
model, which emphasizes the need to understand each patient as an individual. EMG-biofeedback
has been used for the treatment of low back pain, tension headache, temporomandibular
joint disorders, migraine, brain blood flow, temperature-biofeedback for migraine, and Raynaud
syndrome.
Electric neurostimulation includes approved and investigational therapies directed at the spinal
cord, thalamus, periaqueductal, or periventricular grey (PAG or PVG) matter, motor cortex, and
peripheral nerves and has provided short-term relief for untreatable deafferentation and cancer
pain (Coffey et al. 2001). Although preliminary studies support the utility of neurostimulation
methods, randomized control trials with extended follow-up periods are needed. The interactive
method of BCI enables individuals to gain voluntary control over activation in a specific brain
region such as anterior cingulate cortex. These effects can be powerful, providing relief for severe,
chronic clinical pain (DeCharms et al. 2005), possible because the effects of psychophysiological
methods extend beyond palliative coping to the actual amelioration of the pathological mecha-
nisms mediating pain disease (Birbaumer and Schmidt 1996).
8.1.1 Biomechanical models

Back pain may result from physically demonstrable chronic back pain (CBP) disease such as:
(1) degenerative disorders with degeneration of the intervertebral disc(s), spondylolisthesis and
osteoarthritis (Hirsch 1966); (2) structural abnormalities with spina bifida occulta, hyperlordosis,
kyphosis, scoliosis (Cailliet 1981); and (3) muscular and ligamentous dysfunctions with fibrositis or
ligamentous ruptures associated with muscle spasm and muscle degeneration (Jinkins 2003).
These degenerative processes, structural abnormalities and ligamentous ruptures have often been
considered to be pathological entities and possible causal factors for back pain (e.g. Cailliet 1981;
Chataigner et al. 1998, Toyone et al. 1994). However, these physical findings can be found also in
individuals who do not report back pain and accumulating evidence does not support the role of
biomechanical factors as the sole relevant cause of pain (e.g. Beattie and Meyers 1998; Hult 1954;
Jinkins 2004,). For example, the physical findings of muscle spasm and atrophy following
ligamentous ruptures are usually viewed as a consequence, not a cause of the pain, spinal derange-
ments, or postural faults (e.g. Jinkins 2003).
The biomechanical model proposes that CBP may be a result of abnormally low levels of muscle
activity during movement or right-left movement asymmetries revealed by electromyographic
(EMG) recording. In this model, abnormal EMG activity patterns result from poor posture and
guarding that develops in response to the original insult (Cram and Steger 1983). Proponents of
this model believe that irregular muscle activity provides abnormal support for the spine, which
in turn becomes unstable and this instability facilitates infringement upon nerve endings and
resultant pain (Ahern et al. 1988; Dolce and Raczynski 1985; Wolf et al. 1982). There is some
experimental support for the biomechanical model. For example, patients with CBP have been
found to display lower EMG levels compared to controls during movement (Ahern et al. 1988;
Collins et al. 1982, Wolf et al. 1982). In addition, greater EMG asymmetries for CBP patients have
been found compared to controls reported by Cram and Steger (1983). Greater asymmetries have
been observed while sitting, but not while standing (Hoyt et al. 1981) or during dynamic move-
ments (Ahern et al. 1988). Other studies contradict the predictions of the biomechanical model
and provide evidence that supports the reflex-spasm model.
The link between pain and muscle activity is the basis of a reflex-spasm model of pain. This
model describes a cycle that typically begins with an injury that produces reflex-muscle activity. If
sustained, this activity produces more pain and more muscle spasms in a self-perpetuating cycle
(Cobb et al. 1975; Collins et al. 1982). Many of the treatments for CBP, such as heat, cold, trans-
cutaneous electrical nerve stimulation, and relaxation training, are thought to provide relief by
interrupting the cycle, reducing both muscle activity and pain.
Two competing reflex-spasm hypotheses of CBP define different aetiological factors. One
emphasizes the importance of physical stressors, while the other focuses on psychophysiological
factors that underlie the development of CBP. Regardless of whether the initial damage is caused
by physical or psychophysiological factors, both hypotheses agree that eventually a reflex spasm
determines pain (Cobb et al. 1975; deVries 1966; Dolce and Raczynski 1985; Miller 1985; Nouwen
and Bush 1984).
Those traditional medical approaches to chronic pain are characterized by a somatosensory
model that focuses on mechanical factors. These perspectives effectively ignore the physical
heterogeneity of CBP, the presence or non-presence of physical injuries, the association and
non-association between physical changes and pain, and varying levels of muscle activity.
These models narrowly consider pain to be a purely sensory event that is assumed to be directly
proportional to peripheral damage. Several factors support the inadequacy of these reflex-spasm
models, in particular the many instances of pain without discernible peripheral damage
THE PSYCHOPHYSIOLOGY OF CHRONIC BACK PAIN PATIENTS 117
(e.g. McCabe et al. 2005; van Dieën et al. 2003) and the clinical reports of the absence of pain
expression despite severe injury (e.g. Beattie and Meyers 1998; Harris 1999; Jensen et al. 1994;
Jinkins 2004).
Medical research tends to disregard the psychosocial antecedents and consequences of CBP
(Flor and Turk 1984). Linton (2000) and Pincus and colleagues (2002), have emphasized in their
reviews the considerable evidence that psychological factors influence CBP. Stress, affective
distress, anxiety, cognitive functioning and pain behaviour significantly impact the magnitude of
CBP (Linton 2000). CBP is also modulated by learning and conditioning. A number of psycho-
biological process-sensitization, cognitive-affective factors (catastrophizing, depression), classical
conditioning, (Gentry and Bernal 1977) and operant conditioning (Flor et al. 2002; Fordyce
1976) influence the interaction among physiological and psychological factors in persistent pain
(Flor et al. 1990).
8.1.2 Psychological models of chronic pain

Psychological factors are implicated when there is no empirical evidence to support an isomor-
phic relationship between the physiological factors such as inflammatory processes, degenerative
changes, structural deformities, traumatic incidents, and muscular or ligamentous strain (Loeser
1980), and CBP (e.g. Harris et al. 1999; Jinkins et al. 2004; McCabe et al. 2005). More recently, the
lack of association between CBP and physical factors has focused attention on the interaction
between physiological and psychological factors, including subconscious learning and cognitive-
affective variables. Flor et al. (1990) proposed that the development and maintenance of chronic
pain is a function of several interacting components: (1) a predisposition to respond with a spe-
cific bodily system; (2) external or internal aversive stimulation; (3) maladaptive information
processing of and coping with pain-related social and/or physiological stimuli, and (4) operant,
respondent, and observational learning processes. These authors defined psychobiological mecha-
nisms that impact the development and maintenance of somatosensory and affective pain memory
in chronic pain patients (see also, Gracely et al. 2002, 2004). The various models of learning and
their relationship to CBP are presented below.
Sensitization
The term ‘sensitization’ has specific meanings in pain physiology and psychology. It is used by
physiologists to describe an increased peripheral and/or central response to a pain stimulus.
Birbaumer and Schmidt (1996), and Flor et al. (1990), have used the term to describe augmented
pain reports and behaviour caused by a range of factors that include operant and classical condi-
tioning, as well as maladaptive pain cognitions such as catastrophizing and helplessness. We use
this latter meaning in this section.
Classical conditioning
Gentry and Bernal (1977) have proposed a respondent model of CBP in which classical condi-
tioning of pain and muscle tension may occur in acute pain states, leading to a pain-tension cycle
and subsequent chronic back pain. The pain may be exacerbated by conditioned fear of move-
ment, leading to avoidance of activity and immobilization, (e.g. Lethem et al. 1983). Some pre-
liminary tests of the assumptions of the respondent model have been conducted with mixed
results. For example, several investigators have reported that in comparison to pain-free partici-
pants, CBP patients demonstrated higher EMG levels in various body positions, or even during
relaxation (e.g. Hoyt et al. 1981; Kravitz et al. 1981) suggesting a diminished recovery of muscle
tension, although decreased lower back EMG levels have been reported in certain body positions
(Collins et al. 1982).
Operant conditioning
The general assumptions of the operant pain model posit that pain, even though originally a
reflex, is modulated through reinforcement controlled by operant conditioning (Fordyce 1976;
1988). The negative consequences of operant learning are ‘pain behaviours’ that become auto-
matic and subconscious through positive (e.g. excessive solicitous spouse response to pain) and
negative (e.g. avoidance of unpleasant activities) reinforcement. Pain behaviours that coincide
with the perception of increased pain intensity provoke the development of pain memory (Flor
et al. 2002). Operant learning increases avoidance behaviour, such as reducing activities in multi-
ple life areas that leads in turn to physical problems such as muscular deconditioning (Smeets
et al. 2006), medication misuse (Turk et al. 1997), muscle tension and immobility (Flor and
Birbaumer 1994), and finally to mental problems such as anxiety and mood disorders (e.g.
Atkinson et al. 1991; Waxman et al. 2008).
Cognitive-behavioural approach
Cognitive-behavioural approaches to chronic pain emphasize the important role of patients’
cognitions (e.g. appraisals, beliefs, expectancies) as mediators between situational stimuli and
emotional and behavioural responses (Turner and Chapman 1982). Among the cognitive varia-
bles related to individual self-statements, catastrophizing, psychosocial distress, anger, and fear have
received considerable attention in the last decade of CBP research (Flor et al. 1997; Gracely et al.
2004; Sullivan et al. 2001; Vlaeyen et al. 1995). Maladaptive self-appraisals are associated with
pain expectation (e.g. Gracely et al. 2004; Sullivan et al. 2001). Patients with chronic pain are
viewed as having negative expectations about their own ability to control certain motor skills such
as performing specific physical activities (e.g. climbing stairs, lifting objects, bending over). They
tend to attribute their disabilities to a reality characterized by loss of control. The pain expectations
are strongly associated with situation-specific self-statements about the inability to exert control
over aversive sensations (Turk et al. 1983) and may reinforce the experience of demoralization,
inactivity, and overreaction to nociceptive stimulation (Flor and Turk 1988). The expectations
have been associated with lower pain tolerance and higher ratings of pain intensity (Gracely et al.
2002, 2004). Cognitions can also reduce pain. From the psychophysiological view, the conviction
of personal control over pain can ameliorate the experience of experimentally induced nociception
by learning to exert deliberate, voluntary control over the activation in the rostral anterior cingu-
late cortex (rACC; deCharms et al. 2005; LaChapelle et al. 2001). Both self-statements like
catastrophizing vs. coping and situation-specific self-statements like loss of control over pain are
associated with anger, fear, and psychosocial distress and can contribute to significant changes in
perceived pain.
Diathesis–stress model
Flor and Turk (1984) proposed a diathesis–stress model of CBP that integrates many of the dis-
parate findings noted above. The model postulates that CBP may result from the interaction of
personally relevant stressful events with a predisposing organic or psychological condition.
Intense or recurrent potentially aversive stimulation in a predisposed individual with maladap-
tive or inadequate coping abilities is hypothesized to lead to extensive and sustained reactions of
the back muscles. The diathesis–stress model suggests that this increased muscle tension might
lead to ischemia, reflex muscle spasms, oxygen depletion, and the release of pain-eliciting sub-
stances (e.g. histamine, substance P). The augmented pain may act as a new stressor provoking a
vicious circle. The development of movement-related anticipatory anxiety and subsequent immo-
bility and the reinforcement of pain behaviours may exacerbate the pain problem. Moreover, the
anticipatory anxiety may produce hyperactivity (muscle tension) and may further exacerbate
nociception and, subsequently, pain.
The diathesis–stress model (Turk and Flor 1984) predicts that CBP patients will exhibit elevated
and prolonged reactions of the back musculature to stress in comparison to healthy controls or
other pain patients. It predicts that muscular reactions to stress will only occur in response to
personally relevant stressors. Additionally, the model hypothesizes that these muscular hyper-
reactions will be predicted best by psychological variables (e.g. anxiety, helplessness) rather than
by organic variables (e.g. amount of degenerative damage, number of surgeries, duration of the
pain problem) and that anxiety levels and immobility are related in CBP patients.
8.1.3 Psychophysiological models

Physiological responses to stress may be partially responsible for the development and mainte-
nance of chronic pain disorders. Traditional models of the impact of stress on physiological func-
tions assumed the principle of a universal stress response, ‘concept of general activation’ (e.g. Selye,
1956). Alternative models present a more specific view. The ‘concept of response specificity’ (Lacey
and Lacey 1958) describes psychophysiological response patterns characterized by a series of
multiple autonomic functions with varied levels of involvement. The autonomic function with
the greatest involvement defines a stress-reactivity that may be important for the aetiopathogen-
esis of the disease.
The ‘concept of response specificity’ (Lacey and Lacey 1958) describes three different psycho-
physiological response patterns. In the first stimulus-specific response pattern, a defined stimulus
provokes a similar response in different individuals. In the second, individual-specific response
pattern, identical psychophysiological response patterns are provoked by different stress stimuli,
for example mental (mental arithmetic), social (exam), physical (heat or cold), or pain stimuli
(electrical, pressure, thermal). In the third ‘motivation-specific response pattern’, different indi-
viduals produce different patterns independent of the type of stimulus. These patterns are consid-
ered to be reproducible if found in a comparable state that uses the same external experimental
setting and provokes comparable internal variables such as motivation and evaluation of the situ-
ation (Ax 1964; Fahrenberg 1986; Lacey et al. 1953).
Perhaps the earliest description of a stimulus-specific response was outlined by James (1884) in
the ‘theory of emotion’. In this theory, special emotions are provoked by specific physiological
response patterns. This theory was supported by Ax (1953) who investigated the emotions anger
and anxiety—highly relevant for CBP patients—as well as their physiological responses. Anger
was associated with a cardiovascular response pattern of increased diastolic blood pressure (BP)
due to a hypothesized higher norepinephrine production. Anxiety was associated with an increase
in heart rate (HR) and systolic BP attributed to higher epinephrine production. This seminal
study was partially replicated by Berntson and co-workers (1998, 2003) who confirmed the car-
diovascular response pattern for anxiety but not the mediating epinephrine hypothesis (Berntson
et al. 1998). This group also provided mechanistic evidence for the role of norepinephrine, anxi-
ety and pharmacological management. They proposed mediating mechanisms that include
noradrenergic projections from the nucleus of the tractus solitarius and the locus coeruleus both
to the amygdala in memory processes, and the basal forebrain in the processing of anxiety-related
information (Berntson et al. 2003). These studies used only cardiovascular variables to define
anxiety or anger response patterns. Other important parameters (i.e. muscle tension, breathing,
or skin conductance level) were not examined. The concept of a stimulus-specific response pat-
tern offers a new approach that describes the interaction between emotion, cardiovascular system
and pain perception. The empirical results supporting the stimulus-specific response pattern
suggest that psychophysiological response patterns are determined mainly by phylogenetically

significant conditions such as fright, sexual arousal, or pain.
The individual-specific response pattern may reflect increased or decreased response sys-
tems. Lacey and colleagues (1953) defined the increased response system in stress situation as
‘response stereotypy’, proposing a connection to causes of the disease. The individual-specific
response indicates that an individual shows the same stereotypical physiological response pattern
(e.g. sympathetic or parasympathetic) to varied stressors such as cold pressure, mental arithmetic,
hyperventilation, word-association (Lacey et al 1953 ). The organ with the most frequent
responses in these different stress situations has the highest risk for development of a disease
(Lacey et al. 1953).
An important study reported by Malmo and Shagass (1949) showed that painful stimulation
activated the same functional systems that mediate clinical symptoms in psychophysiological
disorders, for example cardiovascular disorders or tension headache. The authors termed this
specific response as a ‘symptom specificity’ that describes the reactivity of one functional system.
Hodapp et al. (1975) and Johannes et al. (2003) demonstrated that patients with hyperten-
sion respond with higher BP increases in stress situations compared to normotensive individuals.
This symptom specificity is not limited to psychophysiological disorders. Patients with non-
psychophysiological diseases, such as the inflammatory rheumatic diseases of rheumatoid arthri-
tis and systemic lupus erythematosus, also respond with BP reactivity and showed a highly
significant correlation with pain sensitivity. Johannes and colleagues (2003), who examined BP
and additionally HR, skin conductance level (SCL), electromyogram (EMG) and voice pitch
found four psychophysiological response patterns.
Thieme and Turk (2006) used a similar design as Johannes et al. (2003) to investigate the stress
reactivity of fibromyalgia (FM) patients and healthy controls. They replicated the same four
response patterns with high BP, low BP, high SCL, and high EMG-reactivity identified by Johannes
et al. Further, this psychophysiological study found that only BP reactivity in patients with chronic
musculoskeletal pain appears to be positively related to pain intensity (Thieme et al. submitted).
These results were confirmed by Bragdon et al. (2002) and Bruehl et al. (2002, 2005) who found
chronic pain-related alterations in the BP–pain sensitivity relationship in patients with chronic
pain. Several studies of patients suffering from CBP and temporomandibular joint disorders
(TMJD) show an EMG-reactivity at the site of pain but not at distal sites (e.g. Burns 1997; Flor
et al. 1992 ; Peters and Schmidt 1991 ). This EMG-reactivity is highly associated with pain
sensitivity. Similar to the psychophysiological results in FM, Flor and Turk (1995) found that
baseline levels are not generally elevated in chronic pain patients. They concluded that the pres-
ence of symptom-specific psychophysiological responses is more commonly observed. The
intriguing results on psychophysiological reactivity patterns (e.g. Flor et al. 1992; Johannes et al.
2003; Thieme and Turk 2006) suggest that stress-related symptom specificity associated with the
main symptom may indicate the aetiology of the painful condition, for example, EMG-reactivity
with pain in CBP or BP-reactivity with pain in FM.
8.2 Assessment of stress-related psychophysiological reactions

8.2.1 Psychological factors in chronification of pain
As noted, stress may facilitate the development of chronic pain states, irrespective of aetiology
(Jensen et al. 1991; Passatore and Roatta 2006). This ‘chronification’ of pain is influenced by
psychological factors such as expectations, cognitive self-statements, learning and by comorbidity
associated with stressful situations reviewed above. Based on the assumptions above, stress is
the interaction between the person’s psychological and physiological response to a stressor.
The following considers the stressors of patients suffering from CBP such as psychosocial risk
factors at work, pain behaviours, cognitions, and comorbidity.
Psychosocial stressors in CBP

CBP stressors include work disability due to physical impairment, reduced free-time, and pain-
related interference in family and social relationships. A growing body of evidence for psychosocial
risk factors at work highlights high task demands, lack of support from colleagues, low job control,
and low influence. The evidence for high job strain, low supervisor support, conflicts at work, low
job security, and limited rest periods is inconclusive (Ariëns et al. 2001). There is interesting evi-
dence for a relation between mental stress at work and upper extremity complaints (Malchaire
et al. 2001; Bongers et al. 2002). An important mental stressor is related to ‘mobbing/bullying’
that is defined as psychological terror that involves hostile and unethical communication, social
exclusion, active and passive obstruction in demanding working tasks. Bullying/mobbing
can initiate and maintain a helpless and defenceless position, usually over a long period of time
(statistical definition: at least 6 months of duration) (Lehmann 1996).
Bullying/mobbing and a loss of social support are thought to increase the risk of developing a
CBP and other musculoskeletal disorders frequently in the neck/shoulder region, particularly in
occupations of low physical demand (Wærsted 2000).
The primary consequences of psychosocial stressors are reduced activity and absence from
work, interpreted as pain-avoidance behaviours mediated by fear and anger (Anderson 1999)
associated with an increased cardiovascular stress response (Berntson et al. 2003). The 1988
National Health Interview Survey (1985–88) indicated that women with CBP experienced 56%
days of restricted activity (Praemer et al 1992). Patients returning to work after 6 months absence
had 68% lost work days (Rossignol et al. 1988). Interestingly, the precipitating event, diagnosis,
and prescribed treatment do not influence treatment outcome, while a specific aetiological diag-
nosis, older age, previous back pain, and psychosocial distress and psychological disturbance and
disorder are negatively associated with recovery (Van Doorn 1995 ; Greenough, 1993 ).
Compensation that is only loosely linked to the presence of physical pathology has been show to
prolong disability in numerous studies. For example, in a retrospective cohort study of 300
patients in Adelaide, Australia, the average time off work for men with compensation was a year
versus a week for men with no compensation, despite comparable levels of objective pathology.
For women with comparable injuries, the corresponding off-work periods were 15 months versus
2 weeks (Greenough et al. 1993). The location of symptom onset is also influential, work-related
back symptoms result in longer absence from work than non-work-related symptoms, even after
controlling for the influence of the physical work environment (Leavitt 1992).
It is not surprising that social problems at the workplace and the associated financial problems,
anger, fear and maladaptive pain cognitions provoke distress in the family (Pienimäki et al.
2002; Schneider et al. 2005). Chronic pain has a negative impact on relationship satisfaction
(Cano 2000, 2004; Flor et al. 1987, 1992; Kerns 1990). This association is bidirectional; the
quality of relationships may affect pain experience associated with psychophysiological responses
(Cano 2004a, Flor et al. 1987, Kerns and Turk 1987) such as increased cortisol and blood pressure
response (Carels et al. 1998; Heffner et al. 2004). Psychological distress impairs activities of
daily life (Ryan et al. 2010) and decreases amount of free-time, affecting individual-specific stress
reactivity.
Personality
A pain- or CBP-personality has not been identified. A specific complex of behaviours seems to
be reflective rather of a general chronification syndrome (Hildebrandt and Franz 1988 ).
Interestingly, many of these behaviours would be described by patients as ‘good’ behaviours.

Using a qualitative analyses of both anamnestic and standardized questionnaires data in 104 CBP
patients, Hildebrandt and Franz (1988) identified three behavioural patterns that appear as if they
would characterize healthy behaviours yet actually maintain chronic pain: (1) consistent per-
formance; (2) readiness to help others; and (3) avoidance of conflicts. Performance pressure and
suppressing of emotions to avoid conflicts are strongly associated with psychophysiological stress
responses such as increased skin conductance level (SCL), heart rate (HR) and BP (see also,
Johannes et al. 2003; Thieme et al. 2006).
These patients described themselves as ‘always busy’, hard-working employees with high per-
formance expectations and their identity is related strongly to their work. They characterized
their lives by early acceptance of responsibility, destitution, and fighting to survive. The fixation
on work as the centre of their life reduced their enjoyment of success and relaxation. Eighty per
cent of the CBP patients ranked ‘ready to help others’ as a primary goal. The prevalent ‘helper
role’ was associated with needs that are ignored and not perceived, and rejecting help from others.
These patients failed to present an adequate, self-protective and assertive approach to environ-
mental demands. Yet, the patients view themselves in a positive way, as they avoid conflicts by
excessive adaptation or by adapting a helper-role. The tendency to avoid conflicts suggests inhibi-
tion and deficits in social competence that has been confirmed by standardized test inventories
(e.g. Asmundson et al. 1995; Hildebrandt et al. 1997). Psychophysiological experiments using
social conflicts to provoke a temporary emotional deprivation showed an increase of EMG, BP,
and HR in those participants who avoided the expression of emotions. In contrast, participants
with lower psychophysiological responses were able to perceive and express anger and fear neces-
sary to solve social conflicts (Flor et al. 1995; Shapiro et al. 1996). The connection of pain and the
cardiovascular system is described below.
Cognitions
Cognitions powerfully influence pain perception. Prominent cognitions in CBP include pain-
related fear (Leeuw et al. 2007; Woby et al. 2007), catastrophizing (Cano 2004; Vowles et al. 2007)
and control over pain (Haythornthwaite et al. 1998; LaChapelle et al. 2001). In comparison to a
number of biomedical indices, Waddell et al. (1993) reported that fear-avoidance beliefs about
work comprised the most specific and powerful factor accounting for disability and work loss.
These pain related avoidance beliefs are related to fear of movements, fear of pain, and fear of
work-related activities. CBP patients may have high or minimal levels of kinesiophobia. Fear of
movement is strongly connected with the presence of pain behaviours as an expression of operant
conditioning. The different level of subconscious learning in patients with chronic pain suggests
the heterogeneity of chronic pain that may be manifest not only in physical but also in pain-
specific fear-avoidance beliefs.
Catastrophizing describes a cognitive style that attributes the most negative interpretation
among a range of possible interpretations of an event. Flor and Turk (1988) found that the com-
bination of situation-specific variables such as catastrophizing and general cognitive variables
explained between 32–60% of the variance in pain and disability, respectively. Turner et al (2002)
replicated these results and reported that pain coping and catastrophizing explain an additional
29% of the variance in pain intensity and 30% of the variance in psychological distress. Hanley
et al. (2008) found that increases in catastrophizing and lack of belief in one’s ability to control
pain were each significantly associated with greater pain interference and poorer psychological
functioning. In contrast, changes in specific coping strategies and social support were not predic-
tors of changes in pain, interference, or psychological functioning (Hanley et al. 2008), particu-
larly social support provided by someone who is involved in a patient’s daily life. Of course, not
all styles of partner support are adaptive!
Perceived control, independent of other variables, is associated with greater use of cognitive
and social coping strategies (i.e. asking for assistance, seeking social support and coping self-
statements) (LaChapelle et al. 2001). Coping self-statements and reinterpreting pain sensations
predicted greater perceptions of control over pain, whereas ignoring pain sensations predicted
lower perceptions of control over pain (Haythornthwaite et al. 1998).
Psychosocial adaptation
Several studies have generally replicated the psychosocial subgroups determined originally by
Turk’s (1996) cluster analysis of responses to the Multidimensional Pain Inventory (MPI).
Asmundson et al. (1997) found three groups: One group, labelled dysfunctional (34.9%), experi-
enced higher than average pain severity, pain interference, affective distress, higher solicitous
response from their spouses, and lower levels of self-efficacy and general activity. A second sub-
group comprised 24.3% of the CBP patients. This interpersonally distressed group reported mod-
erate pain severity and lower perceived social support by significant others who instead responded
with more punishing responses. This group had fewer solicitous or distracting responses from
their spouses and highest levels of activity and affective distress. The third group, adaptive copers
(40.8%), was characterized by lower pain severity, pain interference and affective distress,
and higher levels of self-efficacy and general activity. Spousal behaviour was characterized as
distracting more than solicitous.
An analysis of patients with spinal cord injuries (SCI) also found 3 subgroups (Widerström-
Noga et al. 2007). In addition to the dysfunctional and an adaptive patterns found by Turk et al
(1996), this analysis found an interpersonally supported (33%), instead of an interpersonally dis-
tressed, subgroup characterized by moderately high pain severity, higher life control, support
from significant others with more distracting and solicitous responses, and higher activities
scores. In that injury condition, negative or punishing spouse responses were not observed.
Carels et al. (1998) reported significant increases in BP after stress associated with recalling a
marital conflict. Heffner et al. (2004) who also investigated sympathetic responses in ‘healthy’
couples identified a relationship between lower spousal support satisfaction and higher affective
distress as well as higher cortisol production associated with BP reactivity.
Asmundson et al. (1997) showed that psychosocial adaptation is associated with differ-
ent dimensions of pain-specific cognitions regarding fear and avoidance. In contrast to
interpersonally-distressed and adaptive copers, patients with dysfunctional strategies reported
greater pain-specific cognitive fear, e.g. ‘I can’t think straight when in pain’, and cognitions regard-
ing physiological anxiety, e.g. ‘I begin trembling when engaged in an activity which increases pain’,
more cognitions of escape/avoidance behaviour, e.g. ‘When I feel pain I try to stay as still as pos-
sible’, and elevated fearful appraisals of pain, e.g. ‘When I feel pain, I am afraid something terrible
will happen’. In contrast, negative responses by a partner that provoke less active coping (Thieme
et al. 2004) are frequently associated with depression and both were found to mediate the asso-
ciation between pain and relationship satisfaction, with negative responses emerging as the most
important mediator (Waxman et al. 2008). Considering that cognitions influence pain behav-
iours, the heterogeneity of cognitions in CBP patients suggests heterogeneous pain perception
and avoidance behaviour that determines different subgroups based on those cognitions
and behaviours.
Pain behaviour
Pain behaviours are influenced by significant others and by dysfunctional strategies. Solicitous
responses by significant others have been found to be positively associated with higher ratings of
pain severity, greater disability, decreased activity levels, and more pain and avoidance behaviours
(Flor et al. 1995; Lousberg et al. 1992). Vlaeyen et al. (1995) proposed that avoidance, arising from
beliefs that activity will produce pain and suffering, leads to a vicious cycle characterized by
decreased self-efficacy, fear, and further avoidance as well as disability. This cycle is maintained
by the momentary reduction of anxiety attained through avoidance of feared and/or undesirable
activities. Flor and colleagues (1987) have shown that CBP patients with a solicitous partner show
higher pain behaviour than patients with a distracting or punishing spouse.
CBP patients with dysfunctional strategies reported more pain-specific fear and avoidance as
well as catastrophizing than did those classified as interpersonally distressed or as adaptive copers
(Asmundson et al. 1997). The lower activity in the dysfunctional subgroup is associated with disuse
and deconditioning. The presence of persistent low back pain leads to avoidance of daily activities,
which may lead to physical deconditioning, both generally such as loss of cardiovascular capacity
and as specifically such as loss of strength and endurance of paraspinal muscles (Smeets et al. 2006,
Verbunt et al. 2003), linked with increased muscular and cardiovascular responses in physical
stress situations. Pain catastrophizing and fear of physical activity exacerbate pain, promoting a
self-perpetuating cycle of avoidance, hypervigilance, depression, disuse, and pain (Vlaeyen et al.
1995). Catastrophizing is related to higher SCL and HR response (Bartley and Rhudy 2008). These
deconditioning models are based on classical and operant conditioning, and the influence of
cognitive, affective, and physical variables (Thieme et al. 2004; Turk and Okifuji 1997).
In conclusion, similar to cognitions and psychosocial adaptation, heterogeneous pain behav-
iours result from different psychobiological mechanisms. These differences likely provide a
rational basis for classification of subgroups in CBP.
Comorbidity
CBP is associated with comorbid anxiety (30.9–95%), depression (32–54%) and addictive behav-
iour (64–95%). The first episode of major depression follows pain onset in 58.1% of patients
(Atkinson et al. 1991; Polatin et al. 1993), commencing within the first 2 years of established pain.
Anxiety disorders followed a similar pattern in terms of rates of onset before or after pain, and of
increased risk of ‘new’ cases during the 2 years immediately following back injury. Alcohol use
disorder usually precedes pain onset considerably. However, the pooled life time prevalence of
psychiatric illnesses in chronic pain patients is not significantly different than that of non-pain
subjects. There is little evidence for factors that differentiate depressed from non-depressed
chronic pain patients (Haley et al. 1985; Sullivan et al. 2000). Further studies are needed to exam-
ine the relationship between comorbidity and pain. The data suggest that substance abuse and
anxiety disorders precede chronic low-back pain, whereas depression may develop before or after
onset (Andersson 2000).
Since patients with physical comorbidities appear to adapt to CBP, the presence of psychiatric
disorders likely interferes with this adaptation. Increasing evidence suggests that the tendency to
fear pain and to avoid pain and pain-related situations is mediated in part by anxiety sensitivity
(Asmundson and Norton 1995; Asmundson and Taylor 1996), a dispositional trait-like tendency
to respond fearfully to anxiety-related symptoms, and also to social situations (Asmundson et al.
1996; Vlaeyen 1999).
Summary
Stress variables influencing CBP are psychosocial risk factors at work such as high quantitative
demands, lack of support from colleagues, low job control and low influence, cognitions such as
fear-avoidance beliefs, catastrophizing vs. active coping, and pain and avoidance behaviours rein-
forced by spousal responses at home and causing disuse and deconditioning. These pain-related
variables can be exacerbated by psychiatric disorders such as depression, anxiety, and substance
dependence. The next section presents their relationship to physiological responses.
8.2.2 Psychophysiological variables

Static and dynamic postures
Presently, the high incidence of muscular (myalgia-type) disorders in individuals employed in
light industrial work has no satisfactory explanation. The biomechanical model assumes that
static and dynamic posture influence muscle tension and pain perception, and predicts an
increased EMG associated with the experience of pain after physical activity.
Of 38 studies reviewed, only 17 evaluated both static and dynamic postures (Table 8.1). A con-
cordance between surface electromyographic (sEMG) responses of both conditions was reported
in 52.9% of the studies. Forty-seven per cent reported that sEMG of CBP and pain-free control
participants were similar during static postures such as standing but different during dynamic
postures such as flexion and extension. These results suggest that subgroups based on movement
sEMG may be relevant for diagnosis and treatment. Thirty-five studies reported sEMG responses
during static and dynamic postures. Surprisingly, only 29% of these reported significantly higher
sEMG-responses in CBP patients during dynamic postures (highlighted in grey in Table 8.1),
however 43% showed less erector spinae activity in CBP patients than in pain-free controls, and
29% could not find any differences in muscle tension between CBP patients and pain-free controls
during both static and dynamic postures. Arena et al. (1989) divided the CBP sample into spond-
ylarthritis, intervertebral disc disorder, and unspecified musculoskeletal backache, and found that
spondylarthritis patients and PFCs had lower muscle tension during dynamic postures in com-
parison to intervertebral disc disorder and unspecified musculoskeletal backache patients. The
presence of diagnostic subgroups in CBP may explain these varied EMG responses.
The findings of sEMG-reactivity after dynamic postures in 30% of the studies in Table 8.1 sug-
gest that the reflex-spasm-approach may be relevant for one subgroup of CBP patients.
Unfortunately, sEMG and pain intensity were associated in only one study (Ahern et al. 1988)
that observed a positive correlation between sEMG and pain (Table 8.1). Physiological maladap-
tations may be the reason: the agonist muscle contractions were still maintained while antagonist
muscles were already activated (Radebold et al 2000, 2001). The agonist muscles did not show the
expected pattern of sEMG responses during trunk rotation, most likely because of restricted range
of motion and/or compensatory posturing as pain-maintaining sEMG-pattern (Ahern et al.
1988). Patients with higher sEMG maintained an increased level of muscle activity after lifting
objects (Soderberg et al. 1983). In addition, most patients were not capable to show different
EMG-responses in flexion or relaxation conditions.
Moseley et al. (2004) proposed an altered postural mechanism as a further explanation for the
transition from acute to chronic pain. They suggest that these alterations serve to protect the
spine in the short-term, but increased vulnerability to injury in the long-term. They showed an
increased activity of superficial trunk muscles during anticipation of experimental back pain in
healthy subjects. Moseley et al. (2004) concluded that the anticipation of back pain could evoke a
protective posture that stiffens the spine.
The postural changes are associated with expectations. Anticipatory postural adjustments
(APAs) of trunk-stabilizing muscles result in a feed-forward function in lateral abdominal mus-
cles during rapid arm movements. (Hodges et al. 2001). Sprott and colleagues used a new method
for the non-invasive determination of abdominal muscle feed-forward activity based on tissue
Doppler imaging (TDI, Mannion et al. 2008) as shown in Figure 8.1. The patients with CBP did
not have a delayed-onset of feed-forward activation but rather an earlier activation of abdominal
muscle activity. Further, the onset of muscle activity was not associated with clinical variables
(pain, disability, etc.). The authors proposed that APAs depend on the postural set, therefore the
individual’s perception of their steady-state postural equilibrium and quality of external support
126
FROM ACUTE TO CHRONIC BACK PAIN
Table 8.1 EMG measured during static (SP) and dynamic postures (DP)
Study Sample size Stressor Baseline EMG-reactivity Relationship Relationship Comorbidity

CBP vs HC CBP vs HC between pain between Pain and
and baseline EMG- reactivity
Ahern et al. CBP=40 Standing, flexion/ ns SP: CBP<HC na yes Pain behaviour
1988 HC=40 extension, rotation DP: CBP>HC
Alexiev, CBP=40 Standing, isometric na SP: ns na na na
1994 (acute) muscle activity DP: ns
HC=40 higher EMG on painful site
Ambroz CBP=30 Standing, flexion/ na SP: CBP > HC na na na
et al. 2000 HC=30 extension DP: CBP > HC
Arena et al. CBP=178 Standing, bending, EMG SP: CBP > HC na na na
1989 (3 groups: SA, rising, unsupported HC< IDD, DP: IDD, UMB >
IDD, UMB) sitting, supported UMB HC, SA
HC=29 sitting, and prone
Arena et al. CBP=29 Standing, bending, na SP: ns na na na
1990 HC=20 rising, DP: ns
unsupported sitting,
supported sitting, and
prone during two
testing sessions
Arena et al. CBP=46 Standing, bending, na SP: CBP > HC na na na
1991 (2 groups : IDD, rising, unsupported DP: IDD > HC
UMB) sitting, supported UMB ns HC
HC=20 sitting, and prone. CBP CBP with low compared to
were tested while high pain: ns
having high and low
pain
Capodaglio CBP=4 Isometric contraction na SP: na na na na
and Nilsson, HC=4 DP: ns
1996
Cassisi et al. CBP=21 Sitting, maximal na SP: CBP < HC na na na
1993 HC=12 isometric exertion DP: CBP < HC
Cram and CBP=33 flexion/extension, ns SP: ns na na na
Steger 1983 (3 groups: UBP, rotation DP: CBP > HA
LBP, MBP) For bilateral levels in lumbar and
HA=12 cervical paraspinal muscle groups
DP: CBP < HA
For bilateral levels in frontalis and
masseter groups
Elfing et al CBP=57 Seated isometric trunk na SP: na na na na
2003 HC=55 extension DP: ns
Hoyt et al. CBP= flexion/extension, EMG SP: na na na na
1981 HC= rotation ns DP: CBP > HC
THE PSYCHOPHYSIOLOGY OF CHRONIC BACK PAIN PATIENTS

Jalovaara CBP=18 Standing na SP: CBP > HC DP: na na na na
et al. 1995 HC=11
Kankaanpää CBP=20 Isometric contraction na SP: na na na na
et al. 1998 HC=15 DP: ns
Klein et al. CBP=8 Isometric contraction na SP: na na na na
1991 HC=17 DP: ns
Kravitz et al. CBP= flexion/extension, EMG SP: na na na na
1981 HC= rotation ns DP: CBP > HC
Lee et al. CBP=8 Prone, isometric holds na SP: na na na na
1992 HC=10 with torso unsupported DP: ns
Leinonen CBP=20 Expected and na SP: ns na na na
et al. 2001 HC=15 unexpected loading of DP: ns
the upper extremities ns between expected and
while sitting and nonexpected loading in CBP
standing
(Continued)
127
128
Table 8.1 (continued) EMG measured during static (SP) and dynamic postures (DP)

Lisiński, 2000 CBP=62 Prone, isometric hold SP: na na na na
HC=31 for 2 seconds with DP: CBP < HC
torso unsupported
Lofland et al, CBP (high)=18 Standing na SP: CBP (high) > CBP (low) and na na na
2000 CBP (low)=33 CBP (high) > HC
HC=30 DP: na
Lu et al. CBP=20 Standing in postural na SP: ns na na na
2001 HC=20 restraint device, DP: CBP > HC
performing lifts with
and without trunk
rotation
Mayer et al. CBP=10 Prone, isometric hold na SP: ns na na na
1989 HC=11 for 15 seconds with DP: CBP < HC
torso unsupported.
Two 10-consecutive-
trial sessions
Miller, 1985 CBP=11 Sitting, standing, active na SP: ns na na na
HC=11 sitting DP: ns
Ng et al. CBP=12 Axial rotation while na SP: na na na na
2002 HC=12 standing at different DP: CBP < HC
levels of effort
Pääsuke CBP=12 Prone, isometric hold na SP: na na na Fatigue
et al. 2002 HC=12 with torso unsupported DP: CBP < HC
Lower time endurance in CBP
Peach and CBP=21 Semistanding na SP: na na na na
McGill, 1998 HC=18 DP: ns
Robinson CBP=16 Sitting, repetitive na SP: ns na na na
et al. 1992 HC=12 concentric and DP: CBP < HC
eccentric contractions
with light and heavy
resistance
Roy et al. CBP=12 Standing, isometric na SP: ns na na na
1989 HC=12 muscle activity DP: CBP < HC
Roy et al. CBP=6 Standing, isometric na SP: ns na na na
CBP had higher recovery time
Roy et al. A : CBP=28 Standing, isometric na SP: ns na na na
B : CBP=57

HC=6
Shirado et al. CBP=20 Flexion and extension na SP: na na na na
1995 HC=25 DP: CBP < HC
Sihvonen CBP=87 Flexion and extension na SP: na na na na
et al. 1991 HC=25 DP: CBP > HC
Sihvonen Pregnant Flexion and extension na SP: CBP > HC na na na
et al. 1998 CBP=21 DP: CBP > HC
Pregnant
HC=32
Soderberg CBP=25 Flexion and extension CBP>HC SP: na na na na
and Barr HC=20 DP: CBP < HC
1983
Suter and CBP=25 Prone, isometric hold na SP: na na na na
Lindsay, HC=16 with torso unsupported DP: CBP < HC
2001
Watson et al. CBP=70 Flexion and extension ns SP: na na na na
1997 HC=20 DP: CBP < HC
(Continued)
129
130
Table 8.1 (continued) EMG measured during static (SP) and dynamic postures (DP)

Collins et al. CBP=11 flexion/extension, ns SCL na na na
1982 HC=11 rotation DP: CBP > HC
Radebold CBP=17 Standing, supported na Slower muscle recovery time in na na na
et al. 2000 HC=17 with feet off the CBP in agonist and antagonist
ground muscles
Radebold CBP=16 Standing, supported na na na na
et al. 2001 HC=14 with feet off the
ground
na: no assessed, ns: non-significant, SA: spondyloarthritis, IDD: intervertebral disc disorder, UMB: unspecified musculoskeletal backache, UBP : upper back pain, LBP : low back pain,
MBP : mixed back pain, HA : headache
Fig. 8.1 (Also see Colour Plate 5) Tissue Doppler imaging (TDI).
(Cordo et al. 1982), and that modifications to APAs are centrally mediated (Morris et al. 2006).
Hence, it is conceivable that factors such as fear of pain, fear of movement, weakness, etc.—
factors that might lead patients with CBP to expect that they will be less able to withstand chal-
lenges to postural stability—could precipitate the earlier APAs (Gubler et al. 2010).
The neurophysiological consequence of posture appears to be influenced by both neuropsy-
chological and endocrine variables. Kravitz et al. (1980) showed that a flexed posture is associated
with serotonin production and an extended posture is associated with octopamine production.
An increased serotonin production is connected with emotional factors such as affective distress
and anxiety that can influence the posture. A consequence of postural changes is muscle fatigue
that is associated with significantly shorter endurance time and declining of sEMG spectral mean
power frequency during performance of isometric contractions (Pääsuke et al. 2002). Two fur-
ther studies of recovery time in agonist and antagonist muscles found that CBP need 30% more
time than HC for recovery (Radebold et al. 2000, 2001).
These findings of enhanced sEMG activity during postural changes and an extended time for
recovery provide support for the biomechanical model of chronic pain in only one subgroup of
CBPs. The heterogeneous sEMG activity and earlier APAs indicate the need for further research
pertaining to movement-related lumbar muscle activity and implicate the influence of cognitions
and pain behaviour associated with pain-maintaining central changes (Aherns et al. 1988; Flor
et al. 2002).
An extension to the myogenic model considered variables of HR and SCL as well as frontalis
sEMG in addition to paraspinal sEMG. Those measures were recorded while CBP patients and
pain-free controls were tested by three conditions of different postures, physiological and mental
stress. Though differences were found between different postural categories, the CBP group exhib-
ited similar or significantly less paraspinal muscle activity than the pain-free control group. In
contrast, frontalis sEMG and skin conductance were significantly higher in the CBP group. SCL
in CBP patients was enhanced by mental arithmetic and cold pressor stimulation. This study
provides no support for the biomechanical reflex spasm theory of CBP suggesting an alternative
explanation of an increased arousal response and altered ability to respond to demanding tasks,
leading to pain and eventually to decreased paraspinal muscle activity (Collins et al. 1982).
In summary, there is a considerable incidence of myalgia-type disorders in light industrial

work. In 38 studies of static and dynamic postures, less than 30% confirmed the assumption of
increased muscle tension and pain as a reflex-spasm. Differences in sEMG activity in static and
dynamic postures, in diagnoses of CBP, and the heterogeneity of stress-reactivity in sEMG and in
other variables such as SCL challenge the assumption of a homogenous group of CBP. These
differences and the further findings of altered APAs, the influence of attention and subconscious
operant and classical conditioning, and the association of muscle tension and pain intensity
strongly suggest that the population of CBP patients can be divided into several distinct
subgroups.
Symptom-specific psychophysiological response

Heterogeneous sEMG-responses in response to postural change suggest that the determining
variables extend beyond physical workload to include psychological distress related to time pres-
sure, low job satisfaction and lack of control of muscular tension (Waersted et al. 1991). To fully
understand the high prevalence of musculoskeletal disorders associated with stressful work, it is
important to explore the relationship between muscle activity and psychophysiological stress
responses.
Field studies Rissén et al. (2000) evaluated trapezius sEMG activity, HR, BP, and levels of urinary
catecholamines and salivary cortisol among 31 female supermarket employees suffering from neck
and shoulder disorders. The participants had increased endocrine responses with elevated epine-
phrine, norepinephrine, and cortisol values. They also showed higher BP and HR and slightly
enhanced sEMG and self-reported variables during a 2-hour work period in comparison to 1-hour
defined rest period. The sEMG activity during work was significantly associated with self-reported
negative stress (r=0.38–0.43) and with systolic BP (r=0.38). Associations were found also between
BP and workload (r=0.43–0.57) as well as BP and urinary epinephrine (r=0.40) and norepine-
phrine (r=0.46) during work. No associations were found between sEMG activity and pain or
workload defined as the numbers of customers and as the number and weight of the items that
the participant lifted. Thus, perceived negative stress, but not workload, may have a specific influ-
ence on muscle activity. Further, the data showed a lack of association between clinical pain before
work and higher sEMG activity during work. The association between workload and stress-
induced sEMG may be mediated by a noradrenergic mechanism that impacts the BP response
(e.g. Johannes et al 2003). Future studies may clarify the association between BP and pain in CBP.
The association between sEMG and negative stress is supported by other field studies. A 2-year
prospective study (Heydari et al. 2010) investigated the association of CBP with sEMG variables
over time in 120 healthcare workers. After 2 years, 17.2% of the group with no history of CBP
deteriorated as measured by time off work, disability, reported pain and self-assessment rating.
Several EMG variables were significantly associated with the outcome measures. The value of the
EMG variable half-width (HW, measured as the width of the spectrum at half the maximum
power) at inception demonstrated significant association with changes in outcome measures and
self-assessments of CBP at follow up. CBP patients with a HW of greater than 56Hz were at three-
fold greater risk, and those with an initial median frequency greater than 49Hz were at a 5.8-fold
greater risk of developing back pain.
The finding of increased low-level and long-lasting work-related muscle activity is often inter-
preted as representing solely biomechanical muscle load, ignoring the possibility of extra muscle
activation due to nonbiomechanical factors (Waersted 2000). Several studies tend to support this
assumption but the experimental data remain inconclusive (Waersted et al. 1991; Westgaárd and
Bjøklund 1987).
Experimental studies Flor and Turk (1989) reviewed 59 studies published during 1975–1989
that evaluated psychophysiological responses of patients with chronic pain. This review included
8 studies with CBP patients that evaluated physical variables such as posture and psychological
variables such as maladaptive cognitions. Only 30% of the studies reported an EMG-reactivity in
CBP (Table 8.2). In one study (Flor et al. 1985), pain-free patients and patients with CBP or other
pain disorders were asked to recall a very painful episode and a stressful event. Responses
during this recall were compared to two other conditions: reciting the alphabet backwards and
performing mental arithmetic. In comparison to controls, only the condition of recalling a pain
episode and a stressful event was sufficient to increase the muscle tension of the erector spinae
muscle in CBP patients. Frontalis-sEMG, HR and SCL were not significantly different among the
three groups. In addition, tasks with high personal relevance, but not low personal relevance,
provoked a muscular stress response, a finding described as ‘muscular response stereotypy’ (Flor
et al. 1985).
In contrast to this important result of a symptom-specific stress response related to an emo-
tional highly relevant individual state, two studies (Cohen et al. 1986; Collins et al. 1982) found
similar or less paraspinal muscle activity in CBP and pain-free controls. Both studies used mental
arithmetic as the psychological stressor. The participants did not receive any positive or negative
responses that minimized the personal relevance of this task. This result supports previous data
(Flor et al. 1985) that only personally important states provoke a symptom-specific psychophysi-
ological response. The results suggest that social and emotional factors at work exert a much
greater influence on maintenance of CBP symptoms than the level of work intensity. Collin’s
results (1982) also support the heterogeneity of CBP. Increased SCL, not sEMG, was associated
with pain, suggesting that increased arousal response and altered ability to respond to demanding
tasks leads to pain and to decreased paraspinal muscle activity.
The different sEMG and SCL responses, both at baseline and evoked by physical and psycho-
logical stressors, further suggest heterogeneity in CBP. There was little evidence for the hypothesis
that varied CBP subgroups can be differentiated by levels of muscular activity. Baseline levels of
activity, regardless of type of physiological measure, are not generally elevated in chronic pain
patients, while symptom-specific stress-related psychophysiological responses are commonly
observed. Future studies should examine if the baseline level is increased in the subgroup
with muscular symptom-specific stress response in comparison to the subgroup with sudomotor
stress response.
Individual-specific psychophysiological response

What mediates individual psychophysiological responses to emotionally relevant stressors?
Candidate mediators include anger, fear, and pain behaviour.
Anger appears to be an important consideration in the understanding of chronic pain. Over
70% of CBP patients report angry feelings (Okifuji et al. 1999). Most commonly, patients report
that they are angry with themselves (74%) and with healthcare professionals (62%). The relevance
of anger to chronic pain experience seems to vary across targets. Anger toward oneself is signifi-
cantly associated with pain and depression, whereas overall anger is significantly related to
perceived disability (Okifuji et al. 1999).
It is not just the experience of anger that is important for pain perception, but also the manner
in which anger is typically managed (Burns et al. 1998; Kerns et al. 1994). Two styles of anger
management have received particular attention: anger-in, the tendency to suppress anger when it
is experienced, and anger-out, the tendency to express anger through verbal or physical means
(Spielberger et al. 1985). In comparisons with non-pain participants, CBP patients have elevated
levels of anger (Burns 1997; Fernandez and Turk 1995). Kerns et al. (1994) found that both
134
Table 8.2 EMG measured during mental, physical and physiological stressors as well as pain induction and during sleep

Study Sample size Stressor Baseline Stress-reactivity Relationship Relationship Comorbidity
and baseline Stress reactivity
Controlled studies
Bonnet et al. 2004 D-CBP=9 Acoustic stimulation SCL SCL Yes Yes Depression
ND-CBP=9 ND-CBP>rest ND-CBP>rest in D-CBP in D-CBP
D-HC=9 HC> D-CBP
ND-HC=9
Bruehl et al. 2006 CBP=43 Pressure pain na Absence of na yes Anger-Out
HC=45 Ischemic pain Opioid-Analgesia
Opioid blockade (8mg in CBP
Naloxon)
Placebo blockade
Cohen et al. 1986 CBP=13 Dynamic postures ns EMG na no no psychiatric
HC=13 Mental arithmetic ns comorbidity
Cold pressure
Collins et al. 1982 CBP=11 Dynamic postures ns SCL na na no psychiatric
HC=11 Mental arithmetic CBP > HC comorbidity
DeGood et al. 1994 CBP=20 Mental stress ns EMG na na na
(disabling, CBP > HC
non-disabling) (disabling)
HC=12
Fischer and Chang 1985 CBP=9 Sleep na EMG na na na
HC=12 CBP > HC
Flor et al. 1985 CBP=17 Talk about stress Talk EMG EMG no yes Depression
HC=17 about pain CBP > HC CBP > HC
Mental arithmetic
Reciting an alphabet
Lundberg et al. 1999 CBP=50 Mental Arithmetic na EMG na na Stress
HC=22 STROOP
Cold Pressure Test
Test contractions
Peters and Schmidt 1991 CBP=20 Pressure pain stimuli SCL SCL no Yes na
HC=20 CBP > HC CBP > HC for HC
Peters and Schmidt 1999 CBP=20 Pressure pain EMG EMG no no Anxiety
HC=20 ns. CBP < HC Depression
Sjörs et al. 2009 CBP=18 Baseline rest EMG EMG no yes Anxiety
HC=30 Repetitive low-force CBP > HC CBP > HC
work
Trierer Stress–Test
Recovery
Tousignant-Laflamme and f-CBP=14 Pressure pain as clinical HR ns HR na yes (HR) na
Marchand, 2006 m-CBP=16 pain SCL ns f<m no (SCL)
Thermal pain as SCL
experimental pain f>m

Vlaeyen et al. 1999 HF-CBP=16 Fear-eliciting video EMG EMG na yes Fear
LF-CBT=15 presentation ns ns Negative
vs. neutral nature Affectivity
documentary
Uncontrolled studies
Burns et al. 1997 CBP=107 Mental Arithmetic EMG EMG no yes Depression
Anger recall interview Stress > Baseline
Burns 1997 CBP=102 Mental Arithmetic EMG EMG no yes Anger Hostility
Anger recall interview Stress > Baseline
Geisser at al. 1995 CBP=21 Field study (8.1 h) na EMG na no na
Physical activity
Psychosocial Stress
Rissen et al 2000 CBP=31 Field study: Repetitive EMG EMG na no Negative Stress
physical work Stress > Baseline at work
na: not assessed, ns: non-significant, D-CBP: depressed CBP, ND-CBP: non-depressed CBP, D-HC: depressed HC, ND-HC: non-depressed HC, f-CBP: female CBP, m-CBP: male CBP, HF-CBP:
high-fear CBP, LF-CBP: low-fear CBP
135
‘anger-in’ (toward self) and ‘anger-out’ (toward external agents or events) showed significant
positive associations with ratings of pain intensity in a sample of CBP patients. High levels of
anger-out have been shown to be associated with low levels of improvement in lifting capacity,
and high levels of anger-in with low levels of improvement in activity levels among chronic pain
patients undergoing multidisciplinary pain treatment (Burns et al. 1998). Psychophysiological
studies with pain-free controls showed that experimental induced anger was associated with
increased pain sensitivity to ischemic pain induction, and less control of anger has associated with
lower pain and tolerance thresholds (Gehlkopf et al. 1997; Janssen et al. 2001). Overall, the avail-
able data strongly suggest that anger and anger management style may affect pain and functioning
in chronic pain sufferers.
Anger style influences muscle reactivity. Higher levels of anger-out were associated with greater
increases in lower paraspinal muscle activity, but not trapezius muscle activity, during an anger
recall interview among men with chronic LBP (Burns 1997). Thus, anger-out could influence the
development or maintenance of chronic pain through selective effects of increased muscle ten-
sion at the site of injury or disorder. In contrast, Janssen et al. (2001) found another type of
symptom-specific stress reactivity to anger. They observed increased cardiovascular reactivity in
pain-free subjects after anger induction by harassing comments presented in computer tasks. The
cardiovascular measures and pain sensitivity were higher during harassment than during the
neutral situation. Harassment appears to inhibit pain through physiological mechanisms. This
effect is consistent with evidence that high BP reactivity is associated with lower sensitivity to
pain, possibly resulting from baroreceptor stimulation. According to the results of studies that
show diminished baroreceptor sensitivity in chronic pain patients, CBP patients with cardiovas-
cular stress reactivity in anger situations should experience increased pain related to anger.
The relationship between anger and both acute and chronic pain sensitivity may be related to
endogenous opioid dysfunction. Subjects with higher anger exhibit impaired endogenous opioid
inhibition of cardiovascular responses to a painful stressor (Bruehl et al. 1996). The high anger-
out scores were associated with an absence of opioid analgesia during acute pain tasks; low anger-
out scores were associated with effective opioid analgesia–similar to that of pain-free controls.
These findings suggest that anger-in and anger-out affect pain sensitivity through different mech-
anisms. Anger-out may be mediated by endogenous opioid dysfunction (Bruehl et al. 2002a).
Anger-in may result from depressed affect (Tschannen et al. 1992) and its biochemical sequelae.
These results are consistent with the hypothesis that anger-related variables may affect pain
through an association with endogenous opioid dysfunction and impair the psychological
adaptation to chronic pain.
Fear and pain avoidance behaviour The ‘fear-avoidance model of exaggerated pain perception’
(Lethem et al. 1983), and the more recent ‘cognitive model of fear of movement/(re)injury’
address the important topic of pain-related fear and avoidance behaviour (Vlaeyen et al. 1995).
The central concept of these models is fear of pain, or the more specific fear that physical activity
will cause (re)injury. Two opposing behavioural responses to these fears are confrontation and
avoidance (Crombez et al. 1999). ‘Confrontation’ may lead to adaptive responses which may lead
to the reduction of fear and the promotion of recovery. In contrast, ‘avoidance’ leads to the main-
tenance or exacerbation of fear, possibly resulting in a phobic state. Avoidance behaviour is
strongly connected with fear-avoidance beliefs associated with higher pain perception and less
range of motion seen during a procedure involving a passive but painful straight leg raising test
(McCracken et al. 1992). Fear-avoidance beliefs about work are strongly related to disability in
daily living and lost work days in the past year, and less related to pain variables such as anatomi-
cal pattern of pain, time pattern, and pain severity (Waddell et al. 1993). Using the Tampa-Scale
for Kinesiophobia (TSK), Vlaeyen et al. (1995) found that fear of movement/(re)injury was the
most powerful predictor of the performance in a simple weight lifting task. Patients were asked to
stand and lift a 5.5-kg bag with the dominant arm and hold it until pain or physical discomfort
made it impossible for the patient to continue. This performance test showed differences in
behavioural performance between patients with lower and higher fear of movement, with lower
behavioural performance related to higher fear of movement/(re)injury. Interestingly, fear was
higher after the performance test than before the test. Consistent with the delayed perception of
fear, no significant correlations were found between changes in the psychophysiological variables
of heart rate and skin conductance and fear beliefs. A systematic increase in heart rate and skin
conductance was not observed. Similarly, no significant differences in heart rate and skin con-
ductance were found between patients with low and high levels of fear beliefs. One possible expla-
nation is that behavioural avoidance occurred before psychophysiological arousal levels increased,
i.e. patients may think that short-term avoidance of movement prevents injury or an increase in
pain, a process of ‘cognitive avoidance’. Patients showed avoidance behaviour before they noticed
their anxious feelings. In addition, higher avoidance behaviour was associated with prolonged
pain complaints, possibly because delayed perception of fear permits a subconscious avoid-
ance behaviour associated with increased pain. Both feelings of fear and psychophysiological
stress responses may be delayed, explaining why pain is perceived only after physical activities.
These results are consistent with classical conditioning models in which pain is associated with a
fear response and an urge to avoid the pain. Avoidance and pain behaviour may also be explained
by operant conditioning. An experiment with 30 CBP patients and 30 matched healthy controls
investigated learning rates and extinction of verbal and cortical pain responses after operant
conditioning. Half of each group was reinforced for increased, and half for decreased, pain
reports during recording of EEG and muscle tension. Both groups showed similar learning rates,
however, the CBP patients displayed slower extinction of both the verbal and the cortical (N150)
pain response. In addition, the CBP group displayed prolonged elevated EMG levels to the task.
These data suggest that CBP patients are more easily influenced by operant conditioning factors
than healthy controls and this susceptibility may add to the maintenance of the chronic pain
problem (Flor et al. 2002).
In summary, musculoskeletal pain is often exacerbated by mental and social stress and psycho-
physiological mechanisms may play an important role in the development and maintenance of
chronic pain states. The heterogeneity of pain behaviours, anger expression, and fear beliefs in
CBP patients likely leads to different levels of classical and operant conditioning.
8.3 Interaction between pain regulatory and cardiovascular

systems: a new approach
Several studies indicate that resting BP levels may be elevated in patients with CBP (Brody et al.
1997, Lundberg et al. 1999, Nilsson et al. 1997) including pharmacological studies with the exclu-
sion criteria of ‘hypertension’ in CBP patients (Ongley et al. 1987). In a study of 300 patients with
chronic pain and 300 patients in a non-pain group, 39% of the pain patients were diagnosed with
clinical hypertension, compared with 21% in the non-pain group (Bruehl et al. 2005). Within
chronic pain patients, higher systolic blood pressure, along with higher BMI and more pro-
nounced metabolic disturbances have been observed with more intense pain (Nilsson et al. 1997).
BP is associated with muscle tension in patients suffering from neck and shoulder pain, and with
levels of norepinephrine which plays an important role in BP homeostasis (Lundberg et al. 1999).
These studies support the role of BP in pain chronification.
An important component of the pain regulatory system is the interaction of pain sensitivity and
cardiovascular response. This interaction is influenced by baroreceptor activity, which inhibits
activity in the ascending reticular activating system. The net result is modulation of pain inhibi-
tion by multiple factors (Angrilli et al. 1997 ; Bruehl et al. 2004 ; Droste et al. 1994; Dworkin et al.
1994 ; Maixner 1991; Mini et al. 1995; Rau and Elbert 2001).
In healthy individuals, there is a functional interaction between the cardiovascular and pain
regulatory systems that results in an inverse relationship between resting blood pressure levels and
acute pain sensitivity (Dworkin et al. 1994; Myers et al. 2001; Zamir and Schuber 1980). Baroreceptor
activation leads to the activation of descending pain inhibitory systems that decreases acute pain
sensitivity and result in progressive adaptation to repeated pain stimuli (Villanueva and Bars 1995;
Kleinboehl et al. 1999; Peters et al. 1989; Millan 2002). This baroreflex mechanism may explain
hypertensive hypoalgesia in normotensive individuals (e.g. Bruehl et al. 1992; Ghione et al. 1996).
In contrast, patients with CBP display a lack of the inverse relationship between BP and pain
intensity (Bragdon et al. 2002; Bruehl et al. 2002; Maixner et al. 1997). This lack of inverse rela-
tionship may be related to: (1) impairments in the descending pain inhibitory pathways normally
activated by increased baroreceptor stimulation (Maixner et al. 1995; Millan 2002); (2) increased
activity in descending facilatory mechanisms; or (3) diminished baroreceptor sensitivity (Bruehl
and Chung 2004).
While coordinated cardiovascular-pain regulatory responses may be part of an adaptive mech-
anism that helps the body to face stressful events in pain-free individuals, the dysfunctional
inverse relationship of BP and pain has been found to be maladaptive with central sensitization
and increased wind-up (Eide 2000; Kleinboehl et al. 1999; Li et al. 1999; Peters et al. 1989), con-
sistent with diminished descending pain inhibition (Kosek and Orderberg 2000). Brain regions
involved in descending pain facilitation appear to include the periaqueductal grey (PAG; Berrino
2001), the nucleus raphne magnus (NRM, Wiertelak 1997), the nucleus tractus solitarius (NTS,
Wiertelak 1997), and the rostroventromedial medulla (RVM, Pertovaara 1998). Descending pain
inhibition is likely mediated by both endogenous opioids (Bruehl et al. 2004a) and by nonopioid
mechanisms such as central noradrenergic mechanisms. The alterations of the BP—pain sensitiv-
ity relationship are not related to a dysfunction in opioid analgesic systems (Bruehl et al. 2002).
However, in more disabled CBP patients with higher level of expressed anger (Bruehl et al. 2003),
an endogenous opioid antinociceptive impairment may elevate acute and chronic pain sensitivity
(Bruehl et al. 2004a). Impairments in alpha-2 noradrenergic inhibitory pathways evoked by stress
and anxiety may be of particular relevance for understanding chronic pain-related alterations in
the BP/pain sensitivity relationship (Bruehl et al. 2004).
The net effect of descending modulation reflects the combined effects of descending inhibition
and facilitation. Increased descending facilatory activity may be triggered by baroreceptor activa-
tion via mechanisms involving substance P (Millan 2002; Randich and Gebhardt 1992). This
increased facilitatory activity results in central sensitization and increased temporal summation
(Watkins and Mayer 1999, 1999a) and is also involved in cardiovascular regulation (Ku et al.
1998; Seagard et al. 2000). Substance P pathways and brainstem regions such as NTS and NRM
that are involved in the BP—pain sensitivity modulation appear to overlap, and substance P has
effects on both pain regulation and baroreflex control of cardiovascular function (Gamboa-
Esteves 2001; Wiertelak et al. 1997).
A third contributor of the BP/pain sensitivity relationship is related to alterations of barorecep-
tor sensitivity (BRS). Changes in BRS sensitivity may alter pain regulatory processes (Maixner
et al. 1995). This effect could be due to an altered threshold for baroreceptor firing or altered CNS
gain associated with this firing. The functional consequences of diminished BRS include impaired
inhibition of sympathetic nervous system arousal and impaired activation of parasympathetic
nervous system inhibitory responses evoked by stressful stimuli (Randich and Maixner, 1984).
Only one study activated baroreceptors (by external carotid suction) in CBP patients, showing an
increased sensitivity to electrical pain stimuli (Brody et al. 1997). In contrast, pain-free normo-
tensives respond with analgesia to the same manipulation of baroreceptors (Rau and Elbert
2001). Altered BRS function may have both psychological and physiological consequences.
Decreased BRS is associated with increased anxiety levels (Watkins et al 2002) and increased acute
(Dito et al 1990; Steptoe et al. 1993) and chronic stress (Lawler et al. 1991; Qian et al. 1997). The
decrease in BRS in chronic pain may be due to altered central noradrenergic activity (Lawler et al.
1991; Mitchell and Lawler 1989). Possible interactive effects of substance P and alpha-2 adrenergic
activity in baroreceptor-mediated cardiovascular regulation are combined with chronic pain-
related changes in pathways mediated by these neurochemicals. That might also suggest a role for
baroreceptor changes in the altered BP/pain sensitivity relationship in chronic pain. Baroreceptor
sensitivity can be diminished by acute and chronic stress, and therefore, pain-related stress might
be expected to lead to similar diminished BRS. The resulting non-inverse BP–pain sensitivity
relationship in chronic pain may reflect multiple potential pathways such as baroreceptor, opioid,
norepinephrine, and dopamine pathways, and others located at the brain stem level (Bruehl and
Chung 2004).
8.3.1 CNS and baroreceptor sensitivity

Maixner and co-workers (1995a) suggested that impairments in baroreceptor mediated regula-
tion of nociception may contribute to the development of chronic musculoskeletal pain disorder
by disinhibition of the ascending reticular activating system. This system is a non-specific, cortical
projecting system (Randich and Maixner 1984; Dworkin et al. 1994) that originates from a diverse
number of nuclear groups in the brain stem and basal forebrain (e.g. parabrachial nucleus, locus
coeruleus, raphe system, nucleus basalis) and plays an important role in sculpting sensory, motor,
and autonomic responses to somatosensory input (Rau et al. 1993, Steriade 1988; Steriade and
Llinas 1988). The rostral part of the reticular formation has been described as the head of the
autonomic nervous system (Rohen 1978) from which ascending input is transmitted to the lateral
prefrontal and insular cortices (Bornhovd et al. 2002). The hypothalamus receives direct input
from the nucleus tractus solitarius, but also visceral information via the thalamus. The thalamus
receives baroreceptor input via the reticular formation. Rutecki (1990) suggests that the connec-
tion between the thalamus and the insular cortex mediates important visceral reflexes and this
connection has been mentioned as the possible link to conscious perception of visceral sensations
(Elbert and Schandry 1998).
Specific cardiovascular information reaches cortical areas mainly via the thalamus, the hypoth-
alamus and the tegmentum. In addition to widespread innervation through the reticular forma-
tion, additional regions such as the prefrontal cortex, the insula, and multiple somatosensory
areas process specific cardiovascular input. Of particular interest is the experimental demonstra-
tion that the activity of the ascending reticular formation is regulated, at least in part, by baro-
receptor input: The ascending reticular activating system is normally inhibited by baroreceptor
stimulation and can be disinhibited by deafferentation of baroreceptor pathways (Randich and
Maixner 1984). Disinhibition of this system, by diminishing the efficacy of baroreceptor afferent
input, may contribute to the mosaic of chronic, maladaptive psychological, sensory, motor, auto-
nomic, and neuroendocrine changes associated with CBP. In addition to altering the ascending
reticular activating system, impairments in baroreceptor mediated regulation of nociception may
result in a suppression of descending inhibitory pathways that tonically inhibit trigeminal
and dorsal horn neurons that respond to muscle and cutaneous nociceptive inputs (Mense 1993;
Yu and Mense 1990).
Taken together, these data indicate that stress may activate noradrenergic, serotonergic, and
dopaminergic pathways that diminish baroreceptor sensitivity and thus disinhibit the ascending
reticular activating system that: (1) activates prefrontal cortex, the insula, and somatosensory via
thalamic areas resulting in higher blood pressure and pain perception, and (2) inhibits trigeminal
and dorsal horn neurons resulting in increased EMG response. Thus the heterogenous EMG-
response and the heterogeneous relationship between muscle tension and pain sensitivity may be
mediated by stress-related diminished BRS and enhanced blood pressure. The different patterns
of pain processing in hypotensive persons compared with normotensive (Angrili et al. 1997) sug-
gest that CBP patients with a decreased cardiovascular response will have a lower pain perception
than CBP patients with higher BP response. The hypotensive stress pattern may be associated with
increased baroreceptor sensitivity, decreased muscular response and decreased pain inhibition.
8.3.2 Learning
In addition to these direct physiological effects, psychobiological mechanisms such as operant
and classical conditioning may influence baroreceptor activation. For example, the operant
conditioning of chronic pain articulated by Fordyce (1976) emphasized the influence of social
context on pain perception. This type of operant learning may mediate decreased pain perception
in hypertensive men with parental history of hypertension (Al’Absi 2005; Sandkuehler 2000;
Thieme 2005). There is some evidence that the BP is operantly conditionable, in particular in
patients with a predisposition for essential hypertension (Elbert et al. 1988; Goldstein et al. 1977;
Rau and Elbert 2001). A negative reinforcement of pain behaviour by hypertensive parents may
provoke an operant conditioned increase of baroreceptor sensitivity (Rau and Elbert 2001).
Elbert and co-workers (1994a) investigated the influence of daily life stress on baroreceptor
sensitivity and suggested a classical conditioning mechanism. They found that the increase of BP
was proportional to self-assessed daily life stress. Among the participants reporting the greatest
amount of stress, the pain inhibition effect accounted for more than 80% of the BP variance.
These results support the hypothesis that the reduction in perceived stress produced by baro-
receptor stimulation reinforces increased BP.
Multiple types of learning likely activate multiple pathways of chronic pain. Perceived pain
after operant conditioning is associated with increased activation of the anterior insula, medial
frontal cortex and cerebellum (Ploghaus et al. 1999). In contrast, classical conditioning activates
a different network involving entorhinal cortex of hippocampus, perigenual cingulum and medial
insula (Ploghaus et al. 2001). Classical conditioning is also related to alterations in the primary
somatosensory cortex (SI) that contributes to memory process in associative learning (Diesch
and Flor 2007).
8.4 Conclusion
Psychophysiological studies have disproved the biomechanical model of CBP that proposed that
pain may result from physically demonstrable CBP disease and from poor posture caused by
reflex spasm connected with high muscle tension. Studies that tested static and dynamic postures
found only 29% of the CBP patients with significantly higher sEMG-responses during dynamic
postures in comparison with pain-free controls. Only one study could report a connection
between higher sEMG responses and pain perception.
Studies of anticipatory postural adjustments of trunk-stabilizing muscles showed that postural
changes are connected with expectations that lead to an earlier centrally mediated activation of
abdominal muscle activity caused by fear of pain and movement.
In contrast to the myogenic model, psychophysiological studies showed an important influ-
ence of stress on the aetiopathogenesis of CBP. Classical and operant conditioning as well as
cognitions are relevant factors that play an important role in stress at work and in the partnership.
According to the symptom-specific response pattern, increased muscle tension has also been
proposed to be responsible for maintenance of CBP. However, only 30% of psychophysiological
studies could confirm the assumption of higher muscle tension in CBP. Personally important
situations provoke a symptom-specific psychophysiological response. The results suggest that
social and emotional factors at work exert a much greater influence on maintenance of CBP
symptoms than the level of work intensity.
Studies that reported higher SCL, HR or BP instead of higher EMG suggest psychophysiological
heterogeneity. Higher SCL suggests an increased arousal response influenced by anger, fear, and
comorbidity. A further individual-specific stress responses can be related to higher BP responses
that propose an interaction between cardiovascular and pain system. An inverse relationship
between BP and pain based on diminished baroreceptor sensitivity in CBP patients is assumed as
a further psychophysiological response pattern besides sudomotor (SCL) and muscular response
pattern.
The heterogeneity of the psychophysiological studies with CBP patients is determined by
different groups of characteristics. Physical characteristics such as diagnosis, association, and
non-association between physical changes and pain allow the definition of physical subgroups.
Varying levels of muscle activity and psychophysiological response pattern may similarly contrib-
ute to the determination of psychophysiological subgroups. The role of personal relevant stres-
sors, comorbidity, and close relationships in learned pain behaviour are consistent with previously
described psychosocial subgroups (Turk et al. 1996).
8.4.1 Future implications

Recent results suggest that future studies may investigate different psychophysiological subgroups
with assumed EMG, SCL and BP reactivity as well as their connection to psychosocial subgroups
characterized by pain, cognitions, emotions and behaviour to define different pathways of CBP.
References
Ahern, D. K., Follick, M.J., Council, J.R., Laser-Wolston, N., Litchman, H. (1988). Comparison of lumbar
paravertebral EMG patterns in chronic low back pain patients and non-patient controls. Pain 34(2): 153–60.
Al’ Absi, M., France, C.R., Ring, C., et al. (2005). Nociception and baroreceptor stimulation in
hypertension-prone men and women. Psychophysiology 42: 83–91.
Alexiev, A.R. (1994). Some differences of the electromyographic erector spinae activity between normal
subjects and low back pain patients during the generation of isometric trunk torque. Electromyography
and Clinical Neurophysiology 34: 495–99.
Ambroz, C., Scott, A., Ambroz, A., Talbott, E.O. (2000). Chronic low back pain assessment using surface
electromyography. Journal of Occupational Environmental Medicine 42:660–9.
Anderson, G. B. J. (1999). Epidemiological features of chronic low-back pain. Lancet 354: 581–5.
Angrilli, A., Mini, A., Mucha, R.F., Rau, H. (1997). The influence of low blood pressure and baroreceptor
activity on pain responses. Physiological Behavior 62: 391–97.
Arena, J. G., Sherman, R.A., Bruno, G.M., Young, T.R. (1989). Electromyographic recordings of 5 types of
low back pain subjects and non-pain controls in different positions. Pain 37(1): 57–65.
Arena, J. G., Sherman, R.A., Bruno, G.M., Young, T.R. (1990). Temporal stability of paraspinal
electromyographic recordings in low back pain and non-pain subjects. International Journal of
Psychophysiology 9(1): 31–7.
Arena, J. G., Sherman, R.A., Bruno, G.M., Young, T.R. (1991). Electromyographic recordings of low back
pain subjects and non-pain controls in six different positions: effect of pain levels. Pain 45: 23–8.
Ariëns, G. A., van Mechelen, W., Bongers, P.M., Bouter, L.M., van der Wal, G. (2001). Psychosocial risk
factors for neck pain: a systematic review. American Journal of industrial medicine 39(2): 180–93.
Asmundson, G. J. G., Norton, G.R. (1995). Anxiety sensitivity in patients with physically unexplained
chronic back pain: a preliminary report. Behaviour Research and Therapy 33(7): 771–7
Asmundson, G. J. G., Norton, G.R., Allerdings, M.D. (1997). Fear and avoidance in dysfunctional chronic
back pain patients. Pain 69: 231–6.
Asmundson, G. J. G., Norton, G.R. and Jacobson, S.J. (1996). Social, blood/injury, agoraphobic fears in
patients with physically unexplained chronic pain: are they clinically significant? Anxiety 2: 28–33.
Asmundson, G. J. G., Taylor, S. (1996). Role of anxiety sensitivity in pain related fear and avoidance.
Journal of Behavioral Medicine 19: 573–82.
Atkinson, J. H., Slater, M.A., Patterson, T.L., Grant, I., Garfin, S.R. (1991). Prevalence, onset, and risk of
psychiatric disorders in men with chronic low back pain: a controlled study. Pain 45(2): 111–21.
Ax, A. F. (1953). The physiological differentiation between fear and anger in humans. Psychosomatic
Medicine 75: 433–42.
Ax, A. F. (1964). Goal and methods of psychophysiology. Psychophysiology 1: 8–25.
Bartley, E.J., Rhudy, J.L. (2008). The influence of pain catastrophizing on experimentally induced emotion
and emotional modulation of nociception. Journal of Pain. 9(5): 388–96.
Beattie, P. F., Meyers, S.P. (1998). Magnetic resonance imaging in low back pain: general principles and
clinical issues. Physical Therapy 78(7): 738–53.
Berntson, G. G., Sarter, M., Cacioppo, J.T. (1998). Anxiety and cardiovascular reactivity: the basal
forebrain cholinergic link. Behavioral Brain Research 94(2): 225–48.
Berntson, G. G., Sarter, M., Cacioppo, J.T. (2003). Ascending visceral regulation of cortical affective
information processing. The European Journal of Neuroscience 18(8): 2103–9.
Berrino, L., Oliva, P., Rossi, F., Palazzo, E., Nobili, B., Maione, S. (2001). Interaction between
metabotropic and NMDA glutamate receptors in the periaqueductal grey pain modulation system.
Naunyn Schmiedebergs Archive of Pharmacology 364: 437–43.
Birbaumer, N., Flor, H. (1999). Applied psychophysiology and learned physiological regulation. Applied
psychophysiology and biofeedback 24(1): 35–7.
Birbaumer, N., Ghanayim, N., Hinterberger, T., et al. (1999a). A spelling device for the paralyzed. Nature
25(398(6725)): 297–8.
Birbaumer, N., Murguialday, A.R., Cohen, L. (2008). Brain-computer interface in paralysis. Current
Opinion in Neurology 21(6): 634–8.
Birbaumer, N., Schmidt, R.F. (1996). Biological Psychology [Biologische Psychology]. Berlin, Heidelberg,
New York, Springer.
Bongers, PM, Kremer, A.M., ter Laak J. (2002). Are psychosocial factors, risk factors for symptoms and
signs of the shoulder, elbow, or hand/wrist? : A review of the epidemiological literature. American
Journal of Industrial Medicine 41: 315–42.
Bonnet, A., Naveteur, J. (2004). Electrodermal activity in low back pain patients with and without
co-morbid depression. International Journal of Psychophysiology 53(1): 37–44.
Bornhovd, K., Quante, M., Glauche, V., Bromm, B., Weiller, C., Buchel, C. (2002). Painful stimuli evoke
different stimulus-response functions in the amygdala, prefrontal, insula and somatosensory cortex:
a single-trial fMRI study. Brain 125: 1326–36.
Bragdon, E. E., Light, K.C., Costello, N.L., et al. (2002). Group differences in pain modulation: pain-free
women compared to pain-free men and to women with TMD. Pain 96: 227–37.
Brody, S., Angrilli, A., Weiss, U., Birbaumer, N., Mini, A., Veit, R., Rau, H. (1997). Somatosensory evoked
potentials during baroreceptor stimulation in chronic low back pain patients and normal controls.
International Journal of Psychophysiology 25: 201–10.
Bruehl, S., Burns, J.W., Chung, O.Y., Ward, P., Johnson, B. (2002a). Anger and pain sensitivity in chronic
low back pain patients and pain-free controls: the role of endogenous opioids. Pain 99(1–2): 223–33.
Bruehl, S., Carlson, C.R., McCubbin, J.A. (1992). The relationship between pain sensitivity and blood
pressure in normotensives. Pain 48: 463–7.
Bruehl, S., Chung, O. Y., Ward, P., Johnson, B., McCubbin, J. A. (2002). The relationship between resting
blood pressure and acute pain sensitivity in healthy normotensives and chronic back pain sufferers: the
effects of opioid blockade. Pain 100: 191–201.
Bruehl, S., Chung, O.Y. (2004). Interactions between the cardiovascular and pain regulatory systems: an
updated review of mechanisms and possible alterations in chronic pain. Neuroscience and Biobehavioral
Reviews 28(4): 395–414.
Bruehl, S., Chung, O.Y., Burns, J.W. (2006). Anger expression and pain: an overview of findings and
possible mechanisms. Journal of behavioral medicine 29(6): 593–606.
Bruehl, S., Chung, O.Y., Burns, J.W. (2006). Trait anger and blood pressure recovery following acute pain:
evidence for opioid-mediated effects. International journal of behavioral medicine 13(2): 138–46.
Bruehl, S., Chung, O.Y., Burns, J.W., Biridepalli, S. (2003). The association between anger expression and
chronic pain intensity: evidence for partial mediation by endogenous opioid dysfunction. Pain 106(3):
317–24.
Bruehl, S., Chung, O.Y., Jirjis, J.N., Biridepalli, S. (2005). Prevalence of clinical hypertension in patients
with chronic pain compared to nonpain general medical patients. Clinical Journal of Pain 21(2): 147–53.
Bruehl, S., Chung, O.Y., Ward, P., Johnson, B. (2004a). Endogenous opioids and chronic pain intensity:
Interactions with level of disability. Clinical Journal of Pain 20(5): 283–92.
Bruehl, S., McCubbin, J.A., Carlson, C.R., et al. (1996). The psychobiology of hostility: possible
endogenous opioid mechanisms. International Journal of Behavioral Medicine 3:163–76.
Burns, J.W., Johnson, B.J., Devinem J, Mahoney, N., Pawl, R. (1998). Anger manger management style and
prediction of treatment outcome among male and female chronic pain patients. Behavior and Research
Therapy 36:1051–62.
Burns, J. W., Wiegner, S., Derleth, M., Kiselica, K., Pawl, R. (1997). Linking symptom-specific
physiological reactivity to pain severity in chronic low back pain patients: a test of mediation and
moderation models. Health Psychology 16(4): 319–26.
Cailliet, R. (1981). Low Back Syndrome. Philadelphia, PA, Davis.
Cano, A. (2004a). Pain catastrophizing and social support in married individuals with chronic pain: The
mediating role of pain duration. Pain 110: 656–64.
Cano, A., Gillis, M., Heinz, W., Geisser, M., Foran, H. (2004). Marital functioning, chronic pain, and
psychological distress. Pain 107:99–106.
Cano, A., Weisberg, J., Gallagher, M. (2000). Marital satisfaction and pain severity mediate the association
between negative spouse responses to pain and depressive symptoms in a chronic pain patient sample.
Pain Med 1:35–43.
Capodaglio, P., Nilsson, J. (1996). Functional correlates in the rehabilitation of occupational low back pain.
Gioliane Italiano di Medicina del Lavoro 18:35–9.
Carels, R.A., Szczepanski, R., Blumenthal, J.A., Sherwood, A. (1998). Blood pressure reactivity and marital
distress in employed women. Psychosomatic Medicine 60:639–43.
Cassisi, J.E., Robinson, M.E., O’Connor, P., MacMillian, M. (1993). Trunk strength and lumbar paraspinal
muscle activity during isometric exercise in chronic low-back pain patients and controls, Spine 18:
245–51.
Chataigner, H., Onimus, M., Polette, A. (1998). Surgery for degenerative lumbar disc disease. Should the
black disc be grafted? Revue de chirurgie orthopedique et reparatrice de’l appareil moteur 84(7): 583–9.
Cobb, C. R., DeVries, H.A., Urban, R.T., Luekens, C.A., Bagg, R.J. (1975). Electrical activity in muscle
pain. American Journal of Physical Medicine 54: 80–7.
Coffey, P. J., Girman, S., Wang, S.M., et al. (2001). Long-term preservation of cortically dependent visual
function in RCS rats by transplantation. Nature Neuroscience 5(1): 53–6.
Cohen, M. J., Swanson, G.A., Naliboff, B.D., Schandler, S.L., McArthur, D.L. (1986). Comparison of
electromyographic response patterns during posture and stress tasks in chronic low back pain patterns
and control. Journal of Psychosomatic Research 30(2): 135–41.
Collins, G., Cohen, M., Naliboff, B.D., Schandler, S.L. (1982). Comparative analysis of paraspinal and
frontalis EMG, heart rate and skin conductance in chronic low back pain patient and normals to
various postures and stress. Scandinavian Journal of Rehabilitative Medicine 14(1): 39–46.
Cordo, P.J., Nashner, L.M. (1982). Properties of postural adjustments associated with rapid arm
movements. Journal of Neurophysiology 47:287–302.
Cram, J. R., Steger, J.C. (1983). EMG scanning in the diagnosis of chronic pain. Applied Psychophysiology
and Biofeedback 8(2): 229–41.
Crombez, G., Vlaeyen, J.W.S., Heuts, P.H.T.G., Lysens, R. (1999). Pain-related fear is more disabling than
pain itself: evidence on the role of pain-related fear in chronic back pain disability, Pain 80: 329–39.
DeCharms, R. C., Maeda, F., Glover, G.H., et al. (2005). Control over brain activation and pain learned by
using real-time functional MRI. Proceedings of the National Academy of Sciences of the United States of
America 102(51): 18626–31.
DeCharms, R. C., Maeda, F., Glover, G.H., Ludlow, D., Pauly, J.M., Soneji, D., Gabrieli, J.D., Mackey, S.C.
(2005). Control over brain activation and pain learned by using real-time functional MRI. Proceedings
of the National Academic of Science of the United States of America 102: 18626–31.
DeGood, D.E., Stewart, W.R., Adams, L.E., Dale, J.A. (1994). Paraspinal EMG and autonomic reactivity of
patients with back pain and controls to personally relevant stress, Percept Mot Skills 79:1399–409.
DeVries, H. A. (1966). Quantitative electromyographic investigation of the spasm theory of muscle pain.
Journal of physical medicine 45: 119–34.
Diesch, E., Flor, H. (2007). Central processing of acute muscle pain in chronic low back pain patients: an
EEG mapping study. Journal of clinical neurophysiology 24(1): 76–83.
Ditto, B., France, C. (1990). Carotid baroreflex sensitivity at rest and during psychological stress in
offspring of hypertensives and non-twin sibling pairs. Psychosomatic Medicine 52: 610–20.
Dolce, J. J., Raczynski, J. M. (1985). Neuromuscular activity and electromyography in painful backs:
Psychological and biomechanical models in assessment and treatment. Psychological Bulletin 97: 502–20.
Droste, C., Kardos, A., Brody, S., Greenlee, M.W., Roskamm, H., Rau, H. (1994). Baroreceptor
stimulation: pain perception and sensory thresholds. Biological Psychology 37: 101–13.
Dworkin, B. R., Elbert, T., Rau, H., et al. (1994). Central effects of baroreceptor activation in humans:
Attenuation of skeletal reflexes and pain perception. Proceedings of the National Academy of Sciences of
the United States of America 91(14): 6329–33.
Eide, P. K. (2000). Wind-up and the NMDA receptor complex from a clinical perspective. European Journal
of Pain. 4: 5–17.
Elbert, T., Dworkin, B.R., Rau, H., Pauli, P., Birbaumer, N., Droste, C., Brunia, C.H. (1994). Sensory
effects of baroreceptor activation and perceived stress together predict long-term blood pressure
elevations. International Journal of Behavioral Medicine 1: 215–28.
Elbert, T., Ray, W.J., Kowalik, Z.J., Skinner, J.E., Graf, K.E., Birbaumer, N. (1994a). Chaos and physiology:
deterministic chaos in excitable cell assemblies. Physiological Review 74: 1–47.
Elbert, T., Rockstroh, B., Lutzenberger, W., Kessler, M., Pietrowsky, R. (1988). Baroreceptor stimulation
alters pain sensation depending on tonic blood pressure. Psychophysiology 25(1): 25–9.
Elbert T, Schandry, R.I.E. (1998). Wechselwirkungen zwischen kardiovaskulärem und zentralnervösem
System [Interactions between cardiovascular and central nervous system]. Enzyklopädie der Psychologie,
Serie Biologische Psychologie [Encyclopedia of Psychology, Biological Psychology]. F. Roesler. Göttingen,
Hogrefe: 427–77.
Elfving, B., Dedering, A., Németh, G. (2003). Lumber muscle fatigue and recovery in patients with long-
term back trouble–electromyography and health-related factors, Clin Biomech 18:619–30.
Fahrenberg, J. (1986). Psychophysiological individuality. A pattern analytic approach to personality
research and psychosomatic medicine. Advances in behavioral research and therapy 8: 43–100.
Fernandez, E., Turk, D.C. (1995). The scope and significance of anger in the experience of chronic pain.
Pain 61:165–75.
Fisher, A. A., Chang, C.H. (1985). Electromyographic evidence of muscle spasm during sleep in patients
with low back pain. Clinical Journal of Pain 1: 147–54.
Flor, H., Birbaumer, N. (1994). Acquisition of chronic pain: Psychophysiological mechanisms. American
Pain Society Journal 3: 119–27.
Flor, H., Birbaumer, N., Schugens, M.M., Lutzenberger, W. (1992). Symptom-specific psychophysiological
responses in chronic pain patients. Psychophysiology 29(4): 452–60.
Flor, H., Birbaumer, N., Turk, D. (1990). The psychobiology of chronic pain. Advances in Behaviour
Research and Therapy 12(2): 47–84
Flor, H., Braun, C., Elbert, T., Birbaumer, N. (1997). Extensive reorganization of primary somatosensory
cortex in chronic back pain patients. Neuroscience Letters 224(1): 5–8.
Flor, H., Breitenstein, C., Birbaumer, N., Fürst, M.A. (1995). A psychophysiological analysis of spouse
solitousness towards pain behaviors, spouse interaction and pain perception. Behavioral Therapy 26:
255–72.
Flor, H., Haag, G., Turk, D.C. and Koehler, H. (1983). Efficacy of EMG biofeedback, pseudotherapy and
medical treatment for chronic rheumatic back pain. Pain 17: 21–31.
Flor, H., Kerns, R.D., Turk, D.C. (1987). The role of spouse reinforcement, perceived pain, and activity
levels of chronic pain patients. Journal of Psychosomatic Research 31: 251–9.
Flor, H., Knost, B., Birbaumer, N. (2002). The role of operant conditioning in chronic pain: an
experimental investigation. Pain 95(1–2): 111–18.
Flor, H., Turk, D., Birbaumer, N. (1985). Assessment of stress-related psychophysiological reactions in
chronic back pain patients. Journal of Consulting and Clinical Psychology 53(3): 354–64.
Flor, H., Turk, D.C. (1984). Etiological Theories and Treatments for Chronic Back Pain. I. Somatic Models
and Interventions. Pain 19: 105–21.
Flor, H., Turk, D.C. (1988). Chronic Back Pain and Rheumatoid Arthritis: Predicting Pain and Disability
from Cognitive Variables. Journal of Behavioral Medicine 11(3): 251–65.
Flor, H., Turk, D.C. (1989). Psychophysiology of chronic pain: do chronic pain patients exhibit symptom-
specific psychophysiological responses? Psychological Bulletin 105: 215–9.
Fordyce, W. E. (1976). Behavioral Methods for Chronic Pain and Illness. St. Louis, MO, C V Mosby Co.
Gamboa-Esteves, F.O., Kaye, J.C., McWilliam, P.N., Lima, D., Batten, T.F.C. (2001).
Immunohistochemical profiles of spinal lamina I neurons retrogradely labeled from the nucleus tractus
solitarii in rat suggest excitatory projections. Neuroscience 104: 523–38.
Gelkopf, M. (1997). Laboratory pain and styles of coping with anger. Journal of Psychology 131:121–3.
Geisser, M.E., Robinson, M.E., Richardson, C. (1995). A time series analysis of the relationship between
ambulatory EMG, pain, and stress in chronic low back pain. Biofeedback Self Regul. 20(4): 339–55.
Gentry, W.D., Bernal, A. (1977). Chronic pain. Behavioral approaches to medical treatment . In
W. D. Genty and R.B. Williams (eds). New York, Ballinger: 173–91.
Ghione, S. (1996). Hypertension-associated hypalgesia. Evidence in experimental animals and humans,
pathophysiological mechanisms, and potential clinical consequences. Hypertension 28: 494–504.
Glombiewski, J.A., Hartwich-Tersek, J., Rief, W. (2010). Two psychological interventions are effective in
severely disabled, chronic back pain patients: a randomised controlled trial. International Journal of
Behavioral Medicine 17(2): 97–107.
Goldstein, D.S., Harris, A.H., Brady, J.V. (1977). Baroreflex sensitivity during operant blood pressure
conditioning. Biofeedback Self Regulation 2(2): 127–38.
Gracely, R.H., Geisser, M.E., Giesecke, T., et al. (2004). Pain catastrophizing and neural responses to pain
among persons with fibromyalgia. Brain 127: 835–43.
Gracely, R.H., Petzke, F., Wolf, J.M., Clauw, D.J. (2002). Functional magnetic resonance imaging evidence
of augmented pain processing in fibromyalgia. Arthritis & Rheumatism 46: 1333–43.
Greenough, C.G. (1993). Recovery from low back pain. 1–5 year follow-up of 287 injury-related cases.
Acta Orthopedica Scandinavia 64: (suppl 254): 1–64.
Gubler, D., Mannion, A.F., Schenk, P., et al. (2010). Ultrasound tissue Doppler imaging reveals no delay in
abdominal muscle feed-forward activity during rapid arm movements in patients with chronic low
back pain. Spine (Phila Pa 1976). 35(16): 1506–13.
Guzman, J., Esmail, R., Karjalainen, K.A., Malmivaara, A., Irvin, E., Bombardier, C. (2007). Withdrawn:
Multidisciplinary bio-psycho-social rehabilitation for chronic low-back pain (Review). Cochrane
Database of Systematic Reviews (Online) 18(2): CD000963.
Haley, W. E., Turner, J.A., Romano, J.M. (1985). Depression in chronic pain patients: relation to pain,
activity, and sex differences. Pain 23(4): 337–43.
Hanley, M. A., Raichle, K., Jensen, M., Cardenas, D.D. (2008). Pain catastrophizing and beliefs predict
changes in pain interference and psychological functioning in persons with spinal cord injury. Journal
of Pain 9(9): 863–71.
Harris, A. J. (1999). Cortical origins of pathological pain. Lancet 354: 1464–66
Haythornthwaite, J. A., Menefee, L.A., Heinberg, L.J., Clark, M.R. (1998). Pain coping strategies predict
perceived control over pain. Pain 77(1): 33–39.
Heffner, K.L., Kiecolt-Glaser, J.K., Loving, T.J., Glaser, R., Malarkey, W.B. (2004). Spousal support
satisfaction as a modifier of physiological responses to marital conflict in younger and older couples.
Journal of Behavioral Medicine 27(3): 233–54.
Heydari, A., Nargol, A.V., Jones, A.P., Humphrey, A.R., Greenough, C.G. (2010). EMG analysis of lumbar
paraspinal muscles as a predictor of the risk of low-back pain. European Spine Journal 19:1145–52.
Hildebrandt, J., Franz, C. (1988). Low back pain - pathophysiology of so called ‘idiopathic’ low back pain.
[Kreuzschmerz - Zur Pathophysiologie des sogenannten ‘idiopathischen’ Rueckenschmerzes]. In:
Sehathi-Chafai, G.H. (Editor) Pathophysiology, diagnostic and therapy of low back pain
[Pathophysiologie, Diagnostik und Therapie des Kreuzschmerzes. Winkler, Bochum, 11.
Hildebrandt, J., Pfingsten, M., Saur, P., Jansen, J. (1997). Prediction of success from a multidisciplinary
treatment program for chronic low back pain. Spine 22(9): 990–1001.
Hirsch, C. (1966). Low back pain etiology and pathogenesis. Applied Therapy 8: 857–62.
Hodapp, V., Weyer, G., Becker, J. (1975). Situational stereotypy in essential hypertension patients. Journal
of psychosomatic research 19(2): 113–21.
Hodges, P.W. (2001). Changes in motor planning of feedforward postural responses of the trunk muscles
in low back pain. Experimental Brain Research 141: 261–6.
Hoyt, W. H., Hunt, H.H., Depauw, M.A., et al. (1981). Electromyographic assessment of chronic low-back
pain syndrome. Journal of the American Osteopathic Association 80: 728–30.
Hult, L. (1954). Cervical, dorsal, and lumbar spinal syndromes. A field investigation of a non-selected
material of 1200 workers in different occupations with special reference to disc degeneration and
so-called muscular rheumatism. Acta orthopedica stand 17 (Suppl): 1–102.
Jalovaara, P., Niinimäki, T., Vanharanta, H. (1995). Pocket-size, portable surface EMG device in the
differentiation of low back pain patients. European Spine J 4:210–12.
James, W. (1984). What is an emotion? Mind 9: 188–205.
Janssen, S.J., Spinhoven, P., Brosschot, J.F. (2001). Experimentally induced anger, cardiovascular
reactivity, and pain sensitivity. Journal of Psychosomatic Research 51:479–85.
Jensen, M. C., Brant-Zawadzki, M.N., Obuchowski, N., et al (1994). Magnetic resonance imaging of the
lumbar spine in people without back pain. New England Journal of Medicine 14(331(2)): 69–73.
Jensen, M.P., Turner, J.A., Romano, J.M., Karoly, P. (1991). Coping with chronic pain: a critical review of
the literature. Pain , 47:249–83.
Jinkins, J. R. (2003). Lumbosacral interspinous ligament rupture associated with acute intrinsic spinal
muscle degeneration. JBR-BTR: Organe de la societe royal belge de radiologie (SRBR) 86(4): 226–30.
Jinkins, J. R. (2004). Acquired degenerative changes of the intervertebral segments at and suprajacent to the
lumbosacral junction. A radioanatomic analysis of the nondiscal structures of the spinal column and
perispinal soft tissues. European Journal of Radiology 50(2): 134–58.
Johannes, B., Salnitski, V.P., Thieme, K., Kirsch, K.A. (2003). Differences in the autonomic reactivity
pattern to psychological load in patients with hypertension and rheumatic diseases. Aerospace and
environmental Medicine 37(1): 28–42.
Kankaanpää, M., Taimela, S., Laaksonen, D., Hannien, O., Airksinen, O. (1998). Back and hip extensor
fatigability in chronic low back pain patients and controls. Arch Phys Med Rehabil 79:412–17.
Kerns, R.D., Haythornthwaite, J., Southwick, S., Giller, E. L. (1990). The role of marital interaction in
chronic pain and depressive symptom severity. Journal of Psychosomatic Research 34(4): 401–8.
Kerns, R.D. Rosenberg R, Jacob MC. (1994). Anger expression and chronic pain. Journal of Behavioral
Medicine 17:57–67.
Kerns, R.D., Turk, D.C. (1984). Depression and chronic pain: The mediating role of the spouse. Journal of
Marriage and Family 46: 845–52.
Klein, A.B., Snyder-Mackler, L., Roy, S.H., DeLuca, C.J. (1991). Comparison of spinal mobility and
isometric trunk extensor forces with electromyographic spectral analysis in identifying low back pain,
Phys Ther 71:445–54.
Kleinbohl, D., Holzl, R., Moltner, A., Rommel, C., Weber, C., Osswald, P.M. (1999). Psychophysical
measures of sensitization to tonic heat discriminate chronic pain patients. Pain 81(1–2): 35–43.
Kosek, E., Orderberg, G. (2000). Lack of pressure pain modulation by heterotopic noxious conditioning
stimulation in patients with painful osteoarthritis before, but not following, surgical pain relief. Pain
88: 69–78.
Kravitz, E. A., Moore, M.E., Glaros, A. (1981). Paralumbar muscle activity in chronic low back pain.
Archives of Physical Medicine and Rehabilitation 62: 172–6.
Kravitz, E. A. Glusman, S., Harris-Warrick, R.M., Livingstone, M.S., Schwarz, T., Goy, M.F. (1980).
Amines and a peptide as neurohormones in lobsters: actions on neuromuscular preparations and
preliminary behavioural studies. The Journal of experimental biology 98:159–75.
Ku, Y. H., Tan, L., Li, L.S., Ding, X. (1998). Role of corticotropin-releasing factor and substance P in
pressor responses of nuclei controlling emotion and stress. Peptides 19: 677–82.
Lacey, J., Lacey, B. (1958). Verification and extension of the principle of autonomic response-sterotype.
American Journal of Psychology 71: 50–73.
Lacey, J. I., Bateman, D.E., Van Lehn, R. (1953). Autonomic response specificity: An experimental study.
Psychosomatic medicine 15: 8–21.
LaChapelle, D. L., Hadjistavropoulos, H.D., McCreary, D.R., Asmundson, G.J. (2001). Contributions of
pain-related adjustment and perceptions of control to coping strategy use among cervical sprain
patients. European Journal of Pain 5(4): 405–13.
Lawler, J. E., Sanders, B.J., Cox, R.H., O’Connor, E.F. (1991). Baroreflex function in chronically stressed
borderline hypertensive rats. Physiological Behavior 49: 539–42.
Leavitt, F. (1992). The physical exertion factor in compensable work injuries: a hidden flaw in previous
research. Spine 17: 307–10.
Lee, D.J., Stokes, M.J., Taylor, R.J., Cooper, R.G. (1992). Electro and acoustic myography for noninvasive
assessment of lumbar paraspinal muscle function, Eur J Appl Physiol Occup Physiol 64:199–203.
Leeuw, M., Houben, R.M.A., Severeijns, R., et al. (2007). Pain-related fear in low back pain: A prospective
study in the general population. European Journal of Pain 11: 256–66.
Lehmann, H. (1996). The content and development of mobbing at work. European Journal of Work and
Organizational Psychology 5(2): 165–84.
Leinonen, V., Kankaanpää, M., Luukkonen, M., Hänninen, O., Airaksinen, O., Taimela, S. (2001). Disc
herniation-related back pain impairs feed-forward control of paraspinal muscles. Spine 26: E367–72.
Lethem, J., Slade, P.D., Troup, J.D., Bentley, G. (1983). Outline of a Fear-Avoidance Model of exaggerated
pain perception—I. Behaviour Research and Therapy 21(4): 401–8.
Li, J., Simone, D.A., Larson, A.A. (1999). Windup leads to characteristics of central sensitization. Pain 79:
75–82.
Linton, S. J. (2000). A review of psychological risk factors in back and neck pain. Spine 25: 1148–56.
Lisiński, P. (2000). Surface EMG in chronic low back pain. Eur Spine J 9:559–62.
Loeser, J. D. (1980). Low back pain. New York, Raven Press.
Lofland, K.R., Cassisi, J.E., Levin, J.B., Palumbo, N.L., Blonsky, E.R. (2000). The incremental validity of
lumbar surface EMG, behavioral observation, and a symptom checklist in the assessment of patients
with chronic low-back pain. Appl Psychophysiol Biofeedback 25:67–78.
Lousberg, R., Schmidt, A.J., Groenman, N.H. (1992). The relationship between spouse solicitousness and
pain behavior: searching for more experimental evidence. Pain 51: 75–9.
Lu, W.W., Luk, K.D., Cheung, K.M., Wong, Y.W., Leong, J.C. (2001). Back muscle contraction patterns of
patients with low back pain before and after rehabilitation treatment: an electromyographic evaluation.
J Spinal Disord 14:277–82.
Lundberg, U., Dohns, I.E., Melin, B., et al. (1999). Psychophysiological stress responses, muscle tension,
and neck and shoulder pain among supermarket cashiers. Journal of Occupational Health Psychology
4(3): 245–55.
Maixner, W. (1991). Interactions between cardiovascular and pain modulatory systems: physiological and
pathophysiological implications. Journal of Cardiovascular Electrophysiology 2: S3–S12.
Maixner, W., Fillingim, R., Booker, D., Sigurdsson, A. (1995). Sensitivity of patients with painful
temporomandibular disorders to experimentally evoked pain. Pain 63: 341–51.
Maixner, W., Fillingim, R.B., Kincaid, S., Sigurdsson, A. Harris, M.B. (1997). Relationship between pain
sensitivity and resting arterial blood pressure in patients with painful temporomandibular disorders.
Psychosomatic Medicine 59: 503–11.
Maixner, W., Sigurdsson, A., Fillingim, R.B., Lundeen, T., Booker, D.K. (1995a). Regulation of acute and
chronic orofacial pain. Orofacial pain and temporomandibular disorders. D. R. Fricton JR. New York,
Raven Press.
Malchaire, J., Cock, N., Vergracht, S. (2001). Review of the factors associated with musculoskeletal problems
in epidemiological studies. International Archives of Occupational Environmental Health 74: 79–90.
Malmo, R. B., Shagass, C. (1949). Physiologic studies of reaction to stress in anxiety and early
schizophrenia. Psychosomatic Medicine 11: 9–24.
Mannion, A. F., Pulkovski, N., Schenk, P., et al. (2008). A new method for the noninvasive determination
of abdominal muscle feedforward activity based on tissue velocity information from tissue Doppler
imaging. Journal of Applied Physiology 104(4): 1192–201.
Mayer, T.G., Kondraske, G., Mooney, V., Carmichael, T.W., Butsch, R. (1989). Lumbar myoelectric
spectral analysis for endurance assessment. A comparison of normals with deconditioned patients.
Spine 14: 986–91.
McCabe, C. S., Haigh, R.C., Halligan, P.W. Blake, D.R. (2005). Simulating sensory–motor incongruence in
healthy volunteers: implications for a cortical model of pain. Rheumatology (Oxford) 44: 509–16.
McCracken, L.M., Zayfert, C., Gross, R.T. (1992). The Pain Anxiety Symptoms Scale: development and
validation of a scale to measure fear of pain. Pain 50: 63–7.
Mense, S. (1993). Nociception from skeletal muscle in relation to clinical muscle pain. Pain 54: 241–89.
Millan, M. J. (2002). Descending control of pain. Progressive Neurobiology 66: 355–474.
Miller, D. J. (1985). Comparison of electromyographic activity in the lumbar paraspinal muscles of subjects
with and without chronic low back pain. Physical Therapy 65(9): 1347–54.
Mini, A., Rau, H., Montoya, P., Palomba, D., Birbaumer, N. (1995). Baroreceptor cortical effects,
emotions, and pain. International Journal of Psychophysiology 19(1): 67–77.
Mitchell, V. P., Lawler, J.E. (1989). Norepinephrine content of discrete brain nuclei in acutely and
chronically stressed borderline hypertensive rats. Brain Research Bulletin 22: 545–47.
Morris, S.L., Allison, G.T. (2006). Effects of abdominal muscle fatigue on anticipatory postural adjustments
associated with arm raising. Gait Posture 24:342–8.
Myers, C. D., Robinson, M. E., Riley, J.L. 3rd, Sheffield, D. (2001). Sex, gender, and blood pressure:
contributions to experimental pain report. Psychosomatic Medicine 63: 545–50.
Newton-John, T. R., Spence, S.H., Schotte, D. (1995). Cognitive-behavioural therapy versus EMG
biofeedback in the treatment of chronic low back pain. Behavior Research and Therapy 33(6): 691–7.
Ng, J.K. -F., Richardson, C.A., Parnianpour, M., Kippers, V. (2002). EMG activity of trunk muscles and
torque output during isometric axial rotation exertion: a comparison between back pain patients and
matched controls. Journal of Orthopedic Research 20:112–21.
Nilsson, P. M., Kandell-Collen, A., Andersson, H.I. (1997). Blood pressure and metabolic factors in
relation to chronic pain. Blood Pressure 6: 294–8.
Nouwen, A., Bush, C. (1984). The relationship between paraspinal EMG and chronic low back pain. Pain
20: 109–23.
Okifuji, A., Turk, D.C., Curran, S.L. (1999). Anger in chronic pain: investigations of anger targets and
intensity. Journal of Psychosomatic Research 47:1–12.
Ongley, M. J., Klein, R.G., Dorman, T.A., Eek, B.C., Hubert, L.J. (1987). A new approach to the treatment
of chronic low back pain. Lancet 18(2(8551)): 143–6.
Ostelo, R. W., van Tulder, M.W., Vlaeyen, J.W., Linton, S.J., Morley, S.J., Assendelft, W.J. (2005).
Behavioural treatment for chronic low-back pain. Cochrane Database of Systematic Reviews 25(1):
CD002014.
Passatore, M., Roatta, S. (2006). Influence of sympathetic nervous system on sensorimotor function:
whiplash associated disorders (WAD) as a model. European Journal of Applied Physiology 98:
423–49.
Pääsuke, M., Johanson, E., Proosa, M., Ereline, J., Gapeyeva, H. (2002). Back extensor muscle fatigability
in chronic low back pain patients and controls: Relationship between electromyogram power spectrum
changes and body mass index. Journal of Back Musculoskeletal Rehabilitation 16:17–24.
Peach, J.P., McGill, S.M. (1998). Classification of low back pain with the use of spectral electromyogram
parameters. Spine 23:1117–23.
Pertovaara, A. (1998). A neuronal correlate of secondary hyperalgesia in the rat spinal dorsal horn is
submodality selective and facilitated by supraspinal influence. Experimental Neurology 149: 193–202.
Peters, M. L., Schmidt, A.J.M. (1991). Psychophysiological responses to repeated acute pain stimulation in
chronic low back pain patients. Journal of Psychosomatic Research 35(1): 59–74.
Peters, M. L., Schmidt, A.J.M., Van den Hout, M.A. (1989). Chronic low back pain and the reaction to
repeated pain stimulation. Pain 39: 69–76.
Pienimäki, T., Tarvainen, T., Siira, P., Malmivaara, A., Vanharanta, H. (2002). Associations between pain,
grip strength, and manual tests in the treatment evaluation of chronic tennis elbow. Clinical Journal of
Pain 18(3): 164–70.
Pincus, T., Burton, A.K., Vogel, S., Field, A.P. (2002). A systematic review of psychological factors as
predictors of chronicity/disability in prospective cohorts of low back pain. Review. Spine 27(5):
E109–20.
Ploghaus, A., Narain, C., Beckmann, C.F., et al. (2001). Exacerbation of pain by anxiety is associated with
activity in a hippocampal network. Journal of Neuroscience 21(24): 9896–903.
Ploghaus, A., Tracey, I., Gati, J.S., et al. (1999). Dissociating pain from its anticipation in the human brain.
Science 284(5422): 1979–81.
Polatin, P. B., Kinney, R.K., Gatchel, R.J., Lillo, E., Mayer, T.G. (1993). Psychiatric illness and chronic back
pain. The mind and the spine—which goes first? Spine 18: 66–71.
Praemer, A., Furnes, S., Rice, D.P. (1992). Musculoskeletal conditions in the United States. Rosemont, AAUS.
Qian, Z.M., X. D., Huang, W.Q., Tang, P.L., Xu, B. (1997). Central ANG II receptor involved in carotid
sinus reflex resetting in chronically stressed rats. Physiological Behavior 62: 241–7.
Radebold, A., Cholewicki, J., Panjabi, M., Patel, T. (2000). Muscle response pattern to sudden trunk
loading in healthy individuals and in patients with chronic low back pain. Spine 25: 947–54.
Radebold, A., Cholewicki, J., Polzhofer, G.K., Greene, H.S. (2001). Impaired postural control of the lumbar
spine is associated with delayed muscle response times in patients with chronic idiopathic low back
pain. Spine 26: 724–30.
Randich, A., Gebhart, G.F. (1992). Vagal afferent modulation of nociception. Brain Research Brain Research
Reviews 17(2): 77–99.
Randich, A., Maixner, W. (1984). Interactions between cardiovascular and pain regulatory systems.
Neuroscience & Biobehavioral Review 8: 343–67.
Rau, H., Elbert, T. (2001). Psychophysiology of arterial baroreceptors and the etiology of hypertension.
Biological Psychiatry 57: 179–201.
Rau, H., Pauli, P., Brody, S., Elbert, T., Birbaumer, N. (1993). Baroreceptor stimulation alters cortical
activity. Psychophysiology 30: 322–5.
Rissén, D., Melin, B., Sandsjö, L., Dohns, I., Lundberg, U. (2000). Surface EMG and psychophysiological
stress reactions in women during repetitive work. European Journal of Applied Psychology 83(2–3): 215–22.
Robinson, M.E., Cassisi, J.E., O’Connor, P.D., MacMillian, M. (1992). Lumbar iEMG during isotonic
exercise: chronic low back pain patients versus controls. Journal of Spinal Disorder 5: 8–15.
Rohen, J. W. (1978). Functional anatomy of nervous system. Stuttgart, Schattauer.
Rossignol, M., Suissa, S., Abenhaim, L. (1988). Working disability due to occupational back pain: three-
year follow-up of 2300 compensated workers in Quebec. Journal of Occupational Medicine 30: 502–5.
Roy, S.H., De Luca, C.J., Casavant, D.A. (1989). Lumbar muscle fatigue and chronic lower back pain. Spine
14:992–1001.
Roy, S.H., De Luca, C.J., Emley, M., Buijs, R.J. (1995). Spectral electromyographic assessment of back
muscles in patients with low back pain undergoing rehabilitation, Spine 20:38–48.
Roy, S.H., De Luca, C.J., Snyder-Mackler, L., Emley, M.S., Crenshaw, R.L., Lyons, J.P. (1990). Fatigue,
recovery, and low back pain in varsity rowers. Medical Science of Sports Exercises 22: 463–69.
Rutecki, P. (1990). Anatomical, physiological, and theoretical basis for the antiepileptic effect of vagus
nerve stimulation. Epilepsia 31: S1–S6.
Ryan, C. G., Gray, H.G., Newton, M., Granat, M.H. (2010). The relationship between psychological distress
and free-living physical activity in individuals with chronic low back pain. Manual therapy 15(2): 185–9.
Sandkuehler, J. (2000). Learning and memory in pain pathways. Pain 88: 113–18.
Shirado, O., Ito, T., Kaneda, K., Strax, T. (1995). Flexion-relaxation phenomenon in the back muscles:
a comparative study between healthy subjects and patients with chronic low back pain. American
Journal of Physical and Medical Rehabilitation 74:139–44.
Schneider, S., Schmitt, H., Zoller, S., Schiltenwolf, M. (2005). Workplace stress, lifestyle and social factors
as correlates of back pain: a representative study of the German working population. International
Archives of Occupational and Environmental Health 78(4): 253–69.
Seagard, J.L., Dean, C., Hopp, F.A. (2000). Modulation of the carotid baroreceptor reflex by substance P in
the nucleus tractus solitarius. Journal of Autonomic Nervous System 78: 77–85.
Selye, H. (1956). Stress and disease. New York, McGraw-Hill.
Shapiro, D., Goldstein, I.B., Jamner, L.B. (1996). Effects of cynical hostility, anger out, anxiety, and
defensiveness on ambulatory blood pressure in black and white college students. Psychosomatic
Medicine, 58: 354–64.
Sihvonen, T., Huttunen, M., Makkonen, M., Airaksinen, O. (1998). Functional changes in back muscle
activity correlate with pain intensity and prediction of low back pain during pregnancy. Archive of
Physical and Medical Rehabilitation 79:1210–12.
Sihvonen, T., Partanen, J., Hänninen, O., Soimakallio, S. (1991). Electric behavior of low back muscles
during lumbar pelvic rhythm in low back pain patients and healthy controls, Archive of Physical and
Medical Rehabilitation 72:1080–7.
Sjörs, A., Larsson, B., Dahlman, J., Falkmer, T., Gerdle, B. (2009). Physiological responses to low-force
work and psychosocial stress in women with chronic trapezius myalgia. BMC_Musculoskeletal disorders
7(10): 63.
Smeets, R. J., Wade, D., Hidding, A., Van Leeuwen, P.J., Vlaeyen, J.W., Knottnerus, J.A. (2006). The
association of physical deconditioning and chronic low back pain: a hypothesis-oriented systematic
review. Disability and Rehabilitation 28(11): 673–93.
Soderberg, G.L., Cook, T.M. (1983). An electromyographic analysis of quadriceps femoris muscle setting
and straight leg raising. Physical Therapy 63(9): 1434–8.
Spielberger, C.D., Johnson, E.H., Russell, S.F., Crane, R.J., Jacobs, G.A., Worden, T.J. (1985).
The experience and expression of anger: construction and validation of an anger expression scale.
In: Chesney MA, Rosenman RH, editors. Anger and hostility in cardiovascular and behavioral
disorders, Washington, DC: Hemisphere Publishing, 1985. pp. 5–30.
Steriade, M. (1988). New vistas on the morphology, chemical transmitters and physiological actions of the
ascending brainstem reticular system. Archives Italiennes de Biologie 126(4s): 225–38.
Steriade, M., Llinas, R.R. (1988). The functional states of the thalamus and the associated neuronal
interplay. Physiology Review 68: 649–742.
Sullivan, M. J. L., Rodgers, W.M., Kirsch, I. (2001). Catastrophizing, depression and expectancies for pain
and emotional distress. Pain 91(1–2): 147–54.
Suter, E., Lindsay, D. (2001). Back muscle fatigability is associated with knee extensor inhibition in subjects
with low back pain. Spine 26: E361–66.
Thieme, K., Rose, U., Pinkpank, T., Spies, C., Turk, D.C., Flor, H. (2006). Psychophysiological responses in
patients with fibromyalgia syndrome. Journal of Psychosomatic Research 61(5): 671–79.
Thieme, K., Spies, C., Sinha, P., Turk, D.C., Flor, H. (2005). Predictors of pain behaviors in fibromyalgia
syndrome. Arthritis & Rheumatism 53(3): 343–50.
Thieme, K., Turk, D.C. (2006). Heterogeneity of psychophysiological stress responses in fibromyalgia
syndrome patients. Arthritis Research & Therapy 8(1): R9.
Thieme, K., Turk, D.C., Gracely, R.H., Maixner, W., Flor, H. Interaction of psychological and
psychophysiological characteristics in fibromyalgia patients. Submitted.
Thieme, K., Turk, D.C., Flor, H. (2004). Comorbid depression and anxiety in fibromyalgia syndrome:
relationship to somatic and psychosocial variables. Psychosomatic Medicine 66(6): 837–44.
Tousignant-Laflamme, Y., Marchand, S. (2006). Sex differences in cardiac and autonomic response to
clinical and experimental pain in LBP patients. European Journal of Pain 10:603–14.
Toyone T, T. K., Kitahara H, Yamagata M, Murakami M, Moriya H. (1994). Vertebral bone-marrow
changes in degenerative lumbar disc disease. An MRI study of 74 patients with low back pain. Journal
of Bone and Joint Surgery 46(5): 757–64.
Tschannen, T.A., Duckro, P.N., Margolis, R.B., Tomazic, T.J. (1992). The relationship of anger,
depression, and perceived disability among headache patients. Headache 32:501–3.
Turk, D. C., Flor, H. (1984). Etiological theories and treatments for chronic back pain. II. Psychological
models and interventions. Pain 19(3): 209–33.
Turk, D. C., Meichenbaum, D., Oenest, M. (1983). Pain and Behavioral Medicine. A Cognitive-Behavioral
Perspective. New York, Guilford Press.
Turk, D. C., Okifuji, A. (1997). What factors affect physicians’ decisions to prescribe opioids for chronic
noncancer pain patients? Clinical Journal of Pain 13(4): 330–6.
Turk, D. C., Okifuji, A. (1997a). Evaluating the role of physical, operant, cognitive, and affective factors in
the pain behaviors of chronic pain patients. Behavioral Modification 21: 259–80.
Turk, D. C., Okifuji, A., Starz, T.W., Sinclair, J.D. (1996). Effects of type of symptom onset on
psychological distress and disability in fibromyalgia syndrome patients. Pain 68(2–3): 423–30.
Turner, J. A., Chapman, C. R. (1982). Psychological interventions for chronic pain: A critical appraisal. II.
Operant conditioning, hypnosis, and cognitive-behavior therapy. Pain 12: 23–46.
Turner, J. A., Jensen, M.P., Warms, C.A., Cardenas, D.D. (2002). Catastrophizing is associated with pain
intensity, psychological distress, and pain-related disability among individuals with chronic pain after
spinal cord injury. Pain 98(1–2): 127–34.
van Dieën, J. H., Selen, L.P.J., Choloewicki, J. (2003). Trunk muscle activation in low-back pain patients,
an analysis of the literature. Journal of Electromyography and Kinesiology 13(4): 333–51.
van Doorn, T. W. C. (1995). Low back disability among self-employed dentists, veterinarians, physicians
and physical therapists in the Netherlands. Acta Orthopedica Scandinavia 66 (suppl 263): 1–64.
Verbunt, J. A., Seelen, H.A., Vlaeyen, J.W., et al. (2003). Disuse and deconditioning in chronic low
back pain: concepts and hypotheses on contributing mechanisms. Review. European Journal of Pain
7(1): 9–21.
Villanueva, L., Le Bars, D. (1995). The activation of bulbo-spinal controls by peripheral nociceptive inputs:
diffuse noxious inhibitory controls. Biological Research 28: 113–25.
Vlaeyen J.W.S., Crombez G. (1999). Fear of movement/(re)injury, avoidance and pain disability in chronic
low back pain patients. Manual Therapy 4(4): 187–95.
Vlaeyen, J. W. S., Kole-Snijders, A.M.J., Boeren, R.G.B., van Eek H. (1995). Fear of movement/(re)injury
in chronic low back pain and its relation to behavioral performance. Pain 62(3): 363–72.
Vowles, K. E., McCracken, L.M., Eccleston, C. (2007). Processes of change in treatment for chronic
pain: The contributions of pain, acceptance, and catastrophizing. European Journal of Pain
11: 779–87.
Waddell, G., Newton, M., Henderson, I., Somerville, D., Main, C.J. (1993). A fear-avoidance beliefs
questionnaire (FABQ) and the role of fear avoidance beliefs in chronic low back pain and disability.
Pain 52: 157–68.
Waersted, M. (2000). Human muscle activity related to non-biomechanical factors in the workplace.
European Journal of applied physiology 83(2–3): 151–8.
Waersted, M., Bjøklund, R., Westgaard, R.H. (1991). Shoulder muscle tension induced by two VDT-based
tasks of different complexity. Ergonomics 34: 137–50.
Watkins, L. R., Maier, S.F. (1999). Cytokines and pain: progress in inflammation research. Boston,
Birkhauser.
Watkins, L. R., Maier, S.F. (1999a). Implications of immune-to-brain communication for sickness and
pain. Proceedings of the National Academy of Sciences of the United States of America 96: 7710–13.
Watkins, L.L., Blumethal, J.A., Carney, R.M. (2002). Association of anxiety with reduced baroreflex cardiac
control in patients after acute myocardial infarction. American Heart Journal 143:460–6.
Watson, P. J., Booker, C.K., Main, C.J., Chen, A.C. (1997). Surface electromyography in the identification
of chronic low back pain patients: the development of the flexion relaxation ratio. Clinical
Biomechanics 12(3): 165–71.
Waxman, S. E., Tripp, D.A., Flamenbaum, R. (2008). The mediating role of depression and negative
partner responses in chronic low back pain and relationship satisfaction. Journal of Pain 9(5): 434–42.
Westgaard, R. H., Bjørklund, R. (1987). Generation of muscle tension additional to postural muscle load.
Ergonomics 30(6): 911–23.
Widerström-Noga, E.G., Felix, E.R., Cruz-Almeida, Y., Turk, D.C. (2007). Psychosocial subgroups in
persons with spinal cord injuries and chronic pain. Archives of Physical Medicine and Rehabilitation
88(12): 1628–35.
Wirtelak, E.P., Roemer, B., Maier, S.F., Watkins, L.R. (1997). Comparison of the effects of nucleus tractus
solitarius and ventral medial medulla lesions on illness-induced and subcutaneous formalin-induced
hyperalgesia. Brain Research 748:143–50.
Woby, S., Urmston, M., Watson, P. (2007). Self-efficacy mediates the relation between pain-related fear
and outcome in chronic low back pain patients. European Journal of Pain 11(7): 711–18.
Wolf, S.L., Nacht, M., Kelly, J.R. (1982). EMG feedback training during dynamic movement for low back
pain patients. Behavioral Therapy 13: 395–406.
Yu, X.M., Mense, S. (1990). Response properties and descending control of rat dorsal horn neurons with
deep receptive fields. Neuroscience 39(3): 823–31.
Part 3

Biomechanical Mechanisms
Chapter 9
Electromyographically-Determined
Muscular Fatigue in Low Back Pain
Anne F. Mannion and David O’Riordan
9.1 The theoretical basis for a link between back muscle

function and back pain
The back muscles initiate and control all movements of the vertebral column: they are involved
in minor active movements of the spine, such as the initiation of manoeuvres that are later
assisted by gravity; they contribute to the maintenance of posture when, for example, the move-
ment of other body segments acts to displace the centre of gravity; and they contribute to major
movements during forward bending and lifting (Bogduk and Twomey 1991). By controlling these
movements, the back muscles also act to stabilize and protect the underlying osteoligamentous
spinal structures from potentially harmful stresses that might otherwise be experienced by move-
ments made beyond their optimal functional range and/or for protracted periods of time (Adams
and Hutton, 1986). It is not difficult, therefore, to imagine the potential consequences associated
with malfunctioning of this muscle group.
Over the last two decades many studies have documented an association between suboptimal
back muscle function and the presence of low back pain (LBP). Deficits in the cross-sectional area
(Hultman et al. 1993), muscle density (Hultman et al. 1993), maximal force production (Klein
et al. 1991; Mooney et al. 1997; Elfving et al. 2003; Lariviere et al. 2003a; Crossman et al. 2004) and
local muscular endurance (or fatigue-resistance) (Nicolaisen and Jorgensen 1985; Jorgensen
and Nicolaisen 1987; Roy et al. 1989; Mayer et al. 1989; Biedermann et al. 1991; Klein et al.
1991; Cooper et al. 1993; Tsuboi et al. 1994; Roy et al. 1995; Candotti et al. 2008) of the muscles
have all been observed, particularly in association with chronic pain. However, despite intensive
investigation, a definitive answer to the age-old question of cause and effect—‘what comes first,
the back pain or the muscle dysfunction?’—remains elusive. From a theoretical point of view, it
certainly seems conceivable that inadequate muscular strength could predispose to the develop-
ment of LBP.
The application of a load to the hands requires that the back extensors generate an appropriate
antiflexion moment to balance or raise the load against the force of gravity. If sufficient force
cannot be generated by active muscle contraction, then the loading may be inadvertently dis-
placed to the passive tissues integrated within and surrounding the osteoligamentous spine.
The reduced joint protection, stability, and support may be even more pronounced under
conditions of uncontrolled or sudden loading (Wilder et al. 1996), which can be a prominent
feature of many manual handling environments. The back extensors are primarily ‘slow’ postural
muscles (Mannion et al. 1997c), with a well-developed capacity to sustain static forces, but at the
expense of reaction time and speed of contraction (see later). Particularly when loading is applied
unexpectedly, these muscles may be unable to generate the required force rapidly enough to
prevent excessive bending or twisting of the spine. Rapid trunk flexion movements will generate
particularly high bending moments on the spine, because spinal tissues are viscoelastic and resist
rapid deformations more vigorously than slow ones (Adams and Dolan 2005).
Despite the plausible hypothesis implicating inadequate back muscle strength in the develop-
ment of first-time back pain, there is little evidence to support it (Biering-Sorenson 1984; Mostardi
et al. 1992; Kujala et al. 1996; Adams et al. 2005). This may be because inappropriate tests of
maximal strength have been used, or it might simply suggest that strength requirements are
highly job-specific and that the situations in which the back muscles fail to adequately per-
form their supporting role are mostly random and unpredictable, occurring in connection with
accidental events.
The mechanism by which back muscle fatigability could be causally associated with the develop-
ment of LBP is essentially analogous to that discussed above for insufficient strength, in so far as
fatigued muscles are actually temporarily weakened (Mannion and Dolan 1996b) and slower
(Duchateau and Hainaut 1984) muscles. Thus, during repetitive lifting, the fatigued back muscles
may be less effective in generating the required extensor moment, forcing the intervertebral discs
and ligaments to withstand a relatively greater bending stress and thereby rendering them more
susceptible to injury. Three studies have shown that there is a progressive increase in the degree
to which the lumbar spine is flexed during repetitive lifting (Van Doorn 1995; Sparto et al. 1997b;
Dolan and Adams 1998), sometimes to an extent that affords only a narrow margin of safety with
regards to potential mechanical injury to the intervertebral discs (Dolan and Adams 1998).
Further, with back muscle fatigue significant changes have been reported in the angular displace-
ments at the knee, hip, trunk, and elbow during lifting, and these biomechanical changes were
associated with increased peak torque and forces at the L4–L5 vertebral segment (Bonato et al.
2003).
Other reports have demonstrated that back muscle fatigue is accompanied by an alteration in
anticipatory postural adjustments (Allison and Henry 2002), a reduction in accuracy when trying
to generate a given force (Sparto et al. 1997a), impaired sensory (positional) acuity (Taimela et al.
1999), delayed responses to sudden loading (Magnusson et al. 1996), reduced active muscle stiff-
ness necessitating increased antagonist co-contraction (and hence greater spinal compression) to
maintain spinal stability (Granata et al. 2004), and an altered coordination of trunk muscle
activities, with reduced precise motor control, such that the spine is loaded in a more injury prone
pattern (Parnianpour et al. 1988). Each of these could render the lumbar spine more vulnerable
to injury during repetitive lifting. Whilst people are often aware of their absolute strength limita-
tions, and can make a considered judgement as to whether they should attempt a given manoeu-
vre or not, the same is probably not true in the case of muscle fatigue, the onset and consequences
of which can be considerably more difficult to predict. This is nicely demonstrated by the familiar
scenario in which we judge the weight of a load and deem it manageable to carry, only to realize
a short time later—whilst struggling to sustain the effort—that we have failed to account for the
continuously declining capacity of the fatiguing muscles. The inability to accurately anticipate
fatigue may thus result in individuals being ‘caught off guard’ at a time when they can least afford
to be.
All these findings supported the hypothesis that readily fatigable back muscles might predis-
pose to LBP or be part of a vicious circle in which LBP leads to suboptimal muscle function (con-
sequent to inactivity, disuse or continuous overactivity/overload) which in turn leads to inadequate
protection of the osteoligamentous spine, rendering the individual more liable to subsequent
episodes of pain or the development of chronic symptoms (Figure 9.1). The evidence supporting
this role for back muscle fatigability in LBP will be considered later, once the methods used for its
assessment have been discussed.
EMG-DETERMINED MUSCULAR FATIGUE 157
Low
back pain
Inactivity/
“Tissue
disuse
overload”
Functional Alterations in
Reduced joint deficits structure and
protection, physiology
stability, support E.g. reduced back
muscle endurance
(fatigue-resistance)
Fig. 9.1 The vicious circle of the ‘deconditioning syndrome’, originally described by Mayer et al.
(1985). For more information see Mayer et al. (1985).
9.2 Measurement of back muscle fatigue

Some of the earliest studies examining the predictive role of back muscle endurance and fatigabil-
ity in LBP involved the use of the now well-known Biering-Sorensen test (Biering-Sorenson
1984). The test is shown in Figure 9.2. Basically, it involves placing the individual in a prone posi-
tion on an examination couch with the lower body (i.e. from the anterior superior iliac spine
distally) strapped firmly to the couch. Individuals attempt to maintain the unsupported upper
body in a horizontal position until they are no longer able to overcome the force of gravity. As
soon as the upper body ceases to be horizontal, and the person is unable to rectify the situation
when urged to do so, the test is terminated. Verbal encouragement is typically given throughout,
and the endurance time for the test is recorded to the nearest whole second. Variations on the
method (Demoulin et al. 2006) concern the positioning of the arms and hands, the equipment
used for the test, and the extent to which correct positioning is assessed.
Other methods for assessing the endurance capacity involve the initial determination of the
maximum voluntary contraction (MVC) of the back extensors, using either commercial devices
or custom-built equipment (e.g. Figure 9.2), and the subsequent maintenance of a proportion of
this maximum for as long as the individual can endure. However, when applied to the clinical
population, the need for a maximum contraction in order to set the submaximal load can be
problematic, since issues such as fear of (re)injury, pain, etc., can lead to suboptimal performance
and hence inaccurate load-setting for the submaximal test. Since the endurance time during a
fatiguing task is dependent on the relative force output (Hagberg 1981; Mannion and Dolan
1996b), this has important implications when comparing individuals: if they have not been set the
same test to start with, then the comparison of their endurance capacity is futile. Moreover, since
measurement of MVC may pose a real risk of injury (Hansson et al. 1984), the use of such tests is
necessarily accompanied by various ethical considerations.
Since its introduction, the definition of fatigue as ‘the inability to maintain a predetermined
exercise intensity’ (Ciba Foundation 1981) has been widely accepted—particularly in the study of
metabolic mechanisms responsible for human skeletal muscle fatigue. However, it is not entirely
satisfactory, especially in relation to the conditions typically encountered in the clinical situation.
The definition implies that fatigue is an event that occurs at a specific point in time and it there-
fore fails to consider the changing conditions within the muscle that ultimately precipitate that
failure point. It also necessitates that one be able to drive oneself to this failure point in order to
158
(a) (b)
(d)
(c)
Fig. 9.2 The four test positions most commonly used for performing back muscle fatigue tests: a) upright standing, pushing backwards; b) standing with
flexed trunk pulling up on a bar attached to a load cell between the feet; c) prone, Biering-Sorensen test; d) seated, hips and knees flexed, using back
extension machine.
make the assessment. In practice, when measuring an individual’s ability to sustain a given force
or power output, the endurance time is influenced not only by the intrinsic physiological qualities
of the muscle but also by factors such as motivation, boredom, tolerance of the discomfort of
fatiguing muscles, and in the clinical situation perhaps also by actual pain or fear of pain. It there-
fore adopts a psychological dimension.
Although it may be of importance to know the maximum length of time that an individual can
sustain a contraction, it is also useful to assess the changes in the muscle that lead up to the even-
tual failure point, not least in order to predict when it might occur. In some studies, this has been
done by monitoring the changing metabolic status of the muscle during contraction (e.g. the
increasing lactate concentration, or declining pH and intracellular potassium ion concentra-
tions). However, such measures require invasive techniques or the use of sophisticated equip-
ment such as 31P-NMR, and are hence not suitable for clinical studies. For these reasons, the use
of surface electromyography became an increasingly popular tool for monitoring the progres-
sively declining capacity of the muscles during fatiguing contractions (De Luca 1997; Jurell 1998).
The technique typically involves examination of the change in the frequency distribution of the
power density spectrum of the electromyographic (EMG) signal, obtained by fast Fourier trans-
form of the EMG signal.
During sustained isometric contraction, a compression of the spectrum to lower frequencies is
observed (Figure 9.3), and this is tracked using a characteristic component of the spectrum, such
as its median frequency (MF) or mean power frequency (MPF), plotting this against time using
linear regression techniques. The slope of the decline in MF (in either Hz/s or %/s, normalized to
the initial MF, i.e. the intercept of the regression curve) is then taken to indicate the fatigability of
the muscle. Previous work has shown that the rate of decline in MF is highly correlated with
endurance time (Hagberg 1981; Mannion and Dolan 1994) and also with the accumulation of
certain muscle metabolites that have been implicated in the development of muscle fatigue
(Bouissou et al. 1989; Brody et al. 1991; Vestergaard-Poulsen et al. 1992). The EMG power spectral
MF end
Signal power (arbitrary units)
MF start
100 200 300

Frequency (Hz)
Fig. 9.3 EMG power spectrum at the start and end of a fatiguing contraction, showing the
reduction in median frequency (i.e. the frequency that splits the power spectrum in two parts
containing equal power).
statistics can hence be used to indirectly monitor the development of these underlying metabolic
changes in the muscle, even in the absence of any decline in its given mechanical output.
It has been shown that the decline in EMG MF reflects the muscle’s inability to generate its
maximal force (Mannion and Dolan 1996b) (Figure 9.4). Therefore, it can be stated that the
decline in EMG MF can be used to monitor fatigue, where fatigue is defined as the inability to
generate the maximum force that can be produced by the muscle in its fresh state. The definition
is similar to that originally proposed by Bigland-Ritchie and Woods (1984).
EMG indices other than the MF or MPF are sometimes used in attempts to monitor the devel-
opment of back muscle fatigue. The rise in the average or integrated EMG signal has been
measured (Cooper et al. 1993), as has the power in the low frequency bands (Dolan et al. 1995;
(a)
50% MV C
40% MV C
MV C
30% MV C
(b)
.2
Log rate of decline in force (%/s)
Y = –.056 + .563 * X; R^2 = .716

0
–.2
–.4
–.6
–.8
–1
–1.2
–1.8 –1.6 –1.4 –1.2 –1 –.8 –.6 –.4 –.2 0 .2
Log rate of decline in median frequency (%/s)
Fig. 9.4 Decline in maximum force-generating capacity of the knee extensors at different force
levels (a) and relationship between rate of decline in median frequency of the rectus femoris and
rate of decline in quadriceps isometric maximal force output during sustained contraction (b) (for
further details see Mannion and Dolan 1996b). With kind permission from Springer Science+Business
Media: European Journal of Applied Physiology and Occupational Physiology, Relationship between
myoelectric and mechanical manifestations of fatigue in the quadriceps femoris muscle group,
74(5), 1996, Anne F. Mannion.
Kramer et al. 2005), the spectral power ‘high to low ratio’ (Bradl et al. 2005) and the ‘half-width’
of the EMG power spectrum signal (Oliver et al. 1996; Nargol et al. 1999; Humphrey et al. 2005).
Further, some studies have used nonlinear time series analysis (Sung et al. 2005), wavelet analyses
(Sparto et al. 1999), short-time Fourier transforms, or non-linear recurrence quantification
analysis (Liu et al. 2004) of the EMG signal in order to obviate the difficulties caused by non-
stationary signals, especially when analysing dynamic movements. However, it is not always clear
how such measures (or their changes with fatigue) are to be interpreted on a physiological or
anatomical basis; as such, they are less helpful in establishing the nature of the deficit and the type
of intervention required to remedy it (see later). Of all the EMG parameters, only the slope of the
reduction in MF or MPF of the EMG power spectrum has been linked to known metabolic indi-
ces of fatigue (Bouissou et al. 1989; Brody et al. 1991) and/or the muscle fibre type composition
(Kupa et al. 1995; Mannion et al. 1998).
The establishment of EMG-based indices of back muscle fatigue initially generated great opti-
mism that a measure had been found that was free of the subjectivity of voluntary effort and
would represent an ideal test to assess LBP-related deficiencies in clinical practice. However, ini-
tial enthusiasm was dampened somewhat by later reports that challenged the true objectivity of
these measures. First, it became apparent that the power spectral changes were not entirely inde-
pendent of test contraction time, because the decline in MF was not always linear (Mayer et al.
1989; Mannion and Dolan 1994; Dedering et al. 2000); hence, during an exercise bout of sub-
maximal duration, the choice of test-length became an issue. The initial hope had been that, with
a linear rate of decline in the spectral indices, the test duration would be immaterial, as long as
sufficient data points could be collected to allow accurate enough determination of the slope of
the regression. Ultimately, it was recommended that, in order to circumvent this problem, the
endurance test should be carried out as far as possible to the individual’s endurance limit (Mayer
et al. 1989; Mannion and Dolan 1994; Dedering et al. 2000). In practice, a shortfall of perhaps up
to 20–30 seconds on a 2–3-minute test would, however, be unlikely to lead to grossly inaccurate
values.
Another issue concerned the relative loading chosen for the test, for which still no optimal solu-
tion exists: if the load is set as a percentage maximum voluntary contraction (Lariviere et al.
2003a; Crossman et al. 2004), then it becomes dependent on the generation of a true maximum
in the first place, which can be a challenge in the clinical population (see earlier); if it is expressed
as an absolute force (i.e. the same load for all alike) (Biedermann et al. 1991), then this is disad-
vantageous for the smaller or weaker individuals in the group. If it is set using one’s own body-
weight supported against gravity (e.g. as in the Biering-Sorensen (Biering-Sorenson 1984), Roman
chair test (Mayer et al. 1989)), then it is dependent on the individual’s anthropometric character-
istics and their upper body load (Suuden et al. 2008) as a proportion of their maximum back
strength. There is no satisfactory answer to this problem, although the Biering-Sorensen test
might arguably be considered the ‘best of a bad lot’, since it has the greatest potential for provid-
ing a load-challenge in proportion to the individual’s own body dimensions. Further, even if it
does place some ‘atypically proportioned’ individuals at a disadvantage, those same unfavourable
biomechanics that prevail during performance of the test prevail also during everyday activities
involving repeated or prolonged bending and lifting activities; in this sense, performance in the
Biering-Sorensen test may be considered to better reflect the response of the muscles to the
demands placed upon them in daily life and hence represent a measure of ‘functional fatigability’
(Mannion et al. 1997a).
The aforementioned methodological shortcomings should be considered carefully when
seeking to interpret the results of the various clinical studies on back muscle fatigue. This will be
discussed in more detail later in this chapter.
9.3 Reliability of EMG measurements of back muscle fatigue

If EMG-fatigue indices are to be used to monitor individual performance and its change over
time (e.g. to examine deterioration accompanying the development of a ‘chronic’ back problem,
or improvement following rehabilitation) then it is extremely important that the methods dem-
onstrate adequate test-retest reliability (i.e. low measurement error). A search in PubMed of
studies concerned with EMG measures of back muscle fatigue (search terms ‘EMG and low
back and fatigue’ or ‘EMG and lumbar and muscle fatigue’; search date, 13 August 2008)
revealed many studies on the reliability of the established EMG indices of muscle fatigue (rate of
decline in MF/MPF), in both control and LBP patients. Their main findings are summarized in
Table 9.1. The studies report differing degrees of reliability, which is likely influenced by the test
type, duration of contraction, test equipment, muscles monitored, number of subjects, time
between repeated test sessions, etc. Notably, the data of patients with LBP were just as reliable as
those of controls.
Reliability will always be compromised if good technique following established guidelines (e.g.
Hermens et al. 2000; and www.seniam.org) is not adhered to and if attention is not paid to factors
such as maintaining consistent and accurate electrode alignment between sessions. It is not easy
to condense the findings of the various reliability studies into overarching practical recommenda-
tions; however, in general, it would appear that measures are most reliable when using tests that
demand moderately high force outputs (50–60% MVC), performed for at least 60–90 seconds,
with electrodes positioned more medially (i.e. overlying the multifidi muscles) at lower lumbar
levels (between L3 and L5), using the average of multiple recording sites on both sides of the body.
Whether the MF or MPF data are normalized to the initial value of the MF or MPF (intercept of
the regression of MF/MPF against time) does not appear to make a large difference (Table 9.1).
Following the above recommendations, intraclass correlation coefficients (ICCs) of over 0.8
appear to be achievable, with smallest detectable differences of approx 30% of the absolute mean
value for the given index. The latter figure is particularly important since it influences the sensi-
tivity of the measure to differences between individuals and individual change over time. If
the error of measurement is large, then only interventions with extremely large effect sizes will be
able to be evaluated by the method (i.e. the signal would need to be very large in comparison to
the noise).
As can be seen from Table 9.1, many studies report that the initial median frequency (IMF) is
highly reliable. However, the relevance of this measure per se within the context of muscle fatigue
is uncertain. Certainly it would appear to have more to do with the size, length, relative force
output, and (possibly) relative fibre type sizes of the muscle (Mannion and Dolan 1996a; Mannion
et al. 1998), as well as the degree of subcutaneous fat interposed between the electrodes and the
muscle (Mannion and Dolan 1996a), than with muscle fatigue.
9.4 Physiological/anatomical factors governing fatigability/

muscle endurance
Compared with the number of studies that have examined either the reliability of EMG measures
of back muscle fatigue (see above) or their ability to distinguish between individuals with LBP and
‘normal’ healthy controls (see later), relatively few studies have been carried out to better under-
stand the cause of the changes in the power spectrum and hence the mechanisms that might under-
lie any observed differences between groups, following disuse, or upon training/rehabilitation. If
preventive strategies are to be developed in relation to first-time LBP or its deterioration from the
acute to the chronic condition, then such information must be considered indispensable.
Table 9.1 Summary of results of studies examining the test–retest reliability of EMG-measures of back muscle fatigue using the rate of decline of median or
mean power frequency
Author No. LBP Test condition Measures Retest MF or MPF IMF or IMPF (Hz) Other measures Authors’ comments
year subjects (Y/N) (muscle and EMG interval slope (Hz/s or
index) %/s)
Roy et al. 4 (4M) N Upright standing Longiss L1, 15 min ICC 0.94 ICC 0.98 Technique highly
(1989) extension. 80% iliocost L2 and reliable
MVC for 30s multif L5
——————
IMF, MFslope
(Hz/s)
Biedermann 31 N Upright standing, Multif, iliocost 5 days Test/retest correl IRMS (retest correl) Test/retest reliability is
et al. (1990) weight of 11.6 —————— Multifidus Multifidus better for the
lbs in IMF, IRMS 0.90 (L), 0.95 (R) 0.92 (L), 0.86 (R) multifidus than the
outstretched Iliocostalis Iliocostalis iliocostalis.
arms 45s 0.79 (L), 0.76(R) 0.93 (L), 0.86 (R)
Mannion 5 (2F, N Biering Sorensen Erector spinae, 1–2 wks ICC 0.82–0.98 ICC 0.70–0.77 Greatest MF slope MF slope is a suitable
and Dolan 3M) test to unilateral T10 and CV 14.0–23.3% CV 7.0–9.1% %/s, either region technique for
(1994) exhaustion L3 ICC 0.97, CV 9.7% monitoring back
—————— muscle fatigue.
EMG-DETERMINED MUSCULAR FATIGUE

MF init, MF slope
(%/s)
Ng and 12 (M) N Biering-Sorensen Iliocost, L2; multif, 3 days Iliocost Iliocost Reliable method if use
Richardson test L5 ICC 0.56, CV ICC 0.86, CV 8.5% adequate measures to
(1996) 60s —————— 33.0% Multif minimize cross-talk.
IMF, MFslope Multif ICC 0.79, CV 7.0% Better reliability for the
(Hz/s) ICC 0.78, CV multifidus
27.5%
(continued)
163
164
Table 9.1 (continued) Summary of results of studies examining the test–retest reliability of EMG-measures of back muscle fatigue using the rate of decline of

median or mean power frequenc
index) %/s)
van Dieen 9 (M) N Sitting lumbar Multif, iliocost t2 days ICC 0.84–0.91 RAEMG slope Reproducibility of MPF
and Heijblom extension and longiss. SDD (%) 25–71 ICC 0.31–0.91 superior to RAEMG.
(1996) (Biodex); —————— SDD (%) 61–188 ICCs tended to be
80%MVC to RAEMG (rectified high, but smallest
exhaustion & ave EMG, μV/s) detectable difference
MPF slope (%/s) large, limiting clinical
applicability.
Oliver et al. 10 (M) N Standing pulling Erector spinae, 1 day–4 33% MVC test 33% MVC test Half-width At 33% and 66% of
(1996) up with trunk bilat L4–5 (NB wks ICC 0.44 ICC 0.92 33%, 66% MVC MVC, MF init and
flexed 30°. electrodes widely 66% MVC test SDD (Hz) 8.5 ICC 0.92, 0.94 half-width were
33% and 66% spaced) ICC 0.72 66% MVC test Power slope reproducible; slopes of
MVC for 30s —————— ICC 0.96 33%, 66% MVC the increase in power
IMF SDD (Hz) 6.3 ICC 0.18, 0.43 and decrease in
MF slope (Hz/s) median frequency
Spectral half- were not reproducible
width
Total power slope
Mannion 10 (4M, N 1) Standing Erector spine, t1 day 60% MVC 60% MVC Greatest MF slope Recommend recording
et al. (1997a) 6F) pulling up at bilateral Indiv muscles Indiv muscles %/s from more than one
60%MVC isom T10 ICC 0.73–0. 96 ICC 0.80–0.98 60% MVC test site.
trunk extension, L3 mean ICC T10, BS test ICC 0.94 MF slope, excellent
lumbar spine MF init 0.72 Indiv muscles BS test technique for
60–70% of max MF slope (%/s) mean ICC L3, ICC 0.59–0.94 ICC 0.97 objectively monitoring
ROF 0.95 fatigability of the
2) Biering- BS test erector spinae muscles
Sorensen test. Indiv muscles during sustained
ICC 0.97–0. 99 isometric contractions
Each to mean ICC T10,
exhaustion. 0.99
mean ICC L3,
0.98
Kankaanpaa 10 (8M, N Sitting, submax Paraspinal 1 day Both sides, all Good reproducibility
et al. (1997) 2F) 30 reps/min muscles, L3–4 levels: of the spectral indices
isoinertial trunk and L5/S1 Over 60s (MF)
flex/ext on a —————— ICC 0.36–0.77
David Back Clinic MF slope (%/min) Over 90s (MF)
110 machine MPF slope (%/ ICC 0.58–0.89
min) Over 120s
Recorded over: (MF)
60s, 90s, 120s ICC 0.70–0.94
(MPF similar)
Peach et al. 16 (2 N Semi-standing Erector spinae T9 (1) ICC over all ICC over all sites, REC Apart from IMF,
(1998) grps × 8 30s isometric and L3 levels, every sites, both both groups, R & L ICC over all sites R & spectral parameters
(3M, 5F contraction at multifidus L5 level day for groups, R & L 0.380–0.975 L -.045–0.377 not very reliable within
in each)) 60% MVC —————— 1wk; (2) −0.138–0.656 ICC L3, L5 sites, day a subject.
IMF 1 day/ group L & R 0.889– Experimental condition
MF slope (Hz/s) wk for 0.975 must have sufficiently
REC 4 wks large changes in MF to

constitute a valid
measure (signal>noise)
Elfving et al. 11 (3M, N Sitting, knee and Erector spinae L1 Ave 5 80% MVC test 80% MVC test Borg scale The 95% CIs for the
(1999) 8F) hips flexed, 80% and L5 levels (range ICC 0.04–0.45 ICC 0.41–0.70 ICC 0.84 variables were: IMF ±
MVC isometric —————— 2–13) SEM all sites R & SEM all sites R & L SEM 0.8 10 Hz, slope ± 0.4–
trunk extension, IMF days L 0.20–0.27%/s 4.4–5.5 Hz (CV CV 17% 0.5%/s. Slope may be
David Back 110, MF slope, %/s (CV 35–75%) approx 10%) of limited value
45s because of its large
variability
(continued)
165
166
index) %/s)
Dedering 10 N Biering-Sorensen Erector spinae L1 t 1 day Raw, Hz/s ICC 0.73–0.84 SEM Endurance time ICC Protocol proved to be
et al. (2000) (2M,8F) test to and L5 levels ICC 0.70–0.87 4.6–7.8 Hz 0.89 reliable and is
exhaustion. —————— SEM 0.014– CV%, 7.2–8.9% recommended for
IMF 0.022 further use
MF slope, Hz/s, CV%, 14.4–
%/s 26.2%
Normalized,
%/s
ICC 0.65–0.90
SEM 0.015–
0.025
CV%, 12.0–
23.9%
Koumantakis 16 (7M, N 1) Standing Erector spinae 3 wks 60% MVC all 60% MVC test 60% MVC, Reproducibility of IMF
et al. (2001) 9F) pulling up L2/3, multifidi sites all sites, R & L all sites excellent for both
isometric trunk L4/5 MF Hz/min ICC 0.89–0.96 RMS (%/min) tests.
extension at —————— ICC 0.62–0.91 SDD 12–25% ICC 0.38–0.89 Normalized MF slope
60% MVC IMF SDD 36–69% BS test SDD 86–170% (%/min) more reliable
2) Biering- MF slope Hz/min MF %/min all sites, R & L BS test, all sites with 60% MVC than
Sorensen test MF slope %/min ICC 0.73–0.92 ICC 0.79–0.97 RMS (%/min) BS test.
each for 60s RMS SDD 33–53% SDD 10–27% ICC 0.51–0.74 Both MFinit and
BS test all sites SDD 350–467% MFslope can have
MF Hz/min clinical applicability.
ICC 0.53–0.87 RMS amplitude had a
SDD 46–82% very large between-
MF %/min day error for both
ICC 0.52–0.84 tests
SDD 49–71%
Ebenbichler 14 (all N Standing, knees Paravertebral back 2 hrs MF time Static and dynamic
et al. (2002) M) flexed: static muscles L5, L2, and trends all data lifting MDF time
lifting at 80% T10 2 wks sampled dependent changes,
MVC for 30s; 12 —————— ICC 0.73–0.95 good to excellent
dynamic box-lifts Instantaneous IMDF reliability
(13kg)/min for median frequency (dynamic) all
4.5 min (IMDF) dynamic. data sampled
MF time trends ICC 0.62–0.90
Lariviere et al. 40 (M) Y Standing 75% Bilateral longiss L1 t 2 days MF slope IMF Power in 20–60 Hz Reliable EMG indices
(2002) (20cLBP) MVC fatigue test and T10, iliocost Control Control band (μV2.Hz)/s) obtainable for both
for 30s L3,multif L5 Indiv muscles Indiv muscles, ICC Control controls and cLBP
—————— ICC 0.26–0.77; 0.47–0.85; mean Indiv muscles, ICC patients.
IMF Hz mean (SD) of all (SD) all muscles 0.76 0.54–0.90; mean Better reliability with
MFslope Hz/s, 0.64 (0.14) (0.13) (SD) of all 0.78 (0.13) the medial muscles of
%/s Indiv muscles Indiv muscles Indiv muscles, SEM the back.
RMSslope SEM 22–71%; SEM 6–12%; mean 49–89% Mean (SD) The reliability of EMG
μV/s mean (SD) of (SD) all muscles, 7 of all, 59 (12)% indices might be
REC (recovery of all, 35 (18)% (2)% CLBP increased by averaging
MF) CLBP CLBP Indiv muscles, ICC across recording sites
Indiv muscles Indiv muscles, ICC 0.40–0.68; mean and over repeated
ICC 0.43–0.76; 0.52–0.89; mean (SD) of all 0.52 (0.10) measurements.

mean (SD) of all (SD) all muscles 0.82 Indiv muscles, SEM
0.64 (0.10) (0.12) (%) 74–159; mean
Indiv muscles SEM 6–16%; mean (SD) of all, 114 (26)
SEM 34–87%; (SD) all muscles 8
mean (SD) of (3)%
all, 47(16)%
(values for MF
%/s; MF Hz/s
similar)
(continued)
167
168
index) %/s)
Arnall et al. 10 (6M, N Standing pulling Paraspinal 3 days 50%MVC test 50%MVC test RMS./min Measures at 50%
(2002) 4F) up at 40, 50 & muscles L2/3 and ICC 0.26–0.77 ICC 0.74–0.86 50%MVC test MVC most reliable.
60% MVC isom L4/5 bilaterally. SEM 4. 2–8.3%/ SEM 4.2–12.0 Hz ICC 0.70–0.83 IMF and RMS values
trunk extension, —————— min 60%MVC test SEM 11.3–25.5%/ more reliable than
trunk flexed 30o IMF (Hz) 60%MVC test ICC 0.70–0.89 min MFslope.
to vertical, 60s MFslope ICC 0.24–0.53 SEM 4.1–11.2 Hz 60%MVC test Overall, reliability
(Hz/s and %/s) SEM 3.8–9.1%/ ICC s 0.30–0.87 better at L4/5 than
RMSincline (μV/s min SEM 6.3–15.7%/min L2/3.
and %/min) (values for MF (values for RMS %/s; At L4/5 level all
%/s; MF Hz/s μV/s similar) parameters acceptably
similar) reliable at 50% MVC
No measurable
EMG-fatigue at
40%MVC
Lariviere 40 (M) Y(20M), Standing Longiss T10, L1, t2 days Bilateral ICC 0.68–0.91 Greatest MF slope, Similar reliability,
et al. N (20 M) isometric trunk Illiocost L3, average, SEM (%) d10 any muscle controls and patients.
(2003a) extension fatigue Multif L5. medial ICC 0.74–0.79 RMS slope not eliable.
test 75% MVC —————— muscles SEM (%) 21–26 Averaging MF slope
for 30s IMF ICC 0.68–0.91 over all sites increased
MFslope, Hz/s SEM 5–35% reliability. Bilateral
Average of all medial recording sites
ICC 0.77–0.91 most reliable.
SEM 5–30%
Elfving et al. 20 Y (20) Sitting, knee and Erector spinae L1, 1–6 ICC 0.45–0.71 ICC 0.78–0.89 REC Reliability of IMF good;
(2003) hips flexed, 80% L5 days (L1 mean 0.49, SEM 3.9–6.0 Hz ICC 0–0.36 MF slope less so. ICC’s
MVC isometric —————— L5 mean 0.63) higher for patients
trunk extension, IMF SEM 0.16– than (in previous
David Back 110, MF slope %/s 0.23%/s study) controls; SEM
45s REC (L1 mean 0.17, similar for both. REC
L5 mean 0.21) not reliable enough
for individual
assessment.
ICC, intraclass correlation coefficient; IMF, initial MF at the beginning of the fatigue test; MF, median frequency; MPF, mean power frequency; REC, recovery of median
frequency after fatigue test; RMS, root mean squared amplitude; IRMS, RMS amplitude at the beginning of the fatigue test; SDD, smallest detectable difference; SEM,
standard error of measurement. Longiss, longissimus; iliocost, iliocostalis; multif, multifidus.

169
The fatigue-resistance of skeletal muscles is typically governed by their fibre type area distribu-
tion, capillary density, and enzyme activities and associated metabolism (Saltin and Gollnick
1983). All adult mammalian skeletal muscles contain at least three distinct fibre types, classified on
the basis of their functional and metabolic properties as type I (or ‘slow twitch’), type IIA (‘fast
twitch oxidative’), or type IIX (‘fast twitch glycolytic’) (in previous nomenclature, referred to as the
type IIB fibre; Sant’Ana Pereira et al. 1997). The type I fibre possesses low ATPase activity, a pro-
longed twitch duration (hence, ‘slow twitch’) and a low maximal velocity. In addition, it contains
a higher mitochondrial content and a greater oxidative enzyme complement than the type II fibre.
Type II fibres are characterized by higher ATPase activities and correspondingly shorter isometric
twitch durations and are better endowed with enzymes that support the regeneration of ATP
through anaerobic mechanisms. The type IIX fibre is generally more ‘extreme’ in each of these
respects than the type IIA. During sustained isometric contraction, muscles that possess a pre-
dominance of slow twitch type I fibres contract more ‘economically’ (i.e. they utilize less energy to
maintain the same relative force for the same time, compared with muscles in which the fast twitch
fibre predominates). This reduces the rate of development of inhibitory metabolic end-products,
which have often been implicated in the fatigue process (Bouissou et al. 1989; Laurent et al. 1993).
As such, a greater fatigue-resistance is bestowed upon the ‘slow twitch’ muscle, which allows for a
protracted contraction time (whilst maintaining the same relative force output) in comparison
with ‘faster’ muscle. By virtue of variations in their fibre-type proportions and relative fibre-type
sizes, skeletal muscles have the potential to be specialized for specific types of movement or func-
tion. The back extensors have evolved as primarily postural muscles, with a predominance of large
slow twitch fibres and a superior endurance capacity compared with other skeletal muscles.
As mentioned previously, the rate of decline in the EMG power spectrum MF shows a high
correlation with the force being sustained (Figure 9.5), the task endurance time, and the rate
of reduction of the muscle’s maximum force-generating capacity (Figure 9.4) (Mannion and
Dolan 1996b). This indicates that—even if the common denominator remains elusive—there
85
80
75
Median frequency (Hz)
70
20% MVC
65
30% MVC
60 40% MVC
50% MVC
55
60% MVC
50
45
40
35
–25 0 25 50 75 100 125 150 175 200 225 250
T ime (s )
Fig. 9.5 (See also Colour Plate 6) Rate of decline in median frequency at differing submaximal
contraction force outputs (% MVC) for the quadriceps femoris muscle. See Mannion and Dolan
(1996b) for details. With kind permission from Springer Science+Business Media: European Journal
of Applied Physiology and Occupational Physiology, Relationship between myoelectric and
mechanical manifestations of fatigue in the quadriceps femoris muscle group, 74(5), 1996,
A. F. Mannion.
0.1
y = 0.006x – 0.683 r=0.501 p=0.006
Median frequency slope (%/s) 0.0
–0.1
–0.2
–0.3
–0.4
–0.5
–0.6
–0.7
40 50 60 70 80 90
% area of the muscle occupied by type I fibres
Fig. 9.6 Relationship between proportional area of the muscle occupied by type I fibres and the
slope of the decline in EMG median frequency (for details see Mannion et al. 1998). Reproduced
from Mannion, A.F., Dumas, G.A., Stevenson, J.M. and Cooper, R.G. The influence of muscle fiber
size and type distribution on electromyographic measures of back muscle fatigability. Spine, 23,
pp. 576–84. © 1998, Lippincott, Williams, and Wilkins.
exists a close association between myoelectric and mechanical indices of fatigue for specific tasks.
As the muscle’s fibre type composition influences the rate of build-up of inhibitory metabolites
and these, in turn, seem to be responsible for changes in the muscle’s mechanical function (twitch
tension, rate of rise of twitch tension, twitch relaxation time, etc.) (Sahlin 1992), it is to be expected
that there would also be some relationship between the muscle’s fibre type composition and its
myoelectric manifestations of fatigue. Our own studies conducted on healthy controls in the late
1990s showed that the greater the relative area of the muscle occupied by type I (slow twitch)
fibres, the less rapid was the decline in MF and (generally) the longer was the contraction sus-
tained (Mannion et al. 1998) (Figure 9.6). In two recent studies (Kaser et al. 2001; Crossman et al.
2004) this relationship was confirmed also in LBP patients. If, as suggested, changing metabolic
conditions within the muscle are responsible for eliciting the EMG power spectral shift, then this
would explain the observed association between the decline in MF and the muscle fibre type area
distribution. The correlation appears to be crucially dependent on the relative size of the muscle
fibre types, in addition to their distribution by number.
In one of the above studies, the relationship between the percentage type I area and the EMG
fatigability of the muscle was shown to remain relatively constant before and after 3 months of
exercise therapy (Kaser et al. 2001). The findings of a relatively consistent fatigue level associated
with a given proportional area of the muscle occupied by type I fibres were also supported by the
study of Crossman et al. (2004), in which the values observed for percentage type I fibre area and
the MF decline during fatigue were almost identical in a group of patients with chronic LBP and
in controls.
9.5 The evidence for back muscular fatigue as a predictor of

back pain
At this stage in our examination of the role of EMG-fatigue in LBP, we can conclude that: (1) there
are good theoretical arguments as to why fatigable back muscles might be associated with LBP;
(2) EMG-fatigue can be measured sufficiently reliably; (3) any differences in EMG-determined
back muscle fatigability are likely related to differences in the underlying muscle fibre type size
and composition. The next two sections will examine the evidence for a role of back muscular
fatigability in LBP.
Using the Biering-Sorensen test, three prospective studies have shown that poor back muscle
endurance is associated with an increased risk of developing first-time low back pain (Biering-
Sorenson 1984; Luoto et al. 1995; Adams et al. 1999). However, as mentioned earlier, the ‘time to
fatigue’ during isometric endurance tests is influenced by certain psychological factors, and since
some of these may themselves be independent determinants of the risk of developing LBP (Bigos
et al. 1992; Mannion et al. 1996), this rather complicates things. In one study of risk factors for
first-time LBP, ‘psychological disturbance’ was assessed by questionnaire (a combined score from
the Zung depression questionnaire and the Modified Somatic Perception Questionnaire
(Greenough and Fraser 1991) and the scores were examined in relation to performance on the
Biering-Sorensen test (Mannion et al. 1996 ). For each individual, the time completed on
the endurance test was deducted from the time that would have been predicted on the basis of the
corresponding EMG changes in the back muscles, using the relationship between the greatest
decline in MF at any recording site versus endurance time for the group (Figure 9.7).
The extent to which the endurance time either exceeded or fell short of the predicted time
showed a significant correlation with psychological ‘disturbance’ (Mannion et al. 1996), which
itself was one of the most significant predictors of first time significant LBP (Adams et al. 1999).
Thus, in risk factor studies in which muscle endurance has been examined in isolation, its impor-
tance may have been overestimated and it may have earned its role in predictive models by virtue
of its relationship with some of the important psychological factors. Using a similar ‘performance
Endurance time (log (s))
2.8
Increasing endurance
y = –.665x + 1.761, r2 = .459

2.6
2.4
2.2
2
‘Underperformance’
in endurance time
1.8
1.6
1.4
–1.2 –1 –.8 –.6 –.4 –.2 0
EMG fatigability (log MF%/s decline)
Increasing fatigability
Fig. 9.7 (See also Colour Plate 7) Plot of the endurance time versus EMG-MF slope for individuals
participating in a study of risk factors for LBP (for details, see Mannion et al. 1996). Using the slope
of the relationship between these two measures, it was possible to predict the endurance time
associated with a given EMG fatigability. The subject’s actual performance (e.g. filled circle at
approx. –0.4%/s below) could then be calculated in relation to his predicted endurance time (using
the regression equation) and expressed as the number of seconds of either ‘under-performance’ or
‘over-performance’ compared with his/her peers.
discrepancy score’ for the Biering-Sorensen test, a group of 148 patients with chronic LBP were cat-
egorized as either underperformers (or ‘discomfort-intolerant’) or over-performers (‘discomfort-
tolerant’) compared with their contemporaries (Mannion et al. submitted). In bivariate analyses,
underperformers had significantly higher scores for Roland–Morris disability (Roland and Morris
1983 ), Back Beliefs (Symonds et al. 1996 ), psychological disturbance (Zung and MSPQ)
(Greenough and Fraser 1991), catastrophizing (Rosenstiel and Keefe,1983), and exercise self-effi-
cacy (Schwarzer 1993); their scores for Fear Avoidance Beliefs were of borderline significance
(p=0.055) Underperformers were also older, and more likely to be female. Interestingly, body
mass index (BMI), maximum rate of decline in MF per second (‘intrinsic’ fatigability), average
and worst back pain intensity, duration of LBP and frequency of back pain did not differ signifi-
cantly between the groups. In multiple stepwise linear regression analysis to predict the ‘expected
minus actual endurance time’, female gender, psychological disturbance and ‘negative’ back
beliefs were significant predictors of ‘underperformance’ in the model (adj R-squared, 22.3%; p
<0.001). Once again, this highlights the fact that performance in the Biering-Sorensen, as meas-
ured by endurance time alone, reflects much more than just the intrinsic physiological fatigability
of the trunk musculature and this must be borne in mind when interpreting the results of studies
based on this measure. A similar conclusion was reached in a recent review of the test for spinal
muscle evaluation (Demoulin et al. 2006).
No studies to date have shown that the ‘intrinsic’ physiological fatigability of the back muscu-
lature (measured using EMG-techniques) influences the likelihood of developing LBP. In the
aforementioned study of predictors of first-time LBP (Adams et al. 1999), endurance time was
significantly (though weakly) associated with the development of LBP in bivariate analyses (albeit
failing to make a significant contribution in a multivariate model including the psychological
variables), but the objective EMG-indices of fatigability failed to reach significance at any of the
many follow-ups, in either bivariate or multivariate analyses. Thus, despite the convincing theo-
retical arguments, there is actually little evidence to support the notion that back muscle fatigabil-
ity (as distinct from endurance capacity) is a significant risk factor for the development of low
back pain.
9.6 The evidence for changes in back muscular fatigue as

a consequence of back pain
From the foregoing discussion, it would appear that any associations between muscle fatigability
and LBP are likely the consequence rather than the cause of the development of first-time LBP.
Conceivably, they might be ‘causal’ in relation to the recurrence of LBP, via the vicious circle of
pain, inactivity and altered muscular function described in Figure 9.1, but this has not been well
investigated. In examining any such relationships, the use of objective EMG measures is abso-
lutely essential. As discussed earlier, if only the endurance time for the contraction is measured,
then poor achievers may be demonstrating nothing more than an unwillingness to perform, or a
particularly low pain/discomfort tolerance.
In the PubMed search of the literature referred to earlier (see earlier section ‘Reliability of EMG
measurements of back muscle fatigue’), more than 20 studies were identified in which the rate of
decline in MF (or MPF) was compared in groups of patients with LBP and healthy controls. Brief
methodological details and the main findings of these studies are shown in Table 9.2, ordered by
the date of publication. Interestingly, the first seven studies, carried out between 1989 and 1994
(many of them small studies, on males only, and from the same research group), all showed that
groups of individuals with LBP were significantly more fatigable than controls, and that individu-
als could be classified accurately into their groups based on their EMG-spectral characteristics
Table 9.2 Summary of results of studies examining the difference between groups of controls and
groups of patients with LBP using the rate of decline of the EMG median (or mean) power
frequency.
Author (year) Groups Test procedure Findings for

differences in
MF slope (EMG
fatigability)
Roy et al. 12 LBP, 12 control Back Analysis System (standing isometric LBP > control
(1989) (all male) extension, 40%, 60%, 80% MVC for max 60s; (at L2, L5 80%
15min between efforts); bilateral EMG at L1 MVC only)
longiss, L2 iliocost, L5 multifid
Mayer et al. 10 LBP (8M, 2F) Upper torso unsupported (Roman Chair) for LBP > control
(1989) 11 control (8M, 3F) successive fixed-time trials; EMG, erector
spinae L3
Roy et al. 6 LBP, Back Analysis System (standing isometric LBP > control
(1990) 17 control (all male) extension, 80% MVC for max 30s); bilateral
EMG at L1 longiss, L2 iliocost, L5 multifid
Klein et al. 8 LBP, Back Analysis System (standing isometric LBP > control
(1991) 17 control (all male) extension, 80% MVC for max 30s); bilateral
EMG at L1 longiss, L2 iliocost, L5 multifid
Biedermann 25 LBP (13M, 11F) Standing, in reference frame. Arms outstretched, LBP > control
et al. (1991) 22 cont (10M, 12F) holding a free weight (2×5lb). EMG MF, L2–3, avoiders >
L4–5. Patients classified as ‘avoiders’ or confronters
‘confronters’ using Pain Behaviour Checklist
Cooper et al. 11 LBP (4M, 7F) 28 Biering-Sorensen test; unilat erector spinae at L4; LBP > control
(1993) cont (15M,13F) iEMG rate of change
Tsuboi et al. 9 LBP, Biering-Sorensen test, trunk with lifted further LBP > control
(1994) 10 cont (all male) 5 degrees, 120s; same but with loading of 30% (but no stat
MVC; EMG L1, L2, L4, L5 analyses for
the difference)
Kankaanpaa 20 LBP (F), 15 cont Seated, trunk flexed 30 deg, 50% MVC LBP = control
et al. (1998) (F) isometric back extension (David Back 100) to
fatigue; EMG erector spinae (L3/4 and L5/S1)
Peach and LBP 21 (13M, 8F) Isometric trunk extension with harness around LBP < control
McGill (1998) Cont 18 (3M, 10F). the torso.
NB age 48 (14) and 30 second fatigue test at 60% MVC;
22 (2) respectively EMG T9, L3 (erector spinae), L5 (multifidus)
Suter and 25Y, 16N (all male) BS test; EMG bilat erector spinae at T12 and LBP = control
Lindsay (2001) L4–L5.
Ng et al. LBP 12, Cont 12 (all Fatigue test in standing, 80% MVC, right and LBP = control
(2002) male) left axial rotation.
EMG bilat, illiocostalis lumborum L2, multifidus
L5 (plus rectus abdominus, ext & int oblique,
latissimus dorsi)
Elfving et al. 57 LBP (27M, 30F) EMG bilat erector spinae, L1 and L5 Seated 80% LBP < control
(2003) 55 controls (28M, max isometric extension torque for 45s (in females
27F) only; in males
no difference)
(continued)
Table 9.2 (continued) Summary of results of studies examining the difference between groups of
controls and groups of patients with LBP using the rate of decline of the EMG median (or mean)
power frequency.
Author (year) Groups Test procedure Findings for

differences in
MF slope (EMG
fatigability)
Lariviere et al. LBP 20 controls 20 Standing, static trunk extension 75% MVC LBP < control
(2003a) (all male) for 30s; (for all sites)
EMG bilaterally longissimus T10 and L1,
illiocostalis L3, and multifidus L5; (also rectus
abdominus, external oblique, internal oblique)
Crossman LBP 35, controls 32 Biering-Sorensen and standing pulling up 60% LBP=control
et al. (2004) (all male) MVC fatigue tests; EMG bilat erector spinae, (both tests)
L4/5
Kankaanpaa LBP 17, controls 12 Seated, trunk flexed 30°, submax (wt & LBP=control
et al. (2005) (all male) ht-based) dynamic back extension (David Back
100); 30 reps/min for max 90s; EMG bilat
erector spinae (L4/5)
da Silva et al. LBP 13, controls 15 Standing upright, standing trunk flexed (each LBP=control
(2005) (all male) 50% MVC), Biering-Sorensen test; EMG bilat (all tests)
longiss (T10 and L1), iliocostalis (L3),
multifidus (L5)
Kramer et al. LBP 31, control 31 Seated, trunk flexed 30°, 60% MVC isometric LBP < control
(2005) back extension (David Back 100) for 60s; EMG (signif. for
bilateral multifidus (L4/5 and T11/12) and erector multifidus,
spinae (L4/5/&T11/12) L4/5 and
T11/12)
Humphre y LBP 145 (chronic), Standing, trunk flexed, pulling up at 66% MVC LBP=control
et al. (2005) 30 history, for 30; EMG bilat paraspinal muscles L4/5
control 175
Candotti et al. LBP 30, control 30 Lying prone, 80% max extensor force for 35s; LBP > controls
(2008) (gender not given) EMG bilateral longiss (L1) iliocostalis (L2) (for left side
muscles only)
Suuden et al. LBP 20 (10M, 10F) Biering-Sorensen test to fatigue; EMG bilat LBP=controls
(2008) control 20 (10M, erector spinae, L3
10F)
Sung et al. LBP 40 control 40 Biering-Sorensen test for max 60s; EMG bilat Methods and
(2009) (gender distrib not erector spinae (approx T12, L5) results not
given) clear
during a fatiguing task. These findings in relation to fatigability were compatible with altered
muscle fibre type characteristics in LBP patients, in the direction of an increase in the propor-
tional area of the muscle occupied by fast twitch type II fibres. Such an increase could occur due
to either reflex inhibition of the musculature or pain-induced inactivity and disuse following the
initial acute episode of LBP. In other musculoskeletal systems, it is known that injury to a joint is
associated with a reduction in alpha-motoneuron activity to the muscles surrounding that joint
(Morrissey 1989). It has also been shown that cessation of normal repetitive low-level activity
patterns and stretch, which otherwise serve to maintain the slow twitch fibre population, results
in a transformation of the muscle towards a faster, more fatigable type (Goldspink et al. 1994). It
was hence quite feasible that the observed fatigability of the back muscles of patients reflected
muscular adaptations, consequent to the onset of back pain.
These findings, in turn, stimulated various interventional studies attempting to ‘re-condition’
the deconditioned muscles (see later). However, in the years that followed the initial cross-
sectional studies, few research groups were able to reproduce the findings. A series of studies
carried out throughout the late 1990s up to the present day—using varying types of test position,
test contractions, and electrode recording sites, and some with very large group sizes (making
type II errors unlikely)—reported either no difference between LBP patients and controls, or
even the reverse of the earlier findings, namely LBP patients being less fatigable than controls
(Table 9.2). In all studies reporting the latter, the submaximal load had been set as a %MVC, and
it was questioned whether a true MVC had been obtained in the first place; if not, then the load-
challenge would have been less in the patients, and the fatigue rate naturally lower for these
reasons alone. It is not easy to verify the maximality of voluntary efforts in back strength tests in
patients: twitch interpolation techniques (Mannion et al. 1997b) are not entirely practical and
predictive models based on anthropometric dimensions are not accurate enough, generally
explaining only 30–40% variance in strength (Mannion et al. 1999; Lariviere et al. 2003b; Wang
et al. 2005). Thus, the uncertainties surrounding an individual’s true maximum strength continue
to represent a confounder in any assessments of muscle fatigue that use submaximal contractions
set as a %MVC.
As can be seen in Table 9.2, in all the studies in which the Biering-Sorensen test was used to
assess fatigability, there was no statistically significant difference between the EMG-MF slopes of
patients and controls. Although this test is not without its own limitations, it probably represents
the best available option for making non-MVC-dependent measures of fatigability (see earlier).
Further, one might expect that the ‘trunk fat-mass to lean-mass ratio’ of the theoretically ‘inac-
tive, overweight, and weaker’ patient might, if anything, put them at a disadvantage during
performance of this test. The finding that their back muscles do not fatigue more rapidly than
controls hence indicates that they are definitely not more fatigable, and may even be less fatigable,
than controls.
Possible explanations for the discrepancies of these findings with those of the earlier studies
include the fact that the latter used generally quite athletic LBP-sufferers (that may have been
atypical compared with clinical populations); that generally only men were included; that too
many EMG-variables were used for the classification procedure, such that the model was over-
fitted statistically; and the fact that many of the variables included in the classification models
were actually not related to fatigue per se (e.g. the initial MF; see earlier). Either way, it seems that
the only clear conclusion emerging from the last 20 years’ research on the association between
LBP and EMG-fatigability is that the relationship is highly inconsistent.
9.7 Changes in fatigability through exercise training and their

clinical relevance
In acknowledgement of the early findings on back muscle fatigability in LBP patients, some
studies sought to examine the response of the muscles to various types of training or rehabilita-
tion (Thompson et al. 1992; Moffroid et al. 1993; Roy et al. 1995; Wood et al. 1997; Kankaanpaa
et al. 1999; Mannion et al. 2001b; Sung 2003). In some studies a post-training improvement in
endurance time was recorded without any corresponding change in the EMG-determined muscle
fatigability (Moffroid et al. 1993; Wood et al. 1997; Mannion et al. 2001b) or muscle fibre type
distributional area (Mannion et al. 2001b). This perhaps indicates a determination on the behalf
of the patient to do better after therapy, despite the lack of any real accompanying physiological
change in their muscles enabling them to do so. Some of the research groups that had originally
reported overly-fatigable muscles in LBP patients also reported improvements following exercise
training (Mayer et al. 1989; Thompson et al. 1992; Roy et al. 1995,). Decreases in the steepness of
the MF-slope lasting up to 6 months after therapy, were similarly reported by Kankaanpaa et al.
(1999), although their cross-sectional studies had not shown excessive fatigability in connection
with LBP at baseline, and values for the MF-slope were unrelated to patients’ self-ratings of pain
or disability (Kankaanpaa et al. 1998).
As pointed out by Elfving et al. (2003), in all these studies the submaximal load for the fatigue
test remained the same pre- and post-therapy; any improvements in strength would therefore
have rendered the given force less demanding post-therapy, perhaps explaining the reduced fati-
gability. Later studies failed to reproduce the findings of these earlier studies, sometimes even
finding a post-therapy increase in EMG-determined fatigability in some test situations, commen-
surate with an increased ability to activate the muscles and drive them to fatigue (Mannion et al.
2001b; Sung 2003,). The finding that an increased MF-slope might indicate a ‘back-healthier’
condition was also suggested by the study of Elfving et al. (2003) in which the ability to achieve a
decline in the MF during a sustained contraction was significantly associated with lower scores for
disability in everyday activities.
Few intervention studies have addressed the ultimate question, critical in relation to the clinical
relevance of back muscle fatigability, as to whether improvements observed after therapy are
actually associated with a corresponding reduction in the patient’s self-rated pain and disability.
Ultimately, it is these factors, which influence the quality of life, that drive the patient back into
the clinic and give rise to the enormous societal costs associated with the management of LBP.
There is good reason to suppose that disability might be decreased following active rehabilitation,
but the question of whether an improved muscular capacity should translate to a reduction in
pain is perhaps more complicated. If the muscular deficits are intricately linked with the source of
pain—for example, by their failure to stabilize an injured or unstable joint—then an improve-
ment in muscular support accompanied by a gradual relief from the pain may indeed be antici-
pated. If, however, the decline in muscle function is simply the result of general disuse, then
improvements would not necessarily be expected to relieve the existing pain—although they may
certainly offer a protective effect against the chances of ‘reinjury’ and repeated episodes of low
back pain in the future.
One of the biggest difficulties in interpreting the results of rehabilitation studies concerns iden-
tification of the ‘active element’ of the programme. Typical active therapy programmes include
various elements that aim to ‘recondition’ the patient, some of which are carefully introduced
into the programme and others of which inadvertently accompany it. Specially designed exercises
may aim to restore a particular aspect of physical function, and these may indeed be effective;
however, regular attendance at therapy may coincidentally elicit feelings of well-being and accom-
plishment in achieving an improved physical fitness (Nutter, 1988) as well as antidepressive
effects in connection with the release of natural endorphins (Ransford 1982; Dey 1994). Each of
these elements could independently modify the perception and reporting of pain, and possibly
even the degree of disability that patients allow it to create. In this sense, the recording of group
changes in both function and clinical status does not necessarily confirm that the clinical changes
followed the functional changes; instead, it is necessary to examine comparison groups in which
clinical but no functional changes occurred or to examine the concordance of findings on an
individual basis (i.e. whether improvements in fatigability and clinical status occur in the same
subjects, and whether the magnitude of change in each is related). Even if a correlation between
the changes in two variables can be established, this still doesn’t necessarily prove the existence of
a causal relationship; the converse, however—a reduction in symptoms or disability in the absence
of any significant change in muscle function, or vice versa—would certainly imply that the two
were unrelated. Though difficult, identifying the active ingredient in muscle reconditioning pro-
grammes is of paramount importance in concluding the case for the clinical significance of back
muscle fatigability in LBP.
Only one study has directly addressed this issue, by examining changes in back muscle EMG-
fatigability after different active therapies for chronic LBP in relation to corresponding changes in
pain and disability (Mannion et al. 2001a). The findings were not supportive of any clinical rele-
vance for back muscle fatigability. Firstly, there was no relationship between individual measures
of EMG-fatigability (isometric or dynamic) and self-rated disability at baseline, and no relation-
ship between the changes in back muscle fatigability and changes in self-rated disability after
therapy (Mannion et al. 2001a). Pain and disability were reduced to a similar extent in each of the
three active therapy groups, despite their differing nature (mainly strengthening/reconditioning
vs. aerobic vs. physiotherapeutic). There were no significant changes in the MF slope during the
BS test, and though the MF slope during a dynamic fatigue test changed after therapy (increased),
it did so in a manner unrelated to the changes in pain or disability. No significant changes in the
size or structure of the paraspinal muscles were observed (Kaser et al. 2001). Overall, the results
indicated that neither the intrinsic fatigability of the back muscles, nor the structural characteris-
tics that determined it, bore any meaningful relationship to measures of LBP disability. The
clinical significance of these measures hence remains obscure. Future studies should investigate
whether these aspects of muscle function (and their changes with training) are associated with
future episodes of LBP, either in their frequency, duration or intensity.
In summary, there are many convincing hypotheses linking fatigable back muscles with LBP.
Fatigability of the back muscles, as measured by the rate of decline in the mean or median fre-
quency of the EMG power spectrum, can be measured with acceptable reliability if the basic
principles of EMG are adhered to, and optimized sampling methods are used. The rate of decline
in MF is correlated with the rate of decline in maximum force-generating capacity and is (at least
in part) dependent on the underlying muscle fibre characteristics. The findings on differences in
EMG-determined fatigability in patients with LBP and healthy controls are conflicting, as are
those concerned with changes in EMG-fatigability after rehabilitation or exercise therapy. Back
muscle fatigability does not relate to subjective ratings of pain or disability in LBP patients, and
does not change in parallel to changes in these clinical variables after therapy. Hence, the clinical
significance of EMG-determined back muscle fatigability remains obscure. Future studies should
focus on its potential role in relation to predicting LBP recurrence/chronification.
References
Adams, M.A. and Dolan, P. (2005). Spine biomechanics. J Biomechanics, 38, 1972–83.
Adams, M.A. and Hutton, W.C. (1986). Has the lumbar spine a margin of safety in forward bending?
Clinical Biomechanics, 1, 3–6.
Adams, M.A., Mannion, A.F. and Dolan, P. (1999). Personal risk factors for first-time low back pain. Spine,
24, 2497–505.
Adams, S.A., Matthews, C.E., Ebbeling, C.B., Moore, C.G., Cunningham, J.E., Fulton, J. and Hebert, J.R.
(2005). The effect of social desirability and social approval on self-reports of physical activity. American
Journal of Epidemiology, 161, 389–98.
Allison, G.T. and Henry, S.M. (2002). The influence of fatigue on trunk muscle responses to sudden arm
movements, a pilot study. Clinical Biomechanics, 17, 414–17.
Arnall, F.A., Koumantakis, G.A., Oldham, J.A. and Cooper, R.G. (2002). Between-days reliability of
electromyographic measures of paraspinal muscle fatigue at 40, 50 and 60% levels of maximal
voluntary contractile force. Clinical Rehabilitation, 16, 761–71.
Biedermann, H.J., Shanks, G.L. and Inglis, J. (1990). Median frequency estimates of paraspinal muscles:
reliability analysis. Electromyography Clinical Neurophysiology, 30, 83–8.
Biedermann, H.J., Shanks, G. L., Forrest, W.J. and Inglis, J. (1991). Power spectrum analyses of
electromyographic activity discriminators in the differential assessment of patients with chronic low
back pain. Spine, 16, 1179–84.
Biering-Sorenson, F. (1984). Physical measurements as risk indicators for low-back trouble over a one-year
period. Spine, 9, 106–19.
Bigland-Ritchie, B. and Woods, J.J. (1984). Changes in muscle contractile properties and neural control
during human muscular fatigue. Muscle Nerve, 7, 691–9.
Bigos, S.J., Battie, M.C., Spengler, D.M., et al. (1992). A longitudinal, prospective study of industrial back
injury reporting. Clinical Orthopaedics and Related Research, 279, 21–33.
Bogduk, N. and Twomey, L.T. (1991). Clinical anatomy of the lumbar spine and sacrum (2nd edn.). London:
Bonato, P., Ebenbichler, G.R., Roy, S.H., et al. (2003). Muscle fatigue and fatigue-related biomechanical
changes during a cyclic lifting task. Spine, 28, 1810–20.
Bouissou, P., Estrade, P.V., Goubel, F., Guezennec, C.Y. and Serrurier, B. (1989). Surface EMG power
spectrum and intramuscular pH in human vastus lateralis muscle during dynamic exercise. Journal of
Applied Physiology, 67, 1245–9.
Bradl, I., Morl, F., Scholle, H.C., Grassme, R., Muller, R. and Grieshaber, R. (2005). Back muscle activation
pattern and spectrum in defined load situations. Pathophysiology, 12, 275–80.
Brody, L.R., Pollock, M.T., Roy, S.H., DeLuca, C.J. and Celli, B. (1991). pH induced effects on median
frequency and conduction velocity of the myoelectric signal. Journal of Applied Physiology, 71, 1878–85.
Candotti, C.T., Loss, J.F., Pressi, A.M., et al. (2008). Electromyography for assessment of pain in low back
muscles. Physical Therapy, 88, 1061–7.
Ciba Foundation (1981). Human muscle fatigue: physiological mechanisms. London: Pitman Medical.
Cooper, R.G., Stokes, M.J., Sweet, C., Taylor, R.J. and Jayson, M.I.V. (1993). Increased central drive
during fatiguing contractions of the paraspinal muscles in patients with chronic low back pain. Spine,
18, 610–16.
Crossman, K., Mahon, M., Watson, P.J., Oldham, J.A. and Cooper, R.G. (2004). Chronic low back pain-
associated paraspinal muscle dysfunction is not the result of a constitutionally determined ‘adverse’
fiber-type composition. Spine, 29, 628–34.
da Silva, R.A., Jr., Arsenault, A.B., Gravel, D., Lariviere, C. and de Oliveira, E., Jr. (2005). Back muscle
strength and fatigue in healthy and chronic low back pain subjects: a comparative study of 3 assessment
protocols. Archives of Physical Medicine Rehabilitation, 86, 722–9.
De Luca, C. (1997). The use of surface electromyography in biomechanics. Journal of Applied Biomechanics,
13, 135–63.
Dedering, A., Roos af Hjelmsater, M., Elfving, B., Harms-Ringdahl, K. and Nemeth, G. (2000). Between-
days reliability of subjective and objective assessments of back extensor muscle fatigue in subjects
without lower-back pain. Journal of Electromyography and Kinesiology 10, 151–8.
Demoulin, C., Vanderthommen, M., Duysens, C. and Crielaard, J.M. (2006). Spinal muscle evaluation
using the Sorensen test: a critical appraisal of the literature. Joint Bone Spine, 73, 43–50.
Dey, S. (1994). Physical exercise as a novel antidepressant agent: possible role of serotonin receptor
subtypes. Physiology and Behaviour, 55, 323–29.
Dolan, P. and Adams, M.A. (1998). Repetitive lifting tasks fatigue the back muscles and increase the
bending moment acting on the lumbar spine. Journal of Biomechanics, 31, 713–21.
Dolan, P., Mannion, A.F. and Adams, M. (1995). Fatigue of the erector spinae muscles: A quantitative
assessment using ‘frequency banding’ of the surface electromyography signal. Spine, 20, 149–59.
Duchateau, J. and Hainaut, K. (1984). Training effects on muscle fatigue in man. European Journal of
Applied Physiology and Occupational Physiology, 53, 248–52.
Ebenbichler, G.R., Bonato, P., Roy, S.H., et al. (2002). Reliability of EMG time-frequency measures of
fatigue during repetitive lifting. Medicine and Science in Sports and Exercise, 34, 1316–23.
Elfving, B., Nemeth, G., Arvidsson, I. and Lamontagne, M. (1999). Reliability of EMG spectral parameters
in repeated measurements of back muscle fatigue. Journal of Electromyography and Kinesiology, 9,
235–43.
Elfving, B., Dedering, A. and Nemeth, G. (2003). Lumbar muscle fatigue and recovery in patients with
long-term low-back trouble—electromyography and health-related factors. Clinical Biomechanics, 18,
619–30.
Goldspink, G., Gerlach, G. F., Jaenicke, T. and Butterworth, P. (1994). In Lyall, F., El Haj, A.J. (Eds.)
Biomechanics and cells, pp. 81–95. Cambridge: Cambridge University Press.
Granata, K. P., Slota, G. P. and Wilson, S. E. (2004). Influence of fatigue in neuromuscular control of
spinal stability. Human Factors, 46, 81–91.
Greenough, C.G. and Fraser, R.D. (1991). Comparison of eight psychometric instruments in unselected
patients with back pain. Spine, 16, 1068–74.
Hagberg, M. (1981). Muscular endurance and surface electromyogram in isometric and dynamic exercise.
Journal of Applied Physiology, 51, 1–7.
Hansson, T.H., Bigos, S.J., Wortley, M.K. and Spengler, D.M. (1984). The load on the lumbar spine during
isometric strength testing. Spine, 9, 720–4.
Hermens, H.J., Freriks, B., Disselhorst-Klug, C. and Rau, G. (2000). Development of recommendations for
SEMG sensors and sensor placement procedures. Journal of Electromyography and Kinesiology 10, 361–74.
Hultman, G., Nordin, M., Saraste, H. and Ohlsen, H. (1993). Body composition, endurance, strength,
cross-sectional area, and density of MM erector spinae in men with and without low back pain. Journal
of Spinal Disorders, 6, 114–23.
Humphrey, A., Nargol, A.V., Jones, A.P., Ratcliffe, A.A. and Greenough, C. (2005). The value of
electromyography of the lumbar paraspinal muscles in discriminating between chronic-low-back-pain
sufferers and normal subjects. European Spine Journal, 24, 175–84.
Jorgensen, K. and Nicolaisen, T. (1987). Trunk extensor endurance: determination and relation to low
back trouble. Ergonomics, 30, 259–67.
Jurell, K.C. (1998). Surface EMG and fatigue. Physical Medicine and Rehabilitation Clinics of North America,
9, 933–46.
Kankaanpaa, M., Taimela, S., Webber, C.L., Airaksinen, O. and Hanninen, O. (1997). Lumbar paraspinal
muscle fatigability in repetitive isoinertial loading: EMG spectral indices, Borg scale and endurance
time. European Journal of Applied Physiology. 76, 236–42.
Kankaanpaa, M., Taimela, S., Laaksonen, D., Hanninen, O. and Airaksinen, O. (1998). Back and hip
extensor fatigability in chronic low back pain patients and controls. Archives of Physical Medicine and
Rehabilitation 79, 412–17.
Kankaanpaa, M., Taimela, S., Airaksinen, O. and Hanninen, O. (1999). The efficacy of active rehabilitation
in chronic low back pain. Effect on pain intensity, self-experienced disability, and lumbar fatigability.
Spine, 24, 1034–42.
Kaser, L., Mannion, A.F., Rhyner, A., Weber, E., Dvorak, J. and Muntener, M. (2001). Active therapy for
chronic low back pain: part 2. Effects on paraspinal muscle cross-sectional area, fiber type size, and
distribution. Spine, 26, 909–19.
Kankaanpaa, M., Colier, W.N., Taimela, S., et al. (2005). Back extensor muscle oxygenation and fatigability
in healthy subjects and low back pain patients during dynamic back extension exertion.
Pathophysiology 12, 267–73
Klein, A. B., Snyder Mackler, L., Roy, S.H. and DeLuca, C.J. (1991). Comparison of spinal mobility and
isometric trunk extensor forces with electromyographic spectral analysis in identifying low back pain.
Physical Therapy, 71, 445–54.
Koumantakis, G.A., Arnall, F., Cooper, R.G. and Oldham, J.A. (2001). Paraspinal muscle EMG fatigue
testing with two methods in healthy volunteers. Reliability in the context of clinical applications.
Clinical Biomechanics 16, 263–6.
Kramer, M., Ebert, V., Kinzl, L., Dehner, C., Elbel, M. and Hartwig, E. (2005). Surface electromyography
of the paravertebral muscles in patients with chronic low back pain. Archives Physical Medicine
Kujala, U.M., Taimela, S., Viljanen, T., et al. (1996). Physical loading and performance as predictors of
back pain in healthy adults. A 5-year prospective study. European Journal Applied Physiology
Occupational Physiology, 73, 452–8.
Kupa, E.J., Roy, S.H., Kandarian, S.C. and De Luca, C.J. (1995). Effects of muscle fiber morphology on the
EMG signal. Annals of Biomedical Engineering, 23, S112.
Lariviere, C., Arsenault, A.B., Gravel, D., Gagnon, D. and Loisel, P. (2002). Evaluation of measurement
strategies to increase the reliability of EMG indices to assess back muscle fatigue and recovery. Journal
of Electromyography and Kinesiology 12, 91–102.
Lariviere, C., Arsenault, A.B., Gravel, D., Gagnon, D. and Loisel, P. (2003a). Surface electromyography
assessment of back muscle intrinsic properties. Journal of Electromyography and Kinesiology 13, 305–18.
Lariviere, C., Gravel, D., Gagnon, D., Arsenault, A.B., Loisel, P. and Lepage, Y. (2003b). Back strength
cannot be predicted accurately from anthropometric measures in subjects with and without chronic
low back pain. Clinical Biomechanics, 18, 473–9.
Laurent, D., Portero, P., Goubel, F. and Rossi, A. (1993). Electromyogram spectrum changes during
sustained contraction related to proton and diprotonated inorganic phosphate accumulation: A 31P
nuclear magnetic resonance study on human calf muscles. European Journal Applied Physiology
Occupational Physiology, 66, 263–68.
Liu, Y., Kankaanpaa, M., Zbilut, J.P. and Webber, C.L., Jr. (2004). EMG recurrence quantifications in
dynamic exercise. Biological Cybernetics, 90, 337–48.
Luoto, S., Heliovaara, M., Hurri, H. and Alaranta, H. (1995). Static back endurance and the risk of low-
back pain. Clinical Biomechanics, 10, 323–24.
Magnusson, M.L., Aleksiev, A., Wilder, D.G., et al. (1996). Unexpected load and asymmetric posture as
etiologic factors in low back pain. European Spine Journal, 5, 23–35.
Mannion, A.F. and Dolan, P. (1994). Electromyographic median frequency changes during isometric
contraction of the back extensors to fatigue. Spine, 19, 1223–9.
Mannion, A.F. and Dolan, P. (1996a). The effects of muscle length and force output on the EMG power
spectrum of the erector spinae. Journal of Electromyography and Kinesiology 6, 159–68.
Mannion, A.F. and Dolan, P. (1996b). Relationship between myoelectric and mechanical manifestations of
fatigue in the quadriceps femoris muscle group. European Journal Applied Physiology Occupational
Physiology, 74, 411–19.
Mannion, A.F., Dolan, P. and Adams, M.A. (1996). Psychological questionnaires: do ‘abnormal’ scores
precede or follow first-time low back pain? Spine, 21, 2603–11.
Mannion, A.F., Connolly, B., Wood, K. and Dolan, P. (1997a). The use of surface EMG power spectral
analysis in the evaluation of back muscle function. Journal of Rehabilitation Research and Development
34, 427–39.
Mannion, A.F., Dolan, P., Adam, G.G., Adams, M.A. and Cooper, R.G. (1997b). Can maximal back muscle
strength be predicted from submaximal efforts? Journal of Back and Musculoskeletal Medicine, 9, 49–51.
Mannion, A.F., Weber, B. R., Dvorak, J., Grob, D. and Muntener, M. (1997c). Fibre type characteristics of
the lumbar paraspinal muscles in normal healthy subjects and in patients with low back pain. Journal of
Orthopaedic Research, 15, 881–7.
Mannion, A.F., Dolan, P. and Adams, M.A. (1996). Psychological questionnaires: do ‘abnormal’ scores
precede or follow first time low back pain? Spine, 21, 2603–2611.
Mannion, A.F., Dumas, G.A., Stevenson, J.M. and Cooper, R.G. (1998). The influence of muscle fiber size
and type distribution on electromyographic measures of back muscle fatigability. Spine, 23, 576–84.
Mannion, A.F., Adams, M.A., Cooper, R.G. and Dolan, P. (1999). Prediction of maximal back muscle
strength from indices of body mass and fat-free body mass. Rheumatology, 38, 652–5.
Mannion, A.F., Junge, A., Taimela, S., Muntener, M., Lorenzo, K. and Dvorak, J. (2001a). Active therapy
for chronic low back pain: part 3. Factors influencing self-rated disability and its change following
therapy. Spine, 26, 920–9.
Mannion, A.F., Taimela, S., Muntener, M. and Dvorak, J. (2001b). Active therapy for chronic low back
pain: part 1. Effects on back muscle activation, fatigability, and strength. Spine, 26, 897–908.
Mannion, A.F., O’Riordan, D., Dvorak, J., Masharawi, Y. The dependence of back muscle endurance time
on psychological factors. (Submitted)
Mayer, T.G., Gatchel, R. J., Kishino, N., et al. (1985). Lumbar myoelectric spectral analysis for endurance
assessment. A comparison of normals with deconditioned patients. Spine, 10, 482–93.
Mayer, T.G., Kondraske, G., Mooney, V., Carmichael, T.W. and Butsch, R. (1989). Lumbar myoelectric
spectral analysis for endurance assessment. A comparison of normals with deconditioned patients.
Spine, 14, 986–91.
Moffroid, M.T., Haugh, L.D., Haig, A.J., Henry, S.M. and Pope, M.H. (1993). Endurance training of trunk
extensor muscles. Physical Therapy, 73, 3–10.
Mooney, V., Gulick, J., Perlman, M., et al. (1997). Relationships between myoelectric activity, strength, and
MRI of lumbar extensor muscles in back pain patients and normal subjects. Journal of Spinal Disorders,
10, 348–56.
Morrissey, M.C. (1989). Reflex inhibition of thigh muscles in knee injury. Causes and treatment. Sports
Medicine, 7, 263–76.
Mostardi, R.A., Noe, D.A., Kovacik, M.W. and Porterfield, J.A. (1992). Isokinetic lifting strength and
occupational injury. A prospective study. Spine, 17, 189–93.
Nargol, A.V.F., Jones, A.P.C., Kelly, P.J. and Greenough, C.G. (1999). Factors in the reproducibility of
electromyographic power spectrum analysis of lumbar paraspinal muscle fatigue. Spine, 24, 883–8.
Ng, J.K.F. and Richardson, C.A. (1996). Reliability of electromyographic power spectral analysis of back
muscle endurance in healthy subjects. Archives of Physical Medicine and Rehabilitation, 77, 259–64.
Ng, J.K., Richardson, C.A., Parnianpour, M. and Kippers, V. (2002). Fatigue-related changes in torque
output and electromyographic parameters of trunk muscles during isometric axial rotation exertion:
an investigation in patients with back pain and in healthy subjects. Spine, 27, 637–46.
Nicolaisen, T. and Jorgensen, K. (1985). Trunk strength, back muscle endurance and low-back trouble.
Scandinavian Journal of Rehabiltation Medicine, 17, 121–27.
Nutter, P. (1988). Aerobic exercise in the treatment and prevention of low back pain. Occupational
Oliver, C.W., Tillotson, K.M., Jones, A.P.C., Royal, R.A. and Greenough, C.G. (1996). Reproducibility of
lumbar paraspinal surface electromyogram power spectra. Clinical Biomechanics, 11, 317–21.
Parnianpour, M., Nordin, M., Kahanovitz, N. and Frankel, V. (1988). The triaxial coupling of torque
generation of trunk muscles during isometric exertions and the effect of fatiguing isoinertial
movements on the motor output and movement patterns. Spine, 13, 982–92.
Peach, J.P. and McGill, S.M. (1998). Classification of low back pain with the use of spectral
electromyogram parameters. Spine, 23, 1117–23.
Peach, J.P., Gunning, J. and McGill, S.M. (1998). Reliability of spectral EMG parameters of healthy back
extensors during submaximum isometric fatiguing contractions and recovery. Journal of
Electromyography and Kinesiology 8, 403–10.
Ransford, C.P. (1982). A role for amines in the antidepressant effect of exercise:a review. Medicine &
Science in Sports & Exercise 14, 1–10.
Roland, M. and Morris, R. (1983). A study of the natural history of back pain. Part 1: Development of a
reliable and sensitive measure of disability in low-back pain. Spine, 8, 141–4.
Rosenstiel, A.K. and Keefe, F.J. (1983). The use of coping strategies in chronic low back pain patients:
relationship to patient characteristics and current adjustments. Pain, 17, 33–44.
Roy, S.H., De Luca, C.J. and Casavant, D.A. (1989). Lumbar muscle fatigue and chronic lower back pain.
Spine, 14, 992–1001.
Roy, S.H., De Luca, C.J., Snyder-Mackler, L., Emley, M.S., Crenshan, R.L. and Lyons, J.P. (1990). Fatigue,
recovery and low back pain in varsity rowers. Medicine & Science in Sports & Exercise, 22, 463–9.
Roy, S.H., De Luca, C.J., Emley, M. and Buijs, R.J.C. (1995). Spectral electromyographic assessment of
back muscles in patients with low back pain undergoing rehabilitation. Spine, 20, 38–48.
Sahlin, K. (1992). Metabolic factors in fatigue. Sports Medicine, 13, 99–107.
Saltin, B. and Gollnick, P. (1983). Skeletal muscle adaptability: significance for metabolism and performance.
Washington, DC: American Physiological Society.
Sant’Ana Pereira, J.A.A., Ennion, S., Sargeant, A.J., Moorman, A.F.M. and Goldspink, G. (1997).
Comparison of the molecular, antigenic and ATPase determinants of fast myosin heavy chains in rat
and human: a single-fibre study. Pfluegers Archives - European Journal Physiology, 435, 151–63
Schwarzer, K.A. (1993). Measurement of perceived self-efficacy. Berlin: Forschung an der Freien Universitaet.
Sparto, P.J., Parnianpour, M., Marras, W.S., Granata, K.P., Reinsel, T.E. and Simon, S. (1997a).
Neuromuscular trunk performance and spinal loading during a fatiguing isometric trunk extension
with varying torque requirements. Journal of Spinal Disorders, 10, 145–56.
Sparto, P.J., Parnianpour, M., Reinsel, T.E. and Simon, S. (1997b). The effect of fatigue on multijoint
kinematics and load sharing during a repetitive lifting test. Spine, 22, 2647–54.
Sparto, P.J., Parnianpour, M., Barria, E.A. and Jagadeesh, J.M. (1999). Wavelet analysis of
electromyography for back muscle fatigue detection during isokinetic constant-torque exertions. Spine,
24, 1791–8.
Sung, P.S. (2003). Multifidi muscles median frequency before and after spinal stabilization exercises.
Archives of Physical Medicine and Rehabilitation, 84, 1313–8.
Sung, P.S., Zurcher, U. and Kaufman, M. (2005). Nonlinear analysis of electromyography time series as a
diagnostic tool for low back pain. Medical Science Monitor, 11, CS1–5.
Sung, P.S., Lammers, A.R. and Danial, P. (2009). Different parts of erector spinae muscle fatigability in
subjects with and without low back pain. Spine Journal, 9, 115–20.
Suter, E. and Lindsay, D. (2001). Back muscle fatigability is associated with knee extensor inhibition in
subjects with low back pain. Spine, 26, E361–6.
Suuden, E., Ereline, J., Gapeyeva, H. and Paasuke, M. (2008). Low back muscle fatigue during Sorensen
endurance test in patients with chronic low back pain: relationship between electromyographic spectral
compression and anthropometric characteristics. Electromyography and Clinical Neurophysiology 48,
185–92.
Symonds, T.L., Burton, A.K., Tillotson, K.M. and Main, C.J. (1996). Do attitudes and beliefs influence
work loss due to low back trouble? Occupational Medicine, 46, 25–32.
Taimela, S., Kankaanpaa, M. and Luoto, S. (1999). The effect of lumbar fatigue on the ability to sense a
change in lumbar position. Spine, 24, 1322–27.
Thompson, D.A., Biedermann, H.J., Stevenson, J.M. and MacLean, A.W. (1992). Changes in paraspinal
electromyographic spectral analysis with exercise: two studies. Journal of Electromyography and
Kinesiology, 2, 179–86.
Tsuboi, T., Satou, T., Egawa, K., Izumi, Y. and Miyazaki, M. (1994). Spectral analysis of electromyogram in
lumbar muscles: fatigue induced endurance contraction. European Journal of Applied Physiology 69,
361–6.
van Dieen, J.H. and Heijblom, P. (1996). Reproducibility of isometric trunk extension torque, trunk
extensor endurance, and related electromyographic parameters in the context of their clinical
applicability. Journal of Orthopaedic Research, 14, 139–43.
Van Doorn, J.W.C. (1995). Low back disability among self-employed dentists, veterinarians, physicians and
physical therapists in the Netherlands. A retrospective study over a 13-year period (N = 1,119) and an
early intervention program with 1-year follow-up (N = 134). Acta Orthopaedica Scandinavica,
Supplement, 263, 1–64.
Vestergaard-Poulsen, P., Thomsen, C., Sinkjaer, T., Stubgaard, M., Rosenfalck, A. and Henriksen, O.
(1992). Simultaneous electromyography and 31P nuclear magnetic resonance spectroscopy – with
application to muscle fatigue. Electroencephalography and Clinical Neurophysiology, 85, 402–11.
Waddell, G., Newton, M., Henderson, I., Somerville, D. and Main, C.J. (1993). Fear-Avoidance Beliefs
Questionnaire (FABQ) and the role of fear-avoidance beliefs in chronic low back pain and disability.
Pain, 52, 157–68.
Wang, M., Leger, A.B. and Dumas, G.A. (2005). Prediction of back strength using anthropometric and
strength measurements in healthy females. Clinical Biomechanics 20, 685–92.
Wilder, D., Aleksiev, A.R., Magnusson, M.L., Pope, M.H., Spratt, K.F. and Goel, V.K. (1996). Muscular
response to sudden load. A tool to evaluate fatigue and rehabilitation. Spine, 21, 2628–39.
Wood, K.A., Standell, C.J., Adams, M.A., Dolan, P. and Mannion, A.F. (1997). Exercise training to improve
spinal mobility and back muscle fatigability: a possible prophylaxis for low back pain? Physical
Medicine Research Foundation Symposium: ‘Clinical Approaches to Spinal Disorders’ Prague, Czech
Republic.
Chapter 10
Unmasking the Deconditioning

Paradigm for Chronic Low Back
Pain Patients
Jeanine Verbunt, Rob J.E.M. Smeets, and
Harriet Wittink
10.1 Introduction
Physical activity is behaviour. Like all behaviours it varies widely between (healthy) people and
tends to decrease as people age or develop health impairments. Physical activity and physical
fitness are interrelated; physical activity influences fitness, which in turn may modify the level of
physical activity. Health-related fitness consists of those components of physical fitness that are
affected by habitual physical activity and that are related to health status. An adequate level of
health-related fitness has been defined as a state of being able to perform daily activities with
vigour, and traits and capacities that are associated with a low risk of premature development of
hypokinetic diseases and conditions (Bouchard and Shephard 1994).
The components of health-related fitness are defined as morphological (body composition and
bone strength), muscular (muscular strength, muscular endurance, and flexibility), motor (pos-
tural control), cardiorespiratory and metabolic fitness (carbohydrate and lipid metabolism)
(Bouchard and Shephard 1994; Skinner and Oja 1994). A decrease in the volume and intensity of
physical activities may therefore reduce physical fitness and the capacity for physical activity, thus
creating a downward spiral resulting in physical deconditioning. Physical deconditioning is
defined as an integrated physiological response of the body to a reduction in metabolic rate; that
is, to a reduction in energy use or exercise level (Greenleaf 2004) and affects all aspects of health
related fitness.
Physical inactivity and the resultant physical deconditioning have been hypothesized to be
associated with both the aetiology and consequence of chronic pain (Verbunt et al. 2003). Physical
inactivity, however, has not only become a topic in chronic pain management, but also in public
health. In developed countries, physical inactivity is associated with considerable economic
burden, accounting for 1.5–3.0% of the total direct healthcare costs (Oldridge 2008). This may
give rise to the question whether patients with chronic low back pain (CLBP) exceed the general
population concerning their level of physical inactivity. The extent of the problem of decondi-
tioning in CLBP and its specific perpetuating role in chronicity still seems unclear. Despite this,
the restoration of function (i.e. health related fitness) through intensive physical rehabilitation
intervention continues to form the basis of many physiotherapy-guided training programmes
as well as comprehensive pain management programs. This approach is in part based on the
hypothesis of the ‘deconditioning syndrome’ in which deconditioning itself is seen as a factor
contributing to the intolerance to physical activities and subsequent loss of function and disability
in patients with chronic pain (Mayer et al. 1985, 1986; Vlaeyen and Linton 2000).
A second theoretical model, in which the role of disuse is included as one of the perpetuating
factors of pain is the fear-avoidance model (Vlaeyen and Linton 2000). According to the fear-
avoidance model, patients may interpret their pain as threatening (catastrophizing about their
pain) and this can result in pain-related fear, of which fear of movement/(re)injury is the most
salient. Both this fear and expectation of adverse consequences from keep on performing or
increasing activities may cause avoidance of these physical activities. In the long run, avoidance
behaviour might result in disability, depression and the loss of physical activities in daily life of
which loss of aerobic fitness is a consequence.
Although the influence of pain-related fear on the perceived disability level has currently been
confirmed based on several studies (Klenerman et al. 1995; Vlaeyen et al. 1995; Fritz et al. 2001),
its presumed negative influence on a patient’s level of physical activity in daily life (PAL) is still
inconclusive (Bousema et al. 2007). It can even be debated whether physical deconditioning
in patients with CLBP really exists (Smeets et al. 2007 ; Smeets and Wittink 2007 ; Wittink
et al. 2008).
In this chapter we will examine the deconditioning paradigm by summarizing the existing
literature on this topic. We will start with a section on PALs in patients with CLBP. Next we will
discuss studies describing the various aspects of health related fitness in patients with CLBP, fol-
lowed by the latest theories regarding activity-related behaviour and the measurement problems
in assessing physical fitness parameters in patients with CLBP. And finally, overall conclusions on
the impact of deconditioning in CLBP will be provided.
10.2 The level of physical activity in patients with CLBP

Are patients with CLBP really inactive in comparison with healthy individuals? During recent
years, several cross-sectional studies have been performed to identify inactivity in patients with
CLBP. Whereas both Nielens and Plaghki (2001) and Berg-Emons et al. (2007) reported a sig-
nificantly lower PAL in patients with CLBP, this finding was not confirmed by the studies of
Protas (1999) and Verbunt et al. (2001) that reported a PAL in patients with CLBP that was com-
parable with the PAL of healthy individuals matched for age and gender. In 2007, a literature
review on the presence of disuse in chronic pain was published, which again could not confirm
the existence of disuse. Although a lower mean activity level in patients with pain could not be
confirmed, it seems that in patients with pain activity-fluctuations during the day are more
pronounced (van Weering et al. 2007).
One explanation for the diversity in findings between studies focussing on disuse in CLBP
could be a variation in the percentage of study-participants with paid jobs. Persons with CLBP
who are still working at least will have a PAL that is sufficient to meet the physical demands of
their jobs. A hypothesis could therefore be that patients with CLBP who are working will have
higher PALs than those who are not working. Studies that did not report a decrease in PAL in
patients with CLBP indeed did include a higher percentage of participants who were still engaged
in occupational activities. However, whether the patients having a paid job also performed their
job with the normally required level of activity and effort was not studied.
A second explanation for the diversity of findings in studies on disuse in CLBP could be the use
of different assessment instruments to measure physical activity. Studies in which activity assess-
ment was based on self-report measures more frequently report a lower activity level for patients
with musculoskeletal pain as compared to studies using accelerometry or activity monitoring
(van Weering et al. 2007; Verbunt et al. 2008). Studies using an outcome measure on the basis
of accelerometry (total counts of activity above a threshold) found no difference compared
to healthy matched controls (Verbunt et al 2001; Van Weering et al. 2008), while a study using
UNMASKING THE DECONDITIONING PARADIGM 187
activity monitoring, including a set of three acclerometers which allows differentiating to types of
activity, showed a reduced PAL in patients with CLBP (Van den Berg-Emons et al, 2007).
In 2007, Bousema et al. presented the first prospective cohort study on physical activity assessed
by self-reported questionnaires as well as accelerometry in patients with low back pain (LBP).
Patients were measured at 4–7 weeks after the onset of LBP until 1 year later (Bousema et al.
2007). In individuals who recovered as well as those who developed CLBP, the activity level dur-
ing this year increased. In addition, changes in the level of physical activity and health-related
fitness-parameters such as strength and lean body mass did not differ between both groups.
Unexpectedly, the perceived level of disability of the individuals who developed CLBP was not
associated with the level of registered activity, but was associated with the perceived decline in the
level of activities after the onset of pain (Verbunt et al. 2005b). In summary, although hypothe-
sized, the presence of physical inactivity in patients with chronic pain has still not been confirmed
in the literature on physical activity in chronic musculoskeletal pain.
10.3 Health-related fitness in patients with CLBP

Are patients with musculoskeletal pain really deconditioned in comparison with healthy indi-
viduals? If they are, the components of health-related fitness defined as morphological (body
composition, and bone strength), muscular (muscular strength, muscular endurance, and flexi-
bility), motor (postural control), cardiorespiratory (VO2max) and metabolic fitness (carbohydrate
and lipid metabolism) (Bouchard and Shephard 1994; Skinner and Oja, 1994) should be different
from healthy controls. And, are all components of health-related fitness equally effected in
patients with pain? We will describe the current research on health related fitness in patients with
CLBP next.
10.3.1 Body composition and bone strength

Reduced physical activity while maintaining the same caloric intake tends to lead to weight gain
and changes in body composition. Based on studies using a cross-sectional design, patients with
CLBP have a higher percentage of body fat as compared to age and gender matched individuals
(Toda et al. 2000; Verbunt et al. 2005a). In addition, in chronic pain patients, a significant rela-
tionship between weight gain and decreased physical activity, increased emotional distress, and
accident liability was found (Jamison et al. 1990).
Reduced physical activity can also result in a change in bone strength; the response of the human
skeleton to unloading appears to be a rapid and sustained increase in bone resorption and a more
subtle decrease in bone formation (Zerwekh et al. 1998). The greatest bone loss occurs at weight-
bearing skeletal sites. Recovery of bone mass after immobilization is typically much slower than
the rate at which it is lost. A recent review suggests that there is evidence that back pain and
impaired bone health share common environmental (such as a lower activity level), but also
genetic correlates, indicating that bone health ought to be considered in the context of longer
lasting back pain in otherwise healthy individuals (Briggs et al. 2008). Bone responds to mechan-
ical load and the response tends to be U-shaped, where both reduced physical activity and very
high levels of physical activity are associated with bone loss (Briggs et al. 2008). Changes in bone
strength can also be explained by changes in muscle reactivity due to pain. Individuals with severe
back pain tend to stiffen the trunk and limit normal movement at the intervertebral joints (van
Dieen et al. 2003a). Furthermore, in order to enhance the stability of the lumbar spine, patients
with LBP show altered trunk muscle recruitment patterns (Mok et al. 2007). This altered neuro-
muscular strategy can decrease the opportunity for normal physiologic stresses, necessary for the
maintenance of skeletal integrity, to be transferred through the vertebrae. This might explain
why a history of back pain is reportedly associated with site-specific decreased bone mineral den-
sity in adults.
10.3.2 Muscle strength and endurance

In response to decreased usage (i.e. a decrease in daily activities), skeletal muscle undergoes an
adaptive reductive remodelling. This adaptive response has been found in muscle disuse that
occurs especially in a situation of immobility, such as in detraining, space flight, bed rest and in a
lesser extent in situations of inactivity and ageing (Gogia et al. 1988). A decrease in muscle
strength, or muscle weakening, during detraining or reduced activity is attributable to both a loss
of muscle mass and changes in muscle composition. Muscle mass decreases due to both a general
reduction in muscle cross-sectional fibre area as well as a reduction in the overall number of
muscle fibres. These changes are associated with a decline in force production and decreased
electromyographic (EMG) activity (Mujika and Padilla 2001).
Decreases of 6–40% in muscle strength have been observed within 4–6 weeks of bed rest
(Bloomfield 1997). In micro-gravity simulation models, postural muscles that normally counter-
act the effects of gravity have been reported to become atrophic to a greater extent than fast con-
tracting locomotor muscles (Convertino et al. 1997). This implies that especially muscles situated
on the trunk and lower extremities are affected by physical deconditioning (Berry et al. 1993,
Greenleaf 1997) with a consequence of early back muscle fatigue as presented in chapter 11. This
assumption of physical deconditioning in trunk muscles was indeed shown in an eight-week bed
rest study in 10 healthy males that indicated that bed rest resulted in selective atrophy of the
multifidus muscles. An increased cross-sectional area (CSA) of the trunk flexor musculature
(increases in psoas, anterolateral abdominal, and rectus abdominus muscles) is thought to reflect
muscle shortening or possible overactivity during bed rest (Hides et al. 2007). According to some
researchers, some of the changes resemble those seen in LBP and may partly explain the negative
effects of bed rest seen in LBP sufferers.
Besides changes in muscle mass, physical deconditioning can also result in changes in muscle
composition. As presented in Chapter 9, the human muscles are composed of different muscle
fibres. Muscle disuse is often accompanied by increased fatigability of the muscle because of
the reduced oxidative capacity of disused muscles. Muscle capillary density decreases within
2–3 weeks of disuse (Greenleaf 1997). Furthermore, capillary loss and reductions in blood flow at
rest and arteriolar responsiveness may contribute to this increased fatigability through the resulting
impairment of the supply of energy substrates and oxygen to the muscle (Degens and Alway 2006).
If patients with CLBP are really deconditioned, it would be reasonable to assume that they have
measurable muscle atrophy, and reduced muscle strength and endurance. Over the last two dec-
ades much research has been done on the specific role of changes in muscle strength of the low
back muscles, especially the paraspinal and in particular the multifidus muscles, in the segmental
stability of the lower spine. It is hypothesized that segmental instability is an important contribut-
ing factor in the persistence of low back pain; when the stability at the segmental level is not suf-
ficient, even little disturbances at this level might result in high forces at the region of the
intervertebral disc and surrounding tissues. This could result in repetitive high strains and pos-
sibly re-injury of these tissues and eventually causing pain (Panjabi 1992a, 1992b; Danneels et al.
2001). It has been suggested that specific training of these paraspinal, and more specifically the
lumbar multifidus muscles, is effective in non-specific CLBP (O’Sullivan et al. 1997; Hides et al.
2007) although the evidence is inconclusive (Macedo et al. 2009).
Muscle atrophy in CLBP has been studied based on a number of imaging studies. Several
magnetic resonance imaging (MRI) or computed tomography (CT) studies showed a smaller
CSA of the paraspinal muscles, psoas and multifidus (Gibbons et al. 1997a; Parkkola et al. 1993).
Danneels et al. studied the total paraspinal muscle mass, the isolated multifidus, and the psoas at
three different levels using CT. The results showed that only the CSA of the multifidus and only
at the lowest level (lower end-plate of L4) was found to be statistically smaller in patients with
CLBP (Danneels et al. 2000). Others confirmed these findings of a localized rather than general-
ized pattern of multifidus and psoas muscle atrophy in patients with CLBP (Mayer et al. 1989;
Kader et al. 2000; Barker et al. 2004; Hides et al. 2007).
The greatest asymmetry between sides was seen at the L5 vertebral level in patients with unilat-
eral pain presentations. This asymmetry might be caused, at least partly, by inactivity/disuse and
ageing. One study on patients with unilateral back pain shows clear atrophy of both the multi-
fidus and psoas muscle at the symptomatic side suggesting that not only disuse but also neu-
romuscular changes play a role (Barker et al. 2004). However, there are inconsistencies between
studies. When one corrects for age, gender, body mass index and activity level there is limited
evidence that once LBP originates, atrophy of the multifidus muscles might play a role in the
complex of CLBP and its resulting disability.
The overall results of studies reporting on changes in muscle composition in CLBP based on
biopsy findings are inconclusive; several studies suggest the existence of type II fibres atrophy in
CLBP (Matilla et al. 1986; Zhu et al. 1989; Weber et al. 1997), but others state that type IIX fibres
are generally smaller even in healthy back muscles and decrease with advancing age. Several stud-
ies show that the longer the duration of LBP the more fibre-type transformation seems to occur
(increased IIX:IIA ratio and decreased I:IIX ratio) (Zhu et al. 1989; Mannion et al. 2000). The
observed atrophy may simply represent atrophy of this type IIX fibre (Rantanen et al. 1993;
Mannion et al. 2000). One study with healthy controls matched for age and anthropometry does
not show this transformation, although no further distinction in fibre type II was made (Crossman
et al. 2004). No study has yet been able to convincingly show significant differences in the ratio of
the size of the type I:IIX fibres between patients with CLBP and matched controls.
10.3.3 Postural control

Due to physical immobility or inactivity, changes in motor unit recruitment may occur that
include an impairment of the ability to activate motor units (Sale et al. 1982). Immobility decreases
coordination and balance (Haines 1974). Motor control is reported to be affected after a period
of bed rest. Maintaining a high degree of coordination requires frequent performance of an activ-
ity under conditions in which the sensory perception of the motor performance can be checked
for accuracy and errors may be corrected. Bed rest decreases the amount of proprioceptive stimuli,
which are responsible for regulating neuromuscular performance.
Hides et al. studied the effect of bed rest on trunk muscle activation in 10 healthy males who
underwent 8 weeks of bed rest with a 1-year follow-up (Hides et al. 2007). A repetitive knee move-
ment model at four movement speeds in non-weight bearing was used to assess activity in five
superficial lumbopelvic (LP) muscles and abdominal flexor-lumbar extensor co-contraction for
lumbopelvic stabilization. Increased muscle activity and decreased co contraction during a repet-
itive-movement task were found. During bed rest, increased activity appeared at the highest
movement speed and seemed generalized across all five superficial LP muscles up to 1 year after
bed rest. The same subjects showed significant atrophy of the deep multifidus muscles, but not in
the lumbar erector spinae or abdominal muscles (Hides et al. 2007) The authors suggest that due
to this deep muscle atrophy, in order to maintain the necessary amount of LP stiffness for the
knee movement task, the central nervous system increased the recruitment of the superficial LP
muscle systems. However, research in patients with acute LBP is not available.
There is a wide variability in reported changes in sensorimotor control of the trunk muscles in
low back and pelvic pain. Flexion-relaxation (FR) refers to a pattern of muscle activity during
trunk flexion, in which the lumbar muscles initially contract, but ultimately relax, at what appears
to be a distinct point in the flexion range of motion (ROM). ROM and FR are anatomically cor-
related, and hence, a large ROM will induce FR. The variables are not necessarily independent.
Studies have shown FR of the lumbar muscles to be a consistent and predictable pattern in most
normal subjects without back pain. As compared with normal subjects with no history of back
pain, subjects with CLBP often show elevated muscle activity during full voluntary trunk flexion
and fail to achieve FR (Watson et al. 1997; Neblett et al. 2003). However, it should be noted that
patients with CLBP often fail to reach the ROM necessary for FR. Patients may be anatomically
limited or have fear and/or pain during forward bending which will influence the results of
the test.
During static tasks, such as standing, inconsistent levels of trunk muscle activation have been
found in patients with CLBP. Patients with CLBP have been reported to have increased (Arena
et al. 1989; Healey et al. 2005), decreased (Ahern et al. 1988; Cassisi et al. 1993) and similar surface
electromyography (sEMG) (Kravitz et al. 1981) amplitudes compared with controls. Wong et al;
reported, that workers with LBP spent a significantly higher percentage of time in static trunk
posture when compared to normal (Wong et al; 2009). In a study investigating sEMG during
unsupported ‘usual’ versus ‘slumped’ sitting no differences in trunk muscle activity between
healthy controls and patients with CLBP were observed (Dankaerts et al. 2006). However, when
these patients were subdivided into an active extension pattern group and a flexion pattern group,
differences were identified. A study comparing patients with CLBP to healthy controls while per-
forming slow trunk motions about the neutral posture and isometric ramp contractions while
seated upright, showed greater ratios of EMG amplitudes of antagonists over agonists, and of
lumbar over thoracic erector spinae muscles in patients (van Dieen et al. 2003a).
The recruitment of stabilizing trunk muscles during motion of the upper limbs appears to be
different in persons with and without back pain (Hodges 2001). A number of studies have found
increased superficial lumbar muscle activation in patients with LBP and increased coactivation of
the abdominal muscles enhancing the stability of the lumbar spine. In addition activation of the
deep lumbar muscles is impaired (Hodges et al. 2003; van Dieen et al. 2003b; Geisser et al. 2005).
Several studies showed disturbances of propriosepsis in patients with CLBP (Brumagne et al.
2000; Leinonen et al. 2002; O’Sullivan et al. 2002), and some not (Asell et al. 2006). In most of
these studies PAL of the subjects was not controlled for. It is therefore unclear if deconditioning
influenced these results.
10.3.4 Aerobic capacity

Aerobic capacity is a measure of the ability and efficiency of the body to take up oxygen and use
it to convert fat and carbohydrates into energy. It is an important index of cardiovascular fitness
and increases with sustained rhythmic contractions of a large percentage of muscle mass. The
oxygen uptake system has a central and a peripheral component. The central component is
cardiac output (Q), which is equivalent to heart rate (HR) multiplied by stroke volume (SV).
The peripheral component (a-vO 2) is the difference between arterial and mixed venous
·
blood. Therefore, the level of VO2max depends on both the heart’s ability to pump blood as well
as the capacity of the muscle tissues to extract oxygen from the blood. Aerobic capacity is
expressed in litres/minute (l/min) and often adjusted for bodyweight millilitres/kilogram/minute
·
(ml×kg-1×min-1). The magnitude of reduction in VO2max ml×kg-1×min-1 varies according to the
duration of inactivity and the initial level of aerobic capacity, but appears to be independent of
·
age or gender. In general, fitter people appear to experience relatively greater losses in VO2max,
but retain a more efficient oxygen uptake system than people who are sedentary or who have not
·
exercised for prolonged periods (Saltin et al. 1968; Coyle et al. 1984). The decline in VO2max is the
result of a combination of central (cardiac output) and peripheral factors (arterio-venous differ-
·
ence). The rate at which VO2max changes is determined by two factors: one fast-related to cardiac
output, one slow-related to decrease in muscle oxidative potential (see also ‘Muscle strength and
endurance’ section).
Most physical activities are described in terms of their energy or metabolic cost (Ainsworth
et al. 2000). Physical activities are coded in metabolic equivalent (MET) intensity levels. One
MET is considered a resting metabolic rate obtained during quiet sitting and equals an oxygen
uptake of 3.5ml/kg/min. The oxygen cost for physical activities ranges from 0.9MET for sleeping
to 18MET for running at 10.9mph (Ainsworth et al. 2000). For occupational activities energy cost
ranges from about 1.5MET for sitting in a meeting to 17MET for fast axe chopping in forestry.
·
The higher a person’s VO2max, the more energy is available to perform physical activities. For
an individual with a good aerobic exercise capacity, physical activity generates less fatigue,
which means that he or she is capable of performing at a higher level of energy demand (i.e.
‘do more’).
·
The VO2peak1 attained during a graded maximal exercise to volitional exhaustion is considered
as the single best indicator of aerobic exercise capacity by the World Health Organization
·
(Shephard et al. 1968). The gold standard for determining VO2peak in an individual is by meta-
bolic measurement system analysis of O2 and CO2 gas in expired air at regular intervals and a
plateauing of VO2 during increasing workloads (Oddsson et al. 1997). However, in many subjects
·
including children, patients, and athletes a plateau in V O2 is not observed (Rowland 1993;
Wassermen et al. 1999; Lucia et al. 2006). Several secondary parameters have therefore been estab-
· ·
lished to determine VO2max without a VO2 plateau. These parameters includes a failure of HR to
increase with further increases in exercise intensity, a respiratory exchange ratio (RER) >1.15 and
a rating of perceived exertion of more than 17 (Borg 6–20 scale) (Howley et al. 1995).
The mode of exercise testing and the age and gender of the subject determine oxygen uptake
with exercise testing. In general, the highest oxygen uptake is achieved with the type of exercise
·
that uses the greatest amount of muscle mass. In normal subjects, the highest VO2max is obtained
with treadmill testing due to the quantity of the muscle mass involved, followed by bicycle testing.
·
VO2max achieved by bicycle testing is reported to be 5–15% lower than with treadmill testing in
normal subjects (Hermansen and Saltin 1969; Hermansen et al. 1970). A variety of (submaximal)
tests have been developed estimating aerobic capacity, when direct measurement is not possible
(See, for review, Noonan and Dean 2000.) These tests usually involve running/walking for a given
time or distance, such as the 12min walk/run test (Cooper 1968), the shuttle test (Leger and
Lambert 1982), various step tests (Siconolfi et al. 1985; Francis, 1987) and the 2km walk test (Oja
et al. 1991). Longer distances and shorter test times are associated with higher levels of aerobic
fitness.
·
Other tests estimate VO2max by submaximal testing and extrapolation to maximal heart rate by
treadmill walking or bicycling against a predetermined load with measurement of heart rates
(Astrand and Ryhming 1954). These tests were mostly developed for testing aerobic capacity in
· ·
healthy people and were validated by comparing actual measured VO2max to predicted VO2max or
to the test performance. Nomograms and prediction equations, derived from exercise testing
1 Maximal exercise tests in patients are usually terminated when the subject despite strong verbal encour-
agement from the experimenters, is unwilling or unable to continue. The appropriate term to use is there-
·
fore peak oxygen consumption (VO2peak), which represents the highest oxygen uptake during an exercise
test to volitional exhaustion. Both terms will be used interchanged throughout this chapter.
·
large samples of subjects, were developed to estimate VO2max in healthy populations. Even as
they include age and gender corrections, they reflect a mean aerobic capacity level for a particular
gender at a particular age and can therefore never be as precise as direct measurement of oxygen
uptake. Maximal heart rate may be predicted from age using any of several published equations,
such as 220 minus age, however interindividual variability is quite high (standard deviation
is 10–12 beats/min) (Oddsson et al. 1997). As a result, there is potential for considerable error
in the use of methods that extrapolate submaximal test data to an age-predicted HRmax
(ACSM, 2000).
Whether patients with CLBP have a lower level of aerobic capacity as compared to healthy
controls was a subject of various studies. In a systematic review, nine studies in which at least 50%
of the patients had CLBP were assessed with a submaximal testing procedure to calculate the
·
V O2max were identified (Smeets et al. 2006). The studies are difficult to compare due to the
heterogeneity of protocols, populations and controls used. These studies used different testing
·
procedures and methods to calculate VO2max, sometimes even without adjustments for gender.
Most studies used submaximal exercise-testing. Those studies that attempted maximal testing
found that most patients were not able to reach the criteria for VO2max (Wittink et al. 2000b).
Furthermore, most studies used inappropriate healthy controls (no matching for age and gender)
or used normative data of insufficiently relevant healthy population. Populations differed between
studies in, for example, work status and diagnosis (herniated disc versus non-specific CLBP).
Despite the long existence of the deconditioning theory, this review concluded that it still is not
clear whether symptoms of physical deconditioning, especially a reduced level of aerobic capacity,
develop or even exist in patients with CLBP. However, authors have been consistent in suggesting
that differences in levels of physical activity may contribute to their discrepant findings. For
instance, Nielens and Plaghki (2001) and Wittink et al. (2000a) found male–female differences in
·
VO2max and suggested that these differences are attributable to different levels of physical activity,
especially regarding work, household, sport, and leisure time. More recently, Brox et al. reported
a lower aerobic capacity in patients with low back pain who were sick-listed for 8–12 weeks, but
not in patients with CLBP waiting for lumbar instrumented fusion, compared with healthy per-
sons matched for gender and age (Brox et al. 2005). Unfortunately, they only used four age
cohorts, which limits its generalizability. Rasmussen-Barr et al. showed that male but not female
patients with recurrent low back pain who were still at work have a similar aerobic capacity level
as age- and gender matched controls, indicating that staying active at work might indeed be an
important factor in the prevention of aerobic deconditioning (Rasmussen-Barr et al. 2008).
More recently, Smeets et al. reported that patients with moderately to severely disabling CLBP
had a significantly lower level of aerobic capacity compared to healthy subjects matched for age,
gender, and sport activity (Smeets et al. 2009). Men with CLPB even had a significantly lower level
·
of VO2max than women with CLBP. It appeared that the level of physical activity during leisure
time as well as work including household work were significantly associated with this lower level
of aerobic capacity. A possible explanation for the significantly lower level in aerobic capacity for
men compared to women with CLBP (20% lower versus 13% lower than normative population)
may be the significantly higher level of household activities performed by women, confirming the
hypothesis of Nielens and Plaghki (2001) that women with chronic pain are generally more active
once they are at home because they are still engaged in childcare and various household duties.
Since this loss of aerobic capacity might also be caused by variables of the fear-avoidance model,
Smeets et al. also investigated the role of these factors, and their results clearly showed that pain
intensity, pain-catastrophizing, fear of movement and depression were not significantly associ-
ated with the loss of aerobic capacity (Smeets et al. 2009).
Since all above mentioned studies are based on cross-sectional designs, the possibility that
deconditioning was already present before patients developed back pain cannot be ruled out. This
explanation might by supported by the finding that almost 40% of the level of VO2max can be
explained by genetic factors (McArdle et al. 2001). Nevertheless, prospective research in healthy
subjects has not identified low aerobic capacity levels or physical inactivity as a risk factor for
developing CLBP (Battie et al. 1989; Picavet and Schuit 2003). To further investigate the develop-
ment and impact of loss of aerobic capacity levels, longitudinal studies should be performed in
patients with acute LBP, while monitoring all putative influential factors.
10.3.5 Metabolic factors

Insulin is the major hormonal regulator of energy storage and release. It stimulates glycogen
synthesis, aerobic and anaerobic glycolysis and the synthesis of proteins and fatty acids in the
liver. In numerous bed rest studies, it has been shown that bed rest reduces sensitivity to insulin
(Pavy-Le Traon et al. 2007). Five days of bed rest only, was already associated with the develop-
ment of insulin resistance, dyslipidaemia, increased blood pressure, and impaired microvascular
function in 20 healthy volunteers (Hamburg et al. 2007). In a recent preliminary study, 18 non-
exercising healthy volunteers in their twenties decreased their daily walking for 2–3 weeks to
1500 pedometer-recorded steps from a mean value of more than 6000 steps. During this period
they developed metabolic changes that suggested a decrease in insulin sensitivity and an attenua-
tion of post-prandial lipid metabolism, and physical changes that suggested that calories used to
maintain muscle mass with greater stepping may have been partitioned to visceral fat (Olsen et al.
2008). Extrapolating from these data it would seem evident that patients with CLBP would show
metabolic changes as a result from inactivity. To date, there are however no studies that have
investigated metabolic changes in patients with CLBP.
10.3.6 Conclusion on health-related fitness in patients with CLBP

Currently, no consistent evidence could be found in the literature that the spectrum of changes in
morphological, muscular, motor (postural control), cardiorespiratory (VO2max) or metabolic fac-
tors (carbohydrate and lipid metabolism) in patients with CLBP are due to physical deconditioning.
10.4 Future research

10.4.1 Future research on physical activity in CLBP
What could explain why not all patients with musculoskeletal pain who feel highly disabled even-
tually become inactive? In recent years, several explanatory models (Vlaeyen and Morley 2004;
Hasenbring and Verbunt 2010) have been introduced to explain differences in activity related
behaviour in chronic pain. In addition to the fear avoidance model that introduced fear of move-
ment/re-injury as the explaining factor for a disabling decrease in physical activity, alternative
models have been proposed to explain other strategies concerning activity related behaviour in
chronic pain. Hasenbring and colleagues hypothesized that, in addition to patients using avoid-
ance strategies as a coping mechanism, other patients with pain will have the tendency to cope
with pain by using persistent strategies (Hasenbring et al. 1994). These patients persist in the
performance of activities and appear to ignore their pain and overload their muscles (overuse),
resulting in muscular hyperactivity (further information is presented in Chapter 15). Long-term
muscular hyperactivity and long term false straining of the muscles eventually can result in
chronification of pain.
In accordance with the hypothesis of Hasenbring, Van Houdenhove suggested that, especially
in patients with fibromyalgia and chronic fatigue syndrome, a high level of ‘action proneness’,
promoting an overactive lifestyle, may play a predisposing, initiating and perpetuating role in the
level of disability (Van Houdenhove et al. 1995). People who have an overactive lifestyle may
be at a higher risk of overburdening. If these persons are deprived of overactivity as their favourite
coping strategy, for example due to pain or functional limitations, the level of psychological dis-
tress can increase. An alternative explanation for differences in activity related behaviour in pain
has been introduced by Vlaeyen and Morley in their Mood-as-Input Model (Vlaeyen and Morley
2004). This model explains why some patients will stop, whereas others will continue performing
their activities when confronted by pain. According to the Mood-as-Input Model, the informa-
tional value of the mood in a certain context, rather than the mood itself, determines whether
participants persist at a certain task. For some individuals a negative mood will facilitate task
persistence; whereas for others it will lead to disengagement. Mood can result in different moti-
vational effects leading to inactivity or persistence in activities regardless the level of pain. The
hypothesis of Hasenbring and Vlaeyen and Morley could explain the coexistence of high levels of
physical activity and high levels of disability. Research to confirm different activity related strate-
gies in chronic pain has to be performed.
For future research on activity assessment in pain we urge the incorporation of monitoring of
activities in daily life, preferably combined with methods to measure the total energy consump-
tion (Westerterp 1999; Philippaerts et al. 2001; van den Berg-Emons et al. 2007; Verbunt et al.
2008). And we do need more cohort studies of patients with acute LBP in order to determine
whether changes in the level of activities, and even more important the type of activities, occur,
and whether deconditioning symptoms appear in case the pain persists.
10.4.2 Future research on health related fitness in CLBP

Two important topics have to be addressed in order to be able to answer the question whether
physical deconditioning has a perpetuating role in CLBP. The first, major issue is to agree upon
or develop valid and reliable measurement tools for the assessment of physical fitness parameters
in people who experience pain. Assessment of fitness related parameters most often requires
functional testing, in which patients with pain have to perform an exercise test. To obtain valid
and reliable measurements of aerobic capacity, muscle strength and endurance, maximal effort is
needed. Researchers warn that submaximal performance may play a role in the outcome of fitness
testing in patients with pain and that influences such as motivation, fear that the action will cause
pain or damage and the emotional state of a patient, level of pain, avoidance of pain, self-efficacy
and deconditioning, need to be considered when the results are assessed (Keller et al. 1999;
Watson 1999; Oddsson and De Luca 2003; Crossman et al. 2004; Kramer et al. 2005; Verbunt
et al. 2005a; Smeets et al. 2007). Since voluntary muscle strength is influenced by many of the
above mentioned factors strength and endurance testing is regarded as a invalid method to assess
whether muscular deconditioning is present in patients with low back pain. Also the use of EMG-
studies as a valid and reliable of back muscle weakness and muscle composition based on EMG-
assessment is still controversial (Lariviere et al. 2002; Crossman et al. 2004). More research on
influencing factors during performance testing in chronic pain is needed to be able to improve
assessment of physical fitness related parameters in pain.
Second, current evidence on deconditioning related parameters in pain is especially based
on cross-sectional research. To further investigate the development and impact of loss of the level
of aerobic capacity, longitudinal studies should be performed in patients with acute LBP, while
monitoring all putative influential factors.
10.5 General conclusion

The general assumption that patients with CLBP suffer from disuse and physical deconditioning
still lacks empirical ground. The level of physical activity and fitness-related health of patients
with CLBP seems less or equal as compared to healthy individuals. However, results of functional
testing in patients with pain are often influenced by submaximal during performance testing of
patients as compared to healthy individuals. Further longitudinal research especially is therefore
needed to improve assessment of both physical activity and fitness-related health parameters in
patients with pain, to distinguish different styles of activity-related behaviour in pain, and to
identify and control for factors influencing performance testing in CLBP.
References
ACSM (2000) American College of Sports Medicine: Guideline for exercise testing and prescription,
Philadelphia, PA: Lippincott, Williams and Wilkins.
Ahern, D. K., Follick, M. J., Council, J. R., Laser-Wolston, N. & Litchman, H. (1988) Comparison of
lumbar paravertebral EMG patterns in chronic low back pain patients and non-patient controls. Pain,
34, 153–60.
Ainsworth, B. E., Haskell, W. L., Whitt, M. C., et al. (2000) Compendium of physical activities: an update
of activity codes and MET intensities. Med Sci Sports Exerc, 32, S498–504.
Arena, J. G., Sherman, R. A., Bruno, G. M. & Young, T. R. (1989) Electromyographic recordings of 5 types
of low back pain subjects and non-pain controls in different positions. Pain, 37, 57–65.
Asell, M., Sjolander, P., Kerschbaumer, H. & Djupsjobacka, M. (2006) Are lumbar repositioning errors
larger among patients with chronic low back pain compared with asymptomatic subjects? Arch Phys
Med Rehabil, 87, 1170–6.
Astrand, P. O. & Ryhming, I. (1954) A nomogram for calculation of aerobic capacity (physical fitness)
from pulse rate during sub-maximal work. J Appl Physiol, 7, 218–21.
Barker, K. L., Shamley, D. R. & Jackson, D. (2004) Changes in the cross-sectional area of multifidus and
psoas in patients with unilateral back pain: the relationship to pain and disability. Spine, 29, E515–9.
Battie, M. C., Bigos, S. J., Fisher, L. D., et al. (1989) A prospective study of the role of cardiovascular risk
factors and fitness in industrial back pain complaints. Spine, 14, 141–7.
Berry, P., Berry, I. & Manelfe, C. (1993) Magnetic resonance imaging evaluation of lower limb muscles
during bed rest—a microgravity simulation model. Aviat Space Environ Med, 64, 212–8.
Bloomfield, S. A. (1997) Changes in musculoskeletal structure and function with prolonged bed rest. Med
Sci Sports Exerc, 29, 197–206.
Bouchard, C. & Shephard, R. J. (1994) Physical activity, fitness and health: the model and key concepts.
Physical activity, fitness and health. International Proceedings and consensus statement. Champaign, IL:
Kuman Kinetics.
Bousema, E. J., Verbunt, J. A., Seelen, H. A., Vlaeyen, J. W. & Knottnerus, J. A. (2007) Disuse and physical
deconditioning in the first year after the onset of back pain. Pain, 130, 279–86.
Briggs, A. M., Straker, L. M. & Wark, J. D. (2008) Bone health and back pain: What do we know and where
should we go? Osteoporos Int., 2009, 209–19.
Brox, J. I., Storheim, K., Holm, I., Friis, A. & Reikeras, O. (2005) Disability, pain, psychological factors
and physical performance in healthy controls, patients with sub-acute and chronic low back pain:
a case-control study. J Rehabil Med, 37, 95–9.
Brumagne, S., Cordo, P., Lysens, R., Verschueren, S. & Swinnen, S. (2000) The role of paraspinal muscle
spindles in lumbosacral position sense in individuals with and without low back pain. Spine, 25, 989–94.
Cassisi, J. E., Robinson, M. E., O’Conner, P. & MacMillan, M. (1993) Trunk strength and lumbar
paraspinal muscle activity during isometric exercise in chronic low-back pain patients and controls.
Spine, 18, 245–51.
Convertino, V. A., Bloomfield, S. A. & Greenleaf, J. E. (1997) An overview of the issues: physiological
effects of bed rest and restricted physical activity. Med Sci Sports Exerc, 29, 187–90.
Cooper, K.H. (1968) A means of assessing maximal oxygen intake. JAMA, 203, 201–4.
Coyle, E. F., Martin, W. H., 3rd, Sinacore, D. R., Joyner, M. J., Hagberg, J. M. & Holloszy, J. O. (1984)
Time course of loss of adaptations after stopping prolonged intense endurance training. J Appl Physiol,
57, 1857–64.
Crossman, K. B. M., Mahon, M. M., Watson, P. J. P. M., et al. (2004) Chronic low back pain-associated
paraspinal muscle dysfunction is not the result of a constitutionally determined ‘adverse’ fiber-type
composition. Spine, 29, 628–34.
Dankaerts, W., O’Sullivan, P., Burnett, A. & Straker, L. (2006) Altered patterns of superficial trunk muscle
activation during sitting in nonspecific chronic low back pain patients: importance of subclassification.
Spine, 31, 2017–23.
Danneels, L. A., Vanderstraeten, G. G., Cambier, D. C., Witvrouw, E. E. & De Cuyper, H. J. (2000)
CT imaging of trunk muscles in chronic low back pain patients and healthy control subjects. Eur
Spine J, 9, 266–72.
Danneels, L. A., Cools, A. M., Vanderstraeten, G. G., et al. (2001) The effects of three different training
modalities on the cross-sectional area of the paravertebral muscles. Scan J Med Sci Sports, 11, 335–41.
Danneels, L. A., Vanderstraeten, G. G., Cambier, D. C., Witvrouw, E. E. & De Cuyper, H. J. (2000) CT
imaging of trunk muscles in chronic low back pain patients and healthy control subjects. Eur Spine J,
9, 266–72.
Degens, H. & Alway, S. E. (2006) Control of muscle size during disuse, disease, and aging. Int J Sports Med,
27, 94–9.
Francis, K. T. (1987) Fitness assessment using step tests. Compr Ther, 13, 36–41.
Fritz, J. M., George, S. Z. & Delitto, A. (2001) The role of fear-avoidance beliefs in acute low back pain:
relationships with current and future disability and work status. Pain, 94, 7–15.
Geisser, M. E., Ranavaya, M., Haig, A. J., et al. (2005) A meta-analytic review of surface electromyography
among persons with low back pain and normal, healthy controls. J Pain, 6, 711–26.
Gibbons, L. E., Videman, T. & Battie, M. C. (1997a) Isokinetic and psychophysical lifting strength, static
back muscle endurance, and magnetic resonance imaging of the paraspinal muscles as predictors of
low back pain in men. Scan J Rehab Med, 29, 187–91.
Gibbons, L. E., Videman, T. & Crites Battié, M. (1997b) Determinants of isokinetic and psychophysical
lifting strength and static back muscle endurance: A study of male monozygotic twins. Spine, 23,
2412–21.
Gogia, P., Schneider, V. S., Leblanc, A. D., Krebs, J., Kasson, C. & Pientok, C. (1988) Bed rest effect on
extremity muscle torque in healthy men. Arch Phys Med Rehabil, 69, 1030–2.
Greenleaf, J. (2004) Deconditioning and Reconditioning. Boca Baton, FL: CRC Press LLC.
Greenleaf, J. E. (1997) Intensive exercise training during bed rest attenuates deconditioning. Med Sci Sports
Exerc, 29, 207–15.
Haines, R. F. (1974) Effect of bed rest and exercise on body balance. J Appl Physiol, 36, 323–7.
Hamburg, N. M., McMackin, C. J., Huang, A. L., et al (2007) Physical inactivity rapidly induces insulin
resistance and microvascular dysfunction in healthy volunteers. Arterioscler Thromb Vasc Biol,
27, 2650–6.
Hasenbring, M. & Verbunt, J. (2010). Fear-avoidance and endurance-related responses to pain: new
models of behaviour and their consequences for clinical practice. Clin J Pain, 26(9), 747–53.
Hasenbring, M., Marienfeld, G., Kuhlendahl, D. & Soyka, D. (1994) Risk factors of chronicity in lumbar
outcome. Spine, 19, 2759–65.
Healey, E. L., Fowler, N. E., Burden, A. M. & Mcewan, I. M. (2005) The influence of different unloading
positions upon stature recovery and paraspinal muscle activity. Clin Biomech, 20, 365–71.
Hermansen, L. & Saltin, B. (1969) Oxygen uptake during maximal treadmill and bicycle exercise. J Appl
Physiol, 26, 31–7.
Hermansen, L., Ekblom, B. & Saltin, B. (1970) Cardiac output during submaximal and maximal treadmill
and bicycle exercise. J Appl Physiol, 29, 82–6.
Hides, J. A., Belavy, D. L., Stanton, W., et al. (2007) Magnetic resonance imaging assessment of trunk
muscles during prolonged bed rest. Spine, 32, 1687–92.
Hodges, P. W. (2001) Changes in motor planning of feedforward postural responses of the trunk muscles
in low back pain. Exp Brain Res, 141, 261–6.
Hodges, P. W., Moseley, G. L., Gabrielsson, A. & Gandevia, S. C. (2003) Experimental muscle pain changes
feedforward postural responses of the trunk muscles. Exp Brain Res, 151, 262–71.
Howley, E. T., Bassett, D. R., JR. & Welch, H. G. (1995) Criteria for maximal oxygen uptake: review and
commentary. Med Sci Sports Exerc, 27, 1292–301.
Jamison, R. N., Stetson, B., Sbrocco, T. & Parris, W. C. (1990) Effects of significant weight gain on chronic
pain patients. Clin J Pain, 6, 47–50.
Kader, D. F., Wardlaw, D. & Smith, F. W. (2000) Correlation between the MRI changes in the lumbar
multifidus muscles and leg pain. Clin Radiol, 55, 145–9.
Keller, A., Johansen, J. G., Hellesnes, J. & Brox, J. I. (1999) Predictors of isokinetic back muscle strength in
patients with low back pain. Spine, 24, 275–80.
Klenerman, L., Slade, P. D., Stanley, I. M., et al. (1995) The prediction of chronicity in patients with an
acute attack of low back pain in a general practice setting. Spine, 20, 478–84.
Kramer, M., Ebert, V., Kinzl, L., Dehner, C., Elbel, M. & Hartwig, E. (2005) Surface electromyography of
the paravertebral muscles in patients with chronic low back pain. Arch Phys Med Rehabil, 86, 31–6.
Kravitz, E., Moore, M. E. & Glaros, A. (1981) Paralumbar muscle activity in chronic low back pain. Arch
Phys Med Rehabil, 62, 172–6.
Lariviere, C., Arsenault, B., Gravel, D., Gagnon, D., Loisel, P. & Vadeboncoeur, R. (2002)
Electromyographic assessment of back muscle weakness and muscle composition: reliability and
validity issues. Arch Phys Med Rehabil, 83, 1206–14.
Leger, L. A. & Lambert, J. (1982) A maximal multistage 20-m shuttle run test to predict VO2 max. Eur
J Appl Physiol Occup Physiol, 49, 1–12.
Leinonen, V., Maatta, S., Taimela, S., et al. (2002) Impaired lumbar movement perception in association
with postural stability and motor- and somatosensory-evoked potentials in lumbar spinal stenosis.
Spine, 27, 975–83.
Lucia, A., Rabadan, M., Hoyos, J., et al. (2006) Frequency of the VO2max plateau phenomenon in
world-class cyclists. Int J Sports Med, 27, 984–92.
Macedo, L. G., Maher, C. G., Latimer, J. & McAuley, J. H. (2009) Motor control exercise for persistent,
nonspecific low back pain: a systematic review. Phys Ther, 89(1), 9–25.
Mannion, A. F., Kaser, L., Weber, E., Rhyner, A., Dvorak, J. & Muntener, M. (2000) Influence of age and
duration of symptoms on fibre type distribution and size of the back muscles in chronic low back pain
patients. Eur Spine J, 9, 273–81.
Mattila, M., Hurme, M., Alaranta, H., et al (1986) The multifidus muscle in patients with lumbar disc
herniation. A histochemical and morphometric analysis of intraoperative biopsies. Spine, 11, 732–8.
Mayer, T. G., Gatchel, R. J., Kishino, N., et al. (1985) Objective assessment of spine function following
industrial injury. A prospective study with comparison group and one-year follow-up. Spine, 10, 482–93.
Mayer, T. G., Gatchel, R. J., Kishino, N., et al. (1986) A prospective short-term study of chronic low back
pain patients utilizing novel objective functional measurement. Pain, 25, 53–68.
Mayer, T. G., Vanharanta, H., Gatchel, R. J., et al. (1989) Comparison of CT scan muscle measurements
and isokinetic trunk strength in postoperative patients. Spine, 14, 33–6.
McArdle, W., Katch, F.I. & Katch, V.L. (2001) Exercise Physiology: energy, nutrition, and human
performance. Baltimore, MD: Lippincott Wiliams & Wilkins.
Mok, N. W., Brauer, S. G. & Hodges, P. W. (2007) Failure to use movement in postural strategies leads to
increased spinal displacement in low back pain. Spine, 32, E537–43.
Mujika, I. & Padilla, S. (2001) Cardiorespiratory and metabolic characteristics of detraining in humans.
Med Sci Sports Exerc, 33, 413–21.
Neblett, R., Mayer, T. G., Gatchel, R. J., Keeley, J., Proctor, T. & Anagnostis, C. (2003) Quantifying the
lumbar flexion-relaxation phenomenon: theory, normative data, and clinical applications. Spine,
28, 1435–46.
Nielens, H. & Plaghki, L. (2001) Cardiorespiratory fitness, physical activity level, and chronic pain: are men
more affected than women? Clin J Pain, 17, 129–37.
Noonan, V. & Dean, E. (2000) Submaximal exercise testing: clinical application and interpretation. Phys
Ther, 80, 782–807.
O’Sullivan, P. B., Grahamslaw, K. M., Kendell, M., Lapenskie, S. C., Moller, N. E. & Richards, K. V. (2002)
The effect of different standing and sitting postures on trunk muscle activity in a pain-free population.
Spine, 27, 1238–44.
O’Sullivan, P. B., Phyty, G. D., Twomey, L. T. & Allison, G. T. (1997) Evaluation of specific stabilizing
exercise in the treatment of chronic low back pain with radiologic diagnosis of spondylolysis or
spondylolisthesis. Spine, 22, 2959–67.
Oddsson, L. I. & De Luca, C. J. (2003) Activation imbalances in lumbar spine muscles in the presence of
chronic low back pain. J Appl Physiol, 94, 1410–20.
Oddsson, L. I., Giphart, J. E., Buijs, R. J., Roy, S. H., Taylor, H. P. & De Luca, C. J. (1997) Development of
new protocols and analysis procedures for the assessment of LBP by surface EMG techniques. J Rehabil
Res Dev, 34, 415–26.
Oja, P., Laukkanen, R., Pasanen, M., Tyry, T. & Vuori, I. (1991) A 2-km walking test for assessing the
cardiorespiratory fitness of healthy adults. Int J Sports Med, 12, 356–62.
Oldridge, N. B. (2008) Economic burden of physical inactivity: healthcare costs associated with
cardiovascular disease. Eur J Cardiovasc Prev Rehabil, 15, 130–9.
Olsen, R. H., Krogh-Madsen, R., Thomsen, C., Booth, F. W. & Pedersen, B. K. (2008) Metabolic responses
to reduced daily steps in healthy nonexercising men. JAMA, 299, 1261–3.
Panjabi, M. M. (1992a) The stabilizing system of the spine. Part I. Function, dysfunction, adaptation, and
enhancement. J Spinal Disord, 5, 383–9.
Panjabi, M. M. (1992b) The stabilizing system of the spine. Part II. Neutral zone and instability hypothesis.
J Spinal Disord, 5, 390–6.
Parkkola, R., Rytokoski, U. & Kormano, M. (1993) Magnetic resonance imaging of the discs and trunk
muscles in patients with chronic low back pain and healthy control subjects. Spine, 18, 830–6.
Pavy-Le Traon, A., Heer, M., Narici, M. V., Rittweger, J. & Vernikos, J. (2007) From space to Earth: advances
in human physiology from 20 years of bed rest studies (1986–2006). Eur J Appl Physiol, 101, 143–94.
Philippaerts, R. M., Westerterp, K. R. & Lefevre, J. (2001) Comparison of two questionnaires with a
tri-axial accelerometer to assess physical activity patterns. Int J Sports Med, 22, 34–9.
Picavet, H. S. & Schuit, A. J. (2003) Physical inactivity: a risk factor for low back pain in the general
population? J Epidemiol Community Health, 57, 517–8.
Protas, E. J. (1999) Physical activity and low back pain. In Mitchell, M. (Ed.) An updated review; refresher
course syllabus 9th World Congress on pain. Seattle, WA: IASP Press.
Rantanen, J., Hurme, M., Falck, B., et al. (1993) The lumbar multifidus muscle five years after surgery for a
lumbar intervertebral disc herniation. Spine, 18, 568–74.
Rasmussen-Barr, E., Lundqvist, L., Nilsson-Wikmar, L. & Ljungquist, T. (2008) Aerobic fitness in patients
at work despite recurrent low back pain: a cross-sectional study with healthy age- and gender-matched
controls. J Rehabil Med, 40, 359–65.
Rowland, T. W. (1993) Does peak VO2 reflect VO2max in children?: evidence from supramaximal testing.
Med Sci Sports Exerc, 25, 689–93.
Sale, D. G., Mccomas, A. J., Macdougall, J. D. & Upton, A. R. (1982) Neuromuscular adaptation in human
thenar muscles following strength training and immobilization. J Appl Physiol, 53, 419–24.
Saltin, B., Blomqvist, G., Mitchell, J. H., Johnson, R. L., JR., Wildenthal, K. & Chapman, C. B. (1968)
Response to exercise after bed rest and after training. Circulation, 38, VII1–78.
Shephard, R. J., Allen, C., Benade, A. J., et al. (1968) The maximum oxygen uptake. An international
reference standard of cardiorespiratory fitness. Bulletin of the World Health Organization, 38, 757–64.
Siconolfi, S. F., Garber, C. E., Lasater, T. M. & Carleton, R. A. (1985) A simple, valid step test for
estimating maximal oxygen uptake in epidemiologic studies. Am J Epidemiol, 121, 382–90.
Skinner, J. & Oja, P. (1994) Laboratory and field tests for assessing health-related fitness. Physical activity,
fitness and health. Champaign, IL: Human Kinetics.
Smeets, R. J. & Wittink, H. (2007) The deconditioning paradigm for chronic low back pain unmasked?
Pain, 130, 201–2.
Smeets, R. J., Wittink, H., Hidding, A. & Knottnerus, J. A. (2006) Do patients with chronic low back pain
have a lower level of aerobic fitness than healthy controls?: are pain, disability, fear of injury, working
status, or level of leisure time activity associated with the difference in aerobic fitness level? Spine,
31, 90–7; discussion 98.
Smeets, R. J., Van Geel, A. C., Kester, A. D. & Knottnerus, J. A. (2007) Physical capacity tasks in chronic
low back pain: what is the contributing role of cardiovascular capacity, pain and psychological factors?
Disabil Rehabil, 29, 577–86.
Smeets, R. J., Geel, K. V. & Verbunt, J. A. (2009) Is the fear avoidance model associated with the reduced
level of aerobic fitness in patients with chronic low back pain? Arch Phys Med Rehabil, 90, 109–17.
Toda, Y., Segal, N., Toda, T., Morimoto, T. & Ogawa, R. (2000) Lean body mass and body fat distribution
in participants with chronic low back pain. Arch Intern Med, 160, 3265–9.
Van Den Berg-Emons, R. J., Schasfoort, F. C., De Vos, L. A., Bussmann, J. B. & Stam, H. J. (2007) Impact
of chronic pain on everyday physical activity. Eur J Pain, 11, 587–93.
Van Dieen, J. H., Cholewicki, J. & Radebold, A. (2003a) Trunk muscle recruitment patterns in patients
with low back pain enhance the stability of the lumbar spine. Spine, 28, 834–41.
Van Dieen, J. H., Selen, L. P. & Cholewicki, J. (2003b) Trunk muscle activation in low-back pain patients,
an analysis of the literature. J Electromyogr Kinesiol, 13, 333–51.
Van Houdenhove, B., Onghena, P., Neerinckx, E. & Hellin, J. (1995) Does high ‘action-proneness’ make
people more vulnerable to chronic fatigue syndrome? A controlled psychometric study. J Psychosom
Res, 39, 633–40.
Van Weering, M., Vollenbroek-Hutten, M. M., Kotte, E. M. & Hermens, H. J. (2007) Daily physical
activities of patients with chronic pain or fatigue versus asymptomatic controls. A systematic review.
Clin Rehabil, 21, 1007–23.
Van Weering, M.G., Vollenbroek-Hutten, M.M., Tönis, T.M., & Hermens, H.J. (2009) Daily physical
activities in chronic lower back pain patients assessed with accelerometry. Eur J Pain, 13(6): 649–54.
Verbunt, J. A., Westerterp, K. R., Van Der Heijden, G. J., Seelen, H. A., Vlaeyen, J. W. & Knottnerus, J. A.
(2001) Physical activity in daily life in patients with chronic low back pain. Arch Phys Med Rehabil,
82, 726–30.
Verbunt, J. A., Seelen, H. A., Vlaeyen, J. W., et al (2003) Disuse and deconditioning in chronic low back
pain: concepts and hypotheses on contributing mechanisms. Eur J Pain, 7, 9–21.
Verbunt, J. A., Seelen, H. A., Vlaeyen, J. W., et al (2005a) Pain-related factors contributing to muscle
inhibition in patients with chronic low back pain: an experimental investigation based on
superimposed electrical stimulation. Clin J Pain, 21, 232–40.
Verbunt, J. A., Sieben, J. M., Seelen, H. A., et al. (2005b) Decline in physical activity, disability and
pain-related fear in sub-acute low back pain. Eur J Pain, 9, 417–25.
Verbunt, J. A., Huijnen, I. P. & Koke, A. (2008) Assessment of physical activity in daily life in patients with
musculoskeletal pain. Eur J Pain, 13, 231–42.
Vlaeyen, J. W. & Linton, S. J. (2000) Fear-avoidance and its consequences in chronic musculoskeletal pain:
Vlaeyen, J. W. & Morley, S. (2004) Active despite pain: the putative role of stop-rules and current mood.
Pain, 110, 512–6.
Vlaeyen, J. W., Kole-Snijders, A. M., Boeren, R. G. & Van Eek, H. (1995) Fear of movement/(re)injury in
chronic low back pain and its relation to behavioral performance. Pain, 62, 363–72.
Wasserman, K., Hansen, J. E., Sue, D. Y., Casaburi, R. & Whipp, B. J. (1999) Principles of Exercise Testing
and Interpretation. Baltimore, MD: Williams and Wilkins.
Watson, P. J. (1999) Non-psychological determinants of physical performance in musculoskeletal pain. In
Mitchell, M. (Ed.) Pain 1999- an updated review: Refresher course syllabus 9th World Congress on Pain,
pp. 153–8. Seattle, WA: IASP Press.
Watson, P. J., Booker, C. K., Main, C. J. & Chen, A. C. (1997) Surface electromyography in the
identification of chronic low back pain patients: the development of the flexion relaxation ratio. Clin
Biomech, 12, 165–171.
Weber, B. R., Grob, D., Dvorak, J. & Muntener, M. (1997) Posterior surgical approach to the lumbar spine
and its effect on the multifidus muscle. Spine, 22, 1765–72.
Westerterp, K. R. (1999) Assessment of physical activity level in relation to obesity: current evidence and
research issues. Med Sci Sports Exerc, 31, S522–5.
Wittink, H., Hoskins Michel, T., Wagner, A., Sukiennik, A. & Rogers, W. (2000a) Deconditioning in
patients with chronic low back pain: fact or fiction? Spine, 25, 2221–8.
Wittink, H., Michel, T. H., Kulich, R., Wagner, A., Sukiennik, A., Maciewicz, R. & Rogers, W. (2000b)
Aerobic fitness testing in patients with chronic low back pain: which test is best? Spine, 25, 1704–10.
Wittink, H., Nicholas, M., Kralik, D. & Verbunt, J. (2008) Are we measuring what we need to measure?
Clin J Pain, 24, 316–24.
Wong, K., Lee, R., Yeung, S. (2009) The association between back pain and trunk posture of workers in a
special school for the severe handicaps. BMC Musculoskeletal Dis, 10, 43.
Zerwekh, J. E., Ruml, L. A., Gottschalk, F. & Pak, C. Y. (1998) The effects of twelve weeks of bed rest on
bone histology, biochemical markers of bone turnover, and calcium homeostasis in eleven normal
subjects. J Bone Miner Res, 13, 1594–601.
Zhu, X. Z., Parnianpour, M., Nordin, M. & Kahanovitz, N. (1989) Histochemistry and morphology of
erector spinae muscle in lumbar disc herniation. Spine, 14, 391–7.
Part 4

Sociodemographic and
Psychosocial Mechanisms
Chapter 11
Screening of Psychosocial Risk Factors

(Yellow Flags) for Chronic Back Pain
and Disability
Chris J. Main, Nicholas A.S. Kendall, and
11.1 Introduction
There is a wide range of pain conditions in which there is evidence of significant dysfunction but
little or no evidence of disease or nerve damage. The impact of non-specific pain conditions on
suffering is considerable, as is the cost in terms of healthcare provision and work compromise.
When a clearly identifiable ‘pain generator’ is not apparent, as is often the case, then promise of a
complete cure seems improbable. What then can be done? Two major strategies have emerged:
◆ Prevention of chronic pain becoming unnecessarily disabling.
◆ Prevention of the development of chronic pain/disability in patients with acute/subacute
pain.
Health and safety legislation has improved the safety of the workplace but population surveys and
consultation rates still indicate a high prevalence of back pain in the community. Differences in
methodology have made it difficult to compare directly the results from different studies, e.g. dif-
fering criteria for the identification of ‘new episodes’ lead to different estimates. Waddell (2004)
in a further analysis of the data from the South Manchester population study (Thomas et al.
1999), found that 38% had had back pain in the previous year and of those who had been pain-
free, 19% reported new episodes in the following year. Furthermore, of the 32% of those who had
had intermittent or less disabling pain in the previous year, almost half would have further epi-
sodes during the following year and of the 6% with longstanding or seriously disabling back pain,
one-third would improve to some extent. There is thus a significant minority of people experi-
encing pain who go on to develop persistent pain, and may develop long-lasting disability, in
terms of reduced participation in usual activity and work, and the concomitant psychological
impact on their mood, self-confidence, and personal identity (Pincus and Morley 2001). The goal
of completely preventing the onset of all musculoskeletal pain is simply not achievable and
attempts to clinically alleviate pain, or reduce its severity are often less effective than hoped.
However, reducing the impact of pain by preventing it becoming unnecessary disabling and
enhancing the back sufferer’s quality of life, do appear to be worthwhile endeavours.
There is some evidence for the effectiveness of intervention strategies for people with chronic
musculoskeletal pain problems, and the potential for secondary prevention, has now been recog-
nized for more than a decade (Linton and Anderson 2000). In fact clear guidelines have been
developed for the management of acute non-specific low back pain (LBP; Box 11.1).
Box 11.1 Recommendations for treatment of acute

non-specific low-back pain
◆ Give adequate information and reassure the patient.
◆ Do not describe bed rest as a treatment.
◆ Advise patients to stay active and continue normal daily activities including work if possible.
◆ Prescribe medication, if necessary for pain relief; preferably to be taken at regular intervals;
first choice paracetamol, second choice NSAIDs.
◆ Consider adding a short course of muscle relaxants on its own or added to NSAIDs, if para-
cetamol or NSAIDs have failed to reduce pain.
◆ Consider (referral for) spinal manipulation for patients who are failing to return to normal
activities.
◆ Multidisciplinary treatment programmes in occupational settings may be an option for
workers with subacute low back pain and sick leave for more than 4–8 weeks.
Adapted from Van Tulder et al. (2005).
Although psychosocial factors seem to be stronger predictors of outcome than biomedical/

biomechanical factors (Burton et al. 1999; Crombez et al. 1999), and several randomized control-
led trials (RCTs) in early intervention (Hay et al. 2005; Jellema et al. 2005) have demonstrated the
feasibility of early intervention tackling these risk factors, there is not clear evidence of superiority
of clinical outcomes for biopsychosocial approaches over usual care, either because they don’t
work or because of methodological limitations in the design of the research. Van der Windt et al.
(2008) have identified a number of methodological limitations such as insufficient statistical
power or adoption of a ‘one size fits all approach’ to the intervention, the relative lack of thera-
peutic power in the treatments offered, or the inability to match the right treatment to the right
patient at the right time (see for more detail Chapter 27). A possible solution in terms of an
approach combining screening and targeting approach is described and discussed later in this
chapter.
Estimation of risk is at the core of screening but the literature contains a number of overlapping
and similar terms such as ‘risk factors’, ‘predictive factors’, and ‘prognostic factors’ that at times
are used interchangeably. As a precursor to considering the nature and efficacy of risk factor
identification and screening for improved outcome, we offer our understanding of these terms.
11.2 Terminology
11.2.1 Risk factors, predictive factors, and prognostic factors
Risk factors are usually taken to refer to features associated with the future development or occur-
rence of an event such as a disease of some sort. They may or may not be implicated causally in
the development of the disease, but the disease is not present at the time of risk estimation.
Further investigations may be able to demonstrate a direct causal relationship, but the relation-
ship may be indirect (possibly mediated by other factors) or, in so far as can be investigated, may
turn out to be a chance association.
Predictive factors refer to those that are associated statistically with some sort of future outcome.
Whether or not they are predictive therefore is a matter of statistical association, using whatever
SCREENING OF PSYCHOSOCIAL RISK FACTORS 205
criteria are appropriate. There is no assumption required about the relationship, or lack of,
between the two sets of events. In practice, they can be divided into risk factors and prognostic
factors.
Prognostic factors refer to factors predictive of outcome of a disease or condition already in
progress. In terms of potential interventions however, the distinction between moderators and
mediators is particularly important.
Moderators (or treatment effect modifiers) are baseline characteristics which influence the out-
come of treatment. Thus people with leg pain as well as back pain might improve less with a
particular type of therapy
Mediators are factors which change during or as a consequence of treatment and thereby influ-
ence outcome. Thus, for example, it might be hypothesized that increase in exercise tolerance in
physiotherapy might be mediated by reduction in fear of movement
However, two caveats are in order. Firstly, it should be noted that in understanding the
development of pain, chronic pain, and pain-associated disability, predictors may act as both
risk factors and prognostic factors. For example, psychological distress may be a risk factor
for the development of pain, a prognostic factor in terms of the development of chronic pain,
a risk factor for the development of disability, and a prognostic factor for further change in
disability. Similarly, pain severity may be both a risk factor for sickness absence and a prog-
nostic factor for recovery. Secondly, the distinction between modifiable and unmodifiable
predictive factors (whether risk factors or prognostic factors) although perhaps less impor-
tant from a conceptual point of view is fundamental to the design (and implementation) of
interventions.
11.2.2 Flags as a method of risk identification

The term ‘yellow flags’ has become a familiar term used to describe psychosocial risk factors for
chronicity. The original flag system (Kendall et al. 1997), and its later developments (Main et al.
2005; Kendall et al. 2009) is one such approach to risk identification. It has been described as a
methodological compromise between the inflexibilities of a purely actuarial model and a purely
subjective approach based on clinical judgement (Linton et al. 2005). In this context it has three
important features. Firstly, it offers a ‘systems perspective’ and assumes that an adequate under-
standing of the problem requires consideration of both the injured worker and the individual’s
social and occupational context. Secondly, it contains both clinical and occupational elements.
Thirdly, it makes an important distinction between the individual’s perception of the situation
and the objective features.
The flags approach can be viewed as a conceptual framework potentially capable of including
both actuarial data and individually-assessed risk factors informing different types of specifically
targeted interventions based on modifiable risk. An attempted mapping of the flags onto the
individual predictors of chronic pain and disability is illustrated in Table 11.1. Different flags, and
combinations of flags, require different types on interventions, and a multiflag approach appears
to lend itself flexibly to both individual clinical decision-making and widescale system applica-
tions (Main et al. 2005; Kendall et al. 2009). However, its predictive accuracy in administrative
applications will likely be lower than one arising from a purely statistical or ‘actuarial’ approach
and will likely result in over-identification of individuals at risk. Further reflections on the linking
of screening with targeting are offered below.
As far as individual risk factors for long-term work disability are concerned, Sullivan et al.
(2005) in their review found evidence for fear, beliefs in severity of health conditions, and
catastrophizing (yellow flags). Nicholas et al. (2011) having reviewed the evidence both for the
influence of yellow flags on outcomes in people with acute/subacute LBP, and of yellow flag
targeting on outcomes, concluded:
Overall, from the evidence gathered here the studies that target interventions on known psychological
risk factors for disability do seem to be reporting more consistently positive results relative to those
interventions that either ignore these risk factors or provide omnibus interventions to people regard-
less of psychological risk factors. It seems that the identification of those with these risk factors is an
important precursor to psychological interventions.
Predictors of occupational outcomes have also been investigated. Shaw et al. (2009) in addition
to pain severity and level of depressive symptoms also identified workplace factors such as job
stress, co-worker support, job dissatisfaction, employer attitudes, job autonomy, and availability
of modified work as influences on duration of work disability and return-to-work (RTW) out-
comes. Their findings are consistent with an earlier more widespread review of predictors of
chronic pain and disability (Waddell et al, 2003). It is sometimes difficult, however, to distinguish
clinical outcomes such as increase in activity or postural tolerance from occupational variables
such as RTW rates or indices of work capability.
In fact, the COST B13 prevention guidelines (Burton et al. 2004a) offered a distinction between
the general population and workers. Despite some variation in the level of evidence available, the
similarity among the recommendations is striking.
The guidelines for workers obviously contain a number of recommendations specific to work
settings, but otherwise, in terms of the usefulness of information or educational approaches, the
lack of support for traditional biomechanical and biomedical approaches, and for traditional
clinical interventions is similar. In fact, over the last decade, with the exception of a series of
reviews in the Cochrane Library (http://www2.cochrane.org/reviews/) synthesizing the findings
of earlier conservative management, and studies of outcome of surgery (which are beyond
the remit of this chapter), there appears to have been relatively little new research with clinical
outcomes as a primary focus.
Waddell et al. (2003) appraised the evidence for different sorts of predictors of chronic pain
and disability. In their first set of evidence tables they summarize the findings from published
studies of clinical and psychosocial predictors (Table 11.1) where the strength of evidence
and strength of predictors are shown along with a tentative flag assignment in the right-hand
column.
Aggregation of such a large set of data, with such a wide variety of specific variables is by nature
imprecise, as are the 27 variables under which the predictors have been gathered. Accepting these
strictures, it can be seen that there is evidence for an influence of both sociodemographic and
personal history variables, but the strongest influences appear to be from yellow or blue flags
(most of which are potentially modifiable) and black flags (which are more immutable, although
may present opportunities for a ‘systems solution’; equivalent to Sullivan et al.’s (2005) Type-I
(individually centred) and Type-II (workplace or system-based) solutions.
11.3 The nature and focus of risk identification

The identification of predictors of outcome may enable us to select subgroups, whether in terms
of possible benefit from treatment (‘screening in’) or failure to benefit from treatment (‘screening
out’) (Foster et al. 2011). Furthermore, identification of predictors may lead to the identification
of new and different approaches to treatment for patients with differing presenting characteris-
tics, as in the screening/targeting approach developed by Hay and colleagues (Hay et al. 2008), or
at different stages of pain chronicity.
Table 11.1 Individual clinical and psychosocial predictors of chronic pain and disability
Strength of Strength of Type of Flag assignment

evidence predictor variable
Age *** *** Demog.
Gender * Variable Demog.
Ethnicity ** Not signif.
Marital status * Variable Demog.
Education * * Demog.
Clin. history *** (LBP) *** Clinical ?Red ?Yellow
Clin. exam * * Clinical ?Red
Comorbidity *** * Clinical ?Red
Alc./sub. abuse * * Clinical Orange
Personality * * Orange
Psychol. hist. * * ?Orange
Anxiety * * ?Orange
Stressful life events * * ?Orange
Pain intensity, functional disability *** ** Clinical ?
Poor perceptions of general health *** ** Clinical Yellow
Psychological distress *** *** Clinical Yellow/orange
Depression *** ** Clinical Orange/yellow
Fear avoidance ** ** Clinical Yellow
Maladaptive coping (catastrophizing) *** ** Clinical Yellow
Pain behaviour *** ** Clinical Yellow
Duration sickness absence *** *** Clin./Occ. Yellow/blue ?Black
Employment status *** *** Occ. Blue/black
Job dissatisfaction *** *** Occ. Blue
Expectations re: RTW *** *** Blue
Physical demands of work *** * Occ. Black
Financial incentives *** *** Occ. Black
Unemployment rates ** *** Occ. Black
*** Strong; ** Moderate; * Weak.
Understanding the natural history of chronic low

11.3.1
back pain (CLBP)
For many purposes it is convenient to assume that either acute pain resolves, or it does not, and
in the latter event there is a smooth and clearly identifiable trajectory into chronic pain. This
assumption, however, would appear to be unsafe. Burton et al. (2004b), in a prospective cohort
study of 252 LBP patients attending for manipulative care, were able to follow up 60% of the
sample 4 years later. Of those attending initially with acute LBP, 60.9% had further episodes, with
50% identifying between one and five episodes over the ensuing 4 years, and 10.9% identifying
more than six episodes of persistent pain. The comparable figures for the subacute group were
70.6%, 32.3%, and 38.2%. For the chronic group, the comparable figures are 88.9%, 33.3%, and
55.6%. Admittedly the actual numbers in this study are relatively small, but they suggest firstly
that most back pain is recurrent and secondly that the future course is influenced by the clinical
history at the time of initial consultation.
Conventionally, the development of chronicity is viewed along a continuous timeline, with
pragmatic cut-off points devised primarily as a way of deciding about resource allocation. As can
be inferred from some of the previous discussion, however, there is significant variation in the
clinical course of back pain. Von Korff and Miglioretti (2006, see also Chapter 2 in this volume)
have suggested a new approach to the definition of pain, by classifying it in terms of the probabil-
ity of significant pain being present 1 or more years in the future; making a further distinction
between possible chronic back pain (with a likelihood of occurrence of 50%) and probable
chronic back pain (with a likelihood of occurrence of 80%). They based their estimation on three
variables. Using a baseline score, with cut-offs as a basis for risk estimation, they found that 58.7%
of possible and 82.1% of probable chronic back pain patients had a Chronic Pain Grade of 2–4 1
year later, and the classification also strongly predicted the presence of clinically significant back
pain at years 2 and 5. Furthermore, of those with probable chronic back pain at year 1, 90.9% and
76.5% had clinically significant back pain at years 2 and 5 respectively. This prognostic approach
would seem to merit further investigation since potentially it offers an opportunity for screening
and targeting.
11.3.2 Defining the outcome

In the development of CLBP, the most common outcome variables of interest are pain intensity,
pain-related disability, and pain-related work compromise. Unfortunately in many studies only a
single outcome measure is employed and it is difficult therefore to differentiate between persist-
ence of pain and pain-associated disability. Furthermore, since arguably one of the most appeal-
ing, and to some extent effective, ways of controlling pain is to diminish activities, the variables
are not independent; and there are some patients who suffer from intense pain yet do not mani-
fest significant levels of disability (see Chapter 15). It might be argued further that while an exclu-
sive focus on pain, although consistent with a specific biomedical focus on pain nociception, may
be important particularly in the management of neoplastic and neuropathic pain, it is more
problematic in the management of ‘non-specific’ musculoskeletal pain, since the ‘pain generator’
cannot be clearly identified and targeted. Most evidence for the efficacy of treatment for acute
pain problems lies in pharmacological trials, but there have recently been a number of trials pub-
lished on the Cochrane website (http://www.cochrane.org) of a range of other sorts of treatment.
The need for a broader view was recognized by the European Commission which, under the
COST B13 funding initiative, established three working groups charged with developing LBP
guidelines for acute pain, chronic pain and prevention, respectively. The acute non-specific LBP
guidelines (van Tulder et al. 2005) described in Box 11.1, are targeted primarily on the manage-
ment of pain, although there is recognition that with the passage of time a broader approach
might be considered. A parallel working group (Burton et al. 2004a) considered prevention. It
has extensive specific evidence-based recommendations which can be found on the website but
it also offers a summary of the concepts of prevention accompanied by some overarching
comments (see Box 11.2).
Risk identification in population studies

The linking of harm with exposures is at the heart of population and clinical epidemiology. By their
nature such investigations are descriptive rather than explanatory, and while statistically significant
associations may serve as a foundation for major clinical initiatives (such as immunization) or
Box 11.2 Concepts of prevention and overarching

comments (extracts)
◆ There is limited scope for preventing its incidence (first-time onset).
◆ Prevention in the context of this guideline is focused primarily on reduction of the impact
and consequences of LBP.
◆ There is considerable scope, in principle, for prevention of the consequences of LBP.
◆ Different interventions and outcomes will be appropriate for different populations.
◆ The most promising approaches seem to involve physical activity/exercise and appropriate
biopsychosocial education, at least for adults.
◆ But, no single intervention is likely to be effective to prevent the overall problem of LBP,
owing to its multidimensional nature.
◆ Optimal progress on prevention in LBP will likely require a cultural shift in the way LBP is
viewed, its relation with activity and work, how it may best be tackled and just what is
reasonable to expect from preventative strategies.
◆ It is important to get all the players onside, but innovative studies are required to under-
stand better the mechanisms and delivery of prevention in LBP.
Reprinted from Manual Therapy, 9(1), A. Kim Burton, Timothy D. McClune, Robert D. Clarke, and
Chris J. Main, Long-term follow-up of patients with low back pain attending for manipulative care:
outcomes and predictors, pp. 30–5, Copyright (2004), with permission from Elsevier.
social policy decisions involving the redirection of resources, such risk factors are usually not
sufficiently powerful to inform decision-making on an individual basis.
Epidemiological studies can inform us of the prevalence of pain and other symptoms among
the general (non-clinical) population, using a variety of prevalence estimates. It is possible to
determine the incidence of new pain episodes, and longitudinal studies begin to inform us about
the various (risk) factors that are associated with persistent pain. We can construct a map of the
context of pain and its future development, enabling identification of predictors of chronicity as
well as recovery at a population level. In this manner, demographic and ethnicity variables may
lead to public health initiatives focused on particular subgroups or delivered in a variety of ways.
At a population level even relatively weak levels of prediction may be of considerable importance
in terms of informing public health policy, in terms of prioritizing initiatives or in optimizing use
of resources.
However, population predictors are usually unhelpful in the management of individual patients
since: (1) population predictors are insufficiently robust in terms of sensitivity and specificity
for use in the individual patient; and (2) individual management needs to tackle modifiable risk
factors, and many population factors are difficult to modify or unmodifiable.
Risk identification in clinical epidemiology

Clinical studies are often more narrowly focused, typically on healthcare outcomes, than epide-
miological investigations, and therefore potentially provide a better basis for clinical intervention
in suggesting particular therapeutic targets or in assisting the design of preventative intervention.
However, there are important issues of sensitivity and specificity that need to be evaluated prior
to use in clinical decision-making for individuals. These are discussed below.
Epidemiological data from population studies have provided useful sign-posting for clinical
epidemiology, and studies of pain patients in which much stronger predictors of the natural history
and treatment outcome have been identified. There have been many studies of treatment outcome
for specific neuropathic, neuroplastic and non-cancer pain conditions. Given the importance of
abolishing pain, wherever possible, there has been a major emphasis on the biomedical parameters
associated with outcome of acute pain conditions. Unfortunately, initial biomedical parameters
seem to be relatively weak predictors in contrast with the relative strength of psychosocial predic-
tors. The potential for screening for these psychosocial factors appears to merit attention.
Conceptualizing risk identification for individuals

We believe we have made a strong case in principle for secondary prevention, but is there any way
in which we can capitalize on what we have learned from population and clinical epidemiology?
Can we link our knowledge of populations, and understanding of sub-populations (such as
consulters), in an effort to improve our management of the individual?
There are two further important considerations. Firstly, in considering management of an
individual patient in a clinical setting it is clear not only that there is a wide range of potentially
relevant risk factors to consider, but furthermore, these may differ according to the outcome
of interest. Secondly, not all risk factors are modifiable and so it would seem appropriate, as in
the flags initiative, to narrow the focus to risk factors which are potential obstacles to recovery
(Shaw et al. 2009; Nicholas et al. 2011) as a way of identifying opportunities for change which
could be targeted.
11.4 From risk identification to screening

11.4.1 Principles of screening
The concept of risk factors has entered common parlance, although there is frequent confusion
between the concepts of absolute and relative risk, with incorrect usage in media reports some-
times leading to unnecessary levels of concern and anxiety. Screening was developed initially in
the context of the development of screening programmes for particular diseases and Muir Gray
(2004) emphasizes particularly the need for careful evaluation of both the benefits and harm of
screening as well as quality assurance, in an economically constrained healthcare system.
A number of approaches to screening, however, have been developed on a variety of clinical
populations and it is important to appraise their utility as a part of overall clinical management.
Prior to examining their utility specifically for the identification of risk and prognostic factors for
pain and pain-associated limitations (disability), the nature of screening will be addressed.
The WHO criteria are shown in Box 11.3.
11.4.2 Purposes of screening

The importance of clarifying the purpose of screening, emphasized by Waddell et al. (2003, p. 5)
in consideration of screening for long-term incapacity, identified a number of different purposes
for screening.
Some examples
◆ Identifying people at higher risk of:
● Long-term pain.
● Long-term self-reported disability.
● Long-term sickness absence and early retirement.
Box 11.3 WHO screening criteria (adapted from Wilson and

Junger 1968)
1 Important health problem
2 Accepted treatment for recognized disease.
3 Facilities for diagnosis and treatment.
4 Suitable latent and symptomatic stage.
5 Suitable test or examination.
6 Test acceptable to population.
7 Natural history of condition understood.
8 Agreed on policy on whom to treat.
9 Cost of finding economically balanced with overall health.
10 Case finding should be continuous process.
◆ Identifying people who need:

● Extra educational advice.
● Extra therapeutic or rehabilitation help.
◆ Informing:
● A rehabilitation programme or other work-focused intervention.
● The decision-making processes and case management.
In the case of psychosocial risk factors indicating a high probability of the development of
chronic pain and disability, the offer of an individually targeted intervention focusing on a mod-
ification of these psychosocial factors as early as possible but also as cost-effective as possible is
one of the primary challenges of a yellow-flag screening programme. Turk (1990) has long been
an advocate of customizing patients for treatment on the basis of their initial presenting charac-
teristics. Screening has to be understood in context, not only in terms of the specific purposes of
screening and the stage in illness, but also in terms of the base rates of the characteristic of interest
in the particular group being screened. Most screening, however, can be described in terms of
‘screening in’, ‘screening out’, and ‘screening with targeting’, and the sensitivities/specificities
need to be examined in relationship to the specific purpose of screening. However there is an
important caveat which should be borne in mind: screening out from one treatment does not
necessary imply screening in for another sort of treatment.
Since all screening measures are in effect proxies (i.e. short-cuts) for a full appraisal, the value
of a screening procedure is dependent not only on its accuracy (in terms of the predictive value),
but its cost, and of course increased accuracy usually comes at increased cost, but the worth of
screening is also related to its importance. In clinical practice, screening for a potentially fatal
disease such as cancer, lack of specificity (and therefore increased cost) may be considered a price
worth paying for a higher degree of accuracy in identifying those at risk. The consequences of risk
identification may be costly in terms of resources so decisions are made in terms of the probable
additional benefit likely to accrue from various degrees of effort in improving overall accuracy. In
back pain research, appraisal of cost of treatment has become an important consideration in
healthcare funding and metrics such as the QALY (quality-adjusted life year) are now frequently
employed as a measure of cost-effectiveness (Ratcliffe et al. 2006) In extreme cases, such as

screening for potential terrorists, it may be decided politically that since the cost of failure is so
catastrophic, that almost limitless resources are assigned to identify extremely rare events, but
usually cost is a factor. For example, in screening for clinical depression it would seem reasonable
to consider whether the increased cost of a clinical interview over a simple screening question-
naire, as might be used in a population survey, is justified by the increased accuracy in case iden-
tification. In considering the identification of individuals at risk of a particular clinical or
occupational outcome, it is indeed relevant therefore to make use of all available information and
to consider how much can be learned from population and clinical epidemiology.
11.4.3 Methods of screening

The value of screening depends not only on the content of the items, but the manner in which
the information is collected and the way the information is integrated. Different methods may be
required to elicit different sorts of information. Routinely available administrative data may be
useful at a population, insurance system or organizational level, but usually it has been designed
for different purposes such as audit or population risk estimates. More individualized and clini-
cally relevant information may be collected using questionnaire, telephone or face-to-face inter-
view. The additional potential value has to be offset against much greater cost and practical
difficulties in eliciting and interpreting the information obtained.
Accepting the reality of financially constrained healthcare, insurance, and social security systems,
a comprehensive evaluation on an individual basis, however desirable, is never likely to be funda-
ble, or even acceptable. It is likely that there will always be a need for some sort of filtering or screen-
ing as a precursor to intervention, but before considering screening per se, it is appropriate to
reflect upon the process of clinical decision-making, We offer the view that at the heart of the proc-
ess of clinical decision-making are probabilistic judgements and in this respect, arrival at a decision
involves risk estimation in the context of risk of good and poor outcome (and by implication,
whether to offer pain management, and what to offer). A number of core functions for a system of
screening have been identified. They comprise: (1) identification of items of information that pre-
dict the outcome of interest; (2) development of a practical and usable method of collecting the
information; (3) development of a method of combining the information or scoring the instru-
ment; and (4) measurement of the accuracy of prediction (adapted from Waddell et al. 2003, p. 7).
Actuarial versus clinical methods

Actuarial methods A central issue in screening is the accuracy of predicting future ill health or
disability based on various initial data sets. In clinical practice, treatment recommendations are
usually based on implicit outcome predictions made by a practitioner. An alternative is to adopt
a statistical approach based on population statistics, such that each individual is compared with a
wider group in terms of a number of characteristics and a risk profile thus obtained from data
often gathered for administrative rather than clinical purposes. This empirical approach has been
called actuarial and lends itself to wide administrative applications (Dawes et al. 1989). The exact
nature of the predictors is relatively unimportant, but their strength in terms of accuracy of pre-
diction is critical. There is no inherent ordering of variables in terms of the significance and many
different types of predictors may end up as part of the optimal algorithm, as the selection is based
on each variable’s contribution to the prediction of outcomes, such as duration of disability,
RTW, or costs.
Strengths and limitations of the actuarial model Although there are a number of strengths of
the actuarial model, such as accuracy of data, and maximization of use of the available data, with
the promise of accurate quantification of the relationship between risk factors and future status
and the potential for the construction of algorithms and prediction rules applicable to large pur-
poses, there are also a number of limitations. Linton et al. (2005) have identified some inherent
limitations of the actuarial model:
First, it assumes that the variables are stable and static. Second, there is no room for individual differ-
ences as the same statistical formula is always used, leading to misclassifications. Third, the actuarial
approach is problematic when critically important salient data are not collected. Fourth, the utility of
a strictly actuarial model is limited when the ceiling for predictive accuracy is relatively low and when
the underlying evidence is weak. In addition, the generalizability of such statistical prediction models
to other populations and contexts is often unknown. Lastly, since prediction is the driving factor, the
modifiability of the risk factors may be of secondary importance. Therefore, actuarial models may not
be easily translated into secondary prevention (Linton et al. 2005, p. 462).
In summary, actuarial models as a basis from which to develop secondary prevention applica-
tions are inherently limited.
Clinical methods: the role of clinical decision-making Giving the striking variability in clinical
presentation, an alternative and somewhat more intuitive strategy is to use incorporate clinical
information about the condition or the particular individual into the clinical decision-making
process. At the heart of clinical decision-making is a judgement matching the needs of the patient
with the possible benefits from our intervention and known prognostic factors for a particular
condition seem obvious contenders as a starting point from which to identify factors likely be of
relevance to management of the individual.
Clinical decision-making can be a complex task. We know in clinical practice that there are at
times different opinions expressed, both in terms of clinical formulation (and diagnosis) and
about what treatment is appropriate for a particular patient. However a number of researchers
have advocated the use of clinical prediction rules as an aid to clinical decision-making. These
tools are designed to assist medical decision-making in the management of low back pain, by
combining information from a number of variables, usually from the history, diagnostic tests
and/or physical examination. Typically they quantify the probability of an outcome, e.g. the
persistence/recurrence of LBP after a time period of 1 year as an indicator of the development
of chronic pain and suggest a diagnostic or therapeutic course of action. The use of clinical or
disability prediction rules has intuitive appeal and perhaps some utility as a ‘middle-way’ but is
extremely context dependent, is based on an averaging of scores across individuals and in assum-
ing a fixed relationship amongst the items effectively is restricted to a ‘snap-shot’ derived at a
particular point in time.
Wasson et al. (1985) have specified a number of standards against which such rules should be
appraised. They state inter alia that the outcome being predicted by a rule:
(1) Should be clearly defined and clinically important.
When predicting chronicity of pain the most important outcomes considered should include
pain intensity, pain-related disability, and work absenteeism
(2) Assessed in a blinded manner.
Thus the presence versus absence or the degree of an outcome should ideally be assessed
without knowledge of the status of the predictor variables, especially, when the outcome is
determined by self-reported data or subject to interpretation;
(3) The process of identification and definition of the predictor variables should be reported.
Multivariate predictive models (such as multiple regression) depend on the variables included
in a model. In order to develop reproducible prediction rules, all variables need to be taken
into account otherwise clinicians in daily practice will not be able to confidently apply the
rule with their patients.
(4) Predictor variables as well as the outcomes must be assessed in a standardized format.
(5) Patient characteristics that might have an impact on the predictive value of a prediction rule
should be assessed and reported.
e.g. gender, age, duration of pain.
(6) Aspects of the site where the study was done should be described.
e.g. the type of the institution (primary, secondary, or tertiary care), the setting (office, clinic,
outpatient setting of general practitioners, orthopaedics or physiotherapists.
(7) Mathematical procedures must be described in detail.
e.g. logistic regression, discriminant function analysis used to create the prediction rule,
and
(8) Methods of testing the accuracy and prospective validity must be reported.
Further extensions of these standards for prognostic studies reported by Linton et al. (2005) are
shown in Table 11.2.
The identification of predictors, however, is only a first step. For risk assessment in the context
of clinical application, it is necessary to be able to apply the model on an individual level with a
view on utilizing the screening technique to improve the results of interventions.
Table 11.2 Methodological issues isolated in the prognosis research field
Issue Recommendation
Design Prospective, inception cohorts required
Sampling Strictly outline inclusion/exclusion criteria
State enrolment time point clearly
For many questions, clear, early enrolment is necessary (< 3 weeks following onset)
Representative sampling techniques (random selection or consecutive cases)
Prognostic Strictly define constructs of measure
indicators
Selection should flow from conceptual framework, recognizing the multifactorial nature
of the problem
Use standardized, psychometrically sound instruments
Analysis Multivariable techniques to adjust for all potential confounders
Avoid over-fitting the data (too many covariates for sample size)
Prospective validation in homogenous cohorts required
Follow-up Strictly define outcome(s) of interest
Adequate duration of follow-up (years)
Strive for >80% follow-up rate
Patterns of attrition should be investigated to determine if is random or systematic
Blinded outcome measurement with standardized, psychometrically sound instruments
Conceptual Strictly define the construct of the problem being studied
framework
Account for the recurrent and multifactorial nature of back pain disability
Overarching theory needed identifying specific hypothesized relationships between variables
Broaden the view to include factors over the usual professional/discipline boundaries
With kind permission from Springer Science+Business Media: Journal of Occupational Rehabilitation, Prognosis and the
identification of workers risking disability: Research issues and directions for future research, 15(4), 2005, Steven J. Linton.
Accuracy of screening
In addition to general requirements of reliability and content validity for all measurements, in
consideration of screening, there are a number of key parameters by which accuracy should be
appraised. These are: sensitivity, specificity, and positive and negative predictive value. In prac-
tice, patient screening methods often necessitate a compromise between reliability and predictive
validity on the one hand, and practical utility (in terms of length and ease of administration) on
the other. Reliability may be demonstrated in terms of interobserver agreement, internal consist-
ency, and/or test-retest-reliability. Although reliability does not guarantee high predictive power,
an unreliable test will probably not be sufficiently valid to be useful.
The accuracy of a clinical prediction rule or a simple screening test in identifying people with
subacute back pain who go on to develop a negative outcome is referred to as the sensitivity of
the test. The sensitivity of the test thus is defined as the percentage of cases with a positive screen-
ing score (derived from prospective longitudinal studies) at baseline, who in fact go on to develop
the adverse outcome (whether persistent/recurrent pain, disability and/or work absenteeism) at
follow-up. In contrast, specificity is a measure of the proportion of patients who are classified as
probable cases of good outcome, who do not in fact go on to develop a poor outcome. Ideally a
test will have both a very high sensitivity (accuracy in identifying those who go on to have a poor
outcome) and high specificity (a low rate of ‘false positives’). There often has to be a ‘trade-off’ in
terms of false negatives (missed cases) and false positives (incorrectly identified cases) and the
worth of the test will be determined by the value which is placed on the accuracy of identification,
e.g. a high proportion of false negatives may be considered an acceptable price to pay for a high
level of identification of those at risk of a potentially fatal disease.
The number of false positives of a test is known as its positive predictive value of a test (PPV) and
false negatives termed as the negative predictive value of a test (NPV).
These estimates are also affected by the prevalence of the risk at early stages of the disease. For
example, in case of a very low prevalence of psychosocial risk factors (5%), the PPV of a test,
showing sensitivity of 70% and a specificity of 96%, will yield a PPV of about 48% (and a NPV of
nearly 98%). However with a prevalence of 35% and same sensitivity and specificity, the PNV will
be about 90% and the NPV 86% (Lund et al. 1995). In the case of a yellow flag screening, it is
therefore important to consider both prevalence and length of follow-up (to outcome) in consid-
ering the utility of screening methods. As a number of epidemiological studies, including clinical
studies, have shown, psychosocial risk factors will be of low prevalence during the first 2–4 weeks
of duration of non-specific low back pain, while the prevalence will increase up to 30–40%, 6–12
weeks later (during the subacute phase). Although a precise demarcation between acute and a
subacute phases is probably not merited, recent flag guidelines (Kendall et al. 2009), have recom-
mended screening optimally should be conducted at the subacute stage of the disease.
The importance of timing (lead time of screening)

A further difficulty for screening in addition to method and accuracy is the issue of timing since
different predictors emerge at different stages of illness or disability. When the screening is under-
taken has a major influence on the accuracy of the screening and there may be different challenges
facing screening at different times. Waddell et al. (2003) illustrate this in consideration of the
prediction of incapacity whereby, at the time of initial presentation, the task for a screening tool
may be to identify the 1–2% of individuals who will go on to develop long-term incapacity from
amongst the 98–99% with relatively simple problems, most of whom are likely to return to work
quite rapidly, more or less irrespective of any intervention. However, with benefit claimants iden-
tified only at 26% the task may be to identify the 40% likely to continue on long-term benefit
from those who might (or could) be returned to work.
The interval from screening detection of risk factors (yellow flags) to the time of clinical
diagnosis, or, in the case of LBP, to the time of development of chronic pain, disability, prolonged
work absenteeism and early retirement is defined as ‘lead time’, a term borrowed from early
studies in cancer screening (Hutchison and Shapiro 1968), while the time from disease onset to
clinical diagnosis is labelled a ‘sojourn time’ (Draisma et al. 2009). The advantage gained by
screening of yellow flags as risk factors for the development of chronic pain and disability depends
on the extent to which diagnosis can be advanced by this procedure. The key challenge for screen-
ing is to identify cases before they have become chronic and therefore open the possibility of
interrupting the path to chronicity.
Consider a hypothetical case of the development of chronic low back pain, as shown in
Figure 11.1 which begins with an episode of acute non-specific LBP that leads someone to visiting
a primary care physician.
An acute episode of pain is commonly defined by a new onset of pain following a pain-free
interval of at least 6 months (Von Korff et al. 1992). However, as Von Korff and Miglioretti (2006,
see also Chapter 2 by Dunn et al. in this volume) have shown, a number of patients are character-
ized by a high variability of pain episodes of different intensity, a different degree of disabling
character and with different duration of pain-free phases. In the single case it is therefore difficult
to reliably detect the point of an acute episode of pain that fits into the definition proposed by
Von Korff et al. (1992).
It is important to note that there is a relation between the duration of lead time and the prob-
ability of overdiagnosis (false positives). As Draisma and colleagues have shown convincingly
for the screening of prostate cancer, the longer the duration of lead time, the higher was the prob-
ability of false positive predictions (Draisma et al. 2009). It therefore depends on the utility of
risk-factor based interventions (e.g. the efficacy of treatment psychosocial risk factors) on the one
hand and the costs of these interventions on the other, which of defined lead times or screening
time points might show an optimal cost-effectiveness.
Prevalence Prevalence
of risk factors of risk factors
for chronic pain for chronic pain
5–7% 35–45%
Week 0 Week 6 Week 12 Week 24 Weeks 52–76
Acute pain Sub-acute Sub-chronic Chronic Early retirement due to DLBP

episode DLBP DLBP DLBP is recommended
Yellow-flag
screening
Lead time interval to

clinical diagnosis of
DLBP
Lead time interval to long-term

consequences of DLBP
Fig. 11.1 (See also Colour Plate 8) Phases in the development of chronic disabling low back pain
(DLBP): hypothetical case.
Pain-free
Probability
Chronic pain
Screening score
Fig. 11.2 Typical distribution of screening scores in chronic pain and healthy samples.
The determination of optimal cut-off points

A screening test provides clinicians with a defined cut-off point enabling them to predict positive
(e.g. ongoing moderate or severe pain at the follow-up) or negative (e.g. pain-free) cases.
Depending of the accuracy of a screening test, both populations will overlap to a specific degree.
Patients who develop persistent/recurrent pain at the follow-up will mostly show a greater varia-
bility in the screening scores than the patients developed free of pain (Schwarzer et al. 2002, see
also Figure 11.2).
All patients, who develop persistent pain at the follow-up, but with a screening score below this
cut-off score will be incorrectly classified as free of risk of future pain (false negative). All patients
who become pain-free, but with a screening score above the cut-off will be incorrectly classified
as positive (false positive). With a low cut-off point, the probability of a false negative prediction
is much less than a false positive (as shown in Figure 11.3).
Results from a hypothetical prospective study are shown in Table 11.3 where 94 out of 500
patients with subacute LBP (18.8%) are found to have developed persistent pain at follow-up. In
this example, the low cut-off point for the yellow-flag screening yields three false negative predic-
tion (3.2%) and 91 correct predictions (i.e. an overall sensitivity 96.8%). However, 139 out of 406
patients who did not show a persistent pain at follow-up (34.3%) and were therefore incorrectly
classified as high risk patients (false positive), leading to a specificity of 65.7%).
In contrast, with a high cut-off point, the probability of false negative predictions will increase
and the false positive will decrease (see Figure 11.4).
Thus in the example in Table 11.4, seven out of 94 patients developing chronic pain (7.4%) had
a negative test result (sensitivity 92.5%), while only six out of 406 patients, who did not develop
persistent pain, were false positive tests (1.5%), and with this high cut-off point, the specificity
was 98.5%.
Ultimately, the choice of the cut-off point is a clinical one (Schwarzer et al. 2002). It is deter-
mined, as reasoned above, by the utility and the costs of the consequences of a positive screening
result. Using the procedure of receiver operating characteristic (ROC) curves finding an optimal
Pain-free
False positive
Probability
False negative Chronic pain
Low cut-off
point
Screening score
Fig. 11.3 False positive and false negative predictions in the case of a low cut-off point. Reproduced
from Schwarzer, G., Türp, J.C., Antes, G., EbM-Splitter: Die Vierfeldertafel (in Diagnosestudien):
Sensitivität und Spezifität, Deutsche Zahnärztliche Zeitschrift 2002;57:446–447, Figures 2 and 3,
© with permission from Deutscher Ärzte-Verlag.
cut-off point is a method becoming increasingly popular in judging the discriminative power of
new screening measures (Hanley and McNeil, 1982; Hill et al. 2008).
Combining different screening tools to improve prediction

As it is shown in the later section ‘Methods for screening yellow flags’, currently available screen-
ing tools differ in their ability to predict different outcomes such as future pain or pain-related
disability and days off work due to back pain. Furthermore, single risk factors are of limited
predictive value (McIntosh and Pepe 2002; Wald et al. 2005). A potential solution might be to
combine multiple markers in some way to increase the predictive power of both outcomes, pain
intensity and also disability, but this possibility for yellow flags has not as yet been explored and
accuracy of prediction will have to be considered in terms of feasibility in clinical practice, since
if the system of screening is too long or complex it will simply not be used.
Table 11.3 Prediction in the case of a low cut-off point
Persistent pain at the follow-up

Yes No
Screening Positive 91 139 230
Negative 3 267 270
94 406 500
Reprinted from Manual Therapy, 9 (1), A. Kim Burton, Timothy D. McClune, Robert D. Clarke, and Chris J. Main, Long-
term follow-up of patients with low back pain attending for manipulative care: outcomes and predictors, pp. 30-5,
Copyright (2004), with permission from Elsevier.
Pain-free
Probability
False negative Chronic pain
False positive
High cut-off
point
Screening score
Fig. 11.4 False positive and false negative predictions in the case of a high cut-off point.
Reproduced from Schwarzer, G., Türp, J.C., Antes, G., EbM-Splitter: Die Vierfeldertafel (in
Diagnosestudien): Sensitivität und Spezifität, Deutsche Zahnärztliche Zeitschrift 2002;57:446–447,
Figures 2 and 3, © with permission from Deutscher Ärzte-Verlag.
Feasibility of screening
Ideally, screening procedures must be acceptable in terms of content not only to the population
which are being screened but, in the case of LBP, to healthcare providers as well as to the purchas-
ers of healthcare, whether insurance companies or government departments. However if they are
actually to be used in practice, yellow-flag screening should be convenient to physicians and
patients and thus be easy and economical to administer. Finally of course, perhaps hardest of all,
if it cannot contribute to the development of treatments effective in preventing chronic phases of
disease, then the initiative will be valueless
How then do we decide whether a proposed method of screening is actually worthwhile? High
economic or emotional costs of screening can only be justified if a high level of prediction (i.e.
high predictive value, PV) is achieved, but a somewhat lower level of prediction might be consid-
ered acceptable with a relatively inexpensive screening which is easily tolerated by patients. In
conclusion therefore, optimal yellow flag screening should be easy to administer and acceptable
to patients, offer a high level of detection of those likely to become chronic and in addition facili-
tate effective physician-patient communication in the context of secondary prevention.
Table 11.4 Results for screening using a high cut-off point
Persistent pain at the follow-up

Yes No
Screening Positive 87 6 230
Negative 7 400 270
94 406 500
11.5 Methods for screening yellow flags

During the past two decades, a number of screening measures for the assessment of psychosocial
risk factors, indicating a high probability of chronic pain and disability development, have been
published. For example, Dionne (2005) predicted the level of disability 2 years later with 82%
accuracy from the patients’ initial level of distress.
Studies however differ with respect to the healthcare setting in which they have been validated
(occupational versus clinical setting, primary-care versus specialized care settings), in the degree
of specificity of screening and in the extent to which they have been empirically or theoretically
underpinned.
While approaches to occupational screening are described in more detail in Chapter 19 of this
volume, there are several examples of clinical screening which merit attention in this chapter.
Turk (1990, 2005) has been a long advocate of customizing patients for treatment on the basis of
their initial presenting characteristics. Most screening, can be described in terms of ‘screening in’,
‘screening out’ or ‘screening with targeting’. However, the sensitivities/specificities need to be
examined in relationship to the specific purpose of screening, and it should be emphasized
that screening out from one treatment does not necessary imply screening in for another sort of
treatment.
We have, for purposes of illustration, described four different assessment tools, which differ in
their construction, validation and range of utility. The first tool, the Distress Risk Assessment
Method or DRAM (Main et al. 1992) is a symptom-based measure designed for use in orthopae-
dic clinics (or similar musculoskeletal services). It was developed in response to a specific
challenge, that of incorporating a measure of distress in patients under consideration for ortho-
paedic treatment. At the time of its development it was not considered feasible to incorporate a
complex psychological tool within the context of time-constrained orthopaedic clinics, and so it
was developed specifically as a first-stage screener. The STarT Back Decision Tool or SBDT (Hill
et al. 2008), is also a first-stage screener, but developed for primary care. Although in a shorter
dichotomous format, it captures a wider range of information (using psychosocial and biomedi-
cal predictors of chronicity). The Örebro Musculoskeletal Pain Screening Questionnaire or
ÖMPSQ (Linton and Halden 1998; Boersma and Linton 2002) has a much stronger psychosocial
focus and since its incorporation into the original yellow flag assessment system (Kendall
et al.1997) has come to be widely used in physiotherapy clinics as a secondary prevention screener
in patients already referred from primary care. Although relatively little information has been
provided on its structural properties in terms of construct validity, it has proved to be clinically
robust, is widely used and now available in a shorter form. Finally we appraise the The RIsk
SCreening of Back Pain or RISC-BP (Hallner and Hasenbring 2004), an instrument with a com-
plex aetiology, explicitly derived from research into the psychological processes underpinning
pain coping strategies, and which includes a focus on cognitive, behavioural, and emotional fac-
tors and was developed specifically for the prediction of persistent or recurrent pain.
11.5.1 Distress Risk Assessment Method (DRAM)

DRAM (Main et al. 1992) was developed specifically as a screening tool for the identification of
distress in LBP patients. It comprises a somatic awareness scale (The MSPQ and the Modified
Zung Depression Inventory). DRAM offers a simple classification of patients into those show-
ing no psychological distress, those at risk of developing major psychological overlay, and
those clearly distressed. Four patient types were identified on the basis of scores on the two short
questionnaires.
Classification
N: Modified Zung score 17.
R: Modified Zung score 17–33 and MSPQ score LE 12.
DD: Modified Zung score GE 34.
DS: Modified Zung score 17–33 and MSPQ score GE 13.
In the original validation study, the authors found increased risk of poor outcome across the
distress categories for no improvement of pain (N vs. Type R: odds ratio (OR) 2.0; N vs. DD/DS:
OR 3.5), higher levels of disability (N vs. R: OR 1.9; N vs. DD/DS: OR 5.2), and for not working
(N vs. R: OR 1.5; N vs. DD/DS: OR 2.5). In a study of outcome of osteopathic treatment, Burton
et al. (2004b) found that a higher level of somatic symptoms (MSPQ) inter alia was associated
with poorer outcome. Trief et al. (2000), in a prospective study of psychological predictors of
lumbar surgery found the DRAM classification to be a strong predictor of outcome.
However, a caveat is in order, in patients with significantly troublesome pain problems,
a patient-centred approach to evaluation is always appropriate and it is certainly not appropri-
ate to view distress as a contraindication to treatment with good surgical indications (Hobby
et al. 2001).
Recommendations for use

◆ Recommended in secondary and tertiary care clinics as a screening tool to aid clinical appraisal
of the patient.
◆ Particularly recommended in problem back clinics and pain clinics.
◆ Should be viewed as a first-stage screener, not as a complete psychological assessment or as a

test of malingering.
◆ Distress and its management should always be considered as part of decisions about surgery.
◆ Distress patients not requiring surgery require pain management.
Adapted from Main et al. (1992, p. 52).

Although based on a restricted domain (somatic and depressive symptoms only), this short
screening test has been validated specifically for patients consulting with LBP and does seem to
have predictive validity for a range of outcomes
11.5.2 The STarT Back Decision Tool (SBDT)

Hill et al. (2008), have developed a screening tool, SBDT, as the basis of a system of triaging and
targeting LBP patients presenting with modifiable physical and psychosocial prognostic indica-
tors for persistent pain at the time of consultation to their GP in primary care. An initial set of
items was selected on the basis of secondary analysis of pre-existing data on 1200 LBP consulters
to primary care, an extensive literature review and input from an expert panel of clinicians. All
items for the tool were developed from validated primary-care measurement tools. For prognos-
tic indicators measured using composite instruments, individual questions were chosen using
ROCs with high sensitivity and at least moderate specificity for identifying patients above the
median on full scores of their composite measures. The clinical validity of the items was con-
firmed following appraisal by a consensus group of some 40 GPs and clinical physiotherapists,
and blinded ratings of 12 videotaped 20-minute interviews of patients who had independently
completed the SBDT (Hill et al. 2010).The tool was then tested for feasibility, repeatability and
validity on a consecutive series of 244 patients consulting with LBP in eight GP practices.
Following examination of the distribution of SBDT scores, optimal cut-offs, based on the prog-
nostic indicators included were derived, to allocate patients into one of three groups. An SBDT
score of 0–3 was determined to be low risk; a score of four or more of the five psychosocial items
(items 1, 4, 7, 8 and 9), allocates patients to the high-risk group. The remainder (SBDT score of
more than 3, but less than four of the five psychosocial items) allocates patients to the medium-
risk group. This procedure led to an allocation of 40% into the low-risk group, with 35% into the
medium-risk group, with predominantly physical indicators, and 25% into the high-risk group
with a higher proportion of psychological indicators.
The external validity of the tool was examined in an independent observational sample of 500
LBP primary-care consulters. SBDT scores were similar and recommended cut-offs for allocation
to the three subgroups validated. The SBDT also demonstrated high predictive abilities with
medium- and high-risk groups having relative risks of 3.0 and 4.5 respectively, compared to the
low-risk group for high disability (RMDQ 7) at 6-month follow-up. It forms the basis for the
STarT Back screening and targeted approach (Hay et al. 2008) in which clinicians have been
trained in delivery of interventions on patients identified according to their risk profile. Preliminary
analyses suggest that the screening and targeting approach has proved superior, particularly in the
more complex patients to a non-screening or targeting approach at 4 months after treatment, and
to be more cost-effective in the management of low risk patients.
11.5.3 Örebro Musculoskeletal Pain Screening Questionnaire (ÖMPSQ)

Linton and Hallden (1998)1 developed a screening tool, primarily for psychosocial risk factors for
chronic pain and chronicity. Although not offering clustering as such, the widespread clinical use
of cut-off scores to identify the presence of significant psychosocial risk factors enable patient
screening. It was included in the New Zealand Screening Instrument (Kendall et al. 1997) and was
published later in a slightly modified form as the Örebro Screening Questionnaire for Pain
(Boersma and Linton 2002). The final version comprises 25 items, of which six deal with back-
ground factors, and 19 deal with a variety of background factors. There are extremely minor dif-
ferences in the precise descriptors between the New Zealand and Swedish versions and the way in
which employment status is coded (featuring as an additional numbered item) but in all impor-
tant respects the versions seem to be equivalent. The authors state that in order to give all items
equal weight, they were scaled from 0 to 10 (with some items reversed) and the scores summated
to form a total score. Since the background items (1–4) were not found to contribute statistically,
they were not included in the calculation of the total score. A scoring system was provided such
that a total score could be obtained.
In the earlier publication (Linton and Hallden 1998), it was stated that a cut-off of 105 enabled
75% correct identification of those not requiring alteration to ongoing management, 86% correct
identification of those who would have between 1 and 30 days off work in the next year and 83%
correct identification of those who would have more than 30 days off work.
Hurley et al. (2000), in an investigation of the instrument’s utility in predicting RTW after
physical therapy, found that a cut-off of 112 correctly identified 80% of patients failing to RTW
at the end of treatment (sensitivity) and 59% of those who did RTW (specificity). In further
analysis of the same cohort of patients (Hurley et al. 2001 ) they found that scores on the
instrument (labelled the ALBPSQ) were correlated with the patients’ level of pain and reported
1 The original Linton and Hallden Questionnaire appeared originally in the first yellow flags monograph
(Kendall et al. 1997); and therefter has appeared in various publications in very slightly modified forms as
the ÖMPQ, the ÖMPSQ and the ALBPSQ; and there is now a short-form (Linton et al. 2010).
disability at 1-year follow-up and ‘correctly identified all patients reporting some degree of work
loss but had minimal predictive strength for the other patient-centred variables evaluated’.
In the more recent publication (Boersma and Linton 2002), the authors report satisfactory test–
retest reliability (0.83), using a cut-off of 105 (the maximum is 210), and a specificity of 0.75 with
a sensitivity of 0.88 in the prediction of future absenteeism, and examine the utility of various
cut-off scores in the prediction of sick listing (ranging from 90–120) and for functional ability
(ranging from 80–110) in terms of sensitivity and specificity. As they point out, decisions about
the trade-off between sensitivity and specificity may be different in relation to differing goals
among healthcare professionals.
The questionnaire has considerable appeal as a general psychosocial screener, and is now fairly
widely used in clinical practice. It has both strengths and limitations. The questionnaire targets
important predictors of long-term pain-associated functional limitations and also occupational
outcomes such as duration of sick leave and RTW, but on the evidence currently available would
seem to function best in relation to occupational outcomes. As a screener therefore it is to be
recommended.
It is clear that very different clinical and occupational profiles can yield similar scores on the
questionnaire. This may not be critical in terms of first-level screening, but, as mentioned earlier
in this chapter, screening should only be undertaken with specific purposes in mind. As the prop-
erties of the questionnaire become clearer with its use in a range of settings, it will be possible
hopefully to examine its utility as a stand-alone screener, and also, possibly linking with more
detailed and specific evaluations, to identify individuals likely to benefit from targeted interven-
tions of various types.
It is to be hoped that as part of its further development, there will be further appraisal of its
structural properties with the development of more sensitive scoring systems. As things stand it
represents an extremely worthwhile addition to the psychometric validation.
11.5.4 The Risk Screening of Back Pain (RISC-BP)

Hasenbring and colleagues developed a screening tool, RISC-BP, to identify LBP patients with
and without radiating pain during a subacute phase, who would develop persistent or recurrent
pain in a primary care or a specialized care setting (Hallner and Hasenbring, 2004).
The RISC-BP is part of a large-scale assessment of potential psychosocial risk factors that have
been investigated in different prospective longitudinal studies in patients with subacute low back
and leg pain (Hasenbring et al. 1994; Grebner et al. 1999; Hallner and Hasenbring, 2004). These
studies originally started with a large pool of standardized scales, such as the BDI, a measure of
health-related locus of control, a measure of daily stress in 15 life areas, and the Kiel Pain Inventory
or KPI, Hasenbring 1994), the latter consisting of 18 subscales of pain-related emotional, cogni-
tive, and behavioural responses. The construction of the KPI was theory driven, based on the
behavioural theory of operant conditioning (Fordyce 1976), blended with the cognitive stress and
coping theory of Lazarus (1993), and the integrated cognitive-behavioural model of pain (Turk
et al. 1983). The full assessment battery (excluding pain variables) comprised 162 items. The
predictive power of these scales has been investigated using several statistical prediction rules and
a variety of statistical methods (multiple regression analyses, discriminant analyses, neuronal
network). In the initial study with 111 patients with subacute back and leg pain prior to an inpa-
tient conservative medical treatment, Hasenbring et al. (1994) demonstrated that, after control
for depression, the contrasting behavioural pain responses of high avoidance behaviour and
endurance despite severe pain revealed significant and important predictive power. The psycho-
social predictor variables yielded a sensitivity of 79% and specificity of 81% predicting pain at the
6-months follow-up. Whereas different pain responses were highly predictive of future pain
intensity in this study, application for early retirement due to work absenteeism (the second out-
come of interest) was predicted only by depression and distress at work with sensitivity of 75%, a
specificity of 86% (Hasenbring et al. 1994).
The classification of three high-risk groups focuses explicitly on cognitive, behavioural, and
emotional factors. Specific subgroup analyses based on pain-related avoidance versus endurance
and depression, using both a clinical cut-off procedure (Hasenbring 1993) and an empirical
approach of cluster analysis (Grebner et al. 1999) created the basis for the formulation of a first
version of the avoidance-endurance model (AEM) (Hasenbring 2000). The clinical cut-offs used
in the Hasenbring (1993) study were based on a BDI score of 9, which was published as a highly
reliable and valid cut-off in low back pain patients (Geisser et al. 1997). Using the Thought
Suppression Scale (TSS) and the Behavioural Endurance Scale (BES) of the KPI as further marker
variables, mean scores of ‘3’ (‘this happens to me sometimes’) on a scale ranging from ‘0’ (never)
to ‘6’ (always) were taken as a cut-off score. Patients are characterized in terms of distress endur-
ance responses (DER) if the BDI sum score was equal or above 9 and if the TSS and/or the BES
was above or equal the cut-off score of 3 and as eustress endurance responses (EER) if the BDI was
below 9. When the BDI was above/equal 9 and both, TSS and BES were below 3, patients were
considered to show fear-avoidance pain responses (FARs). Patients whose scores were below the
cut-offs on all three variables were considered to show adaptive responses (ARs). This AEM-
based classification of psychosocial risk for chronic back pain is indicating targeted cognitive-
behavioural interventions, described in more detail in Chapter 27 of this volume.
The current measure of the RISC-BP includes a measure of depression (using the Beck
Depression Inventory (Beck et al. 1961), and measures of both thought suppression and behav-
ioural endurance, derived from the avoidance-endurance model of pain (as described in
Chapter 15) and identified by the Avoidance-Endurance Questionnaire or AEQ (Hasenbring
et al. 2009. All three scales demonstrate sufficient to high internal consistency and are widely
validated in pain patients (Hasenbring 1993; Hasenbring et al. 1994, 2009; Geisser et al. 1997).
The accuracy of the RISC-BP has been shown as to be sufficient or high in two independent
prospective trials. Using an artificial neural network analysis, Hallner and Hasenbring (2004)
calculated the predictive power of the three subscales BDI, TSS, and BES, later named as RISC-BP,
in a sample of 90 subacute back and leg pain patients before and 6 months after an inpatient
medical treatment. In this cross-validated study, persistent/recurrent pain at the 6-months fol-
low-up was correctly classified in 83% of the patients with a sensitivity of 73% and a specificity of
97%. In a recent prospective study in patients with non-specific subacute low back pain (pain
duration <90 days), who underwent an outpatient medical treatment in primary care, the
RISC-BP demonstrated a sensitivity of 80%, specificity of 62% and a positive predictive value
(PPV) of 77.9%, a prevalence (pre-test probability) of 57% presupposed (Hasenbring et al.
in prep).
There is preliminary evidence that avoidance-endurance model based subgroups showing a
pattern of cognitions of thought suppression, anxious/depressive mood and task/pain persistence
behaviour (distress endurance responses) or a pattern of cognitions of ignoring/minimizing pain,
positive mood despite pain and task/pain persistence behaviour (eustress endurance responses)
will develop more pain prospectively and show higher levels of specific strain postures, measured
by accelerometer, than patients showing adaptive pain responses (Hasenbring et al. 2006).
Interestingly, the eustress endurance subgroup has been shown to develop rather low scores of
disability despite severe pain intensity (see Chapter 15).
Finally, the RISC-BP is available both in a paper–pencil version as well as in a computer-based
version (Hasenbring and Hallner, 1999) and therefore is easy to administer and interpret. The
test duration, investigated in a number of primary care practices (GPs as well as orthopaedic
practitioners), on average was about 10 minutes, and found to be both feasible and acceptable by
patients attending a 1-day education programme for primary care providers.
11.6 Summary and conclusion

Based on a huge number of methodically well-controlled prospective longitudinal studies in
subacute LBP patients, conducted during the past two decades, psychosocial factors have been
shown as important predictive or prognostic factors indicating the high risk for developing
chronic pain and pain-related disability in the long term. While high inter-correlations among
some of these variables clearly are of methodological concern (Foster et al. 2010) in terms of con-
ceptual overlap, and of practical concern in terms of the pragmatics of screening, there is convinc-
ing evidence that variables such as depression, distress in daily life, and maladaptive modes of
cognitive/affective and behavioural pain responses play a significant role as so-called yellow flags.
(Blue and black flags, both relevant within occupational settings, are reviewed in Chapter 19 of
this volume).
In the assessment of yellow flags and appraising their utility, we have identified a number of
important influences on their validity and utility (such as accuracy, time and context of identifica-
tion, and the determination of optimal cut-off scores) and more practical issues such as feasibility.
We have specifically examined four different measures which have been used for psychosocial
screening, have been validated on clinical samples and for which some predictive validity has been
demonstrated. Thus while three of these screening tools (DRAM, SBDT, and Örebro) are highly
predictive of pain-related disability (and the ÖMPSQ also for occupational outcomes), the RISC-BP
shows high predictive validity particularly for future pain intensity, perhaps as a consequence of its
specific focus on endurance-related pain-responses, as it has been shown in several well controlled
studies, that pain-related disability is more related to distress and pain-related fear than to pain
intensity itself (Crombez et al. 1999). However, while consistent evidence has confirmed the role
of fear-avoidance responses to pain and pain-related disability measured via self-report, the influ-
ence of fear-avoidance responses on objectively assessed physical activity is less clear.
In considering screening from a public health perspective, it is important to identify all poten-
tial risk factors that may be of relevance. We have drawn a distinction between actuarial screen-
ing, based principally on administrative data, and individual screening in the context of
consideration for pain management. We have emphasized the need for clarity in terms of purpose
for screening, since predictors are, to an extent, outcome specific, and may highlight the potential
value of different types of interventions.
Most screening, however, can be described in terms of ‘screening in’, ‘screening out’ with con-
sideration of the sensitivities/specificities in relationship to the specific purpose of screening. As
aforementioned, screening out from one treatment does not necessary imply screening in for
another sort of treatment.
The attempted link of screening with targeting in the case of yellow flags is found both in with
more general targeted cognitive-behavioural interventions (e.g. Linton and Anderson 2000;
Gatchel et al. 2003; Hill et al. 2008) and also with more specifically targeted interventions, such
as high fears of (re)injury (Vlaeyen et al. 2002), catastrophizing (Sullivan et al. 2006), or fear-
avoidance versus pain-related endurance (Hasenbring et al. 1999). These early interventions are
described in more detail in Chapter 27.
Risk identification and screening has a long tradition in public health, and with the advent of
evidence-based medicine the focus and ‘outcome’ has required a new way of thinking about the
possibilities for pain management. The biomedical and biomechanical models of pain and func-
tion to an extent have served us well, but newer understandings of the influences on and effects of
pain have required a broader perspective, and the biopsychosocial model has served as a vehicle
for this. We have, however, moved beyond descriptions, classifications, and model building to a
focus not only on treatment but towards prevention. In our view we now need a conceptual
framework that allows us to understand the pain patient in context.
In our view, screening has to be linked with appraisal and management of the individual in pain
and the main value of these tools may be in highlighting patients requiring a more comprehensive
assessment or in further clinic assignment. With significant pain problems, this will need to
include a clinical assessment by a competent healthcare professional and may require further
expertise to evaluate and address specific occupational problems.
Although valuable, however, all screening tools are inherently limited in the context of
decision-making at an individual level. Hill et al. (2010) directly compared the use of the StarTBack
Tool with clinical decision-making and concluded:
Clinicians are, therefore, advised to use the instrument as an adjunct to their own decision-making
rather than a replacement to their considered clinical acumen. The strengths of the SBST are likely to
be its systematic and consistent allocation of patients to subgroups, which contrasts with the experts’
overwhelming inconsistencies in decision-making, and their lack of confidence in decision-making
among more complex back pain cases.
The StarTBack initiative (Hay et al. 2008; Hill et al. submitted) in linking specifically tailored
interventions to subgroups identified on the basis of screening, in offering a ‘middle-way’ between
identifying risk factors at the level of clinical epidemiology and the delivery of treatment, is a
model which would seem to merit further investigation in other groups of patients.
The review offered in this chapter clearly demonstrates the potential for yellow flag screening
(discussed further in Nicholas et al. 2011) and we believe serves as an encouragement to strive to
further improve the focus, nature, and quality of our interventions within an evidence-based
framework and with a clear focus on a range of outcomes (as recommended in The Hopkinton
Think Tank meeting in 2005) with its methodological recommendations for improvement in
disability research (Linton et al. 2005).
References
Beck, A.T., Ward, C.H., Mendelson, M., Mock, J., Erbaugh, J. (1961). An inventory for measuring
depression. Archives of General Psychiatry, 4, 561–71.
Boersma, K., Linton, S.J. (2002). Early assessment of psychological factors: the Örebro Screening
Questionnaire for Pain. In Linton, S.J. (Ed.) New avenues for the prevention of chronic musculoskeletal
pain and disability. Pain research and clinical management, vol 12, pp. 205–13. Amsterdam: Elsevier.
Burton, A.K., Battie, M.C., Main, C.J. (1999). The Relative Importance of Biomechanical and Psychosocial
Factors in Low Back Injuries. In Karwoski W., Marras W.S. (Eds.) The Handbook of Industrial
Ergonomics, pp. 1127–38. New York: CRC Press.
Burton, A.K., Balague, F., Cardon, G., et al. (2004a). European guidelines for prevention in low back pain.
www.backpaineurope.org
Burton, A.K., McClune, T.D., Clarke, R.D., et al. (2004b). Long- term follow-up of patients with low back
pain attending for manipulative care: outcomes and predictors. Manual Therapy 9, 30–5.
Crombez, G., Vlaeyen, J.W.S., Heautz, P.H.T.G., Lysens, R. (1999). Pain-related fear is more disabling
than pain itself: evidence on the role of pain-related fear in chronic back pain disability. Pain 80,
329–39.
Dawes, R.M., Faust, D., Meehl, P.E. (1989). Clinical versus actuarial judgment. Science 243, 1668–74.
Dionne, C. (2005). Psychological distress confirmed as predictor of long-term back-related functional
limitations in primary care settings. Journal of Clinical Epidemiology 58, 714–18.
Draisma, G., Etzioni, R., Tsodikov, A. et al. (2009). Lead time and overdiagnosis in prostate-specific
antigen screening: importance of methods and context. Journal of the National Cancer Institute
101, 374–83.
Fordyce, W.E. (1976). Behavioural Methods for Chronic Pain and Illness. St. Louis, MO: C.V. Mosby.
Foster, N.E., Thomas, E., Bishop, A., Dunn, K.M., Main, C.J. (2010). Distinctiveness of psychological
obstacles to recovery in low back pain patients in primary care. Pain, 148, 398–406.
Foster, N.E., Hill, J.C., Hay, E.M. (2011). Subgrouping patients with low back pain in primary care: Are we
getting any better at it? Manual Therapy 16(1), 3–8.
Gatchel, R.J., Polatin, P.B., Noe, C.E., et al. (2003). Treatment- and cost-effectiveness of early intervention
for acute low back pain patients: A one-year prospective study. Journal of Occupational Rehabilitation
13, 1–9
Geisser, M.E., Roth, R.S., Robinson, M.E. (1997). Assessing depression among persons with chronic pain
using the Center for Epidemiological Studies-Depression Scale and the Beck Depression Inventory:
A comparative analysis. Clinical Journal of Pain 13, 163–70.
Grebner, M., Breme, K., Rothoerl, R., Woertgen, C., Hartmann, A., Thomé, C. (1999). Coping and
convalescence course after lumbar disk operations. Der Schmerz 13, 19–30.
Hallner, D., Hasenbring, M. (2004). Classification of psychosocial risk factors (yellow flags) for the
development of chronic low back and leg pain using artificial neural network. Neuroscience Letters
361, 151–4.
Hanley, J.A., McNeil, B.J. (1982). The meaning and use of the area under a receiver operating characteristic
(ROC) curve. Radiology 143, 29–36
Hasenbring, M. (1993). Endurance strategies - a neglected phenomenon in the research and therapy of
chronic pain? Der Schmerz 7, 304–13.
Hasenbring, M. (1994). Das Kieler Schmerzinventar. [The Kiel Pain Inventory-Manual. Three questionnaire
scales for the assessment of pain-related cognitions, emotions and coping strategies]. Bern: Huber.
Hasenbring, M. (2000). Attentional control of pain and the process of chronification. In: J. Sandkühler,
B. Bromm, G.F. Gebhart (Eds.) Progress in Pain Research Vol 129, pp. 525–34. New York:
Elsevier Science.
Hasenbring, M., Hallner, D. (1999). Telemedical system of patient diagnostics. Deutsches Ärzteblatt,
PC-Spektrum, 49–50.
Hasenbring, M., Marienfeld, G., Kuhlendahl, D., Soyka, D. (1994). Risk factors of chronicity in lumbar
Hasenbring, M., Ulrich, H.W., Hartmann, M., Soyka, D. (1999). The efficacy of a risk factor based
cognitive behavioral intervention and EMG-biofeedback in patients with acute sciatic pain: an attempt
to prevent chronicity. Spine, 24, 2525–35.
Hasenbring, M. I., Plaas, H., Fischbein, B. and Willburger, R. (2006). The relationship between activity and
pain in patients 6 months after lumbar disc surgery: do pain-related coping modes act as moderator
variables? European Journal of Pain, 10, 701–9.
Hasenbring, M.I., Hallner, D., Rusu, A.C. (2009). Fear-avoidance- and endurance-related responses to
pain: development and validation of the Avoidance-Endurance Questionnaire (AEQ). European
Journal of Pain, 13, 620–8.
Hay, E.M., Mullis, R., Lewis, M., et al. (2005). Comparison of physical treatments versus a brief pain
management programme for back pain in primary care: a randomised clinical trial in physiotherapy
practice. Lancet 365, 2024–30.
Hay, E.M., Dunn, K.M., Hill, J.C., et al. (2008). A randomised clinical trial of subgrouping and targeted
treatment for low back pain compared with best current care. The STarT Back Trial Study Protocol
BMC Musculskeletal Disorders 9, 58.
Hill, J.C., Dunn, K.M., Lewis, M., et al. (2008). A Primary Care Back Pain Screening Tool: Identifying
Patient Subgroups for Initial Treatment. Arthritis and Rheumatism 9, 632–41.
Hill J.C., Vohora K., Dunn K.M., Main C.J, Hay E.M. (2010). Comparing the STarT Back screening tool’s
subgroup allocation of individual patients with that of independent clinical experts. Clinical Journal of
Pain 26(9), 783–7.
Hill, J.H., Dunn, K., Mason, L., et al. (2011). Comparison of stratified primary care management for low
back pain with current best practice (STarT Back): a randomised controlled trial. The Lancet 29
September 2011 [Epub ahead of print].
Hobby. J.L., Lutchman. L.N., Powell. J.M., et al. (2001). The Distress Risk Assessment Method (DRAM):
Failure to predict the outcome of lumbar discectomy. Journal of Bone and Joint Surgery 83B(1), 19–21.
Hurley, D.A., Dusoir, T., McDonough, S., et al. (2000). Biopsychosocial screening questionnaire for
patients with low back pain: Preliminary report of utility in physiotherapy practice in N Ireland.
Clinical Journal of Pain 16, 214–28.
Hurley, D.A., Dusoir, T., McDonough, S., et al. (2001). How effective is the Acute Low Back Screening
Questionnaire for predicting one-year follow-up in patients with low back pain? Clinical Journal of
Pain 17, 256–63.
Hutchison, G.B., Shapiro, S. (1968). Lead time gained by diagnostic screening for breast cancer Journal of
the National Cancer Institute 41, 665–81.
Jellema, P., van der Windt, D.A.W., vander Horst, H.E., Twisk, J.W.R., Stalman, W.A.B, Bouter, L.M.
(2005). Should treatment of (sub)acute low back pain be aimed at psychosocial prognostic factors?
Cluster randomised clinical trial in general practice. British Medical Journal 331, 84–8.
Kendall, N.A.S., Linton, S.L., Main, C.J. (1997). Guide to assessing psychological yellow flags in acute low back
pain: risk factors for long term disability and work loss. Wellington: Accident Rehabilitation and
Compensation Insurance Corporation of New Zealand and the National Health Committee.
Wellington.
Kendall, N.A.S., Burton, A.K., Main, C.J., Watson, P.J. (2009). Musculoskeletal Problems, A Guide For The
Clinic and Workplace: Identifying Obstacles using the Psychosocial Flags framework. London: The
Stationery Office.
Lazarus, R.S. (1993). Coping theory and research: Past, present, and future. Psychosomatic Medicine, 55,
234–47.
Linton, S.J., Andersson, T. (2000). Can chronic disability be prevented? A randomized trial of a
cognitive-behavioural intervention and two forms of information for patients with spinal pain. Spine
25, 2825–31.
Linton, S.J., Hallden, K. (1998). Can we screen for problematic back pain? A screening questionnaire for
predicting out-come in acute and subacute back pain. Clinical Journal of Pain 14, 209–15.
Linton, S.J., Gross, D., Schultz, I.Z., et al. (2005). Prognosis and the identification of workers risking
disability: Research issues and directions for future research. Journal of Occupational Rehabilitation
15, 459–74.
Linton, S.L., Nicholas, M.K., MacDonald, S. (2010). Development of a Short Form of the Örebro
Musculoskeletal Pain Screening Questionnaire. Spine Dec 29. [Epub ahead of print].
Lund, J.P., Widmer, C.G., Feine, J.S. (1995). Validity of diagnostic and monitoring tests used for
temperomandibular disorders. Journal of Dental Research 74, 1133–43.
Main, C.J., Wood, P.L.R., Hollis, S., et al. (1992). The distress assessment method: A simple patient
classification to identify distress and evaluate risk of poor outcome. Spine 17, 42–50.
Main, C.J., Phillips, C.J., Watson, P.J. (2005). Secondary prevention in health care and occupational
settings in musculoskeletal conditions focusing on low back pain. In: Schultz I Z, Gatchel R J (eds)
Handbook of complex occupational disability claims: Early risk identification, intervention and prevention,
pp. 387–404. New York: Springer.
McIntosh, M.W., Pepe, M.S. (2002). Combining several screening tests: optimality of the risk score.
Biometrics, 58, 657–64.
Muir Gray, J.A. (2004). New concepts of screening. British Journal of General Practice 54, 292–8.
Nicholas, M.K., Watson, P.J., Linton, S.J., Main, C.J. (2011). The identification and management of
psychosocial risk factors (Yellow Flags) in patients with low back pain. Physical Therapy 91(5), 737–53.
Pincus, T., Morley, S.J. (2001). Cognitive-processing bias in chronic pain: a review and integration.
Psychological Bulletin 127, 599–617.
Ratcliffe, J., Thomas, K.J., MacPherson, H., Brazier, J. (2006). A randomized controlled trial of acupuncture
care for persistent low back pain: cost effectiveness analysis. British Medical Journal 333, 626.
Schwarzer, G., Türp, J.C., Antes, G. (2002). EbM-Splitter: Die Vierfeldertafel (in Diagnosestudien):
Sensitivität und Spezifität. Dtsch Zahnärztl Z, 57, 446–7.
Shaw, W.S., van der Windt, D.A., Main C.J., Loisel, P., Linton, S.J. (2009). Now tell me about your work:
the feasibility of early screening and intervention to address occupational factors (“Blue Flags”) in back
disability. Journal of Occupational Rehabilitation 19, 64–80.
Sullivan, M.J.L., Ward, L.C., Tripp, D., et al. (2005). Secondary prevention of work disability:
Community-based psychosocial intervention for musculoskeletal disorders. Journal of Occupational
Rehabilitation 15, 377–92.
Sullivan, M.J.L., Adams, H., Rhodenizer, T., et al. (2006). A psychosocial risk factor targeted intervention
for the prevention of chronic pain and disability following whiplash injury. Physical Therapy 86, 8–18.
Thomas, E., Silman, A.J., Croft, P.R., Papageorgiou, A.C., Jayson, M.I.V., Macfarlane, G. (1999).
Predicting who develops chronic back pain in primary care: a prospective study. Britsh Medical
Journal 318, 1662–7.
Trief, P., Grant, W., Fredrickson. B. (2000). A prospective study of psychological predictors of lumbar
surgery outcome. Spine 25, 2616–21.
Turk, D.C. (1990). Customising treatment for chronic pain patients; who, what, and why. Clinical Journal
of Pain 6, 255–70.
Turk, D.C. (2005). The potential of treatment matching for subgroups of patients with chronic pain:
Lumping versus splitting. Clinical Journal of Pain 21, 44–55.
Turk, D.C., Meichenbaum, D., Genest, D. (1983). Pain and behavioral medicine: a cognitive-behavioural
perspective. New York: Guilford.
Van der Windt, D., Hay, E., Jellema, P., Main C.J. (2008). Psychosocial interventions for low back pain in
Primary Care. Spine, 33, 81–9.
Van Tulder, M., Becker, A., Bekkering, T. et al. on behalf of the COST B23 Working Group on Guidelines
for the management of Acute Low Back pain in Primary Care (2005). www. backpaineurope.org
Vlaeyen, J.W., De Jong, J.R., Onghena, P., et al. (2002). Can pain-related fear be reduced? The application
of cognitive-behavioral exposure in vivo. Pain Research and Management 7, 144–53.
Von Korff, M., Miglioretti, D.L. (2006). A prospective approach to defining chronic pain. In: Flor, H.,
Kalso, E., Dostrovsky, J.O. (eds) Proceedings of the 11th World Congress on Pain, pp. 761–9. Seattle, WA:
IASP Press.
Von Korff, M., Ormel, J., Keefe, F., et al. (1992). Grading the severity of chronic pain. Pain 50, 133–49.
Waddell, G. (2004). The back pain revolution (2nd edn.). Edinburgh: Churchill Livingstone.
Waddell, G., Burton, A.K., Main, C.J. (2003). Screening of DWP clients for risk of long-term incapacity: a
conceptual and scientific review (Royal Society of Medicine Monograph). London: Royal Society of
Medicine Press.
Wald, N.J., Morris, J.K., Rish, S. (2005). The efficacy of combining several risk factors as a screening test.
Journal of Medical Screening, 12, 197–201.
Wasson, J.H., Sox, H.C., Neff, R.K., Golman, L. (1985). Clinical prediction rules–applications and
methodological standards New England Journal of Medicine 313, 793–9.
Wilson, J.M.G., Jungner, J.J. (1968). Principles and practice of screening for disease. World Health
Organization, Geneva.
Chapter 12
Dispositional Fear, Anxiety Sensitivity,

and Hypervigilance
R. Nicholas Carleton and Gordon J.G. Asmundson
12.1 Introduction
Diathesis–stress models are becoming popular and intuitive ways for researchers to describe
the course of psychopathology (McKeever and Huff 2003). Predispositional traits align with
specific environmental stressors resulting in pathological states. In the absence of either the
precedent trait or the appropriate environmental state, psychopathology does not develop.
Theorists have gone on to develop diathesis–stress models in which the stressor—and subsequent
psychopathology—become diatheses for subsequent stressors (Asmundson et al. 2002; Sharp and
Harvey 2001; Turk 2002; Vlaeyen and Linton 2000). The self-perpetuating cycles depicted by
these models are understandably pernicious and difficult to break; however, identifying and
ameliorating the fundamental predispositional trait(s) may be an effective way to provide treat-
ment for diathesis–stress psychopathology (Carleton and Asmundson 2009; Taylor 1993).
Chronic musculoskeletal pain—pain persisting beyond the expected duration required for the
healing of damaged tissue (International Association for the Study of Pain 1986)—is currently
explained using diathesis–stress models. Most often, the stressor is an injurious event followed by
persistent pain despite tissue repair. It is also possible, though presumably less common, for pain
to begin without a definitive injurious event and persist or even worsen in the absence of defini-
tive biological aetiology. Perhaps most importantly, the identification of a definitive injurious
event appears to be a relatively small component of chronic musculoskeletal pain pathology.
Indeed, tissue damage has now long been recognized as having relatively little association with
pain (Beecher 1946, 1959; Melzack and Perry 1975). A similarly venerable notion is that chronic
musculoskeletal pain symptoms can have a basis in organic pathology but nonetheless be main-
tained by psychological factors (Breuer and Freud 1974). Given the relatively small association
between pain and tissue damage, it is reasonable to expect that the psychological factors will
play a substantially greater role in chronic pain pathology. Accordingly, the psychological factors
likely function as the key diatheses that are further mitigated by social factors (Fordyce 1976;
Fordyce et al. 1982). It is within this context that researchers have investigated fundamental pre-
dispositional traits associated with psychopathology as potential diatheses for chronic musculo-
skeletal pain.
Anxiety sensitivity (AS), illness/injury sensitivity (IIS), and fear of negative evaluation (FNE)
were among the first constructs suggested to be fundamental predispositional constructs associ-
ated with more common fearful reactions and psychopathology (Reiss 1991; Reiss and McNally
1985). The fundamental fears have been argued to be inherently noxious stimuli that represent
fundamental predispositions not otherwise logically reducible to other fears (Reiss 1991; Taylor
1993). In contrast, the same investigators have argued that other fears (e.g. fear of snakes, fear
of panic attacks) are common because those fears can be reduced to one or more of AS, FNE,
and IIS. More recent research (Carleton and Asmundson 2009; Carleton et al. 2007b) has sug-
gested that at least two other constructs—pain-related anxiety and intolerance of uncertainty
(IU)—also meet the original requirements for inclusion as fundamental fears (Reiss 1991). Each
of these predispositional psychological constructs has been identified as a potentially critical dia-
thesis for the development and maintenance of chronic musculoskeletal pain (Asmundson and
Carleton 2008; Asmundson et al. 2004a; Carleton et al. 2007b). Across all of these constructs,
pain-related anxiety and fear may be among the most investigated and culpable constructs with
regard to the development and maintenance of chronic musculoskeletal pain; pain-related anxi-
ety and fear are generally thought to be key diathetic variables (Carleton and Asmundson 2009;
Vlaeyen and Linton 2000) and may be more disabling than pain itself (Crombez et al. 1999).
Current fear-anxiety-avoidance models of chronic musculoskeletal pain (Asmundson et al.
2004b; Vlaeyen and Linton 2000) have been developed based on work conducted by several
researchers (Asmundson et al. 1999a; McCracken et al. 1992; Vlaeyen et al. 1995; Waddell et al.
1993). Research on the fear-anxiety-avoidance models has resulted in considerable empirical sup-
port (Asmundson et al. 1999a; Asmundson et al. 2004b; Leeuw et al. 2008; Vlaeyen and Linton
2000). The models were generally intended to describe people with idiopathic chronic musculo-
skeletal pain, but they have also been found applicable to other pain conditions such as headache
(Norton and Asmundson 2004), fibromyalgia (Goubert et al. 2004), and severe burns (Sgroi et al.
2005). These contemporary fear-anxiety-avoidance models of chronic musculoskeletal typically
share similar basic elements (see also Chapter 14). These elements are as follows:
(1) Once the nociceptive stimulus is perceived, a judgement of the meaning or purpose of the
pain is placed on the experience.
(2) For most people, the nociceptive stimulus is judged to be undesirable and unpleasant, but
not catastrophic or suggestive of a major calamity. Accordingly, most people engage in
appropriate behavioural restriction followed by graduated increases in activity until healing
has occurred.
(3) For a significant minority of people, a catastrophic meaning (i.e. ‘This pain means I’m
never going to be able to do the things I like doing’) is placed on the experience of pain. The
quality and intensity of the catastrophizing is influenced by predispositional and current
psychological factors, and results in pain-related anxiety and/or fear of (re)injury.
(4) Pain-related anxiety spirals into a vicious and self-perpetuating cycle that promotes and main-
tains avoidance, activity limitations, disability, pain, as well as additional catastrophizing.
Researchers and theorists continue to develop these models, with proposed modifications and
additions based on evidence that the aforementioned predispositional traits can be associated
with the development and maintenance of chronic musculoskeletal pain. Below we describe each
of the posited predispositional traits–including pain-related anxiety and fear, AS, IU, IIS, and
FNE–as depicted in some fear-avoidance models of chronic pain (see, for example, Figure 12.1–
the predispositions listed below ‘Pain catastrophizing’).
12.2 Pain-related anxiety and fear

Pain-related fear and anxiety are typically referred to collectively as fear of pain (Carleton
and Asmundson 2009); however, there is growing evidence that the difference may be far from
semantic. The precedent distinction between present-tense fear and future-oriented anxiety
has been recognized as important for years (Barlow 2000, 2002), providing theoretical and
clinical utility; nevertheless, in models of chronic musculoskeletal pain the distinction remains
somewhat novel.
DISPOSITIONAL FEAR, ANXIETY SENSITIVITY, AND HYPERVIGILENCE 233
Injury
Disuse
depression Recovery
disability
Avoidance
hypervigilance
Confrontation
Pain experience
Pain-related fear
Pain catastrophizing No fear
Negative affectivity
threatening illness information
Fig. 12.1 Fear-Avoidance Model of chronic pain. Adapted from Vlaeyen and Linton (2000). This
figure has been reproduced with permission of the International Association for the Study of Pain®
(IASP®). The figure may not be reproduced for any other purpose without permission.
The distinction between pain-related fear and anxiety has been suggested to parallel the distinc-
tion drawn in context of anxiety disorders; specifically, pain-related fear can be conceptualized as
a present-oriented emotive state associated with nociceptive stimulation (e.g. pain from an injec-
tion), whereas pain-related anxiety can be conceptualized as a more general, future-oriented
emotive state that does not require nociceptive stimulation, but occurs in anticipation of nocic-
eption (e.g. the possibility of pain from an injection). This formal distinction is relatively novel
within the pain literature (Carleton and Asmundson 2009); however, earlier research identified
distinctions between pain-related fear and anxiety responses within a chronic musculoskele-
tal pain sample (McCracken et al. 1996). In that initial study, self-reports of pain-related fear
were found to be correlated primarily with pain complaints, whereas self-reports of pain-related
anxiety were correlated primarily with pain severity and disability.
Subsequent research has supported the importance of the distinction, possibly extending it to a
neuronal level. For example, hyperalgeisa has been associated with anxiety-provoking unpredict-
able pain, while hypoalgesia has been associated with fear-provoking predictable pain (Ploghaus
et al. 2003). There is also evidence that exposure-based therapies (e.g. graded activity, graded
in vivo exposure) that reduce avoidance by confronting painful anxiety-provoking stimuli can
reduce disabling chronic musculoskeletal pain in a fashion similar to exposure treatments for
anxiety disorders (Bailey et al. 2010).
Despite the apparent association with the development of chronic musculoskeletal pain, pain-
related anxiety plays an important protective role within injury-healing processes. Without some
pain-related anxiety, people would be less motivated to engage in avoidance of behaviours
expected to provoke pain, resulting in higher frequencies of tissue damage and reducing the
evolutionary utility of pain. Accordingly, it makes intuitive sense that situationally-appropriate
pain-related anxiety is beneficial; however, determining what is appropriate is likely to be a
matter for debate. Research evidence suggests that there is substantial variation across the entire
population and no readily apparent qualitative break that distinguishes between normal and
pathological levels of the trait (Asmundson et al. 2007, 2009). Accordingly, nociceptive stimula-
tion in people with increasing pain-related anxiety is likely to result in rumination, thereby fur-
ther increasing pain-related anxiety, and potentially reinforcing unnecessary avoidance
behaviours.
12.2.1 Assessment of pain-related fear and anxiety

Pain-related fear is commonly measured with the 30-item Fear of Pain Questionnaire-III (FPQ-III;
McNeil and Rainwater 1998) and, more recently, the 20-item Fear of Pain Questionnaire-Short
Form (FPQ-SF; Asmundson et al. 2008b). Items on both scales are responded to using a 5-point
Likert scale ranging from 1 (not at all) to 5 (extreme). The short-form resulted from a factor-
analytic investigation that removed unstable items and revised the originally proposed three-
factor structure of the FPQ-III (i.e. minor pain, severe pain, medical pain) into a more robust
four-factor structure. The four factorially distinct subscales of the FPQ-SF are each related to a
specific type of pain: Minor Pain (e.g. biting your tongue while eating), Severe Pain (e.g. having
someone slam a heavy car door on your hand), Injection Pain (e.g. having a blood sample drawn
with a hypodermic needle), and Dental Pain (e.g. having one of your teeth drilled). There remain
questions regarding the additive utility of the Injection Pain and Dental Pain factors, such that
they have been argued to be optional (Asmundson et al. 2008b). Both measures have good inter-
nal consistency and there is evidence of good test-retest reliability for the FPQ-III (McNeil and
Rainwater 1998; Osman et al. 2002).
Pain-related anxiety is most commonly measured with either the original 40-item Pain Anxiety
Symptoms Scale (McCracken et al. 1993; McCracken et al. 1992) or the 20-item Pain Anxiety
Symptoms Scale-20 (PASS-20; McCracken and Dhingra 2002). Items on both scales are responded
to using a 6-point Likert scale anchored from 0 (never) to 5 (always). Both measures consist of
four factorially distinct components of pain-related anxiety: (1) Cognitive Anxiety (e.g. I can’t
think straight when in pain), (2) Pain-related Fear, (e.g. Pain sensations are terrifying), (3) Escape
and Avoidance (e.g. I try to avoid activities that cause pain), and (4) Physiological Anxiety (e.g.
Pain makes me nauseous). The 40-item PASS has been used to assess pain-related anxiety in
clinical (McCracken et al. 1992) and community (Larsen et al. 1997; Osman et al. 1994) samples.
Factorial validity for both the total and subscale scores of the PASS-20 has been demonstrated for
clinical (Coons et al. 2004) and non-clinical (Abrams et al. 2007) samples. Neither the instruc-
tions for the PASS or PASS-20, nor the items themselves, preclude the use of either measure with
participants who do not have current pain.
12.2.2 Treatment of pain-related fear and anxiety

The apparent importance of pain-related fear and anxiety in the development and maintenance
of chronic musculoskeletal pain has resulted in the derivation of at least two specific multidisci-
plinary treatment protocols. Graded in vivo exposure (GivE) is a cognitive-behavioural therapy
that targets pain-related fear and anxiety (Vlaeyen et al. 2002). Proponents of GivE assert that
pain-related fear and anxiety, as well as the associated disability, can be ameliorated by deliberat-
ing exposing people to activities previously avoided because of pain-related anxiety.
Acceptance and commitment therapy (ACT) and related approaches focus on concepts of
mindfulness, acceptance, and values-based action (Hayes 2004; Hayes and Duckworth 2006;
Hofmann and Asmundson 2008), with the ideal result being reduced pain-related anxiety,
catastrophic beliefs, and fewer avoidance behaviours (Vowles et al. 2008). Proponents of ACT
assert that patients with chronic pain need to shift the focus of their life from the pain and onto
things of greater interest or importance. A recent comprehensive review of treatments for pain-
related fear, anxiety, avoidance, and the associated disability suggested that GivE may function as
a faster treatment protocol than ACT; however, ACT may have a lower dropout rate (Bailey et al.
2010).
12.3 Anxiety sensitivity

Paralleling more than a decade of research on pain-related fear and anxiety has been a growing
interest in the relationship between AS and the development and maintenance of chronic muscu-
loskeletal pain. The AS construct represents a heritable propensity (Stein et al. 1999) to appraise
anxiety-related physiological sensations (e.g. increased heart rate, palpitations, trembling), cogni-
tive changes (e.g. difficulty concentrating, racing thoughts), and social consequences (e.g. humil-
iation, rejection) based on the belief that these sensations have harmful consequences (Taylor
1999). The catastrophic appraisal of these sensations, changes, and consequences facilitates
avoidance behaviours which, in turn, reinforce the fearful appraisals. AS is qualitatively different
from trait anxiety (Reiss et al. 1986); specifically, AS focuses on fearing unknown or potential
dangers associated with the sensations of anxiety rather than the anxiety-provoking stimulus itself
(Taylor et al. 1992). In other words, AS focuses on identifiable stimuli—the object of fear is the
anxiety-related sensations themselves. The catastrophic misappraisal of the stimuli associated
with AS requires an anxiety-related sensation accompanied by uncertainty regarding the subse-
quent consequences of the sensation (Reiss et al. 1986). Accordingly, higher levels of AS can
contribute to catastrophic misinterpretations of physical sensations related to pain or general
arousal (Asmundson 1999; Asmundson and Norton 1995; Asmundson and Taylor 1996).
The 16-item Anxiety Sensitivity Index (ASI; Peterson and Reiss 1992) is the most commonly
used measure of AS. Factor analytic investigations of the ASI indicate that it comprises three
internally consistent lower-order factors—i.e. fear of somatic sensations; somatic (e.g. It scares
me when my heart beats rapidly), fear of cognitive dyscontrol; cognitive (e.g. When I cannot keep
my mind on a task, I worry that I may be going crazy), and fear of socially observable anxiety
reactions; social (e.g. It is important to me not to appear nervous)—that load on to a single
higher-order factor (Taylor et al. 1996; Zinbarg et al. 1997). The validity and reliability of using
the ASI total or subscale scores have been well-documented (Peterson and Plehn 1999; Taylor
1999). Despite the relative popularity of the ASI, the measure is not without limitations. AS was
not originally conceptualized as a multidimensional construct and, as such, the ASI was not cre-
ated with multidimensional construct-driven factorial subscales in mind. Accordingly, the factor
structure of the ASI is often found to be unstable. Subsequent research has led to the development
of a new 18-item measure of AS—the ASI-3 (Taylor et al. 2007)—that appears to resolve the
factorial difficulties related to measuring a multidimensional construct. Thus far, the ASI-3 has
proven to be extremely robust, with a replicable factor structure, as well as good performance on
indices of reliability and validity.
The relationships between pain, anxiety, and catastrophizing were established relatively early
(Craig 1994; Rosenstiel and Keefe 1983; Sullivan and D’Eon 1990); however, the role of variant AS
levels in pain was not explicitly investigated until Kuch and colleagues included the ASI in their
investigation of phobias and pain (Kuch et al. 1994; Kuch et al. 1991). The importance of the
relationship between AS and pain — in particular the physical concerns component of AS
(Asmundson 1999)—was implied early by intuitive associations between pain-related anxiety
and pain sensations (McCracken et al. 1993). Subsequent research further supported a relation-
ship between AS and pain (Asmundson et al. 1999a), but primarily as an exacerbating construct
influencing pain-related fear and avoidance (Asmundson 1999). Recent studies have demon-
strated a relationship between AS and both laboratory-induced acute pain and chronic muscu-
loskeletal pain (Keogh and Asmundson 2004).
People with heightened AS are prone to catastrophize about somatic sensations, which include
pain sensations. Pain is naturally unpleasant, but not always interpreted catastrophically, unless
the individual is predisposed to become anxious during the experience of somatic sensation. For
such people, the pain is worse than unpleasant; instead, they are more likely to interpret pain as a
harbinger of protracted physical, social, and financial hardship. The sensation and interpretation
then function to initiate and maintain a spiralling cycle of sensations, catastrophic thoughts, and
avoidance behaviours, which describe disabling chronic musculoskeletal pain.
AS is not consistently elevated in patients with chronic pain relative to non-clinical samples
(Asmundson and Norton 1995; Asmundson and Taylor 1996). The results suggest that AS is
critical for some, but not all, people with chronic pain (Asmundson et al. 2008a). In those with
elevated AS, the fear of anxiety-related sensations, changes, and consequences may facilitate
debilitating chronic pain. To illustrate, AS has been shown to increase the risk of pain-related
avoidance and disability following physical injury (Asmundson and Norton 1995; Asmundson
and Taylor 1996) and headache (Asmundson et al. 1999b). In a sample of patients with compara-
ble injuries, patients reporting high levels of AS and chronic musculoskeletal pain were found to
use more analgesic medication relative to patients with chronic musculoskeletal pain who reported
low levels of AS (Asmundson and Norton 1995). Indeed, pain severity itself may be mediated by
AS, given evidence that caffeine-related hypoalgesia (i.e. reduced pain) may occur specifically for
women with low levels of AS (Keogh and Chaloner 2002). Women with high levels of AS also tend
to demonstrate lower levels of pain tolerance relative to men (Thompson et al. 2008). In general,
however, the evidence regarding whether the relationship between AS and pain is different for
men relative to women is conflicting, with some data indicating differences (Keogh et al. 2006)
and other data indicating no such differences (Conrod 2006; Tsao et al. 2006). In any case, con-
temporary fear-anxiety-avoidance models describe critical direct and indirect paths for AS in the
development and maintenance of chronic musculoskeletal pain (Asmundson et al. 2004a).
12.4 Anxiety sensitivity and pain-related anxiety

Researchers have demonstrated that AS can directly and indirectly exacerbate pain-related fear,
promoting avoidance behaviours even after controlling for differences in pain severity (Asmundson
et al. 1999a; Asmundson and Taylor 1996; Zvolensky et al. 2001). The relationship between AS
and pain-related anxiety is sufficiently interwoven that some researchers have suggested pain-
related anxiety may be a manifestation of AS (Asmundson and Norton 1995; Asmundson and
Taylor 1996). Pain-related anxiety is anxiety related to a somatic sensation and, therefore, argu-
ably no different than anxiety related to other somatic sensations such as heart palpitations. In
support of this argument, researchers have demonstrated that treatments for reducing AS (e.g.
interoceptive exposure) may also reduce pain-related anxiety (Watt et al. 2006). Indeed, AS has
been shown to account for most of the variance in pain-related anxiety (Greenberg and Burns
2003), even after controlling for other potential predictors of pain-related anxiety such as trait
anxiety and pain intensity (Muris et al. 2001).
Pain-related anxiety can also be conceptualized as a fundamental fear (i.e. fears of inherently
noxious stimuli that are not logically reducible to other fears; Reiss 1991; Taylor 1993) unto itself
(Carleton and Asmundson 2009; Carleton et al. 2006b), rather than a manifestation of AS. In
other words, nociceptive stimulation is inherently noxious and other more common fears can be
logically reduced to a pain-related anxiety (e.g. fearing falling because it may cause pain). In this
context pain-related anxiety and AS would be related but not necessarily dependent. The somatic
sensations described in the AS construct can produce arousing and even fearful responses; how-
ever, those sensations are not inherently noxious. For example, a rapid heartbeat can be inter-
preted as exhilarating as easily as it can be interpreted as terrifying, depending on the context (e.g.
lottery win vs. freefall off a building). In contrast, the somatic sensation associated with pain-
related anxiety does not require contextual cues or a catastrophic misinterpretation for the sensa-
tion to be interpreted as noxious. Pain is almost always unpleasant, presumably by evolutionary
design, and associated with specialized nociceptive pathways. Moreover, initial research evidence
from clinical and non-clinical samples (Asmundson et al. 2007; Asmundson et al. 2009; McNeil
and Rainwater 1998) suggests that pain-related anxiety is continuous in nature (i.e. occurring
along a latent continuum ranging from low to high), whereas evidence to date suggests that AS is
taxonic (i.e. having qualitatively distinct normative and pathological forms; Bernstein et al.
2006a; Bernstein et al. 2006b). Based on the available research to date, pain-related anxiety and AS
appear to be related, but distinct (Carleton et al. 2011). Accordingly, it makes sense to continue
to research the interdependent relationship of both constructs with pain.
12.5 Intolerance of uncertainty

Recent research has demonstrated a relationship between the AS and IU constructs, such that the
two are interdependent because IU may be a necessary component of AS (Carleton et al. 2007b).
Indeed, AS and IU appear to share a common basis in experiencing increased anxiety related to
unknown, potentially harmful consequences. IU has been identified as a discriminating disposi-
tional variable associated with worry (Laugesen et al. 2003) and state anxiety (Greco and Roger
2001). Moreover, IU is positively correlated with anxiety-related psychopathologies such as
generalized anxiety disorder (GAD), obsessive–compulsive disorder (OCD), and panic disorder
(Dugas et al. 1998; Dugas et al. 2001; Tolin et al. 2003). Given the established relationship between
AS and pain, it makes sense that IU may play an important role in pain as well.
IU is the tendency for an individual to consider the possibility of a negative event occurring as
unacceptable and threatening irrespective of the probability of its occurrence (Freeston et al.
1994; Holaway et al. 2006). Engaging in situations with uncertain outcomes (e.g. physical activity
following an injury), is likely to induce and perpetuate a heightened level of anxiety in persons
with heightened IU (Dugas et al. 2001). Moreover, people with high IU are likely to interpret
ambiguous information (e.g. muscle tension) as threatening (Heydayati et al. 2003), contributing
to increased arousal (e.g. increased heart rate and blood pressure; Greco and Roger 2001, 2003).
Despite the potential theoretical and clinical importance of understanding the role of IU in
pain (Carleton et al. 2007a), there is a paucity of research exploring the relationship. IU has been
suggested to be related to pain peripherally, as a function of GAD (Roemer and Orsillo 2007),
OCD (Sookman and Pinard 2002), or health anxiety (Abramowitz et al. 2007). To date, there
does not appear to be any research directly evaluating the relationship between IU and pain. The
lack of research may be the result of historical difficulties measuring the construct in a psycho-
metrically robust fashion. For example, the original Intolerance of Uncertainty Scale (IUS) was
fairly long and the factor structures were not robust (Freeston et al. 1994); alterative measures
were either longer than the IUS or had poor internal reliability and convergent validity (e.g.
Intolerance of Ambiguity Scale: Budner 1962; Tolerance of Ambiguity Scale: Kirton 1981).
A psychometrically stable 12-item version of the IUS, the IUS-12, has been proposed and
has become increasingly popular (Carleton et al. 2007a); however, there have also been recent
efforts to expand the scope of the construct using a newly proposed 30-item French measure
(Gosselin et al. 2008). Given the current lack of general consensus and relatively rapid research
developments regarding how to measure the IU construct, interested readers are encouraged
to review the most current literature and evaluate how each measure might best fit with their
research designs.
12.6 Illness/injury sensitivity

The relationship between IIS and pain—which has substantial intuitive face validity—remains
relatively less explored than the relationship between pain and other predispositional variables.
Indeed, most studies evaluate the construct either peripherally or split into fears of illness or fears
of injury. The division makes sense, and the IIS construct does have explicit validity and is based
on fears of becoming ill or injured; however, as a dispositional trait IIS is typically referred to as a
unidimensional construct (Carleton et al. 2006b).
The original measure for IIS was open to confound with stimuli associated with common fears
because the items were taken from the blood-injury items from the Fear Survey Schedule-II
(FSS-II; Geer 1965). A second measure, the ISI, was created to overcome the confound (Taylor
1993). The measure included face-valid items for injury (e.g. I am frightened of being injured)
and illness (e.g. I worry about becoming physically ill) evaluated on a 5-point Likert scale ranging
from 0 (agree very little) to 4 (agree very much). The measure and its latent 2-factor structure
was subsequently revised and confirmed (ISI-R; Carleton et al. 2005, 2006b), and is a common
unitary measure of the IIS construct as a fundamental fear.
The ASI and ISI-R have recently been used to evaluate the predispositional development of AS
and IIS (Watt et al. 2008). Development of AS and IIS both have been related to retrospective
reports of learning through direct and vicarious experiences of chronic musculoskeletal pain. IIS
has been shown to be a stronger predictor than AS of pain catastrophizing, pain-related fear,
pain-related anxiety, and pain tolerance (Vancleef and Peters 2006b; Vancleef et al. 2006). In
contrast, heightened AS appears related to learning to catastrophize about the meaning of somatic
sensations, rather than catastrophizing about the sensations themselves (Stewart et al. 2001; Watt
and Stewart 2000; Watt et al. 1998). In contrast, IIS appears linked to parental modelling and
reinforcement of sick-role behaviour specifically related to aches and pains, which is in line with
precedent research (Vancleef and Peters 2006b; Vancleef et al. 2006) that IIS may be subsumed
within a more general fear of somatic sensations.
Reductions in fear of injury are generally considered to be important in pain management
(Smeets et al. 2006), but whether they are as a result of an independent fear or because they func-
tion as a lower order construct of another fear remains to be determined. For example, pain-
related fear has been proximally assessed using measures of fear of injury (Roelofs et al. 2006;
Samwel et al. 2007), which implies that some researchers may believe either fear of pain or fear of
injury could be derivatives or manifestations of one another. The notion of fear of re-injury con-
tinues to be a critical component of contemporary fear-anxiety-avoidance models (Leeuw et al.
2007a). Patients with chronic lower back pain report higher levels of fear of injury relative to
healthy controls (Leeuw et al. 2007b); however, differences within chronic lower back pain
groups based on fear of injury have ranged from being relatively small (Smeets et al. 2007) to
being the best predictors of physical performance in people with this form of pain (Swinkels-
Meewisse et al. 2006). Despite the lack of robust within-group differences, significant reductions
in fear of injury have been recorded over the course of multidisciplinary treatments (van Wilgen
et al. 2007). Moreover, movement-focused fear of (re)injury has been shown to be a sensitive
measure of change associated with reports of pain severity (Ostelo et al. 2007). Finally, fear of
injury has been related to lengthy bed rest immediately following an injury and increased levels of
disability at 1 year post injury (Verbunt et al. 2008).
There is preliminary evidence that fear of illness is also higher in patients with chronic muscu-
loskeletal pain that healthy controls (Carleton et al. 2006a); however, there is evidence that fear of
illness may be heightened in persons with higher pain-related anxiety, while not being a specific
predictor of pain or disability (Hadjistavropoulos et al. 2004). Furthermore, chronic pain patients
with elevated health anxiety appear to be generally more susceptible to somatic sensations and
catastrophic thoughts; but, this does not necessarily implicate fear of illness as a predictor for pain
(Hadjistavropoulos et al. 2002). Additional hierarchical research is needed to determine what
independent roll, if any, fear of illness plays in pain.
There are alternative measures that assess injury (e.g. the Tampa Scale for Kinesiophobia;
[TSK] Kori et al. 1990; Miller et al. 1991) or illness (e.g. the Illness Attitudes Scale; [IAS] Kellner
1986) more broadly than the IIS. These alternative measures have been used in variety of pain-
related investigations with varying results (see Smeets et al. 2006 and Crossmann and Pauli 2006,
respectively). The items within each measure account for larger cognitive schemas than the items
within the IIS. For example, the TSK measures fear of movement, with the implication being that
such movement will result in re-injury. In other words, the respondent can state they are afraid to
move because they are afraid to be re-injured, rather than having an option to state only that they
are afraid to be injured. Similarly, the IAS measures broad health constructs, from specified dis-
eases (e.g. cancer) to fears of death while simultaneously, but briefly, tapping concerns about
somatic sensations similar to those assessed by the ASI. Researchers and clinicians are, of course,
encouraged to select their illness- and injury-related measures based on intended use. The TSK
and IAS may function better as symptom measures, whereas the IIS may function better as a
measure of predispositional traits.
12.7 Fear of negative evaluation

The FNE construct focuses on fears of being judged disparagingly or hostilely by others (Leary
1983), and has been shown to be partially heritable (Stein et al. 1999). There is relatively solid
support for the notion that social anxiety and Social Anxiety Disorder are driven in part or in
whole by FNE (Rapee and Heimberg 1997). For people with pain, there are at least three major
social foci affected by the pain experience. First, the possibility of chronic musculoskeletal pain
has been suggested to be a significant threat to the person’s sense of self (Morley and Eccleston
2004; Osborn and Smith 2006). Second, the legitimacy of the pain can be called into question,
undermining the person’s sense of self worth and perception of their own self (Aldrich et al. 2000;
Eccleston et al. 1997). Third, either as a function of depression or due to inability to participate in
social activities, chronic musculoskeletal pain serves to isolate people from their social support
networks (Means-Christensen et al. 2008). There is also evidence that the development of pain-
related anxiety may be affected by or partially dependent on social context (Chung 2005). Given
the potentially pervasive relationship between FNE and chronic musculoskeletal pain, there is
relatively little direct research. The research that is available suggests that, relative to the general
population, patients with chronic musculoskeletal pain may be substantially more likely to expe-
rience clinically significant levels of social anxiety (Asmundson et al. 1996a, 1996b). There
certainly appears to be evidence of social changes resulting from chronic pain, including social
withdrawal (Gheldof et al. 2006); however, prospective studies are needed to delineate whether
the increased social anxiety facilitates, or is the result of, chronic musculoskeletal pain.
The paucity of available research on FNE and pain may help to explain some of the conflicting
findings currently available. For example, there are several researchers who have found no direct
association between FNE and pain (Vancleef and Peters 2006a; Vancleef et al. 2007), reporting
instead only peripheral or indirect relationships (Vancleef et al. 2006). In contrast, relatively
higher levels of FNE and social anxiety have been found in samples of children with sickle cell
anaemia (Wagner et al. 2004) and undergraduate adult samples with high levels of pain-related
fear (Asmundson and Carleton 2005). The discrepant findings may also be the result of sampling
differences, wherein some patient groups have come to expect negative evaluation as a result of
their pain. Indeed, chronic disabling pain related to physiopathology has been associated with
fears of negative evaluation, which were thought to drive symptoms of depression and anxiety
(Richards et al. 2004). In any case, the interaction of FNE and pain is likely to be complex, refrac-
tory, and worth additional investigation within the context of psychosocial models of chronic
musculoskeletal pain (Asmundson et al. 2004a).
12.8 Hypervigilance
Hypervigilance refers to a state of relatively heightened vigilance. With respect to pain, hyper-
vigilance refers specifically to continuous scanning for somatic sensations that might be pain or
preface pain (Van Damme et al. 2004). A person who is hypervigilant is not necessarily over-
reporting symptoms. Instead, people who are hypervigilant are likely experiencing a relatively
higher sensitivity to somatic sensations as well as an increased likelihood of catastrophic interpre-
tations once somatic sensations are noticed. Hypervigilance is almost explicitly related to AS and
catastrophizing (discussed in detail above); however, the focus here will be on the sensitivity and
scanning aspects of hypervigilance.
Attending to pain requires at least some application of conscious attention (Eccleston 1995a, b;
Miron et al. 1989); there is also perplexing evidence that attending to pain (Hadjistavropoulos
et al. 2000; Keogh et al. 2000) or being distracted from pain (Villemure and Bushnell 2002) can
both result in lower perceived pain intensities. Furthermore, growing evidence suggests that
people with heightened pain-related fear or fear of re-injury are more likely to attend to pain
sensations (Asmundson and Hadjistavropoulos 2007) without realizing the impact that fear may
be having on their pain experience (LeDoux 1998). Evidence of this bias has been underscored in
people who endure chronic musculoskeletal pain and have heightened pain-related fear or fear of
re-injury (Khatibi et al. 2009). Evidence also exists that cognitively demanding tasks modulate
neuronal activity in several regions involved in pain processing, including insular, cingulate,
and premotor cortices, thalamus, and cerebellum (Bantick et al. 2002). In other words, pain is
necessarily mediated by cognitive attentional biases that can result in hypervigilance.
The results of a recent review of attentional bias research were inconsistent (Pincus and Morley
2001). Only eight studies were available at time of review; nevertheless, the authors concluded
that patients with chronic musculoskeletal pain do appear hypervigilant towards sensory pain-
related stimuli. These results were further supported by a subsequent meta-analysis (Roelofs et al.
2002) that included five studies, all of which were included in the eight studies from the Pincus
and Morley (2001) analysis. Despite general consensus for attentional biases towards threat
stimuli relative to neutral stimuli (e.g. Andersson and Haldrup 2003; Asmundson et al. 2005;
Duckworth et al. 1997; Roelofs et al. 2003), pain-related threat research is conflicting. Some
research has resulted in differences (Beck et al. 2001; Boston and Sharpe 2005; Carleton et al.
2006a; Dehghani et al. 2004) and some has not (Asmundson et al. 2005; Petzke et al. 2003). The
discrepant results may be explained by design differences surrounding the actual pain experi-
enced during research efforts to assess hypervigilance (Vangronsveld et al. 2007). In addition,
the different results are almost certainly due, in part, to the dispositional differences discussed in
this chapter (e.g. pain-related fear and anxiety, AS, IIS, FNE). Such differences may function to
mediate or moderate not only each pain experience, but also individual vigilance towards pain.
There is also substantial evidence that individual differences in life events—particularly traumatic
ones—can serve to maintain and exacerbate chronic pain (Sharp and Harvey 2001); however, a
description of the mutual maintenance model of traumatic life events and pain is beyond the
scope of this chapter.
12.9 Conclusion
Despite the growing body of research into the dynamic, complex, and individual experience of
pain, a great deal remains to be learned. Researchers are continuing to explore the relationships
between dispositional fears, hypervigilance, and chronic musculoskeletal pain. Treatment out-
comes for chronic musculoskeletal pain are often less encouraging than hoped, due largely to a
lack of data on, and understanding of, the mechanisms that underlie chronicity. Recent efforts to
design treatment protocols that reduce pain-related disability by addressing pain-related fear and
anxiety have been promising (Bailey et al. 2010) as have efforts to treat pain-related anxiety
through interoceptive exposure protocols developed to address elevated AS (Watt et al. 2006).
Future research should consider employing Watt and colleagues’ interoceptive treatment exam-
ple with people who have chronic musculoskeletal pain—the study would allow an evaluation of
whether the promising results obtained in undergraduate students with elevated pain-related
anxiety can be achieved in those disabled by their pain experiences. Given the relationship between
AS and IU (Carleton et al. 2007b), researchers might also explore whether ambiguous pain (e.g.
chronic non-specific back pain, chronic unexplained headache) is more threatening and there-
fore more disabling than other types of pain for which there is sometimes a reasonably clear
medical explanation (e.g. arthritis). The AS and IU constructs, both of which drive catastrophiz-
ing, appear to offer promising avenues for future research into ameliorating chronic muscu-
loskeletal pain. Alternatively, researchers might continue with efforts to clearly delineate the
nature and magnitude of attentional biases associated with hypervigilance. If the biases are suffi-
ciently specific, modifications such as those described in recent anxiety disorder research may
provide additional relief for some patients (Amir et al. 2009a; Amir et al. 2009b). In any case,
improving our understanding of each of the predispositional variables described in this chapter
should lead to better a understanding and, thereby, potentially improved therapeutic interven-
tions for people trying to cope with chronic musculoskeletal pain.
References
Abramowitz, J. S., Olatunji, B. O. and Deacon, B. J. (2007). Health anxiety, hypochondriasis, and the
anxiety disorders. Behavior Therapy, 38, 86–94.
Abrams, M. P., Carleton, R. N. and Asmundson, G. J. G. (2007). An exploration of the psychometric
properties of the PASS-20 with a nonclinical sample. The Journal of Pain, 8, 879–86.
Aldrich, S., Eccleston, C. and Crombez, G. (2000). Worrying about chronic pain: vigilance to threat and
misdirected problem solving. Behaviour Research and Therapy, 38, 457–70.
Amir, N., Beard, C., Burns, M. and Bomyea, J. (2009a). Attention modification program in individuals
with generalized anxiety disorder. Journal of Abnormal Psychology, 118, 28–33.
Amir, N., Najmi, S. and Morrison, A. S. (2009b). Attenuation of attention bias in obsessive-compulsive
disorder. Behaviour Research and Therapy, 47, 153–7.
Andersson, G. and Haldrup, D. (2003). Personalized pain words and Stroop interference in chronic pain
patients. European Journal of Pain, 7, 431–8.
Asmundson, G. J. G. (1999). Anxiety sensitivity and chronic pain: Empirical findings, clinical implications,
and future directions. In S. Taylor (ed.) Anxiety sensitivity: theory, research and treatment of the fear of
anxiety, pp.269–85. New Jersey: Erlbaum.
Asmundson, G. J. G., Abrams, M. P. and Collimore, K. C. (2008a). Pain and anxiety disorders. In
M. J. Zvolensky and J. A. J. Smits (eds.) Health behaviors and physical illness in anxiety and its disorders:
Contemporary theory and research, pp.207–35. New York: Springer.
Asmundson, G. J. G., Bovell, C. V., Carleton, R. N. and McWilliams, L. A. (2008b). The Fear of Pain
Questionnaire - Short Form (FPQ-SF): Factorial validity and psychometric properties. Pain, 134, 51–8.
Asmundson, G. J. G. and Carleton, R. N. (2005). Fear of pain is elevated in adults with co-occurring
trauma-related stress and social anxiety symptoms. Cognitive Behaviour Therapy, 34, 248–55.
Asmundson, G. J. G. and Carleton, R. N. (2008). Fear of pain. In M. M. Antony and M. B. Stein (eds.)
Handbook of anxiety and the anxiety disorders, pp. 551–61. New York: Oxford University Press.
Asmundson, G. J. G., Collimore, K. C., Bernstein, A., Zvolensky, M. J. and Hadjistavropoulos, H. D.
(2007). Is the latent structure of fear of pain continuous or discontinuous among pain patients?
Taxometric analysis of the pain anxiety symptoms scale. The Journal of Pain, 8, 387–95.
Asmundson, G. J. G., Coons, M. J., Taylor, S. and Katz, J. (2002). PTSD and the experience of pain:
research and clinical implications of shared vulnerability and mutual maintenance models. Canadian
Journal of Psychiatry, 47, 930–7.
Asmundson, G. J. G. and Hadjistavropoulos, H. D. (2007). Is high fear of pain associated with general
hypervigilance? A categorical re-analysis. The Journal of Pain, 8, 11–18.
Asmundson, G. J. G., Hadjistavropoulos, T., Bernstein, A. and Zvolensky, M. J. (2009). Latent structure of
fear of pain: an empirical test among a sample of community dwelling older adults. European Journal of
Pain, 13, 419–25.
Asmundson, G. J. G., Jacobson, S. J., Allerdings, M. D. and Norton, G. R. (1996a). Social phobia in
disabled workers with chronic musculoskeletal pain. Behavior Reseach and Therapy, 34, 939–43.
Asmundson, G. J. G. and Norton, G. R. (1995). Anxiety sensitivity in patients with physically unexplained
chronic back pain: a preliminary report. Behaviour Research and Therapy, 33, 771–7
Asmundson, G. J. G., Norton, G. R. and Jacobson, S. J. (1996b). Social, blood/injury, and agoraphobic fears
in patients with physically unexplained chronic pain: are they clinically significant? Anxiety, 2, 28–33.
Asmundson, G. J. G., Norton, P. J. and Norton, G. R. (1999a). Beyond pain: the role of fear and avoidance
in chronicity. Clinical Psychology Review, 19, 97–119.
Asmundson, G. J. G., Norton, P. J. and Veloso, F. (1999b). Anxiety sensitivity and fear of pain in patients
with recurring headaches. Behaviour Research and Therapy, 37, 703–13.
Asmundson, G. J. G., Norton, P. J. and Vlaeyen, J. W. S. (2004a). Fear-avoidance models of chronic pain:
An overview. In G. J. G. Asmundson, J. W. S. Vlaeyen and G. Crombez (eds.) Understanding and
Treating Fear of Pain, pp.3–24. Oxford: Oxford University Press.
Asmundson, G. J. G. and Taylor, S. (1996). Role of anxiety sensitivity in pain-related fear and avoidance.
Journal of Behavioral Medicine, 19, 577–86.
Asmundson, G. J. G., Vlaeyen, J. W. S. and Crombez, G. eds. (2004b). Understanding and treating fear of
pain. New York: Oxford.
Asmundson, G. J. G., Wright, K. D. and Hadjistavropoulos, H. D. (2005). Hypervigilance and attentional
fixedness in chronic musculoskeletal pain: consistency of findings across modified stroop and
dot-probe tasks. The Journal of Pain, 6, 497–506.
Bailey, K. M., Carleton, R. N., Vlaeyen, J. W. and Asmundson, G. J. (2010). Treatments Addressing
Pain-Related Fear and Anxiety in Patients with Chronic Musculoskeletal Pain: A Preliminary Review.
Cognitive Behaviour Therapy, 39, 46–63.
Bantick, S. J., Wise, R. G., Ploghaus, A., Clare, S., Smith, S. M. and Tracey, I. (2002). Imaging how
attention modulates pain in humans using functional MRI. Brain, 125, 310–19.
Barlow, D. H. (2000). Unraveling the mysteries of anxiety and its disorders from the perspective of emotion
theory. American Psychologist, 55, 1247–63.
Barlow, D. H. (2002). Anxiety and Its Disorders. 2nd Edition. New York, NY: Guilford Press.
Beck, J. G., Freeman, J. B., Shipherd, J. C., Hamblen, J. L. and Lackner, J. M. (2001). Specificity of Stroop
interference in patients with pain and PTSD. Journal of Abnormal Psychology, 110, 536–43.
Beecher, H. K. (1946). Pain in men wounded in battle. Bulletin. United States Army Medical Department,
5, 445–54.
Beecher, H. K. (1959). Generalization from pain of various types and diverse origins. Science, 130, 267–8.
Bernstein, A., Leen-Feldner, E. W., Kotov, R., Schmidt, N. B. and Zvolensky, M. J. (2006a). A test of a
panicrelevant diathesis-stress model using a taxonic index of anxiety sensitivity. In A. J. Sanfelippo
(ed.) Panic Disorders: New Research, pp.15–40. Hauppauge, NY: Nova Science Publishers Inc.
Bernstein, A., Zvolensky, M. J., Kotov, R., et al. (2006b). Taxonicity of anxiety sensitivity: a multi-national
analysis. Journal of Anxiety Disorders, 20, 1–22.
Boston, A. and Sharpe, L. (2005). The role of threat-expectancy in acute pain: effects on attentional bias,
coping strategy effectiveness and response to pain. Pain, 119, 168–75.
Breuer, J. and Freud, S. (1974). Studies on hysteria. New York: Basic Books.
Budner, S. (1962). Intolerance for ambiguity as a personal variable. Journal of Personality, 30, 29–50.
Carleton, R. N. and Asmundson, G. J. G. (2009). The multidimensionality of fear of pain: Construct
independence for the Fear of Pain Questionnaire-Short Form and the Pain Anxiety Symptoms
Scale-20. The Journal of Pain, 10, 29–37.
Carleton, R. N., Asmundson, G. J. G., Antony, M. M. and McCabe, R. (2011). Pain-related anxiety and
anxiety sensitivity across anxiety and depressive disorders. Journal of Anxiety Disorders, 23, 791–8.
Carleton, R. N., Adams, M. P., Asmundson, G. J. G., Collimore, K. C. and Ellwanger, J. (2006a). Strategic
and automatic threat processing in chronic musculoskeletal pain: a startle probe investigation.
Cognitive Behaviour Therapy, 35, 236–47.
Carleton, R. N., Asmundson, G. J. G. and Taylor, S. (2005). Fear of Physical Harm: Factor Structure and
Psychometric Properties of the Injury/Illness Sensitivity Index. Journal of Psychopathology and
Behavioral Assessment, 27, 235–41.
Carleton, R. N., Norton, M. A. and Asmundson, G. J. G. (2007a). Fearing the unknown: a short version of
the Intolerance of Uncertainty Scale. Journal of Anxiety Disorders, 21, 105–17.
Carleton, R. N., Park, I. and Asmundson, G. J. G. (2006b). The Illness/Injury Sensitivity Index: an
examination of construct validity. Depression and Anxiety, 23, 340–6.
Carleton, R. N., Sharpe, D. and Asmundson, G. J. G. (2007b). Anxiety sensitivity and intolerance of
uncertainty: requisites of the fundamental fears? Behaviour Research and Therapy, 45, 2307–16.
Chung, D. (2005). Something for nothing: understanding purchasing behaviors in social virtual
environments. Cyberpsychological Behavior, 8, 538–54.
Conrod, P. J. (2006). The role of anxiety sensitivity in subjective and physiologic responses to social and
physical stressors. Cognitive Behaviour Therapy, 35, 216–25.
Coons, M. J., Hadjistavropoulos, H. D. and Asmundson, G. J. G. (2004). Factor structure and
psychometric properties of the Pain Anxiety Symptoms Scale-20 in a community physiotherapy clinic
sample. European Journal of Pain, 8, 511–16
Craig, K. D. (1994). Emotional aspects of pain. In P.D. Wall and R. Melzack (eds.) The textbook of pain,
pp.261–74. Edinburgh: Churchill Livingstone.
Crombez, G., Vlaeyen, J. W. S., Heuts, P. H. T. G. and Lysens, R. (1999). Pain-related fear is more
disabling than pain itself: Evidence on the role of pain-related fear in chronic back pain disability. Pain,
80, 329–39.
Crossmann, A. and Pauli, P. (2006). The factor structure and reliability of the Illness Attitude Scales in a
student and a patient sample. BMC Psychiatry, 6, 46.
Dehghani, M., Sharpe, L. and Nicholas, M. K. (2004). Modification of attentional biases in chronic pain
patients: a preliminary study. European Journal of Pain, 8, 585–94.
Duckworth, M. P., Iezzi, A., Adams, H. E. and Hale, D. (1997). Information processing in chronic pain
disorder: A preliminary analysis. Journal of Psychopathology and Behavioral Assessment, 19, 239–55.
Dugas, M. J., Gagnon, F., Ladouceur, R. and Freeston, M. H. (1998). Generalized anxiety disorder: a
preliminary test of a conceptual model. Behavior Reseach and Therapy, 36, 215–26.
Dugas, M. J., Gosselin, P. and Landouceur, R. (2001). Intolerance of Uncertainty and Worry: Investigating
Specificity in a Nonclinical Sample. Cognitive Therapy and Research, 25, 551–8.
Eccleston, C. (1995a). The attentional control of pain: methodological and theoretical concerns. Pain,
63, 3–10.
Eccleston, C. (1995b). Chronic pain and distraction: an experimental investigation into the role of
sustained and shifting attention in the processing of chronic persistent pain. Behavior Reseach and
Therapy, 33, 391–405.
Eccleston, C., Williams, A. C. and Rogers, W. S. (1997). Patients’ and professionals’ understandings of
the causes of chronic pain: blame, responsibility and identity protection. Social Science Medicine, 45,
699–709.
Fordyce, W. E. (1976). Behavioral methods for chronic pain and illness. St. Louis: Mosby., MO
Fordyce, W. E., Shelton, J. L. and Dundore, D. E. (1982). The modification of avoidance learning pain
behaviors. Journal of Behavioral Medicine, 5, 405–14.
Freeston, M., Rhéaume, J., Letarte, H., Dugas, M. J. and Ladouceur, R. (1994). Why do people worry?
Personality and Individual Differences, 17, 791–802.
Geer, J. H. (1965). The development of a scale to measure fear. Behaviour Reseach and Therapy, 3, 45–53.
Gheldof, E. L., Vinck, J., Van den Bussche, E., Vlaeyen, J. W., Hidding, A. and Crombez, G. (2006). Pain
and pain-related fear are associated with functional and social disability in an occupational setting:
evidence of mediation by pain-related fear. European Journal of Pain, 10, 513–25.
Gosselin, P., Ladouceur, R., Evers, A., Laverdiere, A., Routhier, S. and Tremblay-Picard, M. (2008).
Evaluation of intolerance of uncertainty: development and validation of a new self-report measure.
Journal of Anxiety Disorders, 22, 1427–39.
Goubert, L., Crombez, G., Van Damme, S., Vlaeyen, J. W. S., Bijttebier, P. and Roelofs, J. (2004).
Confirmatory Factor Analysis of the Tampa Scale for Kinesiophobia: Invariant Two-Factor Model
Across Low Back Pain Patients and Fibromyalgia Patients. Clinical Journal of Pain, 20, 103–10.
Greco, V. and Roger, D. (2001). Coping with uncertainty: the construction and validation of a new
measure. Personality and Individual Differences, 31, 519–34.
Greco, V. and Roger, D. (2003). Uncertainty, stress and health. Personality and Individual Differences,
34, 1057–68.
Greenberg, J. and Burns, J. W. (2003). Pain anxiety among chronic pain patients: specific phobia or
manifestation of anxiety sensitivity? Behaviour Research and Therapy, 41, 223–40.
Hadjistavropoulos, H. D., Asmundson, G. J. and Kowalyk, K. M. (2004). Measures of anxiety: is there a
difference in their ability to predict functioning at three-month follow-up among pain patients?
European Journal of Pain, 8, 1–11.
Hadjistavropoulos, H. D., Asmundson, G. J., LaChapelle, D. L. and Quine, A. (2002). The role of health
anxiety among patients with chronic pain in determining response to therapy. Pain Research and
Management, 7, 127–33.
Hadjistavropoulos, H. D., Hadjistavropoulos, T. and Quine, A. (2000). Health anxiety moderates the
effects of distraction versus attention to pain. Behavior Reseach and Therapy, 38, 425–38.
Hayes, S. C. (2004). Acceptance and Commitment Therapy, Relational Frame Theory, and the Third Wave
of Behavioral and Cognitive Therapies. Behavior Therapy, 35, 639–65.
Hayes, S. C. and Duckworth, M. P. (2006). Acceptance and Commitment Therapy and Traditional
Cognitive Behavior Therapy Approaches to Pain. Cognitive and Behavioral Practice, 13, 185–7.
Heydayati, M., Dugas, M. J., Buhr, K. and Francis, K. (2003). The relationship between intolerance of
uncertainty and the interpretation of ambiguous and unambiguous information. In Annual
Convention of the Association for Advancement of Behaviour Therapy, Boston, MA.
Hofmann, S. G. and Asmundson, G. J. G. (2008). Acceptance and mindfulness-based therapy: New wave or
old hat? Clinical Psychology Review, 28, 1–16.
Holaway, R. M., Heimberg, R. G. and Coles, M. E. (2006). A comparison of intolerance of uncertainty in
analogue obsessive-compulsive disorder and generalized anxiety disorder. Journal of Anxiety Disorders,
20, 158–74.
International Association for the Study of Pain Subcommittee on Taxonomy. (1986). Classification of chronic
pain. Descriptions of chronic pain syndromes and definitions of pain terms. Pain Suppl, 3, S1–226.
Kellner, R. (1986). Somatization and Hypochondriasis. New York: Praeger Publishers.
Keogh, E. and Asmundson, G. J. G. (2004). Negative affectivity, catastrophizing, and anxiety sensitivity. In
G. J. G. Asmundson, J. W. S. Vlaeyen and G. Crombez (eds.) Understanding and Treating Fear of Pain,
pp. 91–115. Oxford: Oxford University Press.
Keogh, E., Barlow, C., Mounce, C. and Bond, F. W. (2006). Assessing the relationship between cold pressor
pain responses and dimensions of the anxiety sensitivity profile in healthy men and women. Cognitive
Behaviour Therapy, 35, 198–206.
Keogh, E. and Chaloner, N. (2002). The moderating effect of anxiety sensitivity on caffeine-induced
hypoalgesia in healthy women. Psychopharmacology (Berl), 164, 429–31.
Keogh, E., Hatton, K. and Ellery, D. (2000). Avoidance versus focused attention and the perception of pain:
differential effects for men and women. Pain, 85, 225–30.
Khatibi, A., Dehghani, M., Sharpe, L., Asmundson, G. J. and Pouretemad, H. (2009). Selective attention
towards painful faces among chronic pain patients: evidence from a modified version of the dot-probe.
Pain, 142, 42–7.
Kirton, M. J. (1981). A reanalysis of two scales of tolerance of ambiguity. Journal of Personality Assessment,
45, 407–14.
Kori, S. H., Miller, R. P. and Todd, D. D. (1990). Kinesiophobia: A new view of chronic pain behavior. Pain
Management, 3, 35–43.
Kuch, K., Cox, B. J., Evans, R. J. and Schulman, I. (1994). Phobias, panic, and pain in 55 survivors of road
vehicle accidentts. Journal of Anxiety Disorders, 8, 181–7.
Kuch, K., Cox, B. J., Woszczyna, C. B., Swinson, R. P. and Shulman, I. (1991). Chronic pain in panic
disorder. Journal of Behavior Therapy and Experimental Psychiatry, 22, 255–9.
Larsen, D. K., Taylor, S. and Asmundson, G. J. G. (1997). Exploratory factor analysis of the Pain Anxiety
Symptoms Scale in patients with chronic pain complaints. Pain, 69, 27–34.
Laugesen, N., Dugas, M. J. and Bukowski, W. M. (2003). Understanding adolescent worry: the application
of a cognitive model. Journal of Abnormal Child Psychology, 31, 55–64.
Leary, M. R. (1983). A brief version of the Fear of Negative Evaluation Scale. Personality and Social
Psychology Bulletin, 9, 371–5.
LeDoux, J. (1998). The emotional brain. New York: Simon & Schuster.
Leeuw, M., Goossens, M. E., Linton, S. J., Crombez, G., Boersma, K. and Vlaeyen, J. W. (2007a). The
fear-avoidance model of musculoskeletal pain: current state of scientific evidence. Journal of Behavioral
Medicine, 30, 77–94
Leeuw, M., Goossens, M. E., van Breukelen, G. J., de Jong, J. R., Heuts, P. H., Smeets, R. J., Koke, A. J. and
Vlaeyen, J. W. (2008). Exposure in vivo versus operant graded activity in chronic low back pain
patients: results of a randomized controlled trial. Pain, 138, 192–207.
Leeuw, M., Peters, M. L., Wiers, R. W. and Vlaeyen, J. W. (2007b). Measuring fear of movement/(re)injury
in chronic low back pain using implicit measures. Cognitive Behaviour Therapy, 36, 52–64.
McCracken, L. M. and Dhingra, L. (2002). A short version of the Pain Anxiety Symptoms Scale (PASS-20):
preliminary development and validity. Pain Research and Management, 7, 45–50.
McCracken, L. M., Gross, R. T., Aikens, J. and Carnrike, C. L., Jr. (1996). The assessment of anxiety and
fear in persons with chronic pain: a comparison of instruments. Behaviour Research and Therapy, 34,
927–33.
McCracken, L. M., Gross, R. T., Sorg, P. J. and Edmands, T. A. (1993). Prediction of pain in patients with
chronic low back pain: effects of inaccurate prediction and pain-related anxiety. Behaviour Research
and Therapy, 31, 647–52.
McCracken, L. M., Zayfert, C. and Gross, R. T. (1992). The Pain Anxiety Symptoms Scale: development
and validation of a scale to measure fear of pain. Pain, 50, 67–73.
McKeever, V. M. and Huff, M. E. (2003). A diathesis-stress model of posttraumatic stress disorder:

Ecological, biological, and residual stress pathways. Review of General Psychology, 7, 237–50.
McNeil, D. W. and Rainwater, A. J., 3rd. (1998). Development of the Fear of Pain Questionnaire—III.
Journal of Behavioral Medicine, 21, 389–410.
Means-Christensen, A. J., Roy-Byrne, P. P., Sherbourne, C. D., Craske, M. G. and Stein, M. B. (2008).
Relationships among pain, anxiety, and depression in primary care. Depression and Anxiety, 25, 593–600.
Melzack, R. and Perry, C. (1975). Self-regulation of pain: The use of alpha-feedback and hypnotic training
for the control of chronic pain. Experimental Neurology, 46, 452–69.
Miller, R. P., Kori, S. H. and Todd, D. D. (1991). ‘The Tampa Scale for Kinesiophobia.’ Tampa, FL:
Unpublished work.
Miron, D., Duncan, G. H. and Bushnell, M. C. (1989). Effects of attention on the intensity and
unpleasantness of thermal pain. Pain, 39, 345–52.
Morley, S. and Eccleston, C. (2004). The object of pain. In G. J. G. Asmundson, J. W. S. Vlaeyen and
G. Crombez (eds.) Understanding and Treating Fear of Pain pp.163–88. Oxford: Oxford University Press.
Muris, P., Vlaeyen, J. and Meesters, C. (2001). The relationship between anxiety sensitivity and fear of pain
in healthy adolescents. Behavior Reseach and Therapy, 39, 1357–68.
Norton, P. J. and Asmundson, G. J. (2004). Anxiety sensitivity, fear, and avoidance behavior in headache
pain. Pain, 111, 218–23.
Osborn, M. and Smith, J. A. (2006). Living with a body separate from the self. The experience of the body
in chronic benign low back pain: An interpretative phenomenological analysis. Journal for Caring
Sciences, 20, 216–22.
Osman, A., Barrios, F. X., Osman, J. R., Schneekloth, R. and Troutman, J. A. (1994). The Pain Anxiety
Symptoms Scale: psychometric properties in a community sample. Journal of Behavioral Medicine,
17, 511–22.
Osman, A., Breitenstein, J. L., Barrios, F. X., Gutierrez, P. M. and Kopper, B. A. (2002). The fear of pain
questionnaire-III: Further reliability and validity with nonclinical samples. Journal of Behavioral
Medicine, 25, 155–73.
Ostelo, R. W., Swinkels-Meewisse, I. J., Knol, D. L., Vlaeyen, J. W. and de Vet, H. C. (2007). Assessing pain
and pain-related fear in acute low back pain: what is the smallest detectable change? Int J Behav Med,
14, 242–8.
Peterson, R. A. and Plehn, K. (1999). Measuring anxiety sensitivity. In S. Taylor (ed.) Anxiety sensitivity:
Theory, research, and treatment, pp.61–81. Mahwah, NJ: Erlbaum.
Peterson, R. A. and Reiss, S. (1992). Anxiety Sensitivity Index Manual. 2nd Edition. Worthington, OH:
International Diagnostic Systems
Petzke, F., Clauw, D. J., Ambrose, K., Khine, A. and Gracely, R. H. (2003). Increased pain sensitivity in
fibromyalgia: effects of stimulus type and mode of presentation. Pain, 105, 403–13.
Pincus, T. and Morley, S. (2001). Cognitive processing bias in chronic pain: A review and integration.
Psychological Bulletin, 127, 599–617.
Ploghaus, A., Becerra, L., Borras, C. and Borsook, D. (2003). Neural circuitry underlying pain modulation:
expectation, hypnosis, placebo. Trends Cogn Sci, 7, 197–200.
Rapee, R. M. and Heimberg, R. G. (1997). A cognitive-behavioral model of anxiety in social phobia.
Behaviour Research and Therapy, 35, 741–56.
Reiss, S. (1991). Expectancy model of fear, anxiety, and panic. Clinical Psychology Review, 11, 141–53.
Reiss, S. and McNally, R. J. (1985). The expectancy model of fear. In S. Reiss and R. R. Bootzin (eds.)
Theoretical Issues in Behaviour Therapy, pp.107–21. New York, NY: Academic Press.
Reiss, S., Peterson, R. A., Gursky, D. M. and McNally, R. J. (1986). Anxiety sensitivity, anxiety frequency
and the predictions of fearfulness. Behaviour Research and Therapy, 24, 1–8.
Richards, H. L., Herrick, A. L., Griffin, K., Gwilliam, P. D. H. and Fortune, D. G. (2004). Psychological
adjustment to systemic sclerosis: Exploring the association of disease factors, functional ability, body
related attitudes and fear of negative evaluation. Psychology, Health & Medicine, 9, 29–39.
Roelofs, J., Peters, M. L., Patijn, J., Schouten, E. G. and Vlaeyen, J. W. (2006). An electronic diary
assessment of the effects of distraction and attentional focusing on pain intensity in chronic low back
pain patients. Br J Health Psychol, 11, 595–606.
Roelofs, J., Peters, M. L. and Vlaeyen, J. W. (2003). The modified Stroop paradigm as a measure of selective
attention towards pain-related information in patients with chronic low back pain. Psycholoical
Reports, 92, 707–15.
Roelofs, J., Peters, M. L., Zeegers, M. P. and Vlaeyen, J. W. (2002). The modified Stroop paradigm as
a measure of selective attention towards pain-related stimuli among chronic pain patients: a
meta-analysis. European Journal of Pain, 6, 273–81.
Roemer, L. and Orsillo, S. M. (2007). An open trial of an acceptance-based behavior therapy for generalized
anxiety disorder. Behaviour Research and Therapy, 38, 72–85.
Rosenstiel, A. K. and Keefe, F. J. (1983). The use of coping strategies in chronic low back pain patients:
relationship to patient characteristics and current adjustment. Pain, 17, 33–44.
Samwel, H. J., Kraaimaat, F. W., Crul, B. J. and Evers, A. W. (2007). The role of fear-avoidance and helplessness
in explaining functional disability in chronic pain: a prospective study. Int J Behav Med, 14, 237–41.
Sgroi, M. I., Willebrand, M., Ekselius, L., Gerdin, B. and Andersson, G. (2005). Fear-avoidance in
Recovered Burn Patients: Association with Psychological and Somatic Symptoms. Journal of Health
Psychology, 10, 491–502.
Sharp, T. J. and Harvey, A. G. (2001). Chronic pain and posttraumatic stress disorder: mutual
maintenance? Clinical Psychology Review, 21, 857–77.
Smeets, R. J., van Geel, A. C., Kester, A. D. and Knottnerus, J. A. (2007). Physical capacity tasks in chronic
low back pain: what is the contributing role of cardiovascular capacity, pain and psychological factors?
Disability and Rehabilitation, 29, 577–86.
Smeets, R. J., Vlaeyen, J. W., Kester, A. D. and Knottnerus, J. A. (2006). Reduction of pain catastrophizing
mediates the outcome of both physical and cognitive-behavioral treatment in chronic low back pain.
The Journal of Pain, 7, 261–71.
Sookman, D. and Pinard, G. (2002). Overestimation of threat and intolerance of uncertainty in obsessive
compulsive disorder. In R. O. Frost and G. Steketee (eds.) Cognitive Approaches to Obsessions and
Compulsions: Theory, Assessment and Treatment, pp.63–89. Oxford: Elsevier Press.
Stein, M. B., Jang, K. L. and Livesley, W. J. (1999). Heritability of anxiety sensitivity: a twin study. Am
J Psychiatry, 156, 246–51.
Stewart, S. H., Taylor, S., Jang, K. L., et al. (2001). Causal modeling of relations among learning history,
anxiety sensitivity, and panic attacks. Behavior Reseach and Therapy, 39, 443–56.
Sullivan, M. J. and D’Eon, J. L. (1990). Relation between catastrophizing and depression in chronic pain
patients. Journal of Abnormal Psychology, 99, 260–3.
Swinkels-Meewisse, I. E., Roelofs, J., Oostendorp, R. A., Verbeek, A. L. and Vlaeyen, J. W. (2006). Acute
low back pain: pain-related fear and pain catastrophizing influence physical performance and
perceived disability. Pain, 120, 36–43.
Taylor, S. (1993). The structure of fundamental fears. Journal of Behavior Therapy and Experimental
Psychiatry, 24, 289–99.
Taylor, S. ed. (1999). Anxiety sensitivity: Theory, research, and treatment of the fear of anxiety. Mahwah, NJ:
Erlbaum.
Taylor, S., Koch, W. J. and McNally, R. J. (1992). How does anxiety sensitivity vary across the anxiety
disorders? . Journal of Anxiety Disorders, 6, 249–59.
Taylor, S., Koch, W. J., Woody, S. and McLean, P. (1996). Anxiety sensitivity and depression: how are they
related? Journal of Abnormal Psychology, 105, 474–9.
Taylor, S., Zvolensky, M. J., Cox, B. J., et al. (2007). Robust dimensions of anxiety sensitivity: development
and initial validation of the Anxiety Sensitivity Index-3. Psychological Assessment, 19, 176–88
Thompson, T., Keogh, E., French, C. C. and Davis, R. (2008). Anxiety sensitivity and pain: generalisability
across noxious stimuli. Pain, 134, 187–96.
Tolin, D. F., Abramowitz, J. S., Brigidi, B. D. and Foa, E. B. (2003). Intolerance of uncertainty in
obsessive-compulsive disorder. Journal of Anxiety Disorders, 17, 233–42.
Tsao, J. C., Lu, Q., Kim, S. C. and Zeltzer, L. K. (2006). Relationships among anxious symptomatology,
anxiety sensitivity and laboratory pain responsivity in children. Cognitive Behaviour Therapy, 35,
207–15.
Turk, D. C. (2002). A diathesis-stress model of chronic pain and disability following traumatic injury. Pain
Research and Management, 7, 9–19.
Van Damme, S., Crombez, G., Eccleston, C. and Roelofs, J. (2004). The role of hypervigilance in the
experience of pain. In G. J. G. Asmundson, J. W. S. Vlaeyen and G. Crombez (eds.) Understanding and
Treating Fear of Pain, pp.71–90. Oxford: Oxford University Press.
van Wilgen, C. P., Bloten, H. and Oeseburg, B. (2007). Results of a multidisciplinary program for patients
with fibromyalgia implemented in the primary care. Disability & Rehabilitation, 29, 1207–13.
Vancleef, L. M. and Peters, M. L. (2006a). The interruptive effect of pain on attention. The Journal of Pain,
7, 21–2.
Vancleef, L. M. and Peters, M. L. (2006b). Pain catastrophizing, but not injury/illness sensitivity or anxiety
sensitivity, enhances attentional interference by pain. The Journal of Pain, 7, 23–30.
Vancleef, L. M., Peters, M. L., Gilissen, S. M. and De Jong, P. J. (2007). Understanding the role of injury/
illness sensitivity and anxiety sensitivity in (automatic) pain processing: an examination using the
extrinsic affective simon task. The Journal of Pain, 8, 563–72.
Vancleef, L. M., Peters, M. L., Roelofs, J. and Asmundson, G. J. G. (2006). Do fundamental fears
differentially contribute to pain-related fear and pain catastrophizing? An evaluation of the sensitivity
index. European Journal of Pain, 10, 527–36.
Vangronsveld, K., Van Damme, S., Peters, M., Vlaeyen, J., Goossens, M. and Crombez, G. (2007). An
experimental investigation on attentional interference by threatening fixations of the neck in patients
with chronic whiplash syndrome. Pain, 127, 121–8.
Verbunt, J. A., Sieben, J., Vlaeyen, J. W., Portegijs, P. and Andre Knottnerus, J. (2008). A new episode of
low back pain: who relies on bed rest? European Journal of Pain, 12, 508–16.
Villemure, C. and Bushnell, M. C. (2002). Cognitive modulation of pain: how do attention and emotion
influence pain processing? Pain, 95, 195–9.
Vlaeyen, J. W., de Jong, J., Geilen, M., Heuts, P. H. and van Breukelen, G. (2002). The treatment of fear of
movement/(re)injury in chronic low back pain: further evidence on the effectiveness of exposure
in vivo. Clinical Journal of Pain, 18, 251–61.
Vlaeyen, J. W., Kole-Snijders, A. M., Boeren, R. G. and van Eek, H. (1995). Fear of movement/(re)injury in
chronic low back pain and its relation to behavioral performance. Pain, 62, 363–72.
Vlaeyen, J. W. and Linton, S. J. (2000). Fear-avoidance and its consequences in chronic musculoskeletal
pain: a state of the art. Pain, 85, 317–32.
Vowles, K. E., McCracken, L. M. and Eccleston, C. (2008). Patient functioning and catastrophizing in
chronic pain: The mediating effects of acceptance. Health Psychology, 27, S136–S43.
Waddell, G., Newton, M., Henderson, I., Somerville, D. and Main, C. J. (1993). A Fear-Avoidance Beliefs
Pain, 52, 157–68.
Wagner, J. L., Connelly, M. A., Brown, R. T., Taylor, L., Rittle, C. and Cloues, B. (2004). Predictors of
social anxiety in children and adolescents with sickle cell disease. Journal of Clinical Psychology in
Medical Settings, 11, 243–52.
Watt, M. C., O’Connor, R., Stewart, S., Moon, E. and Terry, L. (2008). Specificity of childhood learning
experiences related to anxiety sensitivity and illness/injury sensitivity: implications for health anxiety
and pain. Journal of Cognitive Psychotherapy, 22, 128–42.
Watt, M. C. and Stewart, S. H. (2000). Anxiety sensitivity mediates the relationships between childhood
learning experiences and elevated hypochondriacal concerns in young adulthood. Journal of
Psychosomatic Research, 49, 107–18.
Watt, M. C., Stewart, S. H. and Cox, B. J. (1998). A retrospective study of the learning history origins of
anxiety sensitivity. Behavior Reseach and Therapy, 36, 505–25.
Watt, M. C., Stewart, S. H., Lefaivre, M. J. and Uman, L. S. (2006). A brief cognitive-behavioral approach
to reducing anxiety sensitivity decreases pain-related anxiety. Cognitive Behaviour Therapy,
35, 248–56.
Zinbarg, R. E., Barlow, D. H. and Brown, T. A. (1997). Hierarchical structure and general factor saturation
of the anxiety sensitivity index. Psychological Assessment, 9, 277–84.
Zvolensky, M. J., Goodie, J. L., McNeil, D. W., Sperry, J. A. and Sorrell, J. T. (2001). Anxiety sensitivity in
the prediction of pain-related fear and anxiety in a heterogeneous chronic pain population. Behaviour
Research and Therapy, 39, 683–96.
Chapter 13
Processes Underlying the Relation

between Catastrophizing and Chronic
Pain: Implications for Intervention
Michael J.L. Sullivan and Marc O. Martel
13.1 Introduction
Over the past two decades, pain catastrophizing has emerged as one of the most robust psycho-
logical predictors of pain-related outcomes (Edwards et al. 2006a; Sullivan et al. 2001b; Turk et al.
1983; Weissman-Fogel et al. 2008). Hundreds of studies have documented associations between
pain catastrophizing and adverse pain outcomes, including heightened pain intensity, emotional
distress, and disability (Edwards et al. 2006a; Keefe et al. 2004; Sullivan et al. 2001b; Turk and
Okifuji 2002).
Increasingly, researchers have turned their attention to questions concerning the mechanisms
by which pain catastrophizing impacts on pain outcomes (Seminowicz and Davis 2006; Sullivan
2008; Turner and Aaron 2001). Research in this area has identified psychological, interpersonal
(Cano 2004), physiological (Wolff et al. 2008), and neuroanatomical (Gracely et al. 2004) corre-
lates of pain catastrophizing that might explain how pain catastrophizing impacts on pain experi-
ence. The identification of the mechanisms that link pain catastrophizing to pain outcomes has
both clinical and theoretical implications. From a clinical perspective, understanding the proc-
esses by which pain catastrophizing influences the experience or expression of pain might point
to new avenues for intervention that could reduce the suffering and burden of persistent pain
conditions. From a theoretical perspective, understanding how pain catastrophizing influences
pain might contribute to the elaboration or refinement of conceptual frameworks that address
the linkages between psychology and physiology in the generation of pain experience.
This chapter summarizes research that has addressed the mechanisms implicated in the rela-
tion between pain catastrophizing and pain outcomes. Given the volume of research that has been
conducted in this area, the review of the literature is intended to be illustrative as opposed to
exhaustive. The chapter ends with a discussion of the clinical and theoretical implications of the
research that has accumulated to date.
13.2 Catastrophizing: the construct

Pain catastrophizing has been broadly defined as an exaggerated negative orientation to actual or
anticipated pain comprising elements of rumination, magnification and helplessness (Sullivan
et al. 2001b). In 1995, Sullivan et al. (1995) suggested that different elements of catastrophizing
might be related to primary and secondary appraisal processes (e.g. (Lazarus and Folkman 1984).
At least at a descriptive level, magnification (e.g. exaggerating the threat value of pain) and rumi-
nation (e.g. excessive focus on pain-related stimuli) overlap with the defining features of primary
Fig. 13.1 The mindset of catastrophizing. Original artwork produced by Stephen Read.
appraisals (e.g. evaluating the threat value of a stimulus) (Sullivan et al. 1995). The helplessness
dimension of catastrophizing overlaps with the defining features of secondary appraisals (e.g.
evaluation of one’s ability to effectively deal with the stress situation).
Although the appraisal conceptualization of pain catastrophizing has been debated, it remains
the view espoused by most researchers in this area of research (Jensen et al. 1991b; Keefe et al.
1999; Severeijns et al. 2004; Turner and Aaron 2001). The appraisal conceptualization of pain
catastrophizing places the construct within a cognitive theory framework (Beck et al. 1978; Turk
1996). In its most general form, cognitive theory predicts that exaggerated threat appraisals will
lead to emotional reactions such as anxiety or fear (Lazarus and Folkman 1984). Vlaeyen and his
colleagues have elaborated a cognitive-behavioural model of pain-related disability where pain
catastrophizing is viewed as a key factor in the development of problematic pain outcomes
(Leeuw et al. 2007; Vlaeyen and Linton 2000). The model predicts that catastrophic thinking
following the onset of pain will contribute to heightened fears of movement and increased hyper-
vigilance to pain symptoms. In turn, fear is expected to lead to avoidance or escape of activity that
might be associated with pain (Vlaeyen and Linton 2000). The model is recursive such that
increased pain symptoms, distress and disability become the input for further catastrophic or
alarmist thinking (Vlaeyen and Linton 2000).
It has also been suggested that pain catastrophizing might serve as an interpersonal coping
function (Sullivan et al. 2000). The Communal Coping Model of pain catastrophizing draws on
theoretical perspectives addressing the interpersonal objectives of coping (Coyne and Fiske 1992;
Lackner and Gurtman 2004; Lyons et al. 1995; Taylor 2000). According to the Communal Coping
Model, the pain expressions of high pain catastrophizers serve a social communicative function
aimed at maximizing the probability that distress will be managed within a social/interpersonal
context (Sullivan et al. 2000). Sullivan et al. (2001) suggested that high pain catastrophizers might
engage in exaggerated pain expression in order to maximize proximity, or to solicit assistance or
empathic responses from others. Pain catastrophizers’ expressive pain displays might also be used
to induce others to alter their expectations, reduce performance demands or as a means of man-
aging interpersonal conflict.
CATASTROPHIZING AND CHRONIC PAIN 253
Although the coping style of high pain catastrophizers might appear maladaptive, it is impor-
tant to consider that a communal coping style might only become truly maladaptive under
chronic pain or chronic illness conditions. In response to acute pain, exaggerated pain displays
might result in a precarious, but sustainable, balance between satisfying support or affiliative
needs, and increasing distress. Under acute pain conditions, overall benefits may outweigh costs,
and reinforcement contingencies (e.g. increased support, attention, empathic responses) may
actually serve to maintain the expressive style of high pain catastrophizers. When conditions
become chronic, this balance may be disrupted such that costs begin to outweigh benefits.
Others’ responses may become increasingly negative when distress displays extend over time
(Cano 2004). The disrupted balance may find expression as increased interpersonal conflict,
social rejection and depression (Keefe et al. 2003).
13.3 Processes linking pain catastrophizing to adverse

pain outcomes
An appraisal conceptualization of pain catastrophizing suggests several cognitive and emotional
processes through which pain catastrophizing might impact on pain experience. These might
include expectancies, attentional mechanisms, emotional distress states and the mobilization
of distress reduction strategies (i.e. coping strategies). Research examining the role of these vari-
ables as determinants of the relation between pain catastrophizing and pain outcomes will be
briefly reviewed.
13.3.1 Expectancies
Expectancies are essentially an individual’s predictions about the future. In pain research,
researchers have distinguished between ‘response expectancies’ and ‘efficacy expectancies’
(Bandura 1977; Kirsch 1985). Predictions about non-volitional responses to a particular stimulus
are referred to as ‘response expectancies’. Efficacy expectancies (e.g. self-efficacy) refer to the
confidence individuals’ have that they possess the ability to successfully execute the behaviour
required to yield a desired outcome (Bandura 1977). Considerable research has shown that there
is a high degree of concordance between pain expectancies and pain experience (Jensen et al.
1991a; Lacker et al. 1996).
Response expectancies for pain have been discussed as a significant determinant of actual pain
experience (Crombez et al. 1996; Jensen et al. 1991a). For example, the powerful analgesic effects
of placebos have been discussed in terms of expectancy manipulations (Pollo et al. 2001; Whalley
et al. 2008). It has been suggested that many psychological interventions for pain management
may exert their effects, at least in part, through their influence on pain expectancies (Milling et al.
2006; Milling et al. 2007).
Research has provided support for a relation between pain catastrophizing and response expect-
ancies. In an experimental study, Sullivan et al (Sullivan et al. 2001a) reported that pain catastro-
phizing was associated with expectancies for heightened pain and emotional distress. Van Damme
et al (Van Damme et al. 2002) also found a significant relation between pain catastrophizing
and pain expectancies and suggested that the pain expectancies of high pain catastrophizers
might promote hypervigilance to pain signals. Not only do high pain catastrophizers expect to
experience more pain, but there are findings to suggest that high pain catastrophizers fail to
correct their pain expectancies in the face of disconfirming evidence (Crombez et al. 2002;
Van Damme et al. 2002).
A number of investigations have shown a close association between self-efficacy expectancies
and pain catastrophizing (Keefe et al. 1997b; Thorn et al. 2007). For example, scale items that
assess self-efficacy for coping with pain have frequently loaded on the same factor as scale items
that assess catastrophic thinking (Sullivan et al. 2001b). Albeit a close association, the two con-
structs do not appear to be redundant. One investigation revealed that self-efficacy prospectively
predicted pain behaviour and activity avoidance, even when controlling for pain catastrophizing
(Asghari and Nicholas 2001). In a recent study of patients with osteoarthritis, Shelby et al. (2008)
found that domain-specific self-efficacy mediated the relation between pain catastrophizing and
pain intensity.
There are indications that the impact of expectancies on pain outcomes might be unmediated
(at least with respect to psychological variables). Kirsch (1985) has proposed that response expect-
ancies may represent one of the most basic psychological variables. This position has significant
implications for explicating the underlying basis of several psychological determinants of pain
experience. Expectancies may represent the final common pathway of several psychological influ-
ences on pain perception.
Research suggests that similar brain regions might modulate the effects of pain catastrophizing
and pain expectancies (Gracely et al. 2004; Ploghaus et al. 2003; Ploghaus et al. 1999). In addition,
there are indications that catastrophizing and expectancies might exert influence on endogenous
opioid mechanisms (Peyron et al. 2000). For example, it has been shown that the analgesic effects
of high self-efficacy can be blocked with naloxone (i.e. an opioid antagonist) (Bandura et al.
1987). High levels of pain catastrophizing have been associated with poorer response to opioids
for both clinical and experimental pain (Fillingim et al. 2005; Haythornthwaite et al. 2003;
Jacobsen and Butler 1996).
13.3.2 Attention
It has been suggested that pain catastrophizing might impact on pain experience by increasing
attention to pain sensations. It has long been established that increased attention to pain sensa-
tions augments the intensity of perceived pain (Arntz et al. 1991; Bushnell et al. 2004; Eccleston
and Crombez 1999; McCracken 1997). A number of early studies provided data indicating that
attention diversion strategies such as distraction were less effective when used by individuals with
high levels of pain catastrophizing (Heyneman et al. 1990; Spanos et al. 1979; Sullivan et al. 1995).
Additionally, several studies have reported that among the three subscales of the Pain
Catastrophizing Scale (i.e. rumination, magnification, helplessness), rumination was the best
predictor of pain intensity and pain-related disability (Devoulyte and Sullivan 2003; Sullivan and
Neish 1998; Sullivan et al. 1998).
One study showed that when individuals were asked to suppress thought about an upcoming
pain procedure, individuals with high pain catastrophizing scores reported experiencing
more pain-related thought intrusions than individuals with low pain catastrophizing scores
(Sullivan et al. 1997). In an elegant series of studies, Van Damme et al (Van Damme et al. 2002;
Van Damme et al. 2004) reported findings suggesting that pain catastrophizing might be associ-
ated with an attentional disengagement deficit. The results of the latter studies indicated that
pain catastrophizing does not necessarily lead individuals to be more vigilant to pain, but once
attention has been oriented to a pain stimulus, pain catastrophizing appears to interfere with
disengagement from the pain stimulus.
Neuroimaging research has shown that focusing attention on pain may activate a distributed
network of brain regions, including prefrontal and parietal areas, parts of the anterior cingulate
cortex, and the thalamus (Bushnell et al. 2004; Derbyshire et al. 1997; Peyron et al. 2000). During
painful stimulation, some regions of the ‘attentional network’ have been shown to be significantly
more activated in high pain catastrophizers, particularly the dorsolateral prefrontal cortex, the
anterior cingulate cortex, and the inferior parietal cortex (Gracely et al. 2004; Seminowicz and
Davis 2006). These findings provide neural evidence that attentional mechanisms might account,
at least in part, for the relation between catastrophic thinking and pain experience (Seminowicz
and Davis 2006; Sullivan et al. 2001b).
13.3.3 Emotion
A basic tenet of cognitive theories of emotion is that negative cognitions can lead to negative emo-
tions (Banks and Kerns 1996; Beck et al. 1978; Lazarus and Folkman 1984). Researchers have
appealed to variations of this general framework to understand the relation between catastrophic
thinking and negative emotional reactions (Turner and Aaron 2001; Vlaeyen and Linton 2000).
The relations among pain catastrophizing, fear, and depression have been the focus of numerous
investigations (Keefe et al. 2005; Sullivan and D’Eon 1990). Research has been consistent in show-
ing that measures of catastrophic thinking are significantly correlated with measures of depres-
sion, anxiety, and fear (Borsbo et al. 2008; Drahovzal et al. 2006; Edwards et al. 2006a; Edwards
et al. 2006b; Leeuw et al. 2007). Keefe et al (Keefe et al. 1989) reported that pain catastrophizing
prospectively predicted depressive symptoms in a sample of individuals with arthritis. The
pattern of findings that has emerged suggests that catastrophic thinking might contribute to the
development or maintenance of anxiety, fear or depression associated with pain.
The study of the relation between emotion and pain dates back several decades (Craig 1989;
Schwarz 1962). There is a sizeable literature that has examined the relation between trait meas-
ures of emotional distress and pain outcomes. Numerous investigators have reported significant
cross-sectional and prospective relations between trait measures of depression, anxiety, fear,
anger, and heightened pain experience (Banks and Kerns 1996; Leeuw et al. 2007; Rudy et al.
1988; Sullivan and Neish 1998; Turk 1996; Turk and Okifuji 2002; Vlaeyen et al. 1995). For exam-
ple, Smith and Zautra (2008) reported that anxiety was prospectively related to heightened pain
intensity in a sample of women with arthritis (Smith and Zautra 2008). Carroll et al (2004)
reported that depressive symptoms might increase susceptibility to exacerbation of musculo-
skeletal pain symptoms (Carroll et al. 2004).
Fewer studies have addressed the role of situation-specific or experimentally induced emo-
tional distress on responses to painful stimulation. Findings from experimental studies are not
entirely consistent with the pattern of findings using trait measures of emotional distress. For
example, Meagher et al (Meagher et al. 2001) examined the effects of viewing emotional slides
prior to participating in an experimental pain procedure. Their findings indicated that viewing
slides of fear or disgust resulted in a decrease as opposed to an increase in pain intensity. However,
consistent with the research using trait measures of emotional distress, Carter et al. (2002)
reported that experimental induction of negative emotions (i.e. anxiety, depression) led to
increased pain severity during a cold pressor task. Tang et al. (2008) reported that listening to
sad music led to more intense pain experience and lower pain tolerance in chronic back pain
patients. Thus, although the research on the effects of situation-specific negative mood on pain is
not as consistent as the literature using trait measures of emotional distress, the findings point to
a possible hyperalgesic effect of emotional distress in both healthy individuals and chronic pain
patients.
Studies using functional brain imaging techniques have identified a number of brain areas
responsible for producing emotional/affective responses associated with pain, including feelings
of unpleasantness and distress. For example, studies have been consistent in showing that painful
stimulation leads to increased neural activity in the anterior cingulate and insular cortices, both
part of the limbic system (for a review, see Apkarian et al. 2005). It is generally assumed that
neural activity in limbic areas contributes to heightened pain experience by increasing the
emotional valence attributed to pain sensations.
Recent efforts have been made to examine the neural mechanisms underlying the effects of
emotional states on pain processing. For example, Phillips et al (Phillips et al. 2003) have shown
that experimentally induced negative mood can enhance neural activity in cingulate and insular
cortices during visceral stimulation, leading to increased levels of pain-related discomfort.
Similarly, Ploghaus et al (2001) have shown that experimentally induced anxiety can lead to
hyperalgesic responses and increased neural activity in a number of brain areas associated with
pain processing. Specifically, it has been shown that high levels of anxiety prior to painful heat
stimulation can increase activity in the medial prefrontal cortex, the anterior cingulate cortex,
and parts of the hippocampal formation. These areas are considered to be directly involved in the
amplification of pain experience and provide a neural basis for the effects of emotion on pain
(Schweinhardt et al. 2008; Tracey and Mantyh 2007).
There is reason to believe that pain catastrophizing might influence pain experience through
similar neural mechanisms to those involved in the relationship between emotional distress and
pain. During painful stimulation, Seminowicz and Davis (2006) found that pain catastrophizing
was significantly associated with activity in the medial prefrontal cortex, the anterior cingulate
cortex, the insula, and parts of the hippocampal formation. Pain-evoked neural activity in some
of these regions has been associated with negative affect (Phillips et al. 2003; Ploghaus et al. 2001),
suggesting that pain catastrophizing is likely to overlap with other emotional processes in modu-
lating brain responses to pain. These neuroimaging findings also suggest that pain catastrophiz-
ing might increase emotional distress, facilitating nociceptive processing in cortico-cortical
circuits and augmenting the overall pain experience.
13.3.4 Coping
Coping generally refers to the strategies that individuals use to minimize the impact of life stres-
sors on their psychological well-being (Lazarus and Folkman 1984). In the area of pain, strategies
such as distraction, positive self-statements and re-appraisal have been discussed as ‘adaptive’
coping strategies that might reduce the intensity of pain or the emotional distress associated with
pain (Turk et al. 1983; Turk and Okifuji 2002). Early research on coping and pain catastrophizing
led to the suggestion that pain catastrophizing might represent a ‘maladaptive’ coping strategy
(Keefe et al. 1989). This view was challenged by researchers who argued that pain catastrophizing
was neither strategic or goal directed, and should be considered distinct from coping efforts
(Jensen et al. 1991b; Thorn et al. 2003). As noted earlier, the Communal Coping Model proposed
that pain catastrophizing might indeed be strategic, but the goals of this coping orientation might
be more socially relevant than pain relevant (Sullivan et al. 2001b).
Evidence suggests that pain catastrophizers do not differ from non-catastrophizers in the type
or number of ‘adaptive’ pain coping strategies they use. In one of the earliest studies to address
the relation between coping and pain catastrophizing, Spanos et al. (1979) reported that pain
catastrophizers and non-catastrophizers did not differ in the number of coping strategies they
reported using during a cold pressor procedure. Interestingly however, for non-catastrophizers,
there was an association between number of coping strategies and degree of pain reduction. For
pain catastrophizers, number of coping strategies reported was not associated with pain reduc-
tion. Similar findings were subsequently reported by other investigators suggesting that coping
strategies are less effective when used by pain catastrophizers (Heyneman et al. 1990; Sullivan
et al. 1995).
It is interesting to note that individuals with high pre-treatment scores on measures of pain
catastrophizing are less likely to benefit from pain management or disability management inter-
ventions (Sullivan et al. 2005b). Findings such as these have led investigators to suggest that
reductions in pain catastrophizing might be a pre-requisite to the effective use of pain coping
strategies (Thorn et al. 2002, 2007). Consistent with this perspective, a number of studies have
shown that reductions in catastrophizing mediate the outcomes of pain rehabilitation programmes
(Smeets et al. 2006; Spinhoven et al. 2004; Sullivan et al. 2006b, 2005b).
The Communal Coping Model suggests that pain behaviour displays might be used strategi-
cally by high pain catastrophizers as a means of soliciting attention or support from their social
environment (Sullivan et al. 2001b). Pain catastrophizers’ expressive pain displays might also be
used to induce others to alter their expectations, reduce performance demands or as a means of
managing interpersonal conflict (Keefe et al. 1997a). From this perspective, the expressive pain
displays of pain catastrophizers could be construed as a form of coping.
Support for the Communal Coping Model has come primarily from studies showing that the
pain experience and pain expressions of pain catastrophizers are sensitive to social context
(Giardino et al. 2003; Sullivan et al. 2004). The relation between pain catastrophizing and pain
severity is higher when pain patients are living with a spouse or caregiver (Giardino et al. 2003),
and higher levels of pain catastrophizing are associated with higher levels of short term spousal
support (Cano 2004). Sullivan et al (Sullivan et al. 2006c) found that pain catastrophizers’ expres-
sive displays of pain led observers to infer more intense (but not more accurate) pain experience.
Pain catastrophizing has also been associated with negative interpersonal outcomes where the
emotional demands of high pain catastrophizers seem to tax the support resources of significant
others (Boothy et al. 2004; Keefe et al. 2003; Lackner and Gurtman 2004).
Numerous investigations have shown that pain catastrophizing is associated with heightened
expression of pain behaviour (Keefe et al. 2000; Sullivan et al. 2000; Thibault et al. 2008; Vervoort
et al. 2008). The relation between pain catastrophizing and pain behaviour has been shown to
remain significant even when controlling for pain intensity (Thibault et al. 2008). As such, the
heightened pain expressions of pain catastrophizers cannot be explained as simply being the con-
sequence of experiencing more intense pain. Although a relation between pain catastrophizing
and pain behaviour does not confirm that pain behaviour is being used strategically as a form of
coping, there are indirect indications that pain behaviour might serve such a function. In one
study, the presence of an observer led to increases in pain behaviour for pain catastrophizers but
not for non-catastrophizers participating in an experimental pain procedure (Sullivan et al. 2004).
In the latter study, increases in pain behaviour were associated with decreases in the use of cogni-
tive coping strategies. It has also been shown that emotional disclosure reduces the pain experi-
ence of pain catastrophizers but not that of non-catastrophizers (Sullivan and Neish 1999).
Thus, the literature in this area suggests that pain catastrophizers do not differ from non-
catastrophizers in the types of coping strategies they report using. However, coping strategies
seem to be less effective when used by pain catastrophizers, and pain catastrophizers appear to
benefit less from interventions designed to foster successful adaptation to chronic pain. It is
possible that pain catastrophizers might use pain behaviour as a means of soliciting support to
deal with their distress. Unfortunately, in attaining these interpersonal objectives, pain catastro-
phizers may inadvertently make their pain experience more aversive. Exaggerated display of pain
behaviour may become maladaptive by actually contributing to heightened pain experience,
perhaps through increased attention to pain. In addition, others’ solicitous or reinforcing
responses may serve to trigger, maintain, or reinforce pain catastrophizers’ exaggerated pain
expression (Romano et al. 1995).
13.3.5 Endogenous pain modulation

There are some indications that pain catastrophizing might have a direct impact on endogenous
pain modulation mechanisms. As noted above, research suggests that pain catastrophizers might
benefit less from rehabilitation interventions for chronic pain. There is also research to suggest
that pain catastrophizing might interfere with the effectiveness of pharmacological interventions
for pain. Haythornthwaite et al (2003) reported the findings of a study assessing the efficacy of an
opiate medication for post-herpetic neuralgia. Analyses revealed that initial pain catastrophizing
scores predicted higher post-treatment pain ratings, even when controlling for baseline pain.
Sullivan et al (2008b) reported that catastrophizing was associated with poor response to a topical
analgesic for neuropathic pain. In an experimental study investigating psychological factors
related to pain perception and analgesia, Fillingim et al. (2005) found that catastrophizing in men
was associated with poor overall analgesic responses to intravenous pentazocine.
The mechanisms by which psychological factors interfere with response to analgesics remain
unclear. It has been suggested that individuals high in catastrophizing might produce endogenous
nocebo-like responses due to their negative cognitions (Fillingim et al. 2005). It has also been sug-
gested that catastrophizing might compromise processes involved in descending pain inhibition
(Edwards and Fillingim 2001). For example, in a temporal summation paradigm, Edwards et al.
(2006c) found that individuals with high levels of catastrophizing reported significantly greater
increases in pain ratings than individuals with low levels of catastrophizing during the application
of repeated painful heat stimulations. Similarly, George et al (2006) found that pain catastrophiz-
ing was a significant predictor of increases in pain ratings across repeated noxious heat pulses,
even when controlling for sex and pain-related fear. These findings suggest that pain catastrophiz-
ing may facilitate processes involved in temporal summation of pain or ‘windup’ (Price et al.
2002). The findings also suggest that pain catastrophizing might interfere with descending pain-
inhibitory systems, facilitate neuroplastic changes in the spinal cord during repeated painful
stimulation, subsequently promoting sensitization in the central nervous system.
Other studies have also established a link between pain catastrophizing and the operation of
endogenous pain-modulatory systems. For example, two recently published papers have reported
a negative association between pain catastrophizing and diffuse noxious inhibitory controls, a
psychophysical measure of endogenous pain inhibition (Goodin et al. 2008; Weissman-Fogel
et al. 2008). On the basis of findings such as these, it has been suggested that pain catastrophizing
might directly interfere with the efficacy of endogenous pain-inhibitory mechanisms (Goodin
et al. 2008).
13.4 Processes mediating the relation between pain

catastrophizing and pain outcomes
The literature reviewed in this chapter points to several processes or mechanisms that might
underlie the relation between pain catastrophizing and pain outcomes. Research suggests that
pain catastrophizing is significantly associated with expectancies for heightened pain experience,
increased attention to pain sensations and increased negative mood. In turn, expectancies, atten-
tion and negative mood have been shown to contribute to heightened pain experience. Research
also suggests that pain catastrophizing might interfere with the effectiveness of certain cop-
ing strategies and might be associated with strategies (e.g. distress displays) that inadvertently
contribute to increased pain.
It is important to caution that observed relations between pain catastrophizing and expectan-
cies, attention, negative mood and coping cannot be taken as evidence that these are the processes
by which pain catastrophizing impacts on pain experience. In order to determine which of these
variables underlie the relation between pain catastrophizing and pain outcomes, it is necessary to
examine the degree to which the catastrophizing-pain relation is diminished when controlling,
experimentally or statistically, for potential determinants (or mediators).
Unfortunately, few studies have conducted mediational analyses for variables that might under-
lie the relation between pain catastrophizing and pain outcomes. Still, the available research
permits some speculation about potential mediational candidates. Research addressing the role of
expectancies in mediating the relation between pain catastrophizing and pain has yielded mixed
findings. Sullivan et al. (2001a) reported that expectancies for pain experience prior to an experi-
mental pain procedure partially mediated the relation between catastrophizing and pain. In a
recent study of patients with osteoarthritis, Shelby et al (2008) reported that self-efficacy fully
mediated of the relation between catastrophizing and pain in patients with osteoarthritis.
The research on the relation between pain catastrophizing and attention to pain is strongly sug-
gestive that attention might mediate the catastrophizing-pain relation. Catastrophizing appears
to be associated with a propensity to focus excessively on pain, and might also be associated with
a deficit in mental control over pain-related stimuli such that individuals high in catastrophizing
will have more difficulty disengaging their attention from a pain stimulus. Although the bulk of
this research has been conducted within experimental paradigms, there is little basis for arguing
that different processes might operate in patients with chronic pain. In chronic pain patients, pain
catastrophizing has been shown to correlate significantly with self-reported vigilance to pain
symptoms (Crombez et al. 2004). Surprisingly, research suggests that pain catastrophizing might
mediate the relation between vigilance and pain, as opposed to vigilance mediating the relation
between pain catastrophizing and pain (Crombez et al. 2004).
Questions concerning the role of emotional distress as a mediator of the relationship between
catastrophizing and pain have been addressed in several investigations. Overall, the findings of
this research do not support the mediating role of emotional distress. Many investigations have
reported findings suggesting that instead of emotional distress mediating the relation between
pain catastrophizing and pain outcomes, pain catastrophizing might mediate the relation between
emotional distress and pain. Geisser et al. (1995) found that pain catastrophizing mediated the
relation between depression and pain. Lackner and Quigley (2005) reported that pain catastro-
phizing mediated the relation between worry and pain in individuals with irritable bowel
syndrome. In a study using experimental pain and mood induction, Bartley and Rhudy (2008)
found no evidence that pain catastrophizing exerted its effects on pain indirectly through mood.
Thus, it appears that emotional distress might not be a promising candidate as a mediator of
the relation between pain catastrophizing and pain outcomes. Although emotional distress
co-varies with catastrophic thinking to a significant degree, the research does not support the
view that emotional distress is the vehicle through which pain catastrophizing exerts its impact on
pain outcomes.
The role of coping as a mediator of the relation between catastrophizing and pain outcomes has
been more challenging to address. First, the use of ‘adaptive’ coping strategies does not appear to
co-vary with level of pain catastrophizing, although the effectiveness of coping strategies might be
influenced by level of pain catastrophizing. It is possible that pain catastrophizers might use mala-
daptive strategies, such as the expression of pain behaviour, that inadvertently contribute to
adverse interpersonal or pain outcomes. However, if the use of pain behaviour to solicit attention
or support is not driven by conscious intent, it will be difficult to establish whether such coping
efforts are the process by which pain catastrophizing impacts negatively on pain outcomes.
13.5 Treatment implications

Research examining potential mediators of the relation between catastrophizing and pain
outcomes might have important implications for the development of targeted interventions aimed
at reducing pain catastrophizing, or minimizing the negative impact of pain catastrophizing on
pain outcomes. Although numerous treatment studies have been shown to have an effect on
catastrophic thinking, the critical elements of effective treatments for yielding meaningful change
have yet to be identified.
Considerable research supports the view that pain catastrophizing is a modifiable variable
(Keefe et al. 2005; Sullivan et al. 2005a). In the absence of intervention, pain catastrophizing
shows some degree of stability over time (Sullivan et al. 2001b). However, numerous intervention
studies have shown that catastrophic thinking decreases as a result of participation in treatment
aimed at facilitating recovery or adaptation to chronic pain (Jensen et al. 2001; Smeets et al. 2006;
Spinhoven et al. 2004). Many of these studies have pointed to importance of reducing pain cata-
strophizing as a key factor in determining the success of interventions for chronic pain (Spinhoven
et al. 2004; Sullivan et al. 2005b).
Jensen et al (2001) reported that participation in a 3-week (82 hours) multidisciplinary pain
treatment programme led to a 40% reduction in scores on a measure of catastrophizing (Jensen
et al. 2001). Treatment-related changes in pain catastrophizing rose significantly at 6-month
follow-up, but remained below baseline levels. Sullivan et al. (2003) reported a 33% reduction in
catastrophizing scores following participation in a 10-week (10 hours) psychological intervention
(led by psychologists) designed to target psychosocial risk factors for pain and disability (Sullivan
and Stanish 2003). Sullivan et al. (2006a) reported a 43% reduction in catastrophizing following
participation in a 10-week programme (50 hours) consisting of exercise and a psychosocial inter-
vention (led by occupational therapists and physiotherapists) targeting risk factors for pain and
disability. Adams et al. (2007) reported that reductions in pain catastrophizing following a
10-week (50 hours) treatment programme consisting of exercise and a psychosocial intervention
varied as a function of level of chronicity. For patients in the subacute (4 weeks to 3 months) and
early chronic period (3–6 months) of recovery, pain catastrophizing scores showed a reduction of
39%. For patients whose condition had become chronic (+6 months), pain catastrophizing scores
decreased by only 10%.
There are also indications that psychological intervention might not be essential to yield reduc-
tions in catastrophic thinking. Sullivan et al (2006a) reported a 24% reduction in pain catastro-
phizing scores following participation in a 10-week (45 hours) physical therapy intervention.
Another study reported a 27% decrease in pain catastrophizing scores following a 4-week (100
hours) functional restoration exercise programme (Sullivan et al. 2008a). Smeets et al. (Smeets
et al. 2006) reported no significant difference in the magnitude of reduction in pain catastrophiz-
ing scores for patients who participated in active physiotherapy (−12%), problem-solving therapy
(−10%) or combined treatment (−10%). In the latter study, each treatment programme consisted
of approximately 11 hours of intervention.
The research conducted to date suggests that catastrophic thinking associated with pain can be
reduced through a variety of means. However, some degree of caution must be exercised in the
interpretation of the results of the studies described above. Studies vary in terms of the nature of
the population being treated (recent onset versus long term disability; low back pain versus whip-
lash), the intensity of treatment (10–100 hours), the insurance context within which clients are
treated (no fault versus tort), and the objectives of the intervention (pain management, functional
improvement, or return to work). Initial values (high versus low) on measures of pain catastro-
phizing will play a role in determining the magnitude of reductions that will be observed and the
relation between reductions in pain catastrophizing and clinical outcomes. In a related manner,
there is currently limited information about the magnitude of reduction in pain catastrophizing
scores that is required to impact in a clinically meaningful manner on pain outcomes.
Education, activity resumption and instruction in self-management skills characterize
the content of most multidisciplinary programmes for the management of chronic pain
(Gatchel et al. 2007). It is not unreasonable to assume that each of these elements might impact
directly or indirectly on catastrophic thinking. As noted earlier, catastrophizing has been dis-
cussed as a multidimensional construct comprising rumination, magnification and helplessness.
Intervention techniques that impact on any of these dimensions might yield therapeutic benefit.
Education might permit individuals to re-evaluate or re-appraise the degree of threat they associ-
ate with their condition or their participation in activity (Moseley 2004; Turk 2004). Participation
in exercise or other physical activity might yield benefit by reducing the cognitive resources
that can be allocated to catastrophic thinking. Activity participation and instruction in self-
management skills might increase self-efficacy and, in turn, reduce the helplessness dimension of
catastrophizing.
In recent years, targeted treatments have been designed specifically to reduce catastrophic
thinking (Sullivan et al. 2006a; Thorn et al. 2002). Some of these have taken the form of group
interventions using cognitive-behavioural techniques such as thought monitoring, cognitive
re-structuring and re-appraisal (Thorn et al. 2002). Other interventions have taken the form of
individual treatment where behavioural activation and life role resumption are used to augment
the impact of cognitive-behavioural techniques for pain-related disability (Sullivan et al. 2006a).
To date, limited evidence is available to address whether these targeted interventions have added
value for reducing levels of catastrophic thinking.
Research reviewed in this chapter suggests additional avenues that might be considered in
efforts to reduce catastrophic thinking. The research suggests that intervention techniques spe-
cifically targeting expectancies and attention might augment the impact of treatment programmes
for chronic pain. However, expectancies and attention are broad constructs and might represent
the final common pathways of numerous psychological processes. Indeed, it is difficult to point
to intervention dimensions that will not in some way influence expectancies or attention. The
challenge for future research appears to be the development of interventions that will yield reduc-
tions in catastrophic thinking of sufficient magnitude to have a meaningful impact on pain and
rehabilitation outcomes.
There is a basis for suggesting that activity-based interventions targeting catastrophic thinking
might have advantages over strictly cognitive approaches to reducing catastrophic thinking. The
literature on disengagement deficits suggests that high pain catastrophizers might be limited in
their ability to make effective use of cognitive techniques that might require disengagement such
as distraction or cognitive-restructuring. Activity-based interventions might be useful in reducing
the frequency of catastrophic thinking without appealing to disengagement abilities. The impov-
erished stimulus environment that characterizes sedentary or passive activities might leave the
cognitive content of the high catastrophizer saturated with negative or pessimistic thoughts. Since
activity participation requires some degree of cognitive resource investment, activity participa-
tion might result in fewer attentional resources devoted to catastrophic thinking, without having
to rely on attentional disengagement abilities.
The research that has been cited in support of the Communal Coping Model of pain catastro-
phizing suggests that including techniques designed to target interpersonal or communicative
aspects of pain might hold promise. Disclosure interventions might play a role in reducing pain
severity and as well as reducing pain behaviour (Keefe et al. 2008; Sullivan and Neish 1999).
Increased awareness of emotional or attachment needs and the development of more effective
communication strategies might reduce reliance on the display of pain behaviour to solicit atten-
tion, care or proximity (Cano 2004; Ciechanowski et al. 2003; Gauthier et al. 2008).
The development of more targeted interventions for the reduction of pain catastrophizing
might also have implications for the effectiveness of pharmacotherapeutic interventions for pain.
As discussed earlier, pain catastrophizing has been shown to interfere with the effectiveness
of analgesics. Brief interventions aimed at reducing catastrophic thinking might improve responses
to analgesic medication. Such interventions might be particularly important in domains associ-
ated with treatment resistance such as neuropathic pain or fibromyalgia. Interventions designed
to reduce catastrophic thinking might also prevent the development of chronic pain following
surgical interventions (Forsythe et al. 2008).
The research conducted on the neurophysiological correlates of pain catastrophizing raises
interesting questions about the bi-directionality of relations between cognitive processes and
central mechanisms involved in nociceptive processing. Psychological interventions that yield
reductions in catastrophizing might also impact directly or indirectly on activation patterns of
brain centres associated with pain perception, and might even modify the functioning of endo-
genous opioid mechanisms. It follows that interventions that target central mechanisms of pain
control might also influence catastrophic thinking.
13.6 Summary
Over the past two decades pain catastrophizing has emerged as one of the most robust psycho-
logical predictors of adverse pain outcomes. Recent research has begun to address the psycho-
logical and neurophysiological mechanisms that might underlie the relation between
catastrophizing and pain outcomes. There are indications that psychological variables related to
expectancies and attention might account, at least in part, for the relation between catastrophiz-
ing and pain outcomes. Some research suggests that catastrophic thinking might also impact
directly on central mechanisms of pain control. Challenges for the future include the develop-
ment of more targeted interventions for reducing catastrophic thinking. Future research on the
relations between catastrophizing and central mechanisms of pain control might have implica-
tions for theoretical models that address the linkages between psychology and physiology in the
modulation of pain experience.
References
Adams, H., Ellis, T., Stanish, W. D., et al. Psychosocial factors related to return to work following
rehabilitation of whiplash injuries. J Occup Rehabil (2007) 17(2):305–15.
Apkarian, A. V., Bushnell, M. C., Treede, R. D., et al. Human brain mechanisms of pain perception and
regulation in health and disease. Eur J Pain (2005) 9(4):463–84.
Arntz, A., Dreesen, L., and Merckelbach, H. Attention, not anxiety, influences pain. Behav Res Ther (1991)
29:41–50.
Asghari, A., and Nicholas, M. K. Pain self-efficacy beliefs and pain behavior: A prospective study. Pain
(2001) 94:85–100.
Bandura, A. Self-Efficacy: The Exercise of Control. New York: W.H. Freeman, 1977.
Bandura, A., O’Leary, A., Taylor, C. B., et al. Perceived self-efficacy and pain control: Opioid and
nonopioid mechanims. Journal of Personality and Social Psychology (1987) 53:563–71.
Banks, S., and Kerns, R. Explaining high rates of depression in chronic pain: A diathesis-stress formulation.
Psychological Bulletin (1996) 119:95–110.
Bartley, E. J., and Rhudy, J. L. The influence of pain catastrophizing on experimentally induced emotion
and emotional modulation of nociception. J Pain (2008) 9(5):388–96.
Beck, A. T., Rush, A. J., Shaw, B. F., et al. Cognitive Therapy for Depression. New York: Guilford, 1978.
Boothy, J., Thorn, B., Overduin, L., et al. Catastrophizing and perceived partner responses to pain. Pain
(2004) 109:500–506.
Borsbo, B., Peolsson, M., and Gerdle, B. Catastrophizing, depression, and pain: correlation with and
influence on quality of life and health - a study of chronic whiplash-associated disorders. J Rehabil Med
(2008) 40(7):562–9.
Bushnell, M. C., Villemure, C., and Duncan, G. H. Psychophysical and neurophysiological studies of pain
modulation by attention. In D. D. P. a. M. C. Bushnell., ed., Psychological methods of pain control: Basic
science and clinical perspectives. IASP Press., 2004.
Cano, A. Pain catastrophizing and social support in married individuals with chronic pain: The moderating
role of pain duration. Pain (2004) 110:656–64.
Carroll, L. J., Cassidy, J. D., and Cote, P. Depression as a risk factor for onset of an episode of troublesome
neck and low back pain. Pain (2004) 107(1–2):134–9.
Carter, L. E., McNeil, D. W., Vowles, K. E., et al. Effects of emotion of pain reprots, tolerance and
physiology. Pain Res Manag (2002) 7:21–30.
Ciechanowski, P., Sullivan, M., Jensen, M., et al. The relationship of attachment style to depression,
catastrophizing and health care utilization in patients with chronic pain. Pain (2003)
104(3):627–37.
Coyne, J., and Fiske, V. Couples coping with chronic illness. In T. Akamatsu, J. Crowther, S. Hobfoll, et al.
(eds.), Family Health Psychology, pp. 129–49. Washington DC: Hemisphere, 1992.
Craig, K. Emotional aspects of pain. In R. Melzack and P. Wall (eds.), Textbook of Pain, pp. 220–30.
Edinburgh: Churchill Livingstone, 1989.
Crombez, G., Eccleston, C., Van den Broeck, A., et al. Hypervigilance to pain in fibromyalgia: the
mediating role of pain intensity and catastrophic thinking about pain. Clin J Pain (2004)
20(2):98–102.
Crombez, G., Eccleston, C., Vlaeyen, J. W., et al. Exposure to physical movements in low back pain
patients: restricted effects of generalization. Health Psychol (2002) 21(6):573–8.
Crombez, G., Vervaet, L., Baeyens, F., et al. Do pain expectancies cause pain in chronic low back patients?
A clinical investigation. Behav Res Ther (1996) 34(11–12):919–25.
Derbyshire, S. W. G., Jones, A. K., Gyulai, F., et al. Pain processing during three levels of noxious
stimulation produces differential patterns of central activity. Pain (1997) 73:431–45.
Devoulyte, K., and Sullivan, M. J. Pain catastrophizing and symptom severity during upper respiratory
tract illness. Clinical Journal of Pain (2003) 19(2):125–33.
Drahovzal, D. N., Stewart, S. H., and Sullivan, M. J. Tendency to catastrophize somatic sensations: pain
catastrophizing and anxiety sensitivity in predicting headache. Cogn Behav Ther (2006) 35(4):226–35.
Eccleston, C., and Crombez, G. Pain demands attention: a cognitive-affective model of the interruptive
function of pain. Psychol Bull (1999) 125(3):356–66.
Edwards, R. R., Bingham, C. O., 3rd, Bathon, J., et al. Catastrophizing and pain in arthritis, fibromyalgia,
and other rheumatic diseases. Arthritis Rheum (2006a) 55(2):325–32.
Edwards, R. R., and Fillingim, R. B. Effects of age on temporal summation and habituation of thermal
pain: clinical relevance in health older and yonger adults. J Pain (2001) 2:307–317.
Edwards, R. R., Smith, M. T., Kudel, I., et al. Pain-related catastrophizing as a risk factor for suicidal
ideation in chronic pain. Pain (2006b) 126(1–3):272–9.
Edwards, R. R., Smith, M. T., Stonerock, G., et al. Pain-related catastrophizing in healthy women is
associated with greater temporal summation of and reduced habituation to thermal pain. Clin J Pain
(2006c) 22(8):730–7.
Fillingim, R. B., Hastie, B. A., Ness, T. J., et al. Sex-related psychological predictors of baseline pain
perception and analgesic responses to pentazocine. Biol Psychol (2005) 69(1):97–112.
Forsythe, M. E., Dunbar, M. J., Hennigar, A. W., et al. Prospective relation between catastrophizing and
residual pain following knee arthroplasty: two-year follow-up. Pain Res Manag (2008) 13(4):335–41.
Gatchel, R., Peng, Y. B., Peters, M. L., et al. The biopsychosocial approach to chronic pain: Scientific
advances and future directions. Psychol Bull (2007) 133:581–624.
Gauthier, N., Thibault, P., and Sullivan, M. J. Individual and relational correlates of pain-related empathic
accuracy in spouses of chronic pain patients. Clin J Pain (2008) 24(8):669–77.
Geisser, M. E., Robinson, M. E., Keefe, F. J., et al. Catastrophizing, depression and the sensory, affective
and evaluative aspects of chronic pain. Pain (1995) 59:79–83.
George, S. Z., Wittmer, V. T., Fillingim, R. B., et al. Fear-avoidance beliefs and temporal summation of
evoked thermal pain influence self-report of disability in patients with chronic low back pain. J Occup
Rehabil (2006) 16(1):95–108.
Giardino, N., Jensen, M., Turner, J., et al. Social environment moderates the association between
catastrophizing and pain among persons with a spinal cord injury. Pain (2003) 106:19–25.
Goodin, B. R., McGuire, L., Allshouse, M., et al. Associations between catastrophizing and endogenous
pain-inhibitory processes: sex differences. J Pain (2008) 10:180–90.
Gracely, R. H., Geisser, M. E., Giesecke, T., et al. Pain catastrophizing and neural responses to pain among
persons with fibromyagia. Brain (2004) 127:835–43.
Haythornthwaite, J., Clark, M., Pappagallo, M., et al. Pain coping strategies play a role in the persistence of
pain in post-herpetic neuralgia. Pain (2003) 106:453–60.
Heyneman, N. E., Fremouw, W. J., Gano, D., et al. Individual differences and the effectiveness of different
coping strategies for pain. Cog Ther Res (1990) 14:63–77.
Jacobsen, P. B., and Butler, R. W. Relation of cognitive coping and catastrophizing to acute pain and
analgesic use following breast cancer surgery. Journal of Behavioral Medicine (1996) 19(1):17–29.
Jensen, M. P., Turner, J. A., and Romano, J. M. Self-efficacy and outcome expectancies: relationship to
chronic pain coping strategies and adjustment. Pain (1991a) 44(3):263–9.
Jensen, M. P., Turner, J. A., and Romano, J. M. Changes in beliefs, catastrophizing, and coping are
associated with improvement in multidisciplinary pain treatment. Journal of Consulting and Clinical
Psychology (2001) 69:655 662.
Jensen, M. P., Turner, J. A., Romano, J. M., et al. Coping with chronic pain: a critical review of the
literature. Pain (1991b) 47(3):249–83.
Keefe, F., Lefebvre, J., and Smith, S. Catastrophizing research: Avoiding conceptual errors and maintaining
a balnaced perspective. Pain Forum (1999) 8:176–80.
Keefe, F., Lipkus, I., Lefebvre, J., et al. The social contex of gastrointestinal cancer pain: A preliminary
study examining the relation of patient pain catastrophizing to patient perceptions of social support
and caregiver stress and negative responses. Pain (2003) 103:151–56.
Keefe, F., Rumble, M., Scipio, C., et al. Psychological aspects of persistent pain: Current state of the science.
Journal of Pain (2004) 5:195–211.
Keefe, F. J., Abernethy, A. P., and Campbell, L. C. Psychological approaches to understanding and treating
disease-related pain. . Annual Review of Psychology (2005) 56:601–30.
Keefe, F. J., Anderson, T., Lumley, M., et al. A randomized, controlled trial of emotional disclosure in
rheumatoid arthritis: can clinician assistance enhance the effects? Pain (2008) 137(1):164–72.
Keefe, F. J., Brown, G. K., Wallston, K. A., et al. Coping with rheumatoid arthritis pain: catastrophizing as a
maladaptive strategy. Pain (1989) 37(1):51–6.
Keefe, F. J., Kashikar-Zuck, S., Robinson, E., et al. Pain coping strategies that predict patients’ and spouses’
ratings of patients’ self-efficacy. Pain (1997a) 73(2):191–9.
Keefe, F. J., Lefebvre, J. C., Egert, J. R., et al. The relationship of gender to pain, pain behavior, and
disability in osteoarthritis patients: the role of catastrophizing. Pain (2000) 87(3):325–34.
Keefe, F. J., Lefebvre, J. C., Maixner, W., et al. Self-efficacy for arthritis pain: relationship to perception
of thermal laboratory pain stimuli. Arth Care Res (1997b) 10(3):177–84.
Kirsch, I. Response expectancy as a determinant of experience and behavior. Am Psychol (1985) 40:1189–1202.
Lacker, J. M., Carosella, A. M., and Feuerstein, M. Pain expectancies, pain, and functional self-efficacy
expectancies as determinants of disability in patients with chronic low back disorders. J Consult Clin
Psychol (1996) 64(1):212–20.
Lackner, J., and Gurtman, M. Pain catastrophizing and interpersonal problems: A circumplex analysis of
the communal coping model. Pain (2004) 110:597–604.
Lackner, J. M., and Quigley, B. M. Pain catastrophizing mediates the relationship between worry and pain
suffering in patients with irritable bowel syndrome. Behav Res Ther (2005) 43:943–57.
Lazarus, R., and Folkman, S. Stress, appraisal and coping. New York: Springer, 1984.
Leeuw, M., Goossens, M. E., Linton, S. J., et al. The fear-avoidance model of musculoskeletal pain: current
state of scientific evidence. J Behav Med (2007) 30(1):77–94.
Lyons, R., Sullivan, M., Ritvo, P., et al. Relationships in Illness and Disability. Sage, 1995.
McCracken, L. M. ‘Attention’ to pain in persons with chronic pain: a behavioral approach. Behav Ther
(1997) 28:271–84.
Meagher, M. W., Arnau, R. C., and Rhudy, J. L. Pain and emotion: effects of affective picture modulation.
Psychosom Med (2001) 63:79–90.
Milling, L. S., Reardon, J. M., and Carosella, G. M. Mediation and moderation of psychological pain
treatments: response expectancies and hypnotic suggestibility. J Consult Clin Psychol (2006)
74(2):253–62.
Milling, L. S., Shores, J. S., Coursen, E. L., et al. Response expectancies, treatment credibility, and hypnotic
suggestibility: mediator and moderator effects in hypnotic and cognitive-behavioral pain interventions.
Ann Behav Med (2007) 33(2):167–78.
Moseley, G. L. Evidence for a direct relationship between cognitive and physical change during an
education intervention in people with chronic low back pain. Eur J Pain (2004) 8:39–45.
Peyron, R., Laurent, B., and Garcia-Larrea, L. Functional imaging of brain responses to pain. A review and
meta-analysis (2000). Neurophysiol Clin (2000) 30(5):263–88.
Phillips, M. L., Gregory, L. J., Cullen, S., et al. The effect of negative emotional context on neural and
behavioural responses to oesophageal stimulation. Brain (2003) 126(Pt 3):669–84.
Ploghaus, A., Becerra, L., Borras, C., et al. Neural circuitry underlying pain modulation: expectation,
hypnosis, placebo. Trends in Cog Sci (2003) 7:197–200.
Ploghaus, A., Narain, C., Beckmann, C. F., et al. Exacerbation of pain by anxiety is associated with activity
in a hippocampal network. J Neurosci (2001) 21(24):9896–903.
Ploghaus, A., Tracey, I., Gati, S., et al. Dissociating pain from its anticipation in the human brain. Science
(1999) 284:1979–81.
Pollo, A., Amanzio, M., Arslanian, A., et al. Response expectancies in placebo analgesia and their clinical
relevance. Pain (2001) 93(1):77–84.
Price, D. D., Staud, R., Robinson, M. E., et al. Enhanced temporal summation of second pain and its
central modulation in fibromyalgia patients. Pain (2002) 99(1–2):49–59.
Romano, J., Turner, J., Jensen, M., et al. Chronic pain patient-spouse interactions predict patient disability.
Pain (1995) 65:353–60.
Rudy, T., Kerns, R., and Turk, D. Chronic pain and depression: Toward a cognitive-behavioral mediation
model. Pain (1988) 35:129–40.
Schwarz, B. E. Pain and emotion. J Med Soc N J (1962) 59:517.
Schweinhardt, P., Kalk, N., Wartolowska, K., et al. Investigation into the neural correlates of emotional
augmentation of clinical pain. Neuroimage (2008) 40(2):759–66.
Seminowicz, D. A., and Davis, K. D. Cortical responses to pain in healthy individuals depends on pain
catastrophizing. Pain (2006) 120(3):297–306.
Severeijns, R., Vlaeyen, J., and van den Hout, M. Do we need a communal coping model of pain
catastrophizing? An alternative explanation. Pain (2004) 111:226–29.
Shelby, R. A., Somers, T. J., Keefe, F. J., et al. Domain specific self-efficacy mediates the impact of pain
catastrophizing on pain and disability in overweight and obese osteoarthritis patients. J Pain (2008)
9(10):912–19.
Smeets, R. J., Vlaeyen, J. W., Kester, A. D., et al. Reduction of pain catastrophizing mediates the outcome
of both physical and cognitive-behavioral treatment in chronic low back pain. J Pain (2006)
7(4):261–71.
Smith, B. W., and Zautra, A. J. The effects of anxiety and depression on weekly pain in women with
arthritis. Pain (2008) 138:354–61.
Spanos, N. P., Radtke-Bodorik, H. L., Ferguson, J. D., et al. The effects of hypnotic susceptibility,
suggestions for analgesia, and utilization of cognitive strategies on the reduction of pain. Journal of
Abnormal Psychology (1979) 88:282–92.
Spinhoven, P., Ter Kuile, M., Kole-Snijders, A. M., et al. Catastrophizing and internal pain control as
mediators of outcome in the multidisciplinary treatment of chronic low back pain. European Journal of
Pain (2004) 8(3):211–9.
Sullivan, M., Bishop, S., and Pivik, J. The Pain Catastrophizing Scale: Development and validation.
Psychological Assessment (1995) 7:524–32.
Sullivan, M., Feuerstein, M., Gatchel, R. J., et al. Integrating psychological and behavioral interventions to
achieve optimal rehabilitation outcomes. J Occ Rehab (2005a) 15:475–89.
Sullivan, M. J., and Neish, N. The effects of disclosure on pain during dental hygiene treatment: the
moderating role of catastrophizing. Pain (1999) 79(2–3):155–63.
Sullivan, M. J. L. Toward a biopsychomotor conceptualisation of pain. Clin J Pain (2008) 24:281–90.
Sullivan, M. J. L., Adams, A., Horan, S., et al. The role of perceived injustice in the experience of chronic
pain and disability: Scale development and validation. J Occ Rehab (2008a) 18:249–61.
Sullivan, M. J. L., Adams, A., Rhodenizer, T., et al. A psychosocial risk factor targeted intervention for the
prevention of chronic pain and disability following whiplash injury. Phys Ther (2006a) 86:8–18.
Sullivan, M. J. L., Adams, H., Rhodenizer, T., et al. A psychosocial risk factor—targeted intervention for
the prevention of chronic pain and disability following whiplash injury. Phys Ther (2006b) 86(1):8–18.
Sullivan, M. J. L., Adams, H., and Sullivan, M. E. Communicative dimensions of pain catastrophizing:
social cueing effects on pain behaviour and coping. Pain (2004) 107(3):220–6.
Sullivan, M. J. L., and D’Eon, J. L. Relation between catastrophizing and depression in chronic pain
patients. Journal of Abnormal Psychology (1990) 99(3):260–3.
Sullivan, M. J. L., Lynch, M. E., Clark, A. J., et al. Catastrophizing and treatment outcome: Impact on
response to placebo and active treatment outcome. . Contemporary Hypnosis (2008b) 29:129–40.
Sullivan, M. J. L., Martel, M. O., Tripp, D., et al. The relation between catastrophizing and the
communication of pain experience. Pain (2006c) 122(3):282–8.
Sullivan, M. J. L., and Neish, N. R. Catastrophizing, anxiety and pain during dental hygiene treatment.
Community Dentistry and Oral Epidemiology (1998) 26(5):344–9.
Sullivan, M. J. L., Rodgers, W. M., and Kirsch, I. Catastrophizing, depression and expectancies for pain and
emotional distress. Pain (2001a) 91:147–54.
Sullivan, M. J. L., Rouse, D., Bishop, S. R., et al. Thought suppression, catastrophizing and pain. Cog Ther
Res (1997) 21:555–68.
Sullivan, M. J. L., Stanish, W., Waite, H., et al. Catastrophizing, pain, and disability in patients with
soft-tissue injuries. Pain (1998) 77(3):253–60.
Sullivan, M. J. L., and Stanish, W. D. Psychologically based occupational rehabilitation: the Pain-Disability
Prevention Program. Clin J Pain (2003) 19(2):97–104.
Sullivan, M. J. L., Thorn, B., Haythornthwaite, J. A., et al. Theoretical perspectives on the relation between
catastrophizing and pain. Clin J Pain (2001b) 17(1):52–64.
Sullivan, M. J. L., Tripp, D., and Santor, D. Gender differences in pain and pain behavior: The role of
catastrophizing. Cog Ther Res (2000) 24:121–34.
Sullivan, M. J. L., Ward, L. C., Tripp, D., et al. Secondary prevention of work disability: community-based
psychosocial intervention for musculoskeletal disorders. J Occup Rehabil (2005b) 15(3):377–92.
Tang, N. K., Salkovskis, P. M., Hodges, A., et al. Effects of mood on pain responses and pain tolerance: an
experimental study in chronic back pain patients. Pain (2008) 138(2):392–401.
Taylor, S. Biobehavioral responses to stress in females: Tend-and befriend, not fight-or-flight. Psychological
Review (2000) 107:411–29.
Thibault, P., Loisel, P., Durand, M. J., et al. Psychological predictors of pain expression and activity
intolerance in chronic pain patients. . Pain (2008) 139:47–54.
Thorn, B., Boothy, J., and Sullivan, M. Targeted treatment of catastrophizing for the management of
chronic pain. Cognitive Behavior Practice (2002) 9:127–38.
Thorn, B. E., Pence, L. B., Ward, L. C., et al. A randomized clinical trial of targeted cognitive behavioral
treatment to reduce catastrophizing in chronic headache sufferers. J Pain (2007) 8(12):938–49.
Thorn, B. E., Ward, L. C., Sullivan, M. J., et al. Communal coping model of catastrophizing: conceptual
model building. Pain (2003) 106(1–2):1–2.
Tracey, I., and Mantyh, P. W. The cerebral signature for pain perception and its modulation. Neuron
(2007) 55(3):377–91.
Turk, D. Biopsychosocial perspective on chronic pain. In R. Gatchel and D. Turk (eds.), Psychological
Approaches to Pain Management, pp. 138–58. New York: Guilford, 1996.
Turk, D., Meichenbaum, D., and Genest, M. Pain and Behavioral Medicine: A Cognitive-Behavioral
Perspective. New York: Guilford, 1983.
Turk, D., and Okifuji, A. Psychological factors in chronic pain: Evolution and revolution. Journal of
Consulting and Clinical Psychology (2002) 70:678–90.
Turk, D. C. Understanding pain sufferers: the role of cognitive processes. The Spine J (2004) 4:1–7.
Turner, J., and Aaron, L. Pain-related catastrophizing: What is it? Clinical Journal of Pain (2001) 17:65–71.
Van Damme, S., Crombez, G., and Eccleston, C. Retarded disengagement from pain cues: the effects of
pain catastrophizing and pain expectancy. Pain (2002) 100(1–2):111–18.
Van Damme, S., Crombez, G., and Eccleston, C. Disengagement from pain: the role of catastrophic
thinking about pain. Pain (2004) 107(1–2):70–6.
Vervoort, T., Goubert, L., Eccleston, C., et al. The effects of parental presence upon the facial expression of
pain: the moderating role of child pain catastrophizing. Pain (2008) 138(2):277–85.
Vlaeyen, J. W., Kole-Snijders, A. M., Boeren, R. G., et al. Fear of movement/(re)injury in chronic low back
pain and its relation to behavioral performance. Pain (1995) 62(3):363–72.
Vlaeyen, J. W., and Linton, S. J. Fear-avoidance and its consequences in chronic musculoskeletal pain:
a state of the art. Pain (2000) 85(3):317–32.
Weissman-Fogel, I., Sprecher, E., and Pud, D. Effects of catastrophizing on pain perception and pain
modulation. Exp Brain Res (2008) 186(1):79–85.
Whalley, B., Hyland, M. E., and Kirsch, I. Consistency of the placebo effect. J Psychosom Res (2008)
64(5):537–41.
Wolff, B., Burns, J. W., Quartana, P. J., et al. Pain catastrophizing, physiological indexes, and chronic pain
severity: tests of mediation and moderation models. J Behav Med (2008) 31(2):105–14.
Chapter 14
Fear-Avoidance as a Risk Factor for

the Development of Chronic Back Pain
and Disability
Linda Vancleef, Ida Flink, Steven J. Linton, and
Johan Vlaeyen
14.1 Introduction
Chronic musculoskeletal pain syndromes are responsible for enormous costs for healthcare and
society (Linton 1998; Phillips et al. 2008; Picavet and Schouten 2003; Verhaak et al. 1998).
Nowadays, the biopsychosocial perspective on pain offers a good foundation for a better insight
into how pain can become a persistent problem (Fordyce 1976; Turk and Flor 1999). In this per-
spective, pain and pain disability are influenced by the dynamic interaction among biological,
psychological, and social factors.
The present chapter focuses on the role of fear and avoidance in the development and mainte-
nance of chronic low back pain (CLBP). In the following paragraphs, an overview on the aetiol-
ogy of low back pain (LBP), the conceptualization of fear and avoidance, and the development of
fear-avoidance models will be provided. Furthermore, empirical evidence on the role of fear and
avoidance behaviour in chronic pain, stemming from studies conducted in pain patients and in
healthy volunteers, will be reviewed. This chapter will then continue with a discussion on how to
assess fear of pain both at a direct and indirect level. Finally, this chapter will end with perspec-
tives on the cognitive-behavioural management of chronic pain in patients who are characterized
by increased pain-related fear and avoidance behaviour.
14.1.1 Chronic low back pain

LBP is one of the most prevalent complaints in the musculoskeletal apparatus. LBP refers to pain
in the area of the lower back that often radiates to the buttocks or the legs. Recent reviews indi-
cated that about 60–90% of the general population report to have experienced complaints in the
lower back at some point throughout the course of their lives (Picavet and Schouten 2003).
Population-based studies estimate the lifetime prevalence of LBP between 60–80% (Nachemson
et al. 2000; Walker 2000). A study in the Belgian population showed that 41.8% of the respond-
ents indicated to have experienced LBP for at least one day in the past period of 6 months.
Moreover, 8.2% of this group reported that the pain in their lower back was seriously impeding
daily functioning (Goubert et al. 2004a). These figures are comparable to other Western countries
(e.g. Schmidt et al. 2007; Walker et al. 2004).
Only in a minority of people with LBP specific causes such as nerve injury, fractures, inflamma-
tion or malign disease are found to be responsible for the pain (Waddell 2004). In most cases the
pain is denoted as non-specific, meaning that no specific biomedical cause can be found for LBP.
For the majority of back pain patients who seek care and refrain from work, the problem of pain
resolves within a few weeks. These patients return to work and resume their daily activities within
4–6 weeks after the onset of the complaints. However, there is a small subgroup of patients
(5–10% of the total population), in whom back pain persists for longer than 3 months, and who
develop a chronic pain problem (Waddell 2004). Ironically, this relatively small group of back
pain patients is responsible for the largest amount of healthcare and societal costs of back prob-
lems (Phillips et al. 2008; Waddell 2004). In addition, chronic pain is known to have an enormous
impact on the personal and social relations of these chronic back pain patients (Sullivan et al.
2004; Morley and Eccleston 2004).
14.1.2 General explanatory models

Throughout time, several explanatory models have been put forward to explain why a small
group of back pain patients become chronic pain sufferers. Biomedically oriented specialists have
suggested that these patients have more serious impairments than those who resume daily activi-
ties earlier. However, no research supports this assumption. On the contrary, numerous studies
have shown that there is no perfect relationship between impairment, pain, and disability. Patients
with back problems often show no physical injury, and conversely, it happens that persons who
do show physical injury do not report pain or disability (e.g. Jensen et al. 1994). The biomedical
perspective that any pain is the result of structural and biomechanical abnormalities can also not
account for phantom limb pain, where pain is experienced in missing body parts (Giummarra
et al. 2007) or for the large interindividual differences that exist in the way persons experience,
respond to, and cope with pain.
A biopsychosocial approach offers a better insight into how pain can become a persistent prob-
lem (Turk and Flor 1999; Waddell 2004). According to this approach, pain and pain disability are
not only influenced by organic pathology, if present, but also by psychological and social factors.
The interrelationship between the biological, psychological, and social factors, as well as their
influence on the pain experience can be complex. As such, small content overlap does exist
between the various factors, but different processes are assumed in each individual component of
the model (Turk and Gatchel 2002; Waddel 2004). For example, processes and factors that influ-
ence the biomedical aspects of pain, like an injury, are different from those that influence the daily
functioning of individuals in pain, like the affective evaluation of the sensory experience. In this
multidimensional approach to pain, pain is conceived as a unique experience that can have diver-
gent outcomes in terms of illness, disability and suffering. The major benefit of the biopsycho-
social model concerns its flexibility in allowing a broad variety of factors to influence and
determine each individual pain experience. On the other hand, the biopsychosocial model is
conceived as a theoretical conceptualization only and does not possess explanatory power for the
way in which biological, psychological, and social processes exert their influence on chronic pain.
Under the biopsychosocial umbrella, several specific explanatory models have been developed
that aim to clarify the role of specific factors in pain. Explanatory models on fear of pain are
rooted in one of the most robust explanatory models in psychology, namely conditioning theory.
Fear conditioning theories possess the power to explain why fears can exist and become chronic,
even when the source of threat is not apparent.
14.2 Fear and avoidance

Fear and avoidance are central concepts to contemporary views on pain development and
treatment.
Fear is generally conceived as a basic or pure emotion that represents a present-oriented
state, an emotional reaction that is directed at an identifiable, concrete stimulus (Izard 1992;
FEAR-AVOIDANCE AS A RISK FACTOR FOR CHRONIC BACK PAIN 271
Rachmann 2004). From an evolutionary perspective, fear serves to protect the individual from
immediate threat. However, because fears can be both rational and irrational, fear responses can
occur for either accurately or inaccurately perceived dangers. It is important to note at this point
that fear cannot be equated to anxiety, although both terms are often used interchangeably.
In contrast to fear, anxiety is a future-oriented state that is more diffuse, unfocussed, and less
controllable than fear.
What pain patients are afraid of and what is thus seen as the object of fear in pain has been
divided into three areas; fear of pain sensations (i.e. the pain itself), fear of pain-causing activities
and fear of movement and (re)injury (Vlaeyen and Linton 2000). Other authors have emphasized
that also more general aspects, such as threats to life-goals and identity, might be objects of fear
in pain patients (Morley and Eccleston 2004). Fear of pain encompasses cognitive, physiologic, as
well as motor processes Studies have demonstrated correlations between fear of pain and meas-
ures of anxiety, cognitive errors, depression, and disability (McCracken et al. 1992, 1996). CLBP
patients may fear not only pain, but also activities that are expected to cause pain. In this case, fear
is hypothesized to generalize to other situations that are closely linked to the feared stimulus.
A specific fear is fear of movement and physical activity that is (wrongfully) assumed to cause (re)
injury. Kori et al. (1990) introduced the term ‘kinesiophobia’ (kinesis=movement) for the condi-
tion in which a patient has ‘an excessive, irrational, and debilitating fear of physical movement
and activity resulting from a feeling of vulnerability to painful injury or re-injury’.
Avoidance refers to the performance (or withdrawal) of a behaviour so that an undesirable
experience or situation is delayed or put off. Although in the case of chronic pain it is not possible
to avoid the pain itself, activities that might increase pain or cause (re)injury can be avoided.
Therefore, the suboptimal performance of activities is often taken as an index of avoidance behav-
iour in pain patients.
14.2.1 Fear-avoidance learning

Relying on both classical and operant conditioning principles, the two-factor theory by Mowrer
(1947) has been an influential theory in explaining fear-avoidance acquisition. According to this
theory, classical conditioning accounts for the acquisition of fear responses to aversive stimuli
through learning of associations between stimuli. Neutral stimuli that are associated with uncon-
ditioned aversive stimuli (US) develop fearful qualities and become conditioned fear stimuli
(CS). The likelihood of fear development is increased by exposure to high-intensity pain and/or
fear situations, and by frequent repetitions of the association between the new conditioned
stimulus and the pain/fear. Once objects or situations have acquired fear provoking qualities, they
develop motivating properties and elicit conditioned (defensive) responses, including escape,
avoidance, and safety seeking behaviours. For example, using a differential classical conditioning
paradigm with visual cues as CS and electric shock as the US, Bradley et al. (2008) found that
individuals responded with greater defensive reactivity in the context of threat cues (CS+) as
compared to safe cues (CS−) which were never associated with the US. The operant conditioning
component of the two factor theory describes how defensive responses (e.g. avoidance) become
persistent through learning of associations between behaviour and its consequences. The reduc-
tion of fear that is invoked by these responses serves as their negative reinforcement. Although
Mowrer’s two-factor theory has been very influential in the fear-avoidance literature, it is trou-
bled by a number of shortcomings. For example, it can not explain the persistence of avoidance
behaviour when the aversive stimulus has been withdrawn for a repeated number of times.
Following conditioning principles, the absence of repeated unpleasant experiences should lead to
the extinction of acquired avoidance behaviour. Furthermore, the basic premise of the theory that
all fears are acquired by classical conditioning cannot be sustained, since several instances are
known in which persons develop fears for stimuli they have never encountered before (e.g. fear
of snakes).
Fordyce et al. (1982) described how pain behaviour might result from avoidance learning. In
the case of pain, a patient may no longer perform certain activities because he or she anticipates
that these activities will increase pain and suffering. In the acute phase, avoidance behaviours
such as resting, limping, or the use of supportive equipment are effective in reducing suffering
from nociception. Later on, these protective pain and illness behaviours may persist in anticipa-
tion of pain, instead as of a response to it. Long-lasting avoidance of motor activities may lead to
detrimental consequences, both physically (loss of mobility, muscle strength, and fitness, possibly
resulting in the ‘disuse syndrome’) (Bortz 1984), and psychologically (loss of self-esteem, depres-
sion, and somatic preoccupation).
Although classical and operant learning principles are important mechanisms in the develop-
ment of fear and avoidance behaviour, it seems likely that besides learning principles, other proc-
esses are important in fear-avoidance acquisition as well. Favouring a cognitive theory of avoidance
learning, Philips (1987) takes the view that avoidance is influenced by the expectancy that further
exposure to certain stimuli will promote pain and suffering. This expectancy is assumed to be
based on previous aversive experiences in the same or similar situations. Since the avoidance
behaviour displayed by pain patients and by patients with phobias shows large similarities, Philips
suggested that, ‘chronic pain and chronic fear—both aversive experiences which result in avoid-
ance behaviour—may share important characteristics’ (Philips 1987, p. 277). Several studies have
focussed on the relationship between fear/anxiety and chronic pain, of which the object of fear
has been fear of pain (Lethem et al. 1983), fear of work-related activities (Waddell 1987), and fear
of movement that is assumed to cause (re)injury (Vlaeyen et al. 1995a, 1995b).
Usually, extinction of fear takes place when exposure to the feared stimulus does not lead to
the adverse consequences anymore. In the area of pain, Philips was one of the first to argue
for the systematic application of graded exposure to produce disconfirmations between expected
and the actually experienced pain and harm (Philips 1987). Experimental support for this novel
idea was provided by Crombez et al. (1996) in a sample of CLBP patients who were requested
to perform four exercises (two with each leg) at maximal force. During each exercise, baseline
pain and expected pain before the performance of the movement were recorded as well as
the experienced pain during the movement. As predicted, the CLBP patients initially overpre-
dicted pain, but after repeated exposure to the movements the overprediction was readily
corrected. These findings were replicated with other movements, including bending forward
and straight leg raising (Crombezet al. 2002b; Goubert et al. 2002). However, the later data
also showed that these disconfirmations were context dependent. Indeed, when patients were
exposed to a different movement, again overpredictions were made as if no exposure to a previ-
ous movement had taken place. This restriction of generalization was particularly true for those
patients who catastrophically (mis)interpreted the pain. Exposure to the physical activities did
not result in a fundamental change in the belief that certain movements are harmful or painful,
but rather that the movements involved in the exposure sessions are less harmful or painful than
anticipated.
14.2.2 Fear-avoidance model of pain

With the introduction of a ‘fear-avoidance model of exaggerated pain perception’ in 1983, Lethem
and colleagues reserved a critical role for fear of pain and avoidance behaviour in the explanation
of perpetuating pain complaints in the absence of organic pathology (Lethem et al. 1983). In this
model, ‘confrontation’ and ‘avoidance’ are postulated as two extreme responses to the fear of pain.
Injury/strain
Disuse
disability
depression Recovery
Avoidance/escape
Exposure
Hypervigilance
Pain experience
Fear of movement
(re)injury, pain
Catastrophizing Low fear
Fig. 14.1 Fear avoidance model of chronic pain. This figure has been reproduced with permission of
the International Association for the Study of Pain® (IASP®). The figure may not be reproduced for
any other purpose without permission.
While confrontation will lead to the reduction of fear over time, avoidance leads to the mainte-
nance or exacerbation of fear, possible developing into a phobic-like state. The avoidance results
in the reduction of both social and physical activities, which in turn can lead to a number of
physical and psychological consequences augmenting the disability.
After the introduction of the fear-avoidance model by Lethem and the emphasis on the role of
cognitions in avoidance by Philips, various cognitive-behavioural models of chronic pain have
been proposed. These models are commonly referred to as contemporary fear-avoidance models,
in which pain disability is conceived as the result of a vicious process that is determined by the
interaction between cognitions and behaviour (Asmundson et al. 1999; Vlaeyen et al. 1995b;
Waddell et al. 1993). Subtle differences aside, contemporary fear-avoidance models all share the
same basic tenets, which can be easily understood from the integrated model that is illustrated in
Figure 14.1. Upon the initial perception of pain, individuals assign a certain meaning and purpose
to the painful experience that is based upon current expectations regarding the pain and prior
learning history. Although the majority of individuals will evaluate the pain experience as unde-
sirable and unpleasant at this stage, most persons will not perceive it as an extreme threat or an
insurmountable catastrophe. These individuals will proactively and gradually confront their pain,
and resume their daily activities, promoting health behaviours and early recovery. However, for a
minority of individuals, the painful experiences, which are intensified during movement, will
elicit catastrophizing cognitions. These catastrophic cognitions can then lead to pain-related fear
(fear of pain, fear of movement, fear of (re)injury), which in its turn initiates the avoidance of
potential painful activities and hypervigilance for potential signals of additional pain and bodily
harm. As such, a vicious and self-perpetuating spiral is activated with avoidance of more and
more (daily) activities, leading to functional disability and possibly also to social isolation and
depression. In addition, physical deconditioning and depression may fuel the fear-avoidance
cycle by increasing pain intensity and increasing the fearful appraisal of and selective attention to
pain. In addition to the avoidance of fearful activities, pain disability may also persist because of
the immediate consequences to which it leads, such as diminished pain, increased attention from
others, and the avoidance of social conflicts or responsibility.
14.2.3Empirical support for the main components of the

fear-avoidance model
The fear avoidance model has offered a fruitful framework within which the development and
maintenance of persisting pain complaints can be understood. Empirical support for the model
has been found within the area of CLBP, osteoarthritis, neck pain, and chronic headache (e.g.
Fritz et al. 2001; Leeuw et al. 2007a; Vlaeyen and Linton 2000; Waddell et al. 1993). This para-
graph reviews experimental evidence for the main components of the fear-avoidance model.
Pain catastrophizing
Pain catastrophizing is conceived as a cognitive construct that represents the tendency to make
exaggerated negative or threatening interpretations of pain (Sullivan et al. 1995). Pain catastro-
phizing has often been found closely related to fear of pain. In addition, a few studies have reported
about the predictive value of pain catastrophizing for pain-related fear (Leeuw et al. 2007c;
Vlaeyen et al. 1995b; 2004b). Elevated levels of pain-catastrophizing have been consistently found
related to pain disability in chronic pain samples, acute pain samples, and pain-free volunteers
(Peters et al. 2005; Severeijns et al. 2005; Sullivan et al. 2005; Turner et al. 2004). Furthermore,
persons who tend to catastrophize about pain are found to show more hypervigilance for pain-
related information, less tolerance for pain, and to report increased pain intensity levels when
experiencing pain (Crombez et al. 2002a; Haythornthwaite et al. 2003; Peters et al. 2005; Sullivan
et al. 2005; Turner et al. 2002). In prospective studies, pain catastrophizing has been found pre-
dictive of elevated pain intensity levels during a painful procedure or after an operation (Edwards
et al. 2004; Sullivan et al. 1995; Vlaeyen et al. 2004b). For more detailed information on pain
catastrophizing see also Chapter 13.
Pain-related fear
The importance of pain-related fear has been demonstrated in chronic and acute pain samples as
well as in healthy volunteers possessing elevated levels of pain-related fear (Crombez et al. 1999;
Leeuw et al. 2007a; Swinkels-Meewisse et al. 2006; Vlaeyen and Linton 2000). As such, LBP
patients who are high in pain-related fear perform less well on behavioural (physical) perform-
ance tasks, like a walking test or a lifting task (Al Obaidi et al. 2003; Vlaeyen et al. 1995a; Vlaeyen
and Linton 2000; Vowles and Gross 2003). Fear of movement/(re)injury showed to be a stronger
predictor for detrimental performance on physical tasks then biomedical symptoms or severity
and duration of the pain experience (Vlaeyen and Linton 2000). Furthermore, pain-related fears
about work have been found related to disability of daily living and work lost in the past year,
more so than to biomedical variables such as the anatomic pattern, the time pattern, or the sever-
ity of pain (Fritz et al. 2001; Waddell et al. 1993). In pain-free individuals, elevated levels of pain-
related fear have been found associated with increased pain intensity ratings for experimentally
induced pain (George et al. 2006; Roelofs et al. 2004b; Hirsh et al. 2008). In a number of prospec-
tive studies, pain-related fear proved to be a powerful predictor of disability 6 months or 1 year
after acute pain onset (Klenerman et al. 1995; Swinkels-Meewisse et al. 2006; Turner et al. 2006;
Vlaeyen et al. 1995b).
Avoidance behaviour
In chronic pain research, avoidance behaviour is commonly derived from the suboptimal
performance of activities. Several studies have demonstrated that pain patients perform less well
on behavioural performance tasks like lifting or walking tasks (Vlaeyen and Linton 2000). In
addition, Crombez and colleagues (1999) showed that pain-related fear was associated with
escape/avoidance of physical activities, resulting in poor behavioural performance. More recently,

a number of studies have provided support for the relation between pain-related fear and avoid-
ance. Al Obaidi and colleagues (2003) demonstrated that pain-related fear was associated with
decreased speed in preferred and fast walking. In addition, Geisser and colleagues (2000) and
Vowles and Gross (2003) showed diminished performance on physical tasks in high pain-fearful
individuals. Supporting the tenets of the fear-avoidance model, there is thus compelling evidence
that pain-related fear underlies avoidance behaviours in chronic pain. Nevertheless, it should
be noted that a number of studies failed to demonstrate a relationship between fear and avoidance
(George et al. 2006; Huis in ‘t Veld et al. 2007, Michael and Burns 2004). For example, Reneman
and colleagues found a weak to no relation at all between pain-related fear and performance of
functional activities in a sample of CLBP patients (Reneman et al. 2003; Reneman et al. 2007). In
addition, some studies showed conflicting results, in that pain-related fear was associated with
task persistence rather than avoidance. Pain-related fear may elicit different behaviours depend-
ing on current goal context, and vice versa, outwardly similar behaviours may be driven by differ-
ent motivational strategies. Such an affective-motivational perspective will probably initiate a
next generation of studies addressing new questions including the following: (1) Can higher order
goals inhibit the primary goal to protect the integrity of the body by hypervigilance, escape, and
avoidance behaviour?; (2) What is the effect of life goal interference and goal conflict on pain-
related fear (see e.g. Karsdorp et al. 2010; Van Damme et al. 2008; Vlaeyen and Morley 2004)?
Hypervigilance and attention

Hypervigilance for pain is defined as the excessive orientation towards pain and pain-related
stimuli in both the external and the internal environment. Using Stroop paradigms and dot-
probe tasks, a number of studies have provided evidence that pain fearful individuals do show
attentional bias towards pain-related information or bodily sensations (Asmundson and
Hadjistavropoulos 2007; Keogh et al. 2001; Roelofs et al. 2002, 2003). Furthermore, a series of
studies have shown that pain demands attention and disrupts ongoing activities (Crombez et al.
1998a, 2002a; Eccleston and Crombez 1999). This disruptive effect of pain on attention has been
demonstrated in pain patients as well as in healthy controls, and is dependent upon the novelty,
unpredictability, and intensity of pain (Crombez et al. 1998a). Furthermore, a series of studies
showed that pain-fearful patients attend more to pain-threatening signals, are less able to ignore
pain-related information, and perform worse on cognitive demanding tasks in comparison to
non-fearful patients (Crombez et al. 1998b, 2002a; de Gier et al. 2003; Eccleston et al. 1997; Peters
et al. 2002). Supporting the idea that pain vigilance is dependent on pain-related fear, Deghani
and colleagues (2004) reported a study in which reductions in pain-related fear account for
diminished attentional bias for pain after treatment. In addition, experimental laboratory studies
have repeatedly demonstrated the association between elevated levels of pain catastrophizing and
deteriorated cognitive task performance under conditions of pain (Crombez et al. 2002a; Van
Damme et al. 2002; Vancleef and Peters 2006). Recent evidence has shown that the interruptive
effect of pain on attention results from the difficulty to disengage attention from pain-related
information, rather then from an attentional shift towards pain (Asmundson et al. 2005; Roelofs
et al. 2005; Van Damme et al. 2002). This attentional disengagement has been proposed to be
enhanced by the anticipation of pain (Van Damme et al. 2004).
Depression
Although depressed mood is one of the presumed consequences of longstanding avoidance of
daily activities, a small number of studies have actually examined depression in relation to pain-
related fear. Grotle et al. (2004) found only modest associations between pain-related fear and a
measure of distress, both in acute and CLBP patients. Sullivan et al. (2008) examined the effects
of a pain rehabilitation program aimed at work resumption on changes in levels of pain and
depression. This study revealed that changes in depression were mainly predicted by the level of
chronicity, and that this relationship was mediated by pain reduction, and not by changes in
pain catastrophizing or fear of movement/re-injury. There is a need for more systematic studies
examining the association between pain-related fear and depression.
Disuse
Frequent and enduring avoidance of daily activities may also result in gradual deterioration of a
person’s muscular system and fitness. The term ‘disuse syndrome’ refers to the physiological and
psychological effects of a reduced level of physical activity in daily life (Bousema et al. 2007;
Verbunt et al. 2003). Although often referred to in the literature, the disuse syndrome still is ill-
defined (Wittink et al. 2000). Generally, two different aspects are distinguished: physical decon-
ditioning, which can either be expressed in weakened muscle strength or reduced aerobic fitness,
and disordered muscle coordination during physical activity (also called ‘guarded movements’).
Generally, the physical fitness levels of CLBP patients are found to be either lower or equal to that
of healthy subjects (Verbunt et al. 2003). Only one current study succeeded in demonstrating that
CLBP patients have lower aerobic fitness than matched healthy counterparts, while this could not
be explained by pain-related fear or other relevant variables (Smeets et al. 2006). Evidence for
disordered muscle coordination was found by Geisser and colleagues (2004), who showed that
among CLBP patients pain-related fear was not only associated with reduced lumbar flexion and
greater EMG in full flexion, but also to abnormalities in the muscle activity during flexion from
the standing position. These changes in musculoskeletal functioning and flexion may be impor-
tant for the understanding of how pain may interfere with daily life functioning. In sum, it still is
unclear what the effects of pain-related fear are on physical functioning, both in terms of reduced
aerobic fitness as disordered muscle coordination (see also Chapter 12).
14.2.4 Vulnerabilities for fear of pain

The foregoing paragraphs offer support for the importance of pain-related fear and associated
avoidance behaviour in chronic pain. However, an interesting issue that presents itself concerns
the aetiology of pain-related fear. Complementing fear-learning theories as described in earlier
paragraphs, several factors have been proposed to constitute vulnerability for pain-related fear.
As such, research has indicated that fear of pain is closely related to a number of negative emo-
tional constructs, like negative affectivity, trait anxiety, anxiety sensitivity, and illness injury sen-
sitivity (Keogh and Asmundson 2004; for more information see Chapter 12). Elevated levels of
anxiety sensitivity and illness injury sensitivity have been frequently found related to maladaptive
responses to pain in terms of reduced tolerance for pain, increased pain intensity levels, attention
bias for pain etc (Stewart and Asmundson 2006; Vancleef and Peters 2006; Vancleef et al. 2006).
In addition, trait anxiety and negative affectivity have been found related to increased pain inten-
sity, increased discomfort and disability by the pain, hypervigilance for internal bodily sensations,
and less adequate coping with and perceived control over pain (Keogh and Asmundson 2004;
Stegen et al. 2001). One way to conceptualize the various constructs in relation to each other is by
representing them in a hierarchical tree structure (Keogh and Asmundson 2004; Vancleef et al.
2006; Vancleef et al. 2009). In this hierarchical tree, fear of pain is conceived as the most specific
construct that resides at the lowest level of the tree. Fear of pain is subordinated to anxiety sensi-
tivity and illness injury sensitivity that in their turn subordinate to trait anxiety. One level above
trait anxiety, negative affectivity represents the most general and stable negative emotional
construct at the top of the tree. A hierarchical structure offers an elucidating framework for the
conceptualization of pain-relevant negative emotional constructs. As such, content overlap

between the constructs is accounted for by interrelations between constructs, while each indi-
vidual construct possesses unique explanatory value for pain as well. According to the tree struc-
ture, pain-related fear will exert the most direct influence on pain, while influences of other
constructs, like anxiety sensitivity for example, are understood to run through the specific fears
that are found at the lower positioned levels. It is important to note that the hierarchical structure
serves a mere theoretical conceptualization to date that serves to aid our understanding of the
relative contribution of various negative emotional constructs to pain. Nevertheless, it seems
obvious that general and stable anxiety related constructs entail vulnerability for pain-related fear
and pain catastrophizing. Therefore, it is important that diagnostic and treatment approaches to
pain keep these other pain anxiety constructs into account in understanding and treating indi-
vidual pain problems.
14.3 Assessment of pain-related fear

Patients with fear of pain are at risk for developing persistent back pain and disability, which
implies a need to make a thorough assessment of pain-related fear and avoidance. As pain-related
fear involves cognitions, emotions, overt behaviour and physiological responses, several methods
are needed to fully understand patients’ reactions to pain. This section provides an overview of
assessment tools that are frequently used in current research and clinical practice to assess pain-
related fear. As such, both clinicians and researchers rely heavily on interviews and self-report
measures to assess fear. Several widely used validated questionnaires are also useful tools in
the assessment of fear. Furthermore, another tool based on pictures of every day movements
is helpful in determining fear and above all the avoidance of movements. In order to get a
more complete picture and to be able to grasp different components of the object of pain-related
fear, self-report measures are often combined with observational or physiological measures.
A multimodal assessment presumably helps us to reach a better understanding of patients’ fear
of pain.
14.3.1 Self-report measures

The most common way to assess pain-related fear is through self-report measures which cover
different aspects of fear. Most of these focus on cognitive and affective responsivity that may
indicate fear. A challenge in developing self-report measures has been isolating the emotional
response as opposed to measuring ‘beliefs’. Indeed, many of the instruments available appear to
tap into beliefs that the patient holds rather than things that actually provoke fear. Similarly, some
instruments may measure affect but may be more oriented towards ‘worry’ rather than to actual
fear. Although this may seem to some to be a mute point, the differences may be significant espe-
cially if the proposed treatment (see the section on exposure in vivo) focuses on fear. At present,
a large number of self-report measures are available for the assessment of pain-related fear. The
choice for one measure over the other mainly depends on the object of fear (e.g. fear of pain, fear
of movement, fear of re-injury), and the fear specific concerns (e.g. cognitive concerns, physical
concerns, emotional concerns) one aims to measure. Furthermore, whereas some measures are
more generally applicable for assessing fear in both healthy and pain populations, others are more
appropriate for assessment in specific pain patient groups only (e.g. cervical pain, back pain).
Some of the most commonly used self-report measures of pain-related fear in the current assess-
ment and treatment of pain-related fear are listed below. However, this list is not conclusive, and
other measures, including modified versions of the questionnaires described below are eligible for
use in pain research and clinical practice as well.
It should be noted that self-report instruments are subject to a number of drawbacks that have
to be kept in mind when interpreting results of these measures. As such, respondents might not
always report their genuine thoughts and feelings, because of social desirability, response bias, or
simply because they are unable to access and grasp their real concerns. Furthermore, no one-to-
one relationship between the measure and the latent construct it represents can be assumed.
Nevertheless, keeping these drawbacks in mind, psychometric evaluated self-report measures are
easy to administer, at low-cost, but still offer valuable information to researchers and clinicians.
Fear of pain
The Pain and Impairment Relationship Scale (PAIRS) was an early attempt to capture chronic pain
patients’ beliefs and attitudes towards pain and ability to function despite pain (Riley et al. 1988).
It consists of 15 statements about pain and activity (e.g. ‘I can’t go about my normal life activities
when I am in pain’, ‘An increase in pain is an indication that I should stop what I am doing
until the pain decreases’) which are rated on a seven-point scale (0=completely disagree;
6=completely agree). The scale has been found to have good psychometric properties (DeGood
and Shutty 1992) and may be used to assess beliefs about pain and activity. As this scale was a
relatively early attempt to assess pain-related fear, it is not clear what processes it in fact captures.
Indeed, it appears to assess a wide range of attitudes and beliefs concerning living with persistent
pain that go beyond fear of pain.
The Pain Anxiety Symptoms Scale (PASS) was developed to measure general anxiety and fear in
people with chronic pain (McCracken et al. 1992). The original scale consisted of 40 items
intended to assess four aspects of pain-related anxiety: (1) fearful appraisal of pain (e.g. ‘When
I feel pain I am afraid something terrible will happen’): (2) cognitive anxiety (e.g. ‘During painful
episodes it is difficult for me to think of anything besides the pain’): (3) physiological anxiety (e.g.
‘Pain seems to cause my heart to pound or race’): (4) escape and avoidance behaviour (e.g.
‘I avoid important activities when I hurt’). Responders rate how often they have these thoughts
or sensations on a six-point scale (0=never; 5=always). Later, a shortened version of the scale was
developed and validated (PASS-20) (McCracken and Dhingra 2002). The short form version has
20 items and has been found to be a good reflection of the original PASS (Roelofs et al. 2004a).
One of the advantages with this measure is that it includes bodily indicators of fear through ques-
tions about physiological arousal. However, as these reactions might appear as a direct conse-
quence of pain, it is difficult to determine whether it actually captures indicators of fear.
Furthermore, as bodily reactions are difficult to assess through self-report measures, physiologi-
cal measures might be preferable if available. In addition, the questions in the PASS presume
that the individual has an earlier or present experience of pain problems. This makes the PASS
inappropriate for use with people who are not clinical pain patients.
The Fear of Pain Questionnaire-III (FPQ-III) (McNeil and Rainwater 1998) was developed to
assess fear associated with both acute and chronic pain in healthy individuals. People are asked to
rate their fear of pain in 30 situations, which are divided into: (1) severe pain (e.g. ‘Breaking your
leg’), (2) minor pain (e.g. ‘Having a muscle cramp’) and (3) medical pain (e.g. ‘Having a tooth
pulled’). Answers are given on a five-point scale (1=no fear at all; 5=extreme fear). Recently, a
shortened version of the scale has been developed (FPQ-SF) (Asmundson et al. 2008). The authors
suggest that the FPQ-SF is a more psychometrically sound alternative to the original scale. One
advantage of the FPQ is that it is possible to use it both with pain and non-pain populations as the
questions do not imply any earlier clinical experience of pain. However, the non-clinical focus
of the scale might make it less relevant to use for clinical purposes as some of the items
describe events that are far from clinical situations (e.g. having sand or dust blowing into your
eyes; gulping a hot drink before it has cooled).
Fear of pain-causing activities

The Fear-Avoidance Beliefs Questionnaire (FABQ) (Waddell et al. 1993 ) is used to assess
pain patients’ beliefs about possible harm from work and physical activities. It has 16 items which
are divided into two subscales; (1) fear-avoidance beliefs about physical activity (e.g. ‘Physical
activity might harm my back’) and (2) fear-avoidance beliefs about work (‘My work aggravated
my pain’). Answers are given on a seven-point scale (0=completely disagree; 6=completely
agree). The scale has been found to have good psychometric properties in patients with both acute
and long-term pain (Swinkels-Meewisse et al. 2003; Waddell et al. 1993), and may be used to
identify beliefs that are related to pain-related disability and work loss. One of the disadvantages
with this questionnaire might be in line with what the name of the scale indicates—that it assesses
beliefs rather than fear per se. It might be possible that people hold such beliefs without being
particularly fearful, which is important to take into account when using this scale in a clinical
context.
Fear of movement and (re)injury

The Tampa Scale for Kinesiophobia (TSK) (Kori et al. 1990) has been developed to assess patients’
fear of (re)injury due to movement, and consists of 17 items (e.g. ‘Pain always means I have
injured my body’ and ‘If I try to overcome it, my pain would increase’).
Later, a shorter version of the TSK has been developed and validated, containing only 13 items
as it excludes the four items that are reverse scored. This version has greater internal consistency
and is composed of two lower-order factors (‘Activity Avoidance’ and ‘Somatic Focus’) that also
load on a single higher-order factor (Geisser et al. 2000). A number of studies have corroborated
the two-factor structure of TSK, which also appeared invariant across pain diagnoses and Dutch,
Swedish, and Canadian samples (Goubert et al. 2004b; Roelofs et al. 2007). One concern that has
been highlighted about the TSK is that some items have troublesome wording (Pool et al. 2009).
More specifically, some questions seem to be difficult to understand because they employ words
such as ‘dangerous’ and ‘injury’ that may have different meanings for different patients. Some
items also use negations which may be difficult for some patients to understand. These reflections
might be important to take into consideration when using the TSK with patients.
Perceived harmfulness of movements: the PHODA

When fear of movement is assessed in a clinical context, the Photograph Series of Daily Activities
(PHODA) (Kugler et al. 1999) is a pictorial assessment tool which consists of 98 photographs of
various daily-life activities (e.g. walking, lifting, bending, etc.). Patients are asked to rate how
harmful they think each activity would be to their back by placing the photographs on a ther-
mometer which ranges from 0 (=not at all harmful) to 100 (=extremely harmful). Although the
PHODA mainly is used as a self-report measure, it also provides the clinician with an opportunity
to observe and discuss the patient’s fear when confronted with different activities. The PHODA is
commonly used in the assessment phase of exposure in vivo, a treatment for pain-related fear that
is described later in this chapter. Recently, an electronic and shortened version of the PHODA
(PHODA-SeV) was developed and validated (Leeuw et al. 2007b). It showed excellent psycho-
metric properties, which implies that it is a proper alternative to the original version. One limita-
tion that has been raised is that the PHODA seems to be more appropriate for highly fearful pain
patients and less applicable for those with low or mediate levels of fear (Trost et al. 2009).
Nevertheless, because the PHODA consists of photographs of real activities, it may be a clinically
useful tool for revealing the patients’ beliefs and concerns about specific movements. This is an
important basis for the exposure treatment and one which may be quite difficult without tools
such as the PHODA.
An alternative measure to assess fear of movement is the Pictorial Fear of Activity Scale-Cervical
(PFActS-C) (Turk et al. 2008). This instrument was recently developed to assess fear of move-
ment that is specifically related to neck pain. The PFActS-C differs from the PHODA in its
specific focus on neck pain and the varying degree of mechanical load that is incorporated in the
photographs. The instrument consists of 77 photographs of movements with varying biome-
chanical stress. Respondents are asked, for each photograph, to rate on a scale from 0 (=no fear at
all) to 10 (=extremely fearful) how worried or fearful they would be to perform the activity as
shown in the picture. A first validation study in a group of neck pain patients indicated that the
PFActS-C is a promising tool for assessing fear of movement in patients with cervical pain (Turk
et al. 2008). However, since this tool is still rather new, more research is needed to examine its
applicability.
14.3.2 Interview
Apart from questionnaires, every assessment of pain-related fear also involves a rigorous inter-
view. The semi-structured interview is an important tool for collecting additional information
about the patient’s cognitions, emotions and overt behaviour (e.g. avoidance) in relation to pain.
The clinician also may get a better picture of the role fear and avoidance may have played in the
aetiology and maintenance of the pain and disability problem. However, as the interview provides
a purely subjective picture of the pain problem, it should be complemented by other assessment
tools. Some of the areas that are covered in the interview are listed in Table 14.1.
14.3.3 Self-assessment
Self-assessment is a useful supplement to the interview. Self-assessment is one of the most com-
monly used methods in psychological research and practice (Sigmon and LaMattina 2006). In
self-assessment, the patients systematically observe and record their own behaviour. In pain
Table 14.1 Interview in assessment of pain-related fear
Description of Where do you have pain?

the pain What does the pain feel like?
How intense is the pain? (0–10)
When does it hurt more?
When does it hurt less?
Learning history When did the pain start?
How did it develop?
What do you think has made your pain problem worse?
Cognitions What do you think causes your pain?
When you do something that provokes the pain, what goes through your mind?
What do you think it will lead to in the future?
What is the worst thing that could happen?
Emotions What emotions do you have in relation to the pain?
Are there any activities that you are afraid of doing?
Activities/ What are you able to do despite the pain?
movements vs. What activities do you avoid because of the pain?
avoidance How active are you in daily life?
Coping vs. What do you do to stand the pain?
safety What do you do when the pain gets worse?
behaviours What do you do to be able to perform activities that are difficult because of pain?
patients, diaries are used to keep record of situations that provoke fear and avoidance. Often, the
patients also use the diary to register their daily activities, which may be an indirect measure of
fear and avoidance. In the last years, research about the use of electronic diaries in pain assess-
ment has increased (e.g. Jamison et al. 2001; Peters et al. 2000). Maybe this will become more
common in clinical practice in the future. At the moment, however, paper-and-pencil diaries are
still the most widely used tools in self-assessment of pain-related fear.
14.3.4 Observational measures

Even though it is well known that patients express fear of pain both verbally and non-verbally,
there are few standardized tools to assess overt behaviour. Fear of pain has been associated with a
number of overt behaviours, such as diminished physical performance and escape/avoidance
behaviour (Leeuw et al. 2007a). In a clinical situation, clinicians may instinctively observe how
the patient reacts to pain and movement, and use this information in the further assessment.
Notwithstanding, a thorough observational assessment is seldom accomplished. There are obser-
vational measures which aim to assess pain behaviour (e.g. Weiner et al. 1996). However, these are
not developed to capture fear. In verbal assessment of fear, we do not rely solely on one question
or on what the patients tell us at first glance, but complement this with standardized question-
naires. A comparable scrupulous assessment of overt behaviour would help us to reach a better
understanding of patients’ pain-related fear.
Assessment of other overt behaviours may also help to reach a better understanding of pain-
related fear. Facial expression has been suggested to be a potential target for assessment (McNeil
and Vowles 2004). However, there is still a lack of appropriate assessment tools for facial expres-
sion as a non-verbal manifestation of pain-related fear.
14.3.5 Physiological measures

In the last decades, researchers have paid increasing attention to psychophysiological recordings
as an assessment tool for pain-related fear and anxiety. In particular, Flor and her colleagues have
contributed to the growing knowledge about how psychophysiological measures may be used in
the area of pain (Flor et al. 2001), including as indicators of fear. Muscle tension, skin conduct-
ance, and heart rate are some of the targets for psychophysiological assessments in pain patients
(Andrasik and Flor 2008). Also the startle reflex, an involuntary eye blink movement, has been
used in experiments as an indicator of fear (e.g. Carleton et al. 2006; Naliboff et al. 2008). In
clinical settings, however, it is still uncommon to include physiological measures when assessing
fear. A challenge for the future is the development of sound psychophysiological measures that
may be used in the clinical assessment of pain-related fear.
14.3.6 Additional measures

In addition to explicit and physiological measures of fear of pain, researchers have relied on
indirect cognitive measures of indicators of fear and anxiety related to pain. In these measures,
variables of interest (beliefs, attitudes, cognitive biases) are inferred from behavioural responses
(e.g. reaction times, reading times) within a specific context. Characteristic to these measures is
the fact that they assess cognitive processes, beliefs, and attitudes without the awareness or control
of the participant. Experimental research often uses automatized cognitive processing tasks like
the modified Stroop task and the dot probe paradigm to assess attention bias for pain. In the mod-
ified Stroop task, categories of sensory or affective pain words and neutral words are presented in
different colours. Responders are asked to name the colour of each word and response times are
compared. Typically, it takes longer to name the colour when the information is threatening; in
this case when pain words are presented. In the dot probe paradigm, one neutral word and one
pain-related word are presented simultaneously on a screen. Next, one of the words is replaced by
a dot. Response times for when the dot replaces neutral and pain-related words are compared. In
pain patients and pain-fearful individuals, reaction time is expected to be shorter when the dot
replaces pain words, due to attentional biases. Both the modified Stroop task and the dot probe
paradigm are tests of attentional bias which may presumably indicate fear. There are mixed
results, however, concerning whether these tests are reliable indicators of fear (e.g. Asmundson
et al. 2005; Pincus and Morley 2001; Roelofs et al. 2002).
Furthermore, a few studies have relied on implicit association measures to assess the attitudes
and beliefs that pain-fearful individuals possess regarding pain and disability. These studies dem-
onstrated that CLBP patient who are high in pain-related fear possess stronger associations
between pain and threat than low-fearful patients do. Furthermore, Goubert et al. (2003) reported
about a general implicit negative attitude towards back-stressing pictures in healthy individuals,
using an affective priming task. Recently, Leeuw and colleagues (2007d) used two association
tasks, the implicit association task and the go-no-go association task, to measure implicit fear of
movement(re) injury in a group of CLBP patients and pain-free controls. Results did not indicate
implicit fear of movement(re)injury in both groups, nor did the authors observe differences in
the implicit attitudes between the patient and the control group. Nevertheless, the authors argued
that psychometric and methodological difficulties of the paradigms might explain the lack of
findings. In general implicit measures are promising techniques that might be able to reveal
those aspects of pain-related fear that are inaccessible to explicit measures like self-report and
interviews.
14.4 Treating pain-related fear

Exposure in vivo is a novel treatment approach which has been developed to target fear of pain. As
seen earlier in this chapter, fear and avoidance may be essential factors that maintain pain and
dysfunction in some patients. It has been stated that ‘Fear of pain and what we do about it may be
more disabling than pain itself’ (Waddell et al. 1993, p. 164). Based on this, graded exposure
in vivo recently was developed as a cognitive-behavioural treatment for patients with back pain
and high levels of pain-related fear (Vlaeyen et al. 2004a). Today, it is the only standardized and
evidence-based treatment which exclusively focuses on reducing pain-related fear and avoidance.
Because exposure in vivo is the ‘treatment of choice’ for patients suffering from excessive fears,
e.g. phobias (Barlow 2000), it would also seem to be ideal for patients with pain-related fear.
Accordingly, exposure in vivo for pain-related fear relies on the same principles as exposure
therapy for phobia. In fact, Kori et al. (1990) went so far as to call pain-related fear a phobia,
introducing the term ‘kinesiophobia’ as a description of ‘an excessive, irrational, and debilitating
fear of physical movement and activity resulting from a feeling of vulnerability to painful injury
or reinjury’. Avoidance of the feared stimuli, which here refers to movements, maintains and
exaggerates the fear. During exposure, phobic patients confront the feared stimuli until habitua-
tion, i.e. until the fear decreases and finally dissipates. The same principles may be transferred to
pain patients. During exposure for pain-related fear, the patient is gradually exposed to move-
ments following the hierarchy until extinction of defensive fear responses occur. Besides the effect
of extinction there may also be benefits in terms of cognitions. As the fear and avoidance model
underscores, performing a feared movement may be associated with catastrophic expectations
which, during exposure in vivo, are challenged and disconfirmed (Vlaeyen et al. 2008). The main
goal in exposure in vivo is hence that the patient confronts what (s)he is afraid of as a way to
become less fearful. Recent evidence shows that learned associations between certain movements
and pain are not abolished during exposure, but that inhibitory learning occurs. This means that
the initial associations remain stored in memory, but that the individual has learned to inhibit the
conditioned (defence) responses. Moreover, the inhibitory learning appears to be context depend-
ent, thereby limiting generalization of extinction to other contexts (Craske et al. 2008; Crombez
et al. 2002b).
14.4.1 Main components of exposure

Assessment
The first step in graded exposure in vivo is to perform a thorough assessment of pain-related fear
and avoidance. As assessment has been discussed earlier in this chapter, let us simply review this
briefly here. A proper assessment is a prerequisite for the treatment as it serves to identify target
movements for the exposure. The intention is to get a complete picture of how the patient’s fear
and avoidance maintain the pain problem. The PHODA (Leeuw et al. 2007b), an instrument with
pictures of daily activities that patients rate for fear, is an especially important tool in the assess-
ment. Thus, the PHODA provides information as to which movements the patient is fearful of
and also how afraid they are of each movement. This information is employed to develop a hier-
archy of feared movements. The PHODA may also be used to reveal catastrophic (mis)interpreta-
tions and irrational beliefs that patients have about movements and pain. At the end of assessment,
the therapist should have a clear idea of what movements the patient avoids and what beliefs and
catastrophic thoughts are linked to these movements; this will be the focus of the treatment.
Defining goals
During the assessment phase, the patient and the therapist agree on a limited number of specific
treatment goals. Goal setting serves a number of important functions. First, the goals serve as a
clear statement of the aim of exposure. A primary aim in cognitive behavioural treatments for
pain is to restore functional ability, and not simply to get rid of the pain. Therefore, goals should
focus on activities and not on pain reduction. Second, when the patient has been involved in
developing goals they view as quite valuable; they become more engaged in the treatment and
naturally take a more active role in the treatment. Third, goals help to structure the treatment. By
establishing goals the patient and the therapist share a picture of what they are striving towards.
The goals also make it easier to establish a hierarchy which involves movements that are impor-
tant for the patient to restore. Typically, a hierarchy includes daily activities that the patient
desires to resume or develop such as hobbies, family life, or work. Fourth, goals are also used as
an evaluation tool. With clear goals, both the patient and the therapist may evaluate treatment
progress and see whether the exposure has succeeded or not. Typical example of treatment goals
are shown in Table 14.2.
Table 14.2 Typical examples of treatment goals
Goal Description
Play with my kids Play actively with my kids 30 minutes in the garden twice a week, including
lifting them, bending over etc.
Play soccer Play soccer with my old team 1 hour once a week
Increase my work time Increase my work time from 25% to 75% at my ordinary work place,
including performing all duties.
Do the weekly shopping Do the weekly shopping by myself, including carrying the bags to the car
Table 14.3 An example of a graded fear-hierarchy of pain-related fear stimuli
Goal: To play with my kids in the garden 30 minutes, twice a week

Movement/activity Pre-treatment Post-treatment
PHODA score PHODA score
Running while carrying a child on my shoulders 100 20
Riding off a pavement with a bicycle 90 15
Lifting a child from the floor 80 0
Bending forward to pick up tools from the floor 70 0
Jumping up and down 50 0
Walking up and down the stairs 30 0
Establishing hierarchies
Once treatment goals are identified, the therapist and the patient need to establish a fear hierar-
chy. The hierarchy consists of movements that provoke pain-related fear starting with those
inducing only a little fear to those trigger a maximum of fear. It normally will include 5–10 steps
or movements. An example of a graded fear-hierarchy is shown in Table 14.3. Exposure begins
with a movement that provokes only a small amount of fear. When habituation occurs and the
patient is able to perform the movement without provoking fear, then the next movement in the
hierarchy is tackled. This procedure continues until the patient is able to perform all of the move-
ments in the hierarchy, including the goal activity, without provoking very much fear. In this way,
function is restored and the clear goals of the treatment are achieved.
Education
Education is an important part of the treatment. The purpose of the psychoeducation is to pro-
vide a clear rational and to engage the patient. Before being exposed to any fear-provoking stim-
uli, the patient should understand how and why exposure should reduce fear, and how this will
bring the patient closer to his or her goals. The therapist explains the fear-avoidance model, using
the patient’s own history and wording as much as possible. If the patient does not use the word
‘fear’, other words may be more suitable. An essential message in the education is that fear and
avoidance are natural and normal responses to acute pain, but that they paradoxically maintain
fear and pain in the long term. Another point is that long-term back, neck or shoulder pain typi-
cally signals that muscles are working too hard to protect the back (resulting in muscle pain)
rather than from damage to the spine. There is some evidence that psychoeducation in itself may
have some effect in reducing pain-related fear and disability (Burton et al. 1999; de Jong et al.
2005b; Moore et al. 2000). The psychoeducation aims to enhance the patient’s understanding
of how the exposure treatment will help them to deal with their fears and achieve their goals
concerning function.
Graded exposure
The keystone in exposure is, of course, when the patient actually exposes himself to previously
avoided stimuli according to the fear hierarchy. The patient starts at lower level of the hierarchy,
with a movement that provokes relatively less intense fear. First, the therapist serves as a role
model and demonstrates the movement. Then, the patient performs the movement, with verbal
support from the therapist. The general principles for exposure are followed: the patient is exposed
to the feared stimuli for a prolonged time, until fear and anxiety decreases significantly. In expo-
sure for pain-related fear, the patient consequently performs the movement several times.
The therapist encourages the patient to do the exposure without using safety behaviours, i.e.
subtle strategies that patients use to ‘protect’ themselves (guarded walking, relaxation, seeking
social support, etc.). Safety behaviours are subtle avoidance behaviours which may hinder the
effect of exposure (Tang et al. 2007; Vlaeyen et al. 2004a). In between exposure sessions, the
patient is told to continue the exposure at home by performing homework assignments. Besides
reducing fear and re-establishing physical function, this helps the patient to generalize the effects
and enhance self-efficacy. All treatment sessions follow the same structure, each time at a higher
level in the hierarchy, to systematically decrease the patient’s pain-related fear.
Behavioural experiments
The graded exposure in vivo might be combined with behavioural experiments (Bennett-Levy
et al. 2004; Vlaeyen et al. 2004a). Behavioural experiments have been developed from cognitive
theory, to create a collaborative empiricism. The purpose is to challenge irrational beliefs and
catastrophic (mis)interpretations about pain and movement. In a behavioural experiment, the
patient first formulates the belief and rates its credibility. Then, an alternative belief is formulated
and the patient rates its credibility. Next, the patient performs the movement and describes
what happens. Finally, the outcome is compared to the expectation and conclusions are drawn.
Table 14.4 displays an example of a behavioural experiment in a patient with pain-related fear.
The main goal with behavioural experiments is to challenge and adjust irrational beliefs but may
also be used as a way to encourage patients to view the exposure as an experiment in which new
behaviours are tested.
14.4.2 Evidence for exposure in vivo

Exposure in vivo is a promising treatment approach for patients with pain-related fear. Two
recent reviews concluded that exposure in vivo is currently the treatment of choice for such
patients (Bailey et al. 2010; Lohnberg 2007). The first research about the effect of exposure was a
series of studies using a replicated single-case experimental design. These showed that graded
exposure in vivo was effective in reducing pain-related fear, avoidance behaviour, and to some
extent also pain intensity in patients with back pain (Boersma et al. 2004; Linton et al. 2002;
Vlaeyen et al. 2001). Later, the same type of design was used to evaluate the effect in other groups
of patients. These studies demonstrate the same basic results in patients with complex regional
pain syndrome (de Jong et al. 2005a) and in patients with neck pain after a motor vehicle accident
(de Jong et al. 2008). Further evidence was then provided in randomized control trials (RCTs).
RCTs are generally seen as the most powerful way of demonstrating that an intervention is effec-
tive (Kazdin 2003). Recently, three RCTs have been published, evaluating the effect of exposure
Table 14.4 An example of a behavioural experiment in a patient with pain-related fear
Belief/thought Rating of credibility Rating of credibility

before the experiment after the experiment
Belief/catastrophic thought: ‘If I jump from the 85% 20%
chair, I will get excruciating pain (10 on a scale 0–10)’
Alternative belief: ‘If I jump from the chair, my pain 15% 80%
will only increase slightly (2 on a scale 0–10)’
What happened: The pain increased slightly (from 2 to 3 on a scale 0–10)
Conclusion: The original thought was an overestimation of the harmfulness of the movement.
The alternative belief is more realistic.
in patients with long-term back pain (Leeuw et al. 2008; Linton et al. 2008; Woods and Asmundson
2008). Although the results are not as striking as in the single-subject studies, exposure was found
to result in moderate to large effects for reducing pain-related fear, catastrophic thoughts, disabil-
ity and pain. However, when compared to graded activity, a commonly used treatment in pain
rehabilitation, the results were more modest (Leeuw et al. 2008; Woods and Asmundson 2008).
The exposure groups showed larger improvements on measures of pain-related fear and anxiety,
beliefs, and catastrophizing, but on measures of disability and pain both treatments appeared to
be about equally effective. Overall, these studies show that exposure may be an important treat-
ment, but that it cannot yet be recommended as a ‘stand alone’ treatment. As the authors of one
of the studies suggest (Woods and Asmundson 2008), it is important to explore the use of expo-
sure in applied multidisciplinary treatment settings. Given that patients with high levels of pain-
related fear become noticeably less fearful and anxious through exposure, it has potential to be a
successful addition to ordinary treatment packages for fearful patients suffering from long-term
pain. Likewise, new advances in delivering exposure treatments may also improve its utility.
14.5 Future directions

This chapter shows that the literature on fear-avoidance has strong roots in the learning–
conditioning theory. The classical conditioning account of fear development predicts that neutral
stimuli can develop fearful qualities and become conditioned fear stimuli (CS) when they are
associated with unconditioned aversive stimuli (US). Following this theory, the nature of the CS
and the US determines the kind of fear. Future research should focus more on the nature of both
types of stimuli in fear development. On a continuum from ‘outside’ to ‘inside’ the body—an
exteroceptive-interoceptive continuum—three types of sensorial receptors can be distinguished.
Exteroceptors are situated closely to the bodily surface and are sensitive to mechanical and electro-
mechanical energetic fields surrounding the organism. Proprioceptors are sensitive to the orienta-
tion and the action of parts of the body in space. Interoceptors are located in the cavities of the
body and form the basis for the neural representation of the viscera and the vascular system.
Traditionally, human fear conditioning studies have used paradigms in which affectively neutral,
exteroceptive stimuli (CSs, e.g. a picture or a tone) become predictors for an aversive physical
event (e.g. pain). Pain research has focused more on proprioceptive stimuli (i.e. kinesiophobia).
Certain movements may have acquired the features of a CS as they have been associated with a US
(more pain), and create fear of movement. Strangely enough, there is much less information in
the pain literature on interoceptive fear conditioning. Interoceptive cues (e.g. muscle spasms,
mild pain, dizziness . . .) can also become CSs and this might be a particular form of fear condi-
tioning differentiates the fear of ‘pain’ construct from the fear of ‘movement’ construct (De Peuter
et al. 2010). The fear of pain construct might be more relevant in physical complaints where the
musculoskeletal system is less involved. Besides the theoretical importance of examining this idea
more into detail, findings may also lead to different therapeutic interventions. For example, inte-
roceptive exposure has been reported as an effective treatment in patients with panic disorder,
assumed to be mediated by interoceptive fear conditioning (Walker and Furer 2008). At this
point, it is important to note that the term ‘fear-avoidance’ as introduced by Lethem et al. in 1983
might need reconsideration given the fact that it points only to one defensive mechanism: avoid-
ance. It might be more appropriate to adopt the general term ‘pain-related fear’, which could be
defined as the fear that originated in associations where pain is the US. Depending on the CS, this
general term can then be subdivided in fear of movement (proprioceptive fear), fear of pain (inte-
roceptive fear), fear of sounds (exteroceptieve fear, for example in headaches), etc . . . or more
diffuse pain anxiety.
A second direction for further research on fear-avoidance stems from commonly observed and
complex situations in which the individual cannot identify discrete cues that signal or predict the
US. In that case the context in which the US occurs becomes the US, and theorists have intro-
duced the term ‘contextual’ fear for this form of fear, which has a lot of resemblances with gener-
alized anxiety disorder (Vansteenwegen et al. 2008). Some authors prefer the term ‘anxiety’ above
fear. In the case of contextual fear with pain as the US, one could speak of ‘pain anxiety’. The
characteristics of pain anxiety are that the defensive responses last longer as contextual stimuli
usually are temporally quite stable.
In sum, since the early writings in the 1980s we have made considerable progress in the
understanding and management of pain-related fear and anxiety, and we look forward to further
exploration and fine tuning of these processes in the near future.
Acknowledgements
The contribution of Johan W. S. Vlaeyen was supported by the NWO Social Sciences Research
Council of the Netherlands, Grant No. 453–04-003, and by an Odysseus Grant of the Research
Foundation Flanders, Belgium (FWO).
References
Al Obaidi, S. M., Al Zoabi, B., Al Shuwaie, N., Al Zaabie, N. and Nelson, R. M. (2003) The influence of
pain and pain-related fear and disability beliefs on walking velocity in chronic low back pain.
International Journal of Rehabilitation Research, 26, 101–8.
Andrasik, F. and Flor, H. (2008) Biofeedback, in H. Breivik, W. Campbell and M.K. Nicholas (ed.) Clinical
Pain Management: Practice and procedures. London: Hodder & Stoughton.
Asmundson, G. J. and Hadjistavropoulos, H. D. (2007) Is high fear of pain associated with attentional bias
for pain-related or general threat? A categorical reanalysis. Journal of Pain, 8, 11–18.
Asmundson, G. J. G., Bovell, C. V., Carleton, N. R. and McWilliams, L. A. (2008) The Fear of Pain
Questionnnaire - Short Form (FPQ-SF): Factorial validity and psychometric properties. Pain, 134, 51–58.
Asmundson, G. J. G., Carleton, N. R. and Ekong, J. (2005) Dot-probe evaluation of selective attentional
processing of pain cues in patients with chronic headaches. Pain, 114, 250–6.
Asmundson, G. J. G., Norton, P. J. and Norton, G. R. (1999) Beyond pain: The role of fear and avoidance
in chronicity. Clinical Psychology Review, 19, 97–119.
Bailey, K. M., Carleton, R. N., Vlaeyen, J. W., Asmundson, G. J. (2010) Treatments addressing pain-related
fear and anxiety in patients with chronic musculoskeletal pain: A preliminary review. Cognitive
Behaviour Therapy, 39(1), 46–63.
Barlow, D. H. (2000) Unravelling the mysteries of anxiety and its disorders from the perspective of emotion
theory. American Psychologist, 55, 1247–63.
Bennett-Levy, J., Butler, G., Fennell, M., Hackmann, A., Mueller, M. and Westbrook, D. (2004) Oxford
Guide to Behavioral Experiments in Cognitive Psychology, Oxford: Oxford University Press.
Boersma, K., Linton, S. J., Overmeer, T., Jansson, M., Vlaeyen, J. W. S. and de Jong, J. (2004) Lowering
fear-avoidance and enhancing function through exposure in vivo. A multiple baseline study across six
patients with back pain. Pain, 108, 8–16.
Bortz, W. M. (1984) The disuse syndrome. The Western Journal of Medicine, 141, 691–4.
Bousema, E. J., Verbunt, J. A., Seelen, H. A., Vlaeyen, J. W. and Knottnerus, J. A. (2007) Disuse and
physical deconditioning in the first year after the onset of back pain. Pain, 130, 279–86.
Bradley, M., Silakowski, T. and Lang, P. J. (2008) Fear of pain and defensive activation. Pain, 137, 156–63.
Burton, A. K., Waddell, G., Tillotson, K. M. and Summerton, N. (1999) Information and advice to patients
with back pain can have a positive effect. A randomized controlled trial of a novel educational booklet
in primary care. Spine, 24, 2484–91.
Carleton, N. R., Asmundson, G. J. G., Collimore, K. C. and Ellwanger, J. (2006) Strategic and automatic
threat processing in chronic musculoskeletal pain: a startle probe investigation. Cognitive Behaviour
Therapy, 35, 236–47.
Craske, M. G., Kircanski, K., Zelikowsky, M., Mystkowski, J., Chowdhury, N. and Baker, A. (2008)
Optimizing inhibitory learning during exposure therapy. Behaviour Research and Therapy, 46, 5–27.
Crombez, G., Eccleston, C., Baeyens, F. and Eelen, P. (1998a) Attentional disruption is enhanced by the
threat of pain. Behaviour Research and Therapy, 36, 195–204.
Crombez, G., Eccleston, C., van den Broeck, A., van Houdenhove, B. and Goubert, L. (2002a) The effects
of catastrophic thinking about pain on attentional interference by pain: No mediation of negative
affectivity in healthy volunteers and in patients with low back pain. Pain Research and Management,
7, 31–9.
Crombez, G., Eccleston, C., Vlaeyen, J. W. S., Vansteenwegen, D., Lysens, R. and Eelen, P. (2002b)
Exposure to physical movements in low back pain patients: Restricted effects of generalization. Health
Crombez, G., Vervaet, L., Baeyens, F., Lysens, R. and Eelen, P. (1996) Do pain expectancies cause pain in
chronic low back patients? A clinical investigation. Behaviour Research and Therapy, 34, 919–25.
Crombez, G., Vervaet, L., Lysens, R., Baeyens, F. and Eelen, P. (1998b) Avoidance and confrontation of
painful, back-straining movements in chronic back pain patients. Behavior Modification, 22, 62–77.
Crombez, G., Vlaeyen, J. W. S., Heuts, P. H. T. G. and Lysens, R. (1999) Pain-related fear is more disabling
than pain itself: evidence on the role of pain-related fear in chronic back pain disability. Pain, 80,
329–39.
de Gier, M., Peters, M. L. and Vlaeyen, J. W. (2003) Fear of pain, physical performance, and attentional
processes in patients with fibromyalgia. Pain, 104, 121–30.
de Jong, J. R., Vangronsveld, K., Peters, M. L., et al. (2008) Reduction of pain-related fear and disability in
post-traumatic neck pain: a replicated single-case experimental study of exposure in vivo. Journal of
Pain, 9, 1123–34.
de Jong, J. R., Vlaeyen, J. W. S., Onghena, P., Cuypers, C., den Hollander, M. and Ruijgrok, J. (2005a)
Reduction of pain-related fear in complex regional pain syndrome type 1: the application of graded
exposure in vivo. Pain, 116, 264–75.
de Jong, J. R., Vlaeyen, J. W. S., Onghena, P., Goossens, M. E. J. B., Geilen, M. and Mulder, H. (2005b)
Fear of Movement/(Re)injury in Chronic Low Back Pain: Education or Exposure In Vivo as Mediator
to Fear Reduction? Clinical Journal of Pain, 21, 9–17.
De Peuter, S., Van Diest, I., Vansteenwegen, D., Van den Bergh, O., Vlaeyen, J. W. S. Pain-related fear and
chronic pain: Interoceptive fear conditioning as a novel approach. European Journal of Pain
(submitted manuscript: accepted pending revision).
DeGood, D. E. and Shutty, M. S. (1992) Assessment of pain beliefs, coping and self-efficacy, in D. C.Turk
and D.C. Melzack (ed.) Handbook of pain assessment. New York: Guilford.
Dehghani, M., Sharpe, L. and Nicholas, M. K. (2004) Modification of attentional biases in chronic pain
patients: a preliminary study. European Journal of Pain, 8, 585–94.
Eccleston, C. and Crombez, G. (1999) Pain demands attention: A cognitive-affective model of the
interruptive function of pain. Psychological Bulletin, 125, 356–66.
Eccleston, C., Crombez, G., Aldrich, S. and Stannard, C. (1997) Attention and somatic awareness in
chronic pain. Pain, 72, 209–15.
Edwards, R. R., Haythornthwaite, J. A., Sullivan, M. J. and Fillingim, R. B. (2004) Catastrophizing as a
mediator of sex differences in pain: differential effects for daily pain versus laboratory-induced pain.
Pain, 111, 335–41.
Flor, H., Miltner, W. and Birbaumer, N. (2001) Psychophysiological recording methods. In
D. C. T. Melzack (Ed.) Handbook of Pain Assessment. New York: Guilford.
Fordyce, W. E. (1976) Behavioural methods for chronic pain and illness, St. Louis, Mosby.
Fordyce, W. E., Shelton, J. L. and Dundore, D. E. (1982) The modification of avoidance learning in pain
behaviors. Journal of Behavioral Medicine, 5, 405–14.
Fritz, J. M., George, S. Z. and Delitto, A. (2001) The role of fear-avoidance beliefs in acute low back pain:
Geisser, M., Haig, A., Wallbom, A. and Wiggert, E. (2004) Pain-related fear, lumbar flexion, and dynamic
EMG among persons with chronic musculoskeletal low back pain. Clinical Journal of Pain 20, 61–9.
Geisser, M. E., Haig, A. J. and Theisen, M. E. (2000) Activity avoidance and function in persons with
chronic back pain. Journal of Occupational Rehabilitation, 10, 215–27.
George, S. Z., Dannecker, E. A. and Robinson, M. E. (2006) Fear of pain, not pain catastrophizing, predicts
acute pain intensity, but neither factor predicts tolerance or blood pressure reactivity: An experimental
investigation in pain-free individuals, 10, 457–65.
Giummarra, M. J., Gibson, S. J., Georgiou Karistianis, N. and Bradshaw, J. L. (2007) Central mechanisms
in phantom limb perception: the past, present and future. Brain Research Reviews, 54, 219–32.
Goubert, L., Crombez, G. and De Bourdeaudhuij, I. (2004a) Low back pain, disability and back pain myths
in a community sample: prevalence and interrelationships. European Journal of Pain, 8, 385–94.
Goubert, L., Crombez, G., Hermans, D. and Vanderstraeten, G. (2003) Implicit attitude towards pictures
of back-stressing activities in pain-free subjects and patients with low back pain. European Journal of
Pain, 7, 33–42.
Goubert, L., Crombez, G., Van Damme, S., Vlaeyen, J. W. S., Bijttebier, P. and Roelofs, J. (2004b)
Confirmatory factor analysis of the Tampa Scale for Kinesiophobia: invariant two-factor model across
low back pain patients and fibromyalgia patients. Clinical Journal of Pain, 20, 103–10.
Goubert, L., Francken, G., Crombez, G., Vansteenwegen, D. and Lysens, R. (2002) Exposure to physical
movement in chronic back pain patients: No evidence for generalization across different movements.
Behaviour Research and Therapy, 40, 415–29.
Grotle, M., Vollestad, N. K., Veierod, M. B. and Brox, J. I. (2004) Fear-avoidance beliefs and distress in
relation to disability in acute and chronic low back pain. Pain, 112, 343–52.
Haythornthwaite, J. A., Clark, M. R., Pappagallo, M. and Raja, S. N. (2003) Pain coping strategies play a
role in the persistence of pain in post-herpetic neuralgia. Pain, 106, 453–60.
Hirsh, A. T., George, S. Z., Bialosky, J. E. and Robinson, M. E. (2008) Fear of pain, pain catastrophizing,
and acute pain perception: relative prediction and timing of assessment. Journal of Pain, 9, 806–12.
Huis in ‘t Veld, R. M. H. A., Vollenbroek-Hutten, M. M. R., Groothuis-Oudshoorn, K. C. G. M. and
Hermenss H. J. (2007) The role of the fear-avoidance model in female workers with neck-shoulder
pain related to computer work. Clinical Journal of Pain, 23, 28–34.
Izard, C. E. (1992) Basic emotions, relations among emotions, and emotion-cognition relations.
Psychological Review, 99, 561–5.
Jamison, R. N., Raymond, S. A., Levine, J. G., Slawsby, E. A., Nedeljkovic, S. S. and Katz, N. P. (2001)
Electronic diaries for monitoring chronic pain: 1-year validation study. Pain, 91, 277–85.
Jensen, M. P., Brant-Zawadzki, M., Obuchowski, N., Modic, M., Malkasian, D. and Ross, J. (1994)
Magnetic resonance imaging of the lumbar spine in people without back pain. New England Journal of
Medicine, 331, 69–73.
Karsdorp, P. A., Nijst, S. E., Goossens, M. E. and Vlaeyen, J. W. S. (2009) The role of current mood and
stop rules on physical task performance: An experimental investigation in patients with work-related
upper extremity pain. European Journal of Pain, 14(4), 434–40.
Kazdin, A. E. (2003) Research design in clinical psychology. Boston, MA: Allyn & Bacon.
Keogh, E. and Asmundson, G. J. G. (2004) Negative affectivity, catastrophizing, and anxiety sensitivity. In
G. J. G.Asmundson, J. W. S.Vlaeyen and G. Crombez, G. (ed.) Understanding and treating fear of pain.
New York: Oxford University Press.
Keogh, E., Ellery, D., Hunt, C. and Hannent, I. (2001) Selective attentional bias for pain-related stimuli
amongst pain fearful individuals. Pain, 91, 91–100.
Klenerman, L., Slade, P. D., Stanley, I. M., et al. (1995) The prediction of chronicity in patients with an
acute attack of low back pain in a general practice setting. Spine, 20, 478–84.
Kori, S. H., Miller, R. and Todd, D. (1990) Kinesiophobia: a new view of chronic pain behavior. Pain
Management, Jan/Feb, 35–43.
Kugler, K., Wijn, J., Geilen, M., de Jong, J. and Vlaeyen, J. W. S. (1999) The photograph series of daily
activities (PHODA). CD-rom version 1.0., Heerlen, The Netherlands: Institute for Rehabilitation
Research and School for Physiotherapy.
Leeuw, M., Goossens, M. E., Linton, S. J., Crombez, G., Boersma, K. and Vlaeyen, J. W. (2007a) The
fear-avoidance model of musculoskeletal pain: current state of scientific evidence. Journal of Behavioral
Leeuw, M., Goossens, M. E. J. B., van Breukelen, G. J. P., Boersma, K. and Vlaeyen, J. W. S. (2007b)
Measuring perceived harmfulness of physical activities in patients with chronic low back pain: The
photograph series of daily activities – Short electronic version. Journal of Pain, 8, 840–49.
Leeuw, M., Goossens, M. E. J. B., van Breukelen, G. J. P., et al. (2008) Exposure in vivo versus operant
graded activity in chronic low back pain patients: Results of a randomized controlled trial. Pain,
138, 192–207.
Leeuw, M., Houben, R. M., Severeijns, R., Picavet, H. S., Schouten, E. G. and Vlaeyen, J. W. (2007c)
Pain-related fear in low back pain: a prospective study in the general population. European Journal of
Pain, 11, 256–66.
Leeuw, M., Peters, M. L., Wiers, R. W. and Vlaeyen, J. W. (2007d) Measuring Fear of Movement/(Re)
injury in Chronic Low Back Pain Using Implicit Measures. Cognitive Behaviour Therapy, 36, 52–64.
Lethem, J., Slade, P. D., Troup, J. D. and Bentley, G. (1983) Outline of a Fear-Avoidance Model of
exaggerated pain perception—I. Behaviour Research and Therapy, 21, 401–8.
Linton, S. J. (1998) The socioeconomic impact of chronic back pain: is anyone benefiting? Pain, 75, 163–68.
Linton, S. J., Boersma, K., Jansson, M., Overmeer, T., Lindblom, K. and Vlaeyen, J. W. S. (2008)
A randomized controlled trial of exposure in vivo for patients with spinal pain reporting fear of
work-related activities. European Journal of Pain, 12, 722–30.
Linton, S. J., Overmeer, T., Janson, M., Vlaeyen, J. W. S. and de Jong, J. R. (2002) Graded in-vivo exposure
treatment for fear-avoidant pain patients with functional disability: A case study. Cognitive Behaviour
Therapy, 31, 49–58.
Lohnberg, J. (2007) A Review of Outcome Studies on Cognitive-Behavioral Therapy for Reducing
Fear-Avoidance Beliefs Among Individuals With Chronic Pain. Journal of Clinical Psychology in
McCracken, L. M. and Dhingra, L. (2002) A short version of the Pain Anxiety Symptoms Scale
(PASS—20): Preliminary development and validity. Pain Research and Management, 7, 45–50.
McCracken, L. M., Gross, R. T., Aikens, J. and Carnrike, C. L. M., Jr. (1996) The assessment of anxiety
and fear in persons with chronic pain: A comparison of instruments. Behaviour Research and Therapy,
34, 927–33.
McCracken, L. M., Zayfert, C. and Gross, R. T. (1992) The pain anxiety symptoms scale: development and
validation of a scale to measure fear of pain. Pain, 50, 67–73.
McNeil, D. W. and Rainwater, A. J. (1998) Development of the Fear of Pain Questionnaire—III. Journal of
Behavioral Medicine, 21, 389–410.
McNeil, D. W. and Vowles, K. E. (2004) Assessment of fear and anxiety associated with pain:
Conceptualization, methods, and measures, in G. J. G.Asmundson, J. W. S. Vlaeyen, and G. Crombez
(ed) Understanding and treating fear of pain. New York: Oxford University Press.
Michael, E. S. and Bruns, J. W. (2004). Catastrophizing and pain sensitivity among chronic pain patients:
moderating effects of sensory and affect focus. Annals of Behavioral Medicine, 27, 185–94.
Moore, J. E., Von Korff, M., Cherkin, D., Saunders, K. and Lorig, K. (2000) A randomized trial of
cognitive-behavioral program for enhancing back pain self care in a primary care setting. Pain,
88, 145–53.
Morley, S. and Eccleston, C. (2004) The object of fear in pain, in G. J. G. Asmundson, J. W. S. Vlaeyen, and
G. Crombez (ed.) Understanding and treating fear of pain. New York: Oxford University Press.
Mowrer, O. H. (1947) On the dual nature of learning: a reinterpretation of ‘conditioning’ an
‘problem-solving’. Harvard Educational Review, 17, 102–50.
Nachemson, A. L., Waddell, G. and Norlund, A. L. (2000) Epidemiology of neck and low back pain, in
A. L. Nachemson and E. Johnsson (ed.) Neck and back pain: the scientific evidence of causes, diagnosis,
and treatment. Philadelphia, PA: Lippincott, Williams, and Wilkins.
Naliboff, B. D., Waters, A. M., Labus, J. S., et al. (2008) Increased acoustic startle responses in ibs patients
during abdominal and nonabdominal threat. Psychosomatic Medicine, 70, 920–27.
Peters, M. L., Sorbi, M. J., Kruise, D. A., Kerssens, J. J., Veerhak, P. F. M. and Bensing, J. M. (2000)
Electronic diary assessment of pain, disability and psychological adaptation in patients differing in
duration of pain. Pain, 84, 181–92.
Peters, M. L., Vlaeyen, J. W. and Weber, W. E. (2005) The joint contribution of physical pathology,
pain-related fear and catastrophizing to chronic back pain disability. Pain, 113, 45–50.
Peters, M. L., Vlaeyen, J. W. S. and Kunnen, A. M. W. (2002) Is pain-related fear a predictor of somatosensory
hypervigilance in chronic low back pain patients? Behaviour Research and Therapy, 40, 85–103.
Philips, H. C. (1987) Avoidance behaviour and its role in sustaining chronic pain. Behaviour Research and
Therapy, 25, 273–9.
Phillips, C., Main, C. J., Buck, R., Aylward, M., Wynne-Jones, G. and Farr, A. (2008) Prioristising pain in
policy making: The need for a whole systems perspective. Health Policy, 88, 166–75.
Picavet, H. S. and Schouten, J. S. (2003) Musculoskeletal pain in the Netherlands: prevalences,
consequences and risk groups, the DMC(3)-study. Pain, 102, 167–78.
Pincus, T. and Morley, S. (2001) Cognitive-processing bias in chronic pain: A review and integration.
Pool, J. J. M., Hiralal, S., Ostelo, R. W. J. G., et al. (2009). The applicability of the Tampa Scale of
Kinesiophobia for patients with sub-acute neck pain. Quality and Quantity, 43, 773–80.
Rachmann, S. (2004) Anxiety. Hove: Psychological Press.
Reneman, M. F., Jorritsma, W., Dijkstra, S. J. and Dijkstra, P. U. (2003) Relationship between
kinesiophobia and performance in a functional capacity evaluation. Journal of Occupational
Reneman, M. F., Schiphorst Preuper, H. R., Kleen, M., Geertzen, J. H. B. and Dijkstra, P.U. (2007) Are
pain intensity and pain related fear related to functional capacity evaluation performances of patients
with chronic low back pain? Journal of Occupational Rehabilitation, 17, 247–58.
Riley, J. F., Ahern, D. K. and Follnik, M. J. (1988) Chronic Pain and Functional Impairment: Assessing
Beliefs About Their Relationship. Archives of Physical Medicine and Rehabilitation, 69, 579–82.
Roelofs, J., McCracken, L. M., Peters, M. L., Crombez, G., van Breukelen, G. and Vlaeyen, J. W. S. (2004a)
Psychometric evaluation of the Pain Anxiety Symptoms Scale (PASS) in chronic pain patients. Journal
of Behavioral Medicine, 27, 167–83.
Roelofs, J., Peters, M. L., Fassaert, T. and Vlaeyen, J. W. (2005) The role of fear of movement and injury in
selective attentional processing in patients with chronic low back pain: a dot-probe evaluation. Journal
of Pain, 6, 294–300.
Roelofs, J., Peters, M. L., van der Zijden, M. and Vlaeyen, J. W. (2004b) Does fear of pain moderate the
effects of sensory focusing and distraction on cold pressor pain in pain-free individuals? Journal of
Pain, 5, 250–6.
Roelofs, J., Peters, M. L. and Vlaeyen, J. W. S. (2003) The modified Stroop paradigm as a measure of
selective attention towards pain-related information in patients with chronic low back pain.
Psychological Reports, 92, 707–15.
Roelofs, J., Peters, M. L., Zeegers, M. P. A. and Vlaeyen, J. W. S. (2002) The modified stroop paradigm
as a measure of selective attention towards pain-related stimuli among chronic pain patients: a
Roelofs, J., Sluiter, J. K., Frings-Dresen, M. H. W., et al. (2007) Fear of movement and (re)injury in chronis
musculoskeletal pain: Evidence for an invariant two-factor model of the Tampa Scale for
Kinesiophobia across pain diagnoses and Dutch, Swedish, and Canadian samples. Pain, 131, 181–90.
Schmidt, C. O., Raspe, H., Pfingsten, M., et al. (2007) Back pain in the german adult population:
Prevalence, severity, and sociodemographic correlates in a multiregional survey. Spine, 32, 2005–11.
Severeijns, R., van den Hout, M. A. and Vlaeyen, J. W. S. (2005) The causal status of pain catastrophizing:
An experimental test with healthy participants. European Journal of Pain, 9, 257–65.
Sigmon, S. T. and LaMattina, S. M. (2006) Self-assessment, in M.Hersen (ed.) Clinician’s Handbook of
Adult Behavioral Assessment. Oxford: Academic Press.
Smeets, R. J. E. M., Wittink, H., Hidding, A. and Knottnerus, J. A. (2006) Do patients with chronic low
back pain have a lower level of aerobic fitness than healthy controls? : Are pain, disability, fear of
injury, working status, or level of leisure time activity associated with the difference in aerobic fitness
level? Spine, 31, 90–8.
Stegen, K., Van Diest, I., Van de Woestijne, K. P. and Van den Bergh, O. (2001) Do persons with negative
affect have an attentional bias to bodily sensations? Cognition and Emotion, 15, 813–29.
Stewart, S. H. and Asmundson, G. J. G. (2006) Anxiety sensitivity and its impact on pain experiences and
conditions: a state of the art. Cognitive Behaviour Therapy, 35, 185–8.
Sullivan, M. J., Adams, H. and Sullivan, M. E. (2004) Communicative dimensions of pain catastrophizing:
social cueing effects on pain behaviour and coping. Pain, 107, 220–6.
Sullivan, M. J., Adams, H., Tripp, D. A. and Stanish, W. (2008) Stage of chronicity and treatment response
in patients with musculoskeletal injuries and concurrent symptoms of depression. Pain, 135, 151–9.
Sullivan, M. J. L., Bishop, S. R. and Pivik, J. (1995) The Pain Catastrophizing Scale: Development and
validation. Psychological Assessment, 7, 524–32.
Sullivan, M. J. L., Lynch, M. E. and Clark, A. J. (2005) Dimensions of catastrophic thinking associated with
pain experience and disability in patients with neuropathic pain conditions. Pain, 113, 310–15.
Swinkels-Meewisse, E. J. C. M., Swinkels, R. A. H. M., Verbeek, A. L. M., Vlaeyen, J. W. S. and
Oostendorp, R. A. B. (2003) Psychometric properties of the Tampa Scale for kinesiophobia and the
fear-avoidance beliefs questionnaire in acute low back pain. Manual Therapy, 8, 29–36.
Swinkels-Meewisse, I. E., Roelofs, J., Schouten, E. G., Verbeek, A. L., Oostendorp, R. A. and Vlaeyen, J. W.
(2006) Fear of movement/(re)injury predicting chronic disabling low back pain: a prospective
inception cohort study. Spine, 31, 658–64.
Tang, N. K., Salkovskis, P. M., Poplavskaya, E., Wright, K. J., Hanna, M. and Hester, J. (2007) Increased
use of safety-seeking behaviors in chronic back pain patients with high health anxiety. Behaviour
Trost, Z., France, C. R. and Thomas, J. S. (2009) Examination of the photograph series of daily activities
(PHODA) scale in chronic low back pain patients with high and low kinesiophobia. Pain, 141, 276–82.
Turk, D. C. and Flor, H. (1999) Chronic pain: A biobehavioral perspective. Psychosocial factors in pain:
Critical perspectives. New York: Guilford Press.
Turk, D. C. and Gatchel, R. J. (2002) Psychological approaches to pain management: A practitioner’s
handbook. New York: Guildford.
Turk, D. C., Robinson, J. P., Sherman, J. J., Burwinkle, T. and Swanson, K. (2008) Assessing fear in
patients with cervical pain: development and validation of the Pictorial Fear of Activity Scale-Cervical
(PFActS-C). Pain, 139, 55–62.
Turner, J. A., Franklin, G., Fulton-Kehoe, D., et al. (2006) Worker recovery expectations and
fear-avoidance predict work disability in a population-based workers’ compensation back pain sample.
Spine, 31 682–9.
Turner, J. A., Jensen, M. P., Warms, C. A. and Cardenas, D. D. (2002) Catastrophizing is associated with
pain intensity, psychological distress, and pain-related disability among individuals with chronic pain
after spinal cord injury. Pain, 98, 127–34.
Turner, J. A., Mancl, L. and Aaron, L. A. (2004) Pain-related catastrophizing: a daily process study. Pain,
110, 103–11.
Van Damme, S., Crombez, G. and Eccleston, C. (2002) Retarded disengagement from pain cues: the effects
of pain catastrophizing and pain expectancy. Pain, 100, 111–18.
Van Damme, S., Crombez, G., Eccleston, C. and Roelofs, J. (2004) The role of hypervigilance in the
experience of pain, in G. J. G.Asmundson, J. W. S. Vlaeyen, and G. Crombez (ed.) Understanding and
treating fear of pain. New York: Oxford University Press.
Van Damme, S, Crombez, G. and Eccleston, C. (2008) Coping with pain: A motivational perspective. Pain,
139, 1–4.
Vancleef, L. M. G. and Peters, M. L. (2006) The interruptive effect of pain on attention. Journal of Pain,
7, 21–2.
Vancleef, L. M. G., Peters, M. L., Roelofs, J. and Asmundson, G. J. G. (2006) Do fundamental fears
differentially contribute to pain-related fear and pain catastrophizing? An evaluation of the sensitivity
index. European Journal of Pain, 10, 527–36.
Vancleef, L. M. G., Vlaeyen, J. W. and Peters, M. L. (2009) Dimensional and Componential Structure of a
Hierarchical Organisation of Pain-related Anxiety Constructs. Psychological Assessment, 23, 340–51.
Vansteenwegen, D., Iberico, C., Vervliet, B., Marescau, V. and Hermans, D. (2008) Contextual fear
induced by unpredictability in a human fear conditioning preparation is related to the chronic
expectation of a threatening US. Biological Psychology, 77, 39–46.
Verbunt, J., Seelen, H., Vlaeyen, J., et al. (2003) Disuse and deconditioning in chronic low back pain:
Concepts and hypotheses on contributing mechanisms. European Journal of Pain, 7, 9–21.
Verhaak, P. F., Kerssens, J. J., Dekker, J., Sorbi, M. J. and Bensing, J. M. (1998) Prevalence of chronic
benign pain disorder among adults: a review of the literature. Pain, 77, 231–9.
Vlaeyen, J. W., De Jong, J., Leeuw, M. and Crombez, G. (2004a) Fear reduction in chronic pain: graded
exposure in vivo with behavioural experiments, in G. J. G.Asmundson, J. W. S. Vlaeyen, and
G. Crombez (ed.) Understanding and treating fear of pain. New York: Oxford University Press.
Vlaeyen, J. W., Kole-Snijders, A. M. J., Boeren, A. M. and Van Eek, H. (1995a) Fear of movemement/(re)
injury in chronic low back pain and its relation to behavioral performance. Pain, 62, 363–72.
Vlaeyen, J. W., Timmermans, C., Rodriguez, L. M., et al. (2004b) Catastrophic thinking about pain
increases discomfort during internal atrial cardioversion. Journal of Psychosomatic Research,
56, 139–44.
Vlaeyen, J. W. S., de Jong, J., Heuts, P. H. T. G. and Crombez, G. (2008) Graded exposure in vivo for
pain-related fear, in H. Breivik, W. Campbell and M. K. Nicholas (ed.) Clinical Pain Management:
Practice and procedures. 2nd edn. London, Hodder & Stoughton.
Vlaeyen, J. W. S., de Jong, J. R., Geilen, M., Heuts, P. and van Breukelen, G. (2001) Graded exposure
in vivo in the treatment of pain-related fear: a replicated single-case experimental design in four
patients with low back pain.. Behaviour Research and Therapy, 39, 151–66.
Vlaeyen, J. W. S., Kole-Snijders, A. M. J., Rotteveel, A. M., Ruesink, R. et al. (1995b) The role of fear of
movement/(re)injury in pain disability. Journal of Occupational Rehabilitation, 5, 235–52.
Vlaeyen, J. W. S. and Linton, S. J. (2000) Fear-avoidance and its consequences in chronic musculoskeletal
pain: A state of the art. Pain, 85, 317–32.
Vlaeyen, J. W. and Morley, S. (2004). Active despite pain: The putative role of stop-rules and current
mood. Pain, 110, 512–16.
Vowles, K. E. and Gross, R. T. (2003) Work-related beliefs about injury and physical capability for work in
individuals with chronic pain. Pain, 101, 291–8.
Waddel, G. (2004) The biopsychosocial model, in G. Waddel (ed.) The back pain revolution. Edinburgh:
Waddell, G. (1987) Volvo award in clinical sciences. A new clinical model for the treatment of low back
pain. Spine, 12, 632–44.
Waddell, G. (2004) The epidemiology of back pain, in G. Waddel (ed.) The back pain revolution.
Edinburgh: Churchill Livingstone.
Waddell, G., Newton, M., Henderson, I., et al. (1993) A Fear-Avoidance Beliefs Questionnaire (FABQ) and
the role of fear-avoidance beliefs in chronic low back pain and disability. Pain, 52, 157–68.
Walker, B. F. (2000) The prevalence of low back pain: a systematic review of the literature from 1966 to
1998. Journal of Spinal Disorders, 13, 205–17.
Walker, B. F., Muller, R. and Grant, W. D. (2004) Low back pain in Australian adults. Prevalence and
associated disability. Journal of Manipulative Physiology and Therapy, 27, 238–44.
Walker, J. R. and Furer, P. (2008) Interoceptive exposure in the treatment of health anxiety and
hypochondriasis. Journal of Cognitive Psychotherapy, 22, 366–78.
Weiner, D., Pieper, C., McConnell, E., Martinez, S. and Keefe, F. (1996) Pain measurement in elders with
chronic low back pain: traditional and alternative approaches. Pain, 67, 461–7.
Wittink, H., Hoskins Michel, T. H., Wagner, A., Sukiennik, A. and Rogers, W. (2000) Deconditioning in
patients with chronic low back pain: fact or fiction? Spine, 25, 2221–8.
Woods, M. P. and Asmundson, G. J. G. (2008) Evaluating the efficacy of graded in vivo exposure for the
treatment of fear in patients with chronic back pain: A randomized controlled trial. Pain, 136, 271–80.
Chapter 15
Endurance-Related Pain Responses in

the Development of Chronic Back Pain
Monika I. Hasenbring, Dirk Hallner, and
Adina C. Rusu
15.1 Introduction
During the past two decades, a significant amount of evidence has supported the impact of
pain-related cognitive, affective, and behavioural responses on the development and mainte-
nance of chronic musculoskeletal pain (Turk and Rudy 1992; Waddel 2004; Gatchel et al. 2007).
For example, catastrophizing and pain-related fear were identified as important mediators of pain
and disability (Linton 2000; Sullivan et al. 2001; Turner and Aaron 2001). The fear-avoidance
model (FAM; Vlaeyen and Linton 2000) is a stimulating approach, postulating a pathway between
these pain responses, physical disuse and chronic pain (Verbunt et al. 2003, see detailed discus-
sion of the FAM in Chapter 14). However, recently, there is growing evidence that a second,
potentially opposite pathway will also lead to the development and maintenance of chronic pain.
This alternative pathway is based upon endurance-related responses and includes physical over-
use or overload instead of physical disuse as main mediators (Bousema et al. 2007; Hasenbring
et al. 2006; Smeets et al. 2007). The avoidance-endurance model (AEM) proposes several path-
ways for chronicity of pain based on either endurance- or fear avoidance-related responses. The
following review will at first focus on research concerning the frequency of self-reported endur-
ance-related responses to pain, as a complement to the well-known fear-avoidance responses.
Second, basic assumptions underlying the conceptualization of the AEM will be outlined. This
will be followed by a review of existing empirical evidence supporting hypotheses related to the
AEM and its consequences for the performance of daily activities will be presented. Finally, clini-
cal implications, such as proposed modifications for behavioural interventions will be described.
15.2 The occurrence of endurance-related pain responses

Pain demands attention, specifically when it occurs suddenly. In an acute pain situation this
symptom could be an indicator of a potential medical problem that has to be solved. Pain how-
ever, often occurs in situations where people are engaged in several competing occupational,
house work, and leisure activities. Therefore, at least two goals may be activated simultaneously—
finding an explanation and solution for the pain problem on the one hand and to maintain cur-
rent activities on the other. Due to the model of limited mental resources in goal-directed
behaviour (Norman and Bobrow 1975), these competing goals may result in a goal-conflict trig-
gering a goal-shielding mechanism, in which commitment to a focal goal inhibits the availability
of alternative goals and distracting information (Fishbach and Fergusan 2007; Goschke and
Dreisbach 2008). For example, suddenly occurring, intensive pain that occurs during house-
work may become pre-potent leading to an interruption of ongoing activities and result in calling
a doctor. In contrast, when several weeks later such an acute pain episode is resolving, recurrent
episodes of milder pain may not have this interruptive quality. It more and more depends on the
resulting individual goal interference, potential habitual mechanisms resolving these conflicts, as
well as on situational demands, as to whether attention will be directed to this pain stimulus or to
alternative activities.
Only recently, evidence has been acquired related to the frequency of attentional responses to
pain symptoms, when these occur during other activities. When people are asked to evaluate
coping strategies, distraction is rated as highly effective and is preferred over alternative tech-
niques (McCaul and Haugtvedt 1982; Ahles et al. 1983). In a laboratory experiment, McCaul and
Haugvedt (1982) reported that 80% of their subjects preferred distraction over attending sensa-
tions for coping with cold pressor pain, whereas 92% of subjects who were instructed to attend to
the concrete sensations of the pain stimulus, reported less distress than subjects in the distraction
condition. That means, distraction was preferred by most of the people although it was not an
effective coping strategy, especially for severe pain (McCaul and Mallot 1984). The question
arises, how people behave in their daily life, when confronted with pain? Do they divide their
attention between the pain stimulus and their current activity? How does emotional arousal
impact on the perception of pain, and what do individuals do in order to reduce pain?
Twenty-five years ago, Philips and Jahanshahi (1986) proposed avoidance as the most frequent
and prominent behavioural response to pain. Their evaluation was based on a systematic com-
parison of three different pain related responses (avoidance, complaint, and help-seeking behav-
iour) in patients with chronic headache and low back pain (LBP). This study established the focus
on an important maladaptive pattern of pain responses that was further conceptualized within
several FAMs of pain (Lethem et al. 1983; Philips 1987; Asmundson et al. 1999; Vlaeyen and
Linton 2000). Within this context, fear-avoidance responses (FAR) consist of avoidance of
pain-associated activities on the behavioural level, elicited by fear (e.g. fear of pain-related move-
ments) on the affective level and automatic thoughts, such as catastrophizing or more generalized
appraisals (e.g. fear-avoidance beliefs) on the cognitive level.
However, focusing on coping responses, which encompasses both pain-related thoughts and
behaviour, several studies have shown that task persistent behaviour despite pain and thought
suppression seem to be as frequent as catastrophizing and avoidance. Using the Coping Strategies
Questionnaire (CSQ), Rosenstiel and Keefe (1983) found that patients with chronic pain reported
that they were active despite pain or ignored pain in the same frequency as they reported catastro-
phizing about pain. This finding was confirmed by others (Turner and Clancy 1986 ;
Haythornetwaite et al 1998; Garcia-Campayo et al. 2007). Within the AEM of pain (described
below), cognitions involving thought suppression, distraction or minimization, task persistence
behaviour in spite of severe pain and potentially positive mood despite pain represent central
aspects of pain-related endurance responses (ERs). Pain-related endurance is motivated by the
goal of maintaining current activities despite feeling pain, either by struggling against the poten-
tially alarming character of pain (thought suppression) or by successfully distracting from pain
(Hasenbring 2000).
Hasenbring (1993) and Grebner et al. (1999) investigated the occurrence of fear-avoidance-
versus endurance-related responses in samples of acute and subacute back and leg pain prior to
lumbar disc surgery. Using the Kiel Pain Inventory (Hasenbring 1994), both research groups
reported cognitions of thought suppression with at least the same frequency as helplessness,
hopelessness, and catastrophizing. An identical finding was found for the relation of avoidance
behaviour compared to endurance behaviour despite severe pain. Recently, in a sample of chronic
back pain patients, using a Chinese version of the Chronic Pain Coping Inventory (CPCI, Jensen
et al. 1995), Wong et al. (2010) found different coping strategies, such as guarding, resting, asking
ENDURANCE-RELATED PAIN RESPONSES 297
for assistance, relaxation and task persistence, were rated concerning the number of days over the
past weeks that have been used for each of these strategies at least once. The authors reported the
highest frequency scores for task persistence behaviour (mean 4.32, SD 1.75) compared to typical
FA responses such as guarding (mean 3.06, SD 1.84) or resting (mean 3.64, SD 1.93). Pain
responses that are to be seen as adaptive, such as relaxation (mean 1.94, SD 1.70) and exercise
(mean 2.52, SD 1.89) were rated much lower. Concerning mood, in acute and subacute as well as
in chronic LBP, positive mood despite pain was rated with higher frequency than anxiety and
depressive mood (Hasenbring et al. 1994; Hasenbring et al. 2009).
Taken together, when confronted with acute, subacute, or chronic LBP in daily life, endur-
ance responses such as positive mood despite pain, cognitions of distraction and thought sup-
pression as well as task persistence behaviour seem to be more often occurring than FAR with
anxious and depressive mood, cognitions of catastrophizing, helplessness/hopelessness, and
avoidance behaviour. Whereas there is substantial evidence that FAR leads to increased pain-
related disability and depression over time, the question arises as to the consequences following
endurance responses.
15.3 Theoretical considerations: the Avoidance-Endurance

Model (AEM) of pain
The AEM of pain was developed in order to address FAR as well as ER as important mediators in
the development and maintenance of chronic pain and disability (Hasenbring 2000). While
earlier formulations of this model primarily focused on potential maladaptive consequences
of these different pain responses, the current version of the AEM also hypothesizes adaptive
short-term consequences (positive or negative reinforcement) that may be responsible for the
maintenance of these reactions.
15.3.1 Basic assumptions

Basic assumptions underlying the AEM are as follows. In accordance with the psychophysiologi-
cal model of anxiety (Lang 1970) loose relations are hypothesized for the cognitive, affective, and
behavioural responses to pain. Further, comparable to the Dynamic Model of Affect (DMA, Davis
and Zautra 2004), an opposite pattern of FAR and ER are expected to exist on distinct dimen-
sions, implying that opposite features can be activated independently. This assumption is sup-
ported by evidence of distinct neural systems for opposite affects and cognitions (e.g. Hass and
Canli, 2008). We hypothesize that each individual will show a specific pattern of cognitive, affec-
tive, and behavioural pain responses when confronted with pain and that only specific, highly
rigid and time-stable pattern of cognitive, affective, and behavioural pain responses will contrib-
ute to the maintenance of pain in the long-term run. We suggest that FAR represent a specific and
highly rigid pattern of pain-related fear, cognitions of catastrophizing and helplessness, hopeless-
ness, and behavioural avoidance (FAR pattern). In contrast, the AEM suggests that due to ER at
least two different response pattern can be described influencing the maintenance of pain and
disability. The distress-endurance pattern (DER) refers to marked thought suppression, emotions
of anxiety and depression and to task persistence in spite of pain on the behavioural level. The
eustress-endurance pattern (EER) refers to ignoring pain sensations and minimizing the meaning
of pain experiences and marked task persistence behaviour, often accompanied by high scores on
positive mood despite pain. Furthermore, an adaptive response (AR) pattern is suggested to be
characterized by cognitions of sensory monitoring without ongoing negative affects, by a high
degree of flexibility between avoidance and endurance responses to pain, supporting the effect
of biomedical treatments. Finally, we suggest that there are a number of psychoneurobiological
pathways from acute to chronic pain for FAR and ER that can be explained via learning condi-
tioning theory in the long-term as well as maintaining mechanisms in the short-term.
15.3.2 Fear-avoidance responses and physical disuse

The pathway between catastrophizing, fear of pain, avoidance behaviour, and physical disuse is
conceptualized within several FAMs in the literature (see for more detail Chapter 14). Patients,
that will exclusively show FAR, will tend to avoid all physical and social activities that could elicit
pain. Within this pattern, avoidance behaviour is assumed to be guided by conditioned stimuli
(e.g. the imagination of a painful lifting activity elicits fear), shown preventively in order to avoid
experience of pain (active avoidance). People will avoid physical activities that they believe will
lead to increased back pain (e.g. heavy lifting, walking) as well as social events that are potentially
pain provoking (e.g. sitting in a film theatre, having guests for dinner, playing tennis). In the
short-term, preventive avoidance behaviour will lead to negative reinforcement, when pain is
diminishing or when an increase of pain is lacking. As indicated in Figure 15.1, over time, decreas-
ing physical activity will cause physical deconditioning including negative changes in muscular,
motor, cardiorespiratory and metabolic aspects of physical fitness (Verbunt et al. 2003). Due to
neurobiological processes of central sensitization, physical structures respond with pain during
normal activities (Bortz 1984; see also Chapter 4 in this volume). The AEM assumes a further
pathway between avoidance of social activities and the development of chronic pain via an
increase in depression, mediated by a loss of social reinforcement and increased social isolation
(Lewinsohn and Libet, 1972). These differential hypotheses concerning avoidance of physical
versus social activities were supported by evidence of positive associations between the subscale
Avoidance of Social Activities of the Avoidance-Endurance Questionnaire (AEQ; Hasenbring
et al. 2009) and BDI depression and between the Avoidance of Physical Activities subscale and
fear-avoidance beliefs concerning physical activities, measured with the Fear-Avoidance-Beliefs
Questionnaire (FABQ), Waddell Physical Activity (Hasenbring et al. 2009).
Biomechanical
load
Inactivity/
disability
Disuse syndrome
Pain and disability
Preventive
physical avoidance Sensitivity to normal
biomechanical load
Preventive
social avoidance Biomechanical load
Distress/
Hypervigilance depression
Avoidance
behavior Pacing behavior
Pain
Fear of pain
Coping signal
Catastrophizing ‘Pain is a sign to making a
‘It is not a serious illness, is it?’ short break and continue
afterwards’
Leads to an increase of… Leads to a decrease of…

Fig. 15.1 (See also Colour Plate 9) Maintenance of pain and disability by fear-avoidance responses
to pain.
15.3.3 Endurance-response pattern and physical overuse/overload

Patients that will exclusively show an ER pattern with high scores on endurance behaviour despite
severe pain (irrespective of the distress or eustress version) will have a high risk for physical over-
activity leading to overuse or overload of physical structures. In a situation, in which a pain-
related interruption of daily activities, normally taken into account, will be disdained or
disregarded, overuse (based on repetitive movements) or overload (based on ongoing static strain
postures) will result in a number of damages of muscle fibres, nerves, bones, and ligaments. These
behaviour-related damages as well as the repetitive experience of pain will maintain pain and
contribute to neurobiological mechanisms of peripheral and central sensitization. In the post-
injury acute or subacute phase of pain, an ER pattern will disturb the normal recovery period, and
can consequently result in a higher risk for developing chronic pain problems.
DER and EER patterns differ mainly in their cognitive and affective responses. Patients with a
DER pattern tend to show thought suppression combined with an increased depressive mood.
Thought suppression represents a cognitive response that blocks the perception of pain or the
interruption of daily activities normally demanded by pain (Eccleston and Crombez 1999).
Thought suppression attempts are often disorganized, non-focused, effortful, and at risk for fail-
ure. It may be reasonable, that thought suppression is followed by processes of intermittent rein-
forcement in the short term. Sometimes a successful suppression of a pain experience and
concentration on concurrent daily activities may be possible whereas at other times pain will
repeatedly come into one’s focus of attention. Frequent failure of thought suppression, however,
may lead to a decreased perception of self-efficacy and to an increased emotional distress.
Moreover, a rebound phenomenon was suggested (Wegner et al. 1987) leading to more fre-
quently occurring pain-related thoughts accompanied by the perception of additional failure,
feelings of helplessness-, hopelessness, and depressive mood (see Figure 15.2).
In contrast, patients with an EER pattern tend to pain-related cognitive responses of distraction
from pain or minimizing the meaning of pain experience and marked positive mood despite
severe pain. Distraction from pain might be more successful than thought suppression as it rep-
resents a more organized and focused way of attention diversion (Cioffi and Holloway 1993).
Biomechanical
load
Pain and disability
Endurance behavior
Biomechanical load
Distress/
depressive mood
Distress
Failure in
distraction/ Pain Pacing behavior
Rebound
or
Successful distraction
Coping signal
‘Pain is an important sign to
Thought suppression
making a short break and
‘Pull yourself together!’
continue afterwards’

Fig. 15.2 (See also Colour Plate 10) Maintenance of pain by distress-endurance responses to pain.
Biomechanical
Pain and disability
load
Endurance behavior
Biomechanical load
Positive mood/
feeling of control/
self-efficacy
Pain Pacing behavior

Successful distraction
Coping signal
Minimization ‘Pain is an important sign to
‘Oh, that’s nothing special’ making a short break and
continue afterwards’

Fig. 15.3 (See also Colour Plate 11) Maintenance of pain by eustress-endurance responses (EER) to pain.
Successful distraction and maintaining concentration towards concurrent daily activities may
lead to feelings of high self-control, high self-efficacy, and positive mood in the short term (see
Figure 15.3). A rebound effect will occur less often or never. However, due to ongoing task per-
sistence behaviour, these patients also tend to overlook pain experiences caused by strain-induced
small and repetitive damages of soft tissues, such as muscles, ligaments and tendons and reveal
overuse and overload in the long-term.
15.3.4 Adaptive pain responses and balancing of physical/mental load

The AEM assumes that there is an adaptive way to respond to back pain sensations that leads to a
health-promoting balancing of activities including physical, mental, and social activities. In the
phase of acute/subacute pain, when pain still has the function of a signal indicating some kind of
physical damage, cognitions labelled as coping signal lead to a conscious awareness of individual
needs. In the case, one has sat a long time, pain may be attributed as a signal to making a short
break standing up and move a little. Lingering upon a static strain position, back pain normally
increases over time. The earlier the break is realized, the less time is needed to reach a sensible
relief in pain.
Imagine different people suffering from back pain who are invited to a champagne reception of
a friend where all guests will stay the whole time. Normally, no seats will be available at this sort
of party. All these individuals had experienced a heavy increase of pain in such a situation some
time before. The AEM suggests that a man showing a pattern of FAR will be afraid of this pain
increase and therefore decline this invitation. Another man showing a pattern of DER will visit
this party despite this pain experience. When during standing and talking with friends pain will
increase (for instance from a level of ‘3’ at the beginning of the party up to a level of ‘5’) his
automatic thoughts may be as follows: ‘Don’t make such a fuss, an Indian does not know pain.
You cannot interrupt this talk at this moment!’. He will continue to stay at this party for 2 or more
hours and having a pain intensity level of ‘8’ when he is leaving. He will be angry and depressed
about the weakness of his body, and when pain has the level of ‘8’, he tends to thoughts of cata-
strophizing: ‘What will happen when this pain does not decrease today?’ A third man showing a
pattern of EER will also visit this party and will reach complete cognitive distraction by animated
talking with his friends. At the end of the party he suddenly will sense pain of high intensity (for
instance of level ‘8’) and will wonder why he was not aware of an increase of pain during the time
before. The AEM assumes that a man showing a pattern of adaptive pain responses will know that
a party where all guests will stay a long time will lead to an unnecessary increase of back pain. He
decides not to decline this party visit knowing that meeting his friends will increase positive
mood. He decides to phone to his friend asking him to prepare some chairs giving him the chance
of sitting for a while. During this party, he decides to make short breaks latest at an increase of
pain from ‘3’ to ‘5’.
What ingredients belong to this kind of adaptive pain response? First, the awareness of a slight
pain increase at low levels of pain intensity. Second, the acceptance that standing for a long
time represents a kind of biomechanical load leading to a delay of the recovery of back pain.
Third, the acceptance of individual needs making a short break changing the bodily position.
Fourth, the behavioural competence that enables him, asking his friend providing a chair.
Patients with elevated FAR or ER pattern report a great number of external obstacles (for
instance, high social pressure, specific circumstances), internal cognitive/affective barriers (e.g.
dysfunctional thoughts ‘what will my friends think of me?’) or behavioural deficits regarding
social competence that makes it difficult to realize the above mentioned adaptive responding to
pain. To our best knowledge, there is no empirical evidence concerning a systematic classification
of external and internal obstacles until now. In Chapter 27 of this volume, we will describe differ-
ent obstacles based on extensive clinical work as well as specific cognitive-behavioural treatment
strategies aiming at overcoming individual barriers and promoting adaptive pain responses.
15.4 Evidence for endurance-related components

of AEM
The degree to which single affective, cognitive and behavioural responses, described in the AEM as
well as the more complex response pattern, will be adaptive is a matter of some debate. Whether
single responses or a full pattern will be effective and adaptive or not depends on the outcome
criterion used and on the different time-lags used to measure pain responses and the outcomes.
For example, active avoidance behaviour may lead, in the short term, to a reduction or prevention
of pain. However, in the long term, it may lead to physical disuse and an increase of disability. On
the other hand, task persistence behaviour may lead to a feeling of control and short-term distress
reduction, but in the long term, it can result in physical overuse and overload including pain
increase, a consequence of repetitive micro-damaging of physical structures. Task persistence
behaviour that is part of an exclusive DER is associated with cognitions of thought suppression.
Due to the lack of success of these cognitions within this pattern, patients may even experience a
rapid change between feelings of control and feelings of loss of control. With these considerations
in mind, it is obvious that an evaluation of the effect of pain-related responses based on associa-
tions found in cross-sectional study designs will only deliver vague information that must inter-
preted cautiously. The evaluation of both short-term as well as long-term consequences of different
response patterns need to be performed in prospective study designs. To identify short-term con-
sequences, experimental trials and process-oriented correlation approaches have to be included.
In addition, for identifying long-term consequences prospective cohort studies are necessary.
15.4.1 Measurement of endurance-related pain responses

While measurements of pain-related fear and fear-related beliefs are presented in Chapter 14 of
this volume, the present chapter will mainly focus on measurements of pain-related endurance
with its behavioural feature of task persistence, with the cognitive aspects of thought suppression
and distraction and with the affective feature of positive mood despite pain.
Measurement of behavioural responses

Task persistence behaviour is captured by the CPCI (Jensen et al. 1995), the Kiel Pain Inventory
(KPI, Hasenbring 1994), and the KPI-derived Avoidance-Endurance Questionnaire (AEQ,
Hasenbring et al. 2009). The scale Task Persistence is a subscale of the CPCI-42 (Jensen et al.
1995; Romano et al. 2003) consists of 42 items assessing coping strategies that patients might use
to cope with chronic pain, for instance Guarding, Resting, Asking for Assistance, Relaxation, Task
Persistence, Exercise/Stretch, Seeking Social Support, and Coping Self-Statements. Patients are
asked to rate the number of days (0–7 days) over the past week that they used each of the strategies
at least once. The subscale Task Persistence (TP) consists of 6 items (e.g. ‘I kept on doing what
I was doing’ or ‘I didn’t let the pain interfere with my activities’) and showed sufficient internal
consistence (Cronbach’s alpha between 0.85 and 0.74) but low re-test reliability (r=0.66). Results
concerning validity seem to be inconclusive until now: low to moderate negative cross-sectional
correlations were reported for task persistence and disability, pain intensity and depression in
samples including back pain or mixed pain diagnoses (Jensen et al. 1995; Tan et al. 2001; Romano
et al. 2003). No correlation was found for self-reported activity. Truchon and Coté (2005), how-
ever, did not find cross-sectional or prospective associations between CPCI-TP and the outcomes
pain, disability or return to work, and Garcia-Campayo et al. (2007) reported positive correla-
tions between TP and other so-called illness-focused subscales (e.g. Guarding and Resting) and
also positive correlations with the outcome disability, fatigue and depression in a sample of
402 fibromyalgia patients.
The Behavioural Endurance Scale (BES) of the KPI consists of 11 items that focus on coping
efforts to finish all activities just started in spite of severe pain (e.g. ‘I keep all appointments even
though I don’t feel well’). Patients indicate on a 7-point scale (0 ‘never’, 6 ‘always’) how often they
have acted in such a way in the past 14 days when they experienced pain. Higher scores indicate a
higher number of self-reported endurance cognitions and behaviours. Cronbach’s alpha for BES
was r=0.81 (Hasenbring, 1994) in a sample of acute and chronic pain patients with different diag-
noses. In a sample of 111 sciatic pain patients, no correlation with pain intensity was found cross-
sectional, but positive associations pain intensity prospectively (r=0.25, p <0.01). Patients high
on BES further showed a lower return-to-work rate at the six months follow-up (p <0.05). The
original BES subscale from the KPI was included to the KPI-derived AEQ (Hasenbring et al. 2009)
that was developed in order to measure typical fear-avoidance- as well as endurance-related fea-
tures of pain responses. In a sample of 191 LBP patients, factorial structure indicated that the two
subscales Pain Persistence (PPS, 7 items) and Distraction/Humour (HDS, 5 items) revealed suf-
ficient internal consistency (0.78 and 0.76). The total BES score can be using additionally. Within
this sample, there were positive but low correlations between BES, PPS and HDS and pain inten-
sity and negative associations with self-reported disability, depression, fear of pain and days of
sick listing (Hasenbring et al. 2009). BES is one of the three marker variables used for building the
AEM-based subgroups (see ‘AEM-based subgroups’).
Measurement of cognitive responses

The Thought Suppression Scale (TSS) is a subscale of the KPI as well as of the KPI-derived AEQ.
The TSS consists of 4 items (‘Pull yourself together!’), yielding an internal consistency of 0.78
(Cronbach’s alpha) in the original sample of 405 pain patients with different diagnoses (Hasenbring
1994) and 0.80 in LBP patients (Hasenbring et al. 2009). TSS scores were unrelated to pain inten-
sity in a cross-sectional study of acute sciatic pain patients, but positively related to pain as well to
depression prospectively at a 6-month follow-up (Hasenbring et al. 1994, Klasen et al. 2006). TSS
was more elevated in a dysfunctional subgroup of CLBP patients compared to adaptive copers
(Rusu et al. 2008) using the Multidimensional Pain Inventory (MPI; Kerns et al. (1985) -derived
subgroup analysis (Turk and Rudy 1988)). TSS is further one of the three marker variables used
for building the AEM-based subgroups (see section ‘AEM-based subgroups’).
Cognitive distraction is captured at least by three measurements, the Humour/Distraction
subscale of the AEQ (Hasenbring et al. 2009), the Coping Strategies Questionnaire (CSQ;
Rosenstiel and Keefe, 1983) and the Pain-Coping Inventory (PCI; Kraimaat et al. 2003). The
Diverting Attention subscale of the CSQ consists of 6 items and reveals sufficient internal consist-
ency (e.g. Cronbach’s alpha 0.80 in Hill et al. 1995). Based on exploratory or confirmatory com-
ponents analyses, the Diverting Attention scale is one of 3–4 features summarized as coping
attempts (Hill et al. 1995) or part of so-called active coping (Riley and Robinson, 1997). The same
was seeing for the Distraction subscale of the PCI that consists of five items. Internal consistency,
however, was rather low in different samples (Alpha <0.70; Kraimaat et al. 2003). It is important
to note, that the distraction scale of the PCI contains items that serves to relaxation (‘I distract
myself by reading, listening to music’) on the one hand, and to kinds of endurance behaviour on
the other (‘I distract myself by undertaking a physical activity’). The Humour/Distraction
subscale of the AEQ consists of five items (‘I take it with a laugh’, ‘I distract doing little jobs at
home’), yielding an internal consistency of Alpha=0.78. Preliminary validation indicated a posi-
tive relationship with pain intensity but a negative one with self-reported disability, depression
and fear of pain (Hasenbring et al. 2009).
Measurement of endurance-related affective responses
The Positive Mood Scale (PMS) is a subscale of the KPI as well as of the KPI-derived AEQ. The
KPI-PMS consists of 4 items (‘I feel happy, anyway’) yielding an internal consistency of .85
(Cronbach’s alpha) in the original sample of 513 pain patients with different diagnoses (Hasenbring
1994) and 0.90 for the PMS scale of the AEQ (consisting of 3 items) in a LBP study (Hasenbring
et al. 2009). The AEQ-PMS scores were cross-sectional unrelated to pain intensity but negatively
correlated with disability, depression, and fear of pain measures (Hasenbring et al. 2009). As
described above, the KPI-PMS scores were positively related to pain intensity prospectively, con-
tributing to the prediction of pain condition at the time of discharge from a hospital stay due to
sciatic pain (Hasenbring et al. 1994).
15.4.2 AEM-based subgroups

Research in FAR and ER pattern delivers preliminary support for the fact that subgroups of pain
patients are to be identified by specific AEM-based pattern on the affective, cognitive, and behav-
ioural level. Hasenbring (1993) compared for the first time patients with a DER or EER pattern
with patients showing an adaptive response (AR) pattern. In this study, using the Beck Depression
Inventory (BDI; Beck et al. 1961) and the KPI (Hasenbring, 1994), a cut-off for depression was
defined as BDI >9 and the KPI subscales thought suppression and endurance behaviour (mean
scale scores) were used in a sample of subacute sciatic pain patients. DER patients differed from
AR with respect to higher thought suppression and depression (marker variables), higher pain-
related helplessness, hopelessness, higher nonverbal pain behaviour and a higher degree of chronic
distress in daily life. EER patients differed from AR showing significantly higher positive mood
despite pain, less catastrophizing, less avoidance of physical as well as of social activities and less
nonverbal pain behaviour, besides the higher scores in the marker variable endurance behaviour.
Data concerning the four AEM-based subgroups had been also reported by Grebner and
co-workers (1999). The authors investigated 82 patients prior to a lumbar disc surgery, using a
cluster analysis procedure. DER patients revealed more thought suppression as well as more pain-
related anxiety and less positive mood compared to EER patients, more thought suppression and
pain-related anxiety than FAR and less positive mood compared to AR patients. FAR patients
differed from AR showing less positive mood. With respect to avoidance behaviour, there was
only a significant difference between EER patients and all other subgroups with lower avoidance
in EER. Further data on content validity of AEM-based subgroups have been reporting in a recent
study in patients suffering from subacute non-specific low back pain (Hasenbring et al. under
review). Besides the marker variables depression, thought suppression and behavioural endur-
ance, DER patients reveal significant higher scores in pain-related anxiety and helplessness/
hopelessness as well as in work-related fear-avoidance beliefs. EER patients differ from AR in less
avoidance and less work-related fear-avoidance beliefs. FAR patients differed from AR with
respect to less behavioural endurance, and with higher avoidance due to mild pain from both
ER subgroups.
Thus far, there are preliminary data supporting the content validity of the four subgroups
suggested by the AEM, two times using clinical cut-off scores and one study using cluster analysis.
Further results on criterion related and prospective validity will be reported in the subsections
that follow.
15.4.3 Endurance and pain-related outcomes

Endurance and pain intensity
With regard to single endurance responses, task persistence behaviour seems to be the most often
investigated feature in clinical research. There is first evidence from several prospective studies
concerning short-term as well as long-term effects on the maintenance of pain, mediated by
endurance behaviour. In a prospective study of acute and subacute sciatica patients, who under-
went an inpatient medical treatment, Hasenbring et al. (1994) reported endurance behaviour as a
unique, significant predictor of higher pain intensity at the time of discharge from the hospital as
well as at the 6-month follow-up besides other psychological and biomedical predictor variables.
In a recent prospective study of subacute non-specific back pain patients in an outpatient setting,
endurance behaviour was the main predictor of pain intensity at the 6-month follow-up, after
control for baseline pain variables and depression (Hasenbring et al. under review). Additionally,
Turner and Clancy (1986) found that diverting attention combined with increasing behavioural
activity was predictive of higher pain intensity after a cognitive-behavioural treatment in chronic
back pain patients, using a former version of the CSQ. These first data concerning ER pain behav-
iour provide preliminary support for the AEM-based hypothesis of a link between endurance-
related behaviour and the development of higher levels of future pain intensity.
Cross-sectional there is more inconsistent evidence for the relation between task persistence
behaviour and pain intensity variables. Some studies reported absent or low positive correlations
with pain (Rosenstiel and Keefe 1983; Hadjistavropoulos et al. 1999; Tan et al. 2001; Truchon and
Coté 2005; Hasenbring et al. 2009), whereas others reported moderate positive associations with
pain intensity (Osborne et al. 2007; Hasenbring et al. under review).
Thought suppression as a cognitive aspect of ER pain responses had been exploring within
laboratory experimental studies comparing the effect of different cognitive coping strategies on
pain intensity in healthy individuals as well as in clinical studies. Cioffi and Holloway (1993)
investigated the effect of thought suppression in comparison to distraction and sensory monitor-
ing. While participants in the distraction condition were told to form a vivid mental picture
of their rooms at home, participants in the suppression condition were only told to not think
about their hand sensations, to eliminate awareness of them. Furthermore, the authors studied
short- and mid-term outcomes in the laboratory by assessing pain intensity every 20s after pain
stimulation over a period of 2min and by conducting a further experimental task. Thirty minutes
after pain stimulation a mild vibration stimulus was applied at the participant’s neck which was
experienced as relatively neutral as assessed by a pre-test sample. Results indicated that partici-
pants in the suppression condition initially showed less intense pain but later during the 2-min
interval showed more intense pain compared to the distractors or monitorers. They also rated the
vibration stimulus as less pleasant compared to the distraction and sensory monitoring group.
This differential short- and mid-term effect of suppression on pain intensity mirrors potentially
different operant mechanisms in the maintenance of this cognitive pain response, as suggested in
the AEM above (see Figure 15.2).
As noted previously, Cioffi and Holloway (1993) assumed that especially the instruction to
suppression produces a rebound effect, which is well investigated in research on cognition and
mental control (e.g. Wegner et al. 1987). This research has shown that when people were asked to
suppress awareness of a thought they were, in fact, able to do it but this suppression produced a
rebound effect—with the consequence that this thought occurred subsequently more often than
if it has not been initially suppressed. Wegner and colleagues attributed this phenomenon to the
different goal structures of distraction, sensory monitoring, and suppression. Whereas the goal of
distraction is to replace one thought with another, therefore distraction specifies what to do, the
goal of suppression is to remove a thought from the mind or it specifies what not to do. If people
try to obtain goal-relevant feedback about the success of their mental strategy, in the case of sup-
pression they ask themselves repeatedly how empty their mind will be of the thought X. As a
consequence, in the case of pain stimulation, they focus to this pain stimulus again and again. In
contrast, distraction and also sensory monitoring instructions give a concrete and vivid mental
event, success is evaluated by an awareness of sensations irrelevant of unpleasant features of the
pain stimulus. Furthermore, Wegner and colleagues have found that the instruction to suppress
often induces a search for distractors that was disorganized, non-focused and which has often
failed (Wegner et al. 1991). This search for distractors is more effortful than having a specific
distractor given by the experimenter. Support for the assumption of a mid-term pain-increasing
effect of thought suppression came also from an experimental study conducted by Masedo
and Esteve (2007), comparing suppression, spontaneous coping and instructed acceptance of a
cold-pressor stimulus.
In clinical pain patients, there is preliminary evidence for a pain-increasing or pain-maintaining
effect of thought suppression as a single response. In a cross-sectional study including 191 suba-
cute LBP and chronic LBP (CLBP) patients, a weak but positive correlation was found between
thought suppression and pain intensity (Hasenbring et al. 2009). In another recent study, Rusu
and Hasenbring (2008) found TSS scores significantly more elevated in a dysfunctional sub-
group of CLBP patients compared to adaptive copers, using the MPI clustering methodology
(Turk and Rudy, 1988).
In patient subgroups showing a DER pain response pattern, thought suppression goes along
with enhanced depressive mood as well as with high pain-related task persistence behaviour. In a
sample of acute/subacute sciatic pain patients prior to an inpatient treatment, DER patients
revealed a significant decrease in pain intensity until the discharge from the hospital but they had
shown a new increase at the 1-week and at the 6-month follow-up. At this last follow-up, 53% of
this subgroup had not been return to work compared to 6% in the AR subgroup and 18% in the
EER patients. Despite high positive mood and no depression, the EER patients also had a new
increase in pain intensity at both follow-ups. Furthermore, both ER subgroups revealed signifi-
cantly more often an application for early retirement due to pain (25% in EER, 31% in DER)
compared to AR patients with 3% (Hasenbring 1993). These prospective data regarding pain
intensity and pain-related work absenteeism was supported by the Grebner et al. subgroups study
(Grebner et al. 1999).
Endurance and pain-related disability

Past research has shown that pain intensity and disability are only marginally associated, while
pain-related disability is highly linked with cognitive and affective aspects of pain-related fear,
such as catastrophizing or fear-avoidance beliefs (see Chapter 14). However, the relationship
between self-reported disability and pain-related endurance, such as thought suppression and
task persistence behaviour, remains open. The AEM suggests that a person, who reports high
pain-related endurance, will feel less disabled at that time. This hypothesis is to be investigating
within cross-sectional designs. Reviewing the body of research using cross-sectional designs, this
hypothesis is supported for task persistence behaviour but not for thought suppression. A major-
ity of studies reported significant negative associations between pain-related task persistence
behaviour as a single response and disability (Rosenstiel and Keefe 1983; Hadjistavropoulos et al.
1999; Tan et al. 2001; Truchon and Coté 2005; Hasenbring et al. 2009). First data reveal a signifi-
cant negative association between positive mood as an endurance-related emotional response to
pain and disability (Hasenbring et al. 2009). In contrast, only a few data published data concern-
ing thought suppression as a single response and disability. In a cross-sectional study in non-
specific back pain patients, Hasenbring et al. (2009) reported zero-correlations between thought
suppression, assessed with the AEQ, and pain-related disability, measured with the Pain Disability
Index (PDI; Tait et al. 1990).
As the AEM assumes that thought suppression effectively reduces pain only intermittently (see
Figure 15.2), it may be linking with task persistence as well as with avoidance behaviour, specifi-
cally in the long-term run (Hasenbring and Verbunt, 2010). Future studies investigating bivariate
the relation between thought suppression and disability should control for both behavioural
features, for instance, using partial regression analysis.
15.4.4 Endurance and overt physical activity

A few studies that assess physical activity not only by self-report measures but also by overt behav-
ioural methods found no difference in actual physical activity between CLBP patients and healthy
controls, neither in adults (Verbunt et al. 2001) nor in children (Wedderkopp et al. 2003). They
found also no correlation between self-ratings of pain intensity and the actual physical activity
level (PAL) in CLBP patients (Sanders, 1983; Linton 1985; Vendrig and Lousberg 1997; Verbunt
et al. 2001; Wedderkopp et al. 2003; Wittink et al. 2003). Wittink et al. (2003) had been discussing
the possibility of a subgroup of CLBP patients that will under-report or underestimate their
actual ability to function physically.
Subgroups of LBP patients showing self-reported disability despite a high level of overt physical
activity have been published by Bousema and co-workers (2007). In this study, patients with
subacute back pain were investigated over 1 year. The authors described a subgroup that devel-
oped an increase of back pain despite a significant increase of accelerometer-assessed physical
activity, which was an unexpected finding. Interestingly, these patients with an increase in PAL,
however, suffered of more disability.
As the AEM postulates that patients with high ER pain responses will tend to withstand pain-
related interruptions of daily activities in order to pursuit current work-related goals, it is there-
fore suggested that ER patients will show signs of physical overuse and overload in their daily life.
A first accelerometer study delivered preliminary evidence for this assumption (Hasenbring et al.
2006). A combined subgroup of ER patients revealed a trend to higher PAL compared to adaptive
patients and a significantly higher number of constant strain positions (CP, sitting or standing
upright or bending forward) during an 8-hour assessment period of a working day. It was a strik-
ing finding, that this ER subgroup revealed a higher overt activity level despite higher pain inten-
sity, and, against expectations also despite higher self-reported disability. Due to small sample
sizes, DER and EER patients could not be distinguished. It remains therefore open whether the
high disability score comes more from DER as to be expected and not from EER patients.
Furthermore, a high level of fatigue has been seen in the ER subgroup. Fatigue had been discussed
as a response to repeating, often unsuccessful trials to distract from pain or to suppress pain expe-
riences from consciousness (Eccleston and Crombez 1999). It can be hypothesized that not only
unsuccessful mental trials to suppress pain but also the behavioural components of persisting
daily activities in spite of pain may lead to increased fatigue, mentally and physically.
The prominent role of static physical strain positions in the maintenance of back pain was also
captured by a recent cross-sectional study in teachers in schools for people with severe handicaps.
Using accelerometer data, Wong et al. (2009) have shown that teachers with LBP showed signifi-
cantly more static trunk postures during working days than pain-free colleagues.
Static strain positions have been shown to cause high loads to the spine inducing creep in the
various viscoelastic tissues, such as ligament, discs and fascia (Solomonow et al. 2003a, 2003b).
Sbriccoli et al (2004) have shown that the number of repetitions of a static load on the lumbar
spine is a risk factor in the development of cumulative trauma disorder. The creep developed
becomes more intense as the number of repetitions increases. Furthermore, the authors argue, if
the creep does not fully recover, for instance overnight, people start a new workday with a resid-
ual creep from the previous day. As time goes on, the residual creep may accumulate from day to
day, inducing micro-damages in the lumbar spinal tissues associated with acute inflammation as
an attempt of healing of the damage. Further consequences of repetitive static load positions are
the reduction of oxygenation of lumbar extensors due to the constant isometric contraction
(McGill et al. 2000) as well as an increase of intradiscal pressure (Stokes and Iatridis 2004).
15.4.5 Endurance and neurocognitive functions

Within a recent neurocognitive model of pain (Legrain et al. 2009), attention to pain or to pain-
irrelevant behavioural goals is conceptualized through involuntary (bottom-up) and intentional
(top-down) processes, given the limited processing capacity of the brain. Among top-down proc-
esses, attention may be directed away from nociceptive stimuli in order to fulfil a concurrent
cognitive task, leading to a specific effort due to the primary goal (attentional load hypothesis) or
attention may be facilitated towards stimuli that are associated with defined goal-relevant stimuli,
for instance, towards pain stimuli, when these are experienced as highly threatening (attentional
set hypothesis).
The attentional load hypothesis was investigated in a number of laboratory experiments in
healthy individuals using nociceptive stimulation while instructed to attend to a specific cognitive
task. This kind of distraction seems able to reduce the intensity of experienced pain (Bingel et al.
2007) as well as to reduce brain activity, specifically in areas that are usually involved in the
processing of pain perception and attentional control (Seminowicz and Davis 2006; Bingel et al.
2007; Seminowicz and Davis 2007). Within the context of the AEM model, the question arises
whether people, who usually respond to pain with cognitive distraction, thought suppression,
and task persistence behaviour (ER pattern), will show higher cognitive performance than
individuals low in ERs. In a recent laboratory study, Held et al. (in prep.) instructed healthy
volunteers to attend to a mental arithmetic task while pain was induced by a cold pressor stimula-
tion. Using median split procedure, the participants high in task persistence behaviour, measured
with the AEQ (Hasenbring et al. 2009) modified for healthy individuals (AEQ-H), revealed
shorter response times in the mental task as well as a lower number of cognitive errors.
Furthermore, participants with high endurance revealed less pain intensity scores and signifi-
cantly higher pain tolerance time than low endurance participants. These preliminary findings
indicate that pain persistence behaviour enables a person to direct attention towards pain-
irrelevant goals, which may in the short term be followed by a reduction of pain (negative rein-
forcement) and a better cognitive performance representing a kind of positive reinforcement.
15.4.6 Endurance and feelings of self-control and self-efficacy

The ability to respond with endurance behaviour in unpleasant or difficult situations is undoubt-
edly an important ability and may lead to an increase of feelings of self-control and high self-
efficacy. In pain research, self-efficacy represents a set of cognitions predicting positive outcomes
of medical as well as of psychological treatments (Nicholas 2007). In models of stress and coping
(Zautra et al. 1995), pain is conceptualized as a stressor to which individuals show diverse adapta-
tion patterns. Whereas cognitive responses to pain such as help-hopelessness are important pre-
cursors of feelings of loss of control and depression, several coping responses were positively
associated with self-reported control. Based on a cross-sectional correlation study in 195 chronic
pain patients, positive relations between the CSQ-factors ignoring pain sensations, increasing
behavioural activity, diverting attention as well as for coping self-statements and reinterpreting
pain sensations and feelings of control were reported (Haythornethwaite et al. 1998). Individual
differences in pain coping, predicted the effects of experimentally induced controllability during
heat pain stimulation in healthy individuals and indicate the critical role of prefrontal cortex areas
(Salomons et al. 2004). More recently, two studies investigated the association between pain-
related task persistence behaviour and basal adrenocortical activity represented by a neuroendo-
crinological stress parameter, such as level of salivary cortisol after awakening (Sudhaus et al.
2007, Sudhaus et al. 2009), both supporting the stress-reducing character of behavioural endur-
ance. These data concerning the relation between endurance behaviour, pain-related stress and
control provide preliminary support for the assumption of the AEM that behavioural endurance
associated with effective distraction away from pain may have a stress-reducing short-term effect,
potentially mediated by an increased sense of control and self-efficacy.
15.4.7 Endurance and affective responses

Positive mood despite pain
One of the most surprising and counterintuitive results in psychological pain research refer to the
finding that patients with LBP, irrespective of the duration of pain, report feelings of positive
mood despite pain, besides anxiety, depression, or irritated mood (Hasenbring et al. 1994;
Hasenbring et al. 2009). Positive mood despite pain and endurance behaviour, both assessed with
the KPI or with the AEQ were significantly positively correlated in a sample of 305 pain patients
of different diagnoses (r=0.41; Hasenbring 1994) as well as in a recent studied sample of 191 LBP
patients (r=0.46; Hasenbring et al. 2009). In contrast, PMS was unrelated to thought suppression
in these both validation studies. In a former prospective study in acute or subacute sciatic pain
patients, both, endurance behaviour and positive mood despite pain were positively correlated
with pain intensity prospectively, to a lesser degree but in the same direction as for depression and
avoidance behaviour (Hasenbring et al. 1994). Positive mood despite pain and endurance behav-
iour are the two marker variables representing the EER pattern, seen in a subgroup of patients
who also tend to minimize pain experiences and to cognitive distraction. It remains to be deter-
mined whether attentional distraction in EER patients is more effective in reducing pain experi-
ences and in enhancing the ability to continue with current daily activities and whether this is the
main origin of positive mood. Habitual as well as situational characteristics are also potential
precursors of the ability to maintain positive mood despite pain.
Endurance and pain-related fear and depression

At a first glance, it seems plausible to suggest pain-related helplessness, hopelessness and beliefs
that daily work will lead to an increase of pain as parts of a FAR pattern. It has been shown, how-
ever, that the DER subgroup of LBP patients, revealed even higher scores in pain anxiety, helpless-
ness, hopelessness, and the belief that work will cause pain as well as in the belief of a poor work
prognosis, compared to AR patients (Grebner et al. 1999; Hasenbring et al. under review). In a
recent study in 146 patients with exercise induced leg pain (overuse group) and 154 CLBP patients,
the overuse group revealed as high fear of re-injury pain (measured with the TSK) as the CLBP
group (Lundberg and Styf 2009). These findings are in line with the assumption of the AEM, that
pain-related fear and dysfunctional beliefs may not only be associated with avoidance behaviour
but also with the opposite tendency of task persistence behaviour. Within this context, the ques-
tion arises, whether the beliefs in DER patients that work will cause or increase pain, are really
dysfunctional or alternatively whether these beliefs are based on real experiences and represent
consequences of individual coping by dealing with daily work requirements in these patients.
A positive association between high fear and persistence behaviour has been also reported in
other fields of research, for instance, in military psychology (Rachman 2004). Rachman describes
soldiers trained to realize persistence behaviour in dangerous situations despite subjective feelings
of fear or anxiety, labelling this pattern ‘courageous behaviour’. Comprehensive training pro-
grammes, for instance in parachutists (Walk, 1956), led to a reduction of fear after a few jumps,
representing learning principles of negative reinforcement. Within this context, it is important to
note, that behavioural training programmes in patients suffering from anxiety of pain-related
fear, using the principle of confrontation, also increase the ability of a patient to reveal coura-
geous behaviour, namely persistence despite fear. While this procedure seems to be successful
in FAR patients (see also Chapter 14), it can be questioned whether it would be also effective
and indicated in patients showing a DER pattern, where we assume that high task persistence
behaviour will lead to overuse and physical injury in the long term.
Increased depressive mood is an often seen and theoretically reasoned consequence of pain-
related FAR pattern (see Chapter 14). Moreover, the AEM suggests a link between high thought
suppression and increased depressive mood, due to the often experienced failure of attentional
distraction and the rebound phenomenon reviewed previously. Preliminary, but conclusive
evidence came from a multivariate study in chronic LBP patients, investigating the role of pain
intensity, catastrophizing, helplessness, hopelessness, as well as thought suppression in the
prediction of depression (Klasen et al. 2006). Using path analysis, the authors found a direct rela-
tionship between thought suppression and depressive mood, whereas catastrophizing revealed
only an indirect association via cognitions of helplessness and hopelessness. In this study, pain
intensity by itself had no direct association to depression.
Stability versus variability of pain-related endurance responses

15.4.8
and AEM pattern
Knowledge of stability versus variability over time of response patterns is rare. In addition, no
information is currently available indicating which situational demands, specific motivational
factors and/or aspects of learning history will lead to an individual FAR or ER pattern. The AEM
suggests that exclusive FAR or ER pattern are relatively stable in time unless modification is based
on well-directed behavioural interventions. On the other hand, changes in response pattern may
naturally occur during the long-term process of pain chronification. For example, the experience
of long-term pain, which does not respond adequately to medical therapy, may lead to a change
from an exclusive EER into a DER or even FAR pattern. Longitudinally, ER patients might be at
risk for a decrease of task persistence and an increase of avoidance behaviour. In this situation,
avoidance is assumed as a kind of passive avoidance behaviour, which may have several deterio-
rated consequences: First, patients postpone the interruption of painful daily activities for a long
time. Consequently, when they eventually do interrupt (e.g. a short pause for relaxation) their
activities, this will not lead to a regeneration of damaged physical structures and pain will not
diminish. Therefore, patients will experience such an interruption as unsuccessful. Second, short
pauses that will not result in a pain decrease, will lead to an increase of thought suppression with
the consequences of rebound, feelings of failure, low self-esteem and depression. Third, when
ER patients will only decide to interrupt their daily activities due to very severe pain, eventually,
pain will guide these interruptions consciously or unconsciously. Therefore, although task
persistence behaviour may first lead to feelings of control, in the long-term it can result in feelings
of loss of control.
15.5 Consequences for clinical practice: implication for

differentially targeted approaches
Preliminary clinical consequences for patients with both acute and chronic pain are considered in
the following. Diverse national LBP guidelines advise, in a situation of acute non-specific LBP
(excluding severe structural causes of pain (red flags)), to quickly return to a normal level of daily
activity (Arnau et al. 2006). On a conceptual level, the avoidance and endurance-based research
reported previously in this chapter suggest the need for a definition of the term ‘normal’ activity,
as it is assumed that FAR patients would subjectively define their normal activity level differently
compared to ER patients. Adding to the complexity, patients tend to evaluate their activity-
related problem due to pain by comparing their current activity level in reference to their habitual
activity level before pain-onset. Their perceived decline of activities since pain-onset appeared to
be more disabling as their actual activity level (Verbunt et al 2001; Bousema et al. 2007).
Based on these considerations, the AEM suggests differential interventions for patient educa-
tion, physical therapy, and cognitive-behavioural treatment approaches. Based on a preceding
screening of AEM-derived pain response pattern, multimodal treatment strategies should be
individually targeted dependant on FAR, DER, or EER pattern. For patients with FAR and related
physical disuse, the approach of increasing one’s activity level, based on graded activity and expo-
sure in vivo (a behavioural approach in which fear of movement is the focus of treatment) seemed
both effective, specifically in reducing pain-related disability (Vlaeyen at al. 2002; Leeuw et al.
2008). In contrast, patients showing an endurance-related response pattern should be encouraged
to accept pain experience as a signal that has to be taken seriously in order to prepare short breaks
in their daily activities and to reduce physical postures that cause high load on muscles and the
spine. The general aim of an AEM-based CBT approach is to enable both, FAR and ER patients to
a more flexible response pattern that is comparable to the AR patients.
A first randomized controlled intervention trial in patients suffering from subacute sciatic pain
revealed preliminary evidence that an AEM-scheduled, risk-factor based CBT, offered besides
conservative medical interventions, is significantly more effective in preventing future pain com-
pared to a standardized psychological programme or to a control group receiving only medical
treatment as usual (Hasenbring et al. 1999). Results from an additional randomized trial in suba-
cute non-specific back pain patients provided further evidence concerning the efficacy of an
individually scheduled, AEM-based short physician counselling with a 3-month follow-up
(Hasenbring et al. under review, for more information see also Chapter 27 in this volume).
15.6 Summary and conclusion

Firm evidence identified FAR as an important mediator for the development and maintenance of
chronic pain and disability. In addition, we found increasing evidence for the presence of a
second and opposite pathway for chronicity in pain based on endurance-related responses. The
impact of these opposite pathways on daily activities and their implication for clinical practice
still need further research, but seem promising in the determination of more individualized treat-
ment programmes for patients with chronic pain syndromes.
In conclusion, the AEM provides testable hypotheses concerning the development of pain-
related disability and affective distress, such as increased depressed mood. Most of these hypoth-
eses will require prospective study designs including diary studies exploring short-term
relationships as well as longitudinal designs for the investigation of long-term processes, such as
the recurrence of pain months or years following an acute episode. More research is required to
understand the underlying mechanisms of endurance-related responses and several potential
avenues for future research have been identified including, for instance, activity monitoring,
neuroendocrinological and neurocognitive functions. Future research should also focus on the
motivational mechanisms, situational demands, learning history and the potential impact of
personality features on the development of rigid fear-avoidance and endurance pattern.
References
Ahles, T., Blanchard, E. and Leventhal, H. (1983). Cognitive control of pain: attention to the sensory
aspects of the cold pressor stimulus. Cognitive Therapy and Research, 7, 159–77.
Arnau, J.M., Vallano A, Lopez A, et al. (2006). A critical review of guidelines for low back pain treatment.
European Spine Journal, 15, 543–53.
Asmundson, G.J.G., Norton and P.J., Norton, G.R. (1999). Beyond pain: the role of fear and avoidance in
chronicity. Clinical Psychology Review, 19, 97–119.
Beck, A.T., Ward, C.H., Mendelson, M., Mock, J., Erbaugh, J. (1961). An inventory for measuring
Bingel, U., Rose, M., Gläscher, J. and Büchel, C. (2007). fMRI reveals how pain modulates visual object
processing in the ventral visual stream. Neuron, 55, 157–67.
Bortz, W.M. (1984). The disuse syndrome. Western Journal of Medicine, 141, 691–94.
Bousema, E.J., Verbunt, J.A., Seelen, H.A.M., Vlaeyen, J.W.S., Knottnerus, J.A. (2007). Disuse and physical
deconditioning in the first year after the onset of back pain. Pain, 130, 279–86.
Cioffy, D. and Holloway, J. (1993). Delayed costs of suppressed pain, Journal of Personality and Clinical
Commission on Behavioral and Social Sciences and Education, National Research Council and Institute of
Medicine. (2001). Musculosceletal disorders and the workplace: low back and upper extremities.
Washington, DC: National Academy Press.
Davis, M.C., Zautra, A.J. and Smith, B.W. (2004). Chronic pain, stress, and the dynamics of affective
differentiation. Journal of Personality, 72, 1133–59.
Eccleston, C. and Crombez, G. (1999). Pain demands attention: a cognitive-affective model of the
Fishbach, A. and Ferguson, M.F. (2007). The goal construct in social psychology. In: A.W. Kruglanski and
E.T. Higgins (Eds.) Social Psychology: Handbook of basic principles, pp. 490–515. New York: Guilford.
Garcia-Campayo, J.G., Pascual, A., Alda, M. and Ramirez, M.T.G. (2007). Coping with fibromyalgia:
usefulness of the chronic pain coping inventory-42. Pain, 132, 68–76.
Gatchel, R.J., Peng, Y.B., Peters, M.L., Fuchs, P.N., Turk, D.D. (2007). The biopsychosocial approach to
chronic pain: schientific advances and future directions. Psychological Bulletin, 133, 581–624.
Goschke, T. and Dreisbach, G. (2008). Conflict-triggered goal-shielding: response conflicts attenuate
background monitoring for prospective memory cues. Psychological Science, 19, 25–32.
Grebner, M., Breme, K., Rothoerl, R., Woertgen, C., Hartmann, A. and Thomé, C. (1999). Coping and
convalescence course after lumbar disk operations. Der Schmerz, 13, 19–30.
Hadjistavropoulos HD, Frombach IK, Asmundson GJG. (1999). Exploratory and confirmatory factor
analytic investigations of the Illness Attitudes Scale in a nonclinical sample. Behaviour Research and
Therapy, 37, 671–84.
chronic pain? Der Schmerz, 7, 304–13.
Hasenbring, M. (1994). Das Kieler Schmerzinventar. [The Kiel Pain Inventory-Manual. Three questionnaire
scales for the assessment of pain-related cognitions, emotions and coping strategies]. Bern: Huber.
Hasenbring, M. (2000). Attentional control of pain and the process of chronification. In: J. Sandkühler, B.
Bromm and G.F. Gebhart (Eds.) Progress in Pain Research Vol 129, pp. 525–34. New York: Elsevier
Science.
Hasenbring, M., Marienfeld, G., Kuhlendahl, D. and Soyka, D. (1994). Risk factors of chronicity in lumbar
Hasenbring, M., Ulrich, H.W., Hartmann, M. and Soyka, D. (1999). The efficacy of a risk factor based
Hasenbring, M.I., Plaas, H., Fischbein, B. and Willburger, R. (2006). The relationship between activity and
Hasenbring, M.I., Hallner, D. and Rusu, A.C. (2009). Fear-avoidance- and endurance-related responses to
Hasenbring, M.I., Verbunt, J.A. (2010). Fear-Avoidance and endurance-related responses to pain: new
models of behavior and their consequences for clinical practice. Clin J Pain, 26, 747–53.
Hass, B.W. and Canli, T. (2008). Emotional memory function, personality structure and psychopathology: a
neural system approach to the identification of vulnerability markers. Brain Research Review, 58, 71–84.
Haythornthwaite, J.A., Menefee, L.A., Heinberg, L.J. and Clark, M.R. (1998). Pain coping strategies predict
perceived control over pain. Pain, 77, 33–9.
Hill, A., Niven, C.A. and Knussen, C. (1995). The role of coping in adjustment to phantom limb pain. Pain,
62, 79–86.
Jensen, M.P., Turner, J.A., Romano, J.M. and Strom, S.E. (1995). The Chronic Pain Coping Inventory:
Development and preliminary validation. Pain, 60, 203–16.
Kerns, R.D., Turk, D.C. and Rudy, T.E. (1985). The West Haven-Yale Multidimensional Pain Inventory
(WHYMPI). Pain, 23, 345–56.
Klasen, B.W., Bruggert, J., Hasenbring, M. (2006). The role of cognitive pain coping strategies for
depression in chronic back pain path analysis of patients in primary care. Der Schmerz, 5, 398–410.
Kraimaat, F.W. and Evers, A.W.M. (2003). Pain-coping strategies in chronic pain patients: psychometric
characteristics of the pain-coping inventory (PCI). International Journal of Behavioral Medicine,
10, 343–63.
Lang, P.J. (1970). Stimulus control, response control, and the desensitization of fear. In: D.J. Levis (Ed.)
Learning approaches to therapeutic behavior change, pp. 148–73. Chicago, IL: Aldine Press.
Leeuw, M., Goossens, M.E., van Breukelen, G.J., et al. (2008). Exposure in vivo versus operant graded
activity in chronic low back pain patients: results of a randomized controlled trial. Pain, 138, 192–207.
Legrain, V., VanDamme, S., Eccleston, C., Davis, K.D., Seminowicz, D.A. and Crombez, G. (2009). A
neurocognitive model of attention to pain: behavioral and neuroimaging evidence. Pain, 144, 230–2.
Lethem, J., Slade, P.D., Troup, J.D.G. and Bentley, G. (1983). Outline of a fear-avoidance model of
exaggerated pain perception. Behavior Research and Therapy, 21, 401–8.
Lewinsohn, P.M., Libet, J.M. (1972). Pleasant events, activity schedules and depression. Journal of
Abnormal Psychology, 79, 291–5.
Linton, S. (1985). The relationship between activity and chronic back pain. Pain, 21, 289–94.
Lundberg, M. and Styf, J. (2009). Kinesophobia among physiological overusers with musculoskeletal pain.
Masedo, A.I. and Esteve, R. (2007). Effects of suppression, acceptance and spontaneous coping on pain
tolerance, pain intensity and distress. Behaviour Research and Therapy, 45, 199–209.
McCaul, K.D. and Haugtvedt, C. (1982). Attention, distraction, and cold pressor pain. Journal of
Personality and Social Psychology, 43, 154–62.
McCaul, K.D., and Mallot, J.M. (1984). Distraction and coping with pain. Psychological Bulletin, 95, 516–33.
McGill, S.M., Hughson, R.L. and Parks, K. (2000). Lumbar erector spinae oxygenation during prolonged
contractions: implications for prolonged work. Ergonomics, 43, 486–93.
Nicholas, M.K. (2007). The pain self-efficacy questionnaire: taking pain into account. European Journal of
Pain, 11, 153–63.
Norman, D.A. and Bobrow, D.G. (1975). On data-limited and resource-limited processes. Cognitive
Osborne, T.L., Jensen, M.P., Ehde, D.M., Hanley, M.A. and Kraft, G. (2007). Psychosocial factors
associated with pain intensity, pain-related interference, and psychological functioning in persons with
multiple sclerosis and pain. Pain, 127, 52–62.
Philips, H. (1987). Avoidance behavior and its role in sustaining chronic pain. Behavior Research and
Therapy, 25, 273–9.
Philips, H.C. and Jahanshahi, M. (1986). The components of pain behavior report. Behavior Research and
Therapy, 24, 117–25.
Rachman, S.J. (2004). Fear and courage: a psychological perspective. Social Research, 71, 149–76.
Riley, J.L. and Robinson, M.E. (1997). CSQ: Five factors or fiction? The Clinical Journal of Pain, 13,
156–62.
Romano, J.M., Jensen, M.P.,and Turner, J.A. (2003). The Chronic Pain Coping Inventory-42: Reliability
and validity. Pain, 104, 65–73.
Rosenstiel, A.K. and Keefe, F.J. (1983). The use of coping strategies in chronic low back pain patients:
relationship to patient characteristics and current adjustment. Pain, 17, 33–44.
Rusu, A.C. and Hasenbring, M. (2008). Multidimensional pain inventory derived classifications of chronic
pain: evidence for maladaptive pain-related coping within the dysfunctional group. Pain, 134, 80–90
Salomons, T.V., Johnstone, T., Backonja, M. and Davidson, R.J. (2004). Perceived controllability
modulates the neural response to pain. Journal of Neuroscience, 24, 7199–203.
Sanders, S. (1983). Automated versus self-monitoring of up-time in chronic low-back pain patients: a
comparative study. Pain, 15, 399–405.
Sbriccoli, P., Yousuf, K., Kupershtein, I., Solomonow, M., Zhou, B.H., Zhu, M.P. et al. (2004). Static load
repetition is a risk factor in the development of lumbar cumulative musculoskeletal disorder. Spine,
29, 2643–53.
Seminowicz, D.A. and Davis, K.D. (2006). Cortical responses to pain in healthy individuals depends on
pain catastrophizing. Pain, 120, 297–306.
Seminowicz, D.A. and Davis, K.D. (2007). Interactions of pain intensity and cognitive load: the brain stays
on task. Cerebral Cortex, 17, 1412–22.
Solomonow, M., Hatipkarasulu, S., Zhou, B.H. et al. (2003a). Biomechanics and electromyography of a
common idiopathic low back disorder. Spine, 28, 1235–48.
Solomonow, M., Baratta, R.V., Zhou, B.H. et al. (2003b). Muscular dysfunction elicited by creep of lumbar
viscoelastic tissues. J Electromyogr Kinesiol, 13, 381–93.
Stokes, I.A.F. and Iatridis, J.C. (2004). Mechanical conditions that accelerate intervertebral disc
degeneration: overload versus immobilization. Spine, 29, 2724–32.
Sudhaus, S., Fricke, B., Schneider, S., Stachon, A., Klein, H., vonDüring, M. and Hasenbring, M. (2007).
The cortisol awakening response in patients with acute and chronic low back pain. Relations with
psychological risk factors of pain chronicity. Schmerz, 21, 202–11.
Sudhaus, S., Fricke, B., Stachon, A., Schneider, S., Klein, H., vonDüring, M. and Hasenbring, M. (2009).
Salivary cortisol and psychological mechanisms in patients with acute versus chronic low back pain.
PNEC, 34, 513–22.
Sullivan, M.J.L., Thorn, B., Haythornthwaite, J.A., et al. (2001). Theoretical perspectives on the relation
between catastrophizing and pain. Clin J Pain, 17, 52–64.
Tait, R.C., Chibnall, J.T., & Krause, S. (1990). The Pain Disbility Index: psychometric properties. Pain, 40,
171–82.
Tan, G., Jensen, M.P., Robinson-Whelen, S., Thornby, J.I. and Monga, T.N. (2001). Coping with chronic
pain: A comparison of two measures. Pain, 90, 127–33.
Truchon, M. and Coté, D. (2005). Predictive validity of the chronic pain coping inventory in subacute low
back pain. Pain, 116, 205–212.
Turk, D.C. and Rudy, T.E. (1988). Toward an empirically derived taxonomy of chronic pain patients:
integration of psychological assessment data. Journal of Consulting and Clinical Psychology, 56, 233–8.
Turner, J.A. and Clancy, S. (1986). Strategies for coping with chronic low back pain: relationship to pain
and disability. Pain, 24, 355–64.
Turner, J.A. and Aaron, L.A. (2001). Pain-related catastrophizing: what is it? Clin J Pain, 17, 65–71.
VanDamme, S., Crombez, G., Eccleston, C. and Roelofs, J. (2004). Impaired disengagement from
threatening cues of impending pain in a crossmodal cueing paradigm. European Journal of Pain,
8, 227–36.
Vendrig, A.A., Lousberg, R. (1997). Within-person relationships among pain intensity, mood and physical
activity in chronic pain: a naturalistic approach. Pain, 73, 71–6.
Verbunt, J., Westerterp, K., Van der Heijden, G.J., Seelen, H.A., Vlaeyen, J.W.S., Knottnerus, J.A. (2001).
Physical activity in daily life in patients with chronic low back pain. Arch Phys Med Rehabil, 82, 726–30.
Verbunt, J.A., Seelen, H.A., Vlaeyen, J.W., et al. (2003). Disuse and deconditioning in chronic low back
pain: concepts and hypotheses on contributing mechanisms. European Journal of Pain, 7, 9–21.
Vlaeyen, J.W. and Linton, S.J. (2000). Fear-avoidance and its consequences in chronic musculoskeletal
pain: a state of the art. Pain, 85, 317–32.
Vlaeyen, J.W.S., de Jong, J., Geilen, M., et al. (2002). The treatment of fear of movement/(re)injury in
chronic low back pain: further evidence on the effectiveness of exposure in vivo. Clinical Journal of
Pain, 18, 251–61.
Waddel, G., Newton, M., Henderson, I., Somerville, D., Main, C.J. (1993). A fear-avoidance beliefs
questionnaire (FABQ) and the role of fear-avoidance beliefs in chronic low back pain and disability.
Pain, 52, 157–68.
Waddell, G. (2004). The Back pain revolution. London: Churchill Livingston.
Walk, R. (1956). Self-ratings of fear in a fear-avoking situation. Journal of Social and Abnormal Psychology,
52, 171–8.
Wedderkopp, N., Leboeuf-Yde, D.C., Andersen, L.B., Froberg, K., Hansen, H.S. (2003). Back pain in
children. Spine, 17, 2019–24.
Wegner, D.M., Schneider, D., Carter, S.R. and White, T.L. (1987). Paradoxical effects of thought
suppression. Journal of Personaliyt and Social Psychology, 53, 5–13.
Wittink, H., Rogers, W., Sukiennik, A., Carr, D.B. (2003). Physical functioning: self-report and
performance measures are related but distinct. Spine, 20, 2407–13.
Wong, K.C.H., Lee, R.Y.W. and Yeung, S.S. (2009). The association between back pain and trunk posture
of workers in a special school for the severe handicaps. BMC Musculoskeletal Disorders, 10, 43.
Wong, W.S., Jensen, M.P., Mak, K.H., Tam, B.K.H. and Fielding, R. (2010). Preliminary Psychometric
Properties of the Chinese Version of the Chronic Pain Coping Inventory (ChCPCI) in a Hong Kong
Chinese Population. The Journal of Pain, 11, 672–80.
Zautra, A.J., Burleson, M.H., Smith, C.A. et al. (1995). Arthritis and perceptions of quality of life: an
examination of positive and negative affects in rheumatoid arthritis patients. Health Psychology,
14, 399–408.
Chapter 16
Cognitive Processing and Self-Pain

Enmeshment in Chronic Back Pain
Adina C. Rusu and Tamar Pincus
16.1 Introduction
Given the detrimental effect of chronic pain on the individual and his or her surrounding
environment, research into the factors and mechanisms that contribute to the aetiology, persist-
ence, and effective management of chronic pain has increased significantly in the last decades
(Morley et al. 1999; McCracken and Turk 2002; Turk and Okifuji 2002). This growing research
interest has (amongst other outcomes) resulted in the identification of a number of psychological
factors and mechanisms that may predispose persons toward, or conversely protect them against
the development of persistent pain. This chapter contributes to this line of research by focusing
on the specific contribution of some of these proposed vulnerability factors. The present chapter
focuses on cognitive mechanisms and processes occurring at strategic and automatic levels that
constitute a mediating or moderating role in pain and pain-related disability. Before introducing
the precise research questions that have been explored in this context, in the following paragraphs
a brief overview of cognitive processing theories will be provided. The main part of this chapter
will focus on the theoretical background against which the current research was conducted.
Furthermore, empirical evidence on cognitive processing biases in chronic pain will be reviewed.
Finally, this chapter will end with perspectives on the main clinical implications of this line of
research and future research directions.
16.2 Cognitive processing and schemas

Cognition has been defined as ‘ a generic term embracing the quality of knowing, which
includes perceiving, recognizing, conceiving, judging, sensing, reasoning and imaging’ (Stedman’s
Medical Dictionary 1976). Individuals actively process incoming stimuli using pre-existing
mental structures (schemas) constructed through experience. These schemas play an important
role in organizing and integrating information, including memory processes. Distorted underly-
ing assumptions and errors in information processing can serve to maintain emotional distur-
bance (Beck 1976 ). Many different terms are used when referring to cognitive structures.
A ‘meta-construct’ framework (Ingram et al. 1998) suggests three categories for particular
groups of constructs: cognitive structures (e.g. memory), and propositions or content (e.g.
knowledge) which includes schema-based models; cognitive products (e.g. attributions, thoughts,
beliefs), some of which are thought to be accessible to conscious awareness; and cognitive
processes (e.g. conditioning, retrieval) which occur at a subconscious level. In short, automatic
processes are conceived as largely unconscious and involuntary, requiring a minimum of process-
ing capacity, and being relatively difficult to regulate, whilst controlled processes are conceived
as conscious and voluntary, requiring considerable processing capacity, and being easier to
adjust (Beck and Clark 1997; McNally 1995; Shiffrin and Schneider 1977). Taken together,
the consideration of both implicit and explicit processes in the prediction of any behaviour can
contribute to a more thorough and complete understanding of the various components that
constitute this particular behaviour.
The application of paradigms from cognitive science to clinical studies has been very produc-
tive in contributing to the understanding of the information processing and cognitive content
involved in anxiety and depression (e.g. Alloy et al. 1997). One of the aims of cognitive research
is to understand individual differences in emotionality, and particularly differences in vulnerabil-
ity to pathological emotional states. Two general types of studies are therefore of interest:
(1) Studies comparing groups differing in trait measures of negative emotionality, and (2) studies
contrasting individuals with or without emotional disorders such as anxiety or depression.
Information processing approaches to psychopathology propose that cognitive biases contribute
to the onset and maintenance of psychopathology. More specifically, individuals’ negative affec-
tive state or specific concerns are thought to be associated with favouring the processing of infor-
mation that is congruent to these affective states and concerns, and this selective processing of
information is then assumed to guide subsequent behaviour (Beck and Clark 1997; Mathews and
MacLeod 1994; Williams et al. 1997).
16.2.1 Cognitive biases in chronic pain

Empirical studies of information processing biases in chronic pain have explored selective atten-
tion, interpretation of meaning and memory/recall. It has been argued that information-processing
biases could maintain or exacerbate pain in different ways: Attentional biases may increase
monitoring of physical sensations, hypervigilance, and therefore pain behaviours; interpretation
bias may increase interpretation of sensations as painful; recall bias may increase distress and may
impact on self-image, which can reduce healthy behaviour and increase actual damage (Pincus
1998; Waddell 1996). Biased information processing has been viewed as a risk factor for the devel-
opment of chronic disability and has been shown to predict higher health care costs (Pincus and
Morley 2002; Pincus and Newman 2001).
Attention bias
Attention is probably the most extensively investigated cognitive process in relation to pain. Most
studies on the detection of attention biases towards pain stimuli have relied on experimental
paradigms such as the emotional Stroop task and the visual dot-probe task. Although some stud-
ies found support for attention biases in pain patients using these paradigms, these results seemed
to be difficult to replicate in other studies and seemed to depend on the specific task and the exact
type of stimulus material that were used to assess the bias such as sensory pain words, affective
pain words or pictoral stimuli (Asmundson et al. 1997, 2005; Asmundson and Hadjivropoulos
2007; Roelofs et al. 2002, 2005). However, the most recent studies found attentional biases towards
sensory-related but not affective words using large sample sizes (Dehghani et al. 2003; Sharpe
et al. 2009). In healthy people with elevated levels of pain-related anxiety, results on the occur-
rence of attention bias towards pain-related stimuli showed more consistency (Asmundson et al.
1997, 2005; Keogh and Cochrane 2002; Keogh et al. 2001a, 2001b), although failures in replicat-
ing these findings have been reported as well (e.g. Roelofs et al. 2003a, 2003b). The observation of
more consistent findings in healthy people and inconsistent findings in pain patients might sug-
gest that the paradigms tested were not sufficiently difficult for chronic pain patients who might
have learned to allocate their attentional resources in the continuous presence of actual pain
(Pincus and Morley 2001). In a similar view, Asmundson et al. (2005) proposed that the use of
sensory and affective pain words as threatening stimuli might fail in assessing attention biases in
patients because these stimuli are not threatening enough for pain patients, in contrast to healthy
SELF-PAIN ENMESHMENT IN CHRONIC BACK PAIN 317
persons who are not suffering from persistent pain. In conclusion, emotional Stroop and dot-
probe paradigms might suffer from low ecological validity due to the use of verbal and pictorial
stimulus material. Several studies in which an ecologically more appropriate paradigm has been
used (i.e. the primary task paradigm), have demonstrated the attention demanding properties of
pain, and the interruptive effect that pain can have on the allocation of attention to other tasks
(e.g. Crombez et al. 1998a, 1998b; Eccleston and Crombez 1999; Peters et al. 2002). In the primary
task paradigm, the effect of the administration of electrical pain stimuli on the performance of a
simple cognitive task (e.g. distinguishing high and low pitch tones) is examined. The resulting
interruptive effect of pain on attention has proven to be amplified by the intensity, novelty,
unpredictability, and threat value of the pain stimulus (Eccleston and Crombez 1999), and atten-
tional interference has been found to be pronounced in persons high in pain catastrophizing and
fear of pain (Carleton et al. 2005; Crombez et al. 1998a, 1998b). Other studies have demonstrated
that attention shifts to pain and pain-related cues and, once detected, pain is difficult to disengage
from (e.g. Koster et al. 2006; Van Damme et al. 2004a). Based on this line of research, it can be
concluded that there is initial evidence that pain demands attention, both in pain patients, but
also in healthy individuals with high levels of specific pain-related fears, which has led to the
development of a neurocognitive model of attention to pain, including behavioural and neuroim-
aging evidence (reviewed in Legrain et al. 2009; Van Damme et al. 2010).
In sum, it can be concluded that there is accumulating evidence for a sensory-related atten-
tional bias in chronic pain (Boissevan 1994; Crombez et al. 2000; Pearce and Morley 1989; Snider
et al. 2000; Sharpe et al. 2009), and specifically there is some evidence to suggest that anxiety in
the presence of pain may be associated with interference (Asmundson et al. 1997; Boissevan 1994;
Pincus et al. 1998; Snider et al. 2000).
Interpretation bias
Data from studies of interpretative biases show consistently a bias for illness- and health-related
material; both pain patients and persons with high levels of pain-related fear are inclined to inter-
pret ambiguous and innocuous pain-related stimuli in a threatening or negative fashion (Edwards
and Pearce 1994; Pincus et al. 1994; Pincus, et al. 1996a; Vancleef 2007; Vancleef et al. 2009).
Recently, Keogh and colleagues (Keogh, Ellery et al. 2001; Keogh, Hamid, Hamid and Ellery
2004) have demonstrated that negative interpretative bias for ambiguous situations mediated the
relation between individual levels of anxiety sensitivity and pain tolerance in a cold-pressor test.
Nevertheless, it should be noted that studies on interpretive bias in the context of pain have relied
on explicit measures predominantly, using homophones (words with the same pronunciation,
but a different spelling: e.g. dye/die (McKellar et al. 2003; Pincus et al. 1994), homographs (words
with the same spelling, but different meanings: e.g. needle (Pincus et al. 1996), word-stem com-
pletion tasks (word stem can be completed in different ways: e.g. ten—: tender/tennis (Edwards
and Pearce 1994) or interpretation questionnaires (e.g. Keogh et al. 2001b; Keogh et al. 2004).
Consequently, it remains unclear whether the negative bias results from elaborative, constructive,
integrative processes at the moment of responding to ambiguity, or if negative interpretations
occur actually at an automatic spontaneous level.
Memory bias
The experimental paradigms used to assess memory biases differ from questionnaire methodol-
ogy, in that they are likely to tap preconscious, rather than conscious cognitive processes, and are
therefore less susceptible to response biases. When these types of experiments are computerized,
experimenter effects are additionally minimized (Pincus 1998). These paradigms have been
useful in the investigation of diatheses in other populations (see Pincus and Williams 1999).
Experiments into recall bias and chronic pain typically include an ‘encoding stage’ where partici-
pants are presented with a list of words, and a ‘recall phase’ when they are asked to recall as many
of the words as possible within a given timeframe. The words are selected according to their asso-
ciation with a particular content category, for instance, pain-related content, or neutral content
not associated with pain. The underlying assumption is that people will remember words that are
congruent with their emotional state (see Pincus and Morley 2001).
While evidence for recognition bias in pain patients is currently scarce (Pincus and Morley
2001), there seems to be robust evidence for recall bias towards pain-related words. It has also
been suggested that memory is an important means of developing representations of the self, and
therefore, recall bias can highlight significant information about an individual’s cognitive struc-
tures (Pincus and Morley 2001). Previous investigations have demonstrated that pain patients
recall more pain-related autobiographical memories (Wright and Morley 1995) than people not
in pain; however, Eich et al. (1990) found that this effect was entirely mediated by mood.
Furthermore, as with similar studies in depression, pain patients’ greater autobiographical recall
might simply reflect the greater number of pain experiences available to this group.
Experimental methodology (incidental world-list recall) avoids the issue of differences in actual
experiences. Chronic pain patients with varying diagnoses recalled more words describing sen-
sory aspects of pain than neutral words (Edwards et al. 1992; Johnson and Spence 1997; Pearce
et al. 1990). It is further well established that chronic pain patients show greater recall for sensory-
pain words encoded in reference to the self than to others (Edwards et al. 1992; Koutanji et al.
1999; Pincus et al. 1993, 1995). Koutanji et al. (1999) applied similar methodology and found
similar results in a population of children with chronic pain experience due to arthritis. These
results suggest that preferential recall in pain patients is associated with pain words that had been
encoded in reference to the self and suggest that an individual’s self-schema plays a pivotal role in
processing biases.
Differences have also been found in the types of words recalled by chronic pain patients with
and without depression. Depressed pain patients have been found to show biased recall towards
both sensory and affective (e.g. ‘dependent’) aspects of pain, whilst non-depressed pain patients
show a bias only towards sensory material (Calfas et al. 1997; Clemmey and Nicassio 1997;
Edwards et al. 1992; Koutanji et al. 1999). In order to find out which cognitive biases are attribut-
able to pain and which to mood, Pincus et al. (1995) further refined this methodology by present-
ing depression descriptors and illness-and health-related stimuli that were either positive or
negative in meaning (e.g. ‘suffering’ as a negative health-related word and ‘withdrawn’ as a nega-
tive depression-related word), and found that patients with pain who are depressed or distressed
show a bias for illness- and health-related stimuli, but not for depression-related words. However,
in the absence of a control group with clinically depressed participants without pain, the conclu-
sion that depressed pain patients show a bias towards pain and illness words rather than words
associated with depression cannot be conclusively established. Specifically, people with chronic
pain show preferential recall for pain-sensory words compared to neutral words when rating the
degree to which these words refer to themselves, whereas normal controls show no differential
recall (Koutantji et al. 1999; Pincus et al. 1993).
Wells et al. (2003) investigated the impact of diagnostic status on information processing biases
among chronic pain and ankylosing spondylitis patients. Results showed that diagnosed chronic
pain patients demonstrated a recall bias away from depression-related words, whilst the non-
diagnosed chronic pain patients did not. It was suggested that a diagnosis in chronic pain
patients might serve as a ‘buffering’ device against classic depression-related feelings of guilt,
shame, and self-blame, which were encapsulated in the depression-related words of the above
mentioned study.
Altogether, these findings provide support for the notion that depressed chronic pain patients
experience depression that is qualitatively different from that of depression in the conventional
psychiatric sense, in which the focus is on guilt, shame, and self-denigration. Pincus et al. (1995)
hypothesized that ongoing exposure to pain might result in processing biases towards sensory-
pain information, but in some pain patients, there is also an additional bias for pain-distress
stimuli, which may be associated with poor self-image and increased levels of depression. Finally,
although anxiety has been measured in a number of studies, the relationship between anxiety and
memory bias appears to be insubstantial (Pincus and Morley 2001). One suggestion which might
account for the inconsistent findings in this area is that the differences may be due to the nature
of the words used, which lends further support to the importance of cognitive specificity, sam-
pling biases due to convenience sampling or different recruitment settings (Pincus and Morley
2001). In sum, it can be concluded that memory biases are an important consideration in the
onset and exacerbation of maladaptive responses to pain, although further clarification is neces-
sary on the precise structure and impact of these processes in relation to maladaptive responses to
physical threat.
16.3 Cognitive models of chronic pain

As described in the previous sections of this book, psychological theories work from the perspec-
tive that pain, and the disability related to it, are not only influenced by physiological pathology
but also by psychological and social factors. Chronic pain is not only a somatic problem. It is
influenced by the individual’s attitudes and beliefs about their symptoms and the distress and
behavioural responses initiated by their experience of pain. There are a number of psychological
theories related to chronic pain, however, behavioural and cognitive-behavioural theories are the
most relevant ones which have been discussed in previous chapters. The purpose of the following
section is to describe predominately cognitive models to pain processing and to review the evi-
dence concerning the cognitive mechanisms by which psychological factors modulate pain in
both clinical and experimental contexts. The experience of pain is never an isolated sensory event
and it never occurs without the influence of context and meaning. Pain is influenced by beliefs,
attention, expectations and emotions, regardless of whether it occurs during the most controlled
laboratory circumstances or during circumstances of physical trauma or emotional distress (Price
and Bushnell 2004).
16.3.1 General model of pain processing

Price (1999) has proposed a conceptual model of pain processing to allow the study of temporal
characteristics of sensory-discriminative components, affective, cognitive and behavioural
responses. The first stage of this model consists of an initial sensory-discriminative process, which
leads to the perceived intensity of the pain sensation. Immediate appraisals are associated with the
sensory features of pain, autonomic and somatomotor activation, and perception of the immedi-
ate context surrounding the pain. The second stage, the immediate pain unpleasantness, reflects
immediate affective responses and involves limited cognitive processing. This stage is influenced
by the first stage; therefore the immediate unpleasantness can comprise a moment-by-moment
unpleasantness, distress, annoyance often closely linked to the intensity of the painful sensation
and the accompanying arousal. The third stage of pain processing involves longer-term cognitive
processes that relate to the meanings or implications that pain holds for one’s life and has been
termed extended pain affect and conceptualized as suffering, which might include depression,
frustration, anxiety and anger. It reflects the personal experience, as opposed to overt behaviour
and it is largely autobiographical and directed toward the long-term past and long-term future.
It is influenced by expectations, beliefs, and meanings such as hope for alleviation, belief in one’s
ability to endure the pain, and perceived control over reducing the pain intensity. Thus, an indi-
vidual’s attitudes, beliefs, and memories about the real or imagined long-term consequences of
having pain influence his or her adaptation to this stage. Consequently, these are the mediators
that are the focus of cognitive-behavioural interventions and may explain differing levels of adap-
tation. The final component of the model is the overt behavioural expression of pain, such as
moaning, lying down during the day, or declining to participate in daily responsibilities.
Immediate pain unpleasantness, in turn, causes extended emotions related to pain because it
can provide an immediate cue for the meanings related to pain-related negative emotions once
these meanings have been established over time (Price 1999). For example, the sudden exacerba-
tion of pain in a cancer patient serves as an instant reminder of the progress of the disease, with
its implications for deterioration and death. In addition, the difficulty of having to endure imme-
diate pain unpleasantness over long periods of time (days, weeks, months) can lead to pain-
related negative emotions. Parallel influences on extended pain-related emotions can also occur
as they can be enhanced or diminished by arousal, contextual meanings, and other psychological
factors that are present. Pain itself can be conceptualized as both an exteroceptive and interocep-
tive phenomenon depending on the type of tissue that is stimulated and the types of sensory
qualities that are present during pain. Consistent with Damasio’s (1994) view of the mechanisms
of emotion, pain unpleasantness represents several sources, including pain sensory qualities,
arousal, visceral, and somatomotoric responses, which in combination with appraisal of the
context in which they occur, give rise to a felt meaning of what is happening to the body and the
self, often not necessarily accompanied by specific thoughts. This felt meaning derives largely
form the experience of the body and constitutes the immediate unpleasantness of pain (Price and
Bushnell 2004).
Rainville et al. (1999) demonstrated this serial interaction by specifically targeting hypnotic
suggestions toward unpleasantness, and thereby decreasing or enhancing ratings of the immedi-
ate unpleasantness of a hot stimulus, without changing pain sensation intensity. In contrast, when
suggestions were directed only toward changing pain sensation intensity, both pain sensation
intensity and pain unpleasantness ratings changed in parallel. The combination of these two sets
of results helps to establish that pain sensation intensity is a cause of pain unpleasantness and
not vice versa. The four-stage model of pain processing is consistent with the neural mechanisms
of the sensory and affective dimensions of pain. For example, serial interactions between pain
sensation, immediate pain unpleasantness, and extended pain-related affect are respectively asso-
ciated with serial interactions between somatosensory cortices, insular, and cingulate cortices,
and prefrontal and temporal cortical areas (reviewed in Price 1999; Price and Bushnell 2004).
16.3.2 Enmeshment and chronic pain

Chronic pain is by its nature prolonged and aversive, generally causing considerable disruption to
the sufferer’s life. Pain can present overwhelming obstacles to normal activities: work, leisure, and
relationships with family and friends. The subjectivity of the experience can impose real isolation,
and the personal meaning of pain and its impacts may threaten the individual’s sense of who he
or she is (Clyde and Williams 2002; Morley and Eccleston 2004). Researchers and clinicians have
only recently begun to study the impact of pain on future possible health and self-identity. Reviews
consistently show that depression in pain is significantly higher than in the general population,
and often higher than in other medically ill populations (Banks and Kerns 1996; Fishbain et al.
1997; Sullivan et al. 1992), which leads to the important issue of what it is about chronic pain that
causes such distress.
Pain and the self

Persistent pain can generate interruptions to attention, memory processes and current thinking,
interference with a range of everyday activities, and can pose a threat to identity. Morley and
Eccleston (2004) have therefore proposed the framework of the three levels of interruption, inter-
ference, and identity. They referred to the tremendous value of pain as a threat and the capacity to
threaten a person at different levels and argue that problem-solving similarly needs to occur at
each level. It has been argued that chronic pain establishes itself in the context of an ongoing and
unfolding personal dynamic related to the current stage of a life cycle. Pain will impact on ongo-
ing developmental tasks and particular goals determined by motivational states and earlier expe-
riences (Morley and Eccleston 2004). While a wealth of experimental studies have shown that the
extent of the interruption is a function of stimulus characteristics and individual differences in
threat perception (Eccleston and Crombez 1999), the effect of interference with activities of daily
life can be seen in the frequent reports of frustration made by chronic pain patients (Harris et al.
2003; Morley et al. 2005; Price 1999; Wade et al. 1990). More specifically, Morley and Eccleston
(2004) suggest the interference with mundane ‘low-level’ everyday tasks is psychologically sig-
nificant because they put a burden on others and exert pressure on the sufferer to redefine the self.
In addition, there is evidence linking self-reported depression in chronic pain to interference with
social roles (Harris et al. 2003; Morley and Eccleston 2004). A key issue in any threat to identity is
whether repeated interference with major goals will impact on the self-schemas and therefore on
the person’s identity (Leventhal et al. 1999). In pain, it has been suggested that current pain inter-
feres with an individual’s current tasks, plans, and goals and causes a ‘biographical disruption’
(Bury 1988) that changes the person’s perspective of himself or herself with regard to the past and
future (Hellström et al. 2000, 2001). Biographical disruptions appear to enhance emotional
distress, and might lead to anxious and depressed mood, which in turn might inhibit active
problem-solving (Morley and Eccleston 2004; Morley et al. 2005).
How pain might impact on identity can be further understood by the psychology of motivation
and goal-related behaviour, as encompassed in self-regulation and self-discrepancy theories.
However it would be beyond the scope of the present chapter to review the various conceptualiza-
tions in contemporary psychology (Carver and Scheier 1999; Carver et al. 1999; Karoly 1999;
Oatley 1992; Higgins 1997). Goals are often classified into ‘approach’ goals, a tendency to reduce
the discrepancy between the current position and the goal, and ‘avoidance’ goals, which require
the person to increase the discrepancy between the current state and the goal (Carver and Scheier
1998, 1999). A crucial implication that follows from the frameworks of both Higgins’ and Carver
and Scheier is that individuals who are dominated by avoidance goals are more likely to experi-
ence anxiety and fear when these goals are threatened. In contrast, individuals whose primary
goal state is approach-oriented will experience a sense of loss and sadness when these goals are
threatened. Self-discrepancy theory (Higgins 1987, 1997) proposes that the self is represented in
three domains, actual, ideal, and ought self. The common method used in self-discrepancy research
asks participants to generate a list of descriptors that characterize the various aspects of the
self. Studies have confirmed the relationship between self-discrepancies and experienced
affect (Higgins 1987, 1997; Higgins and Tykocinski 1992; Higgins et al. 1994). However, self-
discrepancy methodology has been criticized on grounds that it does not directly assess particular
explicit goals, and attributes represent rather meta-cognition and the representation of a person’s
motivation and goals (Morley and Eccleston 2004). It has been argued that self-regulation and
self-discrepancy theory might provide a useful framework for understanding the variety of emo-
tional experience in chronic pain patients by providing a starting point for more compelling sys-
tematic research in this area (Morley and Eccleston 2004). In a first attempt, Morley et al. (2005)
developed a new method of assessing future possible selves (Ross and Buehler 2004) and self-pain
enmeshment (Pincus and Morley 2001), which was based on self-discrepancy theory (Higgins
1987, 1997) and the concept of possible selves, which was introduced by Markus and Nurius
(1986). Chronic pain patients generated characteristics describing their current actual self, hoped-
for self, and feared-for self, and made judgements about the degree to which their future possible
selves (hoped-for and feared-for) were dependent on the absence or presence of pain, as an indi-
cator for pain enmeshment. Analyses showed that the degree of role interference attributable to
pain, and the proportion of hoped-for self characteristics that could be achieved even in the pres-
ence of pain predicted the magnitude of depression and acceptance scores. Finally, it should be
noted that self-regulation and self-discrepancy theories, which have not been developed in the
context of chronic pain, do not appear to make predictions concerning the enmeshment of pain
and self as encapsulated in the Schema Enmeshment Model of Pain (SEMP; Pincus and Morley
2001), which is described below, and future research will have to examine its appropriateness for
the field of chronic pain.
The Schema Enmeshment Model of Pain

How can the dynamic interplay between pain, mood and circumstances be represented to under-
stand the observed differences observed in cognitive biases? Clyde and Williams (2002) argued
that there is a need for a new model to aid understanding and treatment of pain and depression.
Such a model needs to incorporate several factors: (1) the relationship between pain and depres-
sion; (2) that the development of depression is not inevitable in the context of chronic pain;
(3) there is a need to take into account both vulnerability to depression and the nature of the
individual experience of pain (stressor); and (4) there may be several pathways by which pain and
depression co-occur, resulting in qualitatively different subtypes of depression.
The SEMP model has been proposed as an extension and refinement of self-discrepancy theory.
In reviewing the existing literature on cognitive biases in chronic pain Pincus and Morley (2001)
proposed that the degree to which aspects of the self were enmeshed with pain contributed to the
observed cognitive bias. In particular the authors proposed that pain patients with concurrent
distress/depression exhibited greater enmeshment of self and pain schema, hence causing suffer-
ing and disability. It is postulated that three self-schemas interact (see Figure 16.1): those pertain-
ing to pain (sensory features), illness consequences (behavioural and emotional), and self
(a dynamic, multifaceted structure including evaluation of self-worth). The SEMP model attempts
to account for the variation in psychological adaptation to chronic pain, and aims to explain the
different information-processing biases found in people with chronic pain. The pain schema
contains the more immediate properties of the experience of pain, and is associated with the
interruption of existing behaviour and the initiation of self-protective behaviours. The illness
schema incorporates affective and behavioural consequences of illness, and is likely to have an
effect on goal attainment, independence, and quality of life. Finally, the self-schema might con-
tain trait-like descriptions of the self or specific behavioural episodes that have been experienced,
and is associated with feelings of self-worth (Bradley and Mathews 1983). Importantly, the schema
enmeshment model takes account of previous vulnerabilities which might affect an individual’s
adaptation. For example, a pre-existing vulnerability to depression might mean that an individu-
al’s self-worth is more likely to be affected by the experience of pain. This diathesis–stress approach
to understanding depression in chronic pain has been described in detail by Banks and Kerns
(1996). The schema enmeshment model, however, focuses more specifically on the role of
schemas in predisposing an individual to developing depression in the presence of chronic pain.
When the content of two or more schemas is activated simultaneously and repetitively, it is
hypothesized that the content of one schema might become incorporated into another schema,
Pain Illness Pain Illness
Self Self
(a) Healthy normal (b) Coping with

enmeshment chronic pain
Pain
Pain Illness
Illness
Self
Self
(c) Non-problematic pain (d) Enmeshment resulting

enmeshment in distress & IPB
Fig. 16.1 The Schema Enmeshment Model (Pincus and Morley 2001).
hence causing an ‘enmeshment’ of the two schemas. Pincus and Morley (2001) suggest that the
three schemas always overlap to some extent, but it is the amount of enmeshment, the content of
the schemas, and the timing and context of the enmeshment that is important. In adaptive
coping, pain and illness are not considered to impact on the person’s sense of self-worth, which
could be explained by a lack of enmeshment between his/her self-schema and the illness/pain
schemas. It is hypothesized that the self-schema is most likely to become enmeshed with the pain
and illness schemas if it already contains information about dependence and distress.
In light of the SEMP model, three possibilities of how the self-schema might become vulnerable
have been proposed (Pincus and Morley 2001). First, there will be a small percentage of patients
who have experienced an episode of depression before the onset of pain. Because of this previ-
ously established vulnerability it is likely that they are at risk for the development of chronic dis-
tress. Second, other patients might have not experienced a previous episode of depression, but
have experienced life events which have compromised their sense of self-worth; previous experi-
ences have made them vulnerable. Even without the onset of pain, people in this group are more
likely to experience a depressive episode, and pain could act as a ‘trigger’ event. This notion is in
line with the work of Banks and Kerns (1996), stating that chronic pain is a highly influential
stressor, which might cause the reactivation of negative beliefs and cognitions among already
vulnerable populations, such as pain patients, resulting in negative cognitive biases. Pincus
and Morley (2001) surmise that both these groups of patients should exhibit their vulnerability
to depression by showing biases towards stimuli incorporating self-denigration when their
self-schema is activated. Finally, the last group consists of those for whom vulnerabilities have not
been established earlier, therefore they are less likely to become depressed as a result of chronic
pain, but they may become distressed. As a consequence, these patients will not show biases
towards material reflecting negative self-worth, but they will show biases towards negative
aspects of pain and illness. A more detailed critique of the SEMP model will be provided in the
following sections.
Empirical support for the main components of the SEMP

It has been shown that the application of information processing approaches to the study of
chronic pain has proven influential by avoiding the limitations associated with classical self-
report measures. Further, it has been shown that conditions such as chronic pain or depression
may be distinguished from one another, on the basis of these associated cognitive biases. Most
importantly, observations of dysfunctional processing associated with chronic pain hold impor-
tant implications for the development of new treatment modalities. The SEMP model has been
developed through close scrutiny of the existing literature and it provides many testable hypoth-
eses (Pincus and Morley, 2001): (1) processing priorities depend on the salience of information
to the content of schemas; (2) all pain patients are proposed to preferentially process pain infor-
mation; (3) self-referential material is also preferentially processed, especially when congruent
with the self-schema; (4) therefore self-referential health-related information would be preferen-
tially recalled, particularly by depressed chronic pain patients, as illness information is enmeshed,
and therefore congruent with the self schema; (5) however, enmeshment does not necessarily
indicate self-denigratory beliefs, so depressed pain patients would not preferentially recall typical
depression-related information, in contrast to clinically depressed patients without chronic pain.
Although the SEMP model has been proposed recently, there are already several existing studies,
which have directly tested predictions from the SEMP (Davies 2003; Gray 2006; Harris et al. 2003;
Morley et al. 2005; Pincus et al. 2007; Read and Pincus 2004; Sutherland and Morley 2008).
As can be seen from the abovementioned research, self-pain enmeshment can be operational-
ized and measured in different ways. A novel approach for studying self-pain enmeshment and
depressive thinking in chronic pain patients has been proposed by Pincus et al. (2007). Beside
questionnaires and information-processing methodologies, a sentence completion method has
been developed to overcome the limitations of endorsement methodology and elicit idiographic
information by describing patients’ individual perspectives (Barton and Morley 1999; Barton
et al. 2005; Rusu et al. 2009). Briefly, the Sentence Completion Test for Chronic Pain (SCP; Rusu
2008) has been illustrated as a promising approach, which might help to clarify the relationship
between pain and depression/distress, contribute to the identification of underlying schemas in
depressed pain patients, and might also be of use for case formulation. A first study using a pre-
liminary version of the SCP showed that depressed pain patients generated more negative health-
related completions (particularly directed towards the future), than non-depressed pain patients
and control participants (Pincus et al. 2007). Non-depressed pain patients focused on health as
well, but not necessarily in a negative way. Moreover, the predominance of danger-related
thoughts in anxiety patients (Beck et al. 1974; Rachman et al. 1988), contrasts with the themes of
loss and self-devaluation in depressive negative automatic thoughts (e.g. Beck et al. 1979), and is
the basis of the content-specificity hypothesis (Beck 1987; Beck et al. 1987). In pain, studies have
demonstrated that the type of depression experienced by chronic pain patients differs qualita-
tively from patients with clinical depression by a tendency for health-related negative processing,
without the component of self-denigration, shame, and guilt often found in clinical depression
(Pincus and Morley 2001; Pincus et al. 1995, 2007). While traditional measures of depressive
thinking are based on Beck’s (1970) cognitive triad of self, world and future and contain lists of
negative thoughts that occur automatically during depressed mood (Beck et al. 1961, 1987;
Hamilton 1967; Hollon and Kendall 1980), it has been argued that this kind of measure will
approximate a patient’s thoughts rather than capture them exactly (Barton and Morley 1999).
Consistent with the SEMP model, Rusu and colleagues (Rusu 2008; Rusu et al. 2008, 2009, in
prep.) manipulated self-reference and future reference as a condition during encoding in a recall
task and generation of idiographic content in a sentence completion task in a series of experi-
ments, and found that recall biases and cognitive content in pain patients was specific to pain-
and illness-related descriptors and themes. More specifically, in depressed pain patients the
negative health-related cognitive biases were firmly connected with patients’ view of themselves
in the future. This might indicate that the experience of chronic pain combined with depressed or
distressed mood might lead to a generalization and amplification of cognitive biases, which also
extends to the future (see Figure 16.2 for the hypothesized enmeshment for different groups). As
both pain groups (whether depressed or not) were involved with pain experiences, one possible
explanation for the findings of the depressed pain group would be that the implications of pain
go beyond the pain itself, into the domain of health and illness, as presumed by the SEMP model.
This explanation would be in line with the results of Pincus et al. (1995), who reported that
depressed pain patients showed a marked bias towards illness-related words, which was stronger
than for pain-related words. At present it is not known which factors are causing cognitive
vulnerabilities, or which processes might lead to the generalization and amplification of cognitive
biases, but it is conceivable that factors such as pain catastrophizing, hypervigilance and anxiety
sensitivity (Van Damme et al. 2004b), in addition to predisposing factors, might play a crucial
role in the development of distress and potentially also the maintenance of chronic pain. There is
Pain
Pain Illness
Illness
Self
Self
Depression Depression
Positive Future
Positive Future
(a) Enmeshment hypothesized for (b) Enmeshment hypothesized for

healthy individuals non-depressed pain patients
Illness Future
Pain
Depression
Self
Positive
(c) Enmeshment hypothesized for

depressed pain patients
Fig. 16.2 Hypothesized enmeshment for different groups (Rusu 2008).
evidence that hypervigilance for pain is associated with a narrowing of the attentional field and
a difficulty in disengaging attention from pain and shifting toward other demands in the envi-
ronment (Van Damme et al. 2002). This effect was more pronounced in persons with high
catastrophic thinking. Further research is needed to clarify issues of risk and vulnerability by
prospective designs.
The above mentioned findings suggest that the concept of enmeshment as captured in the
current measures may provide an alternative approach to explaining distress/depression in
chronic pain. Pincus and Morley (2001) note that enmeshment is not a bias in its own right but a
structural feature relating the self to pain experience and offered no independent way of assessing
the degree of enmeshment, which was essentially a post-hoc explanation for the observed data.
Whether high levels of enmeshment represent a risk factor for depression or conversely whether
low levels of enmeshment act to buffer the individual from depression cannot be distinguished by
the present data. However, it should be recognized that a definitive opinion on this must await
fuller explication of the concept of enmeshment and prospectively designed tests of the compet-
ing theories. Further developments using other methods, e.g. the Implicit Attitude Test (Greenwald
et al. 1998) might also be considered. Additionally, the extent to which enmeshment is specific to
chronic pain rather than a general characteristic of other chronic health conditions is unknown.
It is conceivable that chronic pain is a particularly good model for testing the concept of enmesh-
ment because of the extraordinary capacity of chronic pain to interrupt and interfere with ongo-
ing cognitive and behavioural activity (Banks and Kerns 1996; Eccleston and Crombez 1999;
Morley and Eccleston 2004).
16.4 Clinical implications

The following section discusses the possible implications of this line of research for clinical inter-
ventions. Several implications for the assessment and treatment of distress and depression in the
context of pain could be outlined, including the following: (1) chronic pain patients should be
routinely screened for depressed symptomatology; (2) it may prove useful to distinguish between
distress versus depression in chronic pain; and (3) the potential role of cognitive biases as a
marker for chronicity.
First, it seems reasonable to assume that the introduction of abovementioned research findings
as an educational tool would assist existing strategies of pain management. For example, if pain
patients are presented with results indicating the selective focus on negative health or negative
future experiences, they could employ this knowledge while attempting to focus on their positive
and valued future goals and plans. Presented with the fact that they automatically respond to self-
referent cues with pain- and health-related thoughts, such as illustrated in Figure 16.2c, they may
become more open to the idea that the process can be employed in reverse by focusing on self-
future-positive experiences despite experiencing pain (see Figure 16.2b). In a similar manner, the
SCP might be a useful assessment tool to capture the actual thought content and to initialize
therapeutic strategies for change. Although the results of the studies presented cannot be inter-
preted as evidence for the existence of a self-schema, they do show that pain patients selectively
process and generate information in reference to themselves; their view of themselves, or self-
image is directly connected to their processing and generation of pain- and health-related infor-
mation. The implications for clinical interventions are that it might prove beneficial to concentrate
on patients’ self-image instead of concentrating on the pain itself. This would involve attempts to
create new associative paths between the self-concept, such as more positive concepts of ‘myself
as a coping person’. Although many of the standard interventions may contribute to such a proc-
ess in the long term, there may be more direct routes to contribute to these changes.
Pincus (1998) further proposes that screening for information-processing biases might be a
quick and reliable way to assess whether a chronic pain patient needs to be referred on for further
psychological treatment. Referral might be warranted where an individual has demonstrated a
recall bias for negative health related stimuli, which have been encoded in relation to the self
(Pincus, 1998). Thus, information-processing methodologies might provide an effective way to
identify distressed chronic pain patients, and so improve outcomes, although this proposition is
in need of further research.
From a clinical point of view, it seems important to separate patients with chronic pain who are
distressed about the consequences of chronic pain from those who have the additional burden of
believing that the negative consequences of pain mean that they are fundamentally flawed and
worthless. It is well established that patients with chronic pain rarely endorse self-denigratory
statements (Williams and Richardson 1993; Morley and Wilkinson 1995). They are likely to
respond more rapidly to focused problem solving as their self-efficacy should be relatively intact,
and the more generalized cognitive distortions shown by depressed patients are limited to specific
pain-relevant situations in such patients (Smith et al. 1994). Such patients with pain may be
construed as being depressed about their pain rather than depressed about depression (Pincus
and Morley 2001). The body of literature reviewed in this chapter suggests that traditional
cognitive-behavioural interventions may not be optimal for depressed pain patients and that
their depression is qualitatively different from that of other groups, involving different concerns
and salient concepts. Psychological interventions that focus on these concepts, as many interven-
tions based on Beck’s cognitive therapy do, would prove less beneficial for depressed pain
patients than interventions that focus on retaining a positive sense of self, for instance. However,
cognitive-behavioural interventions have focused so far more on distressed pain patients (Turk
and Rudy 1992; Gatchel and Turk 1996), while interventions to specifically target depressed pain
patients seem to be underdeveloped. This leads to the question of whether there is a need for
psychological therapies such as cognitive-behavioural therapy (CBT) to be specially adapted for
depressed pain patients. At present, the demonstrated efficacy of interventions is, at best, modest
(Van Tulder et al. 2000). Additionally, Morley et al. (1999) have recently reviewed intervention
studies on the effectiveness of CBT in chronic pain and have concluded only modest effect sizes
for CBT in chronic pain compared to other conditions like mood and anxiety disorders, with
room for improvement. Similarly, researchers in the field have noted the importance of identify-
ing distinct subgroups of patients and to tailor interventions to the specific needs of patients in
order to improve treatment outcomes and satisfaction with treatment (Jellema et al. 2005; Turk
2005; Van der Windt et al. 2008; Vlaeyen and Morley 2005).
Despite the recent findings presented in this chapter, it is clear that, as yet, a comprehensive
picture of the function and content of depression in chronic pain is lacking, thus limiting the
potential for optimally targeted interventions. Overall the existing evidence so far, suggests that
there are qualitative differences between clinically depressed patients without pain and depressed
pain patients. As long as depressed pain patients are treated in the same way as clinically depressed
patients without the physical condition of pain, the treatment outcomes will reflect the non-
specificity of the treatment setting for this group of patients. To date, there is little research into
specific cognitions and beliefs, which should be targeted and changed through pain management
for this substantial group of pain patients. Pincus et al. (2002) reviewed evidence from prospec-
tive studies and concluded that current findings constitute a strong indication for the develop-
ment and testing of clinical interventions specifically targeting psychological distress, depressive
mood and somatization. More specifically, Leventhal et al. (1999) argue, in accordance with
Pincus and Morley (2001), that living with a chronic illness, such as severely disabling persistent
pain, has a powerful impact on self-identity: helplessness, despair, and frustration are associated
with a fusion of the self with the disease (‘I am pain’), whereas the ability to compartmentalize
pain and retain other aspects of the self are important for psychological well-being. However,
the main challenge of future research will be to address the question of how to compartmentalize
pain or, to express it in schema theory, how to disentangle the enmeshment of self, pain and ill-
ness schemas in depressed pain patients (see Figure 16.2c). It has been suggested in the previous
chapters that there are a number of plausible mechanisms for how self-pain enmeshment could
be reduced and further research is needed to test their clinical utility. In short, reducing self-
discrepancies between the actual self and ideal or possible self (e.g. Self-System Therapy; SST)
might lead to an improvement in depressed pain patients as well as therapeutic techniques which
emphasize the acceptance of pain. SST has recently been introduced as a new approach for
patients with unipolar depression (Lau 2008) and is based on regulatory focus theory, which
postulates that some depressed individuals are unable to pursue promotion goals effectively
(Higgins 1997). A randomized trial compared SST with cognitive therapy (CT), showing that
there was no overall difference in efficacy between treatments, but patients whose socialization
history lacked an emphasis on promotion goals showed significantly greater improvement with
SST (Strauman et al. 2006). There is currently one study underway, which is examining the effi-
cacy of self-system therapy in chronic pain patients (Kindermans et al. in preparation). Similarly,
accepting the pain as part of one’s present and future life seems to be beneficial for the adjustment
to pain and the reduction of depression (e.g. Acceptance and Commitment Therapy). For
instance, McCracken (1998) found that acceptance of pain was a significant predictor of several
physical and psychosocial variables including depression. The process of acceptance may require
abandonment of previous expectations, roles or goals, or their reassessment. Schmitz et al. (1996)
found less distress among those pain patients who modified existing goals or substituted new ones
(accommodation) than among those who pursed unmodified goals, relying on the eventual relief
of pain to enable them to realize them. Along similar lines, MacLeod and Moore (2000) recom-
mended a two-dimensional approach to mental health, which might be also appropriate for
chronic pain, by incorporating strategies to work on increasing positive as well as decreasing
negative states. Goal setting and planning ability are importantly linked to well-being by having
and progressing towards valued goals (Carver and Scheier 1990; Schmuck and Sheldon 2001;
Sheldon et al. 2002). Loewenstein et al. (2001) suggested that goal progress and goal attainment is
likely to affect well-being though anticipatory effects; people feel good at the moment when they
think about experiencing desirable future outcomes. It has been suggested that having meaning-
ful goals and plans to pursue those goals is also likely to result in higher levels of engagement in
life tasks and roles (Cantor and Sanderson 1999). MacLeod and colleagues (2008) demonstrated
in a recent study that a brief goal-setting and planning skills intervention in a non-clinical sample
significantly increased subjective well-being. Similarly, an alternative way of incorporating a
positive approach to depressed pain patients would be to work with them to develop identities
and interests that are not directly related to their pain problem. However, these kinds of interven-
tions will have to be tested for their pragmatic utility and effectiveness in samples of depressed
pain patients, characterized by low levels of well-being. Future work in this area will have to prove
which of the approaches outlined above will be beneficial under what circumstances for depressed
pain patients.
To conclude, depression has been identified as a determinant of poor rehabilitation outcomes
in individuals with musculoskeletal conditions. However, what remains unclear is why depres-
sion in chronic pain is associated with poor rehabilitation outcomes. One possibility is that
depressive symptoms might become more treatment resistant as the duration of pain-related
disability extends over time. The paucity of research addressing this question has limited concep-
tual advance in this field, and has probably impacted negatively on the development of effective
Box 16.1 Implications of cognitive processing for

clinical practice
In summary, the value of insights to cognitive biases for clinical practice could be described
along the following lines:
◆ Insights into the cognitive content and processing of depressed pain patients can pro-
mote the development of tailored pain management techniques that are directed at coun-
teracting dysfunctional response patterns to pain, and promoting functional and adaptive
processes.
◆ In providing evidence for the need to focus on self-concept, if long-term cognitive changes
are to be achieved.
◆ The identification of assessable psychological vulnerability factors can be helpful for the
early screening of individuals who are at risk for developing chronic pain complaints.
◆ The early identification of the presence of vulnerability for developing chronic pain can
promote the early application of pain management techniques that are specifically tailored
to the needs of the individual at risk.
◆ The assessment of automatic cognitive processing biases might provide a measure to assess
change in the salience of emotional concepts after intervention.
intervention programs for individuals with chronic pain and depressive symptomatology.
Accumulating evidence highlights the importance of early detection and treatment of depressive
symptoms in individuals with painful musculoskeletal conditions. Depression continues to be
under-detected and under-treated in individuals with musculoskeletal pain. Delays in referring
clients for assessment and treatment of depressive symptoms might mean missing a window of
opportunity for effective intervention on depressive symptoms. Surprisingly, there have been few
clinical trials examining the efficacy of pain interventions in depressed patients with musculoskel-
etal pain. Clinicians and researchers alike are encouraged to investigate more systematically the
efficacy of different treatment approaches for the management of depressive conditions associ-
ated with pain, particularly in more chronic cases where options for effective pain reduction are
more limited. In summary, the value of insights into cognitive processing to clinical practice are
described in Box 16.1.
16.5 Recommendations and future directions

Based on the examination of negative cognitive factors involved in the development and mainte-
nance of chronic pain, the following directions for future research may be proposed.
First, in examining pain-related distress and depression in relation to negative future thinking,
it might be valuable to consider the precise object of future-related anxieties and distress, and
its impact on self-identity more thoroughly. It is well acknowledged that distress and anxieties in
the context of pain can be directed at the presence of physical pain and/or the direct functional
consequences of activity or (re)injury, on the exacerbation and continuation of pain. In addition,
one might fear the possibility of putative future pain, disability, or physical health problems, the
impact of pain on one’s social status and relationships, or fears related to the impact of pain on
financial resources (McCracken 2004; McNeil and Vowles 2004; Morley and Eccleston 2004). The
content and meaning of pain-related depression and anxiety may be proposed to depend upon
prior personal learning history, prior knowledge, environmental contexts, and social influences.
For instance, Hadjistavropoulos et al. (2004) suggested that present-oriented fears and anxieties
focused on physical pain might be especially predictive of short-term disability and maladaptive
responses to pain, whereas future-oriented fears or anxieties, directed at putative physical harm,
future long-term disability and the functional and social consequences of injury and illness might
be more predictive of long-term disability.
A second direction for further research concerns the further examination of the role of cogni-
tive processing biases in the aetiology and persistence of chronic pain. One of the main limitations
of the research reviewed earlier concerns the cross-sectional design of the studies in examining
vulnerability factors for developing and maintaining chronic pain problems. Currently, it is
unclear whether processing biases result from long-term exposure to pain, or are vulnerability
indicators that result in maintenance of pain states. An alternative approach for examining cogni-
tive biases which may provide a more complete picture would be within a prospective study
design, which would collect longitudinal data on the development and exacerbation of chronic
pain problems. However, such a research design would be difficult to implement, as it requires
the selection of participants prior to the experience of pain. One alternative to this option would
be to follow up patients from casualty wards, in situations involving acute pain, to see which
patients develop chronic pain syndromes, and at what stage do processing biases become evident.
Furthermore, it might be suggested that studying the combined effects of multiple cognitive
processing biases can add significantly to our understanding of the unique and relative impor-
tance of these separate biases as well. Since individuals with chronic pain and associated high
levels of depression or anxiety are assumed to be characterized by the propensity to process pain-
related information negatively, it seems appropriate to assume that one particular bias will be
influenced by other biases or that several biases exert conjunct influences on the maintenance of
pain. It is also possible that the pain experience and the information processing biases operate in
a cyclical fashion, acting to maintain and strengthen each other (Hirsch et al. 2006). The relation-
ship between processing biases and the prognosis of pain conditions should therefore form the
focus of future work. An additional limitation of the studies reviewed concern power problems
due to the inclusion of relatively small participant samples. Meta-analyses have shown only small
to medium effect sizes for cognitive biases in chronic pain (Pincus and Morley 2001). It is there-
fore advisable that large enough participant groups are concordantly recruited to test for the
influence of these factors.
Another issue that remains to be investigated is the causal linkage between distress/depression,
cognitive biases, and the exacerbation of pain complaints. One way of studying causality would be
to induce cognitive biases for selective processing of pain- or health-related information in
healthy individuals and to study the subsequent effect on pain-related behavioural measures as
well as on mood. Evidence for the practicality of inducing cognitive biases successfully has recently
been provided with respect to automatic interpretation bias and attentional bias in anxiety disor-
ders (Mathews and MacLeod 2002; Salemink et al. 2007; Yiend and Mackintosh 2004). A first
study in student participants has shown that attentional biases towards acute experimentally
induced pain could be altered (McGowan et al. 2009). The findings that anxiety-related cognitive
biases and attentional biases towards pain can be induced leads to the challenge to examine the
extent to which these biases can be conversely reduced, and such techniques may serve as poten-
tial valuable additions to the treatment of persons characterized by increased negative cognitive
biases towards pain and health information. Future work should also concentrate on recovered
groups, as part of outcome studies.
Finally, in addition to studying factors that contribute in a negative fashion to pain and disabil-
ity, it is proposed that further research devotes attention to the role of psychological factors
that might have positive influences on the development and persistence of pain. For instance,
Vancleef (2007) demonstrated, in healthy individuals, the importance of perceived control and
self-efficacy beliefs as cognitive factors that have a beneficial influence on the experience of pain
by reducing pain intensity and fearfulness of pain. Other cognitive factors presumed to exert
positive influences on pain are acceptance (McCracken 2005) as well as several stable personality
traits, which have been suggested to have promising value in protecting against the development
of chronic pain conditions, for example dispositional optimism, general self-efficacy, and dispo-
sitional hope (Arntz and Schmidt 1989; Carver and Scheier 2005; Peters et al. 2007; Peters and
Vancleef 2008; Schwarzer et al. 2005). Accumulating knowledge on processes and factors that
might contribute to adjustment and adequate coping with pain provides useful information for
theoretical and management perspectives on chronic pain.
In conclusion, taking a broader perspective on the role and meaning of depression in pain and
studying the interplay between various constructs on an idiographic and nomothetic level, allows
for the identification of the unique and relative importance of each of these constructs for specific
components of the pain problem. It is suggested that the SEMP model might serve as a useful
framework within which findings of divergent influences can be understood and interpreted, and
further formal testing of this model is warranted.
Acknowledgements
The contribution of Adina C. Rusu and Tamar Pincus was supported by Back Care, United
Kingdom. Work on this chapter was also partially supported by a grant from the Research
Committee, University of Bochum, Germany, awarded to the first author.
References
Alloy, L. B., Abramson, L. Y., Murray, L. A., Whitehouse, W. G. & Hogan, M. E. (1997). Self-referent
information processing in individuals at high and low cognitive risk for depression. Cognition and
Emotion, 11, 539–68.
Arntz, A. & Schmidt, J. M. (1989). Perceived control and the experience of pain. In A. Steptoe & A. Appels
(Eds.), Stress, personal control, and health (pp. 131–62). Chichester: Wiley.
Asmundson, G. J. G., & Hadjivropoulos, H. D. (2007). Is high fear of pain associated with attentional
biases for pain-related or general threat? A categorical reanalysis. Journal of Pain, 8, 11–18.
Asmundson, G. J. G., Kuperos, J. L. & Norton, G. R. (1997). Do patients with chronic pain selectively
attend to pain-related information? Preliminary evidence for the mediating role of fear. Pain, 72, 27–32.
Asmundson, G. J. G., Carleton, N. R. & Ekong, J. (2005). Dot-probe evaluation of selective attentional
processing of pain cues in patients with chronic headaches. Pain, 114, 250–6.
Banks, S. M., & Kerns, R. D. (1996). Explaining high rates of depression in chronic pain: A diathesis stress
framework. Psychological Bulletin, 199, 95–110.
Barton, S. B. & Morley, S. (1999). Specificity of reference patterns in depressive thinking: agency and object
roles in self-representation. Journal of Abnormal Psychology, 108(4), 655–61.
Barton, S. B., Morley, S., Bloxham, G., Kitson, C. & Platts, S. (2005). Sentence completion test for
depression (SCD): An idiographic measure of depressive thinking. British Journal of Clinical Psychology,
44(1), 29–46.
Beck, A. T. (1970). The core problem in depression: The cognitive triad. In J. H. Masserman (Ed.),
Depression: Theories and therapies (pp. 47–55). New York: Grune and Statton.
Beck, A. T. (1976). Cognitive Therapy and the Emotional Disorders. New York: International
Universities Press.
Beck, A. T. (1987). Cognitive models of depression. Journal of Cognitive Psychotherapy, 1, 5–37.
Beck, A. T. & Clark, D. A. (1997). An information processing model of anxiety: Automatic and strategic
processes. Behaviour Research and Therapy, 35, 49–58.
Beck, A. T., Ward, C. H., Mendelson, M., Mock, J. E. & Erbaugh, J. K. (1961). An inventory for measuring
Beck, A. T., Laude, R. & Bohnert, M. (1974). Ideational components of anxiety neurosis. Archives of
General Psychiatry, 31, 319–25.
Beck, A. T., Rush, A. J., Shaw, B. F. & Emery, G. (1979). Cognitive Therapy for Depression. New York:
Guilford.
Beck, A. T., Brown, G., Steer, R. A., Eidelson, J. I. & Riskind, J. H. (1987). Differentiating anxiety and
depression: A test of the cognitive content-specificity hypothesis. Journal of Abnormal Psychology,
96, 179–83.
Boissevan, M. (1994). Information processing in chronic pain. Unpublished doctoral dissertation.
London, Ontario, Canada: University of Western Ontario.
Bradley, B., & Mathews, A. (1983). Negative self-schemata in clinical depression. British Journal of Clinical
Bury, M. (1988). Meaning at risk: The experience of arthritis. In R. Anderson & M. Bury (eds.), Living with
Chronic Illness: The Experience of Patients and their Families (pp. 89–116). London: Unwin Hyman.
Calfas, K. J., Ingram, R. E. & Kaplan, R. M. (1997). Information processing and affective distress in
osteoarthritis patients. Journal of Consulting and Clinical Psychology, 65, 576–81.
Cantor, N. & Sanderson, C. A. (1999). Life task participation and well-being: The importance of taking
part in daily life. In D. Kahneman, E. Diener, & N. Schwartz (Eds.), Well-being: The foundations of
hedonic psychology (pp. 230–43). New York: Russell Sage Foundation.
Carleton, N. R., Asmundson, G. J. G. & Taylor, S. (2005). Fear of physical harm: factor structure and
psychometric properties of the injury/illness sensitivity index. Journal of Psychopathology and
Behavioural Assessment, 27, 235–41.
Carver, C. S. & Scheier, M. F. (1990). Origins and functions of positive and negative affect: A control
process view. Psychological Review, 97, 19–35.
Carver, C. S. & Scheier, M. F. (1998). On the self-regulation of behavior. Cambridge, MA: Cambridge
University Press.
Carver, C. S. & Scheier, M. F. (1999). Themes and issues in self regulation. In R. S. Wyer, Jr. (ed.),
Perspectives on Behavioural Self-regulation, Vol. XII, pp. 1–105. Mahwah, NJ: Lawrence Erlbaum
Associates.
Carver, C. S. & Scheier, M. F. (2005). Optimism. In C. R. Synder & S. J. Lopez (Eds.), Handbook of positive
psychology (pp. 231–43). Oxford: University Press.
Carver, C. S., Lawrence, J. W. & Scheier, M. F. (1999). Self-discrepancies and affect: Incorporating the role
of feared selves. Personality and Social Psychology Bulletin, 25, 783–92.
Clemmey, P. A. & Nicassio, P. M. (1997). Illness self-schemas in depressed and nondepressed rheumatoid
arthritis patients. Journal of Behavioural Medicine, 20, 273–90.
Clyde, Z. K. & Williams, A. C. (2002). Depression and mood. In: S. J. Linton (Ed.), New avenues for the
prevention of chronic musculoskeletal pain and disability. Pain Research and Clinical Management,
Vol. 12 (pp. 105–21). Amsterdam: Elsevier.
Crombez, G., Eccleston, C., Baeyens, F. & Eelen, P. (1998a). Attentional disruption is enhanced by the
threat of pain. Behaviour Research and Therapy, 36, 195–204.
Crombez, G., Eccleston, C., Baeyens, F. & Eelen, P. (1998b). When somatic information threatens,
catastrophic thinking enhances attentional interference. Pain, 75, 187–98.
Crombez, G., Hermans, D. & Andriaensen, H. (2000). The emotional Stroop task and chronic pain: What
is threatening for chronic pain sufferers? European Journal of Pain, 4, 37–44.
Damasio, A. (1994). Descartes Error. New York: Avon Books.
Davies, C. (2003). Self-discrepancy Theory and Chronic Pain. Unpublished Dissertation in Clinical
Psychology, University of Leeds, Leeds.
Dehghani, M., Sharpe, L. & Nicholas, M. K. (2003). Selective attention to pain-related information in
chronic musculoskeletal pain patients. Pain, 105, 37–46.
Eccleston, C. & Crombez, G. (1999). Pain demands attention: a cognitive-affective model of the
Edwards, L. & Pearce, S. A. (1994). Word completion in chronic pain: Evidence for schematic
representation of pain? Journal of Abnormal Psychology, 103, 379–82.
Edwards, L., Pearce, S. A., Collett, B. J. & Pugh, R. (1992). Selective memory for sensory and affective
information in chronic pain and depression. British Journal of Clinical Psychology, 31, 239–48.
Eich, E., Rachman, S. & Lopatka, C. (1990). Affect, pain, and autobiographical memory. Journal of
Abnormal Psychology, 99 (2), 174–78.
Fishbain, D. A., Cutler, R., Rosomoff, H. L. & Rosomoff, R. S. (1997). Chronic pain-associated depression:
antecedent or consequence of chronic pain? A review. Clinical Journal of Pain, 13, 116–37.
Gatchel, R. J. & Turk, D. C. (1996). Psychological approaches to pain management: A practitioner’s handbook.
New York: Guilford Press.
Gray, E. (2006). Cognitive bias and positive affect in chronic pain. Unpublished Dissertation in Clinical
Psychology, Royal Holloway, University of London, London.
Greenwald, A. G., McGhee, D. E. & Schwartz, J. L. K. (1998). Measuring individual differences in implicit
cognition: the implicit association test. Journal of Personality and Social Psychology, 74, 1464–80.
Hadjistavropoulos, T. & Craig, K. D. (2004). An Introduction to Pain: Psychological Perspectives. In
T. Hadjistavropoulos & K. D. Craig (Eds.), Pain: psychological perspectives (pp. 1–12). New Jersey:
Lawrence Erlbaum Associates.
Hamilton, M. (1967). Development of a rating scale for primary depressive illness. British Journal of Social
and Clinical Psychology, 6, 278–96.
Harris, S., Morley, S. & Barton, S. B. (2003). Role loss and emotional adjustment in chronic pain. Pain,
105, 363–70.
Hellström, C., Jansson, B. & Carlsson, S. G. (2000). Perceived future in chronic pain: the relationship
between outlook on future and empirically derived psychological patient profiles. European Journal of
Pain, 4, 283–90.
Hellström, C. (2001). Temporal dimensions of the self-concept: entrapped and possible selves in chronic
pain. Psychology and Health, 16, 111–24.
Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect. Psychological Review, 94, 319–40.
Higgins, E. T. (1997). Beyond pleasure and pain. American Psychologist, 52, 1280–1300.
Higgins, E. T. & Tykocinski, O. (1992). Self-discrepancies and biographical memory: Personality
and cognition at the level of psychological situation. Personality and Social Psychology Bulletin,
18, 527–35.
Higgins, E. T., Roney, C. J. R., Crowe, E. & Hymes, C. (1994). Ideal versus ought predilections for
approach and avoidance distinct self-regulatory systems. Journal of Personality and Social Psychology,
66, 276–86.
Hirsch, C., Clark, D. M. & Mathews, A. (2006). Imagery and interpretation in social phobia: support for
the combined cognitive bias hypothesis. Behavior Therapy, 37, 223–36.
Hollon, S. D. & Kendall, P. C. (1980). Cognitive self-statements in depression: Development on an
automatic thoughts questionnaire. Cognitive Therapy and Research, 4, 383–95.
Ingram, R. E., Miranda, J. & Segal, Z. V. (1998). Cognitive Vulnerability to Depression. New York:
Guilford Press.
Jellema, P., van der Windt, D. A. W. M., van der Horst, H. E., Blankenstein, A., Bouter, L. M. &
Stalman, W. A. B. (2005). Why is a treatment aimed at psychosocial factors not effective in patients
with (sub)acute low back pain? Pain, 118, 350–59.
Johnson, R. & Spence, S. (1997). Pain, affect and cognition in children: recall bias associated with pain. In
G. F. Gebhart, D. L. Hammond & T. S. Jensen (Eds.), Progress in Pain Research and Management 2:
Proceedings of the 7th World Congress on Pain (pp. 877–84). Seattle: IASP Press.
Karoly, P. (1999). A goal systems-self-regulatory perspective on personality, psychopathology, and change.
Review of General Psychology, 3, 264–91.
Keogh, E. & Cochrane, M. (2002). Anxiety sensitivity, cognitive biases, and the experience of pain. Journal
of Pain, 3, 320–29.
Keogh, E., Dillon, C., Georgiou, G. & Hunt, C. (2001a). Selective attentional biases for physical threat in
physical anxiety sensitivity. Journal of Anxiety Disorders, 15, 299–315.
Keogh, E., Ellery, D., Hunt, C. & Hannent, I. (2001b). Selective attentional bias for pain-related stimuli
amongst pain fearful individuals. Pain, 91, 91–100.
Keogh, E., Hamid, R., Hamid, S. & Ellery, D. (2004). Investigating the effect of anxiety sensitivity, gender
and negative interpretative bias on the perception of chest pain. Pain, 111, 209–217.
Koster, E. H. W., Crombez, G., Verschuere, B., Van Damme, S. & Wiersema, J. R. (2006). Components of
attentional bias to threat in high trait anxiety: Facilitated engagement, impaired disengagement, and
attentional avoidance. Behaviour Research and Therapy, 44, 1757–71.
Koutanji, M., Pearce, S., Oakley, D. A. & Feinmann, C. (1999). Children in pain: an investigation of
selective memory for pain and psychological adjustment. Pain, 81, 237–44.
Lau, M. A. (2008). New developments in psychosocial interventions for adults with unipolar depression.
Current Opinion in Psychiatry, 21(1), 30–6.
Legrain, V., Van Damme, S., Eccleston, C., Davis, K. D., Seminowicz, D. A. & Crombez, G. (2009).
A neurocognitive model of attention to pain: Behavioral and neuroimaging evidence. Pain, 144, 230–2.
Leventhal, H., Idler, E. L. & Leventhal, E. A. (1999). The impact of chronic illness on the self system. In
R. J. Contrada & R. D. Ashmore (Eds.), Self, social identity, and physical health. Interdisciplinary
explorations (pp. 185–208). Oxford: Oxford University Press.
Loewenstein, G. F., Weber, E. U., Hsee, C. K. & Welch, N. (2001). Risk as feelings. Psychological Bulletin,
127, 267–86.
MacLeod, A. K. & Moore, R. (2000). Positive Thinking Revisited: Positive Cognitions, Well-being and
Mental Health. Clinical Psychology and Psychotherapy, 7, 1–10.
MacLeod, A. K., Coates, E. & Hetherton, J. (2008). Increasing well-being through teaching goal-setting and
planning skills: results of a brief intervention. Journal of Happiness Studies, 9, 185–96.
Markus, H. & Nurius, P. (1986). Possible selves. American Psychologist, 41, 954–69.
Mathews, A. & MacLeod, C. (1994). Cognitive approaches to emotion and emotional disorders. Annual
Review of Psychology, 45, 25–50.
Mathews, A. & MacLeod, C. (2002). Induced processing biases have causal effects on anxiety. Cognition and
Emotion, 16, 331–54.
McCracken, L. M. (1998). Learning to live with the pain: acceptance of pain predicts adjustment in persons
with chronic pain. Pain, 74, 21–27.
McCracken, L. M. (2004). A behavioural analysis of pain-related fear responses. In G. J. G. Asmundson,
J. W. S. Vlaeyen & G. Crombez (Eds.), Understanding and treating fear of pain (pp. 51–69). New York:
Oxford University Press.
McCracken, L. M. (2005). Contexual Cognitive-Behavioral Therapy for Chronic Pain. Seattle: IASP Press.
McCracken, L. M. & Turk, D. C. (2002). Behavioral and cognitive behavioral treatment for chronic pain:
outcome, predictors of outcome, and treatment process. Spine, 27, 2564–73.
McGowan, N., Sharpe, L., Refshauge, K. & Nicholas, M. K. (2009). The effect of attentional re-training and
threat expectancy in response to acute pain. Pain, 142, 101–107.
McKellar, J. D., Clark, M .E. & Shriner, J. (2003). The cognitive specifity of associative responses in patients
with chronic pain. British Journal of Clinical Psychology, 42, 27–39.
McNally, R. J. (1995). Automaticity and the anxiety disorders. Behaviour Research and Therapy, 33, 747–54.
McNeil, D. W. & Vowles, K. E. (2004). Assessment of fear and anxiety associated with pain:
Conceptualization, methods, and measures. In G. J. G. Asmundson, J. W. S. Vlaeyen & G. Crombez
(Eds.), Understanding and treating fear of pain (pp. 189–211). New York: Oxford University Press.
Morley, S. & Eccleston, C. (2004). The object of fear in pain. In G. J. Asmundson, J. Vlaeyen, &
G, Crombez (Eds.), Understanding and treating fear of pain (pp. 163–88). Oxford: Oxford
University Press.
Morley, S. & Wilkinson, L. (1995). The pain beliefs and perceptions inventory: A British replication. Pain,
61, 427–33.
Morley, S., Eccleston, C. & Williams, A. (1999). Systematic review and meta-analysis of randomized
controlled trails of cognitive behaviour therapy and behaviour therapy for chronic pain in adults,
Morley, S., Davies, C. & Barton, S. (2005). Possible selves in chronic pain: self-pain enmeshment,
adjustment and acceptance. Pain, 115, 84–94.
Oatley, K. (1992). Best Laid Schemes: The Psychology of Emotions. Cambridge: Cambridge University Press.
Pearce, J. & Morley, S. (1989). An experimental investigation of the construct validity of the McGill Pain
Questionnaire. Pain, 39, 115–21.
Pearce, S., Isherwood, S., Hrouda, D., Richardson, P. H., Erskine, A. & Skinner, J. (1990). Memory and
pain: test of mood congruity and state-dependent learning in experimentally induced and clinical pain.
Pain, 43, 187–93.
Peters, M. L. & Vancleef, L. M. G. (2008). The role of personality traits in pain perception and disability.
Reviews in Analgesia, 10, 11–22.
Peters, M. L., Vlaeyen, J. W. S. & Kunnen, A. M. W. (2002). Is pain-related fear a predictor of somatosensory
hypervilance in chronic low back pain patients? Behaviour Research and Therapy, 40, 85–103.
Peters, M. L., Sommer, M., de Rijke, J. M., et al. (2007). Somatic and psychological predictors of long-term
unfavourable outcome after surgical intervention. Annals of Surgery, 245, 487–94.
Pincus, T. (1998). Assessing psychological factors in chronic pain–a new approach. Physical Therapy
Review, 3, 1–5.
Pincus, T. & Morley, S. (2001). Cognitive-processing bias in chronic pain: a review and integration.
Pincus, T. & Morley, S. (2002). Cognitive appraisal. In S.J. Linton (Ed.), New Avenues for the Prevention
of Chronic Musculoskeletal Pain and Disability, Pain Research and Clinical Management, Vol. 12
(pp. 123–41). Amsterdam: Elsevier.
Pincus, T. & Newman, S. (2001). Recall bias, pain, depression and cost in back pain patients. British Journal
of Clinical Psychology, 40, 143–56.
Pincus, T. & Williams, A. C. deC. (1999). Models and measurements of depression in chronic pain. Journal
of Psychosomatic Research, 47, 211–19.
Pincus, T., Pearce, S., McClelland, A. & Turner-Stokes, L. (1993). Self-referential selective memory in pain
patients. British Journal of Clinical Psychology, 32, 365–75.
Pincus, T., Pearce, S., McClelland, A., Farley, S. & et al. (1994). Interpretation bias in responses to
ambiguous cues in pain patients. Journal of Psychosomatic Research, 38, 347–53.
Pincus, T., Pearce, S., McClelland, A. & Isenberg, D. (1995). Endorsement and memory bias of self-
referential pain stimuli in depressed pain patients. British Journal of Clinical Psychology, 34, 267–77.
Pincus, T., Pearce, S. & Perrott, A. (1996a). Pain patients’ bias in the interpretation of ambiguous
homophones. British Journal of Medical Psychology, 69, 259–66.
Pincus, T., Griffith, J., Pearce, S. et al. (1996b). Prevalence of self-reported depression in patients with
rheumatoid arthritis. British Journal of Rheumatology, 35, 879–83.
Pincus, T., Fraser, L. & Pearce, S. (1998). Do chronic pain patients ‘Stroop’ on pain stimuli? British Journal
of Clinical Psychology, 37, 49–58.
Pincus, T., Burton, A.K., Vogel, S. & Field, A.P. (2002). A systematic review of psychological factors as
predictors of chronicity/disability in prospective cohorts of low back pain. Spine, 27 (5), 109–20.
Pincus, T., Santos, R. & Morley, S. (2007). Depressed cognitions in chronic pain patients are focused on
health: Evidence from a sentence completion task. Pain, 130, 84–92.
Price, D. D. (1999). Psychological mechanisms of Pain and Analgesia, Progress in Pain Research and
Management, Vol. 15. Seattle: IASP Press.
Price, D. D. & Bushnell, M. C. (2004). Psychological methods of pain control: basic science and clinical
perspectives, Progress in Pain Research and Management, Vol. 29. Seattle: IASP Press.
Rachman, S., Lopatka, K. & Levitt, L. (1988). Experimental analysis of panic 2: Panic patients. Behaviour
Rainville, P., Carrier, B., Hofbauer, R. et al. (1999). Dissociation of sensory and affective dimensions of
pain using hypnotic modulation. Pain, 82 (2), 159–71.
Read, J. & Pincus, T. (2004). Cognitive bias in back pain patients attending osteopathy: testing the
enmeshment model in reference to future thinking. European Journal of Pain, 8, 525–31.
Roelofs, J., Peters, M. L., Zeegers, M. P.A. & Vlaeyen, J. W. S. (2002). The modified stroop paradigm
as a measure of selective attention towards pain-related stimuli among chronic pain patients: a
Roelofs, J., Peters, M. L. & Vlaeyen, J. W. S. (2003a). The modified Stroop paradigm as a measure of
selective attention towards pain-related information in patients with chronic low back pain.
Psychological Reports, 92, 707–715.
Roelofs, J., Peters, M. L., van der Zijden, M., Thielen, F. G. J. M. & Vlaeyen, J. W. S. (2003b). Selective
Attention and Avoidance of Pain-Related Stimuli: A Dot-Probe Evaluation in a Pain-Free Population.
Roelofs, J., Peters, M. L., Fassaert, T. & Vlaeyen, J. W. (2005). The role of fear of movement and injury in
selective attentional processing in patients with chronic low back pain: a dot-probe evaluation. Journal
of Pain, 6, 294–300.
Ross, M. & Buehler, R. (2004). Identity through time: constructing personal pasts and futures. In
M. B. Brewer, & M. Hewstone (eds.), Self and social identity (pp. 25–51). Oxford: Blackwell,
Rusu, A. C. (2008). Cognitive biases and future thinking in chronic pain. Unpublished doctoral thesis,
London, UK: University of London.
Rusu, A. C. Pincus, T. & Morley, S. J. (2008). Measuring depressive thinking in chronic pain: Development
and preliminary validation of the Sentence Completion Test for Chronic Pain (SCP). Abstracts of the
12th World Congress on Pain, International Association for the Study of Pain, Glasgow, U.K.: PH 317.
Rusu, A.C., Vogel, S., Van der Merwe, J., Pither, C. & Pincus, T. (2009). Testing the Schema Enmeshment
Model of Pain (SEMP): cognitive biases, depressed mood and future thinking in chronic pain.
European Journal of Pain, 13, Suppl.1, S 139.
Rusu, A. C., Barton, S., Morley, S. & Pincus, T. Sentence Completion Test for Chronic Pain (SCP):
Development of an idiographic measure of depressive thinking. Manuscript in preparation.
Salemink, E., van den Hout, M. A. & Kindt, M. (2007). Trained interpretative bias: Validity and effects on
anxiety. Journal of Behavior Therapy and Experimental Psychiatry, 38, 212–24.
Schmitz, U., Saile, H. & Nilges, P. (1996). Coping with chronic pain: flexible goal adjustment as an
interactive buffer against pain-related distress. Pain, 67, 41–51.
Schmuck, P. & Sheldon, K. M. (Eds.). (2001). Life goals and well-being. Towards a positive psychology of
human striving. Seattle: Hogrefe & Huber.
Schwarzer, R., Boehmer, S., Luszczynska, A., Mohamed, N. E. & Knoll, N. (2005). Dispositional
self-efficacy as a personal resource factor in coping after surgery. Personality and Individual Differences,
39, 807–18.
Sharpe, L., Dear, B. F. & Schrieber, L. (2009). Attentional biases on chronic pain associated with
rheumatoid arthritis: hypervigilance or difficulties disengaging? Journal of Pain, 10, 329–35.
Sheldon, K. M., Kasser, T., Smith, K. & Share, T. (2002). Personal goals and psychological growth:
Testing an intervention to enhance goal-attainment and personality integration. Journal of Personality,
70, 5–31.
Shiffrin, R. M. & Schneider, W. (1977). Controlled and automatic human information processing–II.
Perceptual learning, automatic attending, and a general theory. Psychological Review, 84, 127–90.
Smith, T. W., O’Keeffe, J. L. & Christensen, A. J. (1994). Cognitive distortion and depression in chronic
pain: Association with diagnosed disorders. Journal of Consulting and Clinical Psychology, 62, 195–98.
Snider, B. S., Asmundson, G. J. G. & Wiese, K. C. (2000). Automatic and strategic processing of threat cues
in patients with chronic pain: A modified-Stroop evaluation. Clinical Journal of Pain, 16, 144–54.
Stedman’s Medical Dictionary. (1976). 23rd edition. Baltimore: Williams & Wilkins.
Strauman, T. J., Vieth, A. Z., Merrill, K. A., et al. (2006). Self-system therapy as an intervention for self-
regulatory dysfunction in depression: a randomized comparison with cognitive therapy. Journal of
Consulting and Clinical Psychology, 74(2), 367–76.
Sullivan, M. J. L., Reesor, K., Mikail, S. & Fisher, R. (1992). The treatment of depression in chronic low
back pain: review and recommendations. Pain, 50, 5–13.
Sutherland, R. & Morley, S. (2008). Self-pain enmeshment: Future possible selves, sociotropy, autonomy
and adjustment to chronic pain. Pain, 137, 366–77.
Turk, D.C. (2005). The potential of treatment matching for subgroups of chronic pain patients: lumping
vs. splitting. Clinical Journal of Pain, 21, 44–55.
Turk, D. C. & Okifuji, A. (2002). Psychological factors in chronic pain: evolution and revolution. Journal of
Consulting and Clinical Psychology, 70, 678–90.
Turk, D. C. & Rudy, T. E. (1992). Cognitive factors and persistent pain: A glimpse into Pandora’s box.
Cognitive Therapy and Research, 16, 99–122.
Van Damme, S., Crombez, G. & Eccleston, C. (2002). Retarded disengagement from pain cues: The effects
of pain catastrophizing and pain expectancy. Pain, 100, 111–18.
Van Damme, S., Crombez, G. & Eccleston, C. (2004a). Disengagement from pain: the role of catastrophic
thinking about pain. Pain, 107, 70–6.
Van Damme, S., Crombez, G., Eccleston, C. & Roelofs, J. (2004b). The role of hypervigilance in the
experience of pain. In G.J.G. Asmundson, J.W.S. Vlaeyen & G. Crombez (Eds.), Understanding and
treating fear of pain. Oxford: Oxford University Press, p. 71–90.
Van Damme, S., Legrain,V., Vogt, J. & Crombez, G. (2010). Keeping pain in mind: A motivational account
of attention to pain. Neuroscience and Biobehavioral Reviews, 34 (2), 204–213.
Van der Windt, D., Hay, E., Jellema, P. & Main, C. (2008). Psychosocial Interventions for low back pain in
primary care: Lessons learned from recent trials. Spine, 33, 81–89.
Van Tulder, M. W., Ostelo, R., Vlaeyen, J. W., Linton, S. J., Morley, S. J. & Assendelft, W. J. (2000).
Behavioral treatment for chronic low back pain: a systematic review within the framework of the
Vancleef, L. (2007). At Risk for Pain: Pain-related Anxiety, Cognition, and Processing Biases. Maastricht:
Universitaire Pers Maastricht.
Vancleef, L., Peters, M. L. & De Jong, P. J. (2009). Interpreting ambiguous health and bodily threat: are
individual differences in pain-related vulnerability constructs associated with an on-line negative
interpretation bias? Journal of Behavioural Therapy and Experimental Psychiatry, 40, 59–69.
Vlaeyen, J. W. S. & Morley, S. (2005). Cognitive-behavioral treatments for chronic pain: what works for
whom? Clinical Journal of Pain, 21, 1–8.
Waddell, G. (1996). Low back pain: a 20th century health care enigma. Spine, 21, 2820–25.
Wade, J. B., Price, D. D., Hamer, R. M., Schwartz, S. M. & Hart, R. P. (1990). An emotional component
analysis of chronic pain. Pain, 40, 303–10.
Wells, H. J., Pincus, T. & McWilliams, E. (2003). Information processing biases among chronic pain patients
and ankylosing spondylitis patients: the impact of diagnosis. European Journal of Pain, 7, 105–11.
Williams, A. C. deC. & Richardson, J. M. (1993). What does the BDI measure in chronic pain? Pain, 55,
259–66.
Williams, J. M. G., Watts, F. N., MacLeod, C. & Mathews, A. (1997). Cognitive Psychology and Emotional
Disorders (2nd edn.). Chichester: Wiley.
Wright, J. & Morley, S. (1995). Autobiographical memory and chronic pain. British Journal of Clinical
Psychology, 34 (2), 255–65.
Yiend, J. & Mackintosh, B. (2004). The experimental modification of processing biases. In J. Yiend (Ed.),
Cognition, emotion and psychopathology: Theoretical, empirical and clinical directions (pp. 190–210).
New York: Cambridge University Press.
Chapter 17
Significant Others in the Chronicity

of Pain and Disability
Annmarie Cano and Laura Leong
Pain occurs within an interpersonal context. Indeed, interpersonal processes including marital
functioning are associated with physiological and immune processes that affect health and pain
(Kiecolt-Glaser & Newton 2001). Marital distress, as well as maladaptive interaction, is associated
with pain, interference, and depression in persons with chronic pain (Leonard et al. 2006). Family
members can also be affected by pain, in turn contributing to patients’ adjustment (Coyne &
Fiske 1992; Leonard et al. 2006). The purpose of this chapter is to examine the manner in which
the social context might contribute to the chronicity of pain. Unfortunately, this is a topic that has
received little attention in the literature. To address this topic, we provide an overview of the
theoretical and empirical literature that suggests pathways through which relationships with
significant others may foster the development of chronic pain from acute states. We also review
theoretical work that suggests other mechanisms through which relationships may play a role in
the development of chronic pain. We conclude by offering an integrative model and recommen-
dations for further research that might contribute to the knowledge on the extent to which close
relationships impact pain over time.
17.1 Support for theoretical conceptualizations of the role

of close relationships in pain
17.1.1 Operant model
The operant model of pain (Fordyce 1976) is best known as advancing the notion that significant
others including spouses have active roles in the experience of pain. Fordyce (1976) argued that
the reinforcement of pain behaviours and the extinction of well behaviours may explain why
acute pain behaviours from surgeries or injuries can persist over time. Reinforcement may also
explain the persistence of pain behaviours once such behaviours become chronic. Specifically,
spouses may reinforce pain behaviours by providing attention or help. Reinforcing or solicitous
pain behaviours such as these may lead to increases or at least the maintenance of chronic pain
behaviours including verbal expressions of pain, excessive rest, or ambulation problems. As a
result, solicitous behaviours may actually encourage pain behaviours, which can lead to greater
disability. Spouses may also punish pain behaviours by expressing anger or other forms of nega-
tive affect. Finally, spouses may ignore pain behaviours and reinforce well behaviours, both of
which would theoretically lead to an extinction of pain behaviours and an increase in activity.
Very little research has tested the operant model in couples facing acute pain. One study exam-
ined pain, depression, and spouse social support after a laparoscopic radical prostatectomy as a
treatment for prostate cancer (Knoll et al. 2007). Higher levels of pain as reported by patients at
two days post-surgery were associated with a drop in their spouses’ provision of instrumental
support from 2 days to 2 weeks following the surgery. This was only for pain in other body sites,
not at the site of the surgery. The authors suggest that spouses may have interpreted these pains
(e.g. headache, sore throat) as attention-seeking behaviours rather than genuine ailments, and
so they deliberately reduced their instrumental support as a way to prevent the patients from
adopting the sick role.
To date, the strongest evidence for the operant model is found in studies of chronic pain
couples, which have been conducted by Romano and colleagues (1991, 1992, 1995, 2000).
These studies are of particular value because they are observational studies, and thus, they are a
more rigorous test of the model. Couples in which one member had a chronic pain condition
were compared to healthy control couples (Romano et al. 1991, 1992, 1995). Participants were
videotaped in the laboratory doing a series of routine household activities: sweeping the floor,
changing bed sheets, bundling newspapers, and carrying fire logs across the room. They were
instructed to perform the tasks together, stressing that the pain patient must be involved at all
times. Trained raters coded the videos for six categories of behaviours: Non-verbal pain, verbal
pain, solicitous, facilitative, aggressive, and distressed. Non-verbal pain behaviours included
facial expressions, such as grimacing, and bodily actions such as limping or groaning; verbal pain
behaviours included statements that indicated pain or functional limitations; solicitous behav-
iours included offering or giving assistance, and statements that expressed concern for the other;
facilitative behaviours included compliments, encouragement, and humour; aggressive behav-
iours involved negative affect directed towards the other (e.g. disapproval, threats, arguments,
disagreement); and distressed behaviours included complaints not directed towards the spouse,
sadness, and whining. Preliminary analyses revealed that pain patients showed higher rates of
overt verbal and non-verbal pain behaviours, and their spouses showed higher rates of solicitous
behaviour, compared to control couples (Romano et al. 1991). In addition, pain patients and
their spouses both showed significantly lower rates of facilitative behaviours than the control
couples, though groups did not differ on the rate of aggression or distressed behaviours (i.e.
negative affect). In subsequent analyses, the authors found that solicitous spouse behaviours
preceded and followed both verbal and non-verbal pain behaviours more often in the pain
couples than in the control couples (Romano et al. 1992). When they examined additional pain
adjustment variables, they found that the sequence of spouse solicitousness in response to the
patient’s non-verbal pain behaviour was a significant predictor of physical dysfunction, but only
in more depressed patients (Romano et al. 1995).
Another study with similar methodology but using additional household chore tasks (folding
laundry, cleaning up toys, building a bookcase from bricks and boards, then loading books out of
boxes onto the bookcase) confirmed earlier findings: partner solicitous responses to the patient’s
pain behaviours were significantly positively associated with the rate of patient non-verbal pain
behaviours, while negative partner responses were significantly inversely associated with patient
pain non-verbal behaviour rates (Romano et al. 2000). Regression analyses revealed that the rate
of partner solicitous behaviours significantly contributed to the prediction of rates of both patient
verbal and non-verbal pain behaviours, but partner negative behaviours did not. Taken together,
the results of these studies suggest that spouse solicitous behaviours influence patients’ non-
verbal pain behaviours. It was noted that the rate of verbal pain behaviours in these studies were
low, so the results may have been dampened by restriction of range. Different kinds of interaction
tasks may be more appropriate for analysing verbal behaviours, such as a discussion. Another
limitation to these studies is that the sequential data were analysed using z scores, which are
affected by the number of behaviours that are observed. In other words, couples should not be
compared on z scores unless they exhibit exactly the same number of behaviours (Bakeman &
Gottman 1997).
SIGNIFICANT OTHERS IN THE CHRONICITY OF PAIN AND DISABILITY 341
Nevertheless, these observational studies are invaluable because they provide convincing evi-
dence that solicitous spouses can maintain pain and disability in chronic pain patients. The
results were also confirmed by Paulsen and Altmaier (1995) who found, using similar tasks which
only the patient completed, that patient- and spouse-reported spouse solicitousness were consist-
ently associated with higher levels of observed patient pain behaviours. Patients with solicitous
spouses engaged in more pain behaviours than patients with non-solicitous spouses, regardless of
whether the spouse was present or not. A study by Lousberg and colleagues (1992) also sought to
examine this relationship using a different behavioural task. Chronic low back pain patients
engaged in a walking-to-tolerance treadmill test during which they walked on a treadmill that
became gradually steeper. They were instructed to continue until they had to stop due to pain or
fatigue. Each participant performed the test twice: once in the presence of his/her spouse, and
once without the spouse. Pain behaviours were measured as pain intensity on a visual analogue
scale and the total time spent on the treadmill, while spouse solicitousness was reported by both
patients and spouses on a self-report measure. Patients who had solicitous spouses (as rated by
the spouse, after a median split) reported increased pain intensity, and they spent less time on the
treadmill when their spouse was present, compared to when they were alone. Patients with non-
solicitous spouses showed the opposite effect; they spent more time on the treadmill and reported
less pain. The authors interpret this as an increase in pain behaviours in the presence of a solici-
tous spouse. These results, however, were not found when they used the patients’ ratings of
spouse solicitousness: regardless of the patient’s rating of spouse solicitousness, the presence of
the spouse during the walking test resulted in more pain and shorter walking time. It is possible
that patients and spouses had widely different perceptions of the spouse’s solicitousness with
varying degrees of accuracy, which caused these discrepant results.
There are some limitations to how these observational studies can be interpreted. Most impor-
tantly, the data are cross-sectional and cannot be interpreted causally. Also, the patients who
participated already had chronic pain conditions, so it is not certain that the observed behaviours
were related to the progression of an acute condition to a chronic one. Finally, observational stud-
ies do not account for how patients interpret spousal behaviours. Nevertheless, the evidence
generally supports the operant model’s hypothesis that solicitous spouse behaviours should be
detrimental to patients’ pain adjustment.
17.1.2 Cognitive-behavioural models

The next generation of models that includes a role for spouses are the cognitive-behavioural
models of pain (Sharp 2001; Turk & Kerns 1985; Turk et al. 1983). These models have in common
the idea that perceptions and thoughts, including those focused on spouse behaviour, play an
important role in pain adjustment. For instance, a spouse may respond solicitously to a patient’s
pain behaviour. However, the spouse’s response may not reinforce the pain behaviour if the
patient interprets that behaviour as fake or hostile. Similarly, negative spouse responses might
not punish pain behaviour if patients interpret the behaviour as a sign of the spouse’s concern for
the patient’s health.
One cognitive-behavioural model, the Transactional Model of Health (Turk & Kerns 1985),
was developed for health and illness in families more broadly and is easily applied to chronic pain
conditions and spouses. It builds on the work of Turk et al. (1983) by expanding its treatment of
coping with illness and by focusing on the active interactions between spouses and between the
partners and their environment. Couples are said to develop schemata about the world and about
themselves. These, along with their resources, influence how they respond to health threats:
situations are deemed stressful based on the couples’ appraisals and available resources. Coping
is seen as the complement to stress; it is an active process that involves appraising events and
determining what will happen in the future. Spouses’ coping efforts can benefit each partner, or
they can further deteriorate the situation. For example, in pain patients, a solicitous spouse may
ameliorate the situation by providing emotional support, or they may reinforce pain behaviours
and prolong recovery.
Though the cognitive-behavioural models purport to account for unique aspects of the pain
experience, Sharp (2001) has raised some criticisms. He argues that the so-called cognitive-
behavioural theories have failed to successfully incorporate cognitions, and that they are still
heavily based on operant theory. As a result, he proposed a reformulated cognitive-behavioural
model of pain. Sharp asserts that patients who are distressed react to their pain differently than
those who are not distressed. In his model, these reactions include all cognitions as well as observ-
able behaviours. The main tenet of this model, as it is for the other cognitive-behavioural models
of pain, is that patients’ appraisals and interpretations of their pain are fundamental to their
distress. Many factors contribute to their appraisals, including sensations, learning history, cul-
ture, physiological responses, mood, motor behaviours, iatrogenic causes, and environmental
contingencies. Environmental contingencies include responses of family members and significant
others. Sharp explains that these responses can provide evidence for or against patients’ beliefs.
For example, a solicitous spouse may confirm a patient’s belief that he/she is disabled.
Other cognitive models, though not explicitly including significant others, suggest that
environmental factors more generally play an important role in pain adjustment. For example,
fear-avoidance models feature the notion that people with pain have a fear that physical activity
will cause (re)injury (Severeijns et al. 2004; Vlaeyen & Crombez 1999). There are two potential
responses to this fear: confrontation (continuing with motor activities despite the pain, an adap-
tive response), or avoidance (maintaining fear, leading to exacerbated physical and psychological
symptoms). The model starts with a patient experiencing pain (possibly as the result of an injury).
Some patients do not experience a fear of (re)injury, so they confront the pain and recover. Others,
however, enter a detrimental cycle: they catastrophize about the pain, become fearful, and avoid
activities. This avoidance results in increased disability and depression over time (e.g. via loss of
muscle strength and self-esteem). These have been shown to decrease pain tolerance, resulting in
increased pain experiences, and the cycle restarts. One could imagine that fears expressed by
spouses might contribute to patients’ fears and contribute to their avoidance behaviours.
As with the operant model, very little research has tested the cognitive-behavioural perspective
in couples with acute pain. However, one study on postoperative pain and speed of recovery after
coronary bypass surgery may provide some clues (Kulik & Mahler 1989). Men who had bypass
surgery were divided into two groups based on their perceptions of marital quality: those who
rated their marital quality as excellent and those who rated quality as less than excellent. Hospital
staff also counted the number of times they were visited by their wives, and divided this by the
total number of days hospitalized to obtain a measure of support. Men with high levels of support
took significantly fewer pain medications as recorded in their charts and spent less time both in
the surgical intensive care unit and in the hospital overall whereas perceived marital quality was
an insignificant factor. It is possible that much of the support was emotional and that it was per-
ceived by husbands as caring and encouraging behaviour. This interpretation would be consistent
with cognitive-behavioural models of pain, in which spouse encouragement helps to foster the
patient’s own coping skills and autonomy. Follow-up data would be extremely useful for deter-
mining whether these relationship between perceived support and pain hold over time.
Much of the work on perceptions of spousal behaviour has focused on solicitous and punishing
responses in chronic pain couples. Patient perceptions of spousal solicitousness (i.e. getting the
patient medication or something to eat or drink when in pain) are related to poorer pain adjust-
ment including increased pain severity, physical disability, and depression (for a review,
see Leonard et al. 2006). However, patients’ reports of spouses’ punishing responses (e.g. anger,
irritability) to patient pain behaviours are related to increased pain severity and depressive symp-
toms (Cano et al. 2000, 2004a; Kerns et al. 1990; Stroud et al. 2006; Turk et al. 1992; Williamson
et al. 1997). This latter finding is not entirely consistent with the operant model, which suggests
that punishing responses would contribute to a decline in pain behaviours. However, these
responses may represent a loss of reinforcement of positive interaction, thereby contributing to
depression.
More recent research has focused on patients’ perceptions of spousal reinforcement of well
behaviours. Facilitative responses to well behaviours (e.g. encouraging activity and exercise) are
negatively related to physical disability whereas negative responses to well behaviour (e.g. dis-
couraging activity) are positively related to pain behaviours and physical disability in chronic
musculoskeletal pain and headache patient samples (Pence et al. 2008; Schwartz et al. 2005). The
interrelationships between different behaviours were also quite interesting. Facilitative responses
to well behaviour were positively associated with solicitous responses to pain behaviour in both
studies. Facilitative responses to well behaviour and solicitous responses were also negatively
related to punishing spouse responses to pain. Thus, spouses who respond in a supportive man-
ner to both types of behaviours may not respond in a negative manner to pain behaviours.
However, punishing spouse responses to pain behaviours and negative responses to well behav-
iours were not significantly correlated. It is possible that the greater differentiation is between
spouses who respond negatively to pain behaviours versus spouses who respond negatively to
well behaviours. The former may either feel helpless about being able to aid their partners or
may be irritated about how the relationship has changed due to pain. In contrast, spouses who
respond negatively to well behaviours may be fearful that their partners will reinjure themselves.
These studies provide the first evidence that spousal responses to pain and well behaviours must
be taken into account when determining the manner in which spouses may influence patient
outcomes.
In sum, research has demonstrated that perceptions of spouse responses appear to be related to
patients’ pain adjustment. However, the extent to which this research can be applied to the transi-
tion from acute to chronic pain is limited. Like the observational work, much of this research is
cross-sectional in nature and has been conducted with persons experiencing chronic pain. Thus
the findings are correlational, not causal, and for now, only apply to persons experiencing chronic
pain. Furthermore, while patients’ interpretations and attributions of spouse responses are argued
to be crucial to the reinforcement process, very little research has examined these attributions.
17.1.3 Communal coping model

Other models of pain have begun to integrate interpersonal and cognitive-behavioural perspec-
tives. The communal coping model of pain catastrophizing (Sullivan et al. 2001; Thorn et al.
2003) is one such model. Pain catastrophizing is a negative cognitive process that consists of
rumination, magnification, and helplessness about pain (Sullivan et al. 1995). Patients may
engage in catastrophizing to communicate their distress about pain to close others who may be
able to help. Catastrophizing is thought to result in behaviours aimed at communicating distress
and eliciting support. This model has been tested in experimental paradigms of acute pain.
Sullivan and colleagues found that observers’ ratings of participants’ pain severity during a cold
pressor task was explained by the pain behaviour exhibited by participants (Sullivan et al. 2006).
In another cold pressor task study, greater catastrophizing was associated with longer displays of
communicative pain behaviours such as facial expressions of pain when someone was present
during the task than when the participant was alone in the room (Sullivan et al. 2004). A study of
chronic pain patients and their spouses yielded somewhat different results regarding the
communication of pain. In this study, pain behaviours did not enhance the spouse’s ability to
estimate the pain experienced by the patient partner (Gauthier et al. 2008). Rather, pain adjust-
ment variables were correlated with the spouses’ ability to infer patients’ pain. It appears that
spouses attend to patient characteristics to estimate pain but not to behaviours expressed during
the task. This collection of studies shows that observers may use pain behaviours to interpret pain
during the acute phase but not during the chronic phase. However, the acute pain studies inves-
tigated strangers whereas the chronic pain study investigated close others. Therefore, it is not
clear if the differences among studies are due to the relationship of the participants and observers,
differences between acute and chronic pain, or both.
Indirect support for the communal coping model also comes from work conducted on cata-
strophizing and social support. Pain catastrophizing is positively associated with solicitous
responses from significant others in samples of patients with spinal cord injuries and chronic
musculoskeletal pain (Cano 2004; Giardino et al. 2003). Pain duration also matters in the
association between catastrophizing and support. In a study of persons with chronic pain,
catastrophizing was associated with greater solicitous spouse responses at shorter pain durations
(Cano 2004). In contrast, catastrophizing was associated with less social support from the
spouse at longer pain durations. Taken together, these findings suggest that spouses provide
pain-specific support in response to patient catastrophizing; however, catastrophizing over the
long-term may cause partners to withdraw from patients, resulting in a loss of intimacy in cou-
ples. Buenaver et al. (2007) found similar results. Although shorter pain durations in these studies
are still considered chronic, it is possible that the effect of catastrophizing on pain-specific sup-
port occurs in acute pain episodes and contributes to changes in the relationship as pain persists
over time.
17.1.4 Empathy models

Empathy is an interpersonal concept that has received increasing amounts of attention in the pain
literature. Goubert et al. (2005) argued that the pain empathy process involves characteristics of
patients and observers. The observer’s characteristics, including personal experience with pain or
catastrophizing about the observed person’s pain, may contribute to the observer’s understand-
ing or ‘sense of knowing’ about the pain. Patient characteristics including facial expressions or
verbalizations of pain may also contribute to the observer’s knowledge. In turn, the observer may
engage in a variety of behavioural and emotional responses. For example, observers may validate
the patient’s experience of pain, withdraw from the patient, feel personal distress about witness-
ing pain, or feel distress or sympathy for the patient. This model was developed to account for
empathy in acute as well as chronic pain.
Support for this model of pain empathy comes from a study of chronic pain couples. Leonard
and Cano (2006) found that one observer characteristic, spousal catastrophizing about the part-
ner’s pain, was related to greater spousal distress for those spouses who also had personal experi-
ence with pain (i.e. reported chronic pain). This association was not found for spouses without
chronic pain. Patient depression also appears to vary depending on spousal catastrophizing.
Depressive symptoms in patients are most elevated when both partners have high levels of cata-
strophizing about the patient’s pain (Cano et al. 2005). Thus, spousal characteristics such as cata-
strophizing may promote emotional distress in both partners, perhaps because such cognitions
heighten one’s understanding of the other’s pain. In turn, awareness could affect a couple’s inter-
actions and distress. It is possible that spousal catastrophizing promotes distress in acute pain,
which could lead to decreases in activity that contribute to functional impairment and disability
over time.
Some work has also investigated empathic responses toward patients with pain. Newton-John
and Williams (2006) demonstrated that some solicitous spouse responses are interpreted as hos-
tile responses. Furthermore, these researchers have argued that talking about the pain experience
may be helpful for some patients. These ideas are consistent with the pain empathy model
(Goubert et al. 2005). Perhaps talking about pain-related distress is a form of emotional disclo-
sure that provides spouses with the information they need to understand the pain. In turn, the
spouse may be able to engage in empathic and validating responses to the patient. These ideas are
also consistent with emotion regulation models of interaction (Fruzzetti & Iverson 2004, 2006)
and intimacy research (Laurenceau et al. 1998); however, these conceptualizations of emotional
disclosure and empathic response are not entirely consistent with the operant model’s conceptu-
alization of verbal pain behaviour and solicitous responses.
Preliminary research on empathic responses suggests that such responses are indeed conceptu-
ally distinct from solicitous responses. Cano et al. (2008) conducted a study in which couples
attended a lab session and were videotaped while discussing the impact of pain on their lives.
Validating responses were responses that conveyed one spouse’s attempts to understand the
thoughts and feelings of the other. Invalidating responses were responses that communicated a
general disregard for the feelings of the other and included behaviours such as eye rolls and chang-
ing the subject. Marital quality was negatively correlated with invalidation and positively corre-
lated with validation behaviours. An exploratory factor analysis was also conducted to understand
the intercorrlations among spousal validating and invalidating responses, and reports of spousal
solicitous, distracting, and punishing responses. A two-factor solution that accounted for 57% of
the variance in the data was found in which both partners’ reports of solicitous and distracting
spouse responses loaded on a ‘Solicitous Responding’ factor. Couples’ reports of punishing spouse
responses as well as observed spousal invalidation and validation loaded on an ‘Empathic
Responding’ factor. The factors were weakly related to one another (r=−0.17). Both factors cor-
related with marital satisfaction and spousal support in both partners but the Empathic Responding
factor was more strongly correlated with these measures of marital quality. These findings suggest
that validation and solicitousness are related but distinct concepts. Validation may work to enhance
intimacy and emotion regulation in couples whereas solicitous responses may have a rather pain-
specific function. This study points to the importance of using more than one theoretical approach
to understand the function of interaction in couples facing pain. It is likely that similar processes
are present in couples with acute pain but future research will need to address this issue.
While not specifically examining empathic responses, Johansen and Cano (2007) investigated
the role of hostility, one form of invalidation, and sadness in couples’ interactions. Interactions
characterized by hostility or disregard for partners’ emotional responses are thought to disrupt
emotion regulation attempts (Fruzzetti & Iverson 2004, 2006). Couples engaged in a 15-minute
marital problem-solving task. Approximately half of the couples displayed anger/contempt, a
form of emotional invalidation, and sadness whereas humour was normally distributed. The
association between negative interaction codes and depressive symptoms depended on whether
one or both spouses reported chronic pain. Specifically, patient sadness was associated with
greater pain and depression when only he or she reported pain. In contrast, patient sadness was
associated with less pain and depression when patients’ spouses also reported pain. Perhaps,
expressing sadness in the context of an empathic spouse may help patients in their attempts to
regulate their emotions. Spouses’ expressions of anger/contempt were associated with their own
depressive symptoms only when they did not report pain and their partners (i.e. the patients) also
expressed anger/contempt. In these couples, it is possible that the absence of pain in the spouse
interferes with the couple’s ability to talk about pain in an accepting environment. In any event,
these results suggest that empathic processes depend on the partner’s personal experience with
pain, which supports the importance of spouses’ personal characteristics in the empathy process
(Goubert et al. 2005).
In sum, research has borne out that spouses play an important role in pain. Spouses’ behav-
ioural responses to patients as well as patients’ perceptions of these responses may contribute to
pain adjustment and psychological well-being. Furthermore, patients’ and spouses’ thoughts
about the pain, including pain catastrophizing, may influence social support over time. Much of
the research has been conducted on chronic pain and none of the research to date has examined
acute clinical pain in a couples context. These gaps in the research offer exciting new opportuni-
ties for researchers interested in the social factors involved in the progression of acute to chronic
pain over time.
17.2 Treatment research on couples and spouses

We now turn our discussion to pain interventions that involve spouses and couples. Although
there are no chronic pain prevention programmes of which we are aware that involve acute pain
patients and spouses, the effectiveness of couple-based treatments for chronic pain implies that
spouses can help alter the course of pain.
17.2.1 Spouse involvement in education and coping skills training

Keefe and colleagues have developed a programme called spouse-assisted coping skills training
(S-CST), which is a cognitive-behavioural treatment for chronic pain that actively involves
spouses (Keefe et al. 1996, 1999, 2004). S-CST is typically led by a clinician in a group setting with
10–12 2-hour weekly sessions. Clinicians provide education about pain and teach couples various
techniques and use behavioural rehearsal (i.e. practise) so that couples can master skills. Skills
include communication training and coping skills such as relaxation, imagery, and distraction
techniques. In addition, spouses are encouraged to provide feedback to each other and to practise
skills between sessions during activities that might elicit pain.
A series of studies have examined the effectiveness of S-CST in pain management. S-CST has
been compared to coping skills training without spouse involvement (CST) and a control group
that involved education about arthritis and spousal support in a sample of chronic osteoarthritic
knee patients and their spouses (Keefe et al. 1996, 1999). Results showed few differences between
the two CST groups. Results suggested that the S-CST group performed better than the CST
group on self-reported outcome measures at post-treatment and follow-up (i.e. pain severity,
psychological disability, pain behaviour, marital adjustment, coping). However, the CST groups
did not differ significantly on these measures. Both CST groups experienced better outcomes than
the control group.
Additional analyses suggested that differences are more apparent when examining how change
occurs during treatment. Specifically, initial improvements in marital satisfaction were associated
with to better adjustment at 12-month follow-up in the S-CST group (Keefe et al. 1999). This was
not the case in the other two groups, where initial improvement in marital satisfaction was
actually related to poor pain adjustment. Participants in S-CST with or without an exercise com-
ponent reported greater self-efficacy and coping skills compared to participants undergoing
exercise training alone or standard care (Keefe et al. 2004). In this study, however, no significant
group differences were found on marital satisfaction, pain, or psychological disability. It appears
that the S-CST approach may lead to some improvements in marital quality and coping that are
beneficial for patient functioning in chronic pain.
Martire et al. (2007) compared an education plus support intervention that involved couples
with patient-oriented education plus support involving patients only, and usual care in an older
adult sample with osteoarthritis. Participants in the education and support interventions attended
six weekly group sessions that last 2 hours each. The education and support conditions were
similar to S-CST in that the sessions included arthritis education and the teaching of effective
coping skills. Participants were encouraged to rely on each other for support. The couple inter-
vention also presented this information from a couple’s perspective rather than an individual
perspective. Like the S-CST research, there were few differences among the three groups from
baseline to 6-month follow-up. However, participants in the patient-oriented group experienced
greater improvements in pain adjustment than participants in the couple intervention whereas
the couple-oriented group experienced greater improvements in psychological outcomes includ-
ing perceived stress and critical attitudes. Patients’ spouses also experienced some improvements
depending on their gender and relationship satisfaction. Specifically, in the couple-oriented
group, maritally satisfied spouses experienced decreases in depressive symptoms whereas mari-
tally dissatisfied spouses reported increases in depression. In addition, patients’ wives reported
lower stress over time if they participated in the couple-oriented group. These results suggest that
the couples approach is more appropriate for psychologically distressed patients who are married
to relatively satisfied spouses. An individual approach may be more appropriate for patients
experiencing poor pain adjustment.
S-CST has also been adapted as a brief intervention for pain in individuals with advanced can-
cer diagnoses (Keefe et al. 2005). In this study, couples participated in three home-based sessions
over the course of approximately 2 weeks. The sessions included education about cancer pain
management and pain coping skills (e.g. relaxation training, imagery). Instructors taught couples
to pace activity. Behavioural rehearsal techniques were used to train couples on how to maintain
skills over time. This intervention was compared against a usual care condition. Similar to tradi-
tional S-CST for knee pain, there were no group differences in patients’ pain severity. However,
partners in the partner-guided condition reported greater self-efficacy in assisting their partners
with pain and other symptoms. In addition, partners reported slightly less caregiver burden in the
partner-guided group than in the usual care group.
In sum, these studies demonstrate that the involvement of significant others may have benefi-
cial psychological and relational effects on persons with pain and their partners although effects
on pain appear to be limited. Most of these studies have been conducted in chronic pain samples;
however, these findings suggest that education and couples-based coping skills training during
the acute phase of pain may help prevent distress that could contribute to the persistence of pain
over time.
17.2.2 Couple therapy

Insight-oriented therapy has also been tested as a treatment for chronic pain. Insight-oriented
therapy explores historical and current relationship processes rather than focusing on pain
coping skills. Saarijarvi (1991) compared couples who attended five monthly sessions of insight-
oriented couple therapy with couples in a no-treatment control group. Marital satisfaction in
both groups declined over a 12-month follow-up but the decline was significantly smaller in the
insight couple therapy group. The couple therapy group reported that their communication
improved whereas the control group reported communication declines. However, there were no
significant group differences on pain or disability at the 12-month follow-up (Saarijarvi et al.
1991). At 5-year follow-up, the therapy group reported significantly better psychological health
than the control group but the groups did not differ significantly on marital satisfaction, pain, or
disability (Saarijarvi et al. 1992). These results suggest that insight-oriented therapy may be
appropriate to treat psychological distress and relationship satisfaction but it does not appear to
relate to improvement in pain adjustment. Applied to the problem of acute pain, such treatment
may promote relationship and psychological health, which might indirectly relate to pain adjust-
ment over time.
Integrative behavioural couple therapy (IBCT) may also have beneficial effects for couples with
chronic pain (Cano & Leonard 2006). IBCT therapists use behavioural strategies often used in
traditional behavioural couple therapy (e.g. behaviour exchange, communication training). In
addition, therapists promote emotional acceptance, which involves changes in how partners per-
ceive undesirable partner behaviours (Christensen et al. 1995; Jacobson & Christensen 1998). For
example, personality or behaviour patterns (e.g. need for spontaneity, need for order) that have
become irritating to the partner may be reinterpreted in a positive light.
IBCT results in improvements in psychological and marital distress (Christensen et al. 2004,
2006; Jacobson et al. 2000). It has been argued that IBCT is appropriate for chronic pain (Cano &
Leonard 2006). First, IBCT addresses the psychological distress that is reported by couples with
pain (Leonard et al. 2006). Second, such therapy might address spouses’ difficulties in under-
standing the pain of their partners. Spouses under- and overestimate the pain and disability
experienced by their partners (Cano et al. 2005, 2004b) and patients often think their partners do
not understand their pain and emotional suffering (Herbette & Rime 2004). IBCT could be used
to promote empathic understanding of each partner’s experience of the pain and its impact on
the relationship. Greater empathic understanding and improved communication skills might
also alleviate patients’ fear of rejection or lack of self-efficacy in discussing the pain (Morley et al.
2000; Porter et al. 2008). IBCT has not been empirically tested in acute or chronic pain couples.
However, this treatment offers promising avenues through which to improve relationship func-
tioning in pain patients and their partners.
In sum, cognitive-behavioural approaches have dominated the research on couples-based
treatments for pain. The most effective interventions focus on building communication and pain
coping skills and result in improvements in marital and psychological well-being. However,
much is unknown about what is actually being changed in these interventions. For instance, it is
assumed that spousal beliefs about pain are targeted during treatment but these beliefs are often
not assessed. It will be important for researchers to continue this work by measuring spousal pain
beliefs and investigating change in such beliefs (Cano et al. 2009). Furthermore, existing couple-
based treatments, whether they focus on coping skills or insight, have little direct effect on pain
severity and disability and do not consistently outperform similar treatments that focus solely on
patients. It is possible that couples-based treatments should be considered an adjunct to other
forms of pain management that more directly impact pain and disability. Alternatively, the
mechanisms through which these treatments work may need to be further investigated. Perhaps,
couple-based treatment is more important during the acute phase of pain in order to promote
healthy communication and pain coping skills in both partners, set realistic expectations regard-
ing pain, and promote healthy emotion regulation, all of which might prevent pain from adversely
impacting the relationship and resulting in depression and pain behaviour.
17.3 Recommendations and future directions

Operant and cognitive-behavioural theories have received quite a bit of attention in the research
on the social context of pain. Indeed, the empirical literature has demonstrated that spouse
responses to pain behaviours as well as patients’ perceptions of those behaviours are related to
pain adjustment. Other cognitions such as patient and spouse pain catastrophizing also appear to
have a role in patients’ pain adjustment. Existing treatments focusing on cognition and behaviour
appear to be somewhat beneficial in treating the psychological and emotional consequences of
pain. Furthermore, recent research drawing from interpersonal and empathy models suggests
Patient internal process Patient behaviours
Cognitions: e.g. interpretation of Pain behaviours: facial expressions,

spouse behavior, pain catastrophizing, gait, verbalizations
sense of feeling understood Well behaviours: activity, exercise
Motivation: e.g. need for support Emotional disclosure of pain distress
Emotions: e.g. sadness, satisfaction
Partner internal process Partner behaviours
Cognitions: e.g. attributions, pain

Punishing, solicitous, distracting
catastrophizing, cognitive empathy
responses of pain and well behaviours
Motivation: e.g. need for healthy spouse,
Empathic and unempathic responses to
need to support
emotional disclosures
Emotions: e.g. helplessness, sorrow for
partner
Fig. 17.1 Pain-related interaction process that might explain the progression of acute to chronic
pain in couples.
that it is necessary to go beyond a conceptualization of spouse responses that focuses solely on an

operant formulation. Specifically, empathic spouse responses that build intimacy and under-
standing may also enhance adjustment to pain.
Following existing research and theory, we developed a model of the couple interaction process
that may account for the progression of acute to chronic pain (see Figure 17.1). A key feature of the
interaction process is that each partner’s behaviours trigger intrapersonal processes aimed at
understanding partner behaviour. In turn, these intrapersonal processes translate into behaviours.
For instance, a patient may feel helpless about their pain and is motivated to seek support from his
spouse. These cognitions and motivations may lead the patient to engage in more facial expressions
of pain and reduced activity. The wife observes these behaviours and tries to make sense of them.
She may attribute his behaviours to heightened pain and she may be frustrated that his pain is not
diminishing over time. In turn, she may react with behaviours that show her frustration at the
patient for not getting better. This type of interaction fits with the traditional operant approach of
viewing interaction in pain. The model presented in Figure 17.1 can also be used to explain the role
of empathic spouse responses. A patient may engage in facial expressions of pain as well as self-
disclosures about pain-related distress because she is motivated to obtain support from her hus-
band. The husband may understand these behaviours to indicate heightened pain and he may take
her perspective and feel moved to empathize with her. He then may react with empathic responses
such as validation. In this case, intimacy would be enhanced. Of course, some behaviours may
reinforce pain or well behaviours while affecting intimacy levels. For example, some behaviours
may reinforce pain and also enhance intimacy (e.g. solicitousness expressed with sympathy). Other
responses may reinforce pain but lead to deteriorations in intimacy (e.g. hostile solicitousness).
Still other reactions may reinforce well behaviours and enhance intimacy (e.g. praise for activity
and validation for difficulty in initiating activity). Research has yet to demonstrate how the various
unique combinations of responses may influence chronic pain behaviour over time, let alone the
progression of acute pain to chronic pain. This is likely to be an especially fruitful area of research.
As noted several times in this review, the bulk of the research on spousal influences on pain has
been conducted with patients with chronic pain. It is possible that spousal and family influences
would be quite similar in acute pain. Clearly, research is needed in this area to identify how inter-
actions between patients and spouses may actually prevent the progression of acute to chronic
pain. One avenue for this research is to recruit acute pain patients and their families and track
them over time. To minimize cost, patients could be chosen who are at risk for developing persist-
ent pain. For instance, persons who have undergone cardiac or breast surgery (Bruce et al. 2003;
Lahtinen et al. 2006; Vilholm et al. 2008) or who have had traumatic brain injury injuries
(Nampiaparampil 2008) are more likely to develop chronic pain compared to persons with low
back pain (Pengel et al. 2003). However, patients recovering from surgery or brain injuries may
have other difficulties (e.g. difficulty communicating, hospital stay or rehabilitation) that would
preclude their participation in research. Thus, it may be difficult to recruit these participants
within 1–2 months of pain onset. Another possibility is to recruit primary care patients with acute
pain and follow them over time. Because most of these patients are likely to be ambulatory, their
participation might be more feasible. However, a very large initial sample size will be needed to
identify those who transition to chronic pain. Successful research study recruitment may also
be more likely if participation compensation includes some sort of intervention or prevention
programme.
In terms of intervention research, acute pain patients may not be motivated to seek treatment
with their spouses because they might envision that any behavioural changes will be short-lived
until the pain is cured. It is only when the pain becomes chronic and couples come to understand
that long-lasting changes are likely that couples may begin to seek ways to cope with the illness
and its related effects on the relationship. Physicians and nurses may be able to educate acute pain
patients about the importance of maintaining healthy relationships during all phases of treat-
ment, not just when the pain becomes chronic. Most family-based interventions for chronic
illness focus on relationship issues, with many of these being treatments involving spouses
(Martire et al. 2004). Acute pain interventions can build on existing treatments available for cou-
ples facing chronic pain. For instance, couples can be taught about the issues associated with a
particular acute pain problem. The effective use of pain coping skills including the spouse’s role
in encouraging the use of such skills is also likely to be helpful in an acute pain situation.
Communication training and careful use of empathic responses may be beneficial as well. As
discussed earlier, couple-based treatments have direct effects on marital and individual well-
being but more limited effects on pain and disability. Research will be needed to determine if early
couple-based intervention can be used to prevent the acute pain problem from becoming chronic.
Perhaps such interventions do not have direct effects on pain but have an indirect effect on
the pain problem by promoting activity, couples’ coping, and empathic understanding while
preventing depression and marital distress.
Acknowledgements
Work on this chapter was partially supported by grant K01 MH61569 awarded to the first author.
References
Bakeman, R. & Gottman, J. M. (1997). Observing interaction: An introduction to sequential analysis (2nd
edn.). New York: Cambridge University Press.
Bruce, J., Drury, N., Poobalan, A. S., Jeffrey, R. R., Smith, W. C. S., & Chambers, W. A. (2003). The
prevalence of chronic chest and leg pain following cardiac surgery: A historical cohort study. Pain,
104, 265–73.
Buenaver, L. F., Edwards, R. R., & Haythornthwaite, J. A. (2007). Pain-related catastrophizing and
perceived social responses: Inter-relationships in the context of chronic pain. Pain, 127, 234–42.
Cano, A. (2004). Pain catastrophizing and social support in married individuals with chronic pain: The
moderating role of pain duration. Pain, 110, 656–64.
Cano, A. & Leonard, M. T. (2006). Integrative behavioral couple therapy for chronic pain: Promoting
behavior change and emotional acceptance. Journal of Clinical Psychology: In Session, 62, 1409–18.
Cano, A., Weisberg, J., & Gallagher, M. (2000). Marital satisfaction and pain severity mediate the
association between negative spouse responses to pain and depressive symptoms in a chronic pain
patient sample. Pain Medicine, 1, 35–43.
Cano, A., Gillis, M., Heinz, W., Geisser, M., & Foran, H. (2004a). Marital functioning, chronic pain, and
psychological distress. Pain, 107, 99–106.
Cano, A., Johansen, A., & Geisser, M. (2004b). Spousal congruence on disability, pain, and spouse
responses to pain. Pain, 109, 258–65.
Cano, A., Johansen, A., & Franz, A. (2005). Multilevel analysis of spousal congruence on pain, interference,
and disability. Pain, 118, 369–79.
Cano, A., Barterian, J., & Heller, J. (2008). Empathic and nonempathic interaction in chronic pain couples.
Clinical Journal of Pain, 24, 678–84.
Cano, A., Miller, L.R., & Loree, A. (2009). Spouse beliefs about partner chronic pain. Journal of Pain, 10,
486–92.
Christensen, A., Jacobson, N., & Babcock, J. (1995). Integrative behavioral couple therapy. In N. Jacobson
& A. S. Gurman (Eds.), Clinical handbook of marital therapy (2nd edn), pp. 31–64. New York:
Guilford Press.
Christensen, A., Atkins, D., Berns, S., Wheeler, J., Baucom, D., & Simpson, L. (2004). Traditional versus
integrative behavioral couple therapy for significantly and chronically distressed married couples.
Journal of Consulting and Clinical Psychology, 72, 176–91.
Christensen, A., Atkins, D. C., Yi, J., Baucom, D. H., & George, W. H. (2006). Couple and individual
adjustment for 2 years following a randomized clinical trial comparing traditional versus integrative
behavioral couple therapy. Journal of Counseling and Clinical Psychology, 74, 1180–91.
Coyne, J. C. & Fiske, V. (1992). Couples coping with chronic and catastrophic illness. In T. J. Takamatsu,
M. A. P. Stephens, S. E. Hobfall, & J. H. Crowther (Eds.) Family health psychology. Series in applied
psychology: Social issues and questions, pp. 129–49. Washington, DC: Hemisphere Publishing Corp.
Fordyce, W. E. (1976). Behavioral methods for chronic pain and illness. St. Louis, MO: C.V. Mosby.
Fruzzetti A. E. & Iverson, K. M. (2004). Mindfulness, acceptance, validation, and ‘individual’
psychopathology in couples. In S. C. Hayes, V. M. Follette, & M. M. Linehan (Eds.) Mindfulness and
acceptance: expanding the cognitive-behavioral tradition, pp. 168–91. New York: Guilford Press.
Fruzzetti A. E. & Iverson, K. M. (2006). Intervening with couples and families to treat emotion
dysregulation and psychopathology. In D.K. Snyder, J. A. Simpson, & J.N. Hughes (Eds.) Emotion
regulation in couples and families: Pathways to dysfunction and health, pp. 249–67. Washington, DC:
American Psychological Association.
Gauthier, N., Thibault, P., & Sullivan, M. J. L. (2008). Individual and relational correlates of pain-related
empathic accuracy in spouses of chronic pain patients. Clinical Journal of Pain, 24, 669–77.
Giardino, N. D., Jensen, M. P., Turner, J. A., Ehde, D. M., & Cardenas, D. D. (2003). Social environment
moderates the association between catastrophizing and pain among persons with a spinal cord injury.
Pain, 106, 19–25.
Goubert, L., Craig, K. D., Vervoort, T., et al. (2005). Facing others in pain: The effects of empathy. Pain,
118, 285–8.
Herbette, G. & Rime, B. (2004). Verbalization of emotion in chronic pain patients and their psychological
adjustment. Journal of Health Psychology, 9, 661–76.
Jacobson, N. & Christensen, A. (1998). Acceptance and change in couple therapy: A therapist’s guide to
transforming relationships. New York: Norton.
Jacobson, N., Christensen, A., Prince, S., Cordova, J., & Eldridge, K. (2000). Integrative behavioral couple
therapy: An acceptance-based, promising new treatment for couple discord. Journal of Consulting and
Clinical Psychology, 68, 351–5.
Johansen, A. & Cano, A. (2007). A preliminary investigation of affective interaction in chronic pain
couples. Pain, 132, S86–95.
Keefe, F., Blumenthal, J., Baucom, D., et al. (2004). Effects of spouse-assisted coping skills training and
exercise training in patients with osteoarthritic knee pain: A randomized controlled study. Pain, 110,
539–49.
Keefe, F., Caldwell, D., Baucom, D., & Salley, A. (1996). Spouse-assisted coping skills training in the
management of osteoarthritic knee pain. Arthritis Care & Research, 9, 279–91.
Keefe, F., Caldwell, D., Baucom, D., et al. (1999). Spouse-assisted coping skills training in the management
of knee pain in osteoarthritis: Long-term followup results. Arthritis Care & Research, 12, 101–11.
Keefe, F. J., Ahles, T. A., Sutton, L., et al. (2005). Partner-guided cancer pain management at the end of life.
Journal of Pain and Symptom Management, 29, 263–72.
Kerns R. D., Haythornthwaite J., Southwick S., Giller, E. L. (1990). The role of marital interaction in
chronic pain and depressive symptom severity. Journal of Psychosomatic Research, 34, 401–8.
Kiecolt-Glaser J. & Newton, T. (2001). Marriage and health: His and hers. Psychological Bulletin,
127, 472–503.
Knoll, N., Burkert, S., Rosemeier, H. P., Roigas, J., & Gralla, O. (2007). Predictors of spouses’ provided
support for patients receiving laparoscopic radical prostatectomy peri-surgery. Psycho-Oncology,
16, 312–19.
Kulik, J. A. & Mahler, H. I. M. (1989). Social support and recovery from surgery. Health Psychology,
8, 221–38.
Lahtinen, P., Kokki, H., Hynynen, M. (2006). Pain after cardiac surgery – a prospective cohort study of
1-year incidence and intensity. Anesthesiology, 105, 794–800.
Laurenceau, J., Feldman Barrett, L., & Pietromonaco, P. R. (1998). Intimacy as an interpersonal process:
The importance of self-disclosure, partner disclosure, and perceived partner responsiveness in
interpersonal exchanges. Journal of Personality and Social Psychology, 74, 1238–51.
Leonard, M. T. & Cano. A. (2006). Pain affects spouses too: Personal experience with pain and
catastrophizing as correlates of spouse distress. Pain, 126, 139–46.
Leonard, M., Cano, A., & Johansen, A. (2006). Chronic pain in a couples context: A review And integration
of theoretical models and empirical evidence. Journal of Pain, 7, 377–90.
Lousberg, R., Schmidt, A. J. M., & Groenman, N. H. (1992). The relationship between spouse solicitousness
and pain behavior: Searching for more experimental evidence. Pain, 51, 75–9.
Martire, L. M., Lustig, A. P., Schulz, R., Miller, G. E., & Helgeson, V. S. (2004). Is it beneficial to involve a
family member? A meta-analysis of psychosocial interventions for chronic illness. Health Psychology,
23, 599–611.
Martire, L. M., Schulz, R., Keefe, F. J., Rudy, T. E., Starz, T. W. (2007). Couple-oriented education and
support intervention: Effects on individuals with osteoarthritis and their spouses. Rehabilitation
Psychology. 52, 121–32.
Morley, S., Doyle, K., & Beese, A. (2000). Talking to others about pain: Suffering in silence. In M. Devor,
M. Rowbothan, & Z. Wiesenfeld-Hallin (Eds.) Proceedings of the ninth world congress on pain: Progress
in pain research and management, pp. 1123–29. Seattle, WA: IASP Press.
Nampiaparampil, D. D. (2008). Prevalence of chronic pain after traumatic brain injury: A systematic
review. Journal of the American Medical Association, 300, 711–19.
Newton-John, T. R. & Willams, A. C. de C. (2006). Chronic pain couples: Perceived marital interactions
and pain behaviours. Pain, 123, 53–63.
Paulsen, J. S., & Altmaier, E. M. (1995). The effects of perceived versus enacted social support on the
discriminative cue function of spouses for pain behaviors. Pain, 60, 103–10.
Pence L. B., Thorn, B. E., Jensen, M. P., Romano, J. M. (2008). Examination of perceived spouse
responses to patient well and pain behavior in patients with headache. Clinical Journal of Pain,
24, 654–61.
Pengel, L. H. M., Herbert, R. D., Maher, C. G., & Refshauge, K. M. (2003). Acute low back pain: Systematic
review of ifs prognosis. British Medical Journal, 327, 323–7.
Porter, L. S, Keefe, F. J., Wellington, C., & Williams, A. C. de C. (2008). Pain communication in the
context of osteoarthritis: Patient and partner self-efficacy for pain communication and holding back
from discussion of pain and arthritis-related concerns. Clinical Journal of Pain, 24, 662–8.
Romano, J. M., Turner, J. A., Friedman, R. A., Jensen, M. P., & Hops, H. (1991). Observational assessment
of chronic pain patient-spouse behavioral interactions. Behavioral Therapy, 22, 549–67.
Romano, J. M., Turner, J. A., Friedman, L. S., et al. (1992). Sequential analysis of chronic pain behaviors
and spouse responses. Journal of Consulting and Clinical Psychology, 60, 777–82.
Romano, J. M., Turner, J. A., Jensen, M. P., et al. (1995). Chronic pain patient-spouse behavioral
interactions predict patient disability. Pain, 63, 353–60.
Romano, J. M., Jensen, M. P., Turner, J. A., Good, A. B., & Hops, H. (2000). Chronic pain patient-partner
interactions: Further support for a behavioral model of chronic pain. Behavior Therapy, 31, 415–40.
Saarijarvi, S. (1991). A controlled study of couple therapy in chronic low back pain patients: Effects on
marital satisfaction, psychological distress and health attitudes. Journal of Psychosomatic Research, 35,
265–72.
Saarijarvi, S., Rytokoski, U., Alanen, E. (1991). A controlled study of couple therapy in chronic low back
pain patients: No improvement of disability. Journal of Psychosomatic Research, 35, 671–7.
Saarijarvi, S., Alanen, E., Rytokoski, U., & Hyyppa, M. (1992). Couple therapy improves Mental well-being
in chronic low back pain patients: A controlled, five year follow up study. Journal of Psychosomatic
Research, 36, 651–6.
Schwartz, L., Slater, M. A., & Birchler, G. R. (1996). The role of pain behaviors in the modulation of
marital conflict in chronic pain couples. Pain, 65, 227–33.
Schwartz, L., Jensen, M. P., & Romano, J. M. (2005). The development and psychometric evaluation of an
instrument to assess spouse responses to pain and well behavior in patients with chronic pain: The
Spouse Response Inventory. The Journal of Pain, 6(4), 243–52.
Severeijns, R., Vlaeyen, J. W. S., & van den Hout, M. A. (2004). Do we need a communal coping model of
pain catastrophizing? An alternative explanation. Pain, 111, 226–9.
Sharp, T. J. (2001). Chronic pain: A reformulation of the cognitive-behavioural model. Behaviour Research
and Therapy, 39, 787–800.
Stroud, M. W., Turner, J. A., Jensen, M. P., & Cardenas, D. D. (2006). Partner responses to pain behaviors
are associated with depression and activity interference among persons with chronic pain and spinal
cord injury. The Journal of Pain, 7, 91–9.
Sullivan, M. J. L., Bishop, S., & Pivik, J. (1995). The Pain Catastrophizing Scale: Development and
Validation. Psychological Assessment, 7, 524–32.
Sullivan, M. J. L., Thorn, B. E., Haythornthwaite, J. A., Keefe, F. J., Martin, M., & Bradley, L. A. (2001).
Theoretical perspectives on the relation between catastrophizing and pain. Clinical Journal of Pain,
17, 52–64.
Sullivan, M. J. L., Adams, H., & Sullivan, M. E. (2004). Communicative dimensions of pain
catastrophizing: Social cueing effects on pain behaviour and coping. Pain, 107, 220–6.
Sullivan, M. J. L., Martel, M. O., Tripp, D., Savard, A., & Crombez, G. (2006). The relation between
catastrophizing and the communication of pain experience. Pain, 122, 282–8.
Thorn B. E., Ward L. C., Sullivan M. J. L., & Boothby J. L. (2003). Communal coping model of
catastrophizing: Conceptual model building. Pain, 106, 1–2.
Turk, D. C. & Kerns, R. D. (1985). The family in health and illness. In D. C. Turk & R. D. Kerns (Eds.),
Health, illness, and families: A life-span perspective, pp. 1–22. New York: Wiley.
Turk D. C., Meichenbaum D., Genest, M. (1983). Pain and behavioral medicine: A cognitive-behavioral
perspective. New York: Guilford Press.
Turk, D. C., Kerns, R. D., Rosenberg, R. (1992). Effects of marital interaction on chronic pain and
disability: Examining the down side of social support. Rehabilitation Psychology, 37, 259–74.
Vilholm, O. J., Cold, S., Rasmussen, L., & Sindrup, S. H. (2008). The postmastectomy pain syndrome: an
epidemiological study on the prevalence of chronic pain after surgery for breast cancer. British Journal
of Cancer, 99, 604–10.
Vlaeyen, J. W. S. & Crombez, G. (1999). Fear of movement/(re)injury, avoidance and pain disability in
chronic low back pain patients. Manual Therapy, 4, 187–95.
Williamson, D., Robinson, M. E., & Melamed, B. (1997). Pain behavior, spouse responsiveness, and marital
satisfaction in patients with rheumatoid arthritis. Behavior Modification, 21, 97–118.
Chapter 18
Effects of Workers’ Compensation

Systems on Recovery from
Disabling Injuries
James P. Robinson and John D. Loeser
18.1 Introduction
Many observers have argued that disability systems have adverse effects on the people whom
they serve (Harris et al. 2005; Atlas 2007; Gabbe 2007). The purpose of this chapter is to evaluate
these claims, and consider ways in which disability systems might affect the behaviour of their
beneficiaries.
Any discussion of the effects of disability systems is complicated by the fact that there are many
types of disability, and many organizations that provide disability benefits. For example, some
people (e.g. ones with C4 quadriplegia or severe mental retardation) require personal attendants
or long term institutionalization because they cannot perform basic activities of daily living
(ADLs). Others are able to perform basic ADLs, but are disabled in the sense that they are unable
to work. Among work-disabled individuals, some have life-long disabilities, whereas others func-
tion in the work place for an extended period of time, and then develop medical conditions that
disable them. These conditions might be results of their work activities (e.g. a logger who sustains
a spinal fracture when a tree falls on him), or might be unrelated (e.g. a worker who has a myo-
cardial infarction). Corresponding to the multiplicity of types of disability and circumstances
under which people become disabled, there are several organizations that provide assistance to
them. In the USA these include workers’ compensation (WC), programmes run by the Social
Security Administration, disability programmes run through the Veterans Administration sys-
tem, private disability programmes, and welfare programmes. It is not uncommon for a disabled
person to interact with several of these—e.g. he might start with a workers’ compensation claim,
but later receive Social Security Disability Insurance benefits.
For simplicity, this chapter focuses on individuals who sustain injuries at work and receive
benefits through a WC system in the USA. Several other caveats are in order. First, it should be
noted that the majority of work injuries (approximately 70%) (Bureau of Labor Statistics 2008)
resolve uneventfully, and do not lead to work disability. A WC system is more likely to affect the
recovery of an injured worker (IW) if the worker has a significant amount of contact with it.
Thus, this chapter focuses on work injuries that lead to work disability, and especially ones associ-
ated with protracted work disability. Second, there are workers’ compensation systems for each of
the 50 states, as well as federal systems (Williams 1991). These systems vary substantially. It is
beyond the scope of this chapter to explore this variation (Analysis of workers’ compensation
laws 1996). But it is important to note that an IW’s behaviour is likely to be influenced by specific
rules and regulations pertaining to the WC system with which he/she interacts.
Two basic questions need to be asked regarding the influence of the WC system on IWs: (1) are
IWs affected by their interactions with the WC system, and if so, (2) what are the mechanisms by
which the effects take place?
18.2 Is there an effect?

Three kinds of evidence suggest that the behaviour of IWs is affected by their interactions
with WC systems. As noted below, though, each kind of evidence is subject to more than one
interpretation.
18.2.1 Shape of the recovery curve

Informal observation suggests that as time goes on following a disabling work injury, IWs become
progressively more resistant to treatment, and more ‘stuck’ in their disability. Some observers
have used the term ‘disability syndrome’ to describe the constellation of behavioural issues that
chronically disabled IWs often demonstrate (Robinson et al. 1997). The hypothesis that chroni-
cally disabled IWs develop disability syndromes is at least compatible with large-scale data regard-
ing recovery from work injuries. As demonstrated by Cheadle et al. (Cheadle et al 1994) and
several other investigators, the recovery curve following disabling injuries follows a negative
exponential curve (see Figure 18.1). Thus, for example, among workers who go on disability
100
90
80
70
60
Percent
50
40
30
20
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12
Time loss duration (months)
Fig. 18.1 Probability of continued disability payments as a function of time in a cohort of injured
workers who were initially disabled by a work injury. Reproduced from Cheadle A, Franklin G,
Wolfhagen C, et al. Factors influencing the duration of work-related disability: a population-based
study of Washington State workers’ compensation, Am J Public Health, 84(2), 190–6, 1994 with
permission from The Sheridan Press.
EFFECTS OF WORKERS’ COMPENSATION SYSTEMS 357
following an injury, approximately 70% go off disability within 2 months. However, at about
4 months, the recovery curve flattens out. A worker who is on disability at 4 months has about of
50% chance of still being on disability at 1 year following injury. Based on this curve, one might
conclude that workers who are disabled at 4 months become more resistant to rehabilitation,
perhaps because of interactions that they have with the WC system. However, at least two other
explanations of the curve are also possible. One is that IWs who end up in the tail of the curve had
risk factors at the time of injury that made them less likely to recover. Thus, their paths might be
determined by pre-injury or injury factors, rather than by subsequent interactions with a WC
system (Mustard and Hertzman 2001). Second, as time elapses following a disabling injury, IWs
face the kinds of stresses that impact all people who are dislodged from the work place. The
importance of such dislodgement is demonstrated by research on the psychological effects of
prolonged absence from work that has nothing to do with injuries or disability (Bartley 1994;
Dooley et al 1996; Murphy and Athanasou 1999; Fryer and Fagen 2003; Leino-Loison et al. 2004).
It is reasonable to presume that IWs who have lost their confidence and have become depressed
as a result of prolonged separation from the work force will view their prospects for success in the
workplace negatively. Their emotional dysfunction and loss of confidence might delay recovery
from their work injury, regardless of their interactions with WC.
18.2.2 Moral hazards

Studies on the effects of insurance indicate that when people have insurance that indemnifies
them against certain events, they tend to behave in ways that make these events more likely to
occur. The term ‘moral hazard’ has been used to describe this effect. For example, if individuals
are indemnified against the loss of their home in a fire, they may become careless in the safety
procedures they employ at home to prevent fires, or may even set fires deliberately in order to
receive insurance benefits. In the context of WC, there is evidence that as benefits to IWs increase,
the probability of filing WC claims increases, as does the duration of claims (Loeser et al. 1995;
Dembe and Boden 2000). As noted by Dembe and Boden (2000), when the term moral hazard
was introduced in the 19th century, it was widely believed that the changes in behaviour elicited
by insurance benefits were attributable to carelessness, deceit or fraud on the part of insurance
beneficiaries. These authors document that the term continues to be used in ways that disparage
IWs by suggesting that many of them file claims that are unnecessary or fraudulent. Thus, the sta-
tistical finding of an association between insurance benefits and aggregate behaviours of benefici-
aries is congruent with explanations for protracted disability that focus on worker malingering
and/or ‘secondary gain’. These explanations will be described below.
It is beyond the scope of this chapter to discuss the concept of moral hazard in any depth.
However, a few comments are appropriate. First, as Dembe and Boden (2000) point out, much of
the research on moral hazards has been carried out by economists. An implicit assumption in
economic analyses of work behaviour is that workers voluntarily choose employment in order to
make money, and voluntarily choose when to leave their jobs. Thus, when a worker files a WC
claim, this is viewed as a voluntary, calculated act rather than as a response to an injury that actu-
ally incapacitates the worker. If workers’ decisions to file claims are construed as calculated, it is a
short conceptual leap to construe them as fraudulent. Moreoever, as Dembe and Boden forcefully
argue, moral hazard research has tended to focus on the suspicious behaviour of IWs rather than
on the behaviour of employers, insurance adjusters, and others who participate in WC claims.
They state:
In the economist’s model of the labor market, workers freely sell their labor to employers, control their
working conditions, decide when they are injured and how long they need to recuperate, choose their
medical care provider, and decide on the type and amount of care that they receive … They become
injured (or claim that they are injured when they are not), and the defenseless employer or insurer
must pay them until the injured workers decide it’s time to return to work, perhaps after taking a vaca-
tion in sunny Florida, running in a marathon in Hawaii, and painting their houses. Consider another
possibility. Workers who are injured frequently do not file workers’ compensation claims, worrying
that they will be fired, that their fellow workers will think of them as malingerers and cheats, and that
they will carry with them the social stigma formerly only associated with people on welfare. If they file
and take time from work to recover, they may not have a job when they are ready to return. Potential
future employers will know that they have filed workers’ compensation claims and may decide that it
is too risky to hire them. If the employer or insurer contests the claim, they then have to decide to give
up their benefits or deal with their anxiety and discomfort around lawyers and the legal system (Dembe
and Boden 2000, p. 269).
Also, it should be noted that even if generous benefits influence workers to exaggerate their
incapacitation or file fraudulent claims (‘Retirees’ disability epidemic’ 2008), the relevance of this
phenomenon to IWs is questionable. The reason for this is that WC benefits are generally modest,
and do not cover the costs that IWs incur as a result of their injuries (Hunt 2003; Reville et al.
2005). Thus, it is hard to understand the economic benefits that IWs receive when they file claims.
It is also noteworthy that a very substantial proportion of workers do not file claims for work
injuries or occupational illnesses (Morse et al. 2000; Shannon and Lowe 2002; Scherzer et al.
2005). Investigators have attributed this under-reporting of injuries to a variety of factors, such as
fear of retaliation from employers or pressure from co-workers. But it is also likely that workers
realize that they are likely to suffer adverse financial consequences if they file a WC claim and go
off work. These data put the issue of moral hazards in WC in a different light. A reasonable inter-
pretation is that WC benefits are generally so low that they contribute to under-reporting of
injuries by workers, but that the under-reporting is less pronounced in WC systems that provide
relatively generous benefits.
18.2.3 Comparisons between IWs and other patient groups

The most convincing evidence for adverse effects associated with participation in WC systems
comes from studies comparing patients with WC to patients without WC regarding response to
treatment. A recent meta-analysis performed by Harris et al. (2005) included 211 studies. In
175 of them, it was found that compensation/litigation was associated with poorer surgical
outcome. Thirty five studies showed no association between compensation/litigation status
and outcome, while only 1 study found that patients with compensation/litigation had better
outcomes than others. A formal meta-analysis performed on 129 studies with appropriate
data revealed on odds ratio of 3.79 for an unsatisfactory surgical outcome among patients with
compensation/litigation.
It is worth noting that the non-compensation patients were often not clearly described in the
studies reviewed, although it is likely that for the most part they were patients who received treat-
ments under private insurance coverage and were not involved in litigation. Here they are desig-
nated as ‘private pay’ patients. Also, within the compensation cohort, the authors of individual
studies and Harris et al. did not clearly distinguish between compensation status and litigation
status, or between different types of compensation. However, since the ‘compensation’ cohort in
most of the studies consisted of injured workers, the results of the review are relevant to the
present discussion.
At first glance, the above data seem to provide convincing evidence for an adverse effect of
WC on the recovery of IWs. However obvious this conclusion appears to be, it is important to
consider alternative ways of interpreting the findings reviewed by Harris et al. First, studies
that compare recovery trajectories of injured workers vs. ‘private pay’ patients essentially never
control adequately for baseline, pre-injury variables on which these two types of patients may
differ. There is evidence that important differences exist between the two types of patients. For
example, Atlas et al. (2000) found that in comparison to private pay patients, WC patients at
baseline were more likely to be young, male, and employed as labourers. Moreover, there is evi-
dence that compared to IWs in general, ones with protracted WC claims are relatively likely to
have pre-injury histories characterized by high health care utilization and a high risk of receiving
welfare benefits (Mustard and Hertzman 2001) Also, longitudinal studies show that among work-
ers who are initially asymptomatic, ones who report new onset of symptoms and/or file a claim
for a new injury are more likely than other workers to report job-related psychosocial stressors
such as monotony of the job (Harkness et al. 2003) or job dissatisfaction (Bigos et al. 1992). These
data strongly suggest that individuals with risk factors for protracted disability and health prob-
lems are over-represented among workers who file WC claims.
It is also likely that WC and private pay patients differ on a wide range of variables related to
employment. As noted above, Atlas et al. (2000) found that WC patients treated for disc hernia-
tions were more likely than private pay patients to be labourers. Also, they found that 80% of their
private pay patients reported working during the 4 weeks prior to their baseline assessments vs.
only 35% of WC patients. It might seem that these work-related variables have nothing to do with
patients’ clinical status following surgery. However, this assumption ignores the fact that for
many IWs, employment issues, clinical status of the work injury, and access to medical care can
be intertwined in complex ways. The reason for this is that WC claims are typically closed when
an IW’s clinical status has reached medical stability, and disability benefits are terminated if
the worker is judged to be capable of working. Given these rules, IWs who believe that they are
unable to return to the workforce have an incentive to extend their benefits by reporting ongoing
symptoms following treatments.
18.3 How does the workers’ compensation affect the

behaviour of IWs?
Despite the caveats given above, the most parsimonious way to explain the findings of studies on
recovery curves, moral hazards, and surgical outcomes for WC vs. private pay patients is that WC
systems adversely affect the recovery of IWs. Assuming this is true, what are the processes that
mediate this adverse effect?
In addressing this question, it should be noted that the factors influencing IWs’ recovery and
return to work are multiple and complex. They may well include the workers’ interactions with
representatives of their WC carrier, but they also include the severity and functional consequences
of the work injuries, the demands of the workers’ jobs at the time of injury, their opportunities for
alternative employment based on prior education and training, their motivation to return to
work, their resourcefulness in overcoming barriers to return to the job they had at the time of
injury or find alternative employment, their relationships with co-workers and managers at the
job, information they receive from medical professionals about the risks they take by engaging in
various kinds of activity, and advice they receive from their attorneys. Thus, we should not expect
simple explanations when we try to abstract the role of the WC system from the myriad of influ-
ences on an IW’s recovery, and determine exactly how the worker’s interactions with the system
affect him/her.
18.4 Possible explanations for the negative effect of WC

Several hypotheses have been advanced to explain the negative effect of WC on the recovery of
IWs. Many of the studies bearing on these hypotheses combine individuals with different types of
compensation (e.g. WC and Social Security Disability Insurance), and combine compensation
claimants with litigants in tort cases. However, since IWs with WC claims are well represented in
these studies, they will be considered below without any effort to isolate the effect of WC.
18.4.1 Malingering and secondary gain

Malingering
Many publications, often appearing in industry journals, have titles that suggest an epidemic of
fraud/malingering perpetrated by IWs—e.g. ‘Employers must be alert to comp fraud’ (Auth
2005), ‘Reducing workers’ compensation fraud: a deterrent approach’ (Mah 1998), and ‘Screen
patients quickly for workers’ comp fraud’ (Easton 1997). The concern implied by these titles is
consistent with results of a recent review by Aronoff et al. (2007). These authors reported an
astonishingly high prevalence of malingering in settings where compensation or litigation are
involved—for example, some of the studies they reviewed reported a 36–50% rate of malingering
among compensation/litigation patients with chronic pain.
The literature on malingering among compensation recipients dovetails with the previously
described literature on the moral hazard associated with compensation for work injuries. Both
areas of research suggest that the negative effect of compensation on recovery can be explained in
terms of wilful deceit on the part of IWs. If this explanation is valid, it provides a conceptually
simple, though somewhat distasteful, explanation of the compensation effect.
However, serious questions can be raised about the validity of the research on malingering
summarized by Aronoff et al. Malingering is best defined as fraud in a medical context. As such,
it is a criminal activity, and involves planned feigning of incapacitation from a medical condition
in order to receive some kind of reward (typically a monetary reward). Thus, in order to demon-
strate malingering, an investigator must: (1) show evidence that a suspect is deliberately feigning
his/her medical condition, and (2) provide evidence that the feigning is designed to help the sus-
pect secure a monetary award. The malingering research mentioned above runs into problems
with respect to both criteria.
One key issue is that most of the studies cited by Aronoff et al. as proof of malingering involve
the behaviour of claimants on paper and pencil tests of various kinds. In general, malingering is
identified when patients perform very poorly on the tests, or when their test performance is dra-
matically altered by instructions given by an experimenter. It is beyond the scope of this chapter
to discuss the validity of these tests among individuals who claim to have brain injuries or other
work-related conditions that would directly affect their test taking abilities. But regardless of the
validity of the tests among people with purported neurocognitive impairment, their relevance to
work injuries in general is highly suspect. The reason for this is that epidemiologic data clearly
indicate that musculoskeletal injuries dominate all other categories of work related disorders. For
example, the Bureau of Labor Statistics compiles data on various categories of injuries/illnesses
and indicates the percentage of total injuries/illnesses represented by each category. Data for 2005
indicate that 40.8% of all injuries/illnesses were coded as sprains/strains, 7.8% as fractures, 2.9%
as back pain, 1.3% as carpal tunnel syndrome, and 0.5% as tendonitis (Bureau of Labor Statistics
2005). Thus, at least 53.5% of all work injuries/illnesses in 2005 were musculoskeletal in nature.
In these conditions, workers remain disabled because of limitations in physical capacities, rather
than limitations in their ability to perform cognitive tasks. Thus, the detection of malingering
must rest on observations of claimants’ physical activities, as demonstrated, for example, in

surveillance videotapes. Although surveillance videotaping is used in WC, there are very few
publications on the procedure in the medical literature (Kay and Morris-Jones 1998; ‘Phoney
worker’s comp. claim?’ 2007), and the pick up rate of malingering by video surveillance is uncer-
tain. Dembe and Boden (2000) mention a series of newspapers articles on surveillance videotap-
ing in California during the late 1990s—the activity was reportedly disbanded because evidence
of malingering was found so rarely (Fricker 1997).
As far as incentives for malingering are concerned, literature mentioned above in the discussion
of moral hazards is relevant. This literature clearly demonstrates that the monetary benefits that
IWs receive—either during the course of their claims or when their claims are closed—do not
compensate for the financial losses they incur as a result of being out of the work force, or being
re-employed in low paying jobs after their claims have been closed. Thus, assertions that malin-
gering is rampant among IWs fail to give a satisfactory answer to a very basic question: why would
an IW engage in criminal activity in order to gain such a paltry sum of money?
Secondary gain and related terms

There is a substantial literature related to the hypothesis that individuals involved in compensa-
tion or litigation are influenced by ‘secondary gain’ (Fishbain et al. 1995). This term has its origin
in the writings of Freud, who distinguished between the primary gain in a neurosis (management
of unconscious impulses) and various secondary gains (such as enjoying the nurturance that one
can get for being sick). In the context of compensation/litigation, multiple potential rewards
for failure to recover and protracted disability have been proposed (Box 18.1) (Dersh et al.
2004). Most writers, however, discuss secondary gain in relation to financial rewards that a claim-
ant can get.
Some observers emphasize that secondary gain describes the behaviour of claimants who are
influenced unconsciously by potential rewards in their environment (Fishbain 1994). The idea
that secondary gain involves unconscious behaviour makes it morally less objectionable than
malingering, which refers to conscious fabrication in order to achieve some reward. However,
this seemingly clear distinction between secondary gain and malingering has been challenged by
other writers. For example, Dersh et al. (2004) address several conceptual conundrums regarding
the secondary gain construct, and indicate that IWs sometimes consciously engage in behaviours
designed to achieve secondary gains. They do not provide a specific definition of secondary gain,
but, rather describe its features (see Box 18.2).
The disagreements among different observers in the conceptualization of secondary gain high-
light the elusiveness of the concept. It is best to think of the terms ‘malingering’ and ‘secondary
gain’ as alternative ways to describe the effect of environmental rewards on the behaviour of
claimants, with the difference being one of the moral opprobrium assigned to a claimant’s behav-
iour. While neither term is flattering, a clinician or adjudicator who accuses a claimant of malin-
gering is rendering a much harsher judgment than one who says the claimant’s behaviour is
motivated by secondary gain.
A myriad of other terms have been used to describe the behaviours of claimants as they inter-
act with the compensation systems or legal systems. Mendelson (1988) listed several of these (See
Box 18.3). All of them are unflattering, and in some way embody the perception by some observ-
ers that compensation/litigation patients tend to behave in ways that are self-serving.
It is worth noting that some observers have talked about the secondary losses that compensa-
tion/litigation claimants sustain, rather than just the gains they seek (Fishbain 1994; Gatchel et al.
2002). Box 18.4 lists possible losses suggested by Dersh et al. The concept of secondary losses
Box 18.1 Types of secondary gains

Internal
1. Gratification of pre-existing unresolved dependency strivings or affiliation needs.
2. Gratification of pre-existing unresolved revengeful strivings (e.g. revenge directed toward
insurance carriers or adjustors who gave patient a hard time; revenge directed toward spouse/
partner who was perceived as not living up to his or her responsibilities in the relationship).
3. An attempt to elicit care-giving, sympathy, and concern from family and friends.
4. Family anger because of patient’s disability may increase patient resentment and determi-
nation to get his or her due to prove entitlement.
5. Obtaining one’s entitlement for years of struggling, dutiful attention to responsibilities,
and a ‘much-earned’ recompense.
6. Ability to withdraw from unpleasant or unsatisfactory life roles, activities, and responsi-
bilities, including those of ‘breadwinner’, spouse, and parent.
7. Adoption of ‘sick role’ allows the patient to communicate and relate to others in a new,
socially sanctioned manner.
8. Converting a socially unacceptable disability (psychological disorder) to a socially accept-
able disability (injury or disease)
9. Displacing the blame for one’s failures from oneself to an apparently disabling illness
beyond one’s control.
10. Maintenance of status in family.
11. Holding a spouse/partner in a marriage/relationship.
12. Avoiding sex.
13. Contraception.
14. Obtaining drugs.
15. Denial of the randomness of events.
External
1. Obtaining financial awards associated with disability.
a. Wage replacement (short- and long-term disability, social security disability insurance,
workers’ compensation benefits).
b. Settlement (disability- or impairment-based).
c. Disability-based debt protection (e.g. credit cards, mortgage, auto loan).
d. Subsidized child and family care, housing, and food.
2. Protection from legal and other obligations (e.g. child support payments, court appear-
ances, parole or probation demands).
3. Job manipulation (e.g. promotion or transfer, handling personnel or work adjustment
difficulties, prevention of lay-off or termination).
4. Vocational retraining and skills upgrade.
With kind permission from Springer Science+Business Media: Journal of Occupational Rehabilitation, The
Management of Secondary Gain and Loss in Medicolegal Settings: Strengths and Weaknesses, 14(4), 2004,
Jeffrey Dersh.
Box 18.2 Features of secondary gain

◆ Secondary gain issues are prominent factors in illness, particularly chronic illness and illness
being evaluated and treated in a medicolegal context.
◆ Secondary gain issues are rarely suggestive of ‘pure’ malingering or factitious disorder.
◆ Disability/illness exaggeration is frequently associated with secondary gain, and can often be
inferred from abnormal illness behavior.
◆ Conscious, unconscious, and preconscious processes are involved in secondary and tertiary
gain.
◆ Unconscious and conscious processes are better conceptualized from a dimensional
perspective, rather than a dichotomous one.
◆ When conscious processes are primary, the patient may conceal secondary gains from
others (resulting in hidden agendas). On the other hand, the patient may be quite open and
transparent about his or her agendas.
◆ All chronic illnesses (physical and psychiatric) involve secondary gains and losses, as well as
tertiary gains and losses.
◆ In almost all cases, there are multiple gains and losses, resulting in multiple agendas.
◆ Understanding the gain and loss issues (the secondary and tertiary economies) will lead to
the most effective management of these patients.
◆ Operant conditioning (reinforcement) principles will often be effective in altering this
economy in a positive direction.
◆ However, these economies cannot be understood in a purely mechanistic, operant condi-
tioning, or rational manner.
◆ Psychodynamic and sociocultural factors are also important factors to be considered in
understanding gain and loss issues.
Jeffrey Dersh.
associated with injury and participation in compensation systems dovetails with the findings in
studies cited above—i.e. ones demonstrating that the financial remuneration IWs receive from
WC systems almost never matches the financial losses they incur as a result of their injuries.
Cured by a verdict?
Kennedy famously described compensation neurosis as ‘A state of mind, born out of fear, kept
alive by avarice, stimulated by lawyers, and cured by a verdict’ (Kennedy 1946). The idea that the
apparent disability of a person with a compensation/litigation claim suddenly resolves after some
decision or verdict has been reached regarding his/her claim is congruent with the view that the
prior apparent incapacitation reflected malingering or seeking secondary gain. As Mendelson
documented in his 1988 book Psychiatric aspects of injury claims, the empirical evidence to that
point in time did not support Kennedy’s assertion. A later review by Fishbain et al. (1995) reached
the same conclusion. A recent study by Overman et al. (2008) also questions the validity of
the ‘cured by a verdict’ hypothesis. These investigators examined health complaints among a large
cohort of Norwegians several years before, around the time of, and several years after they were
Box 18.3 Terms related to secondary gain

◆ Accident aboulia
◆ Accident neurosis
◆ Accident victim syndrome
◆ Aftermath neurosis
◆ American disease
◆ Attitudinal pathosis
◆ Compensation hysteria
◆ Compensationitis
◆ Compensation neurosis
◆ Entitlement neurosis
◆ Erichsen’s disease
◆ Functional overlay
◆ Fright neurosis
◆ Greek disease
◆ Greenback neurosis
◆ Justice neurosis
◆ Litigation neurosis
◆ Mediterranean back
◆ Mediterranean disease
◆ Neurotic neurosis
◆ Postaccident anxiety syndrome
◆ Postaccident syndrome
◆ Post-traumatic syndrome
◆ Profit neurosis
◆ Railway brain
◆ Railway spine
◆ Secondary gain neurosis
◆ Syndrome of disproportionate disability
◆ Traumatic hysteria
◆ Traumatic neurasthenia
◆ Traumatic neurosis
◆ Triggered neurosis
◆ Unconscious malingering
◆ Vertebral neurosis
◆ Wharfie’s back
◆ Whiplash neurosis
From Mendelson, G., Psychiatric Aspects of Personal Injury Claims, Table 1, pp. 13, Copyright 1988. Courtesy
of Charles C Thomas Publisher, Ltd., Springfield, IL.
Box 18.4 Types of secondary loss

1. Economic loss.
2. Loss of meaningfully relating to society through work.
3. Loss of work social relationships.
4. Loss of social support network.
5. Loss of meaningful and enjoyable family roles and activities.
6. Loss of recreational activities.
7. Loss of respect from family and friends.
8. Negative sanctions from family.
9. Loss of community approval.
10. Loss of respect from those in helping professions (e.g. physicians).
11. New role not comfortable and not well defined.
12. Social stigma of being chronically disabled.
13. Guilt over disability.
14. Communications of distress become unclear.
Jeffrey Dersh.
awarded disability benefits. ‘Somatic symptoms’ and ‘pain distribution’ tended to worsen as the
time when disability was granted approached, with improvement during the 7 years after awards
were granted. However, during the post-award period, the complaints of respondents did not
disappear; rather, they generally returned to the baseline levels that had existed years prior to the
awards. When differences between pre and post award indices were noted, they were in the direc-
tion of being worse following awards than prior to the awards. These data suggest that claimants’
increased symptom levels shortly before disability determinations are made reflect stress associ-
ated with the disability evaluation process. Other large scale studies document relatively high
morbidity and mortality rates among disability beneficiaries (Kivimaki 2003; Vahtera et al 2004;
Wallman et al 2006). In sum, the balance of evidence argues against the conclusion that claimants
are cured by a verdict.
18.4.2 Perceptions of IWs

Whereas research on malingering and secondary gain suggest that workers consciously or
unconsciously exaggerate their disability in order to achieve financial gains, a very different pic-
ture emerges from qualitative studies in which the perceptions of IWs are assessed by interview
(Cole et al. 2002; Cromie et al. 2003; Strunin and Boden 2004; Beardwood et al 2005; Cacciacarro
and Kirsh 2006; Lippel 2007; MacEachen et al. 2007; Soeker et al. 2008). These document recur-
ring themes and concerns of the workers. Since the qualitative studies do not provide statisti-
cal data, quotations from the studies are given below to describe themes that commonly
emerged during interviews with IWs. Four themes were prominent in the majority of qualitative
studies.
Sense of being distrusted

Many subjects felt that they were disbelieved by physicians and WC claims managers. They
expressed the view that they had to prove their incapacitation to sceptical observers. Some of
them expressed concern that they might be put under surveillance, and that any behaviours that
were documented in surveillance tapes might be used against them. For example, Cromie et al.
(2003, pp. 1086–7) write:
Medical examiners made comments in their medical reports that had consequences for the financial
and physical well-being of their patients. In some cases, they were quite judgmental and damning.
Louise reported how a physician attributed her symptoms to a familial condition ‘which was a total
and outright lie’ (p. 6). Denise described a situation where a neurologist decided she was ‘swinging the
lead . . . malingering . . .’ She said: ‘It was written on my history that I was a malingerer . . . they virtually
withdrew all pain-killing drugs . . . and I actually got treated very badly from that point in time’ (p. 10).
Medical examiners, and in particular those to whom participants were referred by insurers, seemed to
reflect some of the stereotypical prejudices toward claimants described by Quintner. . . The attitude of
some medical practitioners led Jane to the point that she wanted to terminate her claim. Jane stated
that a rheumatologist had written in a report that she was using her injury as an excuse not to have
children and that in choosing not to have children she was avoiding her duty to her husband (Jane,
p. 5). The rheumatologist implied that her problem was primarily psychological (Jane, p. 5). The
rheumatology report was ‘amazingly hostile . . . and . . . out of sync [sic] with any of the other reports
I’ve got myself’ (Jane, p. 6). She found the rheumatologist’s report ‘profoundly offensive’ and said
that it made her ‘so angry,’ but at the same time she felt unable to show the report to anybody else
(Jane, p 15).
Conflict with the WC system

Workers repeatedly talked about the difficulties they had in their interactions with WC systems.
Problems ranged from personal insults by claims managers to fears about having benefits cut off.
For example, Beardwood et al. (2005, pp. 40–2) state:
Stories of difficulties with the compensation system were abundant. However, the major areas of
tension could best be categorized as concern regarding the adversarial nature of the relationship with
the compensation system; the lack of information, control, and choice provided by the system; and the
fear and paranoia that were evoked through interaction within this system. Many workers found
themselves entangled in an adversarial relationship, which was characterized by a lack of trust, misin-
terpretation of information provided, and a perception of inappropriate responses of workers within
the system . . . A worker provided an analysis of the power structures in this way: ‘It’s fear. They instill
fear in you. They’re experts at that. They know how to instill fear in the person that’s injured. You’re
already injured so your defenses are down. You’re now sort of backed into a corner because they’re
the ones that are supposed to be paying the money. And yet they’re not paying you properly. Or
they’re threatening to take it away. So you feel like you’re alone and you have no recourse. You know.
And it’s a rotten position to be in.
Stress; personal loss; family loss

IWs repeatedly talked about the emotional stresses they experienced, and about the impact of
their injuries on their finances and their families. Beardwood et al. (2005, p. 43) state:
The final stage was the impact on workers’ lives and the nature of living as an injured worker. Themes
included financial and social impacts, mental and emotional impacts, impacts on activities of daily
living, and living with pain.
Not surprisingly, workers described the financial hardships they experienced as a result of losing
their regular wages. Some incurred medical expenses alongside these losses, which further increased
their debt . . . The effects on family life and social inclusion were severe, as exemplified by this worker’s
report: ‘It has caused the end of my marriage. I do still have the custody of my children, but my injury
affects them. If I have some sort of painful episode, my five-year-old starts to cry. My twelve-year-old
really helps me out a lot by dealing with some things that I physically can’t do. I get a lot of help from
my parents. I’ve been forced to move in with them because of the injury, and I’ll probably be stuck
there for a while until I’m able to be completely independent.’ Many workers revealed that their psy-
chological state had deteriorated through the process of becoming injured and dealing with the system.
They discussed the effects of poverty, role disruption, and stress on their mental and emotional health.
Depression was a common experience.
Anger with the WC system

Interviewees often talked about their anger with the WC system. This was articulated best by an
IW who became an activist, and wrote an entire article about her experiences (Hammer 2000).
She pointed out that although attention was often given to fraud/malingering on the part of IWs,
misbehaviour by other participants in WC claims needed more attention:
Employer fraud is rampant in under-reporting and not having any insurance at all. Workers know
when their employer waits to report their injury way past the required time. The employer hopes that
the injury will improve with care from just the company doctor. Then the employer doesn’t report the
claim at all. Employers are famous for declaring a worker an independent contractor to avoid
Workers’Compensation coverage as well as misclassifying their workers to reduce insurance premi-
ums. Employers have under-reported work as less hazardous than it really is. We’ve seen examples of
employers underestimating employment projections at the beginning of the premium year to receive
a better rate from the insurer, which, in effect, gives them more money to use during the year.
Legitimately injured workers are discouraged and intimidated into not reporting their injuries . . .
Insurance fraud is endless. They often misinform the injured workers of their rights, attempt to trick
them into signing away their rights without their knowledge, deny/delay approval, deny/delay pay-
ment just to work ‘the float.’ We also see high numbers of unpaid benefits as well as inaccurate benefit
notices. In some states, insurance companies have far outspent investigation of worker fraud than any
other fraud in the Workers’ Compensation system. A Texas study of Worker’s Compensation fraud
conducted by the state’s Research and Oversight Council on Workers’ Compensation found that, ‘of
the 4,077 cases of claimant fraud that were investigated, only 18 were prosecuted criminally.’ The
report concluded: ‘It is clear that more resources should be spent fighting the most expensive and
overlooked types of Workers’ Compensation fraud: employer premium and health care provider
fraud.’ However, they do spend plenty of money to pay for video surveillance, private investigators,
and a large staff—all of which is used to intimidate the injured worker out of their rights. When will
they begin to seriously investigate insurance fraud? (pp. 295–6).
Although findings of different qualitative studies are difficult to aggregate, the thematic simi-
larities among several of them are quite striking. In a very general way, the participants in the
studies clearly indicated that they were suffering financially and emotionally, and that they felt
victimized.
The studies do not provide any direct information about reasons why IWs do not recover from
their injuries as rapidly or completely as private pay patients. (The interviewees in these studies
communicated the sense that they were doing the best they could under very adverse circum-
stances.) However, given workers’ descriptions of the severe stress they experienced in relation to
their work injures, it is plausible to infer that their slow recoveries were products of stress. This
hypothesis is consistent with a vast literature demonstrating adverse effects of emotional stress on
health (Finestone et al. 2008; Schmidt et al. 2008).
A key problem in interpreting the many qualitative studies on IWs is that the study participants
may not have been representative of IWs. For example, Beardwood et al. (2005) selected IWs
who were going through training to become peer researchers; Crombie et al. (2003) informally
gathered a cohort of physical therapists with work injuries; Cacciacarro and Kirsh (2006) recruited
participants from an IW support group. It is certainly possible that the IWs who agreed to
participate in the studies were particularly bitter ones who were not representative of IWs as
a whole.
In principle, this possibility could be tested via systematic research on the satisfaction of IWs
with WC systems. Such research could provide insights into whether the negative attitudes
expressed by participants in the qualitative studies were representative of those of IWs in the
aggregate, or reflected the views of a disgruntled minority. Unfortunately, very little research has
been done on this issue. In one population based study that addressed attitudes of IWs toward the
claim process (Wickizer et al. 2004a, 2004b), investigators found that while approximately 70%
of the 804 study participants were very satisfied with the medical care they received, only 42%
were very satisfied with the handling of their claim. These results suggest that IWs were modestly
satisfied with administrative aspects of their claims. It should be noted that the 42% ‘very satis-
fied’ rate may overstate the degree of satisfaction that would be expressed by IWs with protracted
claims. A key issue is that only 60% of the 804 study participants went on work disability because
of their injuries, and only 13% were work disabled more than 6 months. The participants with
prolonged work disability were significantly less satisfied with their care than ones with no or
short term work disability. This subgroup of 13% was probably more comparable to participants
in the qualitative studies than was the cohort as a whole.
If the results of qualitative studies on IWs are compared to results from more rigorous research
done on moral hazards, malingering, and secondary gain, a striking mismatch becomes obvious.
Moral hazard/malingering/secondary gain studies suggest that IWs are doing something deceitful
in order to get financial gain. The behaviour of IWs might be unconscious, but nonetheless it is
somewhat unseemly. The common thread in this research is that IWs (deliberately or uncon-
sciously) drag their feet. In contrast, the qualitative studies document that IWs experience severe
emotional distress and a sense of victimhood as they interact with WC systems. Although these
studies do not address the issue of delayed recovery, they suggest that chronic, severe emotional
distress contributes heavily to the delayed recoveries of IWs.
There is no simple way to integrate these two lines of research into a coherent theory that
explains the delayed recovery of IWs. At first glance, it might seem appropriate to accept conclu-
sions from moral hazards/malingering/secondary gain research, since it is more rigorous and
more extensive than qualitative research on IWs. However, such a conclusion overlooks a funda-
mental point. As Hammer (2000) pointed out, it is quite possible that participants in WC other
than IWs are engaged in unseemly behaviour. But essentially no systematic research has been on
supervisors or owners of businesses, or on WC claims managers. Thus, the unflattering results of
studies on IWs reflect in part the fact that the behaviour of IWs has been carefully scrutinized by
numerous investigators. Representatives of the business community and WC systems have simply
not been scrutinized.
The different pictures of IWs provided by qualitative studies vs. more formal studies on moral
hazards/malingering/secondary gain highlight basic problems in untangling the factors that lead
to delayed recovery among IWs. One basic problem is that the clinical recovery from an injury is
affected not only by objectively measurable medical features of the injury, but also by a variety of
host and environmental factors that are more difficult to measure. Similarly, the ability of an IW
to return to work (or the probability that he/she will return to work) is influenced by a variety of
factors other than strictly objective medical data about the worker’s initial injury or response to
treatment. Thus, while it is administratively convenient for WC carriers to focus on objective
medical data when they make decisions about benefits (Robinson 2005), reliance on these data
vastly oversimplifies the processes involved in recovery from an injury. Subjective data from
IWs—such as reports of continuing pain—can give a more realistic picture of the recovery proc-
ess, but the reports can be manipulated by IWs in order to prolong their claims or receive unwar-
ranted settlements. Physicians, claims managers, and others given the responsibility to make
decisions about IWs have the difficult task of determining the veracity of statements by IWs about
their clinical status and ability to work. Sceptics are likely to see evidence of foot dragging or out-
right fraud in these workers; advocates for IWs are likely to perceive the IWs as doing the best they
can under very difficult circumstances. Evidence to date suggests that determining the ‘true
motives’ of these IWs is difficult for the simple reason that the interpersonal appraisals needed to
make such determinations are fraught with error (Robinson 2010).
18.4.3 Other possible causes of the negative effect of WC

Lack of access to care
Some observers have noted that because of the poor reimbursement offered by WC carriers and
the high burden of paper work imposed on physicians who treat IWs, experienced physicians
with well established practices may refuse to treat these patients (Bellamy 2001; Pourat 2007).
Thus, one possible explanation for poorer outcomes of IWs compared to private pay patients is
that IWs do not receive comparable care. Although this is certainly a possibility, it would not
account for the negative effect of WC documented by Harris et al. (2005), since in most of the
studies in this review, the same physicians treated IWs and private pay patients.
Involvement of unions
There is evidence that workers are more likely to file claims if they are union members (Morse
et al. 2003). The authors of this study suggested several hypotheses to explain this effect, includ-
ing: (1) high filing rates could be spurious byproducts of the fact that industries and occupations
with high injury rates are relatively likely to be unionized; (2) unions might create a culture in
which filing work injuries is views as legitimate; (3) unions might protect workers who file claims
from retaliation.
Involvement of attorneys
Involvement of an attorney may influence the course of a claim. In various studies, attorney
involvement has been shown to be negatively associated with outcomes of a WC claim (Bernacki
2004; Katz et al. 2005; Bernacki et al. 2007; LaCaille et al. 2007; Bernacki and Tao 2008; Welch and
Boden 2009). There is a problem of cause and effect, however. In particular, it is quite possible
that IWs seek assistance from an attorney because they are not recovering from their injury, or are
having adversarial interactions with their WC carrier. From this perspective, attorney involve-
ment should be viewed as a marker of failure to recover rather than as a cause of failure to recover
(Welch and Boden 2009). Consistent with this hypothesis, Wickizer et al. (2004b) found that in a
cohort of 804 IWs, 46 retained an attorney. The median time between onset of the claim and
retention of an attorney was 288 days. Thus, IWs tended to retain attorneys only after their claims
had been opened for an extended period of time.
The need to prove incapacitation

Norton Hadler has written extensively on contradictions within the WC system (Hadler 1995,
2005; Hadler et al. 2007). He has not articulated an easily summarized list of the ways in which
WC negatively impact IWs, but one dynamic he has highlighted involves a paradox that individu-
als face as they interact with disability systems. He points out that it is difficult for individuals with
medical problems to recover if, in order to maintain medical or disability benefits, they must
repeatedly convince others that they are incapacitated (Hadler 1996). Although this assertion is
difficult to prove in relation to WC, it is certainly consistent with research demonstrating that
when people play various roles, they take on attitudes and beliefs consistent with these roles
(Knapp et al. 1989; Smith et al. 1995). In the case of medical disorders, a patient who is forced by
the WC system to play the role of a disabled person is likely to adopt the attitudes, beliefs, and
behaviours of such a person.
18.5 Moral hazards: another look

As noted above, research on moral hazards has focused on the ways in which insurance benefits
influence the behaviour of beneficiaries. This narrow focus obscures the fact all participants in
WC can be influenced by moral hazards.
Employers may have overt or covert policies that influence claims behaviour. Injuries may not
be reported in an attempt to avoid sanctions, increased insurance rates and increased healthcare
costs. Workers who are deemed ready to return to partial or limited employment may be denied
such opportunities by ‘company policy’ and thereby have their disabilities prolonged. Various
forms of retaliation can follow claim for injury submission. In short, employers are just as suscep-
tible to moral hazards influencing their behaviours as are the workers. We live in a capitalist
system whose underlying principle is ‘money motivates behaviour’.
Workers Compensation plan administrators are not immune to moral hazards. It has been our
observation that plan administrators are more likely to identify with and support the employers’
viewpoints than the workers’. They may support a range of policies that work to the disadvantage
of IWs, including delaying treatment authorizations and imposing paperwork hurdles. Also, they
may use a simple but effective method to encourage claims managers to close claims quickly. The
claims managers continue to be assigned new claims regardless of rate at which they close claims.
Thus, claims managers who do not move aggressively to close claims can be inundated by a
mountain of new and old claims.
Physicians are also susceptible to moral hazards when a fee for service system gives them unlim-
ited opportunities to provide services to injured workers. On the other hand, prepaid plans have
the opposite effect of pushing providers to offer services as limited as possible. The effects of
incentives often play out in forensic settings, where each physician expert swears that his testi-
mony is the truth and nothing but the truth, but always reinforces the viewpoint of the attorney
who has hired him.
Attorneys involved in WC have incentives that almost certainly influence their behaviour. Ones
representing IWs typically work on contingency basis. If, for example, an attorney’s client is
awarded a pension by the WC carrier, the attorney gets a proportion of the award. Thus, attorneys
representing IWs have an incentive to emphasize how disabled their clients are. In contrast, attor-
neys representing WC carriers have incentives to minimize the exposure of their clients by doing
what they can to minimize the benefits that IWs receive.
One might think that researchers are above the fray, and impartially seek to determine ‘the
truth’ about the WC system. Unfortunately, new research has convincingly demonstrated that
authors of scientific, economic, and philosophic papers might be influenced by the sources of
funding for their research and writings. One obvious influence involves the choice of topics for
research projects. As noted above, research funded by insurance companies and WC agencies
focuses heavily on the behaviour of IWs. It rarely studies the behaviour of employers, and almost
never examines the adverse effects that the companies and agencies might have on the recovery of
IWs. At a more subtle level, there is evidence that the research results reported by investigators are
statistically linked to the sources of support for their research. This effect has been demonstrated
in relation to studies on the effectiveness of drugs (Bekelman et al. 2003; Bhandari et al. 2004;
Tereskerz and Moreno 2005; Fenton et al. 2007) For example, Lexchin et al. (2005) reviewed
30 publications that reported results from drug studies funded by pharmaceutical companies vs.
ones with other sources of funding. The studies in the review considered a wide range of drugs,
including tacrine, clozapine, third-generation oral contraceptives, erythropoietin, antidepres-
sants, and topical glucocorticosteroids. The authors found that: ‘Studies sponsored by pharma-
ceutical companies were more likely to have outcomes favouring the sponsor than were studies
with other sponsors (odds ratio 4.05; 95% confidence interval 2.98 to 5.51; 18 comparisons).
None of the 13 studies that analysed methods reported that studies funded by industry were of
poorer quality’ (p. 1167). Concerns about the potential for bias as a result of industry sponsorship
have recently reached the political arena, with calls for a sunshine law in the US Senate (Grassley
2008), and requirements by journal editors that authors of papers provide information about
sources of financial support for their work (Fontanarosa 2005).
Although the studies on industry sponsorship cited above deal with medicines and medical
devices, they have clear implications for research on WC. Specifically, they suggest that research
sponsored by WC carriers is likely to be biased in ways that are favourable toward the carriers and
unfavourable toward IWs.
Thus, all of the players in the WC system are influenced by the systems in which they live
and work and play. Workers are not uniquely influenced by the opportunities provided by a
moral hazard. Studies should encompass the full range of such hazardous behaviours, includ-
ing the behaviours of employers, WC claims managers, and other participants in the WC sys-
tem. Moreover, researchers cannot be considered immune to the moral hazard effect. Thus,
future research on the effects of incentives in the WC system should not only be more inclusive
with respect to the topics investigated, but also should be funded by agencies other than WC
carriers.
18.6 Conclusions: implications for social policy

An extensive body of literature supports the conclusion that WC systems exert an adverse effect
on recovery from work injuries. As discussed above, several possible explanations for this negative
effect have been proposed. A key question is: Does the research to date provide insights into how
WC systems might be changed so as to encourage recovery among IWs?
Unfortunately, attempts to improve WC on the basis of the research discussed in this chapter
run into at least 3 problems. First, different participants in and observers of the WC systems have
profoundly different perspectives regarding the reasons why IWs demonstrate delayed recoveries.
On one side of this divide are industry supporters, who in effect argue that IWs are at best careless
and at worst fraudulent. In sharp contrast, advocates for IWs portray these people as victims and
emphasize the suffering that they experience as they try to recover from work injuries. They
express the view that there are abusers in the WC system, but that the abusers are employers and
claims managers rather than employees. The key point here is that questions about causes of the
negative effect of the WC system are anything but questions simply of science and research.
Rather they are fundamentally intertwined with perceptions of social justice and ongoing conflicts
between labour and management.
Second, it is inherently difficult to develop large scale programmes to support IWs that are
efficient and sensitive to the emotional needs of individuals whose livelihoods are threatened
by work injuries. Large, bureaucratic systems have inherent inefficiencies. In this regard, it is
noteworthy that although several observers have criticized the WC systems in the United States
and other Western countries, they have generally not offered concrete ideas about how to improve
the systems.
Finally, research on moral hazards suggests an inherent contradiction in the social goals of WC
systems. The contradiction can be seen when one ponders the question: How can a WC system
provide support to IWs that eases their suffering without creating incentives for them to take
advantage of this support?
Although the above points suggest that dramatic improvements in WC may be hard to achieve,
there are certainly modest steps that could be taken. One important step would be to have research
that is funded by agencies other than WC carriers, and that examines the behaviour of all partici-
pants in the WC system. Second, it would be interesting to examine the breadth of the negative
effects of WC. For example, it would be interesting to see if the negative effect of WC docu-
mented by Harris et al. for surgical treatments also applies to rehabilitative treatments and other
conservative therapies for various injuries.
Thirdly, it would be helpful to study experiments of nature in which fundamental principles of
WC have been altered. For example, during the 1970s New Zealand’s national health system
eliminated the distinction between work-related and non-work-related medical conditions
(McNaughton et al. 2000). It would be interesting to examine the impact of this change in social
policy on the recovery from injuries and illnesses among citizens of New Zealand.
Finally, more systematic research is needed on the perceptions of IWs regarding WC systems.
As noted above, Wickizer et al. (2004a) studied satisfaction of IWs in Washington State with their
medical treatment and their interactions with the WC system. This kind of research should be
replicated in other WC systems. At one extreme, it might demonstrate that from the perspective
of IWs, WC systems usually function quite well. In that event, administrators of WC systems
would be justified in continuing policies that already exist. Alternatively, research might confirm
the negative attitudes voiced by IWs in the qualitative studies described above. If these attitudes
turn out to be widespread among IWs, rather than limited to a few particularly bitter ones, it
would be incumbent upon WC carriers to look carefully at their procedures, and modify them so
as to become more ‘customer-friendly’.
References
Analysis of workers’ compensation laws. (1996). Washington, DC: U.S. Chamber of Commerce, 1996.
Aronoff GM, Mandel S, Genovese E, et al. (2007). Evaluating malingering in contested injury or illness.
Pain Practice 7 (2), 178–204.
Atlas SJ, Chang Y, Kammann E, Keller RB, Deyo RA, Singer DE. (2000). Long-term disability and return
to work among patients who have a herniated lumbar disc: the effect of disability compensation. J Bone
Joint Surg Am 82(1), 4–15.
Atlas SJ, Tosteson TD, Hanscom B, et al. (2007). What is different about worker’s compensation patients?
Spine 32 (18), 2019–26.
Auth J. (2005). Employers must be alert to comp fraud. Occup Health Saf 74(5), 12.
Bartley MU. (1994). Unemployment and ill health: understanding the relationship. J Epidemiol Community
Health 48(4), 333–7.
Beardwood BA, Kirsh B, Clark NJ. (2005). Victims twice over: perceptions and experiences of injured
workers. Qual Health Res 15(1), 30–48.
Bekelman JE, Li Y, Gross CP. (2003). Scope and impact of financial conflicts of interest in biomedical
research. JAMA 289, 454–65.
Bellamy RE. (2001). Workers’ compensation: the patient, the physician, and the system. J Bone Joint Surg
Am 83-A(5), 781–3
Bernacki EJ, Tao XG. (2008). The relationship between attorney involvement, claim duration, and workers’
compensation costs. J Occup Environ Med 50(9), 1013–18.
Bernacki EJ, Yuspeh L, Tao X. (2007). Determinants of escalating costs in low risk workers’ compensation
claims. J Occup Environ Med 49(7), 780–90.
Bernacki EJ. (2004). Factors influencing the costs of workers’ compensation. Clin Occup Environ Med 4(2),
v–vi, 249–57.
Bhandari M, Busse JW, Jackowski D, et al. (2004). Association between industry funding and statistically
significant pro-industry findings in medical and surgical randomized trials. CMAJ 170, 477–80.
Bigos SJ, Battie MC, Spengler DM, et al. (1992). A longitudinal, prospective study of industrial back injury
reporting. Clin Orthop Relat Res 279, 21–34.
Bureau of Labor Statistics. (2008). Workplace injuries and illnesses in 2007. Bureau of Labor Statistics.
http://www.bls.gov/iif/oshwc/osh/os/osnr0030.pdf. [Accessed 25 January 2009.]
Bureau of Labor Statistics. (2005). Distribution of Injuries and Illnesses, by Nature, 2005. http://www.bls.
gov/iif/oshwc/osh/os/osh05_28.pdf [Accessed 30 January 2009.)
Cacciacarro L, Kirsh B. (2006). Exploring the mental health needs of injured workers. Can J Occup Ther
73(3):178–87.
Cheadle A, Franklin G, Wolfhagen C, Savarino J, Liu PY, Salley C, Weaver M. (1994). Factors influencing
the duration of work-related disability: a population-based study of Washington State workers’
compensation. Am J Public Health 84(2), 190–6.
Cole DC, Mondloch MV, Hogg-Johnson S. (2002). Listening to injured workers: how recovery
expectations predict outcomes—a prospective study. CMAJ 166(6), 749–54.
Cromie JE, Robertson VJ, Best MO. (2003). Physical therapists who claimed workers’ compensation:
a qualitative study. Phys Ther 83(12), 1080–9.
Dembe AE, Boden LI. (2000). Moral hazard: a question of morality? New Solut 10(3), 257–79.
Dersh J, Polatin PB, Leeman G, Gatchel RJ. (2004). The management of secondary gain and loss in
medicolegal settings: strengths and weaknesses. J Occup Rehabil 14(4), 267–79.
Dooley D, Fielding J, Levi L. (1996). Health and unemployment. Annu Rev Public Health 17, 449–65.
Easton R. (1997). Screen patients quickly for workers’ comp fraud. Manag Care 6(4), 49–50.
Fenton JJ, Mirza SK, Lahad A, Stern BD, Deyo RA. (2007). Variation in reported safety of lumbar
interbody fusion: influence of industrial sponsorship and other study characteristics. Spine 32, 471–80.
Finestone HM, Alfeeli A, Fisher WA. (2008). Stress-induced physiologic changes as a basis for the
biopsychosocial model of chronic musculoskeletal pain: a new theory? Clin J Pain 24(9):767–75.
Fishbain D. (1994). Secondary gain concept: Definition problems and its abuse in medical practice. Am
Pain Soc J 3(4), 264–73.
Fishbain DA, Rosomoff HL, Cutler RB, et al. (1995). Secondary gain concept: A review of the scientific
evidence. Clin J Pain 11, 6–21.
Fishbain DA, Rosomoff HL, Cutler RB, Rosomoff RS. (1995). Secondary gain concept: A review of the
scientific evidence. Clin J Pain 11(1),: 6–21.
Fontanarosa PB, Flanagin A, DeAngelis CD. (2005). Reporting conflicts of interest, financial aspects of
research, and role of sponsors in funded studies. JAMA 294, 110–11.
M. Fricker M. (1997). , Widespread Fraud: Bogus Claim, The Press Democrat, December 7, 1997. [Cited in;
Dembe AE, Boden LI. (2000). Moral hazard: a question of morality? New Solut 10(3), 257–79.]
Fryer D, Fagan R. (2003). Toward a critical community psychological perspective on unemployment and
mental health research. Am J Community Psychol 32(1–2), 89–96.
Gabbe BJ, Cameron PA, Williamson OD, et al. (2007). The relationship between compensable status and
long-term patient ourtcomes following orthopaedic trauma. MJA 187(1), 14–7.
Gatchel RJ, Adams L, Polatin PB, et al. (2002). Secondary loss and pain-associated disability: theoretical
overview and treatment implications. J Occup Rehabil 12, 99–110.
Grassley C. (2008). Straight talk with . . . Charles Grassley. Interview by Meredith Wadman. Nat Med
14(10):1006–7.
Hadler NM, Tait RC, Chibnall JT. (2007). Back pain in the workplace. JAMA 298(4), 403–4
Hadler NM. (1995). The disabling backache. An international perspective. Spine 20(6), 640–9.
Hadler NM. (1996). If you have to prove you are ill, you can’t get well. The object lesson of fibromyalgia.
Spine 21(20), 2397–400
Hadler NM. (2005). The health assurance-disease insurance plan: harnessing reason to the benefits of
employees. J Occup Environ Med 47(7), 655–7
Hammer A. (2000). Word from the front lines: injured worker organizations speak out. New Solut
10, 293–9.
Harkness EF, Macfarlane GJ, Nahit ES, Silman AJ, McBeth J. (2003). Risk factors for new-onset low back
pain amongst cohorts of newly employed workers. Rheumatology 42(8), 959–68.
Harris I, Mulford J, Solomon M et al. (2005). Association between compensation status and outcome after
surgery: A meta-analysis. JAMA 293 (13), 1644–52.
Hunt HA. (2003).Benefit adequacy in state workers’ compensation programs. Soc Secur Bull 65(4), 24–30.
Katz JN, Amick BC 3rd, Keller R, Fossel AH, Ossman J, Soucie V, Losina E. (2005). Determinants of work
absence following surgery for carpal tunnel syndrome. Am J Ind Med 47(2), 120–30.
Kay N, Morris-Jones H. (1998). Pain clinic management of medico-legal litigants. Injury 29, 305–8.
Kennedy F. (1946). The mind of the injured worker: its effect on disability periods. Compensation Med
1946; 1, 19–24.
Kivimaki M, Head J, Ferrie JE, et al. (2003). Sickness absence as a global measure of health: evidence from
mortality in the Whitehall II prospective cohort study. BMJ 327(7411), 364–8.
Knapp, Patricia A.; Deluty, Robert H. (1989). Relative effectiveness of two behavioral parent training
programs. J Clin Child Psychol 18(4), 314–22.
LaCaille RA, DeBerard MS, LaCaille LJ, Masters KS, Colledge AL. (2007). Obesity and litigation predict
workers’ compensation costs associated with interbody cage lumbar fusion. Spine J 7(3), 266–72.
Leino-Loison K, Gien LT, Katajisto J, Välimäki M. (2004). Sense of coherence among unemployed nurses.
J Adv Nurs 48(4), 413–22.
Lexchin J, Bero LA, Djulbegovic B, Clark O. (2005). Pharmaceutical industry sponsorship and research
outcome and quality: systematic review. BMJ 326, 1167–70.
Lippel K. (2007). Workers describe the effect of the workers’ compensation process on their health:
a Québec study. Int J Law Psychiatry 30(4–5), 427–43.
Loeser JD, Henderlite SE, Conrad DE. (1995). Incentive effects of workers’ compensation benefits:
A literature synthesis. Med Care Res Rev 52(1):34–59.
MacEachen E, Kosny A, Ferrier S. (2007). Unexpected barriers in return to work: lessons learned from
injured worker peer support groups. Work 29(2), 155–64.
Mah DR. (1998). Reducing workers’ compensation fraud: a deterrent approach. Occup Med 13(2), 429–38.
McNaughton HK, Sims A, Taylor WJ. (2000). Prognosis for people with back pain under a no-fualt
24-hour-cover compensation scheme. Spine 2000; 25, 1254–8.
Mendelson G. (1988). Psychiatric aspects of personal injury claims. Springfield, IL: Charles C. Thomas.
Morse T, Dillon C, Warren N. (2000). Reporting of work-related musculoskeletal disorder (MSD) to
workers Compensation. New Solut 10(3), 281–92.
Morse T, Punnett L, Warren N, et al. (2003). The relationship of unions to prevalence and claim filing for
work-related upper-extremity musculoskeletal disorders. Am J Ind Med 2003; 44, 83–93.
Murphy GC, Athanasou JAT. (1999). The effect of unemployment on mental health. J Occup
Organisational Psychol 72, 83–100.
Mustard C, Hertzman C. (2001). Relationship between health services outcomes and social and economic
outcomes in workplace injury and disease: data sources and methods. Am J Ind Med 40(3), 335–43.
Overland S, Glozier N, Henderson M, et al. (2008). Health status before, during and after disabilkty
pension award: the Hordaland Health Study (HUSK). Occup Environ Med 65, 769–73.
Phoney worker’s comp. claim?: smile, you may be on video (2007). Nurs Law Regan Rep 47(11), 4.
Pourat N, Kominski G, Roby D, Cameron M. (2007). Physician perceptions of access to quality care in
California’s workers’ compensation system. J Occup Environ Med 49(6), 618–25.
‘Retirees’ disability epidemic’. (2008). New York Times 21 September, A1.
Reville RT, Seabury SA, Neuhauser FW, Burton JF, Greenberg MD. (2005). An evaluation of California’s
permanent disability rating system. Santa Monica, CA: RAND Corporation, MG-258.
Robinson J.P., Rondinelli R.D., & Scheer, S.J. (1997). Industrial rehabilitation medicine 1: Why is
industrial rehabilitation medicine unique? Arch Phys Med Rehabil, 78(3S), S3–S9.
Robinson, J.P. (2005). Disability management in primary care. In B. McCarberg & SD Passik (Ed.). Expert
guide to pain management. Philadelphia, PA: American College of Physicians.
Robinson JP, Tait RC. (2010). Disability evaluation in painful conditions. In JC Ballantyne JP Rathmell,
SM Fishman (Eds.) Bonica’s Management of Pain (4th edn.). Philadelphia, PA: Lippincott Williams &
Wilkins.
Scherzer T, Rugulies R, Krause N. (2005). Work-related pain and injury and barriers to workers’
compensation among Las Vegas hotel room cleaners. Am J Public Health 95(3), 483–8.
Schmidt MV, Sterlemann V, Müller MB. (2008). Chronic stress and individual vulnerability. Ann N Y Acad
Sci 1148, 174–83.
Shannon HS, Lowe GS. (2002). How many injured workers do not file claims for workers’ compensation
benefits? Am J Ind Med 42, 467–73.
Smith, Mike U.; Katner, Harold P. (1995). Quasi-experimental evaluation of three AIDS prevention
activities for maintaining knowledge, improving attitudes, and changing risk behaviors of high school
seniors. AIDS Educ Prev, 7(5), 391–402.
Soeker MS, Wegner L, Pretorius B. (2008). I’m going back to work: back injured clients’ perceptions and
experiences of their worker roles. Work 30(2), 161–70.
Strunin L, Boden LI. (2004). The workers’ compensation system: worker friend or foe? Am J Ind Med 45,
338–45.
Tereskerz PM, Moreno J. (2005). Ten steps to developing a national agenda to address financial conflicts of
interest in industry sponsored clinical research. Account Res 12, 139–55.
Vahtera J, Pentti J, Kivimaki M. (2004). Sickness absence as a predictor of mortality among male and
female employees. J Epidemiol Community Health 58(4), 321–6.
Wallman T, Wedel H, Johansson S, et al. (2006). The prognosis for individuals on disability retirement. An
18-year mortality follow-up study of 6887 men and women sampled from the general population.
BMC Public Health 6(1), 103.
Welch LS, Boden LI. (2009). Attorney involvement, claim duration, and workers’ compensation costs.
J Occup Environ Med 51(1), 1–2.
Wickizer TM, Franklin G, Fulton-Kehoe D, Turner JA, Mootz R, Smith-Weller T. (2004a). Patient
satisfaction, treatment experience, and disability outcomes in a population-based cohort of injured
workers in Washington State: implications for quality improvement. Health Serv Res 39(4 Pt 1),
727–48.
Wickizer TM, Franklin G, Turner J, Fulton-Kehoe D, Mootz R, Smith-Weller T. (2004b). Use of attorneys
and appeal filing in the Washington State workers’ compensation program: does patient satisfaction
matter? J Occup Environ Med 46(4), 331–9.
Williams CA. (1991). An international comparison of workers’ compensation. Boston, MA: Kluwer
Academic Publishers.
Chapter 19
Work-Related Risk Factors for Transition

to Chronic Back Pain and Disability
William S. Shaw, Glenn S. Pransky, and Chris J. Main
19.1 Introduction
Low back pain (LBP) is a common cause of suffering and disability in workers. As many as 85%
of workers will experience at least one episode of back pain severe enough to limit their ability to
work. Yet, most do recover, returning to normal function at work and at home, although bother-
some symptoms may persist or intermittently recur. Among those recovered from an acute
episode of work-related LBP, the prevalence of recurrent pain over the following year is 50–60%,
but recurring sickness absence from work occurs in only 12–15% of cases (Marras et al. 2007;
Wasiak et al. 2003). Approximately 10% of those with an acute episode of disabling back pain
progress to develop long-term problems that include both significant pain and limitations in their
ability to function at work and at home (Nachemson et al. 2000; Waddell et al. 2002). This small
group accounts for the majority of LBP-related suffering and associated costs in working-age
persons. In the USA, 18% of workers lose an estimated 149 million days of work due to LBP annu-
ally (Guo et al. 1995). There are comparable figures in other industrialized countries (van Tulder
et al. 1995). This significant burden has attracted the attention of workers, employers, providers,
insurers, and governments, who are seeking better ways to prevent pain and disability in those
who would otherwise be contributing to economic productivity.
The workplace has a number of features of specific relevance to LBP occurrence, treatment, and
consequences. First, it is important to distinguish chronic pain (persistent symptoms) from
chronic disability (inability to function at work), because these outcomes are often independent
(Volinn et al. 1988). Indeed, in countries with a workers’ compensation system, the distinction
between injuries caused by work and those of non-occupational origin is of legal necessity,
although this distinction is not always supported by scientific evidence. LBP that is persistent or
intermittent, but does not result in significant pain, functional limitations, or work disability,
may be troublesome but of much less importance to workers as long as they can maintain employ-
ment. Thus return-to-work (RTW) in those off work may be the most salient outcome for work-
ers with acute LBP, as these outcomes are directly linked to key social roles and economic status
for working-age adults (Cats-Baril and Frymoyer 1991).
It is difficult to quantify the precise relationship between LBP and work since working adults
with LBP may encounter issues unique to the nature of their work and workplace. LBP-related
functional limitations may lead to inability to do some jobs, but they may have little or no impact
on other jobs. Similarly, on-site healthcare services, employer accommodations, and co-worker
support can enable rapid recovery and RTW, but there is huge variation in the assistance and
support available to injured workers. Finally, on a macro level, the involvement of a workers’
compensation or disability insurer may also facilitate RTW, but insurance systems can also lead
to legal conflicts that can prolong work disability.
Across disparate insurance and governmental benefit systems, a number of work-related factors
have been consistently associated with poor outcomes in LBP, including progression to chronic
disability, and these are the focus of this chapter. Even after controlling for a number of health,
psychosocial, and demographic variables, characteristics of work and the work environment are
significant predictors of continued symptoms, functional incapacity, and prolonged disability
(Shaw et al. 2001). These factors can be identified early on, before chronicity has developed. We
review the evidence supporting these workplace factors, appraise methods for identifying them
and explore the opportunity for subsequent interventions to prevent long-term, chronic disabil-
ity. Although research on workplace interventions is somewhat limited, enough information is
available to provide some well-supported recommendations to healthcare providers, employers,
and others. Nonetheless, a caveat is necessary: these work factors have been primarily studied in
relationship to workplace-relevant outcomes, commonly RTW, recurrent disability, and function
at work. Their association with other outcomes such as pain or overall quality of life tends to be
weaker (Franche et al. 2005; Hasenbring et al. 1994).
19.2 Workplace risk factors

There are now numerous cross-sectional, retrospective, and prospective cohort studies that have
examined the effects of various factors on the progression from acute to chronic LBP. The most
commonly supported disability risk factors include demographic variables, pain and pain beliefs,
perceptions of function and expected recovery, and workplace physical and psychosocial envi-
ronment (Waddell 2004). For the outcome of back disability, psychosocial variables (including
perceptions of the work environment) seem to be better overall prognostic indicators than either
demographic or clinical exam findings. Thus, individual and workplace psychosocial factors have
been highlighted as crucial in the development of long-term disability (Crook et al. 2002). These
factors reflect some of the greatest challenges facing workers with LBP—managing symptoms
while at work, overcoming the frustrations of having to ask for help, lacking the job flexibility to
modify physical demands, and failing to meet productivity demands (Shaw and Huang 2005).
To identify the most important workplace risk factors for prolonged LBP chronicity and disa-
bility, we searched for English language scientific reviews that summarized workplace risk factors
for LBP and disability, published since the year 2000. Studies of both work-related and non-work
related LBP were included. We identified six reviews that met these criteria. Most of the reviewers
focused on those factors supported by prospective cohort studies. Some reviews were limited to
workplace risk factors, but most included both individual and workplace factors. As these reviews
have applied different methods for synthesizing results, there are some conflicting conclusions.
As examples, job satisfaction was supported in four of six reviews, job stress and social support
were supported in some reviews and not by others; and only one review took magnitude of effect
(relative risk) into account (Hartvigsen et al. 2004).
We compiled a list of the risk factors supported by the scientific evidence in at least one review,
resulting in the 24 variables identified in Table 19.1. For convenience we have grouped these
under four types of factors: physical work demands, social climate at work, perceptions about
health and work, and workplace disability management. Each of the four groupings is described
below:
Physical work demands. The five job physical demand factors all relate to the likelihood that a
worker with diminished capacity due to LBP will be unable to meet significant job demands. The
advantage of jobs in the public sector may relate to larger workforces where additional help is
more readily available; the opposite is true for jobs where a worker has to drive to a remote loca-
tion, and thus may often be working alone. The availability of modified duty can address the
WORK RISK FACTORS FOR CHRONIC LBP 379
Table 19.1 Variables described and supported as modifiable risk factors for chronic LBP disability in
recent literature reviews (2000–2005)
Factors Hartvigsen Crook Hoogendoorn Linton Shaw Steenstra

et al. (2004) et al. et al. (2000) (2001) et al. et al.
(2002) (2001) (2005)
Workplace—physical demands
Fast work pace ✓ ✓ ✓
Heavier physical demands ✓ ✓
Work demands > capacity ✓ ✓
Private industry (versus public) ✓
Driving (as majority of job) ✓
Workplace—social/managerial
Social support or dysfunction ✓ ✓ ✓ ✓
Supervisor support ✓
Lack of control ✓ ✓ ✓ ✓ ✓
Short job tenure ✓ ✓
Conflict at work ✓
Inability to take breaks at will ✓
Workplace—worker perceptions
Job dissatisfaction ✓ ✓ ✓ ✓
Monotonous work ✓ ✓
Job stress ✓ ✓ ✓
Belief work is dangerous ✓
Emotional effort of work ✓
Belief should not work with pain ✓
Work-related disability issues
WC claim (vs. non-wk related) ✓ ✓
Prior work disability claim ✓
Discouraging claim reporting ✓
Delayed report of injury ✓ ✓
Lower quality immediate medical ✓
care
Lack of modified duty ✓ ✓ ✓
Higher wage replacement rate ✓ ✓
Notes: ✓=review found at least moderate evidence supporting this risk factor.
RTW barrier of high work demands, and thus potentially counteract its negative effect (Weir and
Nielson 2001). While there is evidence of an interaction between job demands and psychosocial
work environment in the development of musculoskeletal symptoms (Devereux et al. 1999;
Devereux et al. 2002), few prospective studies of RTW outcomes have tested or reported such
interactions.
Social climate at work. Social or managerial issues may influence disability outcomes in several
ways. Workers who experience low levels of social support at work, from co-workers or supervi-
sors, may be concerned that no one will look after their interests after they return to work. Jobs
that are structured so that workers have little control, or an inability to take breaks, may reflect
adverse or unsupportive labor-management relationships. Conflicts at work might be a result of
poor social relationships. Those with short job tenure may not have had adequate time or oppor-
tunity in the workplace to develop supportive relationships. Some have theorized that lack of
social support leads to more anxiety and arousal in the context of LBP, with higher levels of pain
and dysfunction, and subsequent delays in RTW (Brouwer et al. 2009; Marhold et al. 2002) (see
also Chapter 17).
Perceptions about health and work. Worker perceptions are important, as they can all contribute
to a negative attitude towards the workplace, and subsequent reluctance to pursue RTW. Jobs
that are perceived as stressful or monotonous, emotionally exhausting, unsatisfying or dangerous,
may be undesirable even before onset of low back pain; thus, workers in these positions may
actively seek other alternatives rather than return to the same position. A worker’s beliefs that he
or she should not return to work until completely pain-free may reflect some of these underlying
perceptions about the job (Linton 2001).
Workplace disability management. The last category comprises policies about low back pain in
the workplace- how it is identified, treated, and how the disability consequences are managed.
Workplaces that are most successful in minimizing LBP disability usually have several of the
following characteristics: (1) encouraging workers with LBP concerns to come forward; (2) a
systematic approach that includes supervisor support and access to evidence-based initial care;
(3) early attention to disability issues through communication and offer of alternate duty to avoid
disability; and (4) appropriate organization of economic incentives (for workers and supervisors)
to insure that it’s worthwhile for workers to come back to work as soon as possible (Westmorland
et al. 2005).
19.3 Screening for workplace factors

The evidence-based workplace factors listed above, as well as additional items (both work-related
and non-occupational factors), are included in several available clinical screening tools. Some
tools also guide the interview and evaluation process, to help clinicians find out more about spe-
cific occupational situations, and provide information that can guide individualized efforts to
address these issues. We have identified seven representative approaches including the Örebro
Musculoskeletal Pain Screening Questionnaire (ÖMPSQ) (Linton and Hallden 1998 ), the
Psychosocial Risk for Occupational Disability Instrument (PRODI) (Schultz et al. 2005) the Back
Disability Risk Questionnaire (BDRQ) (Shaw et al. 2005), the ‘blue flags’ screening process (Shaw
et al. 2009), the Work Disability Diagnosis Interview (WoDDI) (Durand et al. 2002), the Obstacles
to Return-to-work Questionnaire (ORQ) (Marhold et al. 2002), and the Participatory Ergonomics
(PE) approach first described by Loisel and colleagues (Loisel et al. 2001a).
The approaches listed above differ in method, content, and specific objectives. Some were
designed to provide a quantification of disability risk in research cohorts, whereas others were
designed as practical tools to guide clinical interviewing. Although brief patient questionnaires
and semi-structured clinical interviews have been the primary mode of screening patients for risk
of chronic disabling LBP, detailed assessment can be more informative. The first three approaches
are questionnaires exclusively based on patient self-report. The ÖMPSQ includes information on
job physical demands and satisfaction, and expectations about RTW and the safety of RTW. It is
the only one of these instruments that has been subjected to formal validation in multiple settings.
The PRODI includes more occupational factors, such as skill discretion, decision authority, job
security, co-worker/supervisor support, control, overall support, resources, RTW expectation,

and employer response to LBP. The BDRQ was designed for early screening of acute LBP, and
includes physical work demands, RTW expectations, availability of modified duty, delay in
reporting the LBP problem, and negative supervisor responses.
The other methods collect more detail through interviews and other assessments. In 1997,
Kendall and colleagues developed a system for early identification and management of clinical
and occupational risk factors (termed Yellow Flags) for unnecessary/prolonged disability in LBP
(Kendall 1999). This list of specific questions was designed to prompt healthcare practitioners to
identify persons at high risk for prolonged disability and to explore these factors. This approach
has been re-conceptualized into blue flags (worker perceptions of an adverse work environment)
and black flags (objective measures of employment conditions, physical job demands, and insur-
ance system characteristics) (Shaw et al. 2009; Main 2002). The ORQ includes questions on dif-
ficulties returning to work, physical workload and perceived harmfulness, social support, worries
about sick leave, work satisfaction, and RTW expectations. The WoDDI includes both self-report
and a semi-structured diagnostic interview to address job satisfaction, work history, occupation
and industry type, prior attempts to RTW, absence duration, job demands, ergonomic risk
factors, work schedule, job control, environmental conditions, diversity of work tasks, and work-
ing relationships. The participatory ergonomics approach advocated by Loisel incorporates the
WoDDI as well as an on-site workplace evaluation by a physical therapist.
The choice of an ideal screening method for important workplace factors depends in part on
the intended application. Unfortunately, data on predictive performance are available only for
screening questionnaires. These can identify the probability of delayed RTW but by themselves
may not be able to specify the issues unique to an individual case (Durand et al. 2002). The reli-
ability, validity, and added value of a more detailed, in-depth but less structured interview are
difficult to quantify, but this approach may be necessary to guide clinical and workplace interven-
tions. We recommend a staged method of screening, involving multiple methods as needed
(questionnaire, interview, worksite visit), as an effective and efficient approach to identifying
obstacles to recovery in the workplace. Risk stratification questionnaires are more appropriate
during an acute stage, to identify risk factors early on. More detailed methods can be used when
screening is positive for workplace factors, or when there is a more prolonged absence from work
(subacute or chronic back pain and disability). This approach is consistent with other screening
methods in primary care that typically start with a patient self-report questionnaire (e.g. depres-
sion, sleeping disorders, irritable bowel syndrome) (Dejesus et al. 2007; Doghramji 2004).
19.4 Intervention strategies based on workplace risk factors

Several disciplines have contributed to the range of strategies now available to address workplace
risk factors, including occupational, rehabilitation and behavioural medicine, clinical and occu-
pational psychology, kinesiology, and occupational and physical therapy. Interventions have
focused on employer efforts to improve support from supervisors and co-workers, or to reduce
ergonomic exposure (Baril et al. 2003; Nicholas 2002; Shaw et al. 2006a). Acute care interventions
include coordinating medical care with RTW efforts, managing the use of various treatments,
modifying provider behaviour (Derebery et al. 2002; Loisel et al. 2005), and improving the readi-
ness of patients to RTW through physical conditioning, education, and often formal or informal
counselling (Gatchel et al. 2003; Linton and Nordin 2006). A recent review of workplace-based
return-to-work interventions led reviewers to conclude moderate or strong evidence for five cat-
egories of intervention: (1) early contact by the workplace with the worker; (2) an offer of job
accommodation; (3) contact between healthcare provider and workplace; (4) ergonomic work
site visit; and (5) presence of a RTW coordinator (Franche et al. 2005). These findings emphasize
the importance of communication among stakeholders and the need for job accommodations in
order to facilitate a successful RTW process. Matching patients to early interventions for LBP
based on individual risk factors may improve outcomes and the cost effectiveness of early inter-
vention (Waddell 2004). Recent studies have shown preliminary evidence of patient clusters by
risk factors in early stages of LBP (Boersma and Linton 2006; Shaw et al. 2009), providing further
support for a risk factor-based approach to early intervention. We have categorized these factors
into groups that connote different intervention strategies—some at an organizational level, oth-
ers at an individual level. A risk factor-based intervention strategy could provide treatments that
map directly onto measured risk factors for individual patients.
Workplace physical demands can be addressed through temporary work restrictions or modi-
fications, and efforts to improve workplace ergonomics, ideally through an inclusive, participa-
tory approach (Anema et al. 2003; Krause et al. 1998). In some instances, directed physical therapy
and exercise can have a role in restoring work ability (Lindström et al. 1992; Cooper et al. 1997).
Cognitive-behavioural strategies may help workers to cope with job strain. Social/managerial
factors suggest interventions to improve relationships and communication between supervisors
and workers, and the importance of flexibility in the workplace in the return to work process
(Shaw, Robertson et al. 2006). Some interventions rely on an ergonomics approach as a process
for safe reintegration, yet also include substantial efforts to improve communication and resolve
conflicts in the workplace (Arnetz et al. 2003).
Personal perceptions that work is dissatisfying, dangerous, or likely to cause re-injury may be
important mediators of back pain disability. These perceptions could result in part from a lack of
support, a mismatch with job requirements, job stress, or an undesirable physical or psychosocial
working environment. It’s unclear whether interventions should attempt to modify these beliefs
in the absence of other workplace modifications or coordination efforts. In most cases, these
perceptions are likely to be well-established over time, based on a realistic appraisal, so psycho-
logical approaches to alter these beliefs would seem daunting and misguided. The optimal strat-
egy would rectify the factors underlying these perceptions, guided by a more detailed and
broad-based inquiry (van Duijn et al. 2004). Interventions that have addressed workplace risk
factors (such as those featuring improved communication, accommodations, RTW policies, pro-
cedures, or support) may actually accomplish this, but studies have not evaluated the impact on
workplace perceptions. Thus it’s not clear whether a possible change in perceptions precedes
improvement in disability outcomes, or whether this is a secondary consequence of intervention.
Nevertheless, these perceptions might be of value to providers outside of the workplace. They can
help to identify significant RTW barriers that might not be amenable to change. This could lead
to more active efforts to secure safe accommodation, to alert the workplace for the need for more
positive communication and support, or to open a dialogue to suggest that a worker may want to
consider an alternative position or employer.
The factors that relate to the approach to health and disability in the workplace suggest the
strategy of a coordinated, positive, supportive and consistent approach to work disability (Shrey
and Hursh 1999). Elements can include encouraging early reporting of potentially disabling back
problems, treatment with selected providers, improved communication between providers and
employers, rapidly providing temporary accommodations if needed, and coordinating the return-
to-work process, with positive incentives to return to work (Bernacki and Tsai 2003).
19.5 Opportunities for future research and practice improvement

Although screening for workplace risk factors offers some promising opportunities, there are
considerable limitations in current knowledge and practice that limit application. Future research
on risk factor identification, targeted intervention, and impacts on outcomes can significantly
contribute to this field.
Improvement in theoretical rigour and consistency in risk factor research are two areas for
further development. The selection of prognostic variables, assessment methods, and choice of
outcome measures varies considerably. General acceptance of a framework that defines categories
of workplace variables that can be routinely incorporated across studies would be an important
advance. Several conceptual models of RTW have been advanced (Loisel et al. 2001b; Young et al.
2005; Schultz et al. 2007), but most risk factor studies rely on readily-available data, rather than
choosing factors based on a conceptual model. Others have selected variables and conducted
analyses based on various classifications—physical versus psychological (Steenstra et al. 2005;
Loisel 2007), modifiable versus non-modifiable (Shaw, Linton et al. 2006), locus of control
(worker, workplace, physician or insurer) (Loisel et al. 2001b), and individual-level versus work-
place-wide factors (Sullivan et al. 2005). In the review by Hartvigsen and colleagues, grouping
variables within four clusters (perception of work, organizational aspects, social support, and
stress) diminished the associations with outcomes, and resulted in their finding of mostly nega-
tive or inconclusive results (Hartvigsen et al. 2004). Thus, retaining the specificity of prognostic
variables seems important. Some of the variables that have been assessed in just one or two stud-
ies may actually be important, but lack wider support. Other problems include inconsistency in
statistical modelling techniques, varying durations of follow-up, and inclusion of different covari-
ates when testing the independent association of factors with outcomes (Pransky et al. 2001;
Linton et al. 2005).
Theoretically, interventions that alleviate causal risk factors should lead to effective disability
prevention, and risk factor screening should provide a reasonable basis for determining who is
likely to benefit from a specific intervention. However, variables commonly referred to as risk
factors for chronic LBP might be more appropriately labelled as ‘risk indicators’ or ‘risk markers’,
as most have not been subjected to rigorous criteria for concluding causation (e.g. demonstrating
dose-response relationships), and many variables overlap in their ability to predict outcomes.
Furthermore, there is some debate as to whether prognostic studies of LBP have been plagued by
significant measurement and confounding biases (Hayden and Cote 2006). While there is wide-
spread agreement that an early RTW reduces long-term disability risk, strategies for facilitating
an early RTW have varied considerably. The diversity of successful RTW interventions calls into
question the mechanisms underlying their effect. The complexities of workplace and worker factors
related to disability, the inevitability of significant changes over time at work, and the impossibil-
ity of identifying identical control situations have seriously hampered efforts to conduct rand-
omized, controlled trials. Instead, return-to-work interventions are sometimes evaluated only
with pre- and post-comparisons within single organizations, sometimes without the benefit of a
comparison non-intervention group. Though less rigorous, these studies can nevertheless pro-
vide an understanding of the key processes that lead to various outcomes (Kristensen 2005).
Significant logistical challenges exist for screening patients and assigning relevant treatments
based on prognostic factors, and there is little consensus as to which intervention strategies match
specific risk factors. One approach is to intervene only for those deemed ‘high risk’ based on a
summation of all empirical risk factors, including both modifiable and non-modifiable (e.g.
demographic) variables (Gatchel et al. 2003). Another approach is to develop an intervention that
targets a single risk factor, then screen patients for those high on that risk factor. For example,
Von Korff and colleagues (Von Korff et al. 2005) developed a brief fear reduction and physical
activation intervention for primary care patients with chronic LBP, then provided the interven-
tion only to those patients reporting significant activity limitations. The results showed improve-
ments in pain-related fear and activity limitations compared with a no-treatment control group
over a 2-year follow-up. Such studies have not been conducted in the acute or subacute stages of
LBP, when there may be even greater potential for reducing work absence.
One new idea in RTW workplace intervention trials is the possibility of bolstering informal
support and accommodation processes that occur naturally in the workplace setting and support-
ing or enhancing the self-management efforts of workers with pain or other health concerns
(Varekamp et al. 2009). Based on the high prevalence of chronic back pain (15–30%) reported by
working-age adults (Hardt et al. 2008; Watkins et al. 2008), most workers apparently find ways to
manage back pain on the job and obtain help from others without the need for any long sickness
absences or formal administrative processes, and this effective use of workplace leeway has been
described in qualitative studies (Durand et al. 2009). This variety of ‘workplace intervention’ is
actually planned and executed by workers themselves, without the need for formal managerial or
administrative processes that document and track an employer’s decision to ‘allow’ individual job
modifications. To promote these self-management processes, employers can designate ‘coping
contacts’ within the workforce, encourage and support worker self-management strategies, and
provide frontline supervisors and employees greater autonomy for organizing and modifying job
tasks (Varekamp et al. 2008; Werner et al. 2007). While the results of early studies look promising
(Werner et al. 2007), such workplace programs based on peer support and worker empowerment
require further study.
Despite some positive research findings, the dissemination of these intervention strategies to
clinical and workplace settings has been slow. Even among specialists in occupational medicine
and rehabilitation, many obstacles exist for intervening in the workplace, including barriers to
employer communication, limited information about job tasks and prospects for modifying
work, and employers unwilling or unable to provide modified or transitional work (Costa-Black
et al. 2007; Loisel et al. 2005; Pransky et al. 2004). Cost-effectiveness, an important factor in the
adoption of such experimental interventions, may be improved if interventions are directed only
to those with demonstrable need based on risk factor assessment. In particular, authors have
emphasized the need to reduce the growing list of workplace variables to a manageable set of core
factors, improve the accuracy and utility of patient screening, and develop effective and plausible
intervention strategies to address workplace concerns (Cedraschi and Allaz 2005; Feldman 2004;
Shaw et al. 2006b).
In summary, we know that workplace factors are of primary importance in determining chronic
work disability outcomes, they can be identified early on in the course of LBP, and certain work-
place interventions appear to be very successful at improving outcomes. Although methods to
screen for workplace factors are still under development, there are several sets of available ques-
tionnaires and interview guides that can help clinicians identify and address these problems.
A number of interventions have led to much better RTW outcomes, mostly targeting the response
of the workplace to a report of a disabling back condition, and the process of communication and
facilitation around RTW. Despite all of this knowledge, there is still much to be learned—how to
best identify and explore workplace factors, how to effectively address these factors, how to over-
come perceptions that workplace factors are unmodifiable, and how to bring along providers,
employers and others to optimize the long-term outcomes of workers with LBP.
References
Anema, J.R., Steenstra, I.A., Urlings, I.J., Bongers, P.M., de Vroome, E.M., and van Mechelen, W. (2003).
Participatory ergonomics as a return-to-work intervention: a future challenge? American Journal of
Industrial Medicine, 44, 273–81.
Arnetz, B.B., Sjogren, B., Rydehn, B., and Meisel, R. (2003). Early workplace intervention for employees
with musculoskeletal-related absenteeism: a prospective controlled intervention study. Journal of
Baril, R., Clarke, J., Friesen, M., Stock, S., and Cole, D. (2003). Work-ready group. Management of
return-to-work programs for workers with musculoskeletal disorders: a qualitative study in three
Canadian provinces. Social Science & Medicine, 57, 2101–14.
Bernacki, E.J. and Tsai, S.P. (2003). Ten years’ experience using an integrated workers’ compensation
management system to control workers’ compensation costs. Journal of Occupational and
Environmental Medicine, 45, 508–16.
Boersma, K. and Linton, S.J. (2006). Psychological processes underlying the development of a chronic
pain problem: a prospective study of the relationship between profiles of psychological variables in the
fear-avoidance model and disability. Clinical Journal of Pain, 22, 160–6.
Brouwer, S., Krol, B., Reneman, M.F. et al. (2009). Behavioral determinants as predictors of return to work
after long-term sickness absence: an application of the theory of planned behavior. Journal of
Occupational Rehabilitation, 19, 166–74.
Cats-Baril, W.L. and Frymoyer, J.W. (1991). The economics of spinal disorders. In J.W. Frymoyer (Ed.)
The Adult Spine, pp. 85–106. New York: Raven.
Cedraschi, C. and Allaz, AF. (2005). How to identify patients with a poor prognosis in daily clinical
practice. Baillieres Best Practice and Research in Clinical Rheumatology, 19, 577–91.
Cooper, J.E., Tate, R., and Yassi, A. (1997). Work hardening in an early return to work program for nurses
with back injury. Work, 8, 149–56.
Costa-Black, K.M., Durand, M.J., Imbeau, D., Baril, R., and Loisel, P. (2007). Interdisciplinary team
discussion on work environment issues related to low back disability: a multiple case study. Work,
28, 249–65.
Crook, J., Milner, R., Schultz, I.Z., and Stringer, B. (2002). Determinants of occupational disability following
a back injury: A critical review of the literature. Journal of Occupational Rehabilitation, 12, 277–95.
Dejesus, R.S., Vickers, K.S., Melin, G.J., and Williams, M.D. (2007). A system-based approach to
depression management in primary care using the Patient Health Questionnaire-9. Mayo Clinic
Proceedings, 82, 1395–402.
Derebery, V.J., Giang, G.M., Saracino, G., and Fogarty, W.T. (2002). Evaluation of the impact of a low
back pain educational intervention on physicians’ practice patterns and patients’ outcomes. Journal of
Devereux, J.J., Buckle, P.W., and Vlachonikolis, I.G. (1999). Interactions between physical and
psychosocial risk factors at work increase the risk of back disorders; an epidemiological approach.
Devereux, J.J., Vlachonikolis, I.G., and Buckle, P.W. (2002). Epidemiological study to investigate potential
interaction between physical and psychosocial factors at work that may increase the risk of symptoms
of musculoskeletal disorder of the neck and upper limb. Occupational and Environmental Medicine,
59, 269–77.
Doghramji, P.P. (2004). Recognizing sleep disorders in a primary care setting. Journal of Clinical Psychiatry,
65(Suppl 16), 23–6.
Durand, M.J., Loisel, P., Hong, Q.N., and Charpentier, N. (2002). Helping clinicians in work disability
prevention: the work disability diagnosis interview. Journal of Occupational Rehabilitation, 12(3), 191–204.
Durand, M.J., Vézina, N., Baril, R., Loisel, P., Richard, M.C., and Nogomo, S. (2009). Margin of
manoeuvre indicators in the workplace during the rehabilitation process: a qualitative analysis. Journal
of Occupational Rehabilitation, 19, 194–202.
Feldman, J.B. (2004). The prevention of occupational low back pain disability: evidence-based reviews
point in a new direction. Journal of Surgical Orthopaedic Advances, 13, 1–14.
Franche, R.L., Cullen, K., Clarke, J. et al. (2005). Research Team. Workplace-based return-to-work
interventions: a systematic review of the quantitative literature. Journal of Occupational Rehabilitation,
15, 607–31.
Gatchel, R.J., Polatin, P.B., Noe, C., Gardea, M., Pulliam, C., and Thompson, J. (2003). Treatment and
cost-effectiveness of early intervention for acute low-back pain patients: A one-year prospective study.
Journal of Occupational Rehabilitation, 13, 1–9.
Guo, H.R., Tanaka, S., and Cameron, L.L. (1995). Back pain among workers in the United States: national
estimates and workers at high risk. American Journal of Industrial Medicine; 28, 591–602.
Hardt, J., Jacobsen, C., Goldberg, J., Nickel, R., and Buchwald, D. (2008). Prevalence of chronic pain in a
representative sample in the United States. Pain Medicine, 9, 803–12.
Hartvigsen, J., Lings, S., Leboeuf-Yde, C., and Bakketeig, L. (2004). Psychosocial factors at work in relation
to low back pain and consequences of low back pain; a systematic, critical review of prospective cohort
studies. Occupational and Environmental Medicine, 61, e2.
Hasenbring, M., Marienfeld, G., Kuhlendahl, D., and Soyka, D. (1994). Risk factors of chronicity in
lumbar disc patients. A prospective investigation of biologic, psychologic, and social predictors of
therapy outcome. Spine, 19, 2759–65.
Hayden, J., Côté, P., and Bombardier, C. (2006). Evaluation of the quality of prognosis studies in
systematic reviews. Annals of Internal Medicine, 144, 427–37.
Hoogendoorn, W.E., van Poppel, M.N., Bongers, P.M., Koes, B.W., and Bouter, L.M. (2000). Systematic
review of psychosocial factors at work and private life as risk factors for back pain. Spine, 25, 2114–25.
Kendall, N.A. (1999). Psychosocial approaches to the prevention of chronic pain: the low back paradigm.
Baillieres Best Practice and Research in Clinical Rheumatology, 13, 545–54.
Krause, N., Dasinger, L.K., and Neuhauser, F. (1998). Modified work and return to work: a review of the
literature. Journal of Occupational Rehabilitation, 8, 113–19.
Kristensen, T.S. (2005). Intervention studies in occupational epidemiology. Occupational and
Lindström, I., Ohlund, C., Eek, C., Wallin, L., Peterson, L.E., and Nachemson, A. (1992). Mobility,
strength, and fitness after a graded activity program for patients with subacute low back pain.
A randomized prospective clinical study with a behavioral therapy approach. Spine, 17, 641–52.
Linton, S.J. (2001). Occupational psychological factors increase the risk for back pain: A systematic review.
Linton, S.J. and Hallden, K. (1998). Can we screen for problematic back pain? A screening questionnaire
for predicting outcome in acute and subacute back pain. Clinical Journal of Pain, 14, 209–15.
Linton, S.J. and Nordin, E. (2006). A 5-year follow-up evaluation of the health and economic consequences
of an early cognitive behavioral intervention for back pain: A randomized, controlled trial. Spine,
31, 853–8.
Linton, S.J., Gross, D., Schultz, I.Z. et al. (2005). Prognosis and the identification of workers risking
disability: research issues and directions for future research. Journal of Occupational Rehabilitation,
15, 459–74.
Loisel, P. (2007). Pain in the workplace, compensation, and disability management. In R.F. Schmidt &
W.D. Willis (Eds.) Encyclopedia of Pain, pp. 1703–5. New York: Springer.
Loisel, P., Durand, M.J., Berthelette, D. et al. (2001). Disability prevention–new paradigm for the
management of occupational back pain. Disease Management and Health Outcomes, 9, 351–60.
Loisel, P., Gosselin, L., Durand, P., Lemaire, J., Poitras, S., and Abenhaim, L. (2001). Implementation of a
participatory ergonomics program in the rehabilitation of workers suffering from subacute back pain.
Applied Ergonomics, 32, 53–60.
Loisel, P., Durand, M.J., Baril, R., Gervais, J., and Falardeau, M. (2005). Interorganizational collaboration
in occupational rehabilitation: perceptions of an interdisciplinary rehabilitation team. Journal of
Occupational Rehabilitation, 15, 581–90.
Main, C.J. (2002). Concepts of treatment and prevention in musculoskeletal disorders. In S.J. Linton (Ed.)
New avenues for the prevention of chronic musculoskeletal pain and disability. Pain research and clinical
management, vol. 12, pp. 47–63. New York: Elsevier.
Marhold, C., Linton, S.J., and Melin, L. (2002). Identification of obstacles for chronic pain patients to
return to work: evaluation of a questionnaire. Journal of Occupational Rehabilitation, 12, 65–75.
Marras, W. S., Ferguson, S. A., Burr, D., Schabo, P. and Maronitis, A. (2007). Low back pain recurrence in
occupational environments. Spine, 32, 2387–97.
Nachemson, A., Waddell, G., and Norlund, A.I. (2000). Epidemiology of neck and low back pain, in
A. Nachemson and E. Jonsson (eds.) Neck and back pain: The scientific evidence of causes, diagnosis, and
treatment, pp. 165–87. Philadelphia, PA: Lippincott Williams & Wilkins.
Nicholas, M.K. (2002). Reducing disability in injured workers: The importance of collaborative
management. In S.J. Linton (Ed.) New avenues for the prevention of chronic musculoskeletal pain and
disability. Pain research and clinical management, vol. 12, pp. 33–46. New York: Elsevier.
Pransky, G., Shaw, W.S., and Fitzgerald, T.E. (2001). Prognosis in acute occupational low back pain:
Methodologic and practical considerations. Human and Ecological Risk Assessment, 7, 1811–25.
Pransky, G., Shaw, W.S., Franche, R.L., and Clarke, A. (2004). Disability prevention and communication
among workers, physicians, employers, and insurers–current models and opportunities for
improvement. Disability and Rehabilitation, 26, 625–34.
Schultz, I.Z., Crook, J., Berkowitz, J., Milner, R., and Meloche, G.R. (2005). Predicting return to work after
low back injury using the Psychosocial Risk for Occupational Disability Instrument: a validation study.
Schultz, I.Z., Stowell, A.W., Feuerstein, M., and Gatchel, R.J. (2007). Models of return to work for
musculoskeletal disorders. Journal of Occupational Rehabilitation, 17, 327–52.
Shaw, W.S., Robertson, M.M., McLellan, R.K., Verma, S., and Pransky, G. (2006). A controlled case study
of supervisor training to optimize response to injury in the food processing industry. Work 26, 107–14.
Shaw, W.S. and Huang, Y.H. (2005). Concerns and expectations about returning to work with low back
pain: identifying themes from focus groups and semi-structured interviews. Disability and
Shaw, W.S., Pransky, G., Fitzgerald, T.E. (2001). Early prognosis for low back disability: Intervention
strategies for health care providers. Disability Rehabilitation, 23, 815–28.
Shaw, W.S., Pransky, G., Patterson, W., and Winters, T. (2005). Early disability risk factors for low back
pain assessed at outpatient occupational health clinics. Spine, 30, 572–80.
Shaw, W.S., Robertson, M.M., Pransky, G., and McLellan, R.K. (2006a). Training to optimize the response
of supervisors to work injuries–needs assessment, design, and evaluation. American Association of
Occupational Health Nurses Journal, 54, 226–35.
Shaw, W.S., Linton, S.J., and Pransky, G. (2006b). Reducing sickness absence from work due to low back
pain: How well do intervention strategies match modifiable risk factors? Journal of Occupational
Shaw, W.S., van der Windt, D.A., Main, C.J. et al. (2009). Early patient screening and Intervention to
address individual-Level occupational factors (‘blue flags’) in back disability. Journal of Occupational
Shrey, De and Hursh, N.C. (1999). Workplace disability management: International trends and
perspectives. Journal of Occupational Rehabilitation, 9, 45–59.
Steenstra, I.A., Verbeek, J.H., Heymans, M.W., and Bongers, P.M. (2005). Prognostic factors for duration
of sick leave in patients sick listed with acute low back pain: a systematic review of the literature.
Sullivan, M.J., Feuerstein, M., Gatche,l R., Linton, S.J., and Pransky, G. (2005). Integrating psychosocial
and behavioral interventions to achieve optimal rehabilitation outcomes. Journal of Occupational
van Duijn, M., Miedema, H., Elders, L., and Burdorf, A. (2004). Barriers for early return-to-work of
workers with musculoskeletal disorders according to occupational health physicians and human
resource managers. Journal of Occupational Rehabilitation, 14, 31–41.
van Tulder, M.W., Koes, B.W., and Bouter, L.M. (1995). A cost-of-illness study of back pain in The
Netherlands. Pain, 62, 233–40.
Varekamp, I., de Vries, G., Heutink, A., and van Dijk, F.J.H. (2008). Empowering employees with chronic
diseases; development of an intervention aimed at job retention and deisgn of a randomised controlled
trial. BMC Health Services Research, 8, 224.
Varekamp, I., Heutink, A., Landman, S., Koning, C.E.M., de Vries, G., and van Dijk, F.J.H. (2009).
Facilitating empowerment in employees with chronic disease: Qualitative analysis of the process of
change. Journal of Occupational Rehabilitation, 19, 398–408.
Volinn, E., La,i D., McKinney, S., et al. (1988). When back pain becomes disabling: A regional analysis.
Pain, 33, 33–9.
Von Korff, M., Balderson, B.H.K, Saunders, K., et al. (2005). A trial of an activating intervention for
chronic back pain in primary care and physical therapy settings. Pain, 113, 323–30.
Waddell, G., Aylward, M., and Sawney, P. (2002). Back pain, incapacity for work and social security benefits:
An international literature review and analysis. London: The Royal Society of Medicine Press.
Waddell, G. (2004). The Back Pain Revolution (2nd edn., pp. 265–82. New York: Churchill-Livingstone.
Wasiak, R., Pransky, G., Verma, S., and Webster, B. (2003). Recurrence of low back pain:
definition-sensitivity analysis using administrative data. Spine, 28, (19), 2283–91.
Watkins, E.A., Wollan, P.C., Melton, L.J. 3rd, and Yawn, B.P. (2008). A population in pain: report from
the Olmstead County health study. Pain Medicine, 9, 166–74.
Weir, R. and Nielson, W.R. (2001). Interventions for disability management. Clinical Journal of Pain,
17 (4 Suppl), S128–32.
Werner, E.L., Laerum, E., Wormgoor, M.E., Lindh, E., and Indahl, A. (2007). Peer support in an
occupational setting preventing LBP-related sick leave. Occupational Medicine, 57, 590–5.
Westmorland, M.G., Williams, R.M., Amick, B.C. 3rd, Shannon, H., and Rasheed, F. (2005). Disability
management practices in Ontario workplaces: employees’ perceptions. Disability and Rehabilitation,
27, 825–35.
Young, A.E., Roessler, R.T., Wasiak, R., McPherson, K.M., van Poppel, M.N.M., and Anema, J.R. (2005).
A developmental conceptualization of return to work. Journal of Occupational Rehabilitation,
15(4), 557–68.
Part 5
Practitioner’s Role in the

Process of Care
Chapter 20
The Physician as Disability Advisor

for Back Pain Patients
James Rainville, Glenn S. Pransky, Sarah Gibson,
and Pradeep Suri
20.1 Introduction
It is a proper and likely inescapable function of physicians to sanction disability in the course of
treating illness (Sullivan and Loeser 1992). In acute or obvious illness, these dual roles result in
little controversy, as recommending a temporary suspension of personal and occupational
responsibilities may be beneficial to the process of healing. Moreover, significantly impaired body
function often necessitates a period of convalescence—recognized by physicians, patients, and
society as reasonable. On the other hand, for the problem of low back pain (LBP), where the
pathology is less overt and where symptoms often endure beyond the expected time required for
healing, physicians may be faced with a difficult dilemma. What level of disability is appropriate
for the pathology, symptoms, and impaired function exhibited by a patient? Here inter-physician
variation increases dramatically regarding the degree of work disablement endorsed, or restric-
tion of activity recommended for a given set of complaints.
The variability in recommendations may have enormous consequences for patients with LBP,
including an impact on the probability of transitioning from acute to chronic disability. It is
recognized that inappropriate disability can lead to decreased quality of life due to inactivity,
while prolonged absence from normal social roles (including work) is deleterious to physical and
mental well-being (Shulman 1994; McGrail et al. 2002). Appropriate and early return to activities
and work avoids the consequences of illness reinforcement: assumption of the sick role, deterio-
ration in family dynamics, dependence on drugs, creation of secondary gain, learned disability, as
well as other negative impacts (Derebery and Tullis 1983).
Do physicians’ disability recommendations make a difference? For LBP, limited evidence
suggests that they do. Several authors have noted that consistent, evidence-based physician rec-
ommendations about disability that encourage resumption of activity independent of pain status
lead to improved disability outcomes in LBP and other disorders (Indahl et al. 1998; Hagen et al.
2002; Rainville et al. 2002) Of importance, physicians may influence patients’ disability in the
opposite direction—advising avoidance of activities to those with back pain. Though the magni-
tude of this effect on disability is unknown, it is probably of consequence. This is supported by the
results from cross-sectional studies of general practitioners (GPs) and their patients with acute
LBP and rheumatologist and their patient with subacute LBP. Both studies observed that high
ratings of pain beliefs in physicians were associated with a higher rate of pain beliefs of their
patients (Poiraudeau et al. 2006; Coudeyre et al. 2007).
Recently, researchers have begun to acknowledge the importance of physicians’ disability rec-
ommendations. Not surprisingly, findings have revealed that a complex interaction of physician,
patient, social, and political factors influence physician recommendations.
This chapter will explore current knowledge about physician disability recommendations for
patients with LBP. It will also review results of attempts to influence these recommendations, and
suggest areas where further research may be beneficial.
20.2 Magnitude of the physician role as disability advisor

Nearly all patient–physician encounters for musculoskeletal disorders include recommendations
about the appropriate level of activity for the illness under treatment. These generally include
advice about use and mobility of the injured body part for personal care, domestic, and recrea-
tional activities. Additionally, physicians are often asked or required to make recommendations
about the ability of their patients to work. Many jurisdictions in the USA mandate such recom-
mendations, when injury is alleged to result from vocational activities or exposures. In the UK
and many other countries, the provision of sickness certification is mandated as part of the
contractual services delivered by GPs.
The actual percentage of patient encounters during which physicians are asked to certify disa-
bility is substantial. Bollag et al. (2007) reported that 4% and Soler and Okkes (2004) reported
that 11% of patient encounters for GPs resulted in the issuance of sickness certificates. Pransky
et al. (2002) and Englund et al. (2000a) reported that primary care providers were asked to supply
opinions concerning work ability in about 9% of all patient encounters. All studies found that
musculoskeletal diagnoses dominated these requests (Englund and Svardsudd 2000). For this
reason, practices specializing in musculoskeletal disorders also have a high number of patient
encounters where definitions of ‘work capacity’ are required (Arrelov et al. 2007).
20.3 The content of physician recommendations

As activity and work recommendations are an essential part of many patient–physician encoun-
ters, it is worthwhile exploring the content of these recommendations. To date, most studies
examining physicians’ recommendations have substituted patient histories or vignettes for actual
patient encounters, with recognized research limitation (Jones et al. 1990). The results demon-
strate a striking lack of consistency in physicians’ assessment of all types of disability. This includes
disability assessment of common musculoskeletal disorders (Haldorsen et al. 1996; Patel et al.
2003), social security disability eligibility (Carey et al. 1988), and ill-health retirement determina-
tions (Elder et al. 1994). Not surprisingly, studies limited to LBP-related disability have also
demonstrated a lack of consensus among physicians (Chibnall et al. 2000; Rainville et al. 2000).
When explored further however, despite high variability between physicians, individual physi-
cians exhibited consistent patterns in their disability recommendations, ranging from lenient to
strict (Getz and Westin 1995; Chibnall et al. 2000; Rainville et al. 2000). The detrimental implica-
tions of this inconsistency and inequality are obvious for both patients with LBP and disability
systems they access. Clearly, the most negative impact is unnecessary disability with associated
overall decline in economic well-being and health (Waddell et al. 2000). Additionally, the incon-
sistency of disability recommendation encourage ‘doctor shopping’ where patients search for
physicians that support their position on disability. Consistent patterns of disability recommen-
dation for individual physicians can establish medical reputations, leading to funnelling of
patients to specific medical providers who will either refute or support disability claims. Ultimately,
this variance in medical opinions probably consumes countless hours of debate as legal systems
try to sort out conflicting physicians’ disability recommendations.
THE PHYSICIAN AS DISABILITY ADVISOR 393
20.4 Physicians’ characteristics that influence disability

recommendations
There are several possible explanations for the high level of observed variability in activity recom-
mendations for LBP. One source of variability may result from differences in the depth of medical
knowledge about LBP. To examine this, Rainville et al. (2000) compared chronic low back pain
(CLBP) work and activity recommendations of family physicians and orthopaedic spine surgeons.
They noted that orthopaedic spine surgeons were only slightly more permissive than family
physicians with work restrictions and very similar with activity restrictions. These finding were
corroborated by Englund et al. (2000a). Apparently, the level of expertise as reflected by medical
specialty has a limited influence on disability recommendations.
A potential, but unexplored source of variability in physicians ’ disability recommenda-
tions may be differences in theories about the aetiology of spinal degeneration and CLBP. For
decades, medical wisdom supported an injury model of LBP, where exposures of the spine
to physical stresses were deemed responsible for spinal degeneration (Nachemson 1975). Indeed,
the injury model spawned empirical recommendation about posture, lifting techniques, and
back health, and inspired decades of ergonomic research. It is logical for physicians that are
strong proponents of this model to advise patients against activities perceived to be harmful to
the spine.
The injury model of LBP has been considerably weakened over the last two decades, in part by
the failure of ergonomic interventions to reduce back injuries, but mainly from epidemiological,
genetic, and neurobiological research that offered alternative explanations for low back disorders
(Martimo et al. 2007). A fundamentally different model of spinal degeneration has emerged in
which the aetiology of degeneration is explained by genetically encoded, age-activated degenera-
tive processes (Ala-Kokko, 2002) As such, this model undermines the alleged role of physical
exposure in spinal degeneration, and weakens the association between persistent pain complaints
and ongoing damage or harm. Theoretically, physicians endorsing this model of LBP would more
likely recommend that activity avoidance is medically unnecessary.
Another possible source of variation in disability recommendations may be inadequate medical
education or poor preparation of physicians for determining disability (Sullivan and Loeser
1992). Indeed, a majority of surveyed physicians indicated that they have learned little about dis-
ability programmes from any source (Carey et al. 1987), do not feel confident in their ability to
determine patient disability (Zinn and Furutani 1996), and felt burdened by participation in this
task (Zinn and Furutani 1996). Furthermore, many physicians feel inadequately educated about
work capabilities and risk for injury from occupation exposures (Pransky et al. 2002). Despite
concerns about their role as disability advisors, the degree of physicians’ dislike of this role or
their feelings of inadequacy towards this process are not reflected in the duration of sickness
certification for patients under their care (Tellness et al. 1990).
A related area of importance is the physician’s ability to identify the small number of cases at
high risk for progression from acute, self-limited injury to a chronically disabled state. These cases
may benefit the most from interventions designed to avoid prolonged disability at an early stage.
Unfortunately, physicians have little training in risk prediction for low back injury and are
often unaware of specific advice or intervention that can alter the progression to disablement
(W.S. Shaw et al. 2001; Haldorsen et al. 2002).
Perhaps reflecting these problems, patients did not view the physician as an active facilitator of
return to work and activity in one study (Roberts-Yates 2003). These patients cited the physician’s
lack of experience, dissociation from the workplace, disinterest in workplace issues and needs,
and failure to devote the necessary time as evidence of their ineffective role in this issue.
One could easily theorize that inherent beliefs, fears, and concerns that healthcare providers
hold about pain and function should be strong determinants of their recommendations for activ-
ity and work after an injury. This association is supported by several studies. Utilizing the
Healthcare Providers’ Pain and Impairment Relationship Scale (Rainville et al. 1995), Rainville
et al. (2000) discovered that pain attitudes and beliefs of physicians were the strongest predictor
of their work/activity recommendations. Similarly, Houben et al. (2004) noted that physical
therapists’ work/activity recommendations were strongly predicted by the strength of their pain
attitudes and beliefs as well. Poiraudeau et al. (2006) noted that rheumatologists with strong fear-
avoidance pain beliefs were less likely to follow guidelines on prescribing physical and occupa-
tional activities for LBP. Linton et al. (2002) reported that physicians’ recommendations for sick
leave were strongly dependent on their own fear-avoidance beliefs with regard to back pain.
20.5 Physician-patient relationship factors

Despite this evidence of a connection between providers’ beliefs and work recommendations,
some studies suggest that other characteristics of physicians may override their personal beliefs.
Gulbrandsen et al. (2007) grouped physicians into four groups based on combinations of charac-
teristics including their perceived burden of sickness certification, doubt, permissiveness, person-
ality, degree of paternalism, job satisfaction, opinion of sick certification as medical task, and
sociopolitical attitudes. Despite prominent differences between groups, no association was found
between group-level differences of physician traits and rates of sickness certification. Watson
et al. (2008) found no relationship between physicians’ pain attitudes and the frequency of issued
illness certificate for non-specific LBP. Instead, they noted that sickness absences certification for
LBP was predicted by the physicians’ sickness certification behaviours for all medical illnesses
in general.
The above findings suggest that there are other characteristics—besides physicians’ pain atti-
tudes and beliefs—that determine physicians’ disability recommendation. Mounting evidence
suggests that the physician recommendations may be largely determined by a seldom-studied
factor—the physician-patient relationship.
Discordance is common between physicians and patients concerning the issue of LBP and work
(Lofgren et al. 2007; von Knorring et al. 2008). Physicians and patients often differ in terms of
expectation of the length of disability for LBP (Kapoor et al. 2006), especially as symptoms
become more chronic (Reiso et al. 2004). Physicians tend to rely more heavily on clinical findings,
whereas patients tend to assess job related factors such as stress and the physical strenuous nature
of work (Reiso et al. 2000).
When differences of opinion occur, it seems that patients’ opinions often win. Pransky et al.
(2002) noted that primary care providers relied mainly on patient input for disability assessment.
Englund and Svardsudd (2000) observed that in cases where primary care physicians could not
medically justify ‘sick-listing’ certification, a certificate was issued anyway in 87% of cases. In a
separate study, Englund and Dahlgren (2002) noted that the most important factor affecting
‘sick-listing’ was the patient’s attitude to ‘sick-listing’, as those wishing ‘sick-listing’ were ‘sick-
listed’ to a greater degree than those who were reluctant. Mayhew and Nordlund (1998) found
that 41% of family physicians reported feeling pressured to write unwarranted work excuses and
had a sense of being manipulated by their patients. Brook (1996) reported that hostile, demand-
ing, threatening and malingering patients, frequently influenced physicians’ completion of disa-
bility requests. Zinn and Furutani (1996) found that 40% of physicians reported a willingness to
exaggerate clinical data for patients they felt deserving of disability.
The observations described above document that significant discordance exists between
patients and physicians about the degree to which illness justifies limitation on activity and work.
For some physicians, the discordance may be trivial—they feel unable to judge the appropriate-
ness of disability more accurately than their patients, and therefore rely upon their patient’s
‘better judgment’. It is also possible that in order to avoid this discordance, many physicians
adapt rather neutral attitudes about pain and function (Rainville et al. 1995). These neutral atti-
tudes may serve the physician well, as it allows the physician to easily acquiesce to the disability
position adapted by their patients, and thus avoid conflict.
For many other physicians, this discordance is troubling. However, they still may acquiesce to
the patient’s requests. One important reason for this is suggested in the study by Zinn and
Furutani (1996), where over 80% of physicians felt that refusing to fill out disability forms could
adversely affect the doctor–patient relationship, and 62% felt that it reflected a conflict of interest.
These findings suggest that most primary care physicians see their role as gatekeepers for com-
pensation systems as subservient to their role as patient advocate, especially when their relation-
ship is long term and includes care for all medical problems (Hussey et al. 2004). From a practical
point of view, it may be much less time-consuming for physicians to simply acquiesce to patients
demands and avoid conflict.
It appears that the interaction of patients’ beliefs, behaviours, and demands, with internal
physician characteristics of professionalism, social desirability, and negotiation strategies can
profoundly influence disability recommendations (Monday et al. 1988). Research is needed on
how physicians cope with their conflicting roles as patient advocate and gatekeeper to disability
benefits, with the goal of identifying potential interventions that would improve their perform-
ance as disability advisors.
20.6 Patients’ perceptions of disability

This discordance may in part result from patients perceiving the disabling effect of their problem
differently than physicians do. For many chronic musculoskeletal disorders, physical factors actu-
ally play a relatively minor role in influencing disability outcomes. Indeed, it appears that patients
with chronic musculoskeletal symptoms who continue to seek medical care may possess attitudes
and beliefs about pain and function that are more restrictive and potential disability levels that are
significantly higher than others in the general population with a similar malady (Szpalski et al.
1995; Waxman et al. 1998). Past research suggests that much of the observed variance in self-
reported disability results from factors that are independent of the physical aspects of the muscu-
loskeletal afflictions. Influential factors include coping abilities (Turner et al. 2000), affect and
personality (Gatchel et al. 1993; Maxwell et al. 1998), fear-avoidance beliefs (Swinkels-Meewisse
et al. 2006), educational level (Roth and Geisser 2002), work issues (Bigos et al. 1991), and con-
cerns about benefit loss and secondary gain (Rainville et al. 1997; Klekamp et al. 1998). Though
frank malingering is probably quite rare, somatization, distortion, and some exaggeration may be
common (Ensalada 2000). These areas of non-medical distress can strongly influence patients’
behaviours and magnify their perception of disability, especially as it relates to work.
Patients relating back pain to exposure to physical activities, fear of re-injury, and the belief that
pain indicates harm are likely important factors influencing disability (Rainville et al. 1993;
Waddell et al. 1993; Jensen et al. 1994; Symonds et al. 1996). People with a history of back pain,
and those with blue-collar occupations more often attribute back pain to external factors such as
physical activities and the work environment, though misconceptions about back pain are wide-
spread, even in the group reporting no back pain (Linton and Wang 1993; Goubert et al. 2004).
Patients may be reluctant to continue activities that are believed to have precipitated their back
pain. With this in mind, a recommendation to remain active and return to work may be ineffec-
tive unless it’s accompanied by information that resolves or addresses fears of re-injury. Physician-
presented patient education, such as that reported by Indahl et al. (1998), that attempts to displace
the injury model of spinal degeneration with a more benign hypothesis, and dissociates pain and
function, may offer some encouraging results. This approach is clearly in need of further study.
20.7 Factors external to the physician-patient relationship

that influence disability
Employers can also have an influence on physician behaviour regarding disability by engaging
more cooperative physicians, communicating their return to work expectations and providing
alternative duty (Habeck et al. 1991; Pransky et al. 2002; L. Shaw et al. 2002; W.S. Shaw 2003; van
Duijn et al. 2004). Baril and Nordqvist (Baril et al. 2003; Nordqvist et al. 2003) conducted qualita-
tive studies of return-to-work programmes, contrasting the perceptions of workers, employers,
physicians, and others involved in the process. Although physicians focused on individual factors
as the most important determinants of return-to-work outcome, workers and health and safety
managers readily identified workplace-specific factors as the most important determinants for
returning to work.
Yet another influence may be the healthcare system itself, including level of reimbursement for
services related to caring for disabled workers, local market effects, and provider-induced utiliza-
tion in work-related musculoskeletal disorders. Johnson et al. (1999) demonstrated that health-
care providers incur a benefit in the form of healthcare utilization in disabling work-related
musculoskeletal disorders. For some providers who offer costly diagnostic and therapeutic proce-
dures, a disincentive may exist to return patients quickly. The burden of salary replacement ben-
efits may motivate some insurers to authorize expensive medical services, especially when the
medical provider insists that return-to-work recommendations are contingent on performing a
procedure or test. Even for clinics specializing in work-related musculoskeletal disorders that are
established specifically to improve care, incentives to increase utilization can lead to a net increase
in disability, as aptly demonstrated in a series of Ontario clinics by Sinclair et al. (1997).
Legal jurisdictional influences on physician behaviour in relation to disability are highly sig-
nificant. This is a key factor in the profound differences across jurisdictions in mean lengths of
disability for equivalent conditions—despite absence of differences in the nature and the severity
of injury, and the availability return to work opportunities (Johnson et al. 1996).
In summary, the recommendations for work/activities made by physicians to patients with
musculoskeletal complaints are the product of the complex interaction of many factors. In addi-
tion to the obvious factors of the medical condition itself, as well as idiosyncratic characteristics
of the physician and patient, factors external to the physician-patient encounter also influence
physicians’ recommendations. These include the work place, medical systems, and legislation.
20.8 Efforts to influence physician recommendations

Interventions have attempted to improve the quality of physician recommendations about work
resumption after injury, with varying degrees of success.
20.8.1 Medical educational efforts

Efforts to improve physicians’ education about the impact of activity recommendations on disa-
bility have been sparse, and the results are unimpressive. Lie (2003) reported on an educational
effort in Norway to decrease the length of sick leaves by augmenting the knowledge base of GPs.
Participating physicians received education about common musculoskeletal disorders and finan-
cial compensation for spending extra time with patients with extended sick leaves. Unfortunately,
no impact on sick listing practices or disability outcomes were noted.
Perhaps it is not the direct knowledge about musculoskeletal disorders that is most important,
but instead the pain attitudes and beliefs of the practitioners that are most critical in disability
recommendations. If so, efforts to directly alter these may be of some benefit. This issue has also
received limited study. Latimer et al. (2004) reported on an educational project designed to
change the attitudes and beliefs about pain and function of physical therapy students. Results
indicated that substantial changes in favour of endorsing function occurred following the teach-
ing module and these changes were maintained 1 year out.
From another perspective, training for physicians should focus on skills that directly address
the findings that it is psychosocial and not medical issues that are the primary determinants of
disability. This implies developing an ability to elicit and address the factors that may inhibit suc-
cessful return to work and knowledge of how to effectively address these factors (L. Shaw et al.
2002). Related skills such as prognostic screening and selective referral to resources based on the
specific barriers to return to work, may also be directly related to producing evidence-based
return-to-work recommendations, inasmuch as these interventions are necessary to insure a high
likelihood of compliance and success (W.S. Shaw et al. 2001; Haldorsen et al. 2002).
20.8.2 Guidelines
In the mid 1990s, leading researchers developed medical guidelines to address disability issues in
musculoskeletal disorders, especially LBP, urging physicians to facilitate return to work (Frank
et al. 1996; Canadian Medical Association 1997). The premise was that evidence-based guidelines,
norms, and other ‘scientific’ material would lead to practice change once physicians were informed
about them (Verbeek et al. 2002). In isolation, educational efforts appear to have minimal impact,
especially in low back pain (Rao et al. 2002; Verbeek et al. 2002).
Direct efforts to mandate that physicians’ work recommendations be based on clinical and not
subjective criteria have shown some promise. In a case-controlled study, Hall et al. (1994) exam-
ined the impact of a mandate that all low back injured workers be released to unrestricted (versus
restricted) work, unless objective medical contraindications were present. This mandate resulted
in twice the number of patients returning to unrestricted work, without an adverse effect on the
probability of successful return to those work activities. Similar finding were noted by Hiebert
et al. (2003). Both studies were undertaken in private business settings where physicians were
employees and modest control over physicians’ actions could be elicited. As such the applicability
of these findings to more traditional medical practice settings is questionable. This is demon-
strated by a study in a less controlled setting by Scheel et al. (2002) which demonstrated that
simply providing physicians with a mechanism to return patients to work sooner was insufficient,
even though most physicians readily acknowledged the importance of improved return to
work outcomes. In part, these results may reflect the highly skewed distribution of length of
disability amongst a minority of patients, the inherent difficulty in engaging physicians, as well
as the multitude of factors besides physician recommendations that shape outcomes (Scheel
et al. 2002).
20.8.3 Systems efforts

Realizing the limitation of interventions that are aimed at medical providers alone, several recent
projects have taken a broad-based, systemic approach, thereby acknowledging that physician
behaviour occurs in a larger context that includes patient knowledge, expectations, and behav-
iours. As observed by Grimshaw, Deyo, and others, most successful approaches: (1) include some
form of medical educational outreach with use of academic detailing by recognized leaders;
(2) include changes in the processes that direct physicians to the desired behaviour; (3) include
simultaneous education of patients; (4) include a high and sustained level of administrative and
financial support; and (5) acknowledge that one technique may not be suitable for all (Oxman
et al. 1995; Deyo et al. 2000; Grimshaw et al. 2001).
Community-wide interventions have been attempted in several countries. These interventions
theorized that by simultaneously supplying both the public and medical providers with similar
information about the benefits of activity and work, the discordance between physicians and
patients could be lessened. One could argue that as the community is brought along, physician-
patient dialogue might change and achieve concordance at a point consistent with evidence-based
guidelines.
Buchbinder et al. (2001a, 2001b) initiated this type of intervention in Australia. The interven-
tion used a multimedia campaign to disseminate information about the benefits of staying active
with back pain, including work. The result was a large, significant improvement in back pain
beliefs and less fear-avoidance regarding physical activity. This was paralleled by improvements
in beliefs and practice intentions of physicians. Of equal importance, the researchers noted a clear
decline in the number of claims for back pain, days of compensation and medical cost for claims
during the campaign.
A similar system wide approach was taken for a public health initiative entitled ‘Working Backs
Scotland’ (Waddell et al. 2007). The initiative utilized a combination of radio and a website to
disseminate information targeting both people with back pain and employers. Concurrently,
information supplied to medical providers advised against recommending bed rest, activity
restrictions and issuing sickness certification for back pain. This initiative also demonstrated a
shift in public beliefs away from rest to staying active, and a comparable shift in professional
advice. However, no changes in work or compensation outcomes were noted. These results
exposed a flaw in the ‘Working Back Scotland’ intervention—it avoided direct discussions of
work The lack of changes in work-related factors suggests that work expectations during a back
pain episode must be addressed directly for these interventions to be successful.
Smaller and more focused systemic changes can also impact physicians’ behaviours and
responses. Bernacki and Tsai (2003) described a successful workers’ compensation management
system at an employer-level that decreased lost time by 73% and the number of medical claims by
61% over a 10-year period. In this system, physicians’ expectations for advising light duty and
return to work are unambiguous; there are appreciable rewards for providers to work within the
system, and employers will accommodate a returning injured worker. Similar programmes have
reported comparable results (Green-McKenzie et al. 2002; McGrail et al. 2002). Eccleston and
Yeager (1998) concluded that these types of programme generated savings of 9–54%, not by
reducing medical costs, but by providing earlier and more effective return-to-work recommenda-
tions from treating physicians.
Some innovative approaches recognize that physicians are trained to perform within a medical
model; to the extent that determinants of disability are primarily psychosocial, not medical, their
training and orientation are poorly suited to effectively address issues of return to work (Hunt
et al. 2002). Based on this presumption, one successful approach places physicians as a contribut-
ing member of a team that provides specific recommendations about return to work and facili-
tates the return to work process (Durand and Loisel 2001; Edlund and Dahlgren 2002). Another
approach trains physicians to become more effective communicators about LBP, its prognosis,
and the optimal approach for management (McGrail et al. 2002).
20.8.4 Legislative efforts

Clearly social systems that compensate individuals for illness and disability have a profound
potential to influence human behaviours. As such, changes in these systems, especially changes
that affect eligibility, should have the ability to change societal expectations and the financial
burden of disability. The positive effects of altering eligibility criteria for financial compensation
are demonstrated in studies by Cassidy et al. (2000) and Claussen (1998).
Even though changes in the public support of disability can be legislated, these may have
limited impact on physicians’ disability recommendation. Englund et al. (2000b) noted no change
in the ‘sick listing’ practices of physicians collected before and after a legislative reform, other
than they completed the forms more thoroughly after the reform. Legislative changes instituted
in Sweden in 1995 aimed at reducing costs of sickness absence by mandating that physicians
exclude non-medical criteria for sick-listing, recommend more part-time sick-listing, and pursue
faster rehabilitation did not meet any of their goals. Indeed, a survey of physicians by Getz and
Westin (1993) reported that a majority of physicians believed that legislation resulting in a higher
percentage of refusals for disability applications would not bring many applicants back to employ-
ment, but would instead transfer the financial burden of disability to other public sources of
support. Apparently, physicians’ attitudes towards disability and their behaviours with patients
requesting disability certification are not easily impacted by legislative efforts, most likely because
they do not lessen the demands of patients for disability certification.
20.9 Summary and conclusions

When attending to patients with musculoskeletal complaints, physicians are required to give
advice about the appropriate level of activities and work for the condition under treatment.
Evidence suggests that this advice varies greatly between physicians. This variation results from
multiple factors. From the perspective of physicians, differences in medical knowledge, attitudes
and beliefs, negotiation abilities, and professionalism surrounding the doctor–patient relation-
ship can all influence their recommendation. Patient, employer, and societal factors may also
influence these recommendations.
Prolonged work avoidance has no identified therapeutic benefit and significant adverse effects
on social well-being. Although some physicians recognize the impact of poor recommendations,
most efforts to persuade physicians to discourage patients from sustained work absence have been
unsuccessful. This may be in part due to an incomplete understanding by physicians of the proc-
esses whereby back pain-related disability becomes chronic (Guzman et al. 2002). Regardless,
efforts to harmonize physicians’ recommendations to promote full return to activity or work
despite chronic musculoskeletal pain seem warranted. One-dimensional interventions such as
physician education or implementation of evidence-based guidelines have had limited effect on
disability recommendations. Based on the available evidence, significant changes in physician
recommendations will require educating both physicians and patients, along with altering soci-
etal expectations about back pain and work (Aylward and Waddell 2002). Successful multidimen-
sional efforts have included educational material to enhance physicians’ knowledge combined
with effective community-based efforts to influence patients’ functional expectations before they
enter the physicians’ office. During the medical encounter, physicians should be motivated to
address the issues where there is disagreement about disability, consistent with the standards of
high quality doctor-patient communication. Accepting the input, and understanding the needs
of employers and governmental organizations can aid the physician in providing appropriate
medical input for these entities.
Thus, future research should concentrate on interventions that can change the attitudes and
actions of key players in the disablement process, within the context of a particular sociopolitical
environment. Even in the setting of expectations for prolonged disability, interventions that
address all of the most significant factors underlying extended LBP-related work absence could be
successful (Loisel et al. 2003). Our future challenge will be to generalize these findings in a way
that will enable physicians, patients and employers to accept early, rapid, and safe return to activ-
ity and/or work as an expected outcome.
Acknowledgements
Dr Suri is supported by the Rehabilitation Medicine Scientist Training K12 Program (RMSTP)
and the National Institutes of Health (K12 HD 01097).
References
Ala-Kokko, L. (2002). Genetic risk factors for lumbar disc disease. Ann Med, 34, 42–7.
Arrelov, B., Alexanderson, K., Hagberg, J., Lofgren, A., Nilsson, G., Ponzer, S. (2007). Dealing with
sickness certification—a survey of problems and strategies among general practitioners and
orthopaedic surgeons. BMC Public Health, 7, 273.
Aylward, M., Waddell, G. (2002) Low back Pain, Incapacity, and Social Security. London: Royal Society of
Medicine Press.
Baril, R., Clarke, J., Friesen, M., et al. (2003). Work-Ready Group. Management of return-to-work
programs for workers with musculoskeletal disorders: a qualitative study in three Canadian provinces.
Soc Sci Med, 57, 2101–14.
Bernacki, E.J., Tsai, S.P. (2003). Ten years’ experience using an integrated workers’ compensation
management system to control workers’ compensation costs. J Occup Environ Med, 45, 508–16.
Bigos, S.J., Battie, M.C., Spengler, D.M., et al. (1991). A prospective study of work perceptions and
psychosocial factors affecting the report of back injury. Spine, 16, 1–6.
Bollag, U., Rajeswaran, A., Ruffieux, C., Burnand, B. (2007). Sickness certification in primary care - the
physician’s role. Swiss Med Wkly, 137, 341–6.
Brook, T.R. (1996). How patients stress, con, and intimidate physicians to file dubious disability reports.
J Natl Med Assoc, 88, 300–4.
Buchbinder, R., Jolley, D., Wyatt, M. (2001). 2001 Volvo Award Winner in Clinical Studies: Effects of a
media campaign on back pain beliefs and its potential influence on management of low back pain in
general practice. Spine, 26, 2535–42.
Buchbinder, R., Jolley, D., Wyatt, M. (2001). Population based intervention to change back pain beliefs
and disability: three part evaluation. BMJ, 322, 1516–20.
Canadian Medical Association (1997). The Physician’s Role in Helping Patients Return to Work after an
Illness or Injury. Can Med Assoc 156, 680A–C.
Carey, T.S., Fletcher, S.W., Fletcher, R., et al. (1987). Social Security disability determinations. Knowledge
and attitudes of consulting physicians. Med Care, 25, 267–75.
Carey, T.S, , Hadler, H.M., Gillings, D., et al. (1988). Medical disability assessment of the back pain patient
for the Social Security Administration: the weighting of presenting clinical features. J Clin Edidemiol,
41, 691–7.
Cassidy, J.D., Carroll, L.J., Cote, P., et al. (2000). Effect of eliminating compensation for pain and suffering
on the outcome of insurance claims for whiplash injury. New Engl J Med, 342, 1179–86.
Chibnall, J.T., Dabney, A., Tait, R,C. (2000). Internist judgment of chronic low back pain. Pain Med, 3,
231–6.
Claussen, B. (1998). Physicians as gatekeepers: will they contribute to restrict disability benefits? Scand
J Prim Health Care, 16, 199–203.
Coudeyre, E., Tubach, F., Rannou, F., et al. (2007). Fear-avoidance beliefs about back pain in patients with
acute LBP. C J Pain, 23, 720–5.
Derebery, V.J., Tullis, W.H. (1983). Delayed recovery in the patient with a work compensable injury.
J Occup Med, 25, 829–35.
Deyo, R.A., Schall, M., Berwick, D.M., et al. (2000). Continuous quality improvement for patients with
back pain. J Gen Intern Med, 15, 647–55.
Durand, M.J., Loisel, P. (2001). Therapeutic return to work: Rehabilitation in the workplace. Work, 17,
57–63.
Eccleston, S.M., Yeager, C.M. (1998). Managed care and medical costs in workers’ compensation: a national
inventory, 1997–1998. Cambridge, MA: Workers Compensation Research Institute.
Edlund, C., Dahlgren, L. (2002). The physician’s role in the vocational rehabilitation process. Disabil
Rehabil, 24, 727–33.
Elder, A.G., Symington, I.S., Symington, E.H. (1994). Do occupational physicians agree about ill-health
retiral? A study of simulated retirement assessments. Occup Med, 44, 231–5.
Englund, L., Svardsudd, K. (2000). Sick-listing habits among general practitioners in a Swedish county.
Scand J Prim Health Care, 18, 81–6.
Englund, L., Tibblin, G., Svardsudd, K. (2000a). Variations in sick-listing practice among male and female
physicians of different specialties based on case vignettes. Scand J Prim Health Care, 18, 48–52.
Englund, L., Tibblin, G., Svardsudd, K. (2000b). Effects on physicians’ sick-listing practice of an
administrative reform narrowing sick-listing benefits. Scand J Prim Health Care, 18, 215–9.
Ensalada, L.H. (2000). The importance of illness behavior in disability management. Occup Med, 15, 739–54.
Frank, J.W., Brooker, A.S., DeMaio, S.E., et al. (1996). Disability resulting from occupational low back
pain. Part II: What do we know about secondary prevention? A review of the scientific evidence on
prevention after disability begins. Spine, 21, 2918–29.
Gatchel, R.J., Polatin, P.B., Mayer, T.G. (1993). The dominant role of psychosocial risk factors in the
development of chronic low back pain disability. Spine, 20, 2702–9.
Getz, L., Westin, S. (1993). Restrictions in connection with disability pensions–physicians’ view on the
revised legislation. Tidsskr Nor Laegeforen, 113, 2133–6.
Getz, L., Westin, S. (1995). Assessment by consulting physicians and general practitioners about complex
disability pension matters. Tidsskr Nor Laegeforen, 115, 1748–53.
Green-McKenzie, J., Rainer, S., Behrman, A., et al. (2002). The effect of a health care management
initiative on reducing workers’ compensation costs. J Occup Environ Med, 44, 100–5.
Goubert, L., Crombez, G., De Bourdeaudhuij, I. (2004). Low back pain, disability and back pain myths in a
community sample: prevalence and interrelationships. Eur J Pain, 8, 385–94.
Grimshaw, J.M., Shirran, L., Thomas, R., et al. (2001). Changing provider behavior: an overview of
systematic reviews of interventions. Med Care, 39(Suppl 2), II2–45.
Gulbrandsen, P., Hofoss, D., Nylenna, M., Saltyte-Benth, J., Aasland, O.G. (2007). General practitioners’
relationship to sickness certification. Scand J Prim Health Care, 25, 20–6.
Guzman, J., Yassi, A., Cooper, J.E., et al. (2002). Return to work after occupational injury. Family
physicians’ perspectives on soft-tissue injuries. Can Fam Physician, 48, 1912–9.
Habeck, R.V., Leahy, M.J., Hunt, H.A., et al. (1991). Employer factors related to workers’ compensation
claims and disability management. Rehab Counseling Bull, 34, 210.
Hagen, K.B., Hilde, G., Jamtvedt, G., Winnem, M.F. (2002). The Cochrane review of advice to stay active
as a single treatment for low back pain and sciatica. Spine, 27, 1736–41.
Haldorsen, E.M.H., Brage, S., Johannesem, T.S., et al. (1996). Musculoskeletal pain: concepts of disease,
illness and sickness certification in health professionals in Norway. Scan J Rheumatol, 4, 224–32.
Haldorsen, E.M., Grasdal, A.L., Skouen, J.S., et al. (2002). Is there a right treatment for a particular patient
group? Comparison of ordinary treatment, light multidisciplinary treatment, and extensive
multidisciplinary treatment for long-term sick-listed employees with musculoskeletal pain. Pain, 95, 49–63.
Hall, H., McIntosh, G., Melles, T., et al. (1994). Effect of discharge recommendations on outcome. Spine,
19, 2033–7.
Hiebert, R., Skovron, M.L., Nordin, M., Crane, M. (2003). Work restrictions and outcome of nonspecific
low back pain. Spine, 28, 722–8.
Houben, R.M.A., Vlaeyen, J.W.S., Peters, M., et al. (2004). Healthcare providers’ attitudes and beliefs
toward common low back pain. Factor structure and psychometric properties of the HC-PAIRS. Clin
J Pain, 20, 37–44.
Hunt, D.G., Zuberbier, O.A., Kozlowski, A.J., et al. (2002). Are components of a comprehensive medical
assessment predictive of work disability after an episode of occupational low back trouble? Spine,
27, 2715–9.
Hussey, S., Hoddinott, P., Wilson, P., Dowell, J., Bardour, R. (2004). Sickness certification system in the
United Kingdom: qualitative study of views of general practitioners in Scotland. BMJ, 328, 88.
Indahl, A., Haldorsen, E.H., Holm, S., Reikeras, O., Ursin, H. (1998). Five-year follow-up study of a
controlled clinical trial using light mobilization and an informative approach to low back pain. Spine,
23, 2625–30.
Jensen, M.P., Turner, J.A., Romano, J.M. (1994). Relationship of pain-specific beliefs to chronic pain
adjustment. Pain, 57, 301–9.
Johnson, W.G., Baldwin, M.L., Burton, J.F., Jr. (1996). Why is the treatment of work-related injuries so
costly? New evidence from California. Inquiry, 33, 53–65.
Johnson, W.G., Burton, J.F., Jr, Thornquist, L., et al. (1999). Why does workers’ compensation pay more
for health care? Benefits Q, 9, 22–31.
Jones, T.V., Gerrity, M.S., Earp, J. (1990). Written case simulations: So they predict physicians’ behaviors.
J Clin Epidemiol, 43, 805–15.
Kapoor, S., Shaw, W.S., Pransky, G., Patterson, W. (2006). Initial patient and clinician expectations of
return to work after acute onset of work-related low back pain. J Occup Environ Med, 48, 1173–80.
Klekamp, J., McCarty, E., Spengler, D.M. (1998). Results of elective lumbar discectomy for patients
involved in the workers’ compensation system. J Spinal Disord, 11, 277–82.
Krause, N., Dasinger, L.K., Neuhauser, F. (1998). Modified work and return to work; a review of the
litersture. J Occ Rehab, 8, 113–39.
Latimer, J., Maher, C., Refshauge, K. (2004). The attitudes and beliefs of physiotherapy students to chronic
back pain. Clin J Pain, 20, 45–50.
Lie, H. (2003). Could sick leaves be reduced by augmenting the knowledge of the general practitioner?
Tidsskr Nor Laegeforen, 123, 2068–71.
Linton, S.J., Vlaeyen, J., Ostelo, R. (2002). The back pain beliefs of health care providers: Are we
fear-avoidant? J Occup Rehabil, 12, 223–32.
Linton, S.J., Wang, L.E. (1993). Attributions (beliefs) and job satisfaction associated with back pain in an
industrial setting. Percept Mot Skills, 76, 51–62.
Lofgren, A., Hagberg, J., Arrelov, B., Ponzer, S., Alexanderson, K. (2007). Frequency and nature of
problems associated with sickness certification tasks: a cross-sectional questionnaire study of 5455
physicians. Scand J Prim Health Care, 25, 178–85.
Loisel, P., Durand, M.J., Diallok, B., et al. (2003). From evidence to community practice in work
rehabilitation: the Quebec experience. Clin J Pain, 19, 105–13.
Martimo, K.P., Verbeek, J., Karppinen, J., et al. (2007). Manual material handling advice and assistive
devices for preventing and treating back pain in workers. Cochrane Database Syst Rev, 3, CD005958.
Maxwell, T.D., Gatchel, R.J., Mayer, T.G. (1998). Cognitive predictors of depression in chronic low back
pain: toward an inclusive model. J Behav Med, 21, 131–43.
Mayhew, H.E., Nordlund, D.J. (1998). Absenteeism certification: the physician’s role. J Fam Pract, 26,
651–5.
McGrail, M.P., Jr., Calasanz, M., Christianson, J., et al. (2002). The Minnesota Health Partnership and
Coordinated Health Care and Disability Prevention: the implementation of an integrated benefits and
medical care model. J Occup Rehabil, 12, 43–54.
Monday, J., Therrien, S., Duguay, M., et al. (1988). The Physician and disability certificates: preconceived
attitudes and behaviors. Can J Psychiatry, 33, 599–605.
Nachemson, A. (1975). Towards a better understanding of low-back pain: a review of the mechanics of the
lumbar disc. Rheumatol Rehabil, 14, 129–4.
Nordqvist, C., Holmqvist, C., Alexanderson, K. (2003). Views of laypersons on the role employers play in
return to work when sick-listed. J Occup Rehabil, 13, 11–20.
Oxman, A.D., Thomson, M.A, , Davis, D.A, , et al. (1995). No magic bullets: a systematic review of
102 trials of interventions to improve professional practice. CMAJ, 153, 1423–31.
Patel, D., Buschbacker, R., Crawford, J. (2003). National variability in permanent partial impairment
ratings. Am J Phys Med Rehabil, 82, 302–6.
Poiraudeau, S., Rannou, F., Baron, G., et al. Fear-avoidance beliefs about back pain in patients with
subacute low back pain. Pain, 124, 305–11.
Poiraudeau, S., Rannou, F., Le Henanff, A., et al. (2006). Outcome of subacute low back pain: influence of
patients’ and rheumatologists’ characteristics. Rheumatology, 45, 718–23.
Pransky, G., Katz, J.N., Benjamin, K. (2002). Improving the physician role in evaluating work ability and
managing disability: a survey of primary care practitioners. Disabil Rehabil, 24, 867–74.
Pransky, G., Katz, J.N., Benjamin, K., Himmelstein, J. (2002). Improving the physician role in evaluating
work ability and managing disability: a survey of primary care practitioners. Disabil Rehabil, 24, 867–74.
Rainville, J., Ahern, D.K., Phalen, L. (1993). Altering beliefs about pain and function in a functionally
oriented treatment program for chronic low back pain. Clin J Pain, 9, 196–201.
Rainville, J., Bagnall, D., Phalen, L. (1995). Health care providers’ attitudes and beliefs about functional
impairments and chronic back pain. Clin J Pain, 11, 287–95.
Rainville, J., Carlson, N., Polatin, P., Gatchel, R.J., Indahl, A. (2000). Exploration of physicians’
recommendations for activities in chronic low back pain. Spine, 25, 2210–7.
Rainville, J., Hartigan, C., Jouve, C., Martinez, E., Hipona, M., Limke, J. (2002). Does medical advice about
pain and function influence patients with chronic low back pain. International Society for Study of the
Lumbar Spine, Cleveland, OH, 15–18 May, 2002.
Rainville, J., Sobel, J.B., Hartigan, C., et al. (1997). The effect of compensation involvement on the
reporting of pain and disability by patients referred for rehabilitation of chronic low back pain. Spine,
22, 2016–24.
Rao, J.K., Kroenke, K., Mihaliak, K.A., et al. (2002). Can guidelines impact the ordering of magnetic
resonance imaging studies by primary care providers for low back pain? Am J Manag Care, 8, 27–35.
Reiso, H., Gulbrandsen, P., Brage, S. (2004). Doctors’prediction of certified sickness absence. Fam Pract,
21, 192–9.
Reiso, H., Nygard, J.F., Frage, S., Gulbrandsen, P., Tellnes, G. (2000). Work ability assessment by patients
and their GPs in new episodes of sickness certification. Fam Pract, 17, 139–44.
Roberts-Yates, C. (2003). The concerns and issues of injured workers in relation to claims/injury
management and rehabilitation: the need for new operational frameworks. Disabil Rehabil,
25, 898–907.
Roth, R.S., Geisser, M.E. (2002). Educational achievement and chronic pain disability: mediating role of
pain-related cognitions. Clin J Pain, 18, 286–96.
Scheel, I.B., Hagen, K.B., Herrin, J., et al. (2002). Blind faith? The effects of promoting active sick leave for
back pain patients: a cluster-randomized controlled trial. Spine, 27, 2734–40.
Shaw, L., Segal, R., Polatajko, H., et al. (2002). Understanding return to work behaviors: promoting the
importance of individual perceptions in the study of return to work. Disabil Rehabil, 24, 185–95.
Shaw, W.S., Pransky, G., Fitzgerald, T.E. (2001). Early prognosis for low back disability: intervention
strategies for health care providers. Disabil Rehabil, 23, 815–28.
Shaw, W.S., Robertson, M.M., Pransky, G., et al. (2003). Employee perspectives on the role of supervisors
to prevent workplace disability after injuries. J Occup Rehabil, 13, 129–42.
Shulman, B.M. (1994). Worklessness and disability; expansion of the biopsychosocial perspective. J Occ
Rehabil, 4, 113–22.
Sinclair, S.J., Hogg-Johnson, S.H., Mondloch, M.V., et al. (1997). The effectiveness of an early active
intervention program for workers with soft-tissue injuries. The Early Claimant Cohort Study. Spine,
22, 2919–31.
Soler, J.K., Okkes, I.M. (2004). Sick leave certification: an unwelcome administrative burden for the family
doctor? The role of sickness certification in Maltese family practice. Eur J Gen Pract, 10, 50–5.
Sullivan, M.D., Loeser, J.D. (1992). The diagnosis of disability. Treating and rating disability in a pain
clinic. Arch Intern Med, 152, 1829–35.
Swinkels-Meewisse, I.E., Raelofs, J., Verbeek, A.L., et al. (2006). Fear-avoidance beliefs, disability, and
participation in workers and non-workers with acute low back pain. Clin J Pain, 22, 45–54.
Symonds, T.L., Burton, A.K., Tillotson, K.M., Main, C.J. (1996). Do attitudes and belief influence work
loss due to low back trouble? Occup Med, 46, 25–32.
Szpalski, M., Nordin, M., Skovron, M.L., et al. (1995). Health care utilization for low back pain in Belgium.
Influence of socio-cultural factors and health beliefs. Spine, 20, 431–42.
Tellness, G., Sandvik, L., Moum, T. (1990). Inter-doctor variation in sickness certification. Scan J Prim
Health Care, 8, 45–52.
Turner, J.A., Jensen, M.P., Romano, J.M. (2000). So beliefs, coping, and catastrophizing independently
predict functioning in patients with chronic pain? Pain, 85, 115–25.
van Duijn, M., Miedema, H., Elders, L., et al. (2004). Barriers for early return-to-work of workers with
musculoskeletal disorders according to occupational health physicians and human resource managers.
J Occup Rehabil, 14, 31–41.
Verbeek, J.H., van Dijk, F.J., Malmivaara, A., et al. (2002). Evidence-based medicine for occupational
health. Scand J Work Environ Health, 28, 197–204.
von Knorring, M., Sundberg, L., Lofgren, A., Alexanderson, K. (2008). Problems in sickness certification of
patients: a qualitative study on views of 26 physicians in Sweden. Scand J Prim Health Care, 26, 22–8.
Waddell, G., Burton, K., Aylward, M. (2000). Work and common health problems. J Insur Med, 39,
109–20.
Waddell, G., Newton, M., Henderson, I., et al. (1993). A Fear-Avoidance Beliefs Questionnaire (FABQ)
and the role of fear- avoidance beliefs in chronic low back pain and disability. Pain, 52, 157–68.
Waddell, G., O’Connor, M., Boorman, S., Torsney, B. (2007). Working Backs Scotland: a public and
professional health education campaign for back pain. Spine, 33, 2139–43.
Watson, P.J., Bowey, J., Purcell-Jones, G., Gales, T. (2008). General practitioner sickness absence
certification for low back pain is not directly associated with beliefs about back pain. Eur J Pain,
12, 314–20.
Waxman, R., Tennant, A., Helliwell, P. (1998). Community survey of factors associated with consultation
for low back pain. BMJ, 317, 1564–7.
Zinn, W., Furutani, N. (1996). Physician perspective on the ethical aspects of disability determination.
J Gen Intern Med, 11, 525–32.
Chapter 21
The Attitudes and Beliefs of Clinicians

Treating Back Pain: Do They Affect
Patients’ Outcome?
Tamar Pincus, Rita Santos, and Steven Vogel
21.1 Introduction
Many common musculoskeletal disorders do not have clear pathophysiological explanations for
pain. Classifications of clinical presentations often rely on broad-based descriptions of symptoms
such as ‘simple low back pain’. Even where guidelines exist, these often focus broadly on patient
management issues rather than the detail of diagnosis and treatment. In the absence of detailed
and specific best practice guidance, clinicians’ interventions are likely to be informed by their
individual beliefs and attitudes (Foster et al. 2003a). The word attitude is derived from Latin,
meaning ‘fit and ready for action’. Attitudes are affectively loaded constructs that precede behav-
iour and guide choices and decisions for action (Hogg and Vaughan 2005). Given the complexi-
ties of musculoskeletal conditions and the proposed role of attitudes and beliefs in informing
treatment, it is important to consider patients and practitioners in the process of care (Foster et al.
2003b). Both attitudes and their associated beliefs are addressed in this chapter with a particular
focus on clinicians working primarily with musculoskeletal conditions. Low back pain (LBP) has
been most thoroughly studied and is used as an exemplar. In complement to the previous chap-
ter, this chapter highlights differences between professional groups, further discusses which
beliefs and attributions in clinicians relate to patient outcome, and reviews measurements of
clinicians’ beliefs.
Back pain and neck pain are directly and indirectly costly to society (Maniadakis and Gray
2000; Picavet and Schouten 2003). Direct costs are often estimated through healthcare utilization,
while indirect costs include days off work etc. Most people who have had an episode of back pain
will have another, but might not seek help (Croft et al. 1998). A significant minority of people
experiencing back pain will progress to chronic and long-term disability. The development of
prevention strategies are required to avoid the high levels of distress and personal impact for these
individuals and to reduce the impact on service utilization and cost to society (Linton 2002). Risk
factors for the development of persistent musculoskeletal pain have been identified, and some of
them include patients’ beliefs and behaviours (Croft et al. 2006). Unhelpful ‘myths’ about back
pain have also been studied in the general population. A large study of Norwegian populations
found that as many as 50% believed that radiography and imaging tests were needed and that
back pain indicated an injury (Ihlebaek and Eriksen 2003).
Another large survey study in Norwegian populations (Werner et al. 2005) found that approx-
imately 20% of respondents, regardless of previous or current experience of back pain, agreed
with the statement ‘Bed rest is the mainstay of therapy for back pain’. In addition, only a third of
respondents agreed with the statement ‘One recovers faster from back pain if one continues to
work or returns as soon as possible’. Similarly, a postal study of over 1600 respondents from the
general population in Belgium found that 77% believe that if one has back pain, wrong move-
ments can lead to serious problems, and 35% believed that back pain required bed rest (Goubert
et al. 2004). These findings suggest that there is still a long way to go in disseminating evidence
and current ideas about recovery from back pain in the general population.
Such beliefs encourage avoiding normal activities during back pain episodes, and are in direct
contradiction to current guidelines (Burton et al. 2004). These beliefs and behaviours are difficult
to change once entrenched. Testing theories about their acquisition is important, in order
to prevent the development of unhelpful beliefs and maladaptive behaviours in future patients.
A much neglected area of research is the complex interaction between patients and clinicians,
including passing on information and messages about the threat posed by back pain, and advice
about protective behaviours which supposedly reduce this threat. This raises the question as
to whether inappropriate advice from clinicians influences the transition from acute pain to
chronic disability?
21.2 How could clinicians increase the risk

of long-term problems?
There are several ways in which clinicians may impact on the long-term well-being of their
patients with musculoskeletal pain in general and back pain specifically. Effects of specific treat-
ment interventions, adverse effects of treatment, failure to diagnose, and practical problems
about timely and correct referral are outside the scope of the current chapter. However, clinicians
almost always offer advice to patients, and some of this advice might slow down recovery. Most
practitioners do not advocate bed rest these days (Ihlebaek and Eriksen 2004; Werner et al. 2005;
Buchbinder and Jolley 2007), suggesting that for the majority of clinicians, in the case of bed rest,
evidence, beliefs, and behaviour are in consensus. However, as myths about back pain are still
prevalent in the general population, the problem may lie in eliciting and identifying harmful
beliefs held by patients, and in communicating more appropriate knowledge as part of an educa-
tional element during treatment.
There are several hypothetical routes by which clinicians’ communications with patients could
be influenced by their own misconceptions, and contribute to patients unhelpful beliefs about back
pain. These in turn may influence behaviours in patients that impede recovery. Individual factors
that have been identified in patients as obstacles to recovery include fear of movement and hyper-
vigilance (Vlaeyen and Linton 2000), described in this volume in Chapters 14 and 16. Clinicians
may inadvertently reinforce fearful beliefs. Suggestions such as ‘listen carefully to your body’,
‘watch out for any signs of pain’, and including minor bodily asymmetries (such as uneven leg
length) in causal explanations of pain may enhance or promote hypervigilance and increase fear.
Increased fear has been shown to result in avoidance, culminating in a vicious circle leading to
increased disability and suffering (Crombez et al. 1999; reviewed in this volume in Chapter 16).
Another hypothetical route is a recommendation for rest from work, which may include formal
sick certification. The evidence on the link between clinicians’ beliefs and sick certification is
reviewed below, and is far from conclusive. Our own studies, described in detail below, suggest
that although sick certification is rare amongst musculoskeletal practitioners in the UK, many
hold the belief that a short rest from work is necessary for the body to allow healing (Pincus et al.
in preparation). Current guidelines recommend moderated work duties or work as normal dur-
ing episodes of musculoskeletal back pain (Waddell and Burton 2001; Staal et al. 2003). However,
to date, there is no evidence that directly evaluates the cost and benefit of short-term time off
work accompanied with advice to stay active and avoid bed rest.
CLINICIANS’ ATTITUDES 407
Clinicians may also offer subtle communications about perceived harm from work. Part of
patient evaluation carried out by most clinicians includes a search for possible causal factors, and
most musculoskeletal practitioners explore work-related factors routinely. While this is a useful
part of the investigation, it may inadvertently convey a message that work is harmful and should
be avoided. Similarly, communication of empathy with suffering, such as ‘Oh yes, I can see that
would really hurt’ may unwittingly convey a message about perceived damage if communicated
in a work-related context.
21.3 Measuring clinicians’ beliefs

A systematic review of current measures of beliefs and attitudes towards back pain in healthcare
practitioners (Bishop et al. 2007) examined the quality of each measure in terms of their concep-
tual model, reliability, validity, burden (respondent and administrative), alternative forms, and
language adaptations. The authors concluded that the theoretical model for most of the measures
lacked validity, as items were commonly adapted from those proposed for patients. The content,
adapted from patient questionnaires, does not necessarily provide a comprehensive representa-
tion of clinicians’ beliefs.
Some measures of clinicians’ attitudes and beliefs are very specific. For example, the Fear
Avoidance Beliefs Questionnaire for Health Care Practitioners (FABQ; Coudeyre et al. 2006), is
an adaptation of the Fear Avoidance Beliefs Questionnaire, developed for patients (Waddell et al.
1993). The questionnaire includes two subscales, examining beliefs about physical exercise and
beliefs about work. It has been criticized for wording items almost exclusively in reference to
exercise, rather than daily activities (Pincus et al. 2006a). A longer questionnaire measuring fear
avoidance beliefs in practitioners was developed by Linton and colleagues (Fear avoidance beliefs
tool; Linton et al. 2002). This instrument combines questions from several questionnaires (includ-
ing the TSK, the FABQ, and the PAIRS) and adds items reflecting beliefs about sick leave as a
treatment for back pain, and items measuring information giving by practitioners. Given the
complexity of the questionnaire and the lack of information about reliability and validity, findings
are difficult to interpret.
A questionnaire with a related underlying model is the Health Care Providers Pain and
Impairment Relationship Scale (HC-PAIRS; Rainville et al. 1995). The HC-PAIRS was adapted
from the Pain and Impairment Relationship Scale (PAIRS; Riley et al. 1988), originally devel-
oped for patients with chronic LBP, to measure beliefs about the link between pain and move-
ment and activities. The items were adapted directly, by changing the reference from ‘I’ to ‘Back
pain patients’ (Rainville et al. 1995). There are several studies reporting good reliability and valid-
ity on this scale. However, subsequent research suggests that the original factor solution needs
adjusting and some items were removed, resulting in a single-factor solution, which has been
shown to be moderately linked to self-report of recommendations about work and activity in
response to a vignette of a patient (Houben et al. 2004).
A more comprehensive approach, albeit still focused on fear avoidance beliefs in clinicians, was
taken in the development of the PABS-PT (Ostelo et al. 2003). The PABS-PT was developed from
existing questionnaires for patients with chronic low back pain, including the Tampa Scale for
Kinesiophobia (TSK, Miller et al. 1991); the Pain Catastrophizing Scale (PCS, Sullivan et al.
1995); the Back Beliefs Questionnaire (BBQ, Symonds et al. 1996); the FABQ (Waddell et al.
1993), plus items that were deemed to have clinical value. The resulting questionnaire consists of
two subscales, labelled ‘biomedical orientation’ and ‘behavioural orientation’. Although the
authors report acceptable clinimetric properties for the questionnaire, the subscale measuring
behavioural orientation has not performed well in subsequent research (Watson et al. 2008).
The Attitudes to Back Pain Scale for Musculoskeletal Practitioners (ABPS-MP; Pincus et al.
2006b) is a comprehensive questionnaire that includes items measuring beliefs about long-term
treatment of patients with back pain, beliefs associated with referring patients elsewhere, and
clinicians’ sense of being connected to a health network. Unlike other instruments, it was devel-
oped through qualitative interviews with clinicians, as opposed to be adapted from instruments
designed for patients. In addition, it includes subcategories measuring factors similar to those
measured in the HC-PAIRS and PABS-PT, such as beliefs in reactivation of patients, beliefs in a
biomedical model, and attitudes to exploring psychological aspects of the patients’ problems.
Although the questionnaire has good reported clinimetric properties (Pincus et al. 2006b; Bishop
et al. 2007) it has not yet been tested in reference to reported clinical decision-making.
In summary, there are a handful of published questionnaires available to measure clinicians’
beliefs and attitudes about back pain, but research in this area is still in its infancy. Even the most
studied questionnaires have only been used in small national cohorts, and there are limitations,
notably in reference to clear underlying models and comprehensiveness. Still missing from cur-
rent measurements are several important areas, including items: (1) reflecting beliefs about clini-
cians’ role in returning/maintaining people at work during episodes of back pain; (2) reflecting
beliefs about the use of analgesia; and (3) reflecting beliefs about the need for hands-on ‘manual’
treatment for back pain.
21.4 The effect of clinician’s beliefs on clinical decisions

and behaviour
Arguably, clinicians’ beliefs drive their clinical decisions and behaviour, although the degree of
the impact of beliefs compared to other influences remains unknown. It has been argued by
researchers in this field that attitudes need to be in situ prior to behaviour (Evans et al. 2005),
based on the theory of planned behaviour. Research that has attempted to link beliefs and behav-
iours in practitioners is often compromised because of the difficulties of measuring clinical deci-
sions and behaviour in real-life situations. Most published research exploring these relationships
employs self-report of behaviour, or reported putative clinical behaviour in response to vignettes
of patients.
A general exploration of beliefs was carried out in a study of 187 physical therapy physicians in
Sweden, which found that the majority held the belief that the patients’ own resources and the
relationship between the clinician and patient were more important to outcome than treatment
techniques. They also endorsed the view that the whole person should be treated, rather than the
site of pain (Stenmar and Nordholm 1994). In a study exploring common ‘myths’ about low back
pain in Norwegian general practitioners (GPs) (n=436) and physiotherapists (n=311), seven
statements were presented to practitioners, corresponding to common mistaken beliefs about
back pain (Ihlebaek and Eriksen 2004). The vast majority of the practitioners did not endorse the
beliefs that tests were useful, that back pain indicated injury or that bed rest was useful. Neither
study explored how beliefs were linked to clinical decisions, recommendation, treatment, and
advice.
Several cross-sectional studies have found a relationship between reported beliefs and reported
practice. A cross-sectional questionnaire study carried out with 60 GPs and 71 physiotherapists in
Sweden (Linton et al. 2002) findings suggested that clinicians high in ‘fear-avoidance’ beliefs were
also more likely to believe that sick leave was a good treatment for back pain. The study is limited
by the choice of measurement, which mixed items from several published scales. Some of the
items are difficult to interpret in the context of fear-avoidance (e.g. giving information about
which activities to do and which to avoid).
In an attempt to clarify the relationship between fear-avoidance beliefs in GPs and their clinical
decisions, questionnaires including a version of the FABQ (Waddell et al. 1993) were sent to
French GPs (Coudeyre et al. 2006). Sixteen per cent (n=139) of the total responders (n=864)
scored above a cut-point indicating high fear-avoidance beliefs, and these were associated
with self-report of giving less information about back pain and higher rates of recommending
sick leave.
The belief that pain should result in reduced activity has also been studied in 142 orthopaedic
surgeons and family practitioners in the USA (Rainville et al. 2000). Responses on the HC-PAIRS
(Rainville 1995) were moderately correlated with responses to three vignettes, and accounted for
around 22% of the variance of recommendations about work. Similarly, a study of 156 Dutch
manual therapists measured beliefs about the perceived harmfulness of physical activity, using the
HC-PAIRS, and correlating responses with responses to a vignette of a patient. Although the
correlations between beliefs and reported recommendations for work and physical activities
were low to moderate(ranging from 0.25–0.62), beliefs were the only significant predictor of these
recommendations when controlling for gender, years of experience, and perceived severity of
symptoms and pathology, albeit accounting for only 16% of the variance (Houben et al. 2004).
In a large study of general practitioners (n=442) and physiotherapists (n=580) in the UK
(Bishop et al. 2008), beliefs were measured using the PABS-PT (Houben et al. 2005). Higher bio-
medical beliefs and lower behavioural beliefs were associated with reported advice to stay off work
in responses to a vignette of a patient with LBP, although the amount of variance accounted for
by these beliefs was not reported. The study demonstrated that the belief that LBP necessitates at
least some avoidance of work was common in both groups (around 30%), but only a few believed
that bed rest was beneficial (Bishop et al. 2008).
One of the only studies that attempted to link beliefs and clinical behaviour in practitioners,
(rather than self-report of such behaviour) explored the correlation between beliefs measured on
the PABS-PT and sick leave certification (SLC) in 83 GPs in the UK (Watson et al. 2008). Neither
biomedical nor psychosocial beliefs predicted SLC, but the findings were compromised by poor
measurement of beliefs. In addition, there is some evidence to suggest that a common reason for
GPs for giving SLC to people with a physical problem is to maintain their interpersonal relation-
ship (Campbell and Ogden 2006). This motivation is currently not measured in questionnaires
about attitudes and beliefs in clinicians.
Long-term treatment of back pain patients might be considered as an obstacle to recovery,
especially in the absence of clear improvement in terms of increased activity. This was studied in
chiropractors, osteopaths, and physiotherapists in the UK using mixed methods (Pincus et al.
2006c). A screening survey was sent to 200 practitioners in each profession. All those who reported
treating at least one patient more than eight times over 3 months with no improvement (n=30,
around 10% of each profession) were interviewed. In addition, 12 practitioners were interviewed
who reported that they never treat long term without improvement. The themes emerging from
the interviews provided a clear rationale for long-term treatment. Across the professions, practi-
tioners believed that ending treatment would result in a void of care for patients. They saw their
role as much wider than dealing with back pain per se, but rather, as providers of counselling and
psychological support, in addition to providing ‘holistic’ treatment, including education about
health in general. Several questionable beliefs also emerged, including the belief that on and off
manual treatment prevented worsening pain and damage, and the belief that a sufficiently exhaus-
tive search would lead to a biomechanical cause for back pain. Above all, a strong belief was
expressed in patients’ right to decide about ending treatment. While the study raises promising
possibilities about expanding the healthcare role of musculoskeletal practitioners, there is also a
danger that long-term treatment fosters dependence and reduced self-management in patients.
Of note during the interviews it emerged that those practitioners, who reported never treating
long-term without improvement, all recalled at least one patient where they did so. This suggests
that the 10% rate of positive responses on the survey may be an underestimate.
Finally, a small qualitative study explored the beliefs of six physiotherapists, found a high
degree of biomedical orientation towards back pain in the clinicians. This was particularly evident
for chronic pain situations where complex psychosocial factors were relevant. Resultant attitudes
divided patients into ‘good’ (active and motivated) and ‘difficult’ to treat patients (passive and
non-participatory). There was some indication that there may be inequality of treatment towards
the latter group and that the physiotherapists associated the difficult to treat group with being
disheartened and frustrated (Daykin and Richardson 2004). A further qualitative study exploring
clinical reasoning in musculoskeletal physiotherapists, suggested that clinicians took a dynamic
approach to reasoning rather than using a single approach or orientation to treatment. Five
categories of reasoning were identified: biomedical, psychosocial, pain mechanisms, chronicity,
and severity/irritability (Smart and Doody 2007). Deconstructing the influence of treatment ori-
entation and clinical reasoning strategies remains a challenge. Identifying the effectiveness of
different strategies and approaches in particular situations would enhance patient care and could
inform the education of clinicians.
In conclusion, the majority of research linking clinicians’ beliefs about back pain to their clini-
cal behaviour has focused on the belief that pain was related to disability and that in the presence
of pain, activities should be reduced. This belief, conceptualized as fear-avoidance, has been
shown to be moderately linked to reported recommendations for reduced activities and work in
several studies, and across several national populations. However, the variance unaccounted for
by beliefs far outweighs that explained by clinicians’ beliefs, suggesting that research should
explore other factors affecting clinical recommendations. In addition, the methodology employed
is limited by failure to establish causal links, limited measurement of beliefs and above all, failure
to measure behaviour, rather than reported or hypothetical behaviour.
21.5 Can practitioners’ beliefs be changed, and if so, is the

change reflected in clinical behaviour?
Campaigns have been carried out aimed at shifting practitioners’ beliefs and increasing guideline
adherent practice. A Norwegian campaign aimed at doctors, physiotherapists, and chiropractors
carried out between 2002–2005, emphasized optimistic advice to stay active during back pain,
and encourage early return to work (Werner et al. 2008). Practitioners’ beliefs were measured in
a questionnaire consisting of seven statements. Measures of behaviour included reported use of
diagnostic imaging and referral patterns. Responses at baseline and post campaign were obtained
only for 25% of those contacted, considerably limiting the value of the findings. The results sup-
ported a shift towards consensus with guidelines between 2002 and 2005, but this shift appeared
independent of exposure to the campaign. However, despite this general shift and the focused
intervention, over 20% of the practitioners failed to endorse an item suggesting that recovery
from back pain is enhanced by early return to work. How such beliefs translate into advice to stay
off work, and in sick certification for back pain is not known.
In contrast to the findings from the Norwegian campaign, a much more costly and high profile
media campaign was carried out in Australia with significant results (Buchbinder et al. 2001). In
addition to the general population, a survey of GPs was carried out before and after the campaign.
Doctors reported awareness of the campaign (89%) and that it had changed their beliefs about
back pain (32%). The survey suggested that doctors in the county in which the campaign was
carried out were almost fourfold more likely to endorse the belief that people could return to
work with back pain. They also were three times more likely to reject bed rest as treatment. These
beliefs were maintained 3 and 4 years post intervention (Buchbinder and Jolly 2005; 2007). Most
important, the changes to beliefs in the general population and in general practitioners corre-
sponded to a reduction in claims for back problems.
In another large randomized control trial (RCT) aimed at musculoskeletal practitioners, the
trial arm participants (n=876) were sent a printed information pack about guideline recommen-
dations for the management of back pain (Evans et al. 2005). Responses to a vignette of a patient
included questions about the practitioners’ recommendations for activity, work, and bed rest,
and these were categorized as consistent/inconsistent with guidelines. Practitioners’ beliefs were
also measured, using a modified version of the HC-PIRS (Rainville et al. 1995). The results indi-
cated a shift in beliefs, albeit of a small order, indicating a reduction in the belief that back pain
should affect daily function. In addition, in the trial arm, reported behaviour became significantly
more guideline consistent in reference to advice about staying active and returning to work,
suggesting that an educational package can affect both beliefs and clinical behaviour.
As part of two larger RCT’s Vonk and colleagues (2004) describe a small exploratory investiga-
tion addressing the relationship between clinicians’ orientation as measured by the PABS-PT and
their professional group. The biomedical orientation subscale of the PABS-PT was reduced
in those practitioners who received training (1.5 days) to deliver a behavioural graded activity
programme. In a larger study of physiotherapy students in Australia exposure to a 16 hour unit of
learning on chronic back pain changed some subscale scores on The HC-PAIRS at immediate and
one year follow up (Vonk et al. 2004).
21.6 Differences between professional groups

As professional philosophy, training, settings, and practice vary between professional groups
treating back pain, it is important to examine difference in the beliefs of such groups. Two studies
set out to test such differences explicitly. In a large survey of Norwegian practitioners, Werner and
colleagues (2005) found that although three-quarters of responding physicians (GPs) endorsed
that statement ‘Back pain recovers best by itself’, none of the responding chiropractors agreed,
and over 70% disagreed with the statement. Physiotherapists were mostly (40%) unsure about
the statement, but 10% of the responders in this group endorsed bed rest as treatment for back
pain, in contrast to the chiropractors, who unanimously rejected the statement. Such differences
in beliefs, if communicated to patients in the course of seeking care from different practitioners,
must cause confusion.
A study of musculoskeletal practitioners (chiropractors, osteopaths, and physiotherapists) in
the UK found overall considerable consensus across the groups (Pincus et al. 2007) and has been
seen elsewhere (Vonk et al. 2009). Endorsement across the groups of a psychosocial component
of treatment, and the emphasis on reactivation was demonstrated (Pincus et al. 2007). There were
small but significant differences between the groups on some of the subscales of the ABS-MP.
Notably, physiotherapists most highly endorsed beliefs in the value of limiting the time span of
treatment and chiropractors scored least highly on this factor. This finding held true even for a
subgroup of physiotherapists who practised privately.
21.7 Summary: current knowledge on clinicians’ beliefs

There is a rising interest in the treatment-extrinsic factors that may affect outcome in patients
with back pain, and in the role that communication with clinicians might play to reduce or
increase obstacles to recovery. Research in this area is still in its early stages. There are some meth-
odological issues related to measuring clinicians’ beliefs. In addition, since it is not clear which
412
Outcome
Clinical factors:
External factors: –perceived pain
–settings –disability
–practical considerations –mood
External factors:
–education
–previous experiences
Professional identity: –social and work related factors
–specialized clinical knowlege
–professional philosophy and theory Clinician’s perception
–comparison to other professions of patient factors:
–experience with patient population individual cost/benefit analysis
Clinician’s occupational Clinician’s clinical Clinician’s Patient’s beliefs

beliefs, attitudes and decisions communications and behaviours
attributions to patients
Other influencing factors:

Personal –time
beliefs and values –language barriers
–communication skills...
Fig. 21.1 The indirect causal path from clinician’s cognition to patient outcome.
clinicians’ beliefs increase obstacles to recovery in patients, interventions have tended to focus
on the messages communicated to patients, and have attempted to increase knowledge and
awareness of guideline recommendations in clinicians. In general, such campaigns have been
moderately successful, but results may be significantly improved if specific beliefs demonstrated
to have a negative effect on patients were addressed directly.
Clinicians’ education, the role of the clinical environment, habitual clinical behaviour, and
practical considerations must also play a considerable role in clinical encounters. There is scope
and a need to investigate these systematically. Most important is the link between non-treatment
interventional behaviour and patient outcomes. This may represent an important component of
the ‘non-specific’ aspects of care that seem to add to the effectiveness of specific interventions.
The difficulty in establishing a clear causal path from clinicians’ beliefs, through their behaviour,
to affect patients’ perceptions and behaviour and culminate in positive or negative outcome is
illustrated in Figure 21.1 (informed by Evans 2007). There is to date no prospective study that
examines the relationship between clinicians’ beliefs and patients’ outcome, along the causal
route indicated in Figure 21.1. Clearly, the methodological complexity of such a study will be
considerable. In addition, there is a need to clearly identify and measure factors extrinsic to beliefs
that play a part in clinicians’ clinical decisions and communication to patients.
Conclusion
Beliefs in clinicians probably affect their behaviour, which in turn may affect the behaviour,
perception of pain and self and ultimately, outcome in patients. Research is needed to improve
measurement, so that beliefs/attitudes in clinicians can be measured and examined in RCTs and
in prospective cohorts.
References
Bishop, A., Foster, N.E., Thomas, E. and Hay, E.M. (2008). How does the self-reported clinical
management of patients with low back pain relate to the attitudes and beliefs of health care
practitioners? A survey of UK general practitioners and physiotherapists. Pain, 135, 187–95.
Bishop, A., Thomas, E. and Foster, N. (2007). Health care practitioners’ attitudes and beliefs about low
back pain: A systematic search and critical review of available measurement tools. Pain, 132, 91–101.
Buchbinder, R., Jolley, D. and Wyatt, M. (2001). Volvo Award Winner in Clinical Studies: Effects of a
media campaign on back pain beliefs and its potential influence on management of low back pain in
general practice. Spine, 26, 2535–42.
Buchbinder, R. and Jolley, D. (2005). Effects of a media campaign on back beliefs is sustained 3 years after
its cessation. Spine, 30, 1323–30.
Buchbinder, R. and Jolley, D. (2007). Improvements in general practitioner beliefs and stated management
of back pain persist 4.5 years after the cessation of a public health media campaign. Spine, 32, 156–62.
Burton, A.K., Balagué, F., Cardon, G., et al. (2006). Chapter 2. European guidelines for prevention in low
back pain : November 2004. Eur Spine J, 15(Suppl 2), S136–68.
Campbell, A. and Ogden, J. (2006). Why do doctors issue sick notes? An experimental questionnaire study
in primary care. Family Practice, 23, 125–30.
Coudeyre, E., Rannou, F., Tubach, F., et al. (2006). General practitioners’ fear-avoidance beliefs influence
their management of patients with low back pain. Pain, 124, 330–37.
Crombez, G., Vlaeyen, J.W., Heuts, P.H., and Lysens, R. (1999). Pain-related fear is more disabling than
pain itself: evidence on the role of pain-related fear in chronic back pain disability. Pain, 80, 329–39.
Croft, P.R., Macfarlane, G.J., Papageorgiou, A.C., Thomas, E. and Silman, A.J. (1998). Outcome of low
back pain in general practice: a prospective study. British Medical Journal, 316, 1356–9.
Croft, P.R., Dunn, K. and Raspe, H. (2006). Course and prognosis of back pain in primary care: The
epidemiological perspective. Pain, 122, 1–3.
Daykin, A.R. and Richardson, B. (2004). Physiotherapists’ pain beliefs and their influence on the
management of patients with chronic low back pain. Spine, 29, 783–95.
Evans, D.W. (2007). ‘Changing the practice of osteopaths, chiropractors and musculoskeletal
physiotherapists, in relation to the management of back pain.’ PhD Thesis. Keele University.
Evans, D.W., Foster, N.E., Underwood, M., Vogel, S., Breen, A.C. and Pincus, T. (2005). Testing the
effectiveness of an innovative information package on practitioner reported behaviour and beliefs. The
UK Chiropractors, Osteopaths and Musculoskeletal Physiotherapists Low Back Pain Management
Trial. BMC Musculoskeletal Disorders, 6, 41.
Foster, N.E., Pincus, T., Underwood, M., Vogel, S., Breen, A. and Harding, G. (2003a). Understanding the
process of care for musculoskeletal conditions—why a biomedical approach is inadequate.
Foster, N.E., Pincus, T., Underwood, M., Vogel, S., Breen, A. and Harding, G. (2003b). Treatment and the
process of care in musculoskeletal conditions: A multidisciplinary perspective and integration.
Orthopedic Clinics of North America, 34, 239–44.
Goubert, L., Crombez, G. and De Bourdeaudhuij, I. (2004). Low back pain, disability and back pain myths
in a community sample: prevalence and interrelationships. European Journal of Pain, 8, 385–94.
Hogg, M. and Vaughan, G. (2005). Social Psychology. London: Pearson Prentice Hall.
Houben, R.M.A., Vlaeyen, J.W.S., Peters, M., Ostelo, R.W.J.G., Wolters, P.M.J.C. and Stomp-van den
Berg, S.G.M. (2004). Health care providers’ attitudes and beliefs towards common low back pain:
factor structure and psychometric properties of the HC-PAIRS. Clinical Journal of Pain, 20, 37–44.
Houben, R.M.A., Ostelo, R.W.J.G., Vlaeyen, J.W.S., Wolters, P.M.J.C., Peters, M. and Stomp-van den
Berg, S.G.M. (2005). Health care providers’ orientations towards common low back pain predict
perceived harmfulness of physical activities and recommendations regarding return to normal activity.
Ihlebaek, C. and Eriksen, H.R. (2003). Are the ‘myths’ of low back pain alive in the general Norwegian
population? Scandinavian Journal of Public Health, 31, 395–8.
Ihlebaek, C. and Eriksen, H.R. (2004). The ‘myths’ of low back pain: status quo in Norwegian general
practitioners and physiotherapists. Spine, 29, 1818–22.
Linton, S.J. (2002). New Avenues for the Prevention of Chronic Musculoskeletal Pain and Disability. Pain
Research and Clinical Management. 12. Amsterdam: Elsevier.
Linton, S.J., Vlaeyen, J. and Ostelo, R. (2002). The back pain beliefs of health care providers: are we fear-
avoidant? Journal of Occupational Rehabilitation, 12, 223–32.
Maniadakis, N. and Gray, A. (2000). The economic burden of back pain in the UK. Pain, 84, 95–103.
Miller, R.P., Kori, S.H. and Todd, D.D. (1991). The Tampa Scale. Unpublished report, Tampa, FL.
Ostelo, R.W.J.G., Stomp-van den Berg, S.G.M., Vlaeyen, J.W.S., Wolters, P.M.J.C. and de Vet, H.H.W.
(2003). Health care provider’s attitudes and beliefs towards chronic low back pain: the development of
a questionnaire. Manual Therapy, 8, 214–22.
Picavet, H.S.J. and Schouten, J.S.A.G. (2003). Musculoskeletal pain in the Netherlands: Prevalences,
consequences and risk groups, the DMC3-study. Pain, 102, 167–78.
Pincus, T., Vogel, S., Burton, A.K., Santos, R. and Field, A.P. (2006a). Fear avoidance and prognosis in back
pain: A systematic review and synthesis of current evidence. Arthritis and Rheumatism, 54, 3999–4010.
Pincus, T., Vogel, S., Santos, R., Breen, A., Foster, N.E. and Underwood, M. (2006b). The attitudes to back
pain scale in musculoskeletal practitioners (ABS-mp): The development and testing of a new
questionnaire. Clinical Journal of Pain, 22, 378–86.
Pincus, T., Vogel, S., Breen, A., Foster, N.E. and Underwood, M. (2006c). Persistent back pain - why do
physical therapy clinicians continue treatment? A mixed methods study of chiropractors, osteopaths
and physiotherapists. European Journal of Pain, 10, 67–76.
Pincus, T., Foster, N.E., Vogel, S., Santos, R., Breen, A. and Underwood, M. (2007). Attitudes to back pain
amongst musculoskeletal practitioners: A comparison of professional groups and practice settings
using the ABS-mp. Manual Therapy, 12, 167–75.
Rainville, J., Bagnall, D. and Phalen, L. (1995). Health care providers’ attitudes and beliefs about functional
impairments and chronic back pain. Clinical Journal of Pain, 11, 287–95.
Rainville, J., Carlson, N., Polatin, P., Gatchel, R.J. and Indahl, A. (2000). Exploration of physicians’
recommendations for activities in chronic low back pain. Spine, 25, 2210–20.
Riley, J.F., Ahern, D.K. and Follick, M.J. (1988). Chronic pain and functional impairment: assessing beliefs
about their relationship. Archives of Physical Medicine and Rehabilitation, 69, 579–82.
Smart, K. and Doody, C. (2007). The clinical reasoning of pain by experienced musculoskeletal
physiotherapists. Manual Therapy, 12, 40–49.
Staal, J.B., Hlobil, H., van Tulder, M.W., Waddell, G., Burton, A.k., Koes, B.W., et al. (2003). Occupational
health guidelines for the management of low back pain: an international comparison. Occupational and
Stenmar, L. and Nordholm, L.A. (1994). Swedish physical therapists’ beliefs on what makes therapy work.
Physical Therapy, 74, 1034–39.
Sullivan, M.J.L., Bishop, S.R. and Pivik, J. (1995). The pain catastrophizing scale: development and
validation. Psychological Assessment, 7, 624–32.
Symonds, T.L., Burton, A.K., Tillotson, K.M. and Main, C.J. (1996). Do attitudes and beliefs influence
work loss due to low back trouble? Occupational Medicine, 46, 25–32.
Vlaeyen, J.W. and Linton, S.J. (2000). Fear-avoidance and its consequences in chronic musculoskeletal
pain: A state of the art. Pain, 85, 317–32.
Vonk, F., Verhagen, A.P., Geilen, M., Vos, C.J. and Koes, B.W. (2004). Effectiveness of behavioural graded
activity compared with physiotherapy treatment in chronic neck pain: design of a randomised clinical
trial. BMC Musculoskeletal Disorders, 5, 34.
Vonk, F., Pool, J.J.M., Ostelo, R. and Verhagen, A. (2009). Physiotherapists’ treatment approach towards
neck pain and the influence of a behavioural graded activity training: An exploratory study. Manual
Therapy, 14, 131–7.
Waddell, G. and Burton, A.K. (2001). Occupational health guidelines for the management of low back pain
at work: evidence review. Occupational Medicine, 51, 124–35.
Waddell, G., Newton, M., Henderson, I., Somerville, D. and Main, C.J. (1993). A Fear-Avoidance Beliefs
Pain, 52, 157–68.
Watson, P.J., Bowey, J., Purcell-Jones, G. and Gales, T. (2008). General practitioner sickness absence
certification for low back pain is not directly associated with beliefs about back pain. European Journal
of Pain, 12, 314–20.
Werner, E.L., Ihlebaek, C., Skouen, J.S. and Laerum, E. (2005). Beliefs about low back pain in the
Norwegian general population: are they related to pain experiences and health professionals? Spine,
30, 1770–6.
Werner, E.L., Ihlebaek, C., Laerum, E., Wormgoor, M.E. and Indahl, A. (2008). Low back pain meia
campaign: No effect on sickness behaviour. Patient Education and Counseling, 71, 198–203.
Part 6
Clinical Implications: New

Approaches to Diagnostics
and Treatment
Chapter 22
International Guidelines for the

Diagnostics and Treatment of Acute,
Subacute, and Chronic Back Pain
Maurits van Tulder and Bart Koes
22.1 Background
Evidence-based medicine has become increasingly more important over the past decade. The
management of low back pain (LBP) has been positively affected by the availability of more
scientific research and better use of critical appraisal techniques to evaluate and apply research
findings (Chou 2005). A large number of systematic reviews are available within and outside the
framework of the Cochrane Back Review Group that have evaluated the diagnostic and therapeu-
tic interventions of LBP (Bombardier et al. 1997; Bouter et al. 2003). This large body of evidence
has greatly improved our understanding of what does and does not work for LBP. However, in
clinical practice this evidence should be integrated with the clinical expertise of the care provider
and patient’s preferences and expectations (Sackett 1996). Recently, the evidence from trials and
reviews have formed the basis for clinical practice guidelines on the management of LBP that have
been developed in various countries around the world (Koes et al. 2001). This chapter provides
an overview of existing guidelines on LBP and discusses the challenges of implementation of these
guidelines.
22.2 Guidelines
Since publication of the Quebec Task Force report in 1987, a number of clinical guidelines for the
management of LBP in primary care have been developed in various countries worldwide. We
have identified 26 clinical guidelines. We did not include guidelines that were not available in
English, Dutch, and German; guidelines that were on prevention of back pain or occupational
healthcare; and guidelines that were not officially approved. The final list of guidelines that we
evaluated is:
1. Quebec Task Force on Spinal Disorders; Canada (Spitzer et al. 1987).
2. Agency for Health Care Policy and Research (AHCPR); United States (Bigos et al. 1994).
3. Dutch College of General Practitioners (NHG); the Netherlands (Faas et al. 1996), update
(Chavannes et al. 2005).
4. Israeli Low back Pain Guideline Group; Israel (Reis & Lahad 2007).
5. Royal College of General Practitioners (RCGP); United Kingdom (Hutchinson et al. 1996;
updates 1999, 2001).
6. National Advisory Committee on Health and Disability; New Zealand (National Health
Committee 1997, update 2004; Kendall et al. 1997).
7. Finnish Medical Association (Duodecim); Finland (Malmivaara et al. 1999).

8. Swiss Medical Society (FMH); Switzerland (Keel et al. 1998).
9. Veterans Health Administration (VHA); United States, 1999.
10. Council on Technology Assessment in Health Care (SBU); Sweden (Nachemson & Johnson
2000).
11. Danish Institute for Health Technology Assessment; Denmark 2000.
12. Drug Committee of the German Medical Society; Germany 2000, update 2007.
13. Florida Agency for Health Care Administration & Department of Health; United States
1999, update 2000.
14. International Paris Task Force, France/Canada (Abenhaim et al. 2000).
15. Dutch Physiotherapy Guidelines (KNGF); the Netherlands (Bekkering et al. 2003).
16. Institute for Clinical Systems Improvement (ICSI); United States 2001, update 2005.
17. Philadelphia Panel; United States 2001.
18. COST B13 Working Group on Guidelines for the Management of Acute Low Back Pain in
Primary Care (van Tulder et al. 2006).
19. COST B13 Working Group on Guidelines for Chronic Low Back Pain in Primary Care
(Airaksinen et al. 2006).
20. American College of Physicians; American Pain Society Low Back Pain Guidelines Panel,
United States (Chou et al. 2007).
21. Center of Excellence for Orthopaedic Pain Management Austria, (Friedrich & Likar 2007).
22. The Care and Research Institute (IRCCS), Italy (Negrini et al. 2006).
23. National Health and Medical Research Council, 2003, Australia.
24. Clinic on Low-Back Pain in Interdisciplinary Practice (Clip) Guidelines. Canada (Rossignol
et al. 2007).
25. Royal Dutch Association for Physiotherapy; Manual Therapy Guidelines; the Netherlands
2005 (Heijmans et al. 2003).
26. Dutch Quality Institute for Healthcare; the Netherlands 2003.
22.3 Quality of guidelines

Since guidelines provide recommendations aimed at influencing or changing clinicians’ behav-
iour, it is important for potential users to critically appraise the internal and external validity
of these recommendations. The AGREE instrument is a helpful tool to help this process (The
AGREE collaborative group 2000; www.agreecollaboration.org). The AGREE instrument consists
of 23 items in six domains (Table 22.2): (1) scope and purpose of the guideline, (2) stakeholder
involvement, (3) rigour of development, (4) clarity and presentation, (5) applicability, and
(6) editorial independence. Each item is rated on a 4-point scale: 1= strongly disagree, 2 = disa-
gree, 3 = agree, and 4 = strongly agree. Domain scores are calculated by dividing the difference
between the obtained score and the minimum possible score by the difference between the maxi-
mum possible and minimum possible score. MacDermid et al. (2005) showed that the AGREE
instrument is reliable and valid for assessing the quality of clinical guidelines relevant to physical
therapy practice.
INTERNATIONAL GUIDELEINES FOR BACK PAIN TREATMENT 421
Table 22.1 Domains and items of the AGREE instrument (www.agreecollaboration.org)
AGREE domain AGREE item

Scope and The overall objective of the guideline is specifically described
purpose
The clinical question covered by the guideline is specifically described
The patients to whom the guideline is meant to apply are specifically described
Stakeholder The guideline development group includes individuals from all the relevant
involvement professional groups
The patients’ views and preferences have been sought
The target users of the guideline are clearly defined
The guideline has been piloted among target users
Rigour of Systematic methods were used to search for evidence
development
The criteria for selecting the evidence are clearly described
The methods used for formulating the recommendations are clearly described
The health benefits, side effects and risks have been considered in formulating the
recommendations
There is an explicit link between the recommendations and the supporting evidence
The guideline has been externally reviewed by experts prior to its publication
A procedure for updating the guideline is provided
Clarity and The recommendations are specific and unambiguous
presentation
The different options for the management of the condition are clearly presented
Key recommendations are easily identifiable
The guideline is supported with tools for application
Applicability The potential organizational barriers in applying the recommendations have been
discussed
The potential cost implications of applying the recommendations have been
considered
The guideline presents key review criteria for monitoring and/or audit purposes
Editorial The guideline is editorially independent from the funding body
independence
Conflicts of interest of guideline development members have been recorded
The AGREE group has stated that it is not yet possible to set a threshold score for the domain
scores to make a distinction between ‘good’ and ‘bad’ guidelines. The AGREE instrument was
developed through consensus among experts. There is little scientific evidence that the criteria
included in the AGREE instrument, regardless of how logical and useful they may seem, are
indeed associated with biased recommendations. Future research in this area may lead to a greater
understanding of the relevance of each item.
In general, the quality of most guidelines was satisfactory (Table 22.2). The domains that were
least often addressed by the guideline development committees were stakeholder involvement,
applicability, and editorial independence, with mean AGREE scores of 52, 40, and 40, respec-
tively. The guidelines scored best on the domains clarity and presentation (mean score 81) and
scope and purpose (mean score 74). Many presented unambiguous recommendations, different
Table 22.2 Scores on the AGREE domains
Guidelines Domains
Scope and Stakeholder Rigour of Clarity and Applicability Editorial
purpose involvement development presentation independence
1. QTF (CAN) 100 42 48 92 22 67
2. AHCPR (US) 100 83 86 83 33 33
3. NHG (NL) 78 83 48 75 33 33
4. ISR 67 42 38 63 33 33
5. RCGP (UK) 89 83 86 83 33 33
6. NZ 81 58 88 100 44 42
7. FIN 56 50 57 67 33 33
8. SUI 67 50 38 63 33 33
9. VHA (US) 100 33 48 92 11 17
10. SWE 100 50 86 83 44 33
11. DK 89 58 33 83 56 33
12. GER 78 38 88 96 96 83
13. Florida (US) 67 33 10 67 0 0
14. Paris 78 42 62 75 44 33
15. KNGF (NL) 44 50 38 42 0 0
16. ICSI (US) 78 50 50 100 89 42
17. Philadelphia 89 75 76 83 33 50
(US)
18. COST B13 61 50 81 67 72 92
acute
19 COST B13 61 46 74 75 44 17
chronic
20. ACP (US) 89 46 95 92 56 83
21. Austria 67 54 48 46 11 8
22. Italy 39 46 62 88 44 17
23. Australia 61 79 98 92 56 50
24. CLIP (CAN) 22 25 43 92 22 25
25. MT (NL) 89 50 93 100 28 58
26. CBO (NL) 61 75 76 100 61 33
1) Spitzer et al. 1987; 2) Bigos et al. 1994; 3) Faas et al. 1996, update Chavannes et al. 2005; 4) Reis and Lahad 2007;
5) Hutchinson et al. 1996; updates 1999, 2001; 6) National Health Committee 1997, update 2004; Kendall et al. 1997;
7) Malmivaara et al. 1999; 8) Keel et al. 1998; 9) Veterans Health Administration (VHA); United States, 1999;
10) Nachemson and Johnson 2000; 11) Danish Institute for Health Technology Assessment; Denmark 2000; 12) Drug
Committee of the German Medical Society; Germany 2000, update 2007; 13) Florida Agency for Health Care
Administration & Department of Health; United States 1999, update 2000; 14) Abenhaim et al. 2000; 15) Bekkering et al.
2003; 16) Institute for Clinical Systems Improvement (ICSI); United States 2001, update 2005; 17) Philadelphia Panel;
United States 2001; 18) van Tulder et al. 2006; 19) Airaksinen et al. 2006; 20) Chou et al. 2007; 21) Friedrich & Likar
2007; 22) Negrini et al. 2006; 23) National Health and Medical Research Council, 2003, Australia; 24) Rossignol et al.
2007; 25) Heijmans et al. 2003; 26) Dutch Quality institute for Healthcare; the Netherlands 2003.
options for management and key recommendations were easily identifiable. Furthermore, almost
every guideline had good scores on the domain of rigour of development (mean score 62).
Potential conflicts of interest were not explicitly described in most of the guidelines. Only the
Quebec Task Force clearly reported that the editorial board was independent from the funding
body, and the Danish and Philadelphia Panel guidelines gave the impression of editorial inde-
pendence. As most guidelines were supported by either a professional association (for example,
the Dutch guidelines by the College of General Practitioners in the Netherlands), a national
organization (for example, the Danish guidelines by the National Institute of Health), or a com-
mercial organization (for example, the Paris Task Force by Mayoly-Spindler Laboratories) it
seems important to explicitly state whether the final recommendations were developed independ-
ently from the supporting agency. Also, other potential conflicts of interest (for example, finan-
cial support from pharmaceutical companies for research, advice, or travelling) should be disclosed
as they may increase the potential risk of biased recommendations.
Few guidelines included tools to facilitate their application in clinical practice. Some included
only one, such as a patient booklet in the UK guidelines, or a screening questionnaire in the New
Zealand guidelines. Guidelines of the Agency for Health Care Policy and Research in the US
provided a set of clinical decision algorithms, a patient booklet and a quick reference guide for
clinicians, and the Dutch General Practitioners’ guidelines included a summary document, a
patient information letter, an ‘expertise improvement package’, and a programme for individual
re-education. The Quebec Task Force included evaluation forms for diagnostic and therapeutic
procedures, consultation, and ergonomic description of work, and matrices for diagnostic and
therapeutic procedures.
In sum, many international guidelines for the management of LBP have been published and
their quality of development and reporting seem to vary (van Tulder et al. 2004). At present, the
AGREE instrument is probably more known among guideline developers than 10 years ago when
the AGREE instrument was just published. The most recently published guidelines seem to meet
more AGREE items compared with the older guidelines. However, there still is room for improve-
ment. We have identified which AGREE items need special attention in future guidelines or
updates of existing guidelines.
22.4 Recommendations on diagnostics and treatment

Although there are some differences between guidelines, most guidelines have similar recommen-
dations for diagnostics and treatment of acute and chronic LBP. Below the recommendations of
the European guidelines are presented, because this was a large initiative in which LBP experts
from many European countries participated. It probably best reflects international consensus on
the management of LBP.
22.5 European guidelines

To increase consistency in the management of non-specific LBP across countries in Europe, the
European Commission approved a programme for the development of European guidelines for
the management of low back pain, called ‘COST B13’. The main objectives of this COST action
were:
◆ Developing European guidelines for the prevention, diagnosis and treatment of non-specific
LBP.
◆ Ensuring an evidence-based approach through the use of systematic reviews and existing
clinical guidelines.
◆ Enabling a multidisciplinary approach; stimulating collaboration between primary healthcare

providers and promoting consistency across providers and countries in Europe.
◆ Promoting implementation of these guidelines across Europe.
This project started in 1999 and the guidelines were finalized in 2004. To ensure an evidence-
based approach, recommendations were based on Cochrane and other systematic reviews and on
existing national guidelines. The European guidelines could be used as a basis for future national
guidelines or future updates of existing national guidelines. The European guidelines could also
help healthcare providers to make evidence-based decisions, improve the quality and outcome of
healthcare, lead to a more rational and efficient use of resources, and identify gaps in the existing
scientific evidence in order to prioritize future research.
The target population of the guidelines consists of healthcare providers who are reached
through individuals or groups that are going to develop new guidelines or update existing guide-
lines, and their professional associations that will disseminate and implement these guidelines.
Indirectly, these guidelines also aim to inform the general public, low back pain patients, and
policy makers in Europe.
Four working groups have been initiated within this COST B13 action that produced the final
guidelines on prevention, management of acute LBP, management of chronic low back pain, and
management of pelvic girdle pain. The experts in the groups represented all countries that had
issued guidelines for low back pain or were developing guidelines and all relevant health profes-
sions. The recommendations for the management of acute and chronic LBP are summarized in
Boxes 22.1 and 22.2.
22.6 Implementation of guidelines

Development and dissemination of guidelines does not automatically mean that healthcare
providers will read, understand, and use the guidelines. Passive dissemination of information is
generally ineffective and specific implementation strategies are necessary to establish changes in
practice. Systematic reviews have shown that a clear and strong evidence base, clear messages,
consistent messages across professions, clear sense of ownership, communication with all relevant
stakeholders, charismatic leadership, continuity of care, continuous education, and continuous
evaluation are successful ingredients for implementation of guidelines (Grimshaw et al. 2001;
Burgers et al. 2003). However, several barriers to implementation of guidelines have been identi-
fied. The practice behaviour of health professionals may be influenced by a lack of knowledge,
a shortage of time, disagreement with the guideline content, or reluctance from colleagues
to adhere to the guideline (Cabana et al. 1999; reviewed in this volume in Chapters 20 and 21).
Furthermore, health professionals may get lost in the large number of different guidelines
received.
Dissemination needs to be supplemented by active implementation strategies. Development of
an implementation plan is an essential first step. It may be helpful to identify local champions for
change, who are influential people that may motivate their peers to accept and start using the
guideline. Setting clear, specific and realistic objectives, and identifying barriers to change behav-
iour of health professionals should have priority in getting evidence into clinical practice. Because
there is not one single intervention that is clearly more effective than others, a multifaceted
approach starting with strategies that are likely to have visible results is recommended. Recently,
a number of trials have evaluated implementation strategies for low back pain guidelines
(Rossignol et al. 2000; McGuirk et al. 2001; Bekkering et al. 2003; Dey et al. 2004; Engers et al.
2005). Four of these studies were randomized trials, one a non-randomized trial. All studies com-
pared the implementation strategy with usual care by general practitioners or physiotherapists.
Box 22.1 European guidelines on the management of acute

low back pain
Summary of recommendations for diagnosis of acute non-specific
low back pain
◆ Case history and brief examination should be carried out.
◆ If history-taking indicates possible serious spinal pathology or nerve root syndrome, carry out
more extensive physical examination including neurological screening when appropriate.
◆ Undertake diagnostic triage at the first assessment as basis for management decisions.
◆ Be aware of psychosocial factors, and review them in detail if there is no improvement.
◆ Diagnostic imaging tests (including X-rays, computed tomography [CT] and magnetic
resonance imaging [MRI]) are not routinely indicated for non-specific LBP.
◆ Reassess those patients who are not resolving within a few weeks after the first visit, or those
who are following a worsening course.
Summary of recommendations for treatment of acute non-specific

low back pain
◆ Give adequate information and reassure the patient.
◆ Do not prescribe bed rest as a treatment.
◆ Advise patients to stay active and continue normal daily activities including work if possible.
◆ Prescribe medication, if necessary for pain relief; preferably to be taken at regular intervals;
first choice paracetamol, second choice non-steroidal anti-inflammatory drugs (NSAIDs).
◆ Consider adding a short course of muscle relaxants on its own or added to NSAIDs, if para-
cetamol or NSAIDs have failed to reduce pain.
◆ Consider (referral for) spinal manipulation for patients who are failing to return to normal
activities.
◆ Multidisciplinary treatment programmes in occupational settings may be an option for
workers with subacute LBP and sick leave for more than 4–8 weeks.
The three randomized trials reported some improvement on process outcomes (an increase in
adherence to recommendations), but did not show any significant differences on patient outcomes.
Although guidelines are expected to improve patient outcomes, the evidence is sparse and not
yet convincing. This is a bit disappointing, because one would expect that successful implementa-
tion of evidence-based guidelines is associated with improved patient outcomes. Failure to find
positive results may be related to insufficient content and/or intensity of the implementation
strategies. But the control groups may also have performed better that expected. General practi-
tioners and physiotherapists who participated in the trials may have volunteered because they
were interested in LBP, and they may have already been familiar with the latest evidence in this
field. This limitation potentially decreased the contrast in guideline adherence between the imple-
mentation and control groups in the trials, and may be the most important reason for the lack of
difference in patient outcomes between groups. It is also possible that general practitioners and
physiotherapists in the control groups read the guidelines better than they would have done if
they had not participated in these trials.
Box 22.2 European guidelines on the management

of chronic low back pain
Summary of diagnosis in chronic low back pain
Patient assessment
◆ Physical examination and case history: the use of diagnostic triage, to exclude specific spinal
pathology and nerve root pain, and the assessment of prognostic factors (‘yellow flags’) are
recommended. We cannot recommend spinal palpatory tests, soft tissue tests and segmental
range of motion, or straight leg raising tests (Laseque) in the diagnosis of non-specific
chronic LBP.
◆ Imaging: we do not recommend imaging (plain radiography, CT or MRI), bone scanning,
SPECT, discography or facet nerve blocks for the diagnosis of non-specific chronic low back
pain unless a specific cause is strongly suspected.
◆ MRI is the best imaging procedure for use in diagnosing patients with radicular symptoms,
or for those in whom discitis or neoplasm is suspected.
◆ Electromyography: we cannot recommend electromyography for the diagnosis of non-
specific chronic low back pain.
Prognostic factors
Assessment of work-related factors, psychosocial distress, depressive mood, severity of
pain and functional impact, prior episodes of LBP, extreme symptom reporting and patient
expectations should be included in the diagnosis of patients with non-specific chronic low
back pain.
Summary of treatment of chronic low back pain

◆ Conservative treatments: cognitive-behavioural therapy, exercise therapy, brief educational
interventions, and multidisciplinary (biopsychosocial) treatment can each be recommended
for non-specific CLBP. Back schools and short courses of manipulation can also be consid-
ered. The use of physical therapy (TENS, heat/cold, traction, laser, ultrasound, short wave,
interferential, massage, corsets) cannot be recommended.
◆ Pharmacological treatments: noradrenergic or noradrenergic-serotoninergic antidepres-
sants, weak opioids, and the short term use of NSAIDs, muscle relaxants, and capsicum
plasters can be recommended for pain relief; strong opioids can be considered in patients
who do not respond to all other treatment modalities.
◆ Invasive treatments: acupuncture, epidural corticosteroids, intra-articular (facet) steroid
injections, local facet nerve blocks, intradiscal injections, trigger point injections, botulinum
toxin, prolotherapy, radiofrequency facet denervation, intradiscal radiofrequency lesioning,
intradiscal electrothermal therapy, radiofrequency lesioning of the dorsal root ganglion, and
spinal cord stimulation cannot be recommended for CLBP. Percutaneous electrical nerve
stimulation (PENS) and neuroreflextherapy can be considered where available.
Surgery for non-specific CLBP cannot be recommended unless 2 years of all other recom-
mended conservative treatment (inclusive multidisciplinary approaches with combined pro-
grammes of cognitive intervention and exercises) have failed, or such combined programmes
are not available, and only then in carefully selected patients.
22.7 Discussion
Since the available evidence is international, one would expect that each country’s guidelines
would give more or less similar recommendations regarding diagnosis and treatment. Comparison
of clinical guidelines for the management of LBP in primary care from 11 different countries
showed that the content of the guidelines regarding the diagnostic classification (diagnostic
triage) and the use of diagnostic and therapeutic interventions is quite similar. However, there
were also some discrepancies in recommendations across guidelines (Koes et al. 2001). Differences
in recommendations between guidelines may be due to the incompleteness of the evidence, dif-
ferent levels of evidence, magnitude of effects, side effects and costs, differences in healthcare
systems (organization/financial), or differences in membership of guidelines committees. More
recent guidelines may have included more recently published trials and, therefore, may end up
with slightly different recommendations compared with older guidelines. Also, guidelines may
have been based on systematic reviews that included trials in different languages; the majority of
existing reviews have considered only studies published in a few languages, and several, only those
published in English.
Recommendations in guidelines are not only based on scientific evidence, but also on consen-
sus. Guideline committees may consider various arguments differently, such as the magnitude of
the effects, potential side effects, cost-effectiveness, and current routine practice and available
resources in their country. Especially as we know that effects in the field of LBP, if any, are usually
small and short-term effects only (Keller et al. 2007), interpretation of effects may vary among
guideline committees. Also, guideline committees may differently weigh other aspects such as
side effects and costs. The constitution of the guideline committee and the professional bodies
they represent, may introduce bias—either for or against a particular treatment. This does not
necessarily mean that one guideline is better than the other or that one is right and the other is
wrong. It merely shows that when translating the evidence into clinically relevant recommenda-
tions more aspects play a role, and that these aspects may vary locally or nationally.
However, most international guidelines on the management of LBP include similar recom-
mendations in diagnostics and treatment. During the last 10–15 years, various guideline commit-
tees around the world have evaluated the available evidence, considered other aspects such as side
effects, costs, patient preferences and expectations, and availability, and developed similar recom-
mendations. So there seems to be consensus on optimal management for acute and chronic LBP.
Changing the behaviour of care providers to result in less variation in practice and improved
adherence to guideline recommendations is the biggest challenge for the next decade.
References
Abenhaim, L., Rossignol, M., Valat, J.P. et al. (2000). The role of activity in the therapeutic management of
back pain. Report of the International Paris Task Force on Back Pain. Spine, 25 (4 Suppl.), S1–33.
Airaksinen, O., Brox, J.I., Cedraschi, C. et al. On behalf of the COST B13 Working Group on Guidelines
for Chronic Low Back Pain. (2006). Chapter 4. European guidelines for the management of chronic
nonspecific low back pain. European Spine Journal, 15 (Suppl 2), S192–300.
Bekkering GE, Hendriks HJM, Koes BW, et al. National Practice Guideline for the physiotherapeutic
management of patients with low back pain. Physiotherapy 2003; 89:82–96 [first published in Dutch
in 2001].
Bigos S, Bowyer O, Braen G. Acute low back problems in Adults. Clinical Practice Guideline no. 14. AHCPR
Publication No. 95–0642. ed. Rockville, MD: Agency for Health Care Policy and Research, Public
Health Service, U.S., Department of Health and Human Services, USA, 1994.
Bombardier C, Esmail R, Nachemson AL, Back Review Group Editorial Board. The Cochrane
Collaboration Back Review Group for spinal disorders. Spine 1997; 22:837–40.
Bouter LM, Pennick V, Bombardier C; The Editorial Board of the Back Review Group. Cochrane back
review group. Spine 2003; 28:1215–18.
Burgers JS, Grol RP, Zaat JO, Spies TH, van der Bij AK, Mokkink HG. Characteristics of effective clinical
guidelines for general practice. Br J Gen Pract 2003; 53:15–19.
Cabana MD, Rand CS, Powe NR, et al. Why don’t physicians follow clinical practice guidelines?
A framework for improvement. JAMA 1999; 282:1458–65.
Chavannes AW, Mens JMA, Koes BW, et al. Dutch general practice guideline for non-specific low back
pain [in Dutch] Huisarts Wet 2005; 48:113–23.
Chou R. Evidence-based medicine and the challenge of low back pain: where are we now? Pain Practice
2005; 5(3):153–78.
Chou R, Qaseem A, Snow V et al. Clinical Efficacy Assessment Subcommittee of the American College of
Physicians; American College of Physicians; American Pain Society Low Back Pain Guidelines Panel.
Diagnosis and treatment of low back pain: a joint clinical practice guideline from the American College
of Physicians and the American Pain Society. Ann Intern Med 2007; 147(7):478–91.
Danish Institute for Health Technology Assessment. Low back pain: frequency, management and
prevention from a health technology perspective. Copenhagen: National Board of Health, 2000.
Dey P, Simpson CWR, Collins SI, et al. Implementation of RCGP guidelines for acute low back pain:
a cluster randomized controlled trial. Br J Gen Pract 2004; 54:33–7.
Drug Committee of the German Medical Society. Recommendations for treatment of low back pain [in
German]. Köln: Drug Committee of the German Medical Society, 2000, update 2007.
Dutch Institute for Healthcare Improvement (CBO). Clinical guideline for non-specific low back pain, 2003
[in Dutch]. Amsterdam: CBO.
Engers AJ, Wensing M, van Tulder MW et al. Implementation of the Dutch low back pain guideline for
general practitioners: a cluster randomized controlled trial. Spine 2005; 30:559–600.
Faas A, Chavannes AW, Koes BW et al. NHG-Standaard ‘Lage-Rugpijn’ (in Dutch). Huisarts Wet 1996;
39:18–31.
Florida Agency for Health Care Administration and the Department of Health. Universe of Florida
patients with low back pain or injury: medical practice guidelines. Tallahassee, FL: Florida Agency for
Health Care Administration and the Department of Health, 1999, update 2000.
Friedrich M, Likar R. Evidenz- und konsensusbasierte österreichische Leitlinien für das Management
akuter und chronischer unspezifischer Kreuzschmerzen. Wien Klin Wochenschr 2007; 119(5–6):189–97.
Grimshaw J, Shirran L, Thomas R. Changing provider behavior: an overview of systematic reviews of
interventions. Medical Care 2001; 29:II2–II45.
Heijmans M, Hendriks HJM, Koes BW et al. National Practice Guideline for manual therapy for patients
with low back pain [in Dutch]. Ned Tijdschr Fys 2003; 113 (Suppl. 3):1–24.
Hutchinson A, Waddell G, Feder G, Breen A, Burton K, Sears C. Clinical Guidelines for the Management
of Acute Low Back Pain. London: Royal College of General Practitioners, UK, 1996 (updated in
1999, 2001).
Institute for Clinical Systems Improvement (ICSI). Adult low back pain. Bloomington, MN: Institute for
Clinical Systems Improvement (ICSI), 2005.
Keel P, Weber M, Roux E, Gauchat M-H, Schwarz H, Jochum H. Kreuzschmerzen: Hintergründe,
Prävention, Behandlung. Bern: Verbindung der Schweizer Ärzte, FMH; Switzerland, 1998.
Keller A, Hayden J, Bombardier C, van Tulder M. Effect sizes of non-surgical treatments of non-specific
low-back pain. Eur Spine J 2007; 16:1776–88.
Kendall N, Linton S, and Main C. Guide to Assessing Psychosocial Yellow Flags in Acute Low Back Pain: Risk
Factors for Long-Term Disability and Work Loss. Wellington: National Advisory Committee on Health
and Disability, and ACC (Accident Rehabilitation and Compensation Insurance Corporation), 1997.
Koes BW, Van Tulder MW, Ostelo R, Kim BA, Waddell G. Clinical guidelines for the management of low
back pain in primary care: an international comparison. Spine 2001; 26:2504–13.
Laerum E, Storheim K, Brox JL. New clinical guidelines for low back pain. Tidsskr Nor Laegeforen 2007;
127(20):2706.
MacDermid JC, Brooks D, Solway S, et al. Reliability and validity of the AGREE instrument used by
physical therapists in assessment of clinical practice guidelines. BMC Health Serv Res 2005; 5:18.
Malmivaara A, Kotilainen E, Laasonen E, Poussa M, Rasmussen M. Clinical Practice Guidelines of the
Finnish Medical Association Duodecim. Diseases of the low back. Duodecim: The Finnish
Medical Association 1999 (Finland) (available in English).
McGuirk B, King W, Govind J, Lowry J, Bogduk N. Safety, efficacy, and cost effectiveness of evidence-
based guidelines for the management of acute low back pain in primary care. Spine 2001; 26:2615–22.
Nachemson AL, Jonsson E. (Eds.) Neck and back pain: the scientific evidence of causes, diagnosis, and
treatment. Philadelphia, PA: Lippincott Williams & Wilkins, 2000.
National Health and Medical Research Council Australian Acute Musculoskeletal Pain Guidelines Group.
Management of acute musculoskeletal pain. Brisbane: National Library of Australia Cataloguing-
in-Publication, 2003.
National Health Committee. National Advisory Committee on Health and Disability, Accident
Rehabilitation and Compensation Insurance Corporation. New Zealand Acute Low Back Pain Guide.
Wellington, New Zealand; 1997, update 2004.
Negrini S, Giovannoni S, Minozzi S, et al. Diagnostic therapeutic flow-charts for low back pain patients:
the Italian clinical guidelines. Euro Medicophys 2006; 42(2):151–70.
Philadelphia panel. Philadelphia panel evidence-based clinical practice guidelines on selected rehabilitation
interventions for low back pain. Phys Ther 2001; 81:1641–74.
Reis S, Lahad A. Clinical guidelines for diagnosis and treatment of acute low back pain. Harefuah 2007;
146(8):631–5, 644.
Rossignol M, Abenhaim L, Seguin P et al. Coordination of primary health care for back pain. A
randomized controlled trial. Spine 2000; 25:251–9.
Rossignol M, Arsenault B, Dionne C et al. Clinic on Low-Back Pain in Interdisciplinary Practice (Clip)
Guidelines. 2007. http://www.santpub-mtl.qc.ca/clip.
Sackett DL, Rosenberg WM, Gray JA, et al. Evidence based medicine: what it is and what it isn’t. BMJ 1996;
312:71–72.
Spitzer WO, LeBlanc FE, Dupuis M et al. Scientific approach to the assessment and management of
activity-related spinal disorders. Spine 1987; 12(Suppl. 7):S1–S59.
The AGREE (Appraisal of Guidelines, Research, and Evaluation in Europe) Collaborative Group.
Guideline development in Europe. An international comparison. Int J Technol Assess Health Care 2000;
16:1039–49.
The Dutch Institute for Healthcare Improvement (CBO). Clinical guideline for non-specific low back pain,
2003 [in Dutch].
Van Tulder MW, Tuut M, Pennick V et al. Quality of primary care guidelines for acute low back pain.
Spine 2004; 29(17):E357–62.
Van Tulder MW, Becker A, Bekkering T et al. European guidelines for the management of acute low back
pain in primary care. Eur Spine J 2006; 15(Suppl 2):S169–91.
Veterans Health Administration, Department of Defense. Clinical practice guideline for the management
of low back pain or sciatica in the primary care setting. Washington (DC): Department of Veterans
Affairs (US); 1999.
Wong DA, Mayer T, Watters W, et al. Unremitting low back pain: North American Spine Society Phase III
Clinical Guidelines for multidisciplinary spine care specialists. La Grange, Ill.: North American Spine
Society, 2000:96.
Part 7
Clinical Approaches for

Patients with Acute and
Subacute Low Back Pain
Chapter 23
Engaging Patients in their Own Care

for Back Care: The Role of Education
and Advice in the Prevention
of Chronic Pain and Disability
Chris J. Main and Kim Burton
23.1 Introduction
There is a spectrum of views about the actual and potential role of education and advice in the
prevention of chronic pain and disability. There is on the one hand an optimism bordering on the
naïve that an essential feature of the problem of chronic low back pain (CLBP) and disability is a
simple knowledge deficit (on the part of patients), the correction of which by provision of knowl-
edge and advice will be sufficient to prevent chronic pain or disability. Then there is an opposing
view that education and advice are valueless or that patients are uneducable. In attempting to find
a middle way between these two extremes, we begin with a review of the evidence, viewed from a
number of perspectives. In considering the role of education for patients we shall also consider
the role of the healthcare professional (HCP). We shall then attempt to integrate the findings into
an action plan, with suggestions for a refocusing of our educational efforts in secondary preven-
tion in patients with recurrent or persistent pain in which our principal objective is the preven-
tion of unnecessary disability or distress rather than the prevention of further pain per se which is
unlikely to be achievable.
In n developing an analysis or our current state of knowledge we shall consider: (1) the educa-
tional process; (2) types of intervention; (3) effectiveness and sustainability; (4) influence of
patient characteristics; and (5) the influence of health professionals, beliefs and treatment orien-
tations. We shall then consider: (1) education content for LBP patients; (2) the educational
context; and (3) the development of an educational strategy. Finally, we will then attempt to draw
a number of conclusions.
We shall, where possible, draw specifically on education/advice in the context of LBP and other
musculoskeletal conditions, in which the provision of education and advice has been examined.
However, we will focus particularly on their role in tacking obstacles to recovery and optimal
function in the context of secondary prevention.
Our primary focus will be on the prevention of unnecessary limitations in functioning and
adverse effects on well-being rather than on the prevention of persistent pain per se. Finally, by
way of introduction, we declare our own starting point, or biases. We view the provision of
education and advice as a core element of clinical management; and we have attempted to be
open-minded about the value of education and advice. Thus we begin with an evaluation of
research findings. We suspect however that education and advice may be a necessary but not
sufficient condition for producing effective behaviour change. We are mindful therefore, in the
appraisal of effectiveness of educational strategies, to require consideration of the effect of knowl-

edge in the context of overall clinical or self-directed management, rather than restrict our atten-
tion to the establishment of improved understanding of low back pain (LBP) and its effects,
important though that is, in addressing obstacles to optimal function and well-being.
23.2 Current state of knowledge

23.2.1 The educational process
Any web-based search for information about LBP or advice on its management will yield an
astonishing quantity and quality of material, ranging from unashamed commercial pitches, via
thinly disguised pseudo-scientific studies promising miracle cures, to properly conducted scien-
tific studies illustrating the limitations of our knowledge. Education and advice have, in fact, been
included explicitly within the behaviourally-based approach to prevention and rehabilitation
developed in the Back Schools, pioneered in companies such as Volvo in Sweden in the early
1970s. Backs schools offered a strategy for primary prevention, but subsequently became a thera-
peutic strategy in the management of LBP for those who had already developed LBP, and later
became an educational ‘vehicle’ widely adopted in clinical treatment.
General educational principles

Wensing and Grol (2005) describe the five different types of educational intervention highlighted
by the Cochrane EPOC group (Thorsen and Makele, 1999): (1) provision of educational materi-
als: (2) large-scale educational meetings: (3) small-scale educational meetings: (4) outreach visits;
and (5) use of opinion leaders. They summarize the conclusions of 39 reviews of controlled trials
and conclude that:
the effects on professional behaviour and health outcomes of reading educational materials or attend-
ing courses are often limited. Small-scale and interactive education may be more effective, but little
research evidence supports this. Factors that may increase the effectiveness of education include longer
time period, an appropriate group compensation, needs assessment before the activity, active partici-
pation, individualisation and the use of local opinion leaders (p. 147).
Wensing and Grol offered, however, the further observation that ‘Although the effects of edu-
cation on behaviour may be limited, it is often a first necessary step in a process of implementa-
tion of innovations. Education is particularly valuable if it is part of a broader implementation
strategy that includes other interventions as well’ (p. 147). Although these reflections are directed
particularly at the education of HCPs they illustrate the general principle that effective education
needs to be embedded within targeted implementation strategies, and suggest that education of
LBP patients similarly might need to be linked with specific behavioural change strategies.
Researching the effectiveness of educational interventions

Hutchinson and Baker (1999) having noted that HCPs were often reluctant to value research into
the effectiveness of educational interventions, and having identified a number of factors which
might influence the effectiveness of intervention, stressed the need for ‘rigorously designed
research’ as a priority in medical education and drew attention to the increase in behavioural
complexity as the focus of the educational endeavour widened
Education about the nature of LBP

Waddell (2004) offers a fascinating historical view of different ways in which LBP has been under-
stood. These differing perspectives have in turn influenced our approach to patient education.
There has been a plethora of educational materials developed about LBP, but traditionally,
PATIENT ENGAGEMENT 435
LBP has been viewed from three distinct, but related, perspectives. Firstly, LBP has been viewed
primarily as a structural problem and much educational material has reflected this. Typically,
illustrations of the spine with its vertebrae, nerves and associated musculature are accompanied
by illustrations of ‘slipped discs’, or other physical effects of trauma and disease. Underpinning
this approach has been an attempt to provide reassurance about safe activity, but (arguably) pre-
senting the back as a weak and vulnerable structure. Until relatively recently, this primarily
‘orthopaedic view’ of the origins of LBP was at the heart of most educational material, and
although not inaccurate was unrepresentative of the much larger proportion of LBP for which no
structural abnormality, physiological abnormality or disease can be identified. Secondly, the ‘bio-
mechanical view’, locating the source of LBP in terms of misalignments of various sorts to be
corrected by spinal manipulation and mobilization, has underpinned the manual therapies, and
evident particularly in chiropractic and osteopathy. This is similar to the orthopaedic view in
requiring some sort of passive treatment, delivered by an HCP as a corrective. Thirdly, and per-
haps most common of all, has been the plethora of pharmacological approaches designed to abol-
ish or ameliorate the pain signal itself, in which the ‘educational’ component is confined primarily
to advice about dose and warnings about side-effects.
Considered together, the impact of these three general approaches has been a primary focus
on ‘abnormality’ and promulgated the view that the back was ‘fragile’ and that every caution
must be taken to avoid damaging it, but none of these general approaches has focused explicitly
on the potential benefits of self-help or equipped patients to embark on more effective self-
management.
Education about the management of LBP

Waddell (2004) contrasts traditional biomedical education with some of the early attempts to
address beliefs about fears of hurting and harming and offer simple messaging about positive
adaptive coping as illustrated in the psychosocial leaflet developed by Symonds et al. (1995), and
its later derivative The Back Book (Roland et al. 2002), an evidence based attempt to provide
biopsychosocial information and advice.
In fact even the earliest pain management programmes advocated a fairly strict behavioural
approach (Fordyce 1976), incorporated educational components on the nature of pain behaviour
and strategies designed to restore function. The later development of cognitive-behavioural
therapy (CBT) to problems of chronic pain (Turk et al. 1983) gave considerable impetus to the
investigation of pain cognitions, with the development of a host of psychometric tools designed to
capture patient beliefs and expectations. It had become evident to pain clinicians that unhelpful or
mistaken beliefs were common features of patients presenting with chronic pain. Iatrogenic
confusion and distress were frequently observed in patients attending pain management pro-
grammes (Main et al. 2008). An essential component in addressing such unhelpful beliefs (and
maladaptive pain behaviours) was the provision of information about the nature of pain, pain
mechanisms, the role of emotion and the relationship between pain and disability as a precursor to
tackling these powerful psychological obstacles to improved function. Education and the provision
of advice therefore have always been at the heart of pain management, but it has not been possible
to disentangle their specific worth within such a complex type of therapeutic intervention.
23.2.2 Types of intervention

Engers et al. (2008) have defined patient education as ‘any set of planned condition-specific
educational activities in a one-to-one situation, designed to improve patients’ health behaviours
and health status in regard to the low back problem’ (p. 21), a reasonable starting point perhaps,
(apart from the exclusion of education delivered on a group basis), but potentially including a
wide range of interventions ranging from a five minute oral information session to multidiscipli-
nary 3-hour sessions, although they acknowledge a ‘thin line between individual patient education
lasting several hours, psycho-education and counselling’. For the purpose of this review we shall
focus firstly on public health initiatives and secondly on initiatives directed at LBP consulters.
Public health initiatives and media campaigns

Prior back trouble is a strong predictor of future recurrence (Burton and Tillotson 1991). This has
led to the proposal that those who have survived previous episodes of back pain with little trouble
are at reduced risk of future disability, compared with those who present for the first time or those
who have had more severe prior episodes (Hazard et al. 1996, McIntosh et al. 2000). Therefore,
interventions that aim to alter community views, targeted at populations as a whole, may be an
effective way of improving outcomes from back pain. Population-based approaches clearly have
potential benefits. Modifying the knowledge or attitudes of a large proportion of the community
simultaneously, provides social support for behavioural change and maintenance of change over
time (Redman et al. 1990); and, because of the ubiquitous nature of back pain, even small or
modest impacts in those at low or medium risk are likely to deliver large improvements on a
population-based scale (Rose, 1992). Importantly, shifting the whole distribution of population
beliefs invariably alters the beliefs of those hard to identify high-risk individuals (Redman et al.
1990), and may prime the population for more targeted approaches
Several countries have performed mass media campaigns in an attempt to modify population
beliefs about back pain, with varying degrees of success (Buchbinder et al. 2001; Waddell et al.
2007; Werner et al. 2008). The outcomes to date have shown concordant results with respect to
shifting public attitudes and beliefs about back pain towards being more optimistic and pro-
active, yet have yielded mixed results with respect to actual behaviour change (Buchbinder et al.
2008). It would appear that more intensive and expensive media campaigns may be more effective
than low-budget campaigns, television may be more effective than radio and print media, and
that explicit recommendations regarding work may be needed if changes in work-related out-
comes are wanted (Buchbinder et al. 2008). Furthermore, to ensure consistency and reinforce-
ment of the messages of a media campaign, it is important to garner widespread support for the
key messages from all the important stakeholders. Indeed social influences have also been shown
to play a more important role than scientific influences in shaping the behaviours and medical
decisions of physicians (Dixon 1990).
Education and advice within a cognitive-behavioural approach

During the past two or three decades, however, there has been increasing recognition of the
importance of cognitive factors and this has been reflected in the design of a range of research
studies, particularly in the treatment of low back pain, which have included a focus to a greater or
lesser extent on the modification of beliefs and behaviour, in the context of early intervention.
Clearly all therapeutic encounters incorporate, at least to some extent, a focus on cognitions and
behaviour, but CBT differs in that the explicit focus of intervention is on the modification of
thoughts and behaviour, rather than medical or biomechanical abnormality. The psychosocial
component within the cognitive-behavioural approach (CBA) is offered within a framework of
reactivation and therefore should be viewed more as ‘psychologically informed physiotherapy’
rather than psychotherapy per se.
Specific evaluations of educational interventions in LBP

An early randomized controlled trial (RCT) of The Back Book versus a traditional biomedical
booklet (Burton et al. 1999) demonstrated a significant shift in inevitability of back problems with
reduction of fear of physical activity in highly fearful LBP patients. Both Von Korff et al. (1998)
and Moore et al. (2000) reduced back-related worries and fear-avoidance beliefs worry in RCTs
of an educational approach that focused on self-management. Albaladejo et al. (2010), in a pri-
mary care RCT, found that the additional of a short education programme on active management
to usual care led to a small but consistent improvement in disability, pain, and quality of life. In
another RCT conducted in Spain, Kovacs et al. (2010) found in a group of institutionalized
elderly subjects that The Back Book and a 20-min group talk improved disability 6 months later.
In the context of early interventions (usually in terms of secondary prevention) the derivation
of simpler psychosocial interventions than ‘full-blown CBT’, containing a specific but narrower
focus on pain- or disability-specific cognitions and behaviour, in patients with less entrenched
disability, has offered an alternative to ‘traditional’ physiotherapy, based on manipulation or
mobilization. Linton and Andersson (2000) conducted an interesting RCT compared a CBT pro-
gramme of six 2-hour sessions with the Symonds et al. (1995) pamphlet (referenced above) and a
conventional packet of ‘biomedical’ material. Although the CBT programme was more effective in
reducing future sickness absence, all three groups improved in pain, fear-avoidance and catastro-
phizing. In such ‘patient-centred’ approaches, the role of education and advice is fundamental.
Education and advice within psychologically informed re-activation.

Education and advice have been core ingredients in the development of CBA approaches to
re-activation delivered in primary and secondary care. Many of the early educational approaches
seemed to be based on assumption of a ‘knowledge deficit’ for which provision of adequate
education was required. Stages in the development of the CBA are shown in Box 23.1.
The psychosocial intervention in the ‘first-wave’ studies included a fairly didactic educational
approach. Thus in an early study comparing back school with an exercise intervention for patients
Box 23.1 Stages in the development of the

cognitive-behavioural approach
Wave 1
◆ Educational approaches (e.g. back schools).
◆ Assumption of a knowledge deficit.
◆ Generic approach.
◆ No assessment of competency in cognitive-behavioural approach.
Wave 2
◆ Specific cognitive and behavioural content.
◆ Generic approach.
◆ No specific training in yellow flag elicitation.
Wave 3
◆ Specific yellow flag training.
◆ Competencies assumed.
◆ Individualized approach.
with chronic back pain, demonstrated that both approaches led to clinical improvement at
6-weeks follow-up (Klaber-Moffett et al. 1986). The back school patients continued to improve
thereafter but the reasons were not fully explored. Education per se thus may be of value. Indeed
in a later controlled study Frost et al. (2004) found that a simple discussion with a physiotherapist
concerning the causes of back pain and advice on how to remain active was as successful in the
long-term as routine physiotherapy, but the level of disability was low and although both groups
improved, the improvements were modest, suggesting that the improvements may have been
suboptimal.
In a second-wave study (including a more developed psychosocial ‘CBT’ component, Klaber-
Moffett et al. 1999), the intervention consisted of a physiotherapist-led exercise class including
strengthening exercises, stretching exercises, relaxation and ‘brief education on back care’. The
results demonstrated that, at 6- and 12-month follow-up, the changes in the intervention group
were significantly greater than in the control group (and at no greater cost), despite the fact that
their intervention was surprisingly brief—only eight 1-hour sessions over 4 weeks. Furthermore
those participants who scored more highly on fear avoidance beliefs were more likely to benefit
from the programme, suggesting that education about fear and avoidance may have been impor-
tant (Klaber et al. 2004). Although it is not known whether such factors were in fact identified and
targeted as such in the intervention group, and the physiotherapists involved seem to have been
given only minimal training in cognitive-behavioural techniques. Nonetheless the study is one of
the first explicitly to recognize the possible influence of cognitive factors on early community-
based interventions for LBP.
George et al. (2003) compared a simple intervention for patients with acute low-back pain,
designed to challenge fear-avoidance beliefs and promote early activation, with standard care.
Although the intervention is described as ‘Fear-Avoidance Physical Therapy’ and would seem to
be a prime candidate for psychologically enhanced physiotherapy, the intervention seems to have
consisted essentially of an initial orientation with The Back Book (Roland et al 2002), followed by
a behaviourally-focused approach to re-activation. Those in the treatment group demonstrated
greater changes in fear-avoidance beliefs although it appears unlikely that there was any sort of
systematic attempt to address cognitions, over and above the educational material. Finally, in the
UK BEAM Trial (UK BEAM Trial Team 2004) exercise, manipulation and manipulation followed
by exercise were compared with ‘best care’ in general practice (which included delivery of The
Back Book and teaching all practice staff the active management approach. Change in patients’
back beliefs in all three ‘experimental’ conditions at 3-month follow-up, with sustained but
reduced change by 12 months suggests that an additional common ‘educational component’ may
have been a feature of all three treatment conditions.
In one of the first third-wave studies, Hay et al. (2005) compared manual therapy with a simple,
individualized CBT pain management approach provided by physiotherapists for patients pre-
senting to their general practitioner (GP) with LBP of less than 12 weeks. A core part of the train-
ing for the physiotherapists included broadly-based training in pain management, with a specific
educational focus on the nature of LBP, pain mechanisms, and the development of disability.
In a later trial, Johnson et al. (2007) compared an active treatment intervention consisting of a
6-week community-based group intervention programme based on active exercise and education
using a CBT approach with a control arm comprising usual GP care supplemented with educa-
tional material. Each of the intervention sessions started with group discussion and ended with
exercise, with the duration spent on exercise increasing throughout the course. Even though ‘the
therapists at times found it difficult to adopt the communication style characteristic of a CBT
approach and some methods, including challenging patients’ beliefs and fears were limited’, the
study showed nonetheless a beneficial influence of patient preference and recommended the
targeting of interventions at subgroups. In the most recent such RCT (Lamb et al. 2010), a cogni-
tive-behavioural intervention for LBP delivered to patients in groups was found to be superior
to ‘best practice advice’, and the difference was sustained at 1yr. follow. The content of the train-
ing programme (Hansen et al. 2010 ) indicates a comprehensive educational package, but
although more cost-effective, the actual clinical difference was relatively small and the research
design did not permit determination of whether the difference was attributable to treatment
content (CBT approach versus best practice advice) or to treatment delivery (group versus indi-
vidualized treatment).
Suggestions for a way forward (Wave 4)

Having reviewed the somewhat disappointing results (in terms of size of effect) of RCTs on early
psychosocial interventions, van der Windt et al. (2008) concluded:
◆ The results of most randomized trials on the effectiveness of psychosocial interventions for
back pain in primary care show only small differences in function or other outcome measures
when compared with active control treatments.
◆ Possible explanations for these ‘negative findings’ include selection of heterogeneous patient
populations; insufficient targeting of interventions on modifiable risk factors; insufficient
competencies of care providers; insufficient intensity or duration of the intervention; inade-
quate adherence to treatment protocols; inadequate assessment of outcome.
In fact Hay et al. (2008) have designed a more individualized and focused approach to interven-
tion linking the screening of patients with a specially developed subgrouping tool (Hill et al. 2008)
designed to identify psychosocial risk factors, with specifically targeted therapeutic objectives
for which the HCPs have been specifically trained. A variety of educational approaches (both
didactic and experimental), including role play with simulated patients and clinical mentoring
were employed. The outcome of this trial is imminent. If successful, it will be the first early inter-
vention trial to have shown that training physiotherapists to identify and target yellow flags can
influence patient outcome in patients at risk of developing chronicity.
23.3.3 Effectiveness and sustainability

Even assuming that an appropriate set of educational messages can be delivered, how effective are
they in effecting desired behaviour change, and are the effects sustained? Engers et al. (2008)
offered a review focusing on advice/information given by an HCP to improve patient under-
standing of LBP problems and what they should do about them. Unfortunately group education
was excluded, as were studies comparing an educational intervention as part of an intervention
programme with other non-educational interventions; such as trials on multidisciplinary treat-
ment or a back school that included patient education compared to manual therapy. The review-
ers were therefore able only to compare individual patient education with either no intervention
or non-educational intervention on the one hand or different types of individual patient educa-
tion on the other. Nonetheless given the perhaps over-restrictive methodological constraints, the
authors felt able to draw a number of conclusions, shown in Box 23.2.
The authors further observe that ‘it is very difficult to evaluate the effects or oral and written
patient education’ and that ‘none of the papers explicitly described the theoretical model on
which the intervention was based’. The recommendations for the design of educational material
are shown in Box 23.3 would still seem to be relevant.
Giving information in the form of a booklet can help. The Back Book (Roland et al. 2002), a
simple low-cost booklet which informs the patient on self-managing back pain and how to keep
active and adopt an active lifestyle, can be used as an aspect of clinical management. The clinician
Box 23.2 Major conclusions from Cochrane Review on

individual patient education for LBP
For individual patients with acute or sub-acute LBP, a 2.5-hour individual patient education
session was more effective than no intervention.
◆ However less intensive patient education did not seem to be more effective than no
intervention.
◆ Individual education appeared to be equally effective to interventions like chiropractic
manipulation and physiotherapy for patients with acute or subacute LBP.
◆ However, for patients with chronic LBP, individual education was less effective than more
intensive treatment and the effectiveness of individual education is unclear.
◆ There is no difference between the effects of various types of individual patient education.
◆ That form of educational intervention is preferred and what content, intensity, and
frequency are best remains unclear.
Adapted from Engers, A. J., Jellema, P., Wensing, M., van der Windt, D.A.W.M., Grol, R., & van
Tulder, M.W. Individual patient education for low back pain (Review). 4. © 2008, The Cochrane
Library: John Wiley and Sons Ltd, with permission.
must ensure, however, that the patients read and understand the information they have been
given. In order to do this it is essential that the clinician is sure about what the patients already
know and understand (or misunderstand) about their pain.
In their review, Burton and Waddell (2002) concluded:
Carefully selected and presented information and advice about back pain in line with current manage-
ment guidelines can [their italics] have a positive effect on patients’ beliefs and clinical outcomes.
Written material is just one way of achieving this and, whilst its effect size may be small, its very
low patient cost may render it highly cost effective. Nevertheless, there is considerable scope for refine-
ment of the messages and their method of delivery within the whole scope of health care and the
contribution of innovative educational interventions for back pain deserves further scientific investi-
gation (p. 256).
Box 23.3 Recommendations for the design of educational

material
◆ Be clear about the core content of the message.
◆ Identify the targeted audience.
◆ Specify learning objectives.
◆ Address style and format as well as content of the material.
◆ Associate learning objectives with recommendations for behavioural change.
◆ Check comprehension and reinforce adherence.
Reprinted from Burton, A.K. & Waddell, G. (2002). Educational and informational approaches. In
S.J. Linton (Ed). New avenues for the prevention of chronic musculoskeletal pain and disability: Pain research
and clinical management. Vol 12, pp. 245–58. Copyright (2002) with permission from Elsevier.
23.3.4 Influence of patient characteristics

The relevance or worth of any approach to guiding self-management will be affected to an extent
by the patient’s ‘starting-point’. Patient beliefs, preferences and expectations influence both treat-
ment seeking and response to treatment. Professional training is predominantly biomedical or
biomechanical in emphasis and, while addressing patient symptoms is at the core of the consulta-
tion, patient beliefs are seldom systematically identified or addressed. As potential obstacles to
behavioural change, they need to be identified and, if necessary, addressed.
The decision to consult

Patients’ attitudes, perceptions, and beliefs about their back pain, its likely course, and the useful-
ness of specific treatments, may influence an individual’s decision to seek healthcare. There is
evidence from community surveys that about half of those who experience LBP in a 1-year period
will consult a HCP, and that while pain severity influences consulting behaviour in the acute
phase (<2 weeks), after this, those who consult are more likely to have externalized locus of con-
trol beliefs with regard to pain management (Waxman et al. 1998). An individual’s attitudes and
beliefs about the relationship between their pain and function appear to be associated with their
level of disability (Poiraudeau et al. 2006). Thus those who believe more strongly that their pain
means they should avoid physical activities and abandon normal roles, report higher levels of
disability than those with opposite beliefs, and are thus more likely to consult, re-consult and use
further healthcare resources.
Influences of beliefs, expectations, and preferences on participation in

treatment and on treatment outcome
Patients’ beliefs, expectations, and preferences about treatments for back pain are likely to influ-
ence their engagement in and adherence to treatment plans, yet empirical data are lacking.
However there is growing evidence from systematic reviews (Crow et al. 1999; Mondloch et al.
2001; Linde et al. 2007) across a wide range of health conditions that patients’ expectations influ-
ence their health outcomes. Data from musculoskeletal pain studies, and in particular, back pain,
demonstrate relationships between treatment preferences and expectations and patients’ clinical
and return-to-work outcomes (Heyman et al. 2006; Johnson et al. 2007; Preference Collaborative
Group, 2008). Positive attitudes towards treatment and confidence in benefit from specific treat-
ments have been shown to lead to between a twofold to fivefold greater likelihood of improve-
ment (Kalauokalani et al. 2001; Linde et al. 2007), although this finding is not consistent across all
studies (Klaber-Moffett et al. 1999; Cherkin et al. 2009).
In addition, general outcome and recovery expectations, irrespective of treatment, have been
shown to influence both clinical outcome (Foster et al. 2008) and occupational outcomes (Turner
et al. 2006, 2008). Nonetheless in evaluating this body of scientific evidence, a degree of caution is
warranted. According to Main et al. (2010a), methodological problems include the challenge of
capturing the beliefs, expectations, or preferences of patients who decline to participate in cohort
studies or clinical trials; the study of these factors is usually a secondary objective, and thus, the
statistical comparisons often lack power. In addition, few studies have measured beliefs, expecta-
tions, or preferences beyond the baseline measurement point and the effects of patients’ treat-
ment preferences and expectations may differ according to the nature, invasiveness, or
unpleasantness of the interventions.
In conclusion, the evidence demonstrates the importance of patients’ beliefs and expectations
to treatment engagement, adherence, and outcome. However, as a precursor to the design of fur-
ther interventions, we still require a clearer understanding of the nature of change and the under-
lying processes involved. Attempting to address such beliefs within a reactivation framework has
become an integral part of new approaches to the prevention associated incapacity both in health-
care settings (Hay et al. 2005) and in occupational of pain-associated settings (Shaw et al. 2009);
but in addressing the secondary prevention agenda, and shifting our focus from distal outcomes
to mediators of behavioural change, the role of education and advice in shifting core beliefs and
expectations seems to be fundamental.
23.3.5 The influence of HCPs’ beliefs and treatment orientations.

In attempting to evaluate the role of education within a self-management approach as a strategy
for behaviour change, however, a further difficulty arises. Even assuming that we know what
needs to be taught and what patient characteristics need to be addressed, there is clear evidence
that HCPs’ beliefs, preferences, and expectations influence both intervention delivery and patient
outcomes and recently, more attention has been directed at these factors.
The nature of HCPs’ beliefs and treatment orientations

The characteristics of the HCP, such as their status as professionals, their therapeutic style, the
words they use with patients, their beliefs about the problem, and their confidence or conviction
in treatments have all been suggested as non-specific effects of treatment (Crow et al. 1999; Ernst
2001; Feinstein 2002). The beliefs, expectations, and preferences of HCPs therefore likely influ-
ence their choice of assessment methods, explanation to patients, and treatment approach.
There is evidence that HCPs, such as primary care doctors, physical therapists, and rheuma-
tologists, hold a wide range of beliefs about pain that correlate with their recommendations to
patients (Coudeyre et al. 2006; Poiraudeau et al. 2006). Studies have emphasized the predomi-
nance of biomedically (or structure) -orientated pain beliefs among HCPs (Ostelo et al. 2003;
Daykin and Richardson 2004; Bishop and Foster 2005; Bishop et al. 2008). They have also shown
that some HCP groups are more biomedical than others in their attitudes and their advice is
characterized by advising patients to restrict activity, be vigilant about their backs and beliefs in a
structural cause of back pain (Pincus et al. 2007).
Influences on design/delivery of interventions and clinical outcomes

Several studies have shown that the attitudes, beliefs and treatment orientations of HCPs are
associated with the advice they give to patients as well as the choice of interventions. Houben et al.
(2004) showed that HCPs’ pain attitudes and beliefs significantly correlated with, and were the
strongest predictor of, their work and activity recommendations to patients. The same authors
subsequently found that HCPs with a more biomedical treatment orientation viewed daily activ-
ities as more harmful for a LBP patient compared with those with a more behavioural orientation,
and that biomedically-orientated therapists were more likely to advise patients to limit their
activities and work (Houben et al. 2005). Bishop et al. (2008) found that these HCPs’ pain atti-
tudes and beliefs were significantly associated with their reported practice behaviour, such as the
type of advice to patients about returning to work. Those with high biomedical orientations
and low behavioural orientations were much more likely to advise continued work absence
(44.9%), than those with high behavioural and low biomedical scores (11.9%). Other studies
have demonstrated that advice to restrict work or daily physical activities is associated with higher
fear avoidance beliefs of HCPs (Linton et al. 2002; Coudeyre et al. 2006; Poiraudeau et al. 2006;)
There is evidence that at the level of the individual back pain patient, HCPs find it difficult to
be consistent in applying best evidence from guideline recommendations. In-depth qualitative
interviews have highlighted that the use of diagnostic investigations may be influenced by patient
demand, avoidance of risk or giving patients ‘peace of mind’ rather than being driven by clinical
need (Corbett et al. 2009). In making decisions with individual patients, HCPs draw heavily on
their own beliefs about the effectiveness of treatments, prior clinical experience and the pre-
eminence of their relationship with the patient (Corbett et al. 2009).
Studies recently have started to incorporate measures of HCPs’ attitudes and beliefs and to
develop educational programmes that attempt to modify these (Jellema et al. 2005; Overmeer
et al. 2009; Sieben et al. 2009). However, to date, few studies have measured HCPs’ attitudes and
beliefs in parallel with patients’ outcomes. Jellema et al. showed that GPs’ attitudes were modifi-
able by a short training session and that their back pain-related attitudes became less biomedical,
but the trial showed no significant differences in patient outcomes. Overmeer et al. (2009) found
that a university-based course of 8 days over 12 weeks resulted in a shift from a biomedical to a
biopsychosocial orientation with a concomitant increase in their understanding of the nature of
psychosocial factors. However, their patients perceived the therapists’ behaviour before and after
the education course as similar and were equally satisfied with their treatment and treatment
outcome
Finally, this growing body of literature suggests that the attitudes, beliefs, and preferences of
HCPs, might serve as barriers to optimal patient outcomes. However, the limitations of available
studies include the frequent use of non-validated measurement tools or tools that have been
adapted from patient measures for use with HCPs. Indeed a recent review highlighted that the
number of tools available with which to measure HCPs’ attitudes and beliefs about musculoskel-
etal pain is limited (Bishop et al. 2007), and that the development and testing of these tools is in
its infancy.
23.4 From evidence to action: some suggestions

for improving the effectiveness of educational strategies
in secondary prevention
What do patients need to know? The central proposition underpinning this chapter is that for
optimal self-management of LBP and its effects, knowledge per se is not what patients need, but
rather they need information needed to develop, embed and maintain optimal pain coping strat-
egies. In developing a strategy to achieve this however, patients need to understand the nature of
their predicament, in this case the nature of LBP and its effects.
A number of key educational messages identified by Main et al. (2010a) are shown in Box 23.4.
These key messages link across to a set of colloquial myths which act as obstacles to recovery and
participation, and are summarized in Box 23.4. Surprisingly perhaps, the myths are not confined
to patients: these beliefs are still held by some HCPs, irrespective of discipline.
23.4.1 Educational content

Patients need to be offered a plausible understanding of LBP, in language which they understand,
which shifts the primary focus from structural mechanisms, pathology and disease, to an under-
standing of LBP in terms of central pain mechanisms and the development of persistent pain/
disability within a biopsychosocial framework (Main et al. 2010b).
They may have unhelpful or mistaken beliefs about LBP and its relationship with pain-associated
limitations and the suffering that accompanies persistent LBP. Main et al. (2010a) have high-
lighted three particularly important types of belief/expectations which need to be considered
Beliefs about the nature of pain
Pain researchers have identified a number of different types of belief or appraisal about the nature
of pain. Beliefs about the extent to which pain can be controlled would appear to be one of the
most powerful determinants of adjustment to pain and the development of incapacity, possibly
Box 23.4 Key educational messages

◆ The perception of pain is a consequence of the interpretation of pain which is shaped by our
memories and prior experience.
◆ Our response to pain is influenced both by our beliefs about it and the emotional signifi-
cance we attribute to it.
◆ Back pain patients’ beliefs, expectations, and preferences should be elicited and used in the
clinical decision-making process to help select treatments that have the best chance of
promoting patient recovery and return to work.
◆ Beliefs about the nature of pain, fears of hurting, harming, and further injury, and self-
efficacy beliefs are the most important beliefs to consider.
◆ The attitudes and beliefs of healthcare practitioners are part of the dynamic interaction
within back pain consultations, and are significantly associated with the advice and recom-
mendations they give to patients and their treatment decisions.
◆ Societal influences play an important role in determining the outcome of back pain and
development of disability.
mediating the influence of pain and depression. These core constructs can also be viewed as
specific therapeutic targets. Correction of fundamental misunderstandings about the nature of
pain, its effects and probable course are likely not only to facilitate optimal management but also
prevent unnecessary iatrogenic misunderstandings and distress. An important stage in optimal
adaptation to a chronic condition seems to depend upon patients’ ability to come to terms
with what they can and cannot control. This is turn may be affected by specific fears of hurting,
harming, and further injury.
Specific fears of hurting, harming, and further injury

Since the early-mid 1990s, there has been an increasing research focus on the role of fear and
avoidance in the development and maintenance of disability, (Leeuw et al. 2007). Among people
with chronic low back pain, pain-related fear has been found to be associated with reduced lum-
bar flexion (Geisser et al. 2004) and pain-related fear and pain catastrophizing have been found
to be stronger predictors of overall disability than pain intensity. One study found that pain-
related fear was also the strongest predictor of performance (van den Hout et al. 2001). However
in acute low back pain only modest correlations between pain intensity, pain-related fear, avoid-
ance behaviour and disability have been detected (Sieben et al. 2005) and more general percep-
tions about illness may also be influential (Foster et al. 2009).
Self-efficacy beliefs
According to self-efficacy theory, once a situation has been perceived as involving harm, loss,
threat, or challenge and individuals have considered a range of coping strategies open to them,
what they do will be dependent on what they believe they can achieve (Bandura, 1977). Asghari
and Nicholas (2001) have shown that pain self-efficacy beliefs are an important determinant of
pain behaviours and disability associated with pain, over and above the effects of pain, distress,
and personality variables. Taken together, clinical and experimental investigations suggest that
perceived coping inefficacy may lead to preoccupation with distressing thoughts and concomitant
physiological arousal, thereby increasing pain, decreasing pain tolerance and leading to increased
use of medication, lower levels of functioning, poorer exercise tolerance, and increased invalidism.
Such beliefs therefore would seem to have considerable potential as targets within clinical
management.
Self-efficacy has been found to account for the greatest proportion of variance in physical per-
formance even after anticipated pain and re-injury have been excluded although pain intensity
was also a significant (albeit limited) predictor of performance (Lackner et al. 1996). Thus expect-
ancies of harm and pain catastrophizing, rather than being primary causal determinants of func-
tion, may be components of one’s confidence of successful task performance (Nicholas 2007).
Treatment recommendations derived from this interpretation emphasize the importance of goal
and quota setting, and monitoring of pain and task performance as components of pain manage-
ment and as such fit well within modern pain management and suggest further specific targets for
advice and education.
23.4.2 The educational context

Communication is at the heart of the clinical consultation and one of the essential ‘vehicles’ by
which information and advice is transmitted. Maguire and Pitceathley (2002) identified a
number of key tasks in communication with pain patients. They are shown in Table 23.1 in which
it can be seen that education and the provision of advice is not only a major component of
HCP-patient communication, but that tailoring the information and ensuring patient under-
standing is fundamental.
23.4.3 Developing an educational strategy

Reconsideration of the role of education and advice in behaviour change
The importance of beliefs and expectations as influences on patient behaviour and clinical outcome
has already been highlighted, and in the modification of beliefs and expectations, education clearly
has a key role. It appears to be assumed often that change in beliefs and expectations will
lead inevitably to a cascade of behaviour change and improved clinical outcome. However both
Table 23.1 Common back pain myths
Myth Reality
◆ Back pain means serious ◆ This is not always the case: pain can occur without injury. Even when
damage and injury specific tissues are affected, activity and work are not precluded.
Temporary discomfort is often part of recovery
◆ Work/activity is the cause: ◆ Normal work/activity can trigger symptoms, but is unlikely to cause
something must be substantial damage. The actual condition is usually not made worse by
damaged so work/activity continuing work (assuming control of significant risks). Work/activity may
will just make it worse be difficult or uncomfortable, but that doesn’t mean it is doing harm
◆ Medical treatment and ◆ Most people, for most episodes of LBP problems, do not seek
tests are necessary healthcare. Tests and imaging are generally unhelpful for LBP
◆ Back problems must be ◆ Quite the contrary—activity leads to faster and more sustained recovery
rested and return to work. Temporary reduction of activity may be required,
but long-term rest is detrimental
◆ Sick leave is needed as ◆ Often sick leave is not needed–staying at work is desirable, perhaps with
part of the treatment some temporary modifications
◆ No return to work/activity ◆ This is clearly unrealistic and unhelpful—many workers can and do go
till 100% fit and pain free back to work with ongoing symptoms, and they come to no harm
Adapted from Kendall et al. (2009).
cross-sectional studies and longitudinal studies have consistently demonstrated that pain-associated
disability has multiple determinants.
Guidelines as a vehicle for education

The role of the HCP traditionally has been focused primarily on assessment, diagnosis, and treat-
ment. Recognition of their specific role as educators is evident in the plethora of guidelines for all
sorts of conditions which are now available, with several developed specifically for back pain
(Burton et al. 2006; Waddell and Burton 2001; NICE 2010). Perhaps then we know what needs to
be done; but we need to implement it.
That provision of information, although perhaps necessary is not sufficient, was illustrated by
Grimshaw et al. (2004) in their systematic review of effectiveness and efficiency of guideline dis-
semination and implementation strategies. Included within the 235 relevant studies they identi-
fied, were 23 comparisons of multifaceted involving educational outreach, yielding modest to
moderate improvements in care with a median absolute improvement in performance across
interventions of 14.1% in 14 cluster randomized comparison of reminders, 8.1% in four cluster
randomized comparison of dissemination of educational materials, 7.0% in five cluster rand-
omized comparisons of audit and feedback, and 6.0% in 13 cluster randomized comparisons of
multifaceted interventions involving educational outreach, with no relationship found between
the number of component interventions and the effects of multifaceted interventions.
The challenge of implementation

According to Michie et al. (2005), evidence-based guidelines are not implemented effectively,
with the result that best health outcomes are not achieved and they suggest that this may be due
to a lack of theoretical understanding of the processes involved in changing the behaviour of
HCPs. There is now a burgeoning literature on linking theory and intervention and Michie et al.
(2008) recommend the incorporation of psychological theory in developing evidence-based prac-
tice and the mapping of theoretically derived behavioural determinants to specific behaviour
change techniques. The task therefore for education is to provide advice not only about the nature
of the condition (e.g. LBP), but to understand the processes underpinning behaviour change and
how to implement it. A number of theories such as the Social Cognitive Theory (Bandura 1977)
and its core concept of self-efficacy; and the concept of illness representation embedded within
the ‘common-sense’ model of self-regulation (Leventhal et al. 1980) are all illuminating, but in
trying to ‘deconstruct’ the educational process, the theory of planned behaviour (Azjen 1991)
perhaps has been most extensively researched. It appears to lend itself well to specify elements in
communication that can be specifically targeted. If indeed it is appropriate to ‘deconstruct’ the
communication process in order to enhance its effectiveness, the focus of education and advice
needs to be not only on the nature of the condition (e.g. LBP) but how to gain mastery over it and
understanding the processes underpinning behavioural change becomes fundamental. In a sense
then education and advice becomes embedded within strategies for cognitive and behavioural
change.
Patient-mediated strategies
Of all the mediators of behavioural change, constructs, self-efficacy appears to have the most
direct relevance in the context of guided self-management in that it can be mapped directly onto
specific behavioural ‘targets’. There remains the question however of whether patients can actu-
ally effect the necessary changes and appropriate education and advice would seem to be a key
component of this process. Wensing et al. (2005) have reviewed the effectiveness of patient-
mediated strategies in implementing healthcare innovations. As a precursor to their review they
state ‘Increasingly patients expect to be well-informed about treatment options and to be

involved in making decisions’ (p. 185). They appraise the effectiveness of four types of patient
strategy: (1) health education or advertising through the mass media; (2) preparation for contact
with care providers; (3) communication through single contacts or within episodes of care; and
(4) feedback on healthcare received. They conclude ‘some strategies to implement innovations
through patients are promising, but our understanding of the effects on professional behaviour
and processes of care is, as yet, limited’ (p. 185). They do however find some evidence for effect
of interventions to enhance self-management in the fields of asthma, cancer, and diabetes.
23.4.4Rolling out LBP education at a systems level: example of the

flags framework
Kendall et al. (1997) coined the term ‘yellow flags’ to encompass psychological and social-
environmental risk factors for prolonged disability and failure to return to work as a consequence
of musculoskeletal symptoms. The particular focus was on LBP. The psychological factors
included fears about pain and injury and unhelpful beliefs about recovery, as well as emotional
distress. The socio-environmental factors included the worker’s perceptions that the workplace
was unsupportive and overly supportive healthcare providers. The monograph provided a guide
to the assessment of yellow flags that included a clinical interview and a psychosocial screening
questionnaire. As with many such guidelines, their impact on clinical practice was unclear (Grol
and Buchan 2006).
Following an international conference and think tank (The Decade of the Flags, 2007), the
further research on flags was reviewed and the overall remit broadened, particularly in respect of
occupational obstacles to recovery (blue flags) and system or contextual obstacles (black flags) and
a completely revised monograph was produced (Kendall et al. 2009). Although this initiative has
not as yet been fully evaluated, and indeed is still something of a ‘work in progress’, the authors
have developed a ‘tool box’ of information and advice about flags as part of a website which is
under continual development (http://www.tsoshop.co.uk/flags). The project has some interest
from an educational point of view in that it represents a specific attempt to link provision of
evidence-based information about the nature of the condition, with specific actions, and a range
of implementation strategies at the levels of the individual, the workplace and the wider social
context so that guided self-management is fully embedded into its biopsychosocial context.
23.5 Conclusion
Understanding chronic and recurrent LBP requires an understanding of the individual concerned
in terms of attitudes and beliefs, emotional responses, and pain coping strategies. The role of edu-
cation and advice is an essential component in the development of a self-management approach.
We might usefully consider development of a range of evidence-based educational materials, using
a range of ‘vehicles’ within an educational ‘stepped care approach’ such as recommended in the
Flags monograph (Kendall et al. 2009), ranging from simple advice (e.g. The Back Book) with more
sophisticated and detailed source material designed to enable the development of specific adaptive
pain coping strategies. The further educational objectives might include enabling the identifi-
cation and management of obstacles to recovery, identifying mediators of outcome in terms
of modifiable risk factors and consideration of strategies to turn obstacles to recovery into oppor-
tunities for change. However to effect sustained impact we need to recognize the importance not
only of the focus, content and format of our educational advice, but also develop effective imple-
mentation strategies. The challenge therefore for education in facilitating effective self-directed
behaviour change is to develop the evidence base of the effectiveness of different behaviour
change strategies, derived from a sharper theoretical understanding of behavioural change theories
(Michie 2005) and evaluate their effectiveness in optimizing the clinical outcomes for our LBP
patients. Our educational focus therefore has to move beyond knowledge about the nature of LBP
and its effects, but to strategies designed to minimize its impact with the development of appropri-
ate self-management strategies. The role of education and advice becomes less focused on provi-
sion of biomedical information to plug a knowledge deficit and more directed at illumination of
ways in which to minimize the impact of LBP within a patient-centred approach (guided self-
management) so that the role of the HCP becomes less of an educator and more of a coach.
References
Albaladejo, C., Kovacs, F.M., Royuela, A., del Pino R., & Zamora, J. (2010). The efficacy of a short
education program and a short physiotherapy program for treating low back pain in primary care.
Spine, 35, 483–96.
Asghari, A., & Nicholas, M.K. (2001). Pain self-efficacy beliefs and pain behaviour. A prospective study.
Pain, 94, 85–100.
Azjen, I. (1991). The theory of planned behaviour. Organisation Behaviour and Human Decision Processes,
50, 179–211.
Bandura, A. (1977) Self-efficacy: Towards a unifying theory of behavioural change. Psychological Review,
84, 191–215.
Bishop, A., & Foster, N.E. (2005) Do physical therapists in the United Kingdom recognize psychosocial
factors in patients with acute low back pain? Spine, 30, 1316–22.
Bishop, A., Thomas, E., & Foster, N.E. (2007) Health care practitioners’ attitudes and beliefs about low
back pain: A systematic search and critical review of available measurement tools. Pain, 132, 91–101.
Bishop, A., Foster, N.E., Thomas, E., & Hay, E.M. (2008). How does the self-reported clinical management
of patients with low back pain relate to the attitudes and beliefs of health care practitioners? A survey of
UK general practitioners and physiotherapists. Pain, 35, 187–95.
Buchbinder, R., Jolley, D., & Wyatt, M. (2001). Population based intervention to change back pain beliefs
and disability: Three part evaluation. British Medical Journal, 322, 1516–20.
Buchbinder, R., Gross, D., Werner, E., & Hayden, J. (2008). Understanding the characteristics of effective
public health interventions for back pain and methodological challenges in evaluating their effects.
Spine, 33, 74–80
Burton, A.K., Balagué, F., Cardon, G., et al. on behalf of the COST B13 Working Group on Guidelines for
Prevention in Low Back Pain. (2006). European guidelines for prevention in low back pain November
2004. European Spine Journal, 15(Suppl. 2), S136–68
Burton, A.K., & Tillotson, K.M. (1991). Prediction of the clinical course of low-back trouble using
multivariable models. Spine, 16, 7–14.
Burton, A.K., & Waddell, G. (2002). Educational and informational approaches. In S.J.Linton (Ed).
New avenues for the prevention of chronic musculoskeletal pain and disability: Pain research and clinical
management. Vol 12, pp. 245–58. Amsterdam, Elsevier.
Burton, A.K., Waddell, G., Tillotson, K.M., & Summerton, N. (1999). Information and advice to patients
with back pain can have a psositive effect: A randomized control trial of a novel educational booklet in
primary care. Spine, 24, 2484–91.
Cherkin, D., Sherman, K., Avins, A., et al. (2009). A randomised trial comparing acupuncture, simulated
acupuncture and usual care for chronic low back pain. Archives of Internal Medicine, 169, 858–66.
Corbett, M., Foster N.E., & Ong, B. (2009). GP attitudes and self-reported behaviour in primary care
consultations for low back pain. Family Practice, 26, 359–64.
Coudeyre, E, Rannou, F., Tubach, F., et al. (2006). General practitioners’ fear-avoidance beliefs influence
their management of patients with low back pain. Pain, 124, 330–7.
Crow, R., Gage, H., & Hampson, S. (1999). The role of expectancies in the placebo effect and their use in
the delivery of health care: A systematic review. Health and Technology Assessment, 3, 3.
Daykin, A.R., & Richardson, B. (2004). Physiotherapists’ pain beliefs and their influence on the
management of patients with chronic low back pain. Spine, 29, 783–95.
Dixon, A. (1990). The evolution of clinical policies. Medical Care, 28, 201–20.
Engers, A. J., Jellema, P., Wensing, M., van der Windt, D.A.W.M., Grol, R., & van Tulder, M.W. (2008).
Individual patient education for low back pain (Review). 4. The Cochrane Library: John Wiley and
Sons Ltd.
Ernst, E. (2001). Towards a scientific understanding of placebo effects. In: D. Peters (Ed.) Understanding
the Placebo Effect in Complementary Medicine: Theory, Practice and Research, pp. 17–30. London:
Feinstein, A. (2002) Post-therapeutic response and therapeutic style: re-formulating the placebo effect.
Journal of Clinical Epidemiology, 55, 427–9.
Fordyce, W.E. (1976). Behavioral Methods for Chronic Pain and Illness. St. Louis, MO: Mosby.
Foster, N.E., Bishop, A., Thomas, E., et al. (2008). Illness perceptions of low back pain patients in primary
care: What are they, do they change and are they associated with outcome? Pain, 136, 177–87.
Foster, N.E., Thomas, E., Bishop A., Dunn K.M., & Main, C.J. (2009) Distinctiveness of psychological
obstacles to recovery in low back pain patuents in primary care. Pain, 148:398–406.
Frost, H., Lamb, S.E., Doll, H.A., Carver, P.T., & Stewart-Brown, S. (2004). Randomised control trial of
physiotherapy compared with advice for low back pain. British Medical Journal, 328, 798.
Geisser, M., Haig, A., & Wallborn, A. (2004) Pain relaled fear, lumbar flexion and dynamic EMG among
persons with chronic musculoskeletal low back pain. Clinical Journal of Pain, 20, 61–9.
George, S.Z., Fritz, J.M., Bialowsky, J., & Donald, D.A. (2003). The effect of a fear-avoidance-based therapy
for patients with acute back pain: Results of a randomised clinical trial. Spine, 28, 2551–60.
Grimshaw, J.M., Thomas, R.E., MacLennan, G., et al. (2004). Effectiveness and efficiency of guideline
dissemination and implementation strategies. Health Technology Assessment, 8, 6
Grol, R., & Buchan, H. (2006). Clinical guidelines: What can we do to increase their use? Medical Journal of
Australia, 185, 301–2
Hansen, Z., Daykin, A., & Lamb, S.E. (2010). A cognitive-behavioural programme for the management of
low back pain in primary care: A description and justification of the intervention used in the Back
Skills Training Trial. Physiotherapy, 96, 87–94.
practice. Lancet, 365, 2024–9.
Hay, E.M, Dunn, K.M, Hill, J.C., et al. (2008). A randomised clinical trial of subgrouping and targeted
treatment for low back pain compared with best current care. The STarT Back Trial Study Protocol
BMC Musculoskeletal Disorders, 9, 58
Hazard, R., Haugh, L., Reid, S., Preble, J.B., & MacDonald, L. (1996). Early prediction of chronic disability
after occupational low back injury. Spine, 21, 945–51.
Heymans, M.W., de Vet, C.W., Knol, D L., Bongers, P.M., Koes B.W., & van Mechelen, W. (2006).
Workers’ beliefs and expectations affect return to work over 12 months. Journal of Occupational
Hill, J.C., Dunn, K.M., Lewis, M., et al. (2008). A Primary Care Back Pain Screening Tool: Identifying
patient subgroups for initial treatment. Arthritis and Rheumatism, 59, 632–41.
Houben, R.M.A., Vlaeyen, J.W.S., Peters, M., et al. (2004). Health care providers’ attitudes and beliefs
towards common low back pain: Factor structure and psychometric properties of the HC-PAIRS.
Houben, R.M.A, Ostelo, R.W.J.G., Vlaeyen, J.W.S, Wolters P., Peters M., & Stomp-van den Berg, S.
(2005). Health care providers’ orientations towards common low back pain predict perceived
harmfulness of physical activities and recommendations regarding return to normal activity. European
Hutchinson, A., & Baker, R. (1999). Making Use of Guidelines in Clinical Practice. Abingdon: Radcliffe
Medical Press.
Jellema, P., van der Windt, D.A.W., van der Horst, H., Blankenstein, A., Bouter, L., & Stalman, W. (2005).
Why is a treatment aimed at psychosocial factors not effective in patients with (sub)acute low back
pain? Pain, 118, 350–9.
Johnson, R.E., Jones, G.T., Wiles, N.J., et al. (2007). Active exercise, education, and cognitive behavioural
therapy for persistent disabling low back pain: A randomised controlled trial. Spine, 32, 1578–85
Kalauokalani, D, Cherkin, D.C., Sherman, K.J., Koepsell, T.D., & Deyo, R.A. (2001). Patients’ preferences.
Lessons from a trial of acupuncture and massage for low back pain: patient expectations and treatment
effects. Spine, 26, 1418–24.
Kendall, N.A.S., Linton, S.J., & Main C.J. (1997). Guide to assessing psychosocial Yellow flags in acute low
back pain: Risk factors for long-term disability and work loss. Wellington: Accident Rehabilitation and
Compensation Insurance Corporation of New Zealand and the National Health Committee.
Kendall, N.A.S., Burton, A.K., Main, C.J., & Watson, P.J. (2009). Tackling musculoskeletal problems, a guide
for the clinic and workplace: Identifying Obstacles using the Psychosocial Flags framework. London:
The Stationery Office.
Klaber-Moffett, J.A., Chase, S.M., Portek, I., & Ennis J.R. (1986). A controlled, prospective study to
evaluate the effectiveness of a back school in the relief of chronic low back pain. Spine, 11, 120–22.
Klaber-Moffett, J.A., Torgerson D., Bell-Syer, S., et al. (1999). A controlled, prospective study to evaluate
the effectiveness of a back school in the relief of chronic low back pain. British Medical Journal,
319, 279–83.
Klaber-Moffett, J., Carr, J., & Howarth, E. (2004). High fear-avoiders of physical activity benefit from an
exercise program for patients with back pain. Spine, 29, 1167–73.
Kovacs, F., Abraira, V., Santos, S., et al. (2010). A comparison of two short education programs for
improving low back-related disability in the elderly. Spine, 32, 1053–9.
Lackner, J.M., Carosella, A.M., & Feuerstein, M. (1996). Pain expectancies, pain, and functional self-
efficacy expectancies as determinants of disability in patients with chronic low back disorders. Journal
of Consulting and Clinical Psychology, 64, 212–20.
Lamb, S.E., Hansen, S., Lall R., et al. (2010). Group cognitive behavioral treatment for low-back pain in
primary care: a randomised controlled trial and cost-effectiveness analysis. Lancet, 375, 916–23.
Leeuw, M, Goossens, M.E.J.B., Linton S.J., Crombez, G., Boersma, K., & Vlaeyen J.W.S. (2007). The fear
avoidance model of musculoskeletal pain: Current state of scientific evidence. Journal of Behavioural
Leventhal, H., Meyer, D., & Nerenz, D. (1980). The common sense representation of illness danger. In:
Rachman S, (Ed.) Contributions to Medical Psychology, pp. 42–65. New York: Pergamon.
Linde, K., Witt, C.M., Streng, A., et al. (2007). The impact of patient expectations on outcomes in four
randomized controlled trials of acupuncture in patients with chronic pain. Pain, 128, 264–71.
Linton, S., Vlaeyen, J., &Ostelo, R. (2002). The back pain beliefs of health care providers: are we
fear-avoidant? Journal of Occupational Rehabilitation, 12, 223–32.
Linton, S.J., & Andersson T. (2000). Can chronic disability be prevented? A randomised trial of a
cognitive-behavioural intervention and two forms of information for patients with spinal pain. Spine,
25, 2825–31.
McIntosh, G, Frank, J, Hogg-Johnson, S., Bombardier, C., & Hall, H. (2000). Prognostic factors for time
receiving workers’ compensation benefits in a cohort of patients with low back pain. Spine, 25, 147–57.
Maguire, P., & Pitceathley, C. (2002). Key communication skills and how to acquire them. British Medical
Journal, 325, 697–700.
Main, C.J., Sullivan, M.J.L., & Watson, P.J. (2008) Pain Management. Practical applications of the
biopsychosocial approach in clinical and occupational settings. Edinburgh:Churchill-Livingstone.
Main, C.J., Foster N., & Buchbinder, R. (2010a). How important are back beliefs and expectations for
satisfactory recovery from back pain? Best Practice and Research Clinical Rheumatology, 24, 205–217.
Main, C.J., Buchbinder, R., Porcheret, M., & Foster, N. E. (2010b). Addressing patient beliefs and
expectations in the consultation. Best Practice and Research Clinical Rheumatology, 24, 219–25.
Michie, S., Johnston, M., Abraham, C., Lawton, R., Parker, D., & Walker, A. (2005) making psychological
theory useful for implementing evidence based practice: A consensus approach. Quality and Safety in
Health Care, 14, 26–33.
Michie, S., Johnston, M., Francis, J., Hardeman, W., & Eccles M. (2008). From theory to Intervention:
mapping theoretically derived behavioural determinants to behaviour change techniques. Applied
Psychology: An International Review, 57, 660–80.
Mondloch, M., Cole, D., & Frank, J. (2001). Does how you do depend on how you think you’ll do? A
systematic review of the evidence for a relation between patients’ recovery expectations and health
outcomes. Canadian Medical Association Journal, 165, 174–9.
Moore, J.E., von Korff, M., Cherkin, D., Saunders, K., & Lorig, K. (2000). A randomised trial of a
cognitive-behavioral program for enhancing back pain self-care in a primary care setting. Pain,
88, 145–53.
NICE (2010) Early management of persistent non-specific low back pain. London: National Institute for
Health & Clinical Excellence.
Nicholas, M.K. (2007). The pain self-efficacy questionnaire: Taking pain into account. European Journal of
Pain, 11, 153–63.
Ostelo, R.WJ.G., Stomp-van den Berg, S.G.M., Vlaeyen, J.W.S., Wolters, M.J.C., & de Vet H.C.W. (2003).
Health care provider’s attitudes and beliefs towards chronic low back pain: The development of a
questionnaire. Manual Therapy, 8, 214–22.
Overmeer, T, Boersma, K, Main C.J., & Linton S. (2009) Do physical therapists change their beliefs,
attitudes, knowledge, skills and behaviour after a biopsychosocially orientated university course?
Journal of Evaluation in Clinical Practice, 15, 724–32.
Pincus, T, Foster N.E., Vogel, S., Santos R., Breen, A., & Underwood, M. (2007). Attitudes to back pain
amongst musculoskeletal practitioners: a comparison of professional groups and practice settings using
the ABS-mp. Manual Therapy, 12, 167–75.
Poiraudeau, S., Rannou, F., Baron, G., et al. (2006). Fear-avoidance beliefs about back pain in patients with
subacute low back pain. Pain, 124, 305–11.
Preference Collaborative Review Group. (2008). Patients’ preferences within randomised trials: systematic
review and patient level meta-analysis. British Medical Journal, 337, a1864–64.
Redman, S., Spencer, E., & Sanson-Fisher, R. (1990). The role of mass media in changing health-related
behaviour: a critical appraisal of two methods. Health Promotion International, 5, 85–101.
Roland, M., Waddell, G., Klaber-Moffett J., Burton, K., & Main, C.J. (2002). The Back Book (2nd Edit.).
Norwich: The Stationery Office.
Rose, G. (1992). The strategy of preventive medicine. Oxford: Oxford University Press.
Shaw, W.S., van der Windt, D.A,Main C.J., Loisel, P., & Linton S.J., Decade of the Flags Working Group
(2009). Early patient screening and intervention to address individual-level occupational factors (‘blue
flags’) in back disability. Journal of Occupational Rehabilitation. 19, 64–80.
Sieben, J.M., Portegijs, P.J.M., Vlaeyen J.W.S., & Knottnerus, J.A. (2005). Pain-related fear at the start of a
new low back pain episode. European Journal of Pain, 9, 635–41.
Sieben, J.M, Vlaeyen, J.W.S, Portegijs, P.J.M, et al. (2009). General practitioners’ treatment orientations
towards low back pain: Influence on treatment behaviour and patient outcome. European Journal of
Pain, 13, 412–8.
Symonds, T.L., Burton, A.K., Tillotson K.M., & Main C.J. (1995). Absence resulting from low back trouble
can be reduced by psychosocial intervention in the workplace. Spine, 20, 2738–45.
Thorsen T., & Makela, M. (1999). Theory and Practice of Clinical Guidelines Implementation. DSI Rapport
99.05. Copenhagen:DSI Danish Institute for Health Services Research and Development.
Turk, D.C., Meichenbaun, D, & Genest, M. (1983). Pain and Behavioral Medicine. A cognitive-behavioral
Turner, J.A., Franklin, G., Fulton-Kehoe, D., et al. (2006). Worker recovery expectations and
fear-avoidance predict work disability in a population-based workers’ compensation back pain sample.
Spine, 31, 682–89.
Turner, J.A., Franklin, G., Fulton-Kehoe, D., et al. (2008). ISSLS prize winner: Early predictors of
chronic work disability: A prospective, population-based study of workers with back injuries. Spine,
33, 2809–2818.
UK BEAM Trial team (2004). United Kingdom back pain and exercise (UK BEAM) randomised trial:
effectiveness of physical treatment for back pain in primary care. British Medical Journal, 329, 1377
van den Hout, J.H., Vlaeyen, J.W.S., Houben, R.M.A., Soeters, A.P.M., & Peters, M.L. (2001). The effects
of failure feedback and pain-related fear on pain report, pain tolerance, and pain avoidance in chronic
low back pain patients. Pain, 92, 247–57.
van der Windt, D.A.W., Hay, E.M., Jellema, P., & Main, C.J. (2008). Psychosocial interventions for low
back pain in primary care: Lessons learned from recent trials. Spine, 33, 81–9.
Von Korff, M., Moore J.E., Lorig, K., et al. (1998). A randomised trial of a lay person-led self-management
group for back pain patients in primary care. Spine, 23, 2608–615.
Waddell G. (2004). The Back Pain Revolution (2nd. Edn.) Edinburgh, Churchill-Livingstone.
Waddell, G., & Burton, A.K. (2001). Occupational health guidelines for the management of low back pain
at work: Evidence review. Occupational Medicine, 51, 124–35.
Waddell, G., O’Connor, M., Boorman, S., & Torsney, B. (2007). Working Backs Scotland. A Public and
Professional Health Education Campaign for Back Pain. Spine 32, 2139–43.
Waxman, R, Tennant, A., & Helliwell, P. (1998). Community survey of factors associated with consultation
for low back pain. British Medical Journal, 317, 1564–7.
Wensing, M., & Grol, R. (2005). Educational Interventions. In R. Grol, M. Wensing M., & M. Eccles (Eds.)
Improving Patient Care: The Implementation of Change in Clinical Practice, pp. 147–57. Edinburgh:
Elsevier.
Wensing, M., Elwyn, G., & Grol R. (2005). Patient-mediated strategies. In R. Grol, M. Wensing, &
M. Eccles. (Eds.) Improving Patient Care: The Implementation of Change in Clinical Practice
(pp 185–96). Elsevier: Edinburgh.
Werner, E.L., Ihlebaek, C., Laerum, E., Wormgoor, M., & Indahl, A. (2008). Low back pain media
campaign: no effect on sickness behaviour. Patient Education and Counselling, 71, 198–203.
Chapter 24
Motivational Issues
in Pain Management
Robert D. Kerns, Mark P. Jensen, and
Warren R. Nielson
Disease self-management has emerged as a dominant approach to keeping chronic illness under
control, reducing its impact on physical health status and functioning, and improving coping
with the psychosocial sequelae of the illness (Clark et al. 1991; Lorig et al. 1999; Newman et al.
2004). Consistent with this zeitgeist, as the field of pain medicine has matured, it has become
increasingly apparent that an approach that emphasizes self-management may be key to control
of acute pain and prevention of the progression to chronic pain (Damush et al. 2003) and to pro-
moting optimal adjustment and adaptation to persistent pain conditions such as arthritis (Keefe
et al. 2000), fibromyalgia (Okifuji and Hare 2011), and chronic back pain (Hoffman et al. 2007),
among others. Given a growing body of research that documents important roles of psychological
factors, including pain-related coping, as predictors of the development of chronic pain, it is
intuitively appealing to predict that early efforts to promote use of adaptive pain coping or self-
management skills among persons with acute pain can play a role in resolving pain and prevent-
ing the development of chronic pain. Adoption of adaptive pain self-management coping skills is
predicted to contribute to positive adjustment and accommodation to acute pain and to facilitate
resolution of the experience of pain and its potential negative impacts.
Optimal self-management involves numerous behaviour changes by persons with pain, includ-
ing taking medications when indicated, physical exercise, undertaking preventive action, learning
stress reduction strategies, moderating or pacing activities, and changing other aspects of their
lifestyles. It is in this context that psychological interventions have been developed and promoted
as important alternatives to more traditional medical and rehabilitation therapies (Turk and
Gatchel 2002). Dominant among the numerous psychological therapies that have been applied to
chronic pain management are those that assist individuals in developing a perspective of personal
control and mastery and adoption of a range of adaptive cognitive and behavioural pain coping
skills (Turk et al. 1983). Significant reductions in pain severity, disability, and affective distress, as
well as reductions in healthcare resource utilization have been repeatedly documented. However,
high rates of treatment refusal and treatment dropout continue to be reported (Turk and Rudy
1991). Equally problematic are reports that many patients fail to adhere to therapist recommen-
dations related to cognitive and behavioural pain management skill acquisition and practice
(Heapy et al. 2005). A variety of contributors to dropout and poor adherence have been suggested
including a failure to take into account important individual differences in treatment planning,
the overall complexity of therapist recommendations for skill development, and patient motiva-
tion for making changes (Jensen 2002; Kerns et al. 1998). After all, because the ability of patients
to manage chronic pain depends much more on what they do than on what is done to them,
motivation can be viewed as a primary issue in pain self-management.
Specific attention to motivation and factors that influence motivation may serve an important
role in improving engagement and participation in pain self-management interventions, includ-
ing interventions for persons experiencing acute pain (Kerns et al. 2006). Motivation for engaging
in pain self-management has emerged as a particularly important focus of research in the past
decade. Of great heuristic importance has been the development of a sound and integrative con-
ceptual motivational model of pain self-management (Jensen et al. 2003) and tools for measuring
the core constructs of the model, especially those designed to measure persons’ ‘pain readiness to
change’ (Kerns and Habib 2004; Nielson et al. 2003, 2008, 2009). In addition, Kerns (1994) has
suggested that a prescriptive approach to pain treatment planning may be an important strategy
for promoting engagement and participation in treatment and for improving outcomes. He calls
for individual tailoring of training in new pain coping skills strategies by matching patients’
preferences for a more limited number of specific treatment components. Jensen (2002) and
Kerns and his colleagues (Kerns et al. 1998) have also suggested that the incorporation of specific
motivational enhancement strategies may also promote engagement of a larger proportion of
otherwise appropriate patients as well as adherence to therapist recommendations for skill acqui-
sition and practice. Frantsve and Kerns (2007) have drawn attention to issues of patient–provider
interactions and shared medical decision-making in pain management and their importance in
considering patient motivation for engaging in pain self-management. These theoretical, clinical,
and empirical developments have contributed to increased attention to issues of motivation in
the field of pain management.
In this chapter we describe a motivational model of pain self-management and discuss its
potential relevance in clinical situations. Briefly, according to the model, patients’ readiness to
engage in pain self-management at the time of acute injury or illness, or even in the context of the
emergence of unexplained pain, is expected to influence their coping responses to pain, their
efforts to acquire and use new adaptive pain coping strategies, and ultimately, the course of pain
and the likelihood that pain will resolve. We also describe methods for assessing motivation and
readiness to adopt a self-management approach to chronic pain, methods for promoting motiva-
tion for pain treatment and for learning and implementing adaptive pain-coping skills, and strat-
egies for tailoring pain treatment to patient preferences, interests, and expectations for treatment.
We conclude the chapter with a more explicit discussion of the implications of the motivational
model and motivational issues for understanding and addressing the process of transitioning
from acute to chronic pain, and for preventing this transition.
24.1 Models of motivation and a motivational model

of pain self-management
Theories of motivation attempt to explain the initiation, direction, persistence, intensity, and
termination of a particular behaviour (Landy and Becker 1987). Motivation can thus be viewed as
a process that involves all of the factors that influence behaviour. Most theories and models of
human behaviour, including a motivational model we are developing for understanding pain
self-management, assume that behavioural change is influenced primarily by two factors: (1) the
perceived importance of behaviour change and (2) the patient’s belief that behaviour change is
possible (i.e. self-efficacy) (Jensen et al. 2003). In fact, because of the high degree of overlap
among the models, as well as the fact that it is often possible to explain changes in motivation or
behaviour from the viewpoint of any one of the models, finding unequivocal support for one
model over the others is quite difficult (Weinstein 1993).
On the other hand, the existence of significant areas of overlap among motivational models
may be used to form a foundation for a general model of motivation for pain self-management.
MOTIVATIONAL ISSUES IN PAIN MANAGEMENT 455
Self-management
Perceived importance behaviours (coping)
Beliefs regarding Exercise
cost/benefit ratio Pacing
Learning history Relaxation
Current contingencies Assertiveness
Readiness to change Task persistence
(or maintain) self- Body mechanics
management behaviours Positive self-statements
Ignoring pain
Self-efficacy Avoid asking for assistance
Personal experience Avoid guarding
Modelling Avoid catastrophizing
Verbal persuasion Avoid pain contingent rest
Perceived barriers Avoid pain contingent
analgesics
Fig. 24.1 Motivational model of pain self-management. Reprinted from The Journal of Pain, 4(9),
Mark P. Jensen, Warren R. Nielson, and Robert D. Kerns. Toward the development of a motivational
model of pain self-management, pp. 477–92, Copyright (2003), with permission from Elsevier.
An initial version of such a model, the Motivational Model for Pain Self-Management, is
presented in Figure 24.1. The primary outcome variable in this model is pain self-management
coping behaviour, which may be defined by a set of behaviours and cognitions that are thought
to reflect ‘adaptive’ pain management, and by avoidance of behaviours or cognitions that are
thought to reflect ‘maladaptive’ pain management. The specific self-management coping behav-
iours listed in Figure 24.1 were drawn from the coping responses that clinicians and researchers
have most closely associated with improved function and with positive outcomes in pain treat-
ment (Jensen et al. 1994; Loeser and Turk 2001; Nielson et al. 2001), but the list is by no means
exhaustive.
Self-management behaviour that is adaptive for one condition may, however, be ineffective or
even harmful for another condition. For example, while patients with low back pain may benefit
from maintaining a programme of regular aerobic exercise, the same exercises might cause fur-
ther joint damage in patients with knee or hip arthritis. As more is learned about the relative
importance of specific coping behaviours and cognitions and about the conditions under which
these are adaptive, maladaptive, or neutral, the operational definition of pain self-management
listed in Figure 24.1 should be updated.
The concept of readiness to self-manage pain (Prochaska and DiClemente 1984a; Kerns et al.
1997) is central to the Motivational Model for Pain Self-Management because it defines motiva-
tion. The model hypothesizes that patients will engage in specific pain self-management strategies
dependent on their readiness, or motivation, to use these strategies. In the model, motivation is
influenced by the two primary variables already mentioned, i.e. beliefs about the importance of
engaging versus not engaging in pain self-management behaviours (‘outcome expectancies’,
‘value’, ‘importance’), and beliefs about one’s ability to engage in pain self-management behav-
iours (‘self-efficacy’, ‘confidence’). Perceived importance is influenced by the value of expected
outcomes, such as pain reduction, increased strength and activity tolerance, increased cognitive
abilities, versus the perceived costs of pain self-management. The outcome expectancies are in
turn affected by the patient’s learning history, since a history of reinforcers or punishers for cer-
tain pain self-management behaviours will respectively increase or decrease the value placed on
pain self-management.
Similarly, a number of factors can contribute to a patient’s confidence in his or her ability to
engage in a specific behavioural response. These include a history of successfully engaging in that
response while undergoing treatments that elicit new behavioural responses to pain (Fordyce
1976; Fordyce et al. 1968), modelling of behaviour by others (Bandura 1986), effective persuasion
(Miller and Rollnick 2002), and the removal of perceived barriers.
Although the Motivational Model for Pain Self-Management may appear static, with its final
endpoint determined by the effects of perceived importance and self-efficacy on readiness to self-
manage pain, we view the model as dynamic because of the many factors described above that
influence motivation. The model provides what we hope is a frame of reference for understanding
patient motivation for self-management, and, more importantly, for identifying ways to improve
this motivation.
Moreover, preliminary evidence supports hypotheses based on this model. Molton and
colleagues (Molton et al. 2008) obtained survey data from 130 adults with spinal cord injury
(SCI) and chronic pain, that included: (1) measures of average pain intensity during the past
week; (2) readiness to self-manage pain (assessed using the Multidimensional Pain Readiness to
Change Questionnaire, see Nielson et al. 2003, and discussion of motivational measures in the
next section); (3) and ratings of perceived importance, self-efficacy, and frequency of use of both
(a) regular exercise and (b) task persistence as pain management strategies. Task persistence was
used by too many of the sample to be able to test the mediation hypotheses of the Motivational
Model of Pain Self-Management; however, the distribution of the amount of exercise reported by
the sample had enough variance to test the model’s hypotheses. Molton et al.’s findings supported
the model. Specifically, patient self-efficacy for exercise, perceived importance of exercise, and
motivation (readiness) were all significantly associated with reported exercise frequency. As pre-
dicted, they also found that the effects of self-efficacy and perceived importance were mediated by
motivation. Thus, patient readiness to make changes in (or maintain) exercise behaviour is influ-
enced by their perception of the importance of exercise and their perception of their own ability
to exercise. In turn, patient readiness to exercise is a strong predictor of exercise behaviour.
24.2 Measurement of motivation in pain self-management

Central to the ability to test the validity of any theoretical or conceptual model is the ability to
reliably measure its core constructs. Similarly, the foundation of effective theoretically-informed
clinical intervention is assessment of the target of the intervention. For over a decade, the authors
of this chapter have embarked on a programme of research designed to develop and test meas-
ures of the concept of pain readiness to change as described in the Motivational Model of Pain
Self-Management described above.
In large measure, this measurement development effort has been informed by the
Transtheoretical Model of Behaviour Change (TTM) of Prochaska and his colleagues (Prochaska
and DiClemente 1984a; 1984b). The TTM incorporates the notion that behaviour modification
involves two interrelated dimensions. Stages of change is based on the observation that individuals
vary to the extent that they are ‘prepared’ or motivated to make changes in a specifically targeted
behaviour. The second dimension, processes of change, focuses on activities and events that con-
tribute to successful behaviour modification. A framework for considering ‘stages of change’ and
the development of reliable strategies for assessing an individual’s ‘stage’ has been pursued. Four
or five stages have been identified across multiple studies in several areas of health behaviour
change such as tobacco use, exercise, and diet (Prochaska et al. 1992). These are generally labelled
‘precontemplation’ referring to individuals who report a low interest in, or consideration of,
changing their behaviour; ‘contemplation’ describing individuals who report thinking about
behaviour changes, but appear unlikely to change in the near future; ‘preparation’ describing
those who are actively considering attempts to change their behaviour and are likely to do so in
the next month; ‘action’ referring to individuals currently engaged in behaviour change efforts;
and finally, ‘maintenance’ describing individuals engaged in maintaining their already changed
health behaviour. Identification of these groups of individuals has been shown to be valuable in
predicting success in treatment designed to modify behaviour, as well as maintenance of treat-
ment effects.
Informed by both a cognitive-behavioural perspective on chronic pain (Turk et al. 1983) and
the TTM, Kerns and his colleagues propose that individuals with chronic pain similarly vary in
their degree of readiness to adopt a self-management approach to their problem (Kerns et al.
1997). It is hypothesized that patients’ degree of readiness to change may predict both engage-
ment in self-management therapies, and outcomes. Based on integration of these models, it has
been hypothesized that self-management interventions such as cognitive-behaviour therapy,
through a process of reconceptualization, skill acquisition, and skill practice, may promote
increased acceptance of a self-management approach. Conversely, increased acceptance of this
perspective may be expected to enhance perceptions of personal control and responsibility for
pain management, promote adherence to treatment recommendations for skill acquisition and
practice, and contribute to improved outcomes.
Application of the pain readiness to change model begins with a consideration of the specific
beliefs that may be held by persons in different ‘stages’ of readiness to adopt a pain self-management
approach. Some persons hold strong beliefs that their pain problem is due to physical injury or
disease that requires medical attention. Such persons often believe that personal responsibility for
the condition or its solution rests with physicians and believe that self-management approaches
are misguided and unlikely to yield benefits. Persons adhering to these beliefs are considered to
be in a ‘precontemplation’ stage and are thought to be unlikely to be interested in engaging in
self-management therapies for pain. Others may believe that there are limits to the utility of a
medical approach and that learning self-management approaches may be useful. Such persons
may actively consider the benefits of acquiring assistance in learning personal skills for improving
pain control and reducing its negative impacts. ‘Contemplation’ is the term that has been used to
characterize a state of emerging interest in engaging in self-management treatment. Within the
model, the active learning of self-management strategies takes place in the ‘action’ stage, while
persons who have already incorporated self-management into their overall approach to chronic
pain are in the ‘maintenance’ stage.
The following sections describe ongoing efforts to develop psychometrically sound strategies
for the measurement of these sets of beliefs consistent with the concept of pain readiness to
change. Ultimately, the value of these tools is believed to be contingent on their ability to reliably
predict engagement, adherence, and outcomes from pain self-management therapies and their
utility in more extensive tests of the Motivational Model of Pain Self-Management described
above.
24.2.1 The Pain Stages of Change Questionnaire (PSOCQ)

The PSOCQ (Kerns et al. 1997) was developed to provide a reliable measure of motivation or
readiness to adopt a self-management approach to chronic pain and to assess the validity and
utility of the pain readiness to change model. Published data largely support the reliability, factor
structure, and criterion-related validity of the PSOCQ (Biller et al. 2000; Dijkstra et al. 2001;
Jensen et al. 2000, 2004; Kerns et al. 1997, 2005; Maurischat et al. 2002, 2006). Four scales of the
PSOCQ have been identified that are consistent with the four stages of the change process
described above. Item examples from the scales that assess precontemplation, contemplation,
action, and maintenance are ‘All of this talk about how to cope better is a waste of time’, ‘Even if
my pain doesn’t go away, I am ready to start changing how I deal with it’, ‘I am developing new
ways to cope with my pain’, and ‘I have learned some good ways to keep my pain problem from
interfering with my life’, respectively. The measure has now been used for research in several
countries and has been translated into other languages including Dutch (Dijkstra et al. 2001),
German (Maurischat et al. 2002), and Norwegian (Strand et al. 2007).
Classification of persons into one of the four stages of change has been most often been accom-
plished by labelling persons based on the highest score on the four PSOCQ scales. More recently,
utilizing cluster analysis and discriminant function analysis, Kerns and his colleagues proposed a
strategy for classification of five subgroups of persons in terms of readiness to change based on
profiles of scores on the PSOCQ (Kerns et al. 2005 ). The subgroups were labelled as
Precontemplation, Contemplation, Noncontemplative Action, Participation, and Ambivalent. It
remains for future research to determine which classification strategy has the greatest utility.
Since the initial publication of the measure and its psychometric properties, numerous addi-
tional reports provide evidence of its predictive validity. Kerns and Rosenberg (2000) demon-
strated that pretreatment scores on the PSOCQ reliably discriminated persons who completed a
10-session programme of cognitive-behavioural therapy for chronic pain. Furthermore, these
investigators found that increased action and maintenance scale scores, inferred as evidence of
‘forward stage movement,’ i.e. enhanced motivation and commitment to a self-management
approach, were correlated with improved outcomes on several key variables. These results have
generally been confirmed (Jensen et al. 2004; Kerns et al. 2005), although concerns have arisen
about the sensitivity and specificity of the PSOCQ in predicting treatment engagement and par-
ticipation (Biller et al. 2000), as well as concerns about the validity of the measure (e.g. Habib
et al. 2003; Strong et al. 2002). Most recently, Burns and his colleagues have demonstrated that
early treatment increases in readiness predict subsequent improvements in outcomes, providing
the strongest evidence to date of a role of enhanced motivation or readiness as a mediator of
improved outcomes associated with self-management treatments (Burns et al. 2005; Glenn and
Burns 2003).
The Multidimensional Pain Readiness to Change

24.2.2
Questionnaire (MPRCQ)
Rationale
In designing strategies to promote patient acquisition of pain-related coping skills, a key question
is whether their readiness to change is best conceptualized as unidimensional or multidimen-
sional. Perhaps the most parsimonious way to view pain-related readiness to change is as a single
dimension that perfuses all aspects of pain-related coping. The PSOCQ provides an example of
this approach as it captures the extent to which individuals are generally ready to adopt a self-
management approach. As noted above, the PSOCQ has been extremely useful in understanding
the role of motivation in psychological pain treatments. An alternative, complementary, approach
would be to assume that considerable variation may exist in the extent to which an individual
would be willing to adopt specific adaptive coping strategies and curtail maladaptive strategies.
Multidisciplinary pain programmes encourage patients to acquire a broad spectrum of coping
skills ranging from exercise to positive self-statements to avoiding pain behaviours. Anecdotally, it
seems that patients may view these various skills quite differently and appear more motivated to
utilize some skills rather than others. There are likely many reasons why such differences can occur
such as previous lack of efficacy, fear, personal beliefs, values and perceived social implications.
For example, if an individual believes that asking others for assistance is necessary, dislikes dis-
continuing work on a task before it is complete (pacing), feels that use of cognitive restructuring
techniques (avoid catastrophizing) implies that they are mentally ill, or has low self-efficacy for
particular coping behaviours they may be less willing to learn or adopt these strategies. Moreover,
it has been observed for some time (e.g. Turk and Rudy 1991), that not all patients appear to
benefit equally from multidisciplinary treatment. Designing interventions to enhance readiness
to change for specific coping behaviours that are associated with lower motivation has the poten-
tial to improve treatment outcomes and adherence. In order to target specific behaviours, they
must be accurately and reliably identified.
Initial development
Nielson, et al. (2003) began development of a MPRCQ (pronounced ‘em-perk’) to provide an
instrument that would allow initial research into a multidimensional model. The items for this
questionnaire were based on the self-management strategies often taught in multidisciplinary
programmes as described in our motivational model. Initial item development included ten pri-
mary scales as well as five cognitive control content scales. During the item analyses one scale,
‘Avoid Guarding’ was dropped because of difficulties in obtaining adequate internal consistency.
The resulting 46-item questionnaire had adequate internal reliability (α = 0.70 to 0.93) and there
was little or no effect of social desirability. The final nine primary scales assess readiness to engage
in the coping strategies of (1) exercise, (2) task persistence, (3) relaxation, (4) cognitive control,
(5) pacing, (6) avoiding pain-contingent rest, (7) avoiding asking for assistance, (8) assertive
communication, and (9) proper body mechanics. The five cognitive control subscales are
(1) diverting attention, (2) self-statements, (3) reinterpreting sensations, (4) avoiding catastro-
phizing, and (5) ignoring pain. Good concurrent validity for the PSOCQ scales was obtained in
the form of moderate correlations with the PSOCQ, the Chronic Pain Coping Inventory (Jensen
et al. 1995, 2008), a measure of pain coping and the Survey of Pain Attitudes (Jensen and Karoly
2008; Jensen et al. 1994), which measures attitudes regarding pain self-management.
Although these data were encouraging, there were a number of elements of the MPRCQ that we
felt warranted improvement. In particular, it is difficult to convey the concept of being ready to
stop engaging in a behaviour. The awkwardness of this concept, along with the combination of
instructions and the wording of some items that created apparent double negatives, made some
items more difficult for patients to understand. In addition, we had based the MPRCQ response
options on the stages of change and felt that, psychometrically and theoretically, it made more
sense to have these options reflect a readiness to change continuum rather than the discrete stages
of change. Indeed, the concept of discrete stages of change has been challenged in the health
behaviour change literature. For example, West (2005) has suggested that stages of change are
arbitrary, oversimplified (stages contain multiple constructs such as time, intent and past attempts
to change) and include only motivational elements of which the patient is consciously aware.
24.2.3 The MPRCQ2

A second version of the MPRCQ, the MPRCQ2 was developed in order to address some of these
concerns (Nielson et al. 2008). The most significant revision was division of items into two cate-
gories. This modification was made to address the ‘double negative’ contained in items involving
discontinuation of maladaptive behaviours. The first category included behaviours that multidis-
ciplinary treatment programmes seek to increase (e.g. exercise) and the second category contained
those behaviours that are discouraged (e.g. contingent rest). Separate sets of instructions were pro-
vided for each of these two sections to increase clarity for patients completing the questionnaire.
Thus, for items designed to measure increases in adaptive behaviours the following instructions
and options were:
Please circle the number that best indicates your intention to use each of the following methods of
coping with or managing your pain by using the 1 to 7 rating scale below:
1 = I am not doing this now, and am not interested in ever doing it.
2 = I might do this someday but I have made no plans to do it.
3 = I will probably start doing this sometime (in the next 6 months).
4 = I have made plans to start doing this soon (within the next month).
5 = I have recently started doing this (within the past month).
6 = I have been doing this for a while (more than a month but less than 6 months).
7 = I have been doing this for a long time (at least 6 months).
Items that measure decreases in maladaptive behaviours were placed in a second section of the
questionnaire and prefaced with the following instructions:
The following questions are slightly different than those that you have already answered. For the
remaining questions, please circle the number that best indicates your intention to stop using each of
the methods of coping with or managing your pain by using the 1 to 7 rating scale below:
1 = I am doing this now and am not interested in ever stopping.
2 = I might stop this someday, but I have made no plans to stop doing it.
3 = I will probably stop doing this sometime (in the next six months).
4 = I have made plans to stop doing this soon (within the next month).
5 = I have recently stopped doing this (sometime in the past month).
6 = I have not done this for a while (more than a month but less than 6 months).
7 = I have not done this for a long time (at least 6 months).
The MPRCQ2 is also considerably longer than the MPRCQ (containing 69 items) and was
constructed utilizing a broad sample of individuals with chronic pain (fibromyalgia, arthritis,
spinal cord injury, acquired amputation). The MPRCQ2 is reliable (α = 0.77–0.94) and shows
adequate construct validity. Confirmatory factor analysis identified two overall factors contained
in the MPRCQ2, which we labelled ‘Active Coping’ and ‘Perseverance’. The former is composed
of scales that tap readiness to actively participate in pain management behaviours (e.g. Pacing,
Exercise, Relaxation Techniques, Assertiveness, Proper Body Mechanics, Cognitive Coping
Techniques) whereas the latter contains scales that involve a readiness to persist in activity
(Avoidance of Pain-Contingent Rest, Avoidance of Asking for Assistance and Task Persistence).
Interestingly, in this model, the Task Persistence scale was the only scale to load on both factors,
suggesting that readiness to persist at tasks may be embedded in both readiness to actively cope
and to persevere. An interesting direction for future research would be to explore the differential
roles of readiness for Active Coping versus Perseverance in treatment participation, adherence
and outcomes.
In addition to these two factors, other aspects of our data strongly supported the multidimen-
sional nature of pain-related readiness for change. The low correlations among the MPRCQ2
scales, significant correlations between these scales and their criterion scales, and high internal
consistencies all support this contention. Our confidence in this view was strengthened inasmuch
as these results replicated the pattern obtain in our construction and analysis of the original
MPRCQ using different clinical samples.
We were somewhat disappointed that during the process of developing the MPRCQ2 we were
unable to create a readiness to Avoid Guarding scale. We had not be able to create a psychometri-
cally acceptable Avoid Guarding scale for the MPRCQ and in the current iteration we had hoped
that including different additional items would have resulted in a reliable, valid scale. Our item
analyses indicated that in the Acquired Amputation and Arthritis samples the items of this scale
were not well intercorrelated, although they were adequate in the Fibromyalgia (α = 0.79) and
Spinal Cord Injured (α = 0.72) samples. Also puzzling was a lack of association between MPRCQ2
Guarding scale and the CPCI Guarding scale, a criterion measure. Clearly our understanding of
readiness to Avoid Guarding requires revision and further exploration.
24.2.4 Brief versions of the MPRCQ2

The last of our triad of studies on the MPRCQ investigated the feasibility of short versions of the
MPRCQ2. The MPRCQ2 is relatively long and in some contexts (e.g. lengthy batteries or ques-
tionnaires, repeated assessment over short intervals) it may be prohibitively long. Excessive
response demands can adversely influence instrument reliability and validity, not to mention
study participation rates. Following the successful development brief versions of other pain-
related psychological questionnaires (Jensen et al. 2003; Tan et al. 2006) we have also developed
one and two-item per scale versions of the MPRCQ2 (Nielson, et al. 2009). Patients attending
Arthritis and Fibromyalgia Day programmes in London, Ontario served as the sample for this
study. For the single-item version, we identified items that were at least moderately correlated
with their parent scale, were sensitive to pre-post treatment change and had at least a modest pat-
tern of association with the PSOCQ scales (concurrent validity). Each author then reviewed the
items that met these criteria and selected those that he or she felt best represented the underlying
construct. Final selection was based upon consensus among the four authors that the item best
met criteria. In the few instances where any disagreement existed, the item that was selected by the
majority was chosen. For the two-items per scale version, in order to maximize validity, regres-
sion analyses were conducted in which parent scale variance was predicted first entering the item
selected using the preceding methodology and then entering other items from the parent scale.
The item that produced the largest additional contribution to the prediction of variance and was
viewed by the investigators as reflecting the breadth of the parent scale was chosen. For both pre
and post-treatment time points, all but one two-item scale (Relaxation, 0.78 at post-treatment)
had a correlation of 0.80 or greater with the parent scale. Change indices indicated that both the
one and two-item scales were sensitive to changes that occurred over the course of the treatment
programmes. Most of the scales correlated well with the PSOCQ criterion measure, in a similar
pattern to that obtained between the parent scales and the PSOCQ. As might be expected, the
two-item versions tended to correlate more highly with the parent scales than the single item
versions. The two versions did not differ in their sensitivity to change.
These 13- and 26-item versions should be especially useful in three circumstances: (1) when
administered with a large number of other questionnaires; (2) in treatment process studies where
frequent re-administration is required in order to access pain-related readiness to change, for
example, as a predictor of adherence; and (3) in predicting success in multidisciplinary treatment
programmes (i.e. treatment outcomes).
24.2.5 Future research

The MPRCQ, the MPRCQ2 and its brief versions provide valid, reliable tools with which to study
motivational processes associated with adoption of, and adherence to, pain self-management
strategies. Motivation to learn and utilize pain self-management strategies is a factor that is well
known to clinicians working in programmes that teach such skills. Although we are not aware of
any formal surveys asking clinicians about the role of motivation, based upon our experiences
discussion with colleagues and the few references to the issue in the literature (e.g. Thomson
2008), there is essentially uniform acknowledgment of the importance of motivational issues. By
definition, the success of self-management strategies is highly dependent upon the motivation
of the individual to learn and apply them. Given the apparent importance of motivational factors,
it is surprising that they have not been the focus of more research interest. An improved under-
standing of the motivational processes that influence these outcomes is perhaps long overdue.
We hope that our motivational model will encourage research in this area and that the
MPRCQ2 and PSOCQ will provide psychometric windows into the processes involved in pain
self-management.
One approach to research using these scales is to examine the variations in readiness to acquire
or discontinue pain coping skills within and between patient groups. Figure 24.2, illustrates the
variation in MPRCQ scale scores among patients seen at St Joseph’s Health Care, London,
Ontario. The MPRCQ profiles of those seen prior to entering the Fibromyalgia Day programme
demonstrated lower scores on some MPRCQ scales compared a sample of rheumatology outpa-
tients (inflammatory arthritis). It is worth noting that this type of inter-scale variation occurred
even though overall readinesses to change scores are similar (in this case 4.6 for the arthritis
sample and 4.3 for the FM sample). In addition, variation among MPRCQ scale scores appears
to exist across scales for both groups such that, for example, in both groups patients were more
willing to change their tendency to persist at tasks but less willing to avoid pain-contingent rest.
Thus, a multidimensional measurement perspective also fits well with our motivational model of
pain self-management (Figure 24.1) inasmuch as there appears to be variation among responses
to different MPRCQ scales within a single group. Obviously, lack of variation would have pro-
vided some disconfirmatory evidence with regard to a multidimensional approach. There may
also be different levels of perceived importance and self-efficacy associated with each self-
management content area as well. That is, it would be expected that scales representing specific
pain self-management content areas would also reflect underlying differences in perceived impor-
tance and/or self-efficacy. At least theoretically, higher MPRCQ scale scores should be associated
with greater self-efficacy, greater perceived importance or both (see Molton et al. 2008).
The ability of the MPRCQ and MPRCQ2 to predict treatment outcomes is not yet established,
but as research in this area progresses, the measure may be used to identify appropriate treat-
ments or to suggest targets for motivational interviewing of patients. Research will also determine
whether the multidimensional MPRCQ/MPRCQ2 represent an improvement over the more
widely used, stage-based, unidimensional indices of motivation such as the PSOCQ. More gener-
ally, whether stage assignment versus simpler continuum-based readiness to change scores offers
any predictive advantage and whether assessment modality-specific readiness offers advantages
over assessment of general readiness to self-manage pain are important issues both theoretically
and practically. The MPRCQ and MPRCQ2 reflect the specific behavioural targets within our
Motivational Model of Pain Self-Management and, in this sense, these instruments are perhaps
ideally suited for future efforts to evaluate that model. However, it is possible that a more trait-
like readiness to change factor, accounting for the majority of variance in readiness to change
individual coping behaviours will provide a more parsimonious account of changes in self-
management behaviours. The specificity of that aspect of our model depends upon which of these
possibilities is borne out by future research. In practice, clinical implications are whether interven-
tions such as Motivational Interviewing should be directed toward individual self-management
behaviours or are more efficiently applied to the broader adoptions of pain self-management
skills as a whole (e.g. tailored vs. general Motivational Interviewing).
MPRCQ scale scores by group
6.5
5.5
5
Scale scores
4.5
3.5
MOTIVATIONAL ISSUES IN PAIN MANAGEMENT

3
2.5
2
Exercise Task Relaxation Cognitive Pacing Avoid PC rest Avoid Assertive Body
persistence control assistance comm. mechanics
Rheum OP clinic (n = 44) FM day program (n = 131)
Fig. 24.2 MPRCQ scale scores by group. Reprinted from The Journal of Pain, 4(9), Mark P. Jensen, Warren R. Nielson, and Robert D. Kerns. Toward the
development of a motivational model of pain self-management, pp. 477–92, Copyright (2003), with permission from Elsevier.
463
24.3 Clinical and research implications

24.3.1 Motivational Interviewing
As indicated previously, we view motivation for pain self-management as a state that can be
altered by a number of factors, many of which can be influenced by clinician responses to patient
behaviours. To enhance patient perceptions of the importance of pain self-management, clini-
cians can encourage positive outcome expectancies and outline the costs of not engaging in self-
management strategies; they can help patients identify and incorporate into their lives positive
contingencies (reinforcers) for self-management coping behaviours; and they can praise patients
for making gradual changes in the direction of pain self-management (i.e. ‘shaping’). To increase
self-efficacy, the clinician can: encourage the patient to practice self-management strategies; pro-
vide the patient with opportunities to observe other patients engaging in pain self-management
strategies; gently challenge ‘distorted’ cognitions, providing directed active listening to encourage
and support self-efficacy beliefs; and help the patient develop a plan to address any perceived or
real barriers to pain self-management.
Many of these suggestions have been incorporated into a set of therapeutic strategies called
Motivational Interviewing (Miller and Rollnick 2002), which was designed specifically to increase
motivation for positive change. Motivational interviewing is both a general approach and a set
of specific therapeutic strategies designed to address ambivalence about adaptive change (e.g.
adaptive pain management and return to work).
In terms of the general approach, Motivational Interviewing is patient-centred (Douaihy et al.
2005), thus what the patient says and does during a motivational interaction guides clinician
responses. Motivational clinicians should seek to assess and address the patient’s worries and
concerns, build and enhance rapport and a collaborative relationship, provide information that
the clinician knows or believes the patient is truly ready to hear, alter the focus of interventions
from making treatment recommendations (the expert model) to supporting and enhancing
patient self-care practices (the coaching model), and foster greater control of decision making
and responsibility for self-care. To help facilitate an atmosphere of positive change and to build
rapport, the single most important clinician response is listening to the patient, especially in a
reflective way, rephrasing the thoughts and feelings that the patient is trying to communicate.
While skilled reflective listening might be considered the foundation of Motivational
Interviewing, other specific responses are involved, including developing discrepancy, avoiding
argumentation, rolling with resistance, and supporting self-efficacy (Miller and Rollnick 2002).
Clinicians help patients see discrepancies by encouraging them to talk about the problem, listen-
ing specifically for discrepancies between their goals and their behaviours, and then reflecting
back those discrepancies that are consistent with adaptive pain management. Arguments with
patients are to be avoided because they could provide patients with an opportunity to talk them-
selves out of positive change. The concept of rolling with resistance refers to a strategy of avoiding
confrontation, and in fact seeking to identify and reflect some adaptive component of any resist-
ant comments a patient might make. Supporting self-efficacy refers to a set of clinician responses
that elicit and reinforce positive patient statements about their capabilities.
When applied to pain self-management, the goal of Motivational Interviewing is to increase
the probability that patients will make adaptive decisions concerning their response to chronic
pain (Jensen 2002). Based on our Motivational Model for Pain Self-Management, to be most
effective, Motivational Interviewing and other strategies should specifically target self-efficacy
beliefs and the perceived importance of change, so as to alter a patient’s readiness or motivation
to change or maintain self-management behaviour, leading to subsequent changes in these
behaviours.
24.4 Implications of the motivational model and motivational

issues for understanding and addressing the process from
acute to chronic pain
The Motivational Model assumes that a key factor in adjustment to chronic pain is pain coping
behaviour; adaptive coping is thought to contribute to positive adjustment (minimal pain, mini-
mal pain impact, and maximum functioning) and maladaptive coping is thought to contribute to
poor adjustment and maladaptation. Along these lines, and according to the Motivational Model,
an individual’s readiness to self-manage pain at the time of acute injury would be expected to
influence his or her coping responses to that pain, and ultimately contribute to the course of that
pain and its influence.
While a number of studies have been performed to determine the predictors of movement
from acute to chronic pain, and psychosocial factors are often identified as the factors most
strongly associated with the development of chronic pain (cf. Schultz et al. 2004 ), coping
responses, and self-efficacy and importance for adaptive coping, are rarely studied. When coping
is studied, it consistently emerges as a significant predictor of future functioning in patients with
acute pain (e.g. Potter and Jones 1992; Prkachin et al. 2007). We hypothesize that the ability to
predict the development of chronic pain in individuals with acute or (sub)acute pain would be
enhanced if researchers assessed a broader range of coping responses, self-efficacy for pain self-
management, and the importance of pain self-management as the predictors in their model, and
readiness to learn new coping strategies.
Similarly, in order to minimize the development of chronic pain, it would be useful to consider
methods for maximizing adaptive coping and for encouraging the development of perceptions of
self-efficacy for pain self-management and beliefs in the importance of pain self-management
very early (or even before) the development of acute pain. For example, general education pro-
grammes provided to the population, through television and radio (public service) announce-
ments, concerning the importance of remaining active following an acute injury (Liddle et al.
2007), and provided that this is supported following an appropriate medical evaluation, might
decrease the incidence of the development of chronic pain in populations of individuals.
In health care settings, physicians and other health care providers who see patients who have
been recently injured should strive to assess the patients’ coping responses, and use interventions,
as appropriate, to increase patient self-efficacy for and perceptions of the importance of adaptive
responding (e.g. activity), in order to increase the changes that the patient will actually respond to
the acute pain in an adaptive way. For these reasons, we view skills in Motivation Interviewing
(briefly described above) as a key skill for health care providers who work with patients with both
acute and chronic pain. In support of this idea, one recent study investigated primary care provid-
ers who were trained to utilize Motivational Interviewing and patient-centred counselling to
increase the likelihood that patients would accept a referral to a multidisciplinary pain treatment
programme (Heapy et al. 2005). In this study, pretreatment PSOCQ scores on the contemplation
scale significantly predicted adherence to the therapist’s recommendations for inter-session prac-
tice of pain-coping skills (e.g. relaxation, activity pacing). Furthermore, inter-session adherence
was correlated with forward stage movement and behavioural goal accomplishment.
In a study in progress at the VA Connecticut Healthcare System, we are currently evaluating the
efficacy of a modified cognitive-behavioural therapy approach that explicitly incorporates patient
preferences, perceived importance, and expectancies of effectiveness for learning specific pain-
coping skills, using a modified version of the MPRCQ and motivational interviewing in an effort
to enhance forward stage movement and adherence. The treatment, labelled Tailored Cognitive-
Behavioural Therapy, is informed by the Motivational Model of Pain Self-Management and by
observations of the inherent flexibility in cognitive-behavioural therapy that allows for prescrip-
tive treatment planning. The model also encourages the development of alternative strategies that
attempt to match patients’ apparent degree of readiness to change with specific interventions.
Preliminary results have begun to help tease apart processes of change during the delivery of this
pain self-management intervention (Burns et al. 2008a, 2008b; Kerns et al. 2008). These results
further highlight the potential value of specifically targeting the beliefs and attitudes emphasized
in the Motivational Model of Pain Self-Management.
It seems reasonable to conclude that theory driven research in the area of motivation has
already proven useful in advancing our understanding of mechanisms and processes of change in
persons adoption of adaptive pain self-management. Central to this work is the availability of reli-
able measures of core constructs of the model and the use of these measures in the context of
trials of self-management treatment programmes in order to better understand these processes.
Continued efforts to apply and study the model in this area appear to have the potential to enhance
the effectiveness of self-management treatment and to promote the engagement and participation
of a larger proportion of persons with chronic pain who may benefit from the interventions. The
model similarly appears to have relevance for understanding the transition from acute to chronic
pain, at least among persons who are vulnerable to this transition. It is intuitive that interventions
informed by the Motivational Model and research designed to complement the provision of reas-
surance by providers when confronted with persons experiencing acute pain may have utility in
promoting an adaptive response to acute pain and in reducing the likelihood of persistent pain
and disability. Research informed by this interesting possibility is encouraged. Ultimately, we
believe that a greater clinical and research focus on motivational issues will result in more adap-
tive adjustment to acute and chronic pain overall, and therefore contribute to less suffering.
References
Bandura, A. (1986). Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ:
Prentice Hall.
Biller, N., Arnstein, P., Caudill, M, Federman, C., Guberman, C. (2000). Predicting completion of
cognitive-behavioral pain management program by initial measurement of chronic pain patients’
readiness to change. Clinical Journal of Pain, 16, 352–9.
Burns, J.W., Glenn, B., Lofland, K., Bruehl, S., Harden, R.N. (2005). Stages of change in readiness to adopt
a self-management approach to chronic pain: the moderating role of early-treatment stage progression
in predicting outcome. Pain, 115, 322–31.
Burns, J., Rosenberger, P., Heapy, A, Shulman, M., Kerns, R. (2008a). Tailored versus Standard
Cognitive-Behavioral Treatment for chronic pain: Differential effects of ‘dose’. Annual meeting of the
Society for Behavioral Medicine, San Diego, CA, 26–29 March.
Burns, J., Rosenberger, P., Heapy, A., Shulman, M., Kerns, R. (2008b). Effects of early-treatment changes
in attitudes toward self-management of pain on treatment outcomes. Annual meeting of the Society for
Behavioral Medicine, San Diego, CA, 26–29 March.
Clark, N.M., Becker, M.H., Janz, N.K., Lorig, K., et al. (1991). Self-management of chronic disease by older
adults: a review and questions for research. Journal of Aging and Health, 3, 3–27.
Damush, T.M., Weinberger, M., Perkins, S.M., et al. (2003). The long-term effects of a self-management
program for inner-city primary care patients with acute low back pain. Archives of Internal Medicine,
163, 2632–8.
Dijkstra, A., Vlayen, J., Rinjen, H., Nielson, W. (2001). Readiness to adopt the self-management approach
to cope with chronic pain in fibromyalgic patients. Pain, 90, 37–45.
Douaihy, A., Jensen, M.P., Jou, R.J. (2005). Motivating behavior change in persons with chronic pain.
In B. McCarberg and S. Passik (Eds.) Expert Guide to Pain Management, pp. 217–32. Philadelphia, PA:
American College of Physicians.
Fordyce, W. (1976). Behavioral Methods for Chronic Pain and Illness. Saint Louis, MO: Mosby.
Fordyce, W., Fowler, R., DeLateur, B. (1968). An application of behavior modification technique to a
problem of chronic pain. Behaviour Research and Therapy, 6, 105–7.
Frantsve, L. M., Kerns, R. D. (2007). Patient-provider interactions in the management of chronic
pain: current findings within the context of shared medical decision making. Pain Medicine,
8(1), 25–35.
Glenn, B., Burns, J.W. (2003). Pain self-management in the process and outcome of multidisciplinary
treatment of chronic pain: evaluation of a stage of change model. Journal of Behavioural Medicine, 26,
417–33.
Habib, S., Morrisey, S.A., Helmes, E. (2003). Readiness to adopt a self-management approach to pain:
evaluation of the pain stage of change model in a non-clinic sample. Pain, 22, 283–90.
Heapy, A., Otis, J., Marcus, K.S., et al. (2005). Intersession coping skill practice mediates the relationship
between readiness for self-management treatment and treatment outcome. Pain, 118, 360–8.
Hoffman, B.M., Papas, R.K., Chatkoff, D.K., Kerns, R.D. (2007). Meta-analysis of psychological
interventions for chronic low back pain. Health Psychol, 26(1), 1–9.
Jensen, M.P. (2002). Enhancing motivation to change in pain treatment. In D.C. Turk, and R.J. Gatchel
(Eds.) Psychological Treatment for Pain: A Practitioner’s Handbook, 2nd ed., pp. 71–93, New York:
Guilford Publications.
Jensen, M.P., Karoly, P. (2008). Survey of Pain Attitudes: Professional Manual. Lutz, FL: Psychological
Assessment Resources.
Jensen, M.P., Turner, J.A., Romano, J.M., Lawler, B.K. (1994). Relationship of pain-specific beliefs to
chronic pain adjustment. Pain, 57, 301–9.
Jensen, M.P., Turner, J.A., Romano, J.M., Strom, S.E. (1995). The Chronic Pain Coping Inventory:
development and preliminary evaluation. Pain, 60, 203–16.
Jensen, M.P., Nielson, W.R., Romano, J.M., Hill, M.L., Turner, J.A. (2000). Further evaluation of the pain
stages of change questionnaire: is the transtheoretical model of change useful for patients with chronic
pain? Pain, 86, 255–64.
Jensen, M.P., Keefe, F.J., Lefebvre, J.C., Romano, J.M., & Turner, J.A. (2003). One- and two-item measures
of pain beliefs and coping strategies. Pain, 104, 453–69.
Jensen, M.P., Nielson, W.R., Turner, J.A., Romano, J.M., Hill, M.L. (2004). Changes in readiness to
self-manage pain are associated with improvement in multidisciplinary pain treatment and pain
coping. Pain, 111, 84–95.
Jensen, M.P., Turner, J.A., Romano, J.M., Nielson, W.R. (2008). Chronic Pain Coping Inventory:
Professional Manual. Lutz, FL: Psychological Assessment Resources.
Keefe, F.J., Lefebvre, J.C., Kerns, R.D., et al. (2000). Understanding the adoption of arthritis self-
management: stages of change profiles among arthritis patients. Pain, 87, 303–13.
Kerns, R.D., Habib, S. (2004). A critical review of the pain readiness to change model. Pain, 5, 357–67.
Kerns, R., Rosenberg, R. (2000). Predicting responses to self-management treatments for chronic pain:
application of the pain stages of change model. Pain, 84, 49–55.
Kerns, R.D., Rosenberg, R., Jacob, M.C. (1994). Anger expression and chronic pain. Journal of Behavioural
Kerns, R., Rosenberg, R., Jamison, R., Caudill, M., Haythornthwaite, J. (1997). Readiness to adopt a
self-management approach to chronic pain: the Pain Stages of Change Questionnaire (PSOCQ). Pain,
72, 227–34.
Kerns, R.D., Bayer, L.A., Findley, J.C. (1998). Motivation and adherence in the management of chronic
pain. In A.R. Block, E.F. Kremer, and E. Fernandez (Eds.) Handbook of Pain Syndromes:
A Biopsychosocial Perspective, pp. 99–121. Mahwah, NJ: Lawrence Erlbaum.
Kerns, R., Wagner, J., Rosenberg, R., Haythornthwaite, J., Caudill-Slosberg, M. (2005). Identification of
subgroups of persons with chronic pain based on profiles on the pain stages of change questionnaire.
Pain, 116, 302–10.
Kerns, R.D., Jensen, M.P., Nielson, W.R. (2006). Motivational issues in pain self-management. In H. Flor,
E. Kalso, and J.O. Dostrovsky (Eds.) Proceedings of the 11th World Congress on Pain, pp. 555–66.
Seattle, WA: IASP Press.
Kerns, R.D., Rosenberger, P., Shulman, M., Heapy, A., Burns, J.W. (2008). Tailored CBT versus
Standard CBT for treatment of chronic pain: Early treatment stage change in attitudes toward pain
self-management and the working alliance. Annual meeting of the American Pain Society, Tampa, FL,
May 8–10, 2008.
Landy, F., Becker, W. (1987). Motivation theory reconsidered. Research on Organizational Behavior,
9, 1–38.
Liddle, S.D., Gracey, J.H., Baxter, G.D. (2007). Advice for the management of low back pain: a systematic
review of randomised controlled trials. Manual Therapies, 12(4), 310–27.
Loeser, J., Turk, D. (2001). Multidisciplinary pain management. In J. Loeser, S. Butler, C. Chapman, and
D. Turk (Eds.) Bonica’s Management of Pain, pp. 2069–79, Philadelphia, PA: Lippincott Williams & Wilkins.
Lorig, K.R., Sobel, D.S., Stewart, A.L, et al. (1999). Evidence suggesting that a chronic disease
self-management program can improve health status while reducing hospitalization. Medical Care,
37, 5–14.
Maurischat, C., Harter, M., Ayclair, P., Kerns, R.D., Bengel, J. (2002). Preliminary validation of a German
version of the pain stages of change questionnaire. European Journal of Pain, 6, 43–8.
Maurischat, C., Härter, M., Kerns, R.D., Bengel, J. (2006). Further support for the Pain Stages of Change
Questionnaire: Suggestions for improved measurement. European Journal of Pain, 10, 41–9.
Miller, W., Rollnick, S. (2002). Motivational Interviewing: Preparing People to Change, 2nd ed. New York:
Guilford Press.
Molton, I.R., Jensen, M.P., Nielson, W., Cardenas, D., Ehde, D.M. (2008). A preliminary evaluation of the
motivational model of pain self-management in persons with spinal cord injury related pain. Journal of
Pain, 9, 606–12.
Newman, S., Steed, L., Mulligan, K. (2004). Self management interventions for chronic illness. Lancet,
364, 1523–37.
Nielson, W.R., Armstrong, J.M., Jensen, M.P., Kerns, R.D. (2009). Two brief versions of the
Multidimensional Pain Readiness to Change Questionnaire (MPRCQ2). Clinical Journal of Pain,
25, 48–57.
Nielson, W.R., Jensen, M.P., Hill, M.L. (2001). An activity pacing subscale for the Chronic Pain Coping
Inventory: development in a sample of patients with fibromyalgia syndrome. Pain, 89, 111–115.
Nielson, W.R., Jensen, M.P., Kerns, R.D. (2003). The Multidimensional Pain Readiness to Change
Questionnaire (MPRCQ): initial development and validation. Journal of Pain, 4,149–59.
Nielson, W.R., Jensen, M.P., Ehde, D.M., Kerns, R.D., Molton, I.R. (2008). Further development of the
Multidimensional Pain Readiness to Change Questionnaire: the MPRCQ2. Journal of Pain, 9, 552–65.
Okifuji, A., Hare, B.D. (2011). Management of musculoskeletal pain. In M.H. Ebert and R.D. Kerns (Eds.)
Behavioral and Psychopharmacologic Pain Management, pp. 286–306. Cambridge: Cambridge
University Press.
Potter, R.G., Jones, J.M. (1992). The evolution of chronic pain among patients with musculoskeletal
problems: a pilot study in primary care. British Journal of General Practice, 364, 462–4.
Prkachin, K.M., Schultz, I.Z., Hughes, E. (2007). Pain behavior and the development of pain-related
disability: the importance of guarding. Clinical Journal of Pain, 23(3), 270–7.
Prochaska, J., DiClemente, C. (1984a). The Transtheoretical Approach: Crossing Traditional Boundaries of
Therapy. Homewood, IL: Dow Jones Irwin.
Prochaska, J. O., DiClemente, C. C. (1984b). Self change processes, self efficacy and decisional balance
across five stages of smoking cessation. Progress in Clinical Biological Research, 156, 131–40.
Prochaska, J. O., DiClemente, C. C., & Norcross, J. C. (1992). In search of how people change. Applications
to addictive behaviors. American Psychologist, 47, 1102–14.
Schultz, I.Z., Crook, J., Meloche, G.R., et al. (2004). Psychosocial factors predictive of occupational low
back disability: towards development of a return-to-work model. Pain, 107(1–2), 77–85.
Strand, E.B., Kerns, R.D., Christie, A., Haavik-Nilsen, K., Klokkerud, M., Finset, A. (2007). Higher levels
of pain readiness to change and more positive affect reduce pain reports–a weekly study of arthritis
patients. Pain, 127, 204–13
Strong, J., Westbury, K., Smith, G., Mckenzie, I., Ryan, W. (2002). Treatment outcome in individuals with
chronic pain: is the Pain Stages of Change Questionnaire (PSOCQ) a useful tool? Pain, 97, 65–73.
Tan, G., Nguyen, Q., Cardin, S.A., Jensen, M.P. (2006). Validating the use of two-item measures of pain
beliefs and coping strategies for a veteran population. J Pain, 7, 252–60.
Thomson, D. (2008). An ethnographic study of physiotherapists’ perceptions of their interactions with
patients on a chronic pain unit. Physiotherapy: Theory and Practice, 24, 408–22.
Turk, D.C., Gatchel, R.J. (Eds.) (2002). Psychological Treatment for Pain: A Practitioner’s Handbook, 2nd
edn. New York: Guilford Publications.
Turk, D.C., Rudy, T.E. (1991). Neglected topics in the treatment of chronic pain patients—relapse,
noncompliance, and adherence enhancement. Pain, 44, 5–28.
Turk, D., Meichenbaum, D., Genest, M. (1983). Pain and Behavioral Medicine: A Cognitive-Behavioral
Perspective. New York: Guilford Press.
Weinstein, N. (1993). Testing four competing theories of health-protective behavior. Health Psychology,
12, 324–33.
West, R. (2005). Time for a change: putting the Transtheoretical (Stages of Change) model to rest.
Addiction, 100, 1036–9.
Chapter 25
Pharmacotherapy of Low Back Pain

Kay Brune and Bertold Renner
25.1 Introduction
Acute low back pain (ALBP) is known to almost every human being aged above 50. Chronic low
back pain (CLBP) is one of the major health problems in Western industrialized nations (Chou
and Huffman 2007). The patients have often been suffering for years, lost their jobs and social ties.
The costs to the healthcare systems are enormous. Consequently, drug research has attempted to
find better remedies, regrettably with limited success (Deshpande et al. 2007; Chou et al. 2009;
Jones and Lamdin 2010; Kuijpers et al. 2010; Roelofs et al. 2008a, 2008b; Staal et al 2008; Van
Tulder et al. 2003, 2006a, 2006b). Other therapeutic attempts (not drug-based) were found to be
even less effective (Deshpande et al. 2007; Staal et al. 2008; Furlan et al. 2009; Dowswell et al. 2009;
Choi et al. 2010; Cui et al. 2010; Haroutiunian et al. 2010; Henschke et al. 2010; Oesch et al. 2010;
Peterson and Hodler 2010; Rubinstein et al. 2010; Van Middelkoop et al 2010). The available
armamentarium is insufficient and should be regarded as a mere walking stick to facilitate at least
temporarily rehabilitation or reduction of pain to an acceptable level, but not as a cure.
To cope with CLBP, the use of drugs has to be integrated in a more complex treatment system,
including physical therapy and psychotherapy as major pillars.
The complexity of the CLBP and the low effectiveness of its drug therapy result from several
components. Among those are (compare Figure 25.1):
◆ The human spine is inadequately engineered for a long life and a predominantly sedentary
lifestyle. Both further the loss of musculature and the decrease of bone density with age.
◆ Chronic structural remodelling of the human spine, often initiated by minor trauma and
perpetuated by degeneration of the small intervertebral joints (zygapophysial joints), leads to
narrowing of the intervertebral foramina, spasticity of the intervertebral muscles, overexposure
of the intervertebral ligaments to tear stress, and—possibly most importantly—damage to the
lumbar nerves and their meninges (Figure 25.1). These nerves and their meningeal sheaths are
particularly vulnerable when leaving the spinal cord and encountering the degenerative and at
times inflammatory environment at the intervertebral foramina which narrows with advanc-
ing age (see Chapter 4, this volume).
◆ Chronic nociceptor activation at the zygapophysial joints and progressing nerve damage
(neurogenic pain) will lead to changes in the pain perception and emotional tinting in the
central nervous system—a phenomenon which may be the major culprit of chronification of
LBP (see Chapter 15).
◆ The fine structures of the zygapophysial joints allow for acute nociceptor activation in differ-
ent anatomical structures. Additional meningeal and neuronal damage furthers chronic
neuropathic pain and thus chronification.
5
Fig. 25.1 (See also Colour Plate 12) Schematic drawing of the lumbar intervertebral joints
(zygapophysial joints), indicating the site of increased nociception in LBP (1–5). Note (outlined in
light blue) cartilage and synovialis of the intervertebral joints, painful when irritated and inflamed;
(red) intervertebral musculature which may—in order to immobilize the joint—become painfully
spastic; the lumbar nerves (yellow) squeezing through the narrow intervertebral foramen, nerve
damage and irritation of the meninges, causing nociceptor activation; the intervertebral disc—when
damaged—leads to invasion of nerves and activation of nociception (pink in the (blue) intervertebral
disk). Pathophysiological "sources" of low back pain: (1) Arthrosis of the vertebral joints (2) Damaged
(extended) intervertebral ligaments (3) Spasticity of the deep intervertebral muscles (4) Raptured disc
with secondary capillarisation and innervation (5) Damaged (irritated) neuronal fibers and sheaths
due to degeneration and deformation of the lumbar spine and inflammatory processes in the
meninges. Image by Christiane Wittek. Reproduced from Kubik, S.: FortbildK. Rheumatol., Bd.6,
pp. 1-29 © Karger, Basel, 1981, with permission.
It is obvious that in acute and even more in CLBP, the complete therapeutic armamentarium
has to be considered and the drugs to be used have to be selected carefully on the basis of the
process (inflammatory or alternatively predominantly neuropathic) as well as on the basis of
the risk factors innate to the individual patients who, in most instances, are elderly and often
multimorbid.
25.2 Strategies for drug therapy

LBP often starts acutely. A trauma to the small joints of the lumbar spine causes an inflamed syn-
ovium. This inflammation may be furthered by a protruding, ruptured annulus fibrosus of the
intervertebral disc of the lumbar joint which irritates the fine meninges surrounding and covering
the lumbar nerves (Figure 25.1). Both tissues produce prostaglandins which in turn further nocic-
eptor activation (wind up), hyperalgesia, and pain (Chapter 4, this volume). Hyperalgesia pro-
vokes spasticity of the local muscles (musculi intervertebrales and others), immobilizing the
painful joint, but adding to the local nociceptor circuitry (Fitzgerald 2005). Under these conditions
cyclo-oxygenase (COX) inhibitors often provide short-term relief which permits resuming physi-
cal activity and, as consequence, disappearance of the local spasticity. A COX inhibitor (selective
PHARMACOTHERAPY OF LOW BACK PAIN 473
Table 25.1 Effects of pharmacological interventions in chronic LBP (only drugs with proven
effect alone1)
Drug classes No. of No. of Statistical method Effect size

studies participants
NSAID
Comparison NSAID vs. placebo for chronic, non-specific LBP, follow up d 12 weeks
Change in pain intensity from 4 1020 Weighted mean difference −12.4
baseline on 100 mm VAS (fixed) 95% CI
Adverse events 4 1034 relative risk (fixed) 95% CI 1.24
Antidepressants2
Comparison antidepressants vs. placebo for chronic, non-specific LBP
Pain 4 292 Standardized mean difference −0.02
(random) 95% CI
adverse effects 2 157 Relative risk (fixed) 95% CI 0.93
Comparison SSRI vs. placebo
Pain 3 199 Standardized mean difference 0.11
(random) 95% CI
Opioids
Comparison opioids (all types) vs. placebo for chronic LBP
Pain 7 2350 Standardized mean difference −0.54
(fixed) 95% CI
Adverse effects 4 1176 Relative risk (fixed) 95% CI 1.28
With kind permission from Springer Science+Business Media: European Spine Journal, A systematic review on the effective-
ness of complementary and alternative medicine for chronic non-specific low-back pain, 19(8), 2010, Sidney M. Rubinstein.
1There is no reliable evidence of effectiveness for muscle relaxants (van Tulder 2003, 2006), antiepileptics, APAP (Davies
2008; Williams 2010), flupirtin (Li 2008, Worz 1996),tapentadol (Etropolski 2010, Wade 2009), topical NSAID (Haroutiunian
2010), glucosamine (Wilkens 2010), herbal remedies (Cui 2010) or other alternative medicines (Rubinstein 2010).
2 For review see also Urquhart (2008).
or non-selective) will lessen pain, facilitate return to mobility, and give relief of muscle spasticity.
The appropriate drug should be chosen on the basis of individual risk factors for each patient as
well as on the risks inherent to the individual drug (Table 25.1; see also 25.4).
If acute low back pain (ALBP) does not resolve within a time period of few weeks, chronifica-
tion is threatening. It may result from damage to the lumbar nerves, including the fine nerves,
innervating the meninges and the relicts of the disrupted intervertebral discs (Freemont et al.
1997; Edgar 2007; Takahashi et al. 2008). These changes in the pain pathology are reflected in the
observation that COX inhibitors tend to lose their effectiveness after a few weeks of use (Figure
25.2). The loss of pain relief often results in a desperate search for alternative therapeutic treat-
ment options which, as outlined before, do not help, but cause unwanted drug effects (UDEs).
In some countries of Europe, flupirtine is believed to be effective under these conditions (Li
et al. 2008). It is a potassium channel opener which might reduce spasticity and/or pain percep-
tion. However, RCT demonstrating reliable effects are lacking (Table 25.2). Flupirtine may cause
liver damage (Arzneimittelkommission der Deutschen Ärzteschaft 2007).
In this (later) phase, three therapeutic options may be explored (Table 25.2):
◆ The use of opioids. They have a limited, but measurable effect even in neuropathic pain
(Table 25.2).
0
Treatment group
–10 Placebo
Etoricoxib 60 mg
LS mean change ± SE
Etoricoxib 90 mg
–20
–30
‡
†
–40
–50
–60
S R 1 2 4 8 12
Study week
Fig. 25.2 The use of COX inhibitors in ALBP (flare period) may lead to pain reduction with
decreasing prevalence with time. Low back pain intensity scale (0 to 100-mm VAS). Least-squares
mean change (mm) from baseline (flare/randomization visit) over the 12-week treatment period
(all-patients-treated approach). Screening (S) to baseline (R) is the washout period from previous
analgesics. Data points for each treatment group were shifted along the x-axis to maximize legibility
at each time point. (Abbreviation: SE, standard error.) •P <0 .001 time weighted average versus
placebo. †P=0.001 time weighted average versus placebo. ‡P=0.018 time weighted average versus
placebo. Adapted from Birbara et al. (2003).
◆ Co-administration of antiepileptic drugs, as gabapentin and pregabalin, with opioids is fre-

quent. However, there is only limited evidence that these compounds, often claimed to be
effective in neuropathic pain, do work in CLBP (Table 25.3).
◆ Other types of combinations are also explored, e.g. opioids with muscle relaxants or antide-
pressants (Table 25.3). Again, strong evidence supporting clinical relevant effects of one or the
other combination is lacking.
If LBP continues, the central nervous system becomes a major player. Indeed, evidence from
functional magnetic resonance imaging (see, e.g. Kobayashi et al. 2009; also Chapter 6) indicates
that the emotional tinting of the pain as well as its intensity changes with time. Constant pain
perception dominates the day, and the patient may become depressive (Urquhart et al. 2008).
Employing antidepressants (adrenalin/serotonin reuptake inhibitors [SSRIs]) may then be
warranted—however, success is not guaranteed, and scientific evidence is scarce (Table 25.3).
It may be added that the infiltration of the small vertebral joints, ligaments, and muscles is
often exercised (Henschke et al. 2010; Peterson and Hodler 2010). Administering local anaesthet-
ics, steroids, and hyaluronic acid to these critical structures is not supported by evidence from
randomized controlled trials (RCTs) (Manchikanti et al. 2010). The local injection into the
complex tissues around lumbar joints may go along with infections, including intrathecal bacte-
rial dissemination with deleterious consequences, as lasting impairment of the ability to walk and
work. Herbal medicines (Cui et al. 2010), glucosamine (Wilkens et al. 2010), and homeopathic
medicines are not of documented clinical value. Acupuncture has been tried, but the interpreta-
tion of the effects seen is controversial (Berman et al. 2010).
Table 25.2 Evidence of therapeutic success in LBP from RCT
Drug/class Chronic LBP pd Reference

NSAID (non-selective)1
Diclofenac − − Dreiser 2003
Aceclofenac − − Roeloffs 2008
Naproxen + 0.01 Berry 1982
NSAID (selective)
Celecoxib − 0.01 Roelofs 2008
Etoricoxib + 0.01 Pallay 2004; Zerbini 2005; Birbara 2003
Acetaminophen − − Davies 2008; Williams 2010
Glucocorticoids − − Chou 2007
Opiods
Oxycodon + 0.01 Webster 2006
Oxymorphon + 0.01 Katz 2007
Tramadol + 0.001 Schnitzer 2000; Deshpande 2007
Muscle relaxants
Benzodiazepines − − van Tulder 2003, 2006
Anticholinergic drugs − − Kuijpers 2010
Flupirtine (+)2 0.01 Worz 1996; Li 20083
1 Effective in ALBP (Roelofs et al 2008a; Dreiser et al 2003).
2 Design questionable.
3 Underpowered, no active controls, no placebo (comp. flupirtine vs. tramadol: no difference, no test of equality, under-
powered).
Table 25.3 Effect of combining drugs in LBP
Drug combination Study type Acute/chronic LBP Outcome Reference

NSAID + opioid vs. NSAID RCT Chronic LBP Combination sig. better Jamison
alone (naproxen/oxycodon) (washout) p d 0.001 1998
NSAID + muscle relaxant RCT Acute LBP Combination better van Tulder
p d 0.01 2003
aceclofenac + tizanidin RCT Acute LBP combination better Pareek 2009
p d 0.01
NSAID + antiepileptic RCT Acute LBP Combination better Romano
(celecoxib + pregabalin) p d 0.01 2009
NSAID + opioid (APAR + RCT Chronic LBP Combination better Ruoff 2003
tramadol) p d 0.01
NSAID + opioid (APAP + RCT Chronic LBP Combination better Peloso 2004
tramadol) p d 0.001
No randomized controlled trials (RCT) were done using opioids + muscle relaxants, opioids + antiepileptics or opioids +
antidepressants (van Tulder et al, 2006a,b).
25.3 COX inhibitors: drugs with proven effectiveness in ALBP

The COX inhibitors are by far the most widely used drugs in ALBP. They divide into four
subgroups:
◆ The non-acidic, non-COX-2 selective inhibitors, namely APAP (paracetamol), phenazone
derivatives (propyphenazone, dipyrone).
◆ Acidic, non-selective COX-2 inhibitors, as, e.g. diclofenac, ibuprofen, meloxicame, naproxen,
and piroxicam.
◆ Selective COX-2 inhibitors, namely celecoxib and etoricoxib.
◆ Selective and acidic COX-2 inhibitors. Lumiracoxib is the only member of the forth still group
available in South America.
There is ample evidence that all COX inhibitors, given at adequate doses and in meaningful
galenic formulations, reduce ALBP and facilitate mobilization. Hard evidence from RCTs is avail-
able for diclofenac and naproxen, but lacking for phenazone derivatives and APAP, despite the
recommendation of the latter as first-line drug in many LBP guidelines (Davies et al. 2008). CLBP
is less amenable to COX inhibitor treatment. RCTs exist only for etoricoxib (Birbara et al. 2003;
Pallay et al. 2004; Zerbini et al. 2005), diclofenac, and naproxen (Berry et al. 1982; Jamison et al.
1998; Dreiser et al. 2003), but others are likely to be effective, too.
As shown in Figure 25.2, the few published, long-term evaluations of COX inhibitors indicate
that there is a loss of effectiveness with time (cf. Pallay et al. 2004). This may be taken as an indica-
tion for an increasing contribution of neuropathic pain components (Chapter 4). Neuropathic
pain is not amenable to COX inhibitors.
The COX inhibitors differ in their pharmacokinetic characteristics and potency (dosage) as
well as in their typical side effects. It is accepted that all COX inhibitors increase the blood pres-
sure (Johnson 1997) and the long-term risk of cardiovascular (CV) events, including cardiac
infarction, thromboembolic events, and stroke (Cannon et al. 2006; Dajani and Islam 2008;
Chang et al. 2010). The same risk exists for the long term-use of APAP (Sudano et al. 2010; for
review see Brune et al. 2009). Given a similar mode of action, proof of the contention that selec-
tive COX-2 inhibitors go along with a higher CV risk is still lacking (Dajani and Islam 2008).
Etoricoxib caused as many CV-events as diclofenac in a long-term study (Cannon et al. 2006). In
any case, it appears meaningful to use these drugs at relatively low doses (Hinz and Brune 2008).
On the other hand, the traditional, acidic COX inhibitors diclofenac, ibuprofen, meloxicame,
naproxen, and piroxicam are more likely to cause gastrointestinal (GI) tract bleeds and perfora-
tions as compared to either non-selective, non-acidic or selective COX-2 inhibitors (Rahme and
Nedjar 2007; Rahme and Bernatsky 2010). It appears meaningful to select the drug in CLBP
according to the risk factors presented by the individual patient and the pharmacokinetic/phar-
macodynamic characteristics of the drugs, i.e. otherwise healthy, aging patients could use all
prototypes; elderly patients with serious CV-risks might be served best with naproxen, and those
with GI-bleeds in their history may resort to COX-2-inhibitors, eventually in combination with
proton pump inhibitors (PPI) (Table 25.4).
A few drug-specific, clinically relevant aspects should be kept in mind:
◆ Diclofenac is often overdosed. Twice daily 50mg or once 75mg (retardation) is often a suffi-
cient dose and should not be exceeded as diclofenac causes liver damage, a dose-dependant
phenomenon (Boelsterli 2003).
◆ Etoricoxib, naproxen, and piroxicam are eliminated slowly from the human organism.
Consequently, giving the same dose daily leads to drug accumulation which may be the reason
for the relatively high toxicity of piroxicam. Also, since naproxen is a non-selective inhibitor
Table 25.4 Pharmacokinetic characteristics of oral COX inhibitors
NSAID/COX inhibitorse
Acidic Non-acidic
t50% Short Medium Long Short Medium Long
(∼1 h) (∼12 h) (t24 h) (∼1 h) (∼12 h) (t24 h)
Nonselective Ibuprofena Lornoxicam – Propyphen- Metamizol –
azone (dipyrone)d
t50% (h) 1–2 4–10 ∼1–4 >4 (MMA)
tmax (h)* ∼1* ∼1* ∼1 ∼0.5–1
FO (%) >80 ∼100 ∼1001 ∼100*
Vd (l/kg) ∼0.15 ∼0.15 ∼1 ∼1
Nonselective Ketoprofenb Naproxen‡,§,b Piroxicam‡,b – – –
t50% (h) 1–2 12–15 34–60
tmax (h) 0.5–1* 2–3* 0.5–2*
FO (%) ∼100 ∼100 >90
Vd (l/kg) ∼0.15 ∼0.15 ∼0.15
Preferential|| Diclofenacc – Meloxicam3 APAP c Celecoxib¶ –
t50% (h) 1–2 18–24 1.5–2.5 4–12
tmax (h) 0.5–1* ∼1* 0.5–1 2–4
FO (%) ∼60 ∼90 ∼80 20–40
Vd (l/kg) ∼0.15 ∼0.15 ∼1 >>1
Selective Lumiracoxibc – – – Rofecoxib§ Etoricoxib
t50% (h) 2–4 17 ∼24
tmax (h) 1–2 ∼3 0.5–1
FO (%) ∼100 ∼100 ∼100
Vd (L/kg) ∼0.15 >1.5 ∼2
* depending on the galenic formulation; ‡ enterohepatic circulation; § drug accumulation; || indicates inhibition of COX2
> COX1, as contrasted with selective inhibitors (COX2 inhibition >> COX1; i. e. lack of COX1 inhibition at therapeutic
doses); ¶ depending on CYP2C9 status. Abbreviations: t50%, plasma elimination half life; tmax, time to maximal plasma
concentration; FO, oral bioavailability; Vd, volume of distribution (comp. Brune et al. 2010; Brune et al. 2009).
a Interferes with platelet activity of ASA.
b High GI-toxicity, inhibits platelets for days.
c Hepatotoxicity.
d Agranulocytosis (rare).
e CV-risks in long-term use.
equally effective on COX-1 (which regulates platelet aggregation), an increased propensity for
bleeds, including GI bleeds, may be the consequence (Rahme and Nedjar 2007). This risk may
be antagonized partially by the co-administration of PPI (not effective in the low GI-tract)
(Chan et al. 2010).
◆ Ibuprofen is generally seen as relatively harmless. However, administering too high doses, i.e.
>1.2g/day, may increase its toxicity. Moreover, ibuprofen interferes with the effects of low-dose
acetylic salicylic acid (ASA) on platelet aggregation. Therefore, higher doses (>1.2g/day) should
be avoided and ibuprofen should not be given within one hour after the intake of ASA.
25.4 Opioids: drugs with moderate effect in CLBP

Opioids should not be used in ALBP: the effect is small, but UDEs are frequent. There are several
RCTs assessing the effectiveness of opioids in CLBP. The data available indicate that oxycodon,
oxymorphone, and tramadol may have therapeutic effects in CLBP (see Table 25.2). These opio-
ids should be used. Others, as buprenorphine, fentanyl, or morphine, appear less suited. All of
them display limited and variable bioavailability. Therefore, fentanyl is often administered as a
patch. The fact that the absorption is slow and not reliable is often not realized. Peak plasma con-
centrations are achieved only about 2 days after the application of the patch. In addition, effective
fentanyl concentrations persist for days after the removal of the patch.
The limited effect is counterbalanced by a relatively high rate of UDEs leading to high (∼50%)
dropout rates (Kuijpers et al. 2010; Machado et al. 2009). The therapeutic effect of the three opio-
ids mentioned is seen as statistically significant, but of minor clinical importance (compare the
effect size of Table 25.1). Consequently, opioids are often combined with other drugs which have
no proven effectiveness of their own in CLBP, but may increase the effect of opioids as well as
UDEs (see Table 25.3).
The risk of addiction going along with the use of opioids in CLBP appears to be low (Fishbain
et al. 2008). Serious side effects are rare. The use of opioids together with other drugs in CLBP is
common practice. But RCTs to document effects are still rare (Table 25.3). A pharmacological
derivative of tramadol is tapentadol which is claimed to combine the opioid effect with noradren-
aline/serotonin reuptake inhibition (Wade and Spruill 2009; Etropolski et al. 2010). The proof of
superiority of this combined effect of tapentadol is not documented by a RCT.
25.5 Muscle relaxants may enhance the effect of analgesics

Muscle relaxants alone do not appear to be useful therapeutic agents in either ALBP or CLBP
(Kuijpers et al. 2010). They are claimed, however, to be useful in combination with either NSAID
in ALBP or with opioids in CLBP (Table 25.3). The culprit of the muscle relaxants is that they are
heavily burdened with UDEs. The anticholinergic agents, as methocarbamol, orphenadrine, or
tizanidine, go along with drowsiness, headaches, and typical anticholinergic effects, as increase of
intraocular pressure, problems of voiding urine, and changes of the GI-tract motility. Alternatively,
benzodiazepines, including diazepam and tetrazepam, may be used. They also increase sedation,
tiredness, and drowsiness. They have no anticholinergic effects, but often induce drug abuse and
drug dependency. Worldwide, there is a dramatic overuse of benzodiazepines in elderly people
which gives rise to many fractures resulting of drowsiness and instability, particularly at night
(Pariente et al. 2008).
Taken together, muscle relaxants may be used for short periods of time only. Benzodiazepines
often induce abuse and cause fractures in elderly. In several European countries, flupirtine is used
in ALBP. It is claimed to have a spasmolytic effect due to opening potassium channels. The effect
of flupirtine in ALBP is not well documented (Li et al. 2008; Williams et al. 2010). Flupirtine
should not be used in specific or CLBP as it may cause liver toxicity (Arzneimittelkommission
der Deutschen Ärzteschaft 2007).
25.6 Antidepressants: sometimes useful as co-therapeutics

in CLBP
Antidepressants, in particular SSRIs, have no place in ALBP. In CLBP, their effect, when given
alone, is not different from placebo ( Urquhart et al. 2008 ). One RCT claims effectiveness
(Skljarevski et al. 2010). They appear to be beneficial in selected cases of CLBP when given in
combination with opioids (Chou and Huffman 2007). It is still an unsettled question if this effect
is due to the depressive mood of many patients with CLBP or whether it is based on an independ-
ent mechanism. Fluoxetine and duloxetine are used, but have as yet no proven indication in
CLBP (Staal et al. 2008).
25.7 Conclusion
It becomes clear that LBP is and will remain a major health problem. There is little proven evi-
dence of the therapeutic effects of many drugs and other treatments (Bouwmeester et al. 2009). In
ALBP, only COX inhibitors, used cautiously, may help to regain mobility and social functions
(Kroenke et al. 2009) which in turn allows for exercise that furthers recovery (Oesch et al. 2010).
As COX inhibitors lose effectiveness with time, opioids may step in. Their effect may be
enhanced by co-medication with either muscle relaxants or antiepileptics. However, opioids +
muscle relaxants or antidepressants show a high rate of UDEs. Most patients find these side effects
intolerable which leads to a dropout-rate (in studies) around 50% (Kuijpers et al. 2010).
Obviously, the treatment of CLBP requires therapeutic planning and long-term therapy,
including psychotherapy and physiotherapy as main pillars whilst the different drugs may be
regarded as auxiliary.
The biggest problem is the inability of many patients to cope and wait. Some resort to remedies
of unproven or even disproven value (effect), as herbal medicines, local steroids, topical non-
steroidals, acupuncture, an injection therapy using steroids, etc., and surgical interventions.
Whilst the former are unlikely to harm the patient seriously, attempts to inject into the fine zyga-
pophysial joints and surgical interventions in patients who have no hard indication (nerve decay
and paralysis) may add to the negative long-term outcomes of LBP.
References
Arzneimittelkommission der Deutschen Ärzteschaft. 2007. ‘Mitteilungen: Aus der UAW-Datenbank’:
Leberschäden unter Flupirtin. Deutsches Ärzteblatt, 104, A 3200.
Berman, B. M., Langevin, H. M., Witt, C. M. & Dubner, R. 2010. Acupuncture for chronic low back pain.
N Engl J Med, 363, 454–61.
Berry, H., Bloom, B., Hamilton, E. B. & Swinson, D. R. 1982. Naproxen sodium, diflunisal, and placebo in
the treatment of chronic back pain. Ann Rheum Dis, 41, 129–32.
Birbara, C. A., Puopolo, A. D., Munoz, D. R., et al. 2003. Treatment of chronic low back pain with
etoricoxib, a new cyclo-oxygenase-2 selective inhibitor: improvement in pain and disability—a
randomized, placebo-controlled, 3-month trial. J Pain, 4, 307–15.
Boelsterli, U. A. 2003. Diclofenac-induced liver injury: a paradigm of idiosyncratic drug toxicity. Toxicol
Appl Pharmacol, 192, 307–22.
Bouwmeester, W., Van Enst, A. & Van Tulder, M. 2009. Quality of low back pain guidelines improved.
Spine, 34, 2562–7.
Brune, K. & Hinz, B. 2004. Selective cyclooxygenase-2 inhibitors: similarities and differences. Scand
J Rheumatol, 33, 1–6.
Brune, K., Hinz, B. & Otterness, I. 2009. Aspirin and acetaminophen: should they be available over the
counter? Curr Rheumatol Rep, 11, 36–40.
Brune, K., Renner, B. & Hinz, B. 2010. Using pharmacokinetic principles to optimize pain therapy. Nat Rev
Rheumatol, 6, 589–98.
Cannon, C. P., Curtis, S. P., Fitzgerald, G. A., et al. 2006. Cardiovascular outcomes with etoricoxib and
diclofenac in patients with osteoarthritis and rheumatoid arthritis in the Multinational Etoricoxib and
Diclofenac Arthritis Long-term (MEDAL) programme: a randomised comparison. Lancet, 368,
1771–81.
Chan, F. K., Lanas, A., Scheiman, J., Berger, M. F., Nguyen, H. & Goldstein, J. L. 2010. Celecoxib versus
omeprazole and diclofenac in patients with osteoarthritis and rheumatoid arthritis (CONDOR): a
randomised trial. Lancet, 376, 173–9.
Chang, C. H., Shau, W. Y., Kuo, C. W., Chen, S. T. & Lai, M. S. 2010. Increased risk of stroke associated
with nonsteroidal anti-inflammatory drugs: a nationwide case-crossover study. Stroke, 41, 1884–90.
Choi, B. K., Verbeek, J. H., Tam, W. W. & Jiang, J. Y. 2010. Exercises for prevention of recurrences of
low-back pain. Cochrane Database Syst Rev, CD006555.
Chou, R., Baisden, J., Carragee, E. J., Resnick, D. K., Shaffer, W. O. & Loeser, J. D. 2009. Surgery for low
back pain: a review of the evidence for an American Pain Society Clinical Practice Guideline. Spine,
34, 1094–109.
Chou, R. & Huffman, L. H. 2007. Medications for acute and chronic low back pain: a review of the
evidence for an American Pain Society/American College of Physicians clinical practice guideline.
Ann Intern Med, 147, 505–14.
Cui, X., Trinh, K. & Wang, Y. J. 2010. Chinese herbal medicine for chronic neck pain due to cervical
degenerative disc disease. Cochrane Database Syst Rev, CD006556.
Dajani, E. Z. & Islam, K. 2008. Cardiovascular and gastrointestinal toxicity of selective cyclo-oxygenase-2
inhibitors in man. J Physiol Pharmacol, 59 Suppl 2, 117–33.
Davies, R. A., Maher, C. G. & Hancock, M. J. 2008. A systematic review of paracetamol for non-specific low
back pain. Eur Spine J, 17, 1423–30.
Deshpande, A., Furlan, A., Mailis-Gagnon, A., Atlas, S. & Turk, D. 2007. Opioids for chronic low-back
pain. Cochrane Database Syst Rev, CD004959.
Dowswell, T., Bedwell, C., Lavender, T. & Neilson, J. P. 2009. Transcutaneous electrical nerve stimulation
(TENS) for pain relief in labour. Cochrane Database Syst Rev, CD007214.
Dreiser, R. L., Marty, M., Ionescu, E., Gold, M. & Liu, J. H. 2003. Relief of acute low back pain with
diclofenac-K 12.5 mg tablets: a flexible dose, ibuprofen 200 mg and placebo-controlled clinical trial.
Int J Clin Pharmacol Ther, 41, 375–85.
Edgar, M. A. 2007. The nerve supply of the lumbar intervertebral disc. J Bone Joint Surg Br, 89, 1135–9.
Etropolski, M. S., Okamoto, A., Shapiro, D. Y. & Rauschkolb, C. 2010. Dose conversion between
tapentadol immediate and extended release for low back pain. Pain Physician, 13, 61–70.
Fishbain, D. A., Cole, B., Lewis, J., Rosomoff, H. L. & Rosomoff, R. S. 2008. What percentage of chronic
nonmalignant pain patients exposed to chronic opioid analgesic therapy develop abuse/addiction and/
or aberrant drug-related behaviors? A structured evidence-based review. Pain Med, 9, 444–59.
Fitzgerald, M. 2005. The development of nociceptive circuits. Nat Rev Neurosci, 6, 507–20.
Freemont, A. J., Peacock, T. E., Goupille, P., Hoyland, J. A., O’Brien, J. & Jayson, M. I. 1997. Nerve
ingrowth into diseased intervertebral disc in chronic back pain. Lancet, 350, 178–81.
Furlan, A. D., Imamura, M., Dryden, T. & Irvin, E. 2009. Massage for low back pain: an updated systematic
review within the framework of the Cochrane Back Review Group. Spine, 34, 1669–84.
Haroutiunian, S., Drennan, D. A. & Lipman, A. G. 2010. Topical NSAID therapy for musculoskeletal pain.
Pain Med, 11, 535–49.
Henschke, N., Kuijpers, T., Rubinstein, S. M., et al. 2010. Injection therapy and denervation procedures for
chronic low-back pain: a systematic review. Eur Spine J, 19, 1425–49.
Hinz, B. & Brune, K. 2008. Can drug removals involving cyclooxygenase-2 inhibitors be avoided? A plea for
human pharmacology. Trends Pharmacol Sci, 29, 391–7.
Jamison, R. N., Raymond, S. A., Slawsby, E. A., Nedeljkovic, S. S. & Katz, N. P. 1998. Opioid therapy for
chronic noncancer back pain. A randomized prospective study. Spine, 23, 2591–600.
Johnson, A. G. 1997. NSAIDs and increased blood pressure. What is the clinical significance? Drug Saf,
17, 277–89.
Jones, P. & Lamdin, R. 2010. Oral cyclo-oxygenase 2 inhibitors versus other oral analgesics for acute soft
tissue injury: systematic review and meta-analysis. Clin Drug Investig, 30, 419–37.
Katz, N., Rauck, R., Ahdieh, H., et al. 2007. A 12-week, randomized, placebo-controlled trial assessing the
safety and efficacy of oxymorphone extended release for opioid-naive patients with chronic low back
pain. Curr Med Res Opin, 23, 117–28.
Kobayashi, Y., Kurata, J., Sekiguchi, M., et al. 2009. Augmented cerebral activation by lumbar mechanical
stimulus in chronic low back pain patients: an FMRI study. Spine (Phila Pa 1976), 34, 2431–6.
Kroenke, K., Krebs, E. E. & Bair, M. J. 2009. Pharmacotherapy of chronic pain: a synthesis of
recommendations from systematic reviews. Gen Hosp Psychiatry, 31, 206–19.
Kuijpers, T., Van Middelkoop, M., Rubinstein, S. M., et al. 2010. A systematic review on the effectiveness
of pharmacological interventions for chronic non-specific low-back pain. Eur Spine J.
Li, C., Ni, J., Wang, Z., LI, M., Gasparic, M., Terhaag, B. & Uberall, M. A. 2008. Analgesic efficacy and
tolerability of flupirtine vs. tramadol in patients with subacute low back pain: a double-blind
multicentre trial. Curr Med Res Opin, 24, 3523–30.
Machado, L. A., Kamper, S. J., Herbert, R. D., Maher, C. G. & Mcauley, J. H. 2009. Analgesic effects of
treatments for non-specific low back pain: a meta-analysis of placebo-controlled randomized trials.
Rheumatology (Oxford), 48, 520–7.
Manchikanti, L., Singh, V., Falco, F. J., Cash, K. A. & Pampati, V. 2010. Evaluation of lumbar facet joint
nerve blocks in managing chronic low back pain: a randomized, double-blind, controlled trial with
a 2-year follow-up. Int J Med Sci, 7, 124–35.
Oesch, P., Kool, J., Hagen, K. B. & Bachmann, S. 2010. Effectiveness of exercise on work disability in
patients with non-acute non-specific low back pain: Systematic review and meta-analysis of
randomised controlled trials. J Rehabil Med, 42, 193–205.
Pallay, R. M., Seger, W., Adler, J. L., et al. 2004. Etoricoxib reduced pain and disability and improved
quality of life in patients with chronic low back pain: a 3 month, randomized, controlled trial. Scand
J Rheumatol, 33, 257–66.
Pareek, A., Chandurkar, N., Chandanwale, A. S., Ambade, R., Gupta, A. & Bartakke, G. 2009.
Aceclofenac-tizanidine in the treatment of acute low back pain: a double-blind, double-dummy,
randomized, multicentric, comparative study against aceclofenac alone. Eur Spine J, 18, 1836–42.
Pariente, A., Dartigues, J. F., Benichou, J., Letenneur, L., Moore, N. & Fourrier-Reglat, A. 2008.
Benzodiazepines and injurious falls in community dwelling elders. Drugs Aging, 25, 61–70.
Peloso, P. M., Fortin, L., Beaulieu, A., Kamin, M. & Rosenthal, N. 2004. Analgesic efficacy and safety of
tramadol/ acetaminophen combination tablets (Ultracet) in treatment of chronic low back pain: a
multicenter, outpatient, randomized, double blind, placebo controlled trial. J Rheumatol, 31, 2454–63.
Peterson, C. & Hodler, J. 2010. Evidence-based radiology (part 1): Is there sufficient research to support
the use of therapeutic injections for the spine and sacroiliac joints? Skeletal Radiol, 39, 5–9.
Rahme, E. & Bernatsky, S. 2010. NSAIDs and risk of lower gastrointestinal bleeding. Lancet, 376, 146–8.
Rahme, E. & Nedjar, H. 2007. Risks and benefits of COX-2 inhibitors vs non-selective NSAIDs: does their
cardiovascular risk exceed their gastrointestinal benefit? A retrospective cohort study. Rheumatology
(Oxford), 46, 435–8.
Roelofs, P. D., Deyo, R. A., Koes, B. W., Scholten, R. J. & Van Tulder, M. W. 2008a. Non-steroidal
anti-inflammatory drugs for low back pain. Cochrane Database Syst Rev, CD000396.
Roelofs, P. D., Deyo, R. A., Koes, B. W., Scholten, R. J. & Van Tulder, M. W. 2008b. Nonsteroidal
anti-inflammatory drugs for low back pain: an updated Cochrane review. Spine (Phila Pa 1976), 33,
1766–74.
Romano, C. L., Romano, D., Bonora, C. & Mineo, G. 2009. Pregabalin, celecoxib, and their combination
for treatment of chronic low-back pain. J Orthop Traumatol, 10, 185–91.
Rubinstein, S. M., Van Middelkoop, M., Kuijpers, T., et al. 2010. A systematic review on the effectiveness of
complementary and alternative medicine for chronic non-specific low-back pain. Eur Spine J, 19, 1213–28.
Ruoff, G. E., Rosenthal, N., Jordan, D., Karim, R. & Kamin, M. 2003. Tramadol/acetaminophen
combination tablets for the treatment of chronic lower back pain: a multicenter, randomized,
double-blind, placebo-controlled outpatient study. Clin Ther, 25, 1123–41.
Schnitzer, T. J., Gray, W. L., Paster, R. Z. & Kamin, M. 2000. Efficacy of tramadol in treatment of chronic
low back pain. J Rheumatol, 27, 772–8.
Skljarevski, V., Zhang, S., Desaiah, D., Alaka, K. J., Palacios, S., Miazgowski, T. & Patrick, K. 2010.
Duloxetine versus placebo in patients with chronic low back pain: a 12-week, fixed-dose, randomized,
double-blind trial. J Pain, 11(12), 1282–90.
Staal, J. B., De Bie, R., De Vet, H. C., Hildebrandt, J. & Nelemans, P. 2008. Injection therapy for subacute
and chronic low-back pain. Cochrane Database Syst Rev, CD001824.
Staal, J. B., De Bie, R. A., De Vet, H. C., Hildebrandt, J. & Nelemans, P. 2009. Injection therapy for
subacute and chronic low back pain: an updated Cochrane review. Spine (Phila Pa 1976), 34, 49–59.
Sudano, I., Flammer, A. J., Periat, D., et al. 2010. Acetaminophen increases blood pressure in patients with
coronary artery disease. Circulation, 122, 1789–96.
Takahashi, K., Aoki, Y. & Ohtori, S. 2008. Resolving discogenic pain. Eur Spine J, 17 Suppl 4, 428–31.
Urquhart, D. M., Hoving, J. L., Assendelft, W. W., Roland, M. & VAN Tulder, M. W. 2008.
Antidepressants for non-specific low back pain. Cochrane Database Syst Rev, CD001703.
Van Middelkoop, M., Rubinstein, S. M., Kuijpers, T., et al. 2010. A systematic review on the effectiveness
of physical and rehabilitation interventions for chronic non-specific low back pain. Eur Spine J.
Van Tulder, M. W., Koes, B. & Malmivaara, A. 2006a. Outcome of non-invasive treatment modalities on
back pain: an evidence-based review. Eur Spine J, 15 Suppl 1, S64–81.
Van Tulder, M. W., Koes, B., Seitsalo, S. & Malmivaara, A. 2006b. Outcome of invasive treatment
modalities on back pain and sciatica: an evidence-based review. Eur Spine J, 15 Suppl 1, S82–92.
Van Tulder, M. W., Touray, T., Furlan, A. D., Solway, S. & Bouter, L. M. 2003. Muscle relaxants for
non-specific low back pain. Cochrane Database Syst Rev, CD004252.
Wade, W. E. & Spruill, W. J. 2009. Tapentadol hydrochloride: a centrally acting oral analgesic. Clin Ther,
31, 2804–18.
Webster, L. R., Butera, P. G., Moran, L. V., Wu, N., Burns, L. H. & Friedmann, N. 2006. Oxytrex
minimizes physical dependence while providing effective analgesia: a randomized controlled trial in
low back pain. J Pain, 7, 937–46.
Wilkens, P., Scheel, I. B., Grundnes, O., Hellum, C. & Storheim, K. 2010. Effect of glucosamine on
pain-related disability in patients with chronic low back pain and degenerative lumbar osteoarthritis:
a randomized controlled trial. JAMA, 304, 45–52.
Williams, C. M., Latimer, J., Maher, C. G., et al. 2010. PACE—the first placebo controlled trial of
paracetamol for acute low back pain: design of a randomised controlled trial. BMC Musculoskelet
Disord, 11, 169.
Worz, R., Bolten, W., Heller, B., Krainick, J. U. & Pergande, G. 1996. [Flupirtine in comparison with
chlormezanone in chronic musculoskeletal back pain. Results of a multicenter randomized
double-blind study]. Fortschr Med, 114, 500–4.
Zerbini, C., Ozturk, Z. E., Grifka, J., Maini, M., Nilganuwong, S., Morales, R., Hupli, M., Shivaprakash, M.
& Giezek, H. 2005. Efficacy of etoricoxib 60 mg/day and diclofenac 150 mg/day in reduction of pain
and disability in patients with chronic low back pain: results of a 4-week, multinational, randomized,
double-blind study. Curr Med Res Opin, 21, 2037–49.
Part 8
Subgroup-Specific
Approaches for Patients
at Risk For or With
Chronic Pain
Chapter 26
Reviewing the Concept of Subgroups

in Subacute and Chronic Pain and the
Potential of Customizing Treatments
Adina C. Rusu, Katja Boersma, and Dennis C. Turk
26.1 The conundrum of pain

The tradition in medicine is to make a differential diagnosis of patients on the basis of the
presence of a set of signs and symptoms that meet the defining criteria for a specific disease and
that are inconsistent with alternative diagnoses that may have some similarities. Once a patient
receives a diagnosis a specific treatment or set of treatments will be prescribed preferably to cure
the disease or at least to alleviate some if not all of the symptoms. However, not all patients
exposed to the same pathogens or traumas meeting criteria for the same disease will present to the
healthcare provider in the same way if they seek treatment at all and they may not respond to the
same treatment in the same way. For example, only a small percentage exposed to an ankle frac-
ture will develop complex regional pain syndrome, similarly a minority of people exposed to
motor vehicle collision develop fibromyalgia but some do. Moreover, although a percentage of
people who have amputations will develop phantom limb pain many do not.
In many pain syndromes (e.g. fibromyalgia syndrome (FM), chronic headache, low back pain
(LBP), temporomandibular disorders (TMDs)), the definitive identification of causative agents is
rare and physical findings are only minimally related to patient-reported symptomatology (e.g.
pain severity). Even for pain syndromes with relatively clear-cut causative lesions or processes
(e.g. osteoarthritis, postherpetic neuralgia, chronic postsurgical pain), the severity of the insult
often bears only a modest relationship to the probability or intensity of pain report. For example,
the presence of a severe zoster rash, or a unilateral mastectomy, or a near-complete loss of carti-
lage in the knee joint may produce no pain at all in one individual and severe chronic pain in
another, indicting the presence of substantial individual differences in the processing of informa-
tion related to noxious stimulation. Even when they receive identical treatments for the same
diagnosis, responses are variable. In clinical trials, the emphasis tends to be on mean results with
much less attention given to the variability in responses. Why do not all individuals exposed to
the same insult demonstrate the same physical and psychological responses to the disease or
trauma and how can the variability in response to treatment be explained?
26.2 Contributions to individual differences

26.2.1 Genotype
The traditional biomedical approach, however, loses sight of the fact that patients have unique
genotypes and the phenotypic manifestation of an insult (i.e. disease or trauma). Mogil et al. (1999)
demonstrated marked variability in the responses of different inbred genetic strains of mice.
Since some of these strains were bred to have an absence of certain genes they implicate the
role of genetic factors in the variability. The demonstration of individual genetic variability of
rodents can reasonably be extrapolated to humans. Assume normal distribution in both pain
sensitivity (thresholds) and endogenous pain control the combination of both being lower
may place some individuals at risk (Edwards 2005). Human genetic studies suggest that single-
nucleotide polymorphisms of specific genes contribute significantly to basal pain sensitivity.
Single-nucleotide polymorphisms of the δ-opioid receptor gene (Kim et al. 2004) and the catechol-
O-methyltransferease (COMT) gene (Zubieta et al. 2003) are associated with pain ratings in
response to noxious stimuli and pain-induced μ-opioid receptor binding in the CNS (Zubieta
et al. 2003). Polymorphisms of μ-opioid receptor gene (OPRM1), specifically the single nucle-
otide polymorphisms 118 A>G (substitution) have been shown to influence the efficacy of intra-
venous morphine (Chou et al. 2006; Landau et al. 2008; for more information see also Chapter 3).
Thus, any two individuals with different polymorphisms of this gene may respond with wide
variability to the same treatment for the same traumatic event-type of surgery. Response to the
insult and response to treatment will be influenced by genetic characteristics.
26.2.2 Prior learning histories and psychological variations

In addition to individual uniqueness attributable to genotype, prior to symptom onset, patients
have extensive life experiences and learning histories that that shape their personalities and ways
of coping with trauma, injury, symptoms and that will influence their symptom presentation and
response to both the treatment and the insult. Pain intensity is influenced by a range of psycho-
social processes such as catastrophizing, which is associated with lower pain thresholds (Sullivan
et al. 2001) and enhanced CNS processing (Gracely et al. 2004).
There has been a growing awareness of and empirical support for the importance of psycho-
logical factors in reports of pain, suffering, and disability (e.g. Gatchel and Turk 1999). For exam-
ple, Flor and Turk (1988) found that physical factors in LBP and rheumatoid arthritis patients did
not predict pain severity, life interference, or physician visits; whereas cognitive appraisals of
helplessness and hopelessness were predictive of both self-report of pain impact and behavior in
response to pain. Similarly, studies provide evidence in support of the role of variables other than
medical diagnoses and identified physical pathology in treatment outcome (Jensen et al. 2001;
Burns et al. 2003). They found that perceptions of control over pain and decreased beliefs about
being disabled, and catastrophizing were associated with reductions in pain intensity, depression,
number of physician visits, and physical disability.
Over the past two decades there has been growing interest in the presence of individual differ-
ences in endogenous pain modulation. Psychological factors may interact with physiological and
genetic ones. People with chronic pain may demonstrate central augmentation (CA), key features
of which are increased vigilance to noxious stimuli, facilitation of such stimuli, and impaired
inhibition of them (Clauw 2010). CA can be manifested by reduced ability to inhibit noxious
stimuli (frequently assessed via conditioned pain modulation, CPM) and by facilitation of noxious
input (frequently assessed via temporal summation, TS).
CA has been evaluated in a variety of chronic pain populations, and although results have not
been entirely consistent (Leffler et al. 2002; Staud et al. 2003; Sutton et al. 2008); increased CA has
been identified in patients with various pain conditions, e.g. FM (Lautenbacher and Rollman
1997) and LBP (Peters and Schmidt 1992). It is not clear whether CA should be viewed as a cata-
lyst for the development of chronic pain or a result of chronic pain. However at least one study
demonstrated that impaired CA predicted postoperative pain severity and persistence up to
6 months among patients undergoing thoracotomy (Yarnitsky et al. 2008). Greater basal pain
sensitivity and decreased pain-inhibitory capacity may serve as a diathesis for the development of
chronic pain (Edwards et al. 2003).
CUSTOMIZING TREATMENTS 487
If CA is a risk factor for the development of chronic pain, what is the association between CA
and other risk factors? Since pain processing involves complex cerebral-spinal-peripheral connec-
tions, CA is likely to be associated with psychological risk factors. Although research on associa-
tions between indices of CA and psychological measures is in its infancy, there is some evidence
for an association between CA and anxiety (Granot et al. 2008) and catastrophizing (Edwards
et al. 2003; Edwards 2005; Granot et al. 2008; Goodin et al. 2009).
Studies of healthy, pain-free individuals reveal intact diffuse noxious inhibitory control (DNIC)
functioning: the application of one noxious stimulus inhibits the perception of pain produced by
the application of a second noxious stimulus (Edwards et al. 2003). However, studies of patients
with chronic pain conditions (e.g. FM, chronic back pain, irritable bowel syndrome), suggest that
they have either a diminished or non-existent DNIC response, and thus, experience increased
pain intensity (Staud et al. 2003; Wilder-Smith et al. 2004). DNIC has also been demonstrated to
predict who is likely to develop chronic pain following a specific trauma such as surgery (Yarnitsky
et al. 2008). Thus it appears that efficiency in the DNIC response may explain individual variation
in response to objectively comparable noxious stimulation and be a potential risk factor for
chronic pain.
26.2.3 Environmental contributions

To complicate things further, people with specific diseases live in social environments not in
isolation and responses from significant others contribute to the identified patient’s response to
symptoms, impact of symptoms the patients’ lives, treatments prescribed and initiated, and
response to treatments prescribed and self-initiated. Thus, the symptoms do not just occur in
isolated body parts or body systems but in a unique individual. The individual differences present
may explain the variation in how patients manifest, respond to, and adapt to their symptoms and
disease as well as their compliance with and response to treatment. These individual differences
are masked in the presentation of group averages in responses to treatment. Environmental influ-
ences exert profound effects on basal pain sensitivity with early exposure (Taddio et al. 1995;
Anand et al. 1999; Taddio et al. 2002) and trauma (Scarini et al. 1994) altering subsequent pain
sensitivity and placing individuals at risk for actue (Fillingim, Wilkinson, and Powell 1999) and
chronic pain (Aaron et al. 1997; Lampe et al. 2003).
Consider the role of contextual factors on pain perception and response. Back pain may have
been initiated by a trauma working in the garden or a trauma at work, a similar cause but a dif-
ferent context and with different meaning may lead to different responses such as how to treat,
whether to treat, from whom to seek treatment, and appropriate actions to take. The context in
which the symptoms began, at least to some extent, guide the response.
Now consider a man who awakens one morning with a severe headache. His first thought may
be to question why does he have this headache. It is possible that his interpretation of the cause
and the meaning of the symptoms may influence his level of emotional arousal and behaviours
related to treatment. If he recalls that his father, who died of a brain aneurism, described a severe
headache before his death and believes that his own headache is similar to that described by his
father, it is likely that his emotional arousal will greatly increase, potentially exacerbating the
headache, and he may rush to the emergency department at a local hospital. An alternative sce-
nario may lead to very different responses to the same symptom, headache. If instead of relating
the headache to his father’s death but rather relates it to having had too much alcohol to drink the
previous night, his level of emotional arousal might be significantly less elevated. He may decide
to take an over-the-counter analgesic and lie down for a while. Here the same symptom, severe
headache, was interpreted differently and resulted in different responses—levels of emotional
arousal and treatment behaviours. Thus, variations observed may be related to premorbid
individual differences. Response to treatments and adaptation to the disease are quite variable.
This variation is lost when mean differences are reported.
26.3 A diathesis–stress–environment conceptualization

Based on the brief overview, a number of explanations might account for the differences in evolu-
tion of diseases, maintenance, and exacerbation of symptoms, and response to the pathology and
to treatment as well. There is likely to be a normal distribution of sensory sensitivity, such that
individuals will vary in their detection thresholds, and perception of noxious stimuli, this distri-
bution is likely to be determined by individuals’ genotypes. People also have unique learning
histories that will also guide their beliefs, attitudes, and expectation regarding symptoms and
treatments. Treatments are filtered through individual lens comprised of individual, pathophysi-
ological, and environmental factors that result in differential experiences, responses, and beliefs
and expectations. Thus, these factors—genotype, phenotype, environment, and learning–interact
in such a way that considering them in isolation will provide an inadequate understanding. A more
comprehensive perspective that we might consider is a complex diathesis–stress–environment
model (Turk and Wilson 2010, see Figure 26.1). Individuals may be predisposed to the develop-
ment of pain based on genetics, prior learning history and culture (diatheses). Initiation of a dis-
ease, trauma, or noxious sensory input (stresses) is filtered through the lens of this diathesis
producing differential experiences, responses, and beliefs and expectations. The experience and
responses to the stresses occur in a social content (environment) which can both directly and indi-
rectly impact on diatheses and stresses altering the subsequent experience, beliefs and expectations.
This model takes a longitudinal rather than cross-sectional perspective. Regardless of the causes
of individual variations—genetics, prior learning history, environmental factors, one size most
Stress
• Disease
• Trauma
• Noxious sensory input
Diathesis
• Genes • Treatments
• Anatomy • Healthcare providers
• Neurophysiology
• Learning Hx
Environment
• Interpersonal support
• Economic resources
• Healthcare system
Time
Fig. 26.1 The diathesis–stress–environment model.
assuredly does not fit all when it comes to treating people with chronic pain even if they have the
same diagnosis.
There are a number of subacute and chronic pain syndromes that are diagnosed in a large
number of patients based on ill-defined criteria often without any clear understanding of the
underlying mechanisms. For example, it is estimated that 3–6 million American seek treatment for
FM each year (Wolfe et al. 1997). The current criteria for diagnosing FM consist of: (1) widespread
pain of at least 3 months’ duration and (2) pain present upon palpation of at least 11 of 18 specific
locations, hence tender points (Wolfe et al. 1990). The two most common types of chronic head-
ache are migraine and tension-type and for both the pathogenesis are unclear. Historically,
migraines were attributed to abnormal vascular activity and the mechanism for tension-type head-
aches was assumed to be muscle contraction. However, studies have demonstrated that the muscle
tension levels of people with migraine were actually higher than recorded in tension-type head-
aches leading some to suggest that these two types of headaches should be viewed along a contin-
uum of severity rather than be viewed as categorically different (Marcus 1992).
Subacute and chronic pain syndromes create a great deal of distress and disability; yet, despite
being prevalent, the aetiology and pathophysiology have remained elusive. Although no patho-
physiological factors have been identified, it is not for lack of trying. Various biological and psy-
chological factors have been investigated, but no definitive associations have been consistently
demonstrated. However, to date, these syndromes remain recalcitrant to treatment. In fact the
lack of understanding of the aetiology and maintaining factors has served as the impetus for the
diversity of interventions that have been used. What seems apparent is that the treatment of these
chronic pain syndromes has been based on the traditional model where the intervention is
matched to medical diagnoses even when crudely defined. Prescribing treatment based on diag-
noses is appropriate when there is a known aetiology but this approach has little basis when the
cause is unknown. When there is no consensus regarding what treatment should be prescribed
the choice becomes empirical, delivered on a trial-and-error basis depending on healthcare pro-
viders’ predilections and experience. The lack of universal success may explain why so many
treatments have been prescribed for these patients (Turk, 2002).
Lack of knowledge of the underlying mechanisms of the vague chronic pain syndromes has not
inhibited efforts to treat these patients with diverse modalities. There are a range of treatments for
subacute and chronic pain diagnosis. For example, there are over 50 pharmacological and more
than 50 additional non-pharmacological treatments for fibromyalgia syndrome (Turk 2004).
These treatments range from various classes of drugs (e.g. anticonvulsants, antidepressants,
opioids) to very diverse somatic and psychological therapy exercise, spa therapy, magnetized mat-
tresses, written emotional disclosure, and nerve blocks. Published studies of these disparate
treatment approaches reveal that on average 30–40% of patients obtain 20–50% improvement in
some symptoms (Turk 2004). This is a perplexing state of affairs, how could such a range of varied
treatments produce similar effects for patients with the same diagnosis?
26.4 Patient uniformity myth

The absence of empirical data linking pathophysiology or patients’ characteristics to successful
outcome has impeded the design of individual treatment plans. Consequently chronic pain
patients have tended to be treated as a homogeneous group based either on the medical diagnosis
or the vaguely defined—‘chronic pain syndrome’—for whom the same treatment is appropriate.
For example patients with TMDs have been treated with a variety of intraoral appliances,
biofeedback, oral surgery, and so forth, depending upon the conceptualization of the cause of the
symptoms and the practitioner’s training and experience (Turk et al. 1995). Chronic LBP patients
have been treated with steroid injections, physical therapy modalities, chiropractic manipula-
tions, and in multidisciplinary pain clinics with varying emphases ranging from relatively pure
programmes based on operant conditioning (Vlaeyen et al. 1996) to those with heavy emphasis
on physical conditioning (e.g. Mayer et al. 1987).
If the rationale for a treatment is that it addresses some underlying pathology—the ‘pain
generator’—then there should be some evidence implicating the source of the symptoms. For
example, the basis for prescribing non-steroidal anti-inflammatory drugs (NSAIDS) is that an
inflammatory process plays a causal role in the symptoms; consequently, these drugs should only
be prescribed in the presence of inflammation. If the rationale for treating tension-type headaches
with biofeedback is related to the presence of maladaptive muscular activity, then in would seem
that biofeedback should only be offered to patients who demonstrate some level of excessively
high muscular activity. If the use of operant conditioning treatment is predicated on alteration of
maladaptive environmental contingencies reinforcing pain behaviours, then some effort should
be made to offer this form of treatment only to those patients in whom such reinforcement con-
tingencies have been demonstrated (Fordyce 1976). Few attempts, however, have been made to
customize treatments to the mechanisms involved for specific patients. In short, we have adopted
the patient uniformity myth—all patients with the same diagnosis, no matter how vague, are
treated in a similar fashion as if they were a homogeneous group.
It is common to see patients with a wide range of diagnoses and locations of pain, not to men-
tion demographics and individual differences, treated within the same rehabilitation programme
with largely the same components (e.g. somatic, neuropathic, visceral, and psychogenic localized
in extremities, back, head, abdominal, thoracic and rectal; Becker et al. 2000). Many pain reha-
bilitation treatment outcome studies even fail to mention the composition of the patient sample
but simply refer to ‘consecutively referred patients’ or patients with ‘non-malignant pain’ or
‘musculoskeletal disorders’ (e.g. Becker et al. 1998; Cipher et al. 2001).
An implicit assumption of pain treatment facilities seems to be that providing multicomponent
treatments to all chronic pain patients is appropriate since some component(s) will prove effec-
tive, and perhaps different ones for patients with different characteristics. Thus, a pain centre may
include physical exercise, education in body mechanics, assertion training, systematic medication
reduction, stress management, biofeedback, family counselling, the use of transcutaneous nerve
stimulation, nerve blocks, and so forth. This is like treating a set of patients with a range of vita-
min deficiencies with multivitamins with the hope that each patient will derive benefits from the
appropriate vitamin contained within the collection comprising the multivitamin regimen rather
than prescribing specific vitamin therapy based on individual patient needs.
Recent data suggest that greater attention should be given to identifying the characteristics of
patients who improve and those who fail to improve when treated with the same intervention.
Treatment should be prescribed only for those patients who are likely to derive significant benefit.
The ability to predict improvement should be more efficient producing improved outcomes and
reduction in costs. Furthermore, the ability to identify characteristics of patients who do not
benefit from specific treatment should facilitate the development of treatment approaches tailored
to the needs of those patients.
There has been a growing recognition that pain is a complex perceptual experience influenced
by a wide range of psychosocial factors, including emotions, social and environmental context,
sociocultural background, beliefs, attitudes, and expectations, as well as biological factors. Pain
that persists for months and years, chronic pain, will influence all aspects of a person’s function-
ing: emotional, interpersonal, avocational, and physical. Consequently, successfully treating
chronic pain patients requires attention not only to the organic basic of the symptom but also
to a range of factors that modulate nociception and moderate the pain experience and related
disability.
It would be impossible to review the research on all of the attempts to identifying individual
differences in patients with persistent along with there responses to different sets of treatments.
We have narrowed the focus of the remainder of this chapter to several areas that we feel are
representative of the developments over the past decade and that we believe have promise for
the future evolution of understanding pain sufferers and treatment processes. In accordance with
the aims of this volume, we focus on chronic non-cancer pain in adults, with a special interest on
back pain.
Despite major advances in medical and surgical treatment, the incidence of low back pain dis-
ability continues to escalate (Waddell 1998). Several recent studies support the efficacy of reha-
bilitation programmes that incorporate psychological interventions for chronic back pain. For
instance, in a study designed to evaluate psychological treatments in the prevention of chronicity
in patients with sciatic pain, Hasenbring et al. (1999) compared two psychological treatments
(biofeedback vs. cognitive-behavioural therapy, CBT) for patients at risk for chronicity with usual
care and treatment refusers. One of the best predictors of disability was refusal of treatment.
These results lend support to the notion that efficacy will be enhanced if cognitive-behavioural
treatments are initiated earlier in the course of treatment and scheduled in accordance with
individual psychological patterns of coping (Turk 1990a).
26.5 Identification of patient subgroups

Chronic pain syndromes are made up of heterogeneous groups of people even if they have the
identical medical diagnosis. A common pitfall in clinical research and practice is the assumption
of patient homogeneity. Inclusion of a diverse group of patients into the same category simply
because they present a common set of symptoms may result in inconsistent research results.
Several attempts have been made to identify groups of patients who differ on important variables
and to evaluate differential treatment responses (e.g. Turk et al. 1996; Thieme et al. 2006; Verra
et al. 2009). A significant problem with rehabilitation-oriented treatments is that: (1) a large
number of patients refuse treatments offered; (2) a large number of patients drop out of treat-
ment; and (3) the rates of relapse are relatively high (Richmond and Carmody 1999; Turk and
Rudy 1990a).
A number of studies have focused on empirically identifying patient subgroups based on psy-
chological characteristics and psychopathology. Numerous attempts to identify subgroups have
relied on the use of traditional psychological measures such as the Minnesota Multiphasic
Personality Inventory (MMPI) and the Symptom Checklist-90R (Swimmer et al. 1992; Hutten
et al. 2001). Others have tried to identify subgroups of patients based on behavioural communi-
cations of pain—‘pain behaviours’ (Keefe et al. 1990), physical performance (Hutten et al. 2001),
patient perceived aetiology (Waylonis and Perkins 1994), and psychophysiological response pro-
files (Thieme et al. 2006). Up to this point, the attempts to identify subgroups of patients with
chronic pain reviewed have focused on single factors. Outcomes, however, are likely to be deter-
mined by the interactive effects of multiple factors, as single factors may not account for a statisti-
cally significant or clinically meaningful proportion of the variance in outcome. Several studies
that included measures of both physical and psychological functioning have reported interactive
effects of biopsychosocial factors on the outcomes (Mayer et al. 1987). Some authors have pro-
posed that multidimensional rather than unidimensional classifications should be attempted for
a problem as complex as chronic pain (Turk and Rudy 1988; Dworkin and LeResche 1992).
26.6 Empirically-derived classification of patients with pain

A number of studies have identified subgroups of patients according to psychosocial and
behavioural characteristics (e.g. Turk and Rudy 1988; 1990a,b; Turk et al. 1998a; Johansson and
Lindberg 2000). Several studies found that patients classified into different subgroups on the basis
of their psychosocial and behavioural responses responded differently to identical treatments
(Dahlstrom et al. 1997; Thieme et al. 2006; Verra et al. 2009). Subgroups of chronic pain patients
characterized by a number of psychosocial and behavioural characteristics seem to be fairly con-
sistently observed across different pain syndromes (e.g. LBP, FM, TMD, headaches, cancer, spinal
cord injuries, whiplash associated disorders; Turk et al. 1998b; Turk and Rudy 1990b; Soderlund
and Denison 2006; Widerström-Noga et al. 2007; see Table 26.1) suggesting the independence of
psychosocial factors from the physical pathology. Distinctiveness of the psychosocial profiling
implies that patients in different subgroups may exhibit differential responses to treatment, which
has been demonstrated in several outcome studies.
Turk and Rudy (1988) performed a cluster analysis on patients’ responses to the West Haven-
Yale Multidimensional Pain Inventory (MPI, Kerns et al. 1985), a frequently used measure in the
chronic pain literature (Piotrowski 2007) that has been shown to be reliable, valid, and sensitive
to change (Lousberg et al. 1999; Bergström et al. 2001). The MPI consists of a set of empirically-
derived scales designed to assess the perceptions of: (1) pain severity; (2) interference with family
and marital functioning, work, and social and recreational activities; (3) support received from
significant others; (4) life control incorporating perceived ability to solve problems and feelings
of personal mastery and competence; (5) affective distress; and (6) performance of activities.
Turk and Rudy (1988) identified three relatively homogenous groups and labelled one subgroup
characterized by severe pain, compromised life activities and enjoyment, reduced sense of control,
and high level of emotional distress as ‘dysfunctional’ (DYS). Another subgroup, also marked with
relatively high degrees of pain and affective distress but further characterized by low levels of per-
ceived support from significant others, was labelled ‘interpersonally distressed’ (ID). The third
group consisted of chronic pain patients who appeared to be coping relatively well despite their
long-standing pain. This group, which experienced low levels of pain, functional limitations, and
emotional distress, was labelled ‘adaptive copers’ (AC). Reliable, external scales supported the
uniqueness of each of the three subgroups of patients. The subgroups have been replicated in sev-
eral studies conducted in Sweden (Bergström et al. 2001, Hellström and Jansson 2001), Finland
(Talo et al. 2001), the Netherlands (Lousberg et al. 1997), and Germany (Flor et al. 1990) and using
different measures of the constructs assessed by the MPI (e.g. Jamison et al. 1994).
26.7 Subgrouping on the basis of the MPI

To ascertain whether the three MPI subgroups differ only in the areas in which they are supposed
(discriminant validity), a series of analyses were performed on physical functioning, pain severity,
depressive mood, and perceived functional limitation (Turk and Rudy 1988). Interestingly, the
Table 26.1 Distributions of the MPI profiles in various disorders
Profiles MPI FMS1 LB2 HA2 TMD2 Regional/local Cancer3 Metastatic Cancer3
(91)* (200) (245) (200) (47) (137)
DYS 26% 62% 44% 46% 44% 64%
ID 39% 18% 26% 22% 18% 5%
AC 35% 20% 30% 32% 35% 18%
* numbers in parentheses = sample size
1 Turk et al. (1996);
2 Turk & Rudy (1990);
3 Turk et al. (1998)
three subgroups did not differ in physical functioning (e.g. lumbar flexion, straight leg raise, and
cervical range of motion). Similarly, Bergstrom et al. (2001) found that the subgroups did not
differ on spinal and neck mobility, impaired reflexes, or Lasègue sign. These results reinforce the
suggestion that psychosocial dimension of chronic pain syndromes may be independent of
physical pathology. The independence of the psychologically based classification from medical
diagnosis is particularly important, as it would be plausible to assume that patients with greater
physical pathology have more to cope with and thus are more likely to be dysfunctional, whereas
those who have less pathology would have less to cope with and therefore would more likely be
adaptive copers. The abovementioned studies refute this possibility and support the notion that
medical and psychosocial classifications are independent.
Using external measures of relevant constructs, Turk and Rudy (1988) found that: (1) the DYS
patients reported significantly higher levels of pain than AC patients; (2) the DYS and ID groups
reported significantly higher levels of depressed mood and perceived disability; and (3) the ID
patients rated their interpersonal relationships with significant others to be significantly lower in
quality compared with the DYS and AC patients. These data provide convergent validity for the
three subgroups.
Comparing the taxonomic structure and profiles of distinct chronic pain groups can help to
establish the generalizability of the MPI taxonomy. The three MPI subgroups have been repli-
cated with a number of different pain populations. Turk and Rudy (1990b) directly compared the
power and stability of the MPI taxonomy by studying samples of patients with chronic low back
pain (CLBP), headache (HA), and temporomandibular disorders (TMD). The scale interrelation-
ships for the three diagnostic groups were equivalent. All three syndrome groups were repre-
sented in each of the subgroups (see Table 26.1). Thus, it is possible that patients with CLBP, HA,
and TMD who are classified within the same subgroup may be psychologically more similar to
each other than patients with the same diagnosis but who are classified within different subgroups
(see Table 26.1).
Similarly in another study, Turk et al. (1998) examined the covariance structures among the
MPI scales (profile patterns) of three very different patient groups: chronic pain, FMS, and cancer
patients. They discovered that the covariance structures were remarkably consistent among the
three MPI profiles. Based on the results of these studies it appears that CP, FMS and cancer
patients who are classified within the same MPI profile may be more similar to each other in
subjective experience and response to symptoms than patients with the same diagnosis but who
are classified to a different MPI profile (see Figure 26.2). These results suggest that although
different physical diagnostic groups may require common biomechanical treatment targeting the
pathophysiological mechanisms underlying each disorder, they may also benefit from specific
psychosocial interventions tailored to their psychosocial-behavioural characteristics based on the
MPI profiles.
Studies comparing the MPI subgroups have yielded evidence supporting differential response
to the same intervention (e.g. Dahlstrom et al. 1997; Epker and Gatchel 2000; Rudy et al. 1995;
Verra et al. 2009). For instance, when a rehabilitation pain management programme for FMS
patients was tested, the DYS group improved in most areas, whereas the ID patients, who reported
levels of pain and disability comparable to the DYS group, failed to respond to the treatment
(Turk et al. 1998). There was little change in the AC patients, owing possibly to a floor effect.
Strategier et al. (1997) reported similar results with a sample of patients with low back pain. Both
these studies found that the AC group improvements were intermediate between the DYS and ID.
The results further support the need for different treatments targeting characteristics of sub-
groups and suggest that psychosocial characteristics of patients with chronic pain are important
predictors of treatment responses and may be used to customize treatment. For example, DYS
DYS profile FMS

Cancer
CP
6
Mean scores 4
0
PS I LC AD S NR SR DR GA
MPI scale
ID profile FMS
Cancer
6 CP
Mean scores
0
MPI scales
AC profile FMS
Cancer
CP
6
Mean scores
0
MPI scales
Note. DYS = dysfunctional; ID = interpersonally distressed;

AC = adaptive coper.
FMS = fibromyalgia syndrome; CP = chronic pain.
PS = pain severity; I = interference; LC = life control;
AD = affective distress; S = support;
NR = negative responses; SR = solicitous responses;
DR = distracting responses; GA = general activity
Fig. 26.2 Mean MPI scale scores in three groups of patients with pain by MPI profile.
patients are characterized by high levels of emotional distress, feelings of little control, and high
levels of inactivity and may warrant treatment with the focus on cognitive factors associated with
their maladaptive beliefs (e.g. cognitive restructuring), whereas the ID patients may require addi-
tional treatment components addressing clinical needs specific to this group (e.g. interpersonal
skills) and some components of the standard interdisciplinary treatment may not be essential for
the AC patients.
Patient attrition is a major problem in pain rehabilitation programmes. It is possible that
patients who receive a treatment that does not match with their specific needs (patterns of coping
and adapting) would be more likely to terminate treatment. In one small-scale treatment study,
Carmody (2001) observed that ID (47%) and DYS (33%) patients were significantly more likely
to drop-out of group rehabilitation treatments than the AC group (11%). Data such as these
reinforce the idea that treatments that are prescribed need to be congruent with patient charac-
teristics. If patients are not satisfied with the treatment they are receiving and do not believe that
the treatment components are addressing their particular concerns, especially when rehabilita-
tion treatments can be very demanding, they will not be committed to the treatment. Clinicians
will need to spend time developing tailored interventions to match patient needs and expecta-
tions. Moreover, they will need to expend extra effort motivating patients by educating them
regarding the relevance of the components of the treatments being offered (see also Chapter 24).
The results described above implicate the importance of matching patients’ characteristics to
treatment. By specifically addressing patients’ psychological needs, clinicians are likely to be able
to enhance both the cost-effectiveness and the clinical effectiveness of interventions. Additionally,
early identification of patients’ modes of adapting to subacute pain may lead to the development
of interventions that can prevent chronicity and long-term disability (Gatchel and Epker 1999;
Johansson and Lindberg 2000).
Research by Vlaeyen et al. (1995) has emphasized the prominent role played by fear of injury
and pain in patients with chronic pain. The presence of high levels of fear avoidance might serve
as a target for treatment. The presence of other patient characteristics (e.g. endurance responses,
catastrophizing, negative affectivity) might also serve as foci of treatment. The essential point is
that they can target important patient differences. Psychosocial subgroups on the basis of pain-
related fear and endurance will be reviewed in the following.
26.8 Subgrouping on the basis of pain-related fear

and emotional distress
Although numerous studies underscore the importance of psychological factors in the develop-
ment and perpetuation of pain problems, most do not explicitly propose any mechanisms of how
these variables might interact. For example, a large range of cross-sectional, experimental, and
prospective studies show that pain-related fear (e.g. catastrophizing, fear of movement) and more
general emotional distress (e.g. depressed mood, anxiety) are related to pain and disability (Pincus
et al. 2002; Leeuw et al. 2007). Yet, we have only just begun to understand how these and other
important variables may interrelate within individuals as well as how these variables relate sequen-
tially across time (Boersma and Linton 2006; Wideman et al. 2009). Understanding how risk
factors interact within individuals is important because these variables might operate differently
for different people, an assumption that is not being made within traditional correlational designs
(Von Eye and Bergman 2003). For example, Clyde and Williams (2002) suggest in an overview of
depression in chronic pain, that there is a need to take into account that there may be several
pathways by which pain and depression co-occur, possibly resulting in different subtypes. They
suggest that there is a need to take into account both vulnerability to depression, and the nature
of the experience of pain as a stressor. In line with this, there could, for example, be a subgroup
that has a general vulnerability for negative cognitive processing while, for another subgroup,
depressive symptoms could be a mere consequence of the prolonged exposure to pain and of its
interference with valued life goals. Related to this second possibility is the role that depressed
mood has been assigned within the fear avoidance model (Vlaeyen et al. 1995). Here, depression
is largely seen as a consequence of disability and subsequent lack of positive reinforcement, even
though it is incorporated in the model that general vulnerability factors may influence and fuel
pain catastrophizing. Even though the continuous development of theoretical models that
propose interrelationships between risk factors is pivotal for furthering our understanding, there
is also a need for empirical studies that focus on the interrelationships between, for example,
depressive mood and pain-related fear within individuals and across time.
Studies that focus on discovering subpopulations could give important information of possible
differential pathways by which pain, depression and pain-related fear co-occur in the process
towards prolonged disability. In a line of studies investigating the interrelationship between pain-
related fear variables and emotional distress, we consistently found that pain-related fear, and
emotional distress can appear in isolation within individuals as well as they can appear in combi-
nation. Individuals with a combination of pain-related fear and emotional distress show by far
the highest levels of disability, and these psychological patterns appear to be relatively stable over
time (Boersma and Linton 2005, 2006a; Westman et al. 2011). For example, we used primary care
patient’s ratings of pain intensity, fear of movement, function, and depressed mood to distinguish
possible patterns (Boersma and Linton 2005). We found four distinct profiles: a group of patients
characterized by high pain, fear of movement and dysfunction, a group that was similar but in
addition suffered from low mood, a group that was mostly characterized by low mood only, and
a group that was characterized by relatively low scores on all variables. These four subgroups had
clearly distinct outcomes at follow up with the subgroups characterized by fear of movement
alone and fear of movement plus low mood developing the highest levels of sick leave. To repli-
cate and extend the findings of this study, we performed another study where the aim was to
identify possible distinctive patterns based only on psychological process variables (fear of move-
ment, catastrophizing and depressed mood) and to test whether these patterns are in fact related
to the development of disability (Boersma and Linton 2006a). In this study five meaningful sub-
groups were extracted that display differential profiles on measures of disability cross sectional as
well as prospective.
These clusters, graphically displayed in Figure 26.3, largely replicated the results from the previ-
ous study (Boersma and Linton 2005) and give support to the fear avoidance model as there
are strong relations between fear of movement and catastrophizing. However, as can be seen,
the exact pattern in which these factors are organized varies between subgroups. Further, fear of
movement and catastrophizing can be (cluster 2) but are not necessarily (cluster 1) accompanied
by signs of depressed mood. Signs of depressed mood on the other hand, do not have to be accom-
panied by fear of movement and catastrophizing (cluster 4). In summary, in both these studies
profiles emerged that were characterized by pain-related fear with- and without depressed mood,
by low pain-related fear and no depressed mood (‘low risk’) and by depressed mood alone.
A recent study replicated and extended these findings by following a sample of 110 primary care
patients with spinal pain over a period of 3 years (Westman et al. submitted). Three separate sub-
grouping analyses were performed at baseline, 1-year follow-up and 3-year follow-up. The results
replicated the existence of subgroups characterized by emotional distress only, pain-related fear
only, and emotional distress in combination with pain-related fear. Moreover, the results showed
that these risk profiles displayed significant stability across the 3-year period, and showed increas-
ingly distinct patterns of disability. In fact, individuals with an initial profile of pain-related fear
and distress were three times more likely to remain fearful and distressed across the 3-year time
period. Furthermore, the results showed that individuals displaying one of the three risk profiles
at baseline had significantly worse recovery on all outcome variables (pain, function and sick
leave) at 1 year, as well as 3-year follow-up, compared to individuals not displaying one of these
risk profiles.
The results of these studies are in line with previous research that focuses on classifying pain
patients in homogeneous subgroups based on their adjustment to pain, such as the studies using
the MPI (Kerns et al. 1985; Turk and Rudy 1988; Bergström et al. 2001). As mentioned above,
1.5
0.5 tsk
z score
pcs
0 had-dep
Cluster 1 Cluster 2 Cluster 3 Cluster 4 Cluster 5
–0.5 (10%) (13%) (30%) (24%) (23%)
–1
–1.5
Fig. 26.3 Subgroups based on fear of movement (TSK), catastrophizing (PCS), and depression
(HADS-depression scale).
these studies report on three different subgroups: DYS, ID, and AC. When comparing the level of
disability, the DYS group and the AC group from the MPI show some similarities with respec-
tively the pain-related fear with and without depression and the ‘low score profile’ groups. The
studies that have investigated pain-related fear within the three MPI subgroups support this idea;
in these studies the DYS group had significantly more pain-related fear than the AC group
(Asmundson et al. 1997; McCracken et al. 1999). Thus it appears that results from these studies
converge on one another, and that where classifications derived from the MPI are made on the
basis of maladjustment on a broad range of variables (pain-related, functional as well as psycho-
social) the studies discussed above employ a more narrow theoretical framework with a sole
psychological focus.
In summary, these studies elucidate the variations in the variable relationships between sub-
groups. These variations may have considerable implications for intervention, as the unfavoura-
ble outcome in relation to function and sick-leave for the risk profile groups suggests that the
received healthcare is not helping. It is possible that the important underlying psychological proc-
esses are not dealt with in a sufficient way. An approach that would include a focus on pain-
related fear on the one hand and more general emotional distress on the other hand could greatly
improve patient functioning and enhance the secondary prevention of chronic disability includ-
ing sick leave. Specific as well as more general cognitive-behavioural interventions have proven to
be of use in rehabilitation as well as in preventive intervention (Linton and Andersson 2000;
Sullivan and Stanish 2003; van den Hout et al. 2003; Woods and Asmundson 2008). However,
since CBT is a collection of a wide range of techniques, knowledge on the existence of subgroups
may further inform which components within CBT are most effective for whom. For example, it
is possible that individuals characterized by mood disturbances that is not merely a consequence
of pain-related fear, may not be helped by an intervention merely focused on, for example, in vivo
exposure to feared and avoided movements. Indeed, even though exposure based treatments have
shown effect for individuals with high levels of fear of movement, they also show that some
patients respond better to exposure than others (Leeuw et al. 2008; Linton et al. 2008; Woods and
Asmundson, 2008).
In a recent study, we analysed the individual differences between those with successful and
non-successful outcome of an exposure based intervention for high pain-related fear (Flink et al.
2010). The results showed that individuals with a non-successful outcome where characterized
by high levels of catastrophizing. Possibly, the exposure treatment was hindered by, or merely
did not sufficiently address, pain related distress and could be complemented with other mood
regulation techniques to improve outcome.
26.9 Subgrouping on the basis of endurance-responses

and emotional distress
A number of prospective studies have shown that ongoing pain and disability are mediated by
psychosocial factors (Linton 2000; Shaw et al. 2001; Pincus et al. 2002). As mentioned above,
depression and fear-avoidance variables seem to be the most important psychosocial risk factors
of chronic pain and disability (Fritz et al. 2001; Boersma and Linton 2006). However, researchers
have noted limitations in the application of the fear-avoidance model to subgroups of patients
where pain-related disability is associated with high task performance or overuse (Vlaeyen and
Morley 2004). In particular, there is evidence for the role of endurance-related responses as pre-
dictors of chronic pain in patients (Hasenbring 1993; Hasenbring et al. 1994; Grebner et al. 1999).
The avoidance-endurance model (AEM; described in more detail in Chapter 15) provides testable
hypotheses about subgroups of patients with endurance-related responses and will therefore be
briefly summarized in the following. The avoidance-endurance model of pain postulates opposite
pathways into chronic pain: one via pain-related fear-avoidance responses (FAR), and at least
another one other via endurance-related responses (ER). FAR including misinterpreting pain as
a sign of serious injury, increasing fear, hypervigilance and avoidance behaviour may lead to
physical deconditioning and disability (Smeets et al. 2007). ER behaviour despite severe pain is
suggested to lead to chronic pain by physical overload of physical structures (Hasenbring et al.
2006; Hasenbring et al. 2009). Thought suppression (TS) represents a cognitive ER where patients
suppress the perception of pain or the interruption of daily activities normally demanded by pain
(e.g. Eccleston and Crombez 1999). A rebound effect was suggested which leads to more fre-
quently occurring pain-related thoughts accompanied by the perception of failure and increased
depression.
In order to identify pain patients at risk for developing chronic pain and disability at an early
stage and to develop individually targeted treatments, it is important to identify specific sub-
groups of patients with homogenous and highly rigid responses pattern to pain. In a first study in
chronic back pain patients, we investigated whether ER play a significant role in the phenomenol-
ogy of the MPI subgroups (Rusu and Hasenbring 2008). Although it is assumed that the DYS
group is associated with dysfunctional coping (Turk and Rudy 1988), only a few studies have
investigated the relationship between DYS group and FAR. These studies have consistently shown
that the DYS group reported significantly more catastrophizing, pain-related anxiety, avoidance,
less control, and positive future thinking (Geisser et al. 1994; Asmundson et al. 1997; Hellström
et al. 2000). However, little is known about the interrelation of the DYS group, ER and pain com-
munication. It can be suggested that identification of maladaptive pain-related coping strategies
strongly associated with MPI subgroups is important for their modification in treatment.
Investigating the interrelationships between the MPI subgroups with regard to pain-related FAR,
ER and pain communication, we found that, even after control for pain intensity and depression,
patients classified as DYS reported more anxiety/depression, help-/hopelessness, and catastro-
phizing than did those classified as AC (Rusu and Hasenbring 2008). Furthermore, the DYS
group showed more thought suppression compared to AC; however, subgroups did not differ
significantly with regard to avoidance of social and physical activity, and endurance behaviour. In
line with our expectations, DYS as well as ID patients showed more non-verbal pain behaviour
compared to AC, which refers to the special role of operant conditioning.
4
DYS
3 ID
AC
2
0
PS IN AD SS LC PR SR DR
Fig. 26.4 MPI-D scale scores for the DYS, ID, and AC profile (Rusu and Hasenbring, 2008).
PS = Pain severity; IN = Interference; AD = Affective distress; SS = Social support; LC = Life control;
PR = Punishing responses; SR = Solicitous responses; and DR = Distracting responses. This figure
has been reproduced with permission of the International Association for the Study of Pain®
(IASP®). The figure may not be reproduced for any other purpose without permission.
The results provided further verification of the ability of the MPI classification system (Turk
and Rudy 1988; Rudy 2004) to successfully identify subgroups of patients. Only 11.6% of patients
were not classified into primary subgroups. However, the distribution of profiles differed from
previous studies. That is, studies of chronic pain patients consistently showed that the DYS pro-
file described the largest proportion of patients (mean across study samples = 39%), the AC
profile described approximately 38% of patients, and the ID profile described the smallest pro-
portion of patients (24%). However, we found that the AC profile classified 61.7% and the DYS
profile only 15.8% of patients in our sample. These differences may be related to varying study
designs and methods, for instance, we collected data from general practices with a number of pre-
determined selection criteria. Conversely, the majority of previous studies, in which DYS patients
represented the largest group, took place in secondary care settings (Jamison et al. 1994 ;
McCracken et al. 1999).
As indicated in Figure 26.4, DYS patients reported higher levels of pain severity and pain-re-
lated interference than the other two subgroups. In comparison to AC patients, DYS patients had
higher levels of affective distress and perceived less ability to control their lives. The ID patients
were distinguishable from the other two patient profiles by their marked differences on the social
support scale and significant other response scales. These patients perceived their family mem-
bers not to be very supportive of them and their pain problem.
A more focused examination revealed that the MPI subgroups were associated with distinct
pattern of FAR and ER. The DYS group displayed more affective distress as well as more maladap-
tive pain-related cognitions; however, we found no differences in avoidance of physical and
avoidance of social activities, in our sample. Again, this might be related to varying measures. For
instance, the Kiel Pain Inventory (KPI; Hasenbring 1994) avoidance scales consider more general
every-day behaviours related to avoidance behaviour, whereas measures like the Pain Anxiety
Symptoms Scale (PASS; McCracken et al. 1992) or the Tampa Scale of Kinesiophobia (TSK;
Vlaeyen et al. 1995 ; Clark et al. 1996 ) assess more specific escape/avoidance behaviours.
Consequently, patients in our sample may have endorsed the KPI avoidance scales more,
irrespective of the subgroup characteristics, due to the larger behavioural units considered by the
KPI. As mentioned above, DYS patients exhibited more thought suppression, which is in line
with recent experimental studies showing that thought suppression produced shorter tolerance
time, more pain, and distress in a cold pressor test compared to acceptance based thoughts
(Masedo and Esteve 2007). The results suggest that the application of techniques specifically
geared toward the reduction of automatic and frequent thought suppression may be an impor-
tant component in the treatment of DYS patients. Evidence from a randomized controlled
intervention study for patients with acute sciatic pain has shown that a risk factor-based
cognitive-behavioural intervention (RCBT; more information can be found in Chapter 27),
which was aimed to specifically target dysfunctional endurance responses, was more effective
compared to a standardized treatment (Hasenbring et al 1999). Alternatively, it might be useful
to apply acceptance-based approaches to the treatment of the DYS group to reduce thought sup-
pression. Both approaches might help patients to realize that their natural response to escape or
avoid by suppressing the pain experience cognitively has been ineffective and may increase the
urge for strategies of pain control, which undermines the understanding that pain reduction in
the long-term is in most cases not achievable (Hasenbring et al. 1999; McCracken et al. 1999;
Nicholas et al. 2003). Acceptance and RCBT might lead patients from pain reduction attempts
(thought suppression as one of them) to acceptance of pain for the foreseeable future and satisfy-
ing lives despite the pain (McCracken et al. 1999; McCracken, 2005). Future research will
be necessary to evaluate the additional contribution of these specific treatment possibilities to
standardized protocols.
Contrary to our expectations, the MPI group differences in endurance behaviour were found
to be insignificant. The AEM delineates the role of endurance responses, and particularly endur-
ance behaviour plays in the transition from acute to chronic pain. We know from prospective
studies of patients with acute sciatic pain that endurance behaviour and positive mood in spite of
severe pain are important risk factors for the development of chronic pain (Hasenbring et al.
1994). However, the generalizability of these results in acute and subacute patients to populations
of chronic pain patients is potentially questionable due to different conditions (acute and suba-
cute versus persistent pain and primary care versus secondary care) applied. Seeing that the AC
group displayed significantly higher scores on positive mood compared to both other subgroups
could lead to the assumption that endurance behaviour in conjunction with positive mood may
be helpful for adjustment to chronic pain. New treatment and assessment approaches emphasize
the positive association between positive mood, acceptance, and adjustment (McCracken and
Eccleston 2003; Pincus et al. 2004). However, further replication studies are needed to investigate
the contribution of thought suppression and endurance behaviour within the MPI subgroups and
to test the usefulness of the avoidance-endurance model in chronic pain populations.
In our previous studies we reported evidence for the prospective validity of AEM-based
subgroups for back pain, assessed by a short screening instrument including depression, thought
suppression and behavioural endurance (Risk Screening for Back Pain, RISC-BP; Hasenbring
1993; Grebner et al. 1999; Hallner and Hasenbring 2004). Besides a FAR subgroup, which
showed high fear and avoidance, patients of the ER subgroup responded with thought suppres-
sion, depression and endurance behaviour (distress-endurance pattern DER), while a second
ER subgroup showed endurance behaviour, minimization and positive mood in spite of pain
(eustress-endurance pattern EER). A fourth subgroup was found (adaptive responses to pain AR)
which was low on depression, thought suppression and behavioural endurance.
Hasenbring (1993) investigated in a prospective study these four subgroups and found that the
DER as well as the EER subgroup reported higher pain intensity at the 6-month follow-up, less
return to work and a higher rate of application for early retirement compared to the adaptive
group. Consistent results were also reported by Grebner and colleagues (1999) in 82 patients
undergoing lumbar disc surgery, showing that EER and FAR subgroups had a higher rate of early
retirement at the 6-month follow-up compared to the adaptive group, and against their expecta-
tions also compared to the DER group.
In a recent study replicating and extending these findings, 162 primary care patients with suba-
cute non-specific back pain were investigated at the start of medical treatment and at a 6-months
follow-up (Hasenbring et al. submitted). Back pain patients were classified into the four AEM-
based subgroups using RISC-BP and pre-defined cut-off points. Results revealed that both ER
subgroups showed higher pain intensity at the 6-month follow-up compared to the adaptive
subgroup. In contrast, subgroup analyses revealed that the adaptive group showed prospectively
less pain chronicity compared to all other groups and less disability compared to FAR and DER
subgroups. Preliminary support for the hypothesis that ER patients reveal a higher level of physi-
cal activity and a higher rate of static strain postures indicated by physical overload was found in
a cross-sectional study measuring overt physical activity using accelerometry in patients 6 months
after lumber disc surgery (Hasenbring et al. 2006). Additionally, Chapter 15 gives a comprehen-
sive overview of further studies investigating AEM-based subgroups using different methodo-
logical approaches.
The purpose of another recent study in chronic back pain patients was to further test the
content validity of the AEM subgroups and to elucidate the variations in the relationships between
AEM-based subgroups and MPI scales; fear-avoidance and endurance responses (Rusu et al.
submitted). Looking back at the MPI scales, the results revealed that the adaptive group scored
lowest on the MPI scales pain severity, interference, and affective distress, while the FAR group
showed by far the highest scores on these scales. DER patients reported significantly more pain
intensity and disability (as assessed by the Graded Chronic Pain Scale; Von Korff et al. 1992)
compared to the EER group and the adaptive group. When an independent measure of depres-
sion was used to assess depressive symptomatology (Beck Depression Inventory; Beck et al. 1961),
as expected both subgroups with a depressive component, DER and FAR patients had signifi-
cantly higher depression scores compared to the EER group, and DER patients scored signifi-
cantly higher compared to the adaptive group. Multivariate analyses of variance further indicated
that groups differed significantly for pain-related fear-avoidance and endurance responses, even
after control of pain intensity and depression.
With regard to fear-avoidance responses, FAR and DER patients revealed significantly higher
scores on the majority of fear-avoidance associated scales (e.g. helplessness, hopelessness, cata-
strophizing, and avoiding of social and physical activity) compared to the other two subgroups.
EER patients differed significantly with regard to endurance variables (positive mood, thought
suppression, minimizing thoughts, and behavioural endurance) compared to FAR patients even
after control for depression and pain intensity. DER patients differed as well from FAR patients
with regard to more thought suppression, minimization and behavioural endurance.
In summary, the studies reviewed in this section have advanced one way of assessing psycho-
logical subgroups based on the AEM and it clearly requires replication. In conclusion, the results
of the studies reported indicate that AEM-based subgroups are useful for the identification of
fear-avoidance as well as endurance-related responses to pain, underline the notion that the
differentiation between the two endurance-related subgroups is pivotal and that new therapeutic
strategies are strongly needed for patients with endurance-related coping, especially for those
with an EER pattern. Future prospective studies should investigate the AEM subgroups in the
prediction of long-term adjustment to pain using different outcomes, which are independent
from the KPI and may better elucidate the conditions under which endurance-related responses
may become adaptive or maladaptive. Finding further evidence for differences in psychosocial
subgroups, whether classified by the MPI or other clinically relevant psychological measures, may
improve treatment evaluation as well as the process of patient-therapy matching.
26. 10 Future directions and conclusion

We have attempted to provide an overview of areas that contribute to our understanding of the
person with chronic pain and psychological variables that have implications for improvements in
treatment. We noted that clinical outcomes tend to support addressing cognitive, affective, and
behavioural contributors to the experience of and response to chronic pain. We now consider
future directions beyond what we have already suggested.
26.10.1 Dual-diagnostic approach

Several authors have proposed the use of a dual-diagnostic approach, whereby two diagnoses are
assigned concurrently, a biomedical and a psychosocial diagnosis (e.g. Turk 1990; Dworkin and
LeResche 1992; Scharff et al. 1995). Treatment could then be directed toward both simultane-
ously. A patient diagnosed with LBP who was also classified on the MPI as DYS might require a
different set of treatments than a patient with back pain who was classified on the MPI as being
ID. Both patients with back pain would benefit from physical therapy, but the former might
require attention to their perceptions of low self-efficacy and control as well as beliefs about being
disabled, whereas the latter might benefit from interpersonal problem solving (Jensen et al. 1994;
Jensen et al. 2001). On the other hand, two patients diagnosed with different medical syndromes,
for example, FMS and TMD both of whom were classified as ID, might have very different physi-
cal treatments but might benefit from a similar psychological intervention (e.g. interpersonal
problem solving, assertion training, and social support). Adopting a dual-diagnostic approach
would serve the valuable function of encouraging clinicians as well as researchers to think concur-
rently in terms of these two different biomedical and psychosocial-behavioural but complemen-
tary diagnostic systems. Empirical studies would be needed to evaluate the efficacy of
individualized treatments based on the dual-diagnostic approach.
26.10.2 Customizing treatments to patients’ needs

Multiple studies, using a broad range of assessment instruments, have identified subgroups of
patients with chronic pain. To date, there have been none that have customized treatment to the
psychosocial and behavioural characteristics of patients with pain and particularly none of back
pain patients. The studies that have been conducted have used retrospective methods rather than
looking prospectively at differential response to the same treatment (e.g. Moore et al. 1986).
A handful of studies, described above, support the potential value of treatment matching, although
the results are not uniformly consistent. There is a critical need to make use of these results in
designing treatments and evaluating their differential efficacy. Clinical investigations should be
conducted to determine the relative utility of different treatment modalities based on the match
of treatment to patient characteristics and classification to predict which patients are most likely
to benefit from what combination of therapeutic modalities. The field might be advanced by
asking a set of questions, namely, ‘What treatment, by whom, is most effective for this individual,
with that specific problem, under which set of circumstances?’ (Paul 1967).
Treatments might be provided in a modular fashion, where there are separate components that
are woven together into an overall treatment regimen based on individual patient characteristics.
All patients with back pain may require physical therapy and psychological support and encour-
agement; however, as indicated by the results reported above, some may need additional modules
such as treatment of depression (Ingram et al. 1990) or the treatment of rigid endurance behav-
iour (Hasenbring et al. 1999). The combination of physical treatments, pharmacological treat-
ments, and specific psychological interventions may provide unique and complementary
advantages in addressing different symptoms.
It is important to acknowledge that the identification of subgroups, regardless of the methods
used, does not mean that the resulting classification will incorporate all features of patients.
Subgroups should be viewed as prototypes with significant room for individual variability. Thus,
treatments matched to subgroup characteristics will also need to consider and address unique
characteristics of the individual patient. It is noteworthy to mention that the transparent items
contained in the MPI may facilitate denial of problems by patients classified as AC. Patients in the
subgroups should be examined more carefully before assuming that they are truly adapting in a
specific manner (e.g. Rusu and Hasenbring 2008). The MPI might be viewed as a screening instru-
ment, and based on patient subgroup, the clinician would follow-up with additional assessment
measures and procedures. The subgroup customization fits somewhere between the exclusively
idiographic approach evaluated by single case treatment designs and the generic ‘pain patient
homogeneity’ approach that has characterized much of the pain treatment outcome studies.
The transition from acute to chronic pain:

26.10.3
secondary prevention
The vast majority of people who are injured recover in a reasonable amount of time and do not
develop chronic disorders. Similarly, a significant number of people who develop chronic disor-
ders associated with pain do not become physically and emotionally disabled. As noted previ-
ously, a number of efforts have been made to identify the predictors of disability among these
groups (e.g. Hasenbring et al. 1994; Linton and Hallden 1997; Gatchel and Epker 1999; Johansson
and Lindberg 2000 Riipinen et al. 2005; Jellema et al. 2007). There are, however, few longitudinal
studies, and replications are the exception rather than the rule (Turner et al. 2000).
As noted earlier, a number of studies have begun to identify predictors of disability for injured
workers with acute pain states and also predictors of response to treatment. Solely, identifying
predictors is insufficient. The next step is to determine whether knowledge of predictors can
guide treatment design and the development of strategies for improving outcome. Prevention
and earlier interventions hold promise for reduction in the extent of disability (for more informa-
tion see Chapters 11 and 30). The average person treated at a multidisciplinary pain centre has
had symptoms of pain for an average of 7 years (Flor et al. 1992). By this time, patients have
become so disabled that rehabilitation becomes a very difficult task, the outcomes, although rea-
sonably good, could have been improved if implemented at an earlier stage. Linton and Bradley
(1996) have reviewed a number of studies that may be viewed as secondary prevention, namely,
treatments used with those who have already had a first pain episode and designed to prevent
long-term disability. Given the natural history of many musculoskeletal pain disorders, it is
important that early interventions are reasonably inexpensive. Providing expensive interventions
for groups with high rates of natural recovery is inefficient and costly (Von Korff 1999). Efforts to
prevent, or at least minimize, the consequences of exacerbations and relapse may be particularly
cost-effective. The effects of early interventions are currently mixed (Jellema et al. 2005; Van der
Windt et al. 2008), but the costs of chronicity are so extreme that research in this area is definitely
warranted.
26.10.4 Overgeneralization based on pain clinic samples

Minimal attention has been given to those people who recover spontaneously or who make
adequate and often exceptional accommodations to their conditions regardless of physical
impairments and limitations. Much of what is known about chronic pain syndromes is based on
people who seek treatment. These individuals are not a representative group, therefore it is likely
that people will seek treatment when they have an exacerbation of their symptoms (e.g. Bradley
et al. 1994). As a result, almost any treatment will appear reasonably successful. It is essential
that research extends beyond the clinical population to community samples that are not seek-
ing care.
26.10.5 Chronic pain should be treated like other chronic diseases

Even when successful, pain rehabilitation does not cure pain but emphasizes self-control and self-
management of symptoms. Examination of the results of long-term opioid therapy reveals
significant residual pain; similarly, the majority of people who have surgery or have spinal cord
stimulators or drug delivery systems implanted continue to report substantial pain (Turk and
Okifuji 1998). Because these pain sufferers are not cured, they require regular care and follow-up.
We have tended to view chronic pain more as an acute condition that will resolve following treat-
ment, but this is patently not the case. If we view chronic pain as a lifelong disease, then, as with
other chronic diseases such as diabetes, we should expect treatment to be ongoing, requiring
regular check-up and continuing care. From this perspective, treatment is not over after a few
sessions or a 3–4 weeks rehabilitation programme. Instead, we should expect and plan for the
need to include booster sessions (Lanes et al. 1995; Bendix et al. 1998). We need to give attention
to the development and evaluation of treatments of pain as a chronic disease that is not cured but
requires maintenance-enhancement strategies.
In conclusion, we noted the importance of considering a range of factors that contribute to
individual differences, and posed a diathesis–stress–environment heuristic model. We focused on
psychosocial factors to illustrate the importance of understanding meaningful groups of patients.
Similar analyses focusing on genetics, pathophysiology, and environmental variables are also
relevant and a need to be considered when developing a treatment plan. At this point, whether
treatment tailoring will produce greater effects than providing completely idiographic or generic
treatments can only be viewed as a reasonable hypothesis. The fact that a significant proportion
of patients with chronic pain are not successfully treated by current general approaches and the
identification of various subgroups of patients makes investigation of treatment matching of
particular importance.
References
Aaron, L.A., Bradley, L.A., Alarcon, G.S. et al. (1997). Perceived physical and emotional trauma as
precipitating events in fibromyalgia. Associations with health care seeking and disability status but not
pain severity. Arthritis and Rheumatism, 40, 453–60.
Anand K.S., Coskun V., Thrivikraman K.V., Nemeroff C.B., Plotsky P.M. (1999). Long-term behavioral
effects of repetitive pain inneonatal rat pups. Physiology and Behavior, 66, 627–37.
Asmundson, G.J., Norton, G.R. and Allerdings, M.D. (1997). Fear and avoidance in dysfunctional chronic
back pain patients. Pain, 69, 231–6.
Beck, A.T., Ward, C.H., Mendelson, M., Mock, J. and Erbaugh, J. (1961). An inventory for measuring
Becker, N., Hojsted, J., Sjogren, P. and Eriksen J. (1998). Sociodemographic predictors of treatment
outcome in chronic non-malignant pain patients. Do patients receiving or applying for disability
pension benefit from multidisciplinary pain treatment? Pain, 77, 279–87.
Becker, N., Sjogren, P., Bech, P. et al. (2000). Treatment outcome of chronic non-malignant pain patients
managed in a Danish multidisciplinary pain center compared to general practice: A randomised
controlled trial. Pain, 84, 203–11.
Bendix, A.F., Bendix, T., Haestrup, C. and Busch, E. (1998). A prospective, randomized 5-year follow-up
study of functional restoration in chronic low back pain patients. European Spine Journal, 7, 111–19.
Bergström, G., Bodin, L., Jensen, I.B., Linton, S.J. and Nygren, A. L. (2001). Long-term, non-specific spinal
pain: reliable and valid subgroups of patients. Behaviour Research and Therapy, 39, 75–87.
Bradley, L.A., Alarcon, F.S., Triana, M. et al. (1994). Health care seeking behavior in fibromyalgia:
Association with pain thresholds, symptom severity, and psychiatric morbidity. Journal of
Musculoskeletal Pain, 2, 79–87.
Boersma, K. and Linton, S.J. (2005). Screening to identify patients at risk: profiles of psychological risk
factors for early intervention. Clinical Journal of Pain, 21, 38–43.
Boersma, K. and Linton, S.J. (2006a). Psychological processes underlying the development of a chronic
pain problem: a prospective study of the relationship between profiles of psychological variables in the
fear-avoidance model and disability. Clinical Journal of Pain, 22, 160–6.
Boersma, K. and Linton, S.J. (2006b). Expectancy, fear and pain in the prediction of chronic pain and
disability: A prospective analysis. European Journal of Pain, 10, 551–7.
Burns, J.W., Glenn, B., Bruehl S, Harden, R.N. and Lofland, K. (2003). Cognitive factors influence
outcome following multidisciplinary chronic pain treatment: a replication and extension of a
cross-lagged panel analysis. Behaviour Research and Therapy, 41, 1163–82.
Carmody, T.P. (2001). Psychosocial subgroups, coping, and chronic low-back pain. Journal of Clinical
Psychology in Medical Settings, 8, 137–48.
Chou, W.Y., Wang, C.H., Liu, P.H., Liu, C.C., Tseng, C.C. and Jawan, B. (2006). Humanopioid receptor
A118G polymorphism affects intravenous patient-controlled analgesia consumption after abdominal
hysterectomy. Anesthesiology, 5, 334–37.
Cipher, D.J., Fernandez, E. and Clifford, P.A. (2001). Cost-effectiveness and health care utilization in a
multidisciplinary pain center: Comparison of three treatment groups. Journal of Clinical Psychology in
Clark, M.E., Kori, S.H. and Brockel, J. (1996). Kinesiophobia and chronic pain: psychometric
characteristics and factor analysis of the Tampa Scale. American Pain Society Abstracts, 77.
Clauw, D. (2010). Fibromyalgia. In Ballantyne JC, Rathmell JP and Fishman SM. (Eds). Bonica’s
Management of Pain, 4th ed. Philadelphia, PA: Lippincott Williams & Wilkins.
Clyde, Z. and Williams, A. (2002). Depression and mood. In S. J. Linton (Ed.) New avenues for the
prevention of chronic musculoskeletal pain and disability (Vol. 12, pp. 105–21). Amsterdam: Elsevier.
Dahlstrom, L., Widmark, G. and Carlsson, S. G. (1997). Cognitive-behavioral profiles among different
categories of orofacial pain patients: diagnostic and treatment implications. European Journal of Oral
Science, 105, 377–83.
Dworkin, S.F. and LeResche, L. (1992). Research diagnostic criteria for temporomandibular disorders:
review, criteria, examination and specifications, critique. Journal of Craniomandibular Disorders, 6,
301–55.
Eccleston, C. and Crombez, G. (1999). Pain demands attention: a cognitive-affective model of the
Edwards, R.R. (2005). Individual differences in endogenous pain modulation as a risk factor for chronic
pain. Neurology, 65, 437–43.
Edwards, R.R., Fillingim, R.B. and Ness, T.J. (2003). Age-related differences in endogenous pain
modulation: a comparision of diffuse noxious inhibitory controls in healthy older and younger adults.
Pain, 101, 155–65.
Edwards, R.R., Ness, T.J., Weigent, D.A. et al. (2003). Individual differences in diffuse noxious inhibitory
controls (DNIC): association with clinical variables. Pain, 106, 427–37.
Epker, J. and Gatchel, R.J. (2000). Coping profile differences in the biopsychosocial functioning of patients
with temporomandibular disorders. Psychosomatic Medicine, 62, 69–75.
Fillingim, R.B., Wilkinson, C.S. and Powell T. (1999). Self-reported abuse history and pain complaints
among young adults. Clinical Journal of Pain, 15, 85–91.
Flink, I., Boersma, K. and Linton, S. J. (2010). Catastrophizing moderates the effect of exposure in vivo for
back pain patients with pain-related fear. European Journal of Pain 14(8), 887–92.
Flor, H., Rudy, T.E. and Birbaumer, N. (1990). Zur Anwendbarkeit des West Haven-Yale
Multidimensional Pain Inventory im deutschen Sprachraum. Schmerz, 4, 82–87.
Flor, H., Fydrich, T. and Turk, D.C. (1992). Efficacy of multidisciplinary pain treatment centers:
A meta-analytic review. Pain, 49, 221–30.
Flor, H. and Turk DC. (1988). Chronic back pain and rheumatoid arthritis: predicting pain and disability
from cognitive variables. Journal of Behavioral Medicine, 11, 251–65.
Fordyce W. E. (1976). Behavioral Methods for Chronic Pain and Illness. St. Louis, MO: Mosby.
Fritz, J.M., George, S.Z. and Delitto, A. (2001). The role of fear-avoidance beliefs in acute low back pain:
Gatchel, R.J. and Epker, J. (1999). Psychosocial predictors of chronic pain and response to treatment. In
R.J. Gatchel and D.C. Turk (Eds.) Psychosocial factors in pain: Critical perspectives, pp. 412–34.
Gatchel, R.J. and Turk, D.C., Eds. (1999). Psychosocial Factors in Pain: Critical Perspectives. New York:
Guilford Press.
Geisser, M.E., Robinson, M.E. and Henson, C.D. (1994). The Coping Strategies Questionnaire and chronic
pain adjustment: a conceptual and empirical reanalysis. Clinical Journal of Pain, 10, 98–106.
Goodin, B.R., McGuire L., Allshouse M. et al. (2009). Associations between catastrophizing and
endogenous pain-inhibitory processes: sex differences. Journal of Pain, 10, 180–90.
Gracely, R.H., Geisser, M.E., Giesecke, T. et al. (2004). Pain catastrophizing and neural responses to pain
among persons with fibromyalgia. Brain, 127, 835–43.
Granot, M., Weissman-Fogel, I., Crispel, Y. et al. (2008). Determinants of endogenous analgesia magnitude
in a diffuse noxious inhibitory control (DNIC) paradigm: do conditioning stimulus painfulness,
gender and personality variables matter? Pain, 136, 142–9.
Grebner, M., Breme, K., Rothoerl, R., Woertgen, C., Hartmann, A and Thomé, C. (1999). Coping and
convalescence course after lumbar disk operations. Der Schmerz, 13, 19–30.
Hallner, D. and Hasenbring, M. (2004). Classification of psychosocial risk factors (yellow flags) for the
development of chronic low back and leg pain using artificial neural network. Neuroscience Letters,
361, 151–4.
Hasenbring, M. (1993). Endurance strategies–a neglected phenomenon in the research and therapy of
Hasenbring, M. (1994). The Kiel Pain Inventory–Manual. Three questionnaire scales for assessment of
pain-related cognitions, emotions and coping strategies. Bern: Huber.
Hasenbring, M.I., Hallner, D. and Rusu, A.C. (2009). Fear-avoidance- and endurance-related responses to
Hasenbring, M., Marienfeld, G., Kuhlendahl, D. and Soyka, D. (1994). Risk factors of chronicity in lumbar
Hasenbring, M., Ulrich, H. W., Hartmann, M. and Soyka, D. (1999). The efficacy of a risk factor-based
cognitive behavioral intervention and electromyographic biofeedback in patients with acute sciatic
pain. An attempt to prevent chronicity. Spine, 24, 2525–35.
Hasenbring, M. I., Plaas, H., Fischbein, B. and Willburger, R. (2006). The relationship between activity and
Hasenbring, M. I., Hallner, D., Klasen, B., Streitlein-Böhme, I., Willburger, R. and Rusche, H.
(submitted). Pain-related fear-avoidance and endurance responses versus response pattern in the
prediction of future pain and disability: validation of the Risc Screening for Back Pain (RISC-BP).
Hellström, C. and Jansson, B. (2001). Psychological distress and adaptation to chronic pain:
symptomatology in dysfunctional, interpersonally distressed and adaptive copers. Journal of
Musculoskeletal Pain, 9, 51–67.
Hellström, C., Jansson, B. and Carlsson, S.G. (2000). Perceived future in chronic pain: the relationship
between outlook on future and empirically derived psychological patient profiles. European Journal of
Pain, 4, 283–90.
Hutten, M.M.R., Hermens, H.J. and Zilvold, G. (2001). Differences in treatment outcome between
subgroups of patients with chronic low back pain using lumbar dynamometry and psychological
aspects. Clinical Rehabilitation, 15, 479–88.
Ingram, R.E., Atkinson, J.H., Slater, M.A., Saccuzzo, D.P. and Garfin, S.R. (1990). Negative and positive
cognition in depressed and non-depressed chronic pain patients. Health Psychology, 9, 300–14.
Jamison, R.N., Rudy, T.E., Penzien, D.B. and Mosley, T.H., Jr. (1994). Cognitive-behavioral classifications
of chronic pain: Replication and extension of empirically derived patient profiles. Pain, 57, 277–92.
Jellema, P., van der Windt, D.A.W.M., van der Horst, H.E., Blankenstein, A., Bouter, L.M. and
Stalman, W.A.B. (2005). Why is a treatment aimed at psychosocial factors not effective in patients with
(sub)acute low back pain? Pain, 118, 350–59.
Jellema, P. van der Windt, D.A.W.M., van der Horst, H.E., Stalman, W.A.B., and Bouter, L.M. (2007).
Prediction of an unfavorurable course of low back pain in general practice: Comparison of four
instruments. British Journal of General Practice, 57, 15–22.
Jensen, M.P., Turner, J.A. and Romano, J.M. (1994). Correlates of improvements in multidisciplinary
treatment of chronic pain. Journal of Consulting and Clinical Psychology, 62, 172–79.
Jensen, M.P., Turner, J.A. and Romano, J.M. (2001). Changes in beliefs, catastrophizing, and coping are
associated with improvements in multidisciplinary pain treatment. Journal of Consulting and Clinical
Johansson, E. and Lindberg, P. (2000). Low back pain patients in primary care: Subgroups based on the
Multidimensional Pain Inventory. International Journal of Behavioral Medicine, 7, 340–52.
Keefe, F.J., Bradley, L.A. and Crisson, J.E. (1990). Behavioral assessment of low back pain: identification of
pain behavior subgroups. Pain, 40, 153–60.
Kerns, R.D., Turk, D.C. and Rudy, T.E. (1985). The West Haven-Yale Multidimensional Pain Inventory
(WHYMPI). Pain, 23, 345–56.
Kim, H., Neubert, J.K., San Miguel, A. et al. (2004). Genetic influence on variability in human acute
experimental pain sensitivity associated with gender, ethnicity and psychological temperament. Pain,
109, 488–96.
Lampe, A., Doering, S., Rumpold, G. et al. (2003). Chronic pain syndromes and their relationship to
childhood abuse and stressful live events. Journal of Psychosomatic Research, 54, 361–7.
Landau, R., Kern, C., Columb, M.O., Smiley, R.M. and Blouin, J.L. (2008). Genetic variability of the
μ-opioid receptor influences intrathecal fentanyl analgesia reqluirements in laboring women. Pain,
139, 5–14.
Lanes, T.C., Gauron, E.F., Spratt, K.F., Wernimont, T.J., Found, E.M. and Weinstein, J.N. (1995).
Long-term follow-up of patients with chronic back pain treated in a multidisciplinary rehabilitation
program. Spine, 20, 801–6.
Lautenbacher, S. and Rollman, G.B. (1997). Possible deficiencies of pain modulation in fibromyalgia.
Leeuw, M., Goossens, M.E.J.B., Linton, S.J., Crombez, G., Boersma, K. and Vlaeyen, J.W. (2007). The
fear-avoidance model of musculoskeletal pain: current state of scientific evidence Journal of Behavioral
Leeuw, M., Goossens, M.E.J.B., van Breukelen, G.J. et al. (2008). Exposure in vivo versus operant graded
activity in chronic low back pain patients: results of a randomized controlled trial. Pain, 138(1), 192–207.
Leffler, A.S., Hansson, P. and Kosek, E. (2002). Somatosensory perception in a remote pain-free area and
function of diffuse noxious inhibitory controls (DNIC) in patients suffering from long-term trapezius
myalgia. European Journal of Pain, 6, 149–59.
Linton, S.J. (2000). A review of psychological risk factors in back and neck pain. Spine, 25 (9), 1148–56.
Linton, S.J. and Bradley, L.A. (1996). Strategies for the prevention of chronic pain. In R.J. Gatchel and D.C.
Turk (Eds.), Psychological approaches to pain management: A practitioner’s handbook, pp. 458–285.
Linton, S.J. and Hallden, K. (1997). Risk factors and the natural course of acute and recurrent
musculoskeletal pain: Developing a screening instrument. In T.S. Jensen, J.A. Turner, and
Z. Wiesenfeld-Hallin (Eds.) Proceedings of the Eighth World Congress on Pain, pp. 527–36. Seattle, WA:
IASP Press.
Linton, S. J. and Andersson, T. (2000). Can chronic disability be prevented? A randomized trial of a
cognitive-behavior intervention and two forms of information for patients with spinal pain. Spine,
25, 2825–31.
Linton, S. J., Boersma, K., Jansson, M., Overmeer, T., Lindblom, K. and Vlaeyen, J. W. (2008).
A randomized controlled trial of exposure in vivo for patients with spinal pain reporting fear of
work-related activities. European Journal of Pain, 12(6), 722–30.
Lousberg, R., Schmidt, A.J., Groenman, N.H. et al. (1997). Validating the MPI-DLV using experience
sampling data. J Behav Med, 20, 195–206.
Lousberg, R., Van Breukelen, G.JP., Schmidt, A.J.M. et al. (1999). Psychometric properties of the
Multidimensional Pain Inventory, Dutch Language version (MPI-DLV). Behavior Research and
Therapy, 7, 167–82.
Marcus, D.A. (1992). Migraine and tension-type headaches: the questionable validity of current
classification systems. Clinical Journal of Pain, 8, 28–36.
Masedo, A. and Esteve, M.R. (2007). Effects of suppression, acceptance and spontaneous coping on pain
tolerance, pain intensity and distress. Behavior Research and Therapy, 45, 199–209.
Mayer, T.G., Gatchel, R.J., Mayer, H. et al. (1987). A prospective two-year study of functional restauration
in industrial low back injury. JAMA, 258, 1763–7.
McCracken, L.M. (2005). Contextual cognitive-behavioral therapy for chronic pain. Progress in pain research
and management, Vol. 33. Seattle, WA: IASP Press.
McCracken, L.M. and Eccleston, C. (2003). Coping or acceptance: what to do about chronic pain? Pain,
105, 197–204.
McCracken, L.M., Zayfert, C. and Gross, R.T. (1992). The pain anxiety symptoms scale: Development and
validation of a scale to measure fear of pain. Pain, 50, 67–73.
McCracken, L.M., Spertus, I.L., Janeck, A.S., Sinclair, D. and Wetzel, F.T. (1999). Behavioral dimensions
of adjustment in persons with chronic pain: pain-related anxiety and acceptance. Pain, 80, 283–9.
Mogil, J.S., Wilson, S.G., Bon, K. et al. (1999). Heritability of nociception I: responses of 11 inbred mouse
strains on 12 measures of nociception. Pain, 80, 67–82.
Moore, J.F., Armentrout, D.P., Parker, J.C. and Kivlahan, D. R. (1986). Empirically derived pain-patients
MMPI subgroups: prediction of treatment outcome. J Behav Med, 9, 51–63.
Nicholas, M., Molloy, A., Tonkin, L. and Beeston, L. (2003). Manage your pain: practical and positive ways
of adapting to chronic pain. London: Souvenir Press.
Paul, G.L. (1967). Outcome research in psychotherapy. Journal of Consulting and Clinical Psychology, 31,
109–18.
Peters, M.L. and Schmidt, A.J. (1992). Differences in pain perception and sensory discrimination between
chronic low back pain patients and healthy controls. Journal of Psychosomatic Research, 36, 47–53.
Pincus, T., Burton, K. Vogel, S. and Field, A. (2002). A systematic review of psychological factors as
predictors of chronicity/disability in prospective cohorts of low back pain. Spine, 27, 109–20.
Pincus, T., Williams, A., Vogel, S. and Field, A. (2004). The development and testing of the depression,
anxiety and positive outlook scale (DAPOS). Pain, 109, 181–8.
Piotrowski, C. (2007). Review of the psychological literature on assessment instruments used with pain
patients. North American Journal of Psychology, 9, 303–6.
Richmond, R.L. and Carmody, T.P. (1999). Dropout from treatment for chronic low back pain.
Professional Psychology, 30, 51–5.
Riipinen, M., Niemisto, L., Lindgren, K.-A., and Hurri, H. (2005). Psychosocial differences as predictors
for recovery from chronic low back pain following manipulation, stabilizing exercises and physician
consultation or physician consultation alone. Journal of Rehabilitation Medicine, 37, 152–8.
Rudy, T.E. (2004). Multiaxial assessment of pain: Multidimensional pain Inventory computer program user’s
manual. Version 3.0. Pittsburgh, PA. University of Pittsburgh.
Rudy, T.E., Turk, D.C., Kubinski, J.A. and Zaki, H.S. (1995). Differential treatment responses of TMD
patients as a function of psychological characteristics. Pain, 61, 103–12.
Rusu, A.C. and Hasenbring, M. (2008). Multidimensional Pain Inventory derived classifications of chronic
pain: Evidence for maladaptive pain-related coping within the dysfunctional group. Pain, 134, 80–90.
Rusu, A.C., Scholich, S.L., Harenbrock, S. and Hasenbring, M. (submitted). Psychological factors
discriminate clinical groups of chronic back pain patients: The role of endurance and fear-avoidance
responses for distress and disability.
Scarinci, I.C., McDonald Haile, J., Bradley, L.A. and Richter, J.E. (1994). Altered pain perception and
psychosocial features among women with gastrointestinal disorders and history of abuse: a preliminary
model. American Journal of Medicine, 97, 108–18.
Scharff, L., Turk, D.C. and Marcus, D.A. (1995). Psychosocial and behavioral characteristics in chronic
headache patients: support for a continuum and dual-diagnostic approach. Cephalagia, 15, 216–23.
Shaw, W.S, Pransky, G. and Fitzgerald, T.E. (2001). Early prognosis for low back disability: Interventions
strategies for health care providers. Disability and Rehabilitation, 23, 815–28.
Smeets, R.J.E.M. and Wittink, H. (2007). The deconditioning paradigm for chronic low back pain
unmasked? Pain, 130, 201–2.
Soderlund, A. and Denison, E. (2006). Classification of patients with whiplash associated disorders (WAD):
Reliable and valid subgroups based on the Multidimensional Pain Inventory (MPI-S). European
Staud, R., Robinson, M.E., Vierck, C.J. Jr. et al. (2003). Diffuse noxious inhibitory controls (DNIC)
attenuate temporal summation of second pain in normal males but not in normal females or
fibromyalgia patients. Pain, 101, 167–74.
Strategier, L.D., Chwalisz, K., Altmaier, E.M. et al. (1997). Multidimensional assessment of chronic low
back pain: predicting treatment outcomes. Journal of Clinical Psychology in Medical Settings, 4, 91–110.
Sullivan, M.J.L. and Stanish, W.D. (2003). Psychologically based occupational rehabilitation: the
pain-disability prevention program. Clinical Journal of Pain, 19, 97–104.
Sullivan, M.J., Thorn, B., Haythornthwaite, J.A. et al. (2001). Theoretical perspectives on the relation
between catastrophizing and pain. Clinical Journal of Pain, 17, 52–64.
Sutton, J.T., Bachmann, G.A., Arnold, L.D. et al. (2008). Assessment of vulvodynia symptoms in a sample
of U.S. women: a follow-up national incidence survey. Journal of Women’s Health, 17, 1285–92.
Swimmer, G.I., Robinson, M.E. and Geisser, M.E. (1992). Relationship of MMPI cluster types, pain coping
strategy, and treatment outcome. Clinical Journal of Pain, 8, 131–7.
Taddio, A., Goldbach, M., Ipp, M., Stevens, B. and Koren G. (1995). Effect of neonatal circumcision on
pain response during vaccination in boys. Lancet, 345, 291–2.
Taddio, A., Shah, V., Gilbert-MacLeod, C. and Katz J. (2002). Conditioning and hyperalgesia in newborns
exposed to repeated heel lances. Journal of the American Medical Association, 288, 857–61.
Talo, S., Forssell, H., Jeikkonen, S. et al. (2001). Integrative group therapy outcome related to psychosocial
characteristics in patients with chronic pain. International Journal of Rehabilitation Research, 24, 25–33.
Thieme K., Flor H. and Turk D.C. (2006). Psychological treatment in fibromyalgia syndrome: efficacy of
operant behavioural and cognitive behavioural treatments. Arthritis Research and Therapy, 8, R121.
Thieme K., Rose U., Pinkpank T., Spies C., Turk D.C. and Flor H. (2006). Psychophysiological responses
in patients with fibromyalgia syndrome. Journal of Psychosomatic Research, 61, 671–80.
Turk, D.C. (1990). Customizing treatments for chronic pain patients; who, what, and why. Clinical Journal
of Pain, 6, 255–70.
Turk DC. (2002). Clinical effectiveness and cost effectiveness of treatments for chronic pain patients.
Turk, D.C. (2004). Fibromyalgia: a patient-oriented perspective. In Dworkin, R.H., Breitbart, W.S. (Eds).
Psychosocial and Psychiatric Aspects of Pain: A Handbook for Health Care Providers, pp. 309–38. Seattle,
WA: IASP Press.
Turk, D.C. and Rudy, T.E (1988). Toward an empirically derived taxonomy of chronic pain patients:
Integration of psychological assessment data. Journal of Consulting and Clinical Psychology, 56, 233–38.
Turk, D.C. and Rudy, T. E. (1990a). Neglected factors in chronic pain treatment outcome studies–Referral
patterns, failure to enter treatment, and attrition. Pain, 43, 7–25.
Turk, D.C. and Rudy, T. E. (1990b). The robustness of an empirically derived taxonomy of chronic pain
patients. Pain, 42, 27–35.
Turk, D.C. and Okifuji, A. (1998). Treatment of chronic pain patients: Clinical outcome, cost-effectiveness,
and cost-benefits. Critical Reviews in Physical Medicine & Rehabilitation, 10, 181–208.
Turk, D. C. and Wilson, H. D. (2010). Fear of activity as a prognostic factor in patients with chronic pain:
conceptual models, assessment, and treatment implications. Current Pain and Headache Reports, 14,
88–95.
Turk, D.C., Penzien, D.B. and Rains, J.C. (1995). Temporomandibular disorders. In Goreczny, ed.
Handbook of Recent Advances in Behavioral Medicine. New York, NY: Plenum Press (pp. 55–77).
Turk, D.C., Okifuji, A., Sinclair, J.D. et al. (1996). Pain, disability, and physical functioning in subgroups of
fibromyalgia patients. Journal of Rheumatology, 23, 1255–62.
Turk, D.C., Okifuji, A., Sinclair, J.D. et al. (1998a). Differential responses by psychosocial subgroups of
fibromyalgia syndrome patients to an interdisciplinary treatment. Arthritis Care Research, 11, 397–404.
Turk, D.C., Sist, T.C., Okifuji, A. et al. (1998b). Adaptation to metastatic cancer pain, regional/local cancer
pain, and non-cancer pain: role of psychological and behavioral factors. Pain, 74, 247–56.
Turk, D.C., Okifuji, A., Sinclair, J.D. and Starz, T.W. (1998). Interdisciplinary treatment for fibromyalgia
syndrome: clinical and statistical significance. Arthritis Care & Research, 11, 186–95.
Turner, J.A., Franklin, G. and Turk, D.C. (2000). Predictors of long-term disability in injured workers:
A systematic literature synthesis. American Journal of Industrial Medicine, 38, 707–22.
Van den Hout, J.H., Vlaeyen, J., Heuts, P.H., Zijlema, J.H. and Wijnen, J.A. (2003). Secondary prevention
of work-related disability in non-specific low back pain: does problem-solving therapy help?
A randomized clinical trial. Clinical Journal of Pain, 19, 87–96.
Van der Windt, D., Hay, E., Jellema, P. and Main, C. (2008). Psychosocial interventions for low back pain
in primary care: Lessons learned from recent trials. Spine, 33, 81–9.
Verra, M.L., Angst, F., Brioschi, R., et al. (2009). Does classification of persons with fibromyalgia into
Multidimensional Pain Inventory subgroups detect differences inoutcome after a standard chronic
pain management program? Pain Research & Management, 14, 445–53.
Vlaeyen, J.W.S. and Morley, S. (2004). Active despite pain: the putative role of stop-rules and current
mood. Pain, 110, 512–16.
Vlaeyen, J.W.S., Kole-Snijders, A.M.J., Boeren, R.G.B. and van Eek, H. (1995). Fear of movement/(re)
injury in chronic low back pain and its relation to behavioral performance. Pain, 62, 363–72.
Vlaeyen, J.W.S., Haazen, I.W.C.J., Kole-Snijders, A.M.J. et al. (1996). Behavioural rehabilitation of chronic
low back pain: Comparison of an operant treatment, and operant cognitive treatment and an operant
respondent treatment. British Journal of Clinical Psychology, 34, 95–118.
Von Eye, A. and Bergman, L.R. (2003). Research strategies in developmental psychopathology: dimensional
identity and the person oriented approach. Development and Psychopathology, 15, 553–80.
Von Korff, M. (1999). Pain management in primary care: An individualized stepped-care approach. In
R.J. Gatchel and D.C. Turk (Eds.), Psychosocial factors in pain: Critical perspectives (pp. 360–73).
Von Korff, M., Ormel, J., Keefe, F.J. and Dworkin, S. F. (1992). Grading the severity of chronic pain. Pain,
50, 133–49.
Waddell, G. (1998). The back pain revolution. Edinburgh: Churchill Livingstone.
Waylonis, G.W. and Perkins, R.H. (1994). Post-traumatic fibromyalgia. A long-term follow-up. Am J Phys
Med Rehabil, 73, 403–12.
Westman, A., Boersma, K., Leppert, J. and Linton, S.J. (2011). Fear-avoidance beliefs, catastrophizing, and
distress: a longitudinal subgroup analysis on patients with musculoskeletal pain. Clin J Pain, 7, 567–77.
Wideman, T.H., Adams, H.E. and Sullivan, M.J.L. (2009). A prospective sequential analysis of the
fear-avoidance model of pain. Pain, 145(1–2), 45–51.
Widerström-Noga, E.G., Felix, E.R., Cruz-Almeida, Y. and Turk. D.C. (2007). Psychosocial subgroups in
persons with spinal cord injuries and chronic pain. Archives of Physical Medicine and Rehabilitation,
88, 1628–35.
Wilder-Smith, C.H., Schindler, D., Lovblad, K., Redmond, S.M. and Nirkko, A. (2004). Brain functional
magnetic resonance imaging of rectal pain and activation of endogenous inhibitory mechanisms in
irritable bowel syndrome patient subgroups and healthy controls. Gut, 53, 1595–601.
Wolfe, F., Smythe, H.A., Yunus, M.B. et al. (1990). The American College of Rheumatology 1990 criteria
for the classification of fibromyalgia: report of the multicenter criteria committee. Arthritis &
Rheumatism, 36, 160–72.
Wolfe, F., Russell, I.J., Vipraio, G. et al. (1997). Serotonin levels, pain threshold, and fibromyalgia
symptoms in the general population. J Rheumatology, 24, 555–9.
Woods, M.P. and Asmundson, G.J. (2008). Evaluating the efficacy of graded in vivo exposure for the
treatment of fear in patients with chronic back pain: a randomized controlled clinical trial. Pain,
136, 271–80.
Yarnitsky, D., Crispel, Y., Eisenberg, E., Granovsky, Y., Ben-Nun, A., Sprecher, E., Best, L.A. and
Granot M. (2008). Prediction of chronic post-operative pain: Pre-operative DNIC testing identifies
patients at risk. Pain, 138, 22–8.
Zubieta, J.K., Heitzeg, M.M., Smith, Y.R. et al. (2003). COMT val158met genotype affects mu-opioid
neurotransmitter responses to a pain stressor. Science, 299, 1240–43.
Chapter 27
Risk Factor-Based Cognitive-

Behavioural Therapy for Acute and
Subacute Back Pain
Monika I. Hasenbring, Bernhard W. Klasen,
and Adina C. Rusu
27.1 Introduction
There is substantial evidence that treatment of patients with chronic back pain reveals limited
success. Although several multimodal approaches including pharmacological, physiotherapeuti-
cal, and psychological features lead to statistical significant reduction of pain, pain-related disa-
bility, and work absenteeism, effect sizes are only small to moderate. Effects concerning clinical
significance are rather marginal (Van der Windt et al. 2008). A number of systematic reviews
concluded that offering patients multimodal treatment only when pain and disability had evolved
into a chronic state and offering treatment in the same manner to each patient represent two of
the central problems in treatment allocation (Morley et al. 1999; van Tulder et al. 2001; Hoffmann
et al. 2007; Henschke et al. 2010). Turner and colleagues (1996) recommended the implementa-
tion of early educational and cognitive-behavioural interventions in the phase of subacute pain,
when patients are still in primary care as well as more interventions targeted to the individual
needs of the patients. Treating low back pain (LBP) patients not as a homogeneous group and
identifying specifiable subgroups due to clinical, psychological, neurophysiological, and perhaps
genetic risk variables represents an important topic for research (Turk 2005; Brennan et al. 2006;
see Chapter 26). In line with this, Haldorsen (2003) pointed out that there is a challenge to find
‘the right treatment to the right patient at the right time’.
In this chapter, we will present some basic considerations concerning: (1) the choice of an
‘optimal’ time point, starting with more comprehensive, multimodal treatment approaches at a
cost-effective level; (2) potential criteria defining features of patients with back pain that may
describe appropriate patients; and (3) finding the right treatment for a specific patient. A qualita-
tive review of empirical evidence improving efficacy and effectiveness of current attempts in early,
risk factor-based psychosocial interventions in the treatment of subacute back pain will follow. In
the third section of this chapter, we will provide an illustrative description of current treatment
options. Finally, we will try to draw some conclusions towards the need for future research and
clinical practice.
27.2 Preliminary considerations

27.2.1 Finding the right time point for treatment
Bearing in mind that although a substantial number of acute low back pain (ALBP) patients will
recover with little or no medical care, close to one-third of these patients tend to develop
continuous or recurrent pain one year after an initial episode (Waddell 2004). As the duration of
pain is one of the most important predictors of the development of severely chronic pain states,
this leads to the conclusion that an effective treatment should start as early as possible. One of the
first interventional strategies that focused on primary prevention was aimed at the prevention of
the occurrence of back pain in healthy people at the workplace, since a number of risk factors for
the occurrence of LBP have been shown to be work-related (see Chapter 19). However, despite
the reduction of several ergonomic stressors at the workplace, the effects of these interventions on
the development of pain and disability in healthy workers remained more than marginal (Shaw
et al. 2006). The concept of secondary prevention—starting treatment in a phase of acute LBP
with a duration of not more than 4 weeks—should be more cost-effective, because interventions
are offered to fewer patients with a potentially greater effect. Within this context, it is less likely
that a person who is healthy and who may have no risk of developing LBP or chronic states will
receive an intervention that causes no effect but does cause costs. As described in more detail in
the second section of this chapter, a number of randomized trials have been published during the
past 10 years which aimed at secondary prevention of a recurrence of pain, pain-related disability,
and work-related issues, conducted at the workplace, within primary care, and also on a popula-
tion-based level. Although the results of these trials indicate that acute and subacute LBP is an
optimal time point for starting multidisciplinary treatment, we recognize that there is substantial
room for improvement of treatment offers.
27.2.2 Finding the right patient

Up to 60% of the patients, with ALBP will recover spontaneously with only little or no medical
treatment. Therefore, a useful strategy is to optimize treatments by identifying those patients
whose symptoms are likely to persist. Indeed, with the increasing knowledge about risk factors
and mechanisms that may be indicative of the development of chronic back pain conditions,
investigators and clinicians seek to develop new and even more effective treatment strategies
early, in an effort to prevent chronicity and in as cost-effective a manner as possible.
As several chapters of this volume have shown, there is no doubt that psychosocial risk factors
make a substantial contribution to the prediction of chronic pain, disability, and inability to
work. This is true for variables such as depression and distress at work. Maladaptive responses
to pain including catastrophizing about pain, fear of pain-related activities, more generalized
fear-avoidance beliefs, and endurance-related responses such as thought suppression and task
persistence behaviour despite severe pain all appear to contribute to disability (see chapters of
Part 4 for further details). As most of these factors are related to each other, we are left with the
question of their independent contribution (Foster et al. 2010). Additional important questions
concern the difficulty of modifying proven risk factors and which risk factors will yield a maxi-
mum effect if modified successfully. For instance, it might be easier to modify the tendency of
catastrophizing pain experiences by informing a patient in an acute or subacute pain episode
about the harmless character of his pain condition than to reach a reduction of distress at
work arising from long-lasting conflicts between the patient and his colleagues. Answering such
questions will require intervention studies targeting specified psychological risk factors.
A pre-condition of an individually targeted treatment is the identification of psychosocial risk
factors that are assessed with high reliability and predictive validity. Although it is desirable that a
screening tool reveals adequate sensitivity and specificity, most of the currently available psycho-
social screening measures reveal either high sensitivity and moderate specificity, or vice versa (see
Chapter 11). Consequently, whether a screening tool should show higher specificity or sensitivity
depends on costs. In the case of high-cost treatments, greater specificity is necessary in order not
to treat patients that did not require this treatment. If more low-cost interventions are available,
RISK FACTOR-BASED COGNITIVE-BEHAVIOURAL THERAPY 515
using an assessment measure or procedure with high sensitivity but only moderate specificity may
be justifiable.
27.2.3 Finding the right treatment

The next step after identifying the relevant psychological risk factors indicating the development
of ongoing or recurrent pain and increasing disability, is finding the right treatment approaches.
Intervention-based psychological pain research over the past decades seems to follow a continu-
ous alternation of deductive, theory driven, and inductive designs. In the 1960s, Fordyce (1976)
extended the operant condition model conceptualizing functional mechanisms leading to the
maintenance of so-called ‘pain behaviours’. For instance, facial expression of acute pain, caused by
a sudden damage of physical structures, may come under environmental control. A decade later,
stimulated by an extensive laboratory experimental research on attention and pain (Ahles et al.
1983; McCaul and Mallot 1984), Turk and colleagues (1983) developed a cognitive-behavioural
model of pain emphasizing the role of dysfunctional cognitions, such as maladaptive attributions
and perceptions of low self-efficacy. This model stimulated the cognitive-behavioural perspective
and the application of a range of cognitive and behavioural techniques for the treatment of
patients with persistent pain (Turk and Rudy 1986; Turk 1998). This perspective served as an
important stimulant for numerous experimental and clinical studies searching for specific mala-
daptive cognitions, such as catastrophizing, fear of injury/re-injury, and fear-avoidance beliefs,
leading to a more specific theoretical model of psychological mediators of pain, the fear-avoid-
ance model (FAM; Vlaeyen and Linton 2000). New cognitive-behavioural treatment approaches,
such as guided exposure, were developed, specifically treating patients suffering from high fear of
pain, pain-related activities, and avoidance behaviour (e.g. Vlaeyen et al 2001; George et al. 2008).
Recently, the avoidance-endurance model of pain was developed, focusing on additional cogni-
tive, affective, and behavioural responses to pain that seem to act as risk factors for subsequent
pain and disability. Pain-related fear may not only be followed by avoidance behaviour but also
by a pattern of pain persistence with cognitions of thought suppression and marked endurance
behaviour shown despite severe pain conditions. Moreover, there seems to be a subgroup of back
pain patients impressing by very low fear, positive mood despite pain, and elevated endurance
behaviour that also has a high probability to developing chronic pain conditions (Hasenbring
et al. 2001, Hasenbring and Verbunt 2010, see Chapter 15 for more details). These high-risk pat-
terns might optimally benefit from individually tailored cognitive-behavioural therapy (CBT)
approaches that incorporate different kinds of activity pacing interventions.
27.3 Early psychosocial interventions for LBP patients

in primary care
During the past decade, a number of randomized intervention trials in primary care settings have
been conducted, in which psychological treatments were offered to LBP patients who had back
pain of less than 12 weeks’ duration. While duration and content of the psychological interven-
tions differed between group discussions based on cognitive-behavioural principles (e.g. Linton
and Andersson 2000; Hay et al. 2005), and guided exposure sessions developed to decreasing high
fear of pain-related activities, most trials revealed significant reduction of pain, disability, and
catastrophizing (George et al. 2008; Linton et al. 2008). However, no or only small differences
between these psychological interventions and active control treatments have been reported.
Although a number of reasons may be responsible for these results (Van der Windt et al. 2008),
the most critical issue seems to be offering these interventions to a large proportion of patients
who might not need them. These patients show few psychosocial risk factors and they will develop
a high rate of spontaneous recovering, mirrored by the significant time effects. Subanalyses in
some of these secondary prevention studies revealed low levels of psychological distress (e.g. Hay
et al. 2005) and baseline disability (George et al. 2008), supporting this hypothesis. Therefore, it
will be important to improve the indication of psychosocial interventions, for instance by focus-
ing only on patients who show signs of psychosocial risk factors.
Jellema and co-workers (2005) performed a cluster-randomized trial (41 practices of general
medicine) of 143 patients with LBP of less than 12 weeks’ duration. Patients received a minimal
interventions strategy, carried out by general practitioners (GPs), aimed at the assessment and
modification of psychosocial prognostic factors compared to patients receiving usual care
(N=171). The GPs received two training sessions of 2.5 hours each, provided by a GP with exper-
tise in psychosocial interventions. The training consisted of theory, role-playing, and feedback on
the practised skills. The minimal intervention arm consisted of a 20-minute session with the
patient including three phases: exploration, information, and self-care. The authors designed this
session to identify and modify psychosocial risk factors. That means risk factors are assessed after
randomization and only in the intervention group. Consistent with our previous suggestion, the
results of this trial, revealed significant reduction of disability, pain intensity, and sick leave due
to back pain in both groups with no significant group difference. Within 6–13 weeks, 60–70% of
the patients’ symptoms had been resolved, which very likely mirrors the high rate of spontaneous
recovering in patients with acute and subacute back pain in general. Since all patients in the
minimal intervention group received the psychosocial interventions irrespective of the results of
the physicians’ rating, this trial covers a large number of low-risk patients known to showing a
high rate of spontaneous recovering. Therefore, despite relative large sample sizes, those trials will
be underpowered showing a significant group difference.
Another study conducted by the authors (Jellema et al. 2005) showed that, despite physician
training, the risk assessment based on global physician ratings, are of doubtful reliability and low
prospective validity (see also Streitlein-Böhme et al. 2008).
27.3.1 Risk factor-based cognitive-behavioural interventions (RCBIs)

RCBIs are guided by assessment of psychosocial risk factors. Randomized-controlled trials (RCTs)
usually assess risk factors before randomization. The procedures published to date, differ in how
the assessment of psychosocial risk factors (‘yellow flags’, see also Chapter 11) was handled. Some
studies used a patient or physician rating of the perceived risk for the development of chronic
pain, while other approaches used an independent, standardized assessment of psychosocial risk
factors known to be reliable and valid. RCBI were allocated randomly only to high-risk patients.
Low-risk patients were used for further comparison.
RCBI guided by a global self-rating of individual risk for chronicity
Linton and colleagues published a number of early intervention studies in order to prevent work
disability and they are leading in the development of psychosocial risk factor diagnostics (see
Chapter 11). In 2000, they reported first results of a randomized trial with a 5-year follow-up
published in 2006 (Linton and Andersson 2000; Linton and Nordin 2006). Although not explic-
itly named as a risk factor-based trial, it is, to our knowledge, one of the first early intervention
studies that has targeted only high-risk patients. The authors used a global self-rating of their
patients evaluating their personal risk for the development of chronic pain.
In the first study (Linton and Andersson 2000), 272 patients who fulfilled the inclusion crite-
rion of less than 3 months of cumulated sick leave due to back pain were randomly allocated to
three interventions groups (pamphlet, information-package, and CBT). Patients were recruited
from local primary care facilities and via an advertisement in a local newspaper. Primary outcomes
(all self-reported) were long-term sick-leave, healthcare use, and the self-perceived risk for devel-
oping long-term pain. The investigators included pain variables and different psychosocial char-
acteristics, assessed with standardized questionnaires as secondary outcomes. In the pamphlet
group participants received short written information based on the fear-avoidance model recom-
mending positive thinking and remaining active and confronting with pain rather avoiding pain-
related activities. In the information package group participants received more conservative
written information presented once a week for 6 weeks with relatively broad and general informa-
tion. Investigators typically conduct RCBI using a 6-session structured programme with a 2-hour
session once a week and 6–10 patients per group. Certified behaviour therapists provided a treat-
ment based on a written manual. The six sessions were concerned with the following topics:
(1) causes of pain and prevention of chronic problems with information about pain, learning
problem-solving skills and applied relaxation; (2) managing pain with activity scheduling and
maintaining activities; (3) promoting good health and controlling stress at home and at work,
with detecting warning signals and considering health beliefs; (4) adapting for leisure and work,
with communication skills and assertiveness training in risk situations; (5) controlling flare-ups;
and (6) maintaining and improving results making plans for adherence. Each of the sessions
included initial information for a maximum of 15 minutes, structured exercises, and individual-
ly-tailored homework assignments. Results demonstrated significant group differences concern-
ing sick absenteeism between CBT and the pamphlet group, using an intention-to-treat analysis.
Both information groups showed an increase in the number of sick days from baseline to the
1-year follow-up, whereas CBT participants showed a slight decrease. The difference between
CBT and pamphlet was still significant after 1 year. Regarding the number of visits to the physi-
cian and physical therapist per year, CBT patients showed a significant decrease, whereas both
information groups showed an increase. The third main outcome, self-rated perception of
long-term disability, revealed a significant decrease in CBT and a stable level in both information
groups, although group differences at the follow-up were not significant. Furthermore, no group
differences were found at follow-up concerning pain intensity, activity level, depression, and the
fear-avoidance variables such as catastrophizing, fear-avoidance beliefs, or fear of pain/injury.
Overall, the sample of this study has shown relatively mild problems at baseline level, perhaps due
to the specific selection procedure of patients. However, the investigators reported a significant
time effect with a decrease of scores in the fear-avoidance beliefs in all groups and a decrease in
catastrophizing in both information groups, while the level of fear of pain/injury remained stable
over time in all groups. The results on general anxiety and depression were comparable.
In a 5-year follow-up of the 2000 sample, Linton and Nordin (2006) conducted a reassessment
by now comparing the CBT group with the combined information group due to the fact that in
most variables there had been no differences between the pamphlet and the information package
group. At this long-term follow-up, the CBT group was superior to the information group in
most of the outcomes, specifically with respect to pain intensity, activities of daily living, quality
of life, and medication use. A detailed cost analysis revealed lower pain and disability-related costs
for the CBT group compared to the information group. Although the CBI sessions were six times
more expensive than the information approach, the total costs were lower due to higher produc-
tivity (fewer days of sick leave) and a lower rate of healthcare costs due to fewer healthcare visits
in the CBT group.
RCBI guided by the results of a comprehensive psychosocial risk assessment

During the past decade, only a few studies have conducted a standardized and validated assess-
ment of psychosocial risk factors in order to guide the indication of psychosocial interventions.
For instance, Gatchel and colleagues (Gatchel et al. 2003; Whitfill et al. 2010) evaluated the
efficacy of CBI as part of a multimodal functional restoration programme in patients suffering

from acute low back pain. Hasenbring et al. (1999) investigated risk factor-based CBI in a sample
of subacute sciatic pain patients who were offered (in addition to conservative medical treatment
such as pain medication), relaxation and physiotherapy. In a recent RCT, Hasenbring et al. (sub-
mitted) compared a short physician-delivered counselling based on risk screening with group
therapy interventions and treatment as usual in a sample of subacute non-specific LBP from
different primary care practices. In all approaches, the investigators allocated patients with high
psychosocial risk factors in the previous screening, randomly to targeted interventions and
compared them with treatment as usual groups or with different standardized active treatments.
Furthermore, both approaches used the low-risk group, matched carefully by gender, age, and
diagnosis, as a further comparison group. We will describe these in more detail in the next section
(see ‘RCBI guided by a short theory-based screening of psychosocial risk factors’).
Gatchel and co-workers (Gatchel et al. 2003) used a comprehensive assessment, consisting of
two pain and disability visual analogue scales, Scale 3 (‘Hysteria’) of the Minnesota Multiphasic
Personality Inventory (MMPI), and the fact of existing workers compensation in identifying
patients showing a high psychosocial risk of not returning to work at a 12-month follow-up. In a
prospective, longitudinal study of 421 patients with acute LBP (less than 10 weeks duration) this
risk assessment yielded a correct classification of more than 90% of the cases, who did not return
to work at the follow-up (Gatchel et al. 1995). In a second validation study using a number of
standardized measurements of psychopathological disorders, coping styles, and specific traits in
the domain of personality disorders, Pulliam et al. (2001) showed that high- and low-risk patients
could be classified with 80% sensitivity and 81.5% specificity. High-risk patients were low on
positive temperament and showed more workaholic tendencies measured with the Schedule for
Nonadaptive and Adaptive Personality (SNAP, Clark 1993). High scorers on this scale are char-
acterized by their placing work above leisure time and neglecting social contact such as friends
and family in favour of work. They indicate that they do not finish a job until it is perfect, and they
will drive themselves until exhausted. This pattern of pain-related cognitions, emotions and
behaviour is comparable to the endurance pattern, described within the avoidance-endurance
model of pain (Hasenbring and Verbunt 2010; see also Chapter 15 of this volume), and the
concept of a maladaptive ‘work style’, described by Feuerstein and colleagues (2005).
Based on this psychosocial risk assessment, conducted within a number of orthopaedic prac-
tices, high-risk patients, randomized into the interventions group (HR-I), Gatchel and colleagues
(2003) offered an early interdisciplinary intervention programme consisting of psychological
components, physical therapy, occupational therapy, and case management. The protocol con-
sisted of three physician and one physical therapy evaluation, nine physical therapy sessions of
15 minutes’ individual training, nine physical therapy group sessions of 30 minutes’ duration,
nine 30-minute biofeedback/pain management sessions, nine group didactic lesions each lasting
45 minutes, nine case manager/occupational therapy sessions lasting 30 minutes each, and three
interdisciplinary team conferences. Trained and licensed therapists administered all treatments.
All sessions were ideally spaced over a 3-week period; however most patients did not need all
sessions. The outcome assessed by structured telephone interviews at 3-, 6-, 9-, and 12-month
follow-up, included return-to-work status, number of healthcare visits in general and related to
LBP, number of disability days, pain intensity, and medication use. In all variables, the 22 HR-I
patients showed a highly significant better outcome compared to high-risk patients without this
interdisciplinary treatment (HR-NI, N=48). HR-I patients delivered a similar positive outcome as
the low-risk patients at the 1-year follow-up on pain intensity and return to work. In a rough cost-
comparison analysis, the authors reported significantly lower costs ($12,721 in HR-I compared to
$21,843 in HR-NI) despite the costs for this early intervention programme (Gatchel et al. 2003).
In a recent RCT, this kind of early intervention programme was as effective as a new protocol that
consisted of these interventions given in addition to a specific work-transition module (Whitfill
et al. 2010).
RCBI guided by a short theory-based screening of psychosocial risk factors
During the past two decades, Hasenbring and colleagues developed pain risk screening-
based cognitive-behavioural approaches (PARIS-CBA) that were evaluated within two RCTs
(Hasenbring et al. 1999, submitted). PARIS-CBA contains of different kinds of cognitive-
behavioural interventions, such as short counselling sessions (PARIS-CBC), group therapy
(PARIS-CBGT), and individual therapy interventions (PARIS-CBIT). A previous ‘yellow-flag’
screening that is based on the avoidance-endurance model (AEM) of pain (see Chapters 15 for the
AEM and 11 for the screening RISC-BP) guides all approaches. The investigators developed
standardized modules aimed at the modification of AEM-based dysfunctional pain response
pattern, such as fear-avoidance-, distress-, or eustress-endurance pattern.
In the following, we will outline some basic principles of PARIS-CBA and different steps of
interventions and modules, referencing the kind of intervention to which it is relevant. We rec-
ommend that these interventions be administered within a stepped-care programme due to
limited financial resources. This ranges from short physician and/or physiotherapist counselling
(15 minutes’ duration) as a first step to cognitive-behavioural group therapy (6–12 weekly ses-
sions of 100 minutes each) and cognitive-behavioural individual therapy (25–45 or more weekly
sessions of 50 minutes each) as further steps. We consider the reliable and valid indication of the
need to these more extensive approaches. Finally, we report on the results of two RCTs conducted
in two different settings of medical care.
Basic principles of PARIS-CBA Irrespective of the kind of CBA, several principles guide the dif-
ferent procedures as well as the relationship between patient and CBA-provider. These include:
(1) all interventions and modules will follow an individually tailored schedule; (2) the relation-
ship between patient and CBA provider is based on principles of shared decision-making that
have been shown to increase patient satisfaction and adherence (Joosten et al. 2008); and (3) all
interventions are based on ‘third wave’ cognitive-behavioural methods, including learning prin-
ciples, dysfunctional cognitions perspective, supporting individual resources including an increase
of acceptance as well as interpersonal approaches.
The principles of individualization and shared decision-making are realized within several steps
of PARIS-CBA interventions. The primary concern to: (1) the personal risk profile of a patient;
(2) the patients’ illness model; (3) the patients’ individual treatment goals; and (4) the patients’
level of motivation for behavioural changes, following the trans-theoretical model of behavioural
change (Norcross et al. 2011; Prochaska et al. 2001). At first, the AEM-based yellow-flag screening
(RIsc SCreening for Back Pain, RISC-BP) informs a patient whether screening data reveal a high-
risk pattern for the development of chronic back pain and second, which precise pattern of high-
risk behaviour is present—a fear-avoidance pattern or a distress or eustress-endurance-pattern.
An important question for the provider is to reconfirm as to whether patients will recognize
themselves when informed about their risk profile or which part of the risk behaviour will be
familiar to them. We tailor this reinsurance to the individual model of illness representation that
may be more somatic in some cases or more psychosocial in others. Our clinical impressions
reveal that many patients with subacute pain report substantial psychosocial stress (e.g. conflicts
with their partner or with colleagues at work) but they do not recognize a relation between these
stressors and their pain problem. Further, a number of patients do recognize aspects of a distress-
endurance pain pattern and they can imagine that this is a dysfunctional way to respond to pain
but they do not know how to change this pattern.
We extend the principle of individualization and shared-decision-making in the development

of treatment goals. CBA providers will propose different treatment goals to their patients and
encouraging them to develop their own way. Furthermore, providers will continuously harmo-
nize their own perspectives with the ideas, goals and wishes of their patients.
In addition to consideration of learning principles and modification of dysfunctional pain-
related cognitions and behaviour, PARIS-CBA aims to promote the individual awareness of physi-
cal signs and individual needs, especially when pain increase will be only slight, as well as to
increase the acceptance of the need for an individual balancing of physical needs and external
demands. As a great number of individual pain situations will occur within an interpersonal con-
text (e.g. with relatives, friends, or colleagues at work), interpersonal approaches including meth-
ods of social competence training will be included.
PARIS-CBA: procedure and steps of intervention In the following, standardized procedures of

the PARIS-CBA stepped-care programme will be presented, beginning with instructions con-
cerning a preliminary provision of the patient with a biopsychosocial perspective in further diag-
nostics, yellow-flag assessment, individual goal assessment, and finally an overview of
endurance-related interventions.
Step 1: providing the patient with a biopsychosocial model of pain PARIS-CBA aimed at a harmo-
nization of the illness models of different healthcare providers acting at the different settings of a
stepped care model. As it is proposed to offer a yellow-flag screening by physicians in primary
care, presenting such an illness model in a standardized manner should be part of the introduc-
tion of the screening procedure. PARIS-CBA proposes the following instruction the physician
may give to his patient: ‘During the past weeks your pain did not respond (enough) to the treat-
ment I had proposed to you. Our careful diagnostics ruled out any serious illness. Therefore, we
have to look for further potential influences that can cause or maintain back pain. First, it could
be that biomechanical stress (e.g. long-lasting and/or recurrent dysfunctional body postures) may
cause or maintain back pain via an increase of muscular load and muscular tension. Daily psycho-
logical stress, such as conflicts with colleagues or superiors at work may also increase muscular
tension. Furthermore, if someone has to care for an ill or elderly relative, biomechanical (physi-
cal) and psychological stress may act simultaneously. Second, when pain does not respond to
medical treatment as we would expect and when it lasts longer than anticipated, it very often leads
to sleep disturbances, leading to increased fatigue during the day and increasing effort to main-
tain daily activities in spite of fatigue. This, in the long-term, may cause additional muscular
tension and more and more depressive mood, leading to a vicious circle, which influences pain
and disability. Third, how we do respond to pain emotionally is important, in our spontaneous
automatic cognitions and concerning our behaviour, there are more functional and dysfunctional
ways to respond to pain.’ As a first step, we might ask patients to complete a short screening
questionnaire and tell them that their responses will be used as a basis for discussion of further
treatment options.
The physician may observe the reactions of his patients towards these potential pain factors and
pathways. One patient will quickly respond positively when exploring psychological stress, for
instance when he is suffering from some kind of stress. Another patient will look highly sceptical
when mentioning psychological stress, but respond by nodding his head when mentioning sleep
disturbances and depression as a consequence of pain or dysfunctional coping. This is one of the
first steps in which the physician may explore the personal illness model of his patient. As all
directions of a potential vicious circle are involved, the probability that patients may accept one
aspect of the illness model is high.
Step 2: Yellow-flag assessment and counselling PARIS-CBA offers different kinds of presentation
modes for the yellow-flag screening RISC-BP—such as a paper-and-pencil form or in a compu-
ter-based form, which might be conducted within a primary care practice or via Internet to
be filled in at home (Hasenbring and Hallner, 1999). In the latter form, patients will receive a
specific identification number and be asked to bring this to their physicians after answering the
RISC-BP via computer at home. The physician has access to the results, will explain these to his
patient, and will discuss the further steps of treatment with him. It is important to initially train
physicians on how to explain screening results to the patient by using standardized instructions,
which may be individually adapted to each patient.
In the following, we provide an example of these suggested instructions for the classification as
high risk of development of chronic pain via the ‘distress-endurance response’ (DER) pattern:
‘The screening revealed high scores in negative, depressive mood together with a tendency for
thought suppression. This means that you mostly try to suppress pain and all thoughts about
pain, in order not to interrupt your daily activities. The depression score is not on pathological
level and many people do respond with automatic thoughts trying to suppress thoughts about
pain sensations, but in most cases, this is not very successful. Unwanted thoughts will persist
thoroughly and your mood might become more and more irritated and depressed. In your
behaviour, you are scoring high on endurance behaviour, which might mean that you try to finish
all activities of daily life despite your severe pain. On the one hand, being able to show endurance
behaviour despite pain or other unwanted negative feelings is an important strength at times with
a lot of daily stress. However, if we do chronically respond with thought suppression and endur-
ance behaviour and are not able to realize short breaks during daily activities in order to relax,
these responses will lead to long-term overload or overuse of muscles, discs and other physical
structures, causing or maintaining your current pain. Is there anything you recognize and you
may agree with?’
Step 3: Individual goal assessment Goal assessment differs between short physician counselling,
group treatment, and individual treatment with respect to the increasing degree of individualiza-
tion and further concerning the extent of potential changes in patients’ dysfunctional cognitions
and behaviour that may be possible. However, potential causes of dysfunctional pain pattern,
such as motivational factors (Van Damme et al. 2010) or the kind of learning history are still a
matter of research, which is just beginning. Motivational factors of persistent dysfunctional pain
pattern may vary from more situational influences, such as short-term occupational pressure, to
more individual factors. The latter may result, for example, from kinds of self-discrepancies
(Kindermans et al. 2011) and long-term learning mechanisms. It is further a matter of research
whether the resulting dysfunctional pattern are easy to change (e.g. merely by a short education
or whether motivation, learning history, and other possible causes will afford more intense treat-
ment methods in order to reach an adaptive modification). It seems to be plausible, that offering
screening and counselling in the very early phases of subacute LBP will increase the probability of
more situational influences of dysfunctional pain response pattern and that these will be more
easily changeable by low-dose interventions. Before the physician decides to provide the patient
with solutions, he should ask his patient whether he has ideas how to change his situation the
next time it occurs. By this procedure, the physician may get a first impression of the level of
motivation for behaviour change the patient may represent.
Following a first physician counselling, primary care providers may make one or two more
appointments with their patients in order to evaluate their level of motivation for behaviour
change. Providers should attempt to determine which among the transtheoretical levels of:
(1) precontemplation in which there is no intention to change behaviour; (2) contemplation, in
Box 27.1 Guiding principles of ‘graded balancing’

1. Interruption of pain-related endurance should be realized very early, when pain increases
from 3 to 4 (in case of an 11-numeric self-rating ranging from 0 to 10). Only in the case of
early breaks, relaxation has a chance to be successful.
2. Only short breaks of current activity should be realized, finding an individual alternative
leading to muscular and mental relaxation.
3. Automatically occurring thoughts as obstacles for balancing should be observed as a
primary hindrance.
4. Individual motivation background and learning history of thought suppression and
endurance behaviour has to be explored.
which patients are aware that a problem exists but have not yet made a commitment to take
action; (3) preparation, in which patients are intending to take action and reporting some small
behavioural changes; (4) action, in which they modify their behaviour or environment to over-
come their problems; and (5) maintenance, in which patients work to prevent relapse and main-
tain the gains attained during action (Prochaska et al. 2001) is best applied to each patient. When
level 2 or 3 are recognizable but the patient signals persistent failure to reach behavioural changes
by his own effort, when psychological risk factors indeed persist, and when also pain persists for
a couple of weeks, the next step of a CBA group intervention should be considered.
Step 4: Endurance-related interventions following the principle of ‘graded balancing’ In the case of a
marked fear-avoidance-related pain response pattern, treatment strategies derived by the princi-
ples of ‘graded exposure’ will be offering to the patient. Vancleef and co-workers describe this kind
of treatment in more detail in Chapter 14 of this volume. Furthermore, this chapter will present
some essential parts of the principle of ‘graded balancing’ offered to patients with marked endur-
ance-related pain response pattern (see Box 27.1).
Within the short physician counselling of 15 minutes’ duration, only a few minutes are available
for discussing alternatives of the individual dysfunctional pain pattern. ‘Graded balancing’
encourages the patient to have short breaks for relaxation when pain increases above an accept-
able level and to observe automatic thoughts that normally occur when pain threatens to inter-
rupt a current activity, thoughts that can be labelled as obstacles for realizing an effective short
break. For example: ‘I do not have time for a break right now’, ‘You can have a break later on,
when you have finished your work’, ‘If you have a break right now, you will never finish your
work’. The guiding principle is to find a balance between an acceptable degree of endurance
(maintaining a current pain-irrelevant activity) and an increase of relaxation-inducing activities
(e.g. interruption of current activities for physical activity in case of long-term sitting at work). In
principle, graded balancing seems to be related to the concept of ‘activity pacing’, which has been
described by other authors (e.g. Nicholas et al. 2003; McCracken 2005). The kind of relaxation-
inducing activities depends on the mode of current activity. When pain increases following a
long-term phase of sitting at work (e.g. in a forward bend position as a biomechanical strain posi-
tion), a short break lying down (only for a few minutes) and later cycling or walking for a while
may represent the situation-specific kind of relaxation. An important add-on of the physician
counselling is a back book describing dysfunctional pain response pattern as well as alternative
responses in the sense of ‘graded balancing’, based on the avoidance-endurance-model of pain.
Within six or more sessions of PARIS-CBA group treatment, more time is available for: (1) a
more intense discussion of illness- versus health-related cognition and behaviour; (2) exercises
for the detection of individual maladaptive automatic cognitions and behaviour; and
(3) exercises, for example, role plays, for the modification of maladaptive automatic cognitions
and behaviour. However, within a six-session PARIS-CBA group treatment including seven or
eight patients, each patient will have two or three times the opportunity for individual exercises
during these sessions.
In the case of a long learning history of maladaptive endurance coping and of additional psy-
chosocial risk factors such as depression due to unsolved emotional problems, a PARIS-CBA
individual treatment may be appropriate. Using a 25-session short-term therapy, there is time for
analysis and change of persistent and stubborn kinds of meta-cognitions (e.g. ‘If I continue with
this therapy, my whole household will be in a mess’). These cognitions may act as powerful obsta-
cles against adaptive cognitions (e.g. ‘I will have a short break and relax for a little while before
I continue with my activities’).
Empirical evidence for PARIS-CBA In a first RCT, Hasenbring and colleagues (1999) evaluated
the efficacy of a PARIS-CB individual therapy trial conducted in a sample of subacute sciatic pain
patients during an inpatient hospital stay for conservative medical treatment. At the time of
admission all patients underwent the RISC-BP psychosocial screening consisting of the Beck
Depression Inventory (BDI) depression score and two subscales of the Kiel Pain Inventory,
embedded in a more comprehensive psychosocial assessment. The evaluation of increased
psychosocial risk as well as the classification of high-risk patients as individuals showing fear-
avoidance responses (FARs), distress-endurance responses (DERs) or eustress-endurance
responses (EERs) to pain followed the procedure published by Hasenbring (1993, see also Chapter
11 in this volume). The purpose of this intervention study was to evaluate the efficacy of a risk
factor-based cognitive-behavioural regimen (in this study labelled as RCBT) in comparison with
another active treatment and with a third group that only received treatment as usual, therefore
conservative medical treatment (group HRIS). The second active treatment consisted of electro-
myographic biofeedback (BFB), a standardized behavioural treatment in patients with chronic
pain, shown as effective in previous studies (Flor and Birbaumer 1993). These three groups were
compared with a self-selected group of low-risk patients (LRIS) undergoing treatment as usual.
The investigators hypothesized that both behavioural interventions, in addition to conservative
medical treatment, would be more effective in the prevention of chronic pain than medical treat-
ment without psychological intervention. Further, that high-risk patients who attended the
behavioural interventions would demonstrate an improvement similar to that of low-risk patients,
and finally, an individually scheduled risk factor-based cognitive behavioural treatment would be
more effective than a standardized biofeedback regimen. The study provided follow-up assess-
ments at 3, 6, 12, and 18 months after baseline. Results of repeated measurement ANOVAs have
shown that according to the first hypothesis both active treatments (RCBT and BFB) revealed
greater improvement in pain intensity than HRIS, while furthermore, RCBT exhibited greater
pain reduction than BFB (see Figures 27.1–27.3).
Only the RCBT group revealed significantly better outcomes than HRIS patients; the BFB
group scored as insignificant between RCBT and HRIS on the secondary outcomes of disability,
depression, medication intake, and application for early retirement. A strength of this study was
the calculation of different measures of clinical significance following the suggestions by Jacobson
et al. (1984). For example, the group of low-risk patients, known to show a good outcome from
previous studies (e.g. Hasenbring et al. 1994), was treated as a functional reference group and the
percentage of high-risk patients showing an improvement within the range of LRIS for each
experimental condition was assessed. The authors defined as two standard deviations of the mean
within that functional group. At the time of discharge from the hospital, 83% of the RCBT
10
9
8
Pain intensity (0 –10) 7
6 Low risk
5 High risk/no CBT
4 High risk/RCBT
3 High risk/BFB
High risk/RBT
2
1
0
Admission Discharge 3 months
Fig. 27.1 Pain intensity over time in different treatment and control groups from the time of
admission to the hospital, discharge, and the 3-month follow-up. Reproduced from Hasenbring, M.,
Ulrich, H.W., Hartmann, M. & Soyka, D. The efficacy of a risk factor based cognitive behavioral
intervention and EMG-biofeedback in patients with acute sciatic pain: an attempt to prevent
chronicity. Spine, 24, pp.2525–35 © 1999 Lippincott, Williams, and Wilkins with permission.
patients showed a decline of pain intensity that fulfilled this criterion in contrast to 18% in BFB
and 0% in HRIS. At the 6 month follow-up, 63% in both RCBT and BFB groups fulfilled this
criterion compared to 8% in HRIS. However, 66% in RCBT showed a clinically significant reduc-
tion in depression compared to 33% in BFB and 25% in HRIS. The investigators calculated the
reliability of change index (RC; Jacobson et al. 1984) as a second measure of clinical significance.
Values of RC exceeding 1.96 are unlikely to occur (p <0.05) unless an actual change in scores (of
at least two standard deviations) occurs between pre-treatment and follow-up assessments. At the
6-month follow-up, 91% of the RCBT group showed a RC improvement in pain intensity com-
pared to 78% in BFB, and 33% in HRIS. No patient in RCBT revealed a RC deterioration compared
10
High risk/RCBT
9
High risk/BFB
8
High risk/RBT
Pain intensity (0 –10)
7
6
5
4
3
2
1
0
on
ge
th
th
h
t
t
on
on
on
on
si
ar
is
-m
-m
ch
m
3-
6-
is
12
18
Ad
Fig. 27.2 Pain intensity over time in the two treatment groups and the comparison group from the
time of admission to the hospital, discharge, and the 3-, 6-, 12-, and 18-month follow-ups.
100
90
80
Physical function (%) 70

60
50
40
30
High risk/RCBT
20
High risk/BFB
10 High risk/RBT
0
on
th
th
th
th
on
on
on
on
si
is
-m
-m
m
3-
6-
12
18
Ad
Fig. 27.3 Physical function (%) over time in the two treatment groups and the comparison group
from the time of admission to the hospital, discharge, and the 3-, 6-, 12-, and 18-month follow-ups.
to 8%, 22% and 33% at the long-term follow-ups in the BFB group. Finally, a self-selected group
of patients who had refused an additional BT (RBT) showed no improvement over time, compa-
rable to HRIS.
27.4 Conclusion and future directions

The recommendations not to treat LBP patients as a homogenous group of patients had a stimu-
lating impact on the development of specific clinical, psychological, and neurobiological features
that serve as a basis for subgrouping patients and targeting treatments more individually.
Especially in the phase of acute and subacute back pain, a number of psychological features are
identifiable as high-risk factors (‘yellow flags’) indicating a high probability of developing chronic
pain and pain-related disability. Recent advances in the further development of approaches
directed more and more towards subgroups of patients showing specific psychosocial risk pro-
files. However, controlled randomized trials, based on a preliminary reliable and valid yellow flag
screening and offering CBT only to subgroups of patients who indeed show high psychosocial risk
factors, are still rare. This research design would increase the probability that not too many
patients receive CBT who would not need this kind of intervention. Systematic comparisons with
the short- and long-term follow-up of low-risk patients, who have a high probability of good
recovery and may serve as a functional reference group, should be conducted. Functional refer-
ence groups deliver norms of psychosocial functioning and should act as a further approach test-
ing the clinical significance of a trial. Cognitive-behavioural interventions conducted as risk
factor-based treatments, should draw on validated theoretical models, as recommended once
again by several authors (e.g. Michie et al. 2008).
References
Ahles, T.A., Blanchard, E.B. & Leventhal, H. (1983). Cognitive control of pain: attention to the sensory
aspects of cold-pressor stimulus. Cogn Res Ther, 7, 159–78.
Brennan, G.P., Fritz, J.M. & Hunter, J. (2006). Identifying subgroups of patients with acute/subacute
‘nonspecific’ low back pain. Results of a randomized clinical trial. Spine, 31(6), 623–31.
Clark, L. (1993). Schedule for Nonadaptive and Adaptive Personality (SNAP). Minneapolis, MN: University
of Minneapolis Press.
Feuerstein, M., Nicholas, R.A., Huang, G.D., Haufler, A.J., Pransky, G. & Robertson, M. (2005).
Workstyle: development of a measure of response to work in those with upper extremity pain. J Occup
Rehabil, 15, 87–104.
Flor, H. & Birbaumer, N. (1993). Comparison of the efficacy of electromyographic biofeedback,
cognitive-behavioral therapy, and conservative medical interventions in the treatment of chronic
musculoskeletal pain. J Consult Clin Psychol, 61, 653–8.
Fordyce, W.E. (1976). Behavioral methods for chronic pain and illness. St. Louis, MO: Mosby.
Foster, N.E., Thoams, E., Bishop, A., Dunn, K.M., & Main, C.J. (2010). Distinctiveness of psychological
obstacles to recovery in low back pain patients in primary care. Pain, 148, 398–406.
Gatchel, R.J., Polatin, P. & Mayer, T. (1995). The dominant role of psychosocial risk factors in the
development of chronic low back pain disability. Spine, 20(24), 2702–9.
Gatchel, R.J., Polatin, P.B., Noe, C., Gardea, M., Pulliam, C. & Thompson, J. (2003). Treatment- and
cost-effectiveness of early intervention for acute low-back pain patients: a one-year prospective study.
J Occup Rehabil, 13(1), 1–9.
George, S. Z., Zeppieri, G. Jr., Cere, A.L., et al. (2008). A randomized trial of behavioral physical therapy
interventions for acute and sub-acute low back pain. Pain, 140(1), 145–57.
Haldorsen, E.M.H. (2003). The right treatment to the right patient at the right time. Occupational
Hasenbring, M. & Hallner, D. (1999). Telemedizinisches Patienten-Diagnose-System. Deutsches Ärzteblatt/
Praxis Computer, 6, 49–50.
Hasenbring, M.I. & Verbunt, J.A. (2010). Fear-avoidance and endurance-related responses to pain: new
models of behaviour and their consequences for clinical practice. Clin J Pain, 26, 747–53.
Hasenbring, M., Marienfeld, G., Kuhlendahl, D. & Soyka, D. (1994). Risk factors of chronicity in lumbar
Hasenbring, M., Ulrich, H.W., Hartmann, M. & Soyka, D. (1999). The efficacy of a risk factor based
Hasenbring, M., Hallner, D. & Klasen, B. (2001). Psychological mechanisms in the transition from acute to
chronic pain: over- or underrated? Schmerz, 15(6), 442–7.
Hasenbring, M.I., Klasen, B.W., Streitlein-Böhme, I. & Rusche, H. Risk-factor based cognitive-behavioural
approaches in sub-acute low back pain patients: first results of a randomized, controlled trial
(submitted).
Henschke, N., Ostelo, R.W., van Tulder, et al. (2010). Behavioral treatment for chronic low back pain.
Cochrane Database Syst Rev, 7, CD002014.
Hoffmann, B.M., Papas, R.K., Chatkoff, D.K. & Kerns, R.D. (2007). Meta-analysis of psychological
interventions for chronic low back pain. Health Psychol, 26, 1–9.
Jacobson, N.S., Follette, W.C. & Revensdorf, D. (1984). Psychotherapy outcome research: Methods for
reporting variability and evaluating clinical significance. Behav Ther, 15, 336–52.
Jellema, P., van der Windt, D.A.W.M., van der Horst, H.E., et al. (2005). Should treatment of (sub)acute
low back pain be aimed at psychosocial prognostic factors? Cluster randomised clinical trial in general
practice. BMJ, 331, 84.
Joosten, E.A.G., DeFuentes-Merillas, L., DeWeert, G.H., Sensky, T., vander Staak, C.P.F. & deJong, C.A.J.
(2008). Systematic review of the effects of shared-decision making on patient satisfaction, treatment
adherence and health status. Psychotherapy and Psychosomatics 77, 219–26.
Kindermans, H.P., Huijnen, I.P., Goossens, M.E., Roelofs, J., Verbunt, J.A. & Vlaeyen, J.W. (2011). ‘Being’
in pain: the role of self-discrepancies in the emotional experience and activity patterns of patients with
chronic low back pain. Pain, 152(2), 403–9.
Linton, S.J. & Andersson, T. (2000). Can chronic disability be prevented? A randomised trial of a
cognitive-behavioural intervention and two forms of information for patients with spinal pain. Spine,
25, 2825–31.
Linton, S.J. & Nordin, E. (2006) A 5-year follow-up evaluation of the health and economic consequences of
an early cognitive behavioral intervention for back pain: a randomized, controlled trial. Spine, 31, 853–8.
Linton, S.J., Boersma, K., Jansson, M., Overmeer, T., Lindblom, K. & Vlaeyen, J.W. (2008). A randomized
controlled trial of exposure in vivo for patients with spinal pain reporting fear of work-related
activities. Eur J Pain, 12, 722–30.
McCaul, K.D. & Malott, J. (1984). Distraction and coping with pain. Psychol Bull, 95, 516–33.
McCracken, L.M. (2005). Contextual cognitive-behavioral therapy for chronic pain (Progress in pain research
and management, vol. 33). Seattle, WA: IASP Press.
Michie, S., Johnston, M., Francis, J., Hardeman W. & Eccles, M. (2008). From theory to intervention:
Mapping theoretically derived behavioural determinants to behaviour change techniques. Applied
Morley, S., Eccleston, C. & Williams A. (1999). Systematic review and meta-analysis of randomized
Nicholas, M., Molloy, A., Tonkin, L. & Beeston, L. (2003). Manage your pain: practical and positive ways of
adapting to chronic pain. London: Souvenir Press.
Norcross, J.C., Krebs, P.M. & Prochaska, J.O. (2011). Stages of change. J Clin Psychol, 67, 143–54.
Prochaska, J.O., Redding, C.A. & Evers, K. (2001). The transtheoretical model and stages of change. In
K. Glanz, F.M. Lewis, & B.K. Rimer (Eds.), Health behavior and health education (3rd edn.), pp. 60–84.
San Francisco, CA: Jossey-Bass.
Pulliam, C.B., Gatchel, R.J. & Gardea, M.A. (2001). Psychosocial differences in high risk versus low risk
acute low-back pain patients. J Occup Rehabil, 11(1), 43–52
Shaw, W.S., Linton, S.J. & Pransky, G. (2006). Reducing sickness absence from work due to low back pain:
how well do intervention strategies match modifiable risk factors? J Occup Rehab, 16, 591–605.
Streitlein-Böhme, I., Hasenbring, M., Hallner, D. & Rusche H. (2008). Do psychosocial risk factors
(‘yellow flags’) influence the diagnosis and therapy of acute and subacute low back pain in the general
practitioner (GP) practice? Z Allg Med, 84, 538–42.
Sullivan, M.J.L., Bishop, S.R. & Pivik, J. (1995). The pain catastrophizing scale: development and
validation. Psychol Assess, 7, 524–32.
Turk, D.C. (1998). Cognitive factors and persistent pain: A glimpse into Pandora’s box. Presented at the
‘Pain Management ’98: Failed Back Surgery Syndrome Conference’, Rotterdam, The Netherlands, 12
November, 1998.
Turk, D.C. (2005). The potential of treatment matching for subgroups of patients with chronic pain:
lumping versus splitting. Clin J Pain, 21, 44–55.
Turk, D.C. & Rudy, T.E. (1986). Assessment of cognitive factors in chronic pain: a worthwhile enterprise?
J Consult Clin Psychol, 54(6), 760–8.
Turk, D.C., Meichenbaum, D. & Genest, M. (1983). Pain and behavioral medicine: A cognitive-behavioral
Turner, J.A. (1996). Educational and behavioral interventions for back pain in primary care. Spine,
21, 2851–9.
Van Damme, S., Legrain, V., Vogt, J. & Crombez, G. (2010). Keeping pain in mind: a motivational account
of attention to pain. Neurosci Biobehav Rev, 34, 204–13.
Van der Windt, D.A.W., Hay, E.M., Jellema, P., & Main, C.J. (2008). Psychosocial interventions for low
back pain in primary care: Lessons learned from recent trials. Spine, 33, 81–9.
Van Tulder, M.W., Ostelo, R., Vlaeyen, J.W.S., Linton, S.J., Morley, S.J. & Assendelft, W.J.J. (2001).
Behavioral treatment for chronic low back pain: A systematic review within the framework of the
Vlaeyen, J.W., de Jong, J., Geilen, M., Heuts, P.H. & Van Breukelen, G. (2001). Graded exposure in vivo in
the treatment of pain-related fear: a replicated single-case experimental design in four patients with
chronic low back pain. Behav Res Ther, 39, 151–66.
Vlaeyen, J.W. & Linton, S.J. (2000). Fear-avoidance and its consequences in chronic musculoskeletal pain:
Waddell, G. (2004). The Back Pain Revolution (2nd. edn.). Edinburgh: Churchill-Livingstone.
Whitfill, T., Haggard, R., Bierner, S.M., Pransky, G., Hassett, R.G. & Gatchel, R.J. (2010). Early
intervention options for acute low back pain patients: a randomized clinical trial with one-year
follow-up outcomes. J Occup Rehabil, 20(2), 256–63.
Part 9
Clinical Approaches for

Patients with Established
Pain and Disability
Chapter 28
Physical Exercise Interventions and

Low Back Pain
J. Bart Staal, Chris G. Maher, and William S. Shaw
28.1 Introduction
Although low back pain (LBP) is usually considered a benign disorder with a good prognosis, a
spontaneous relapse of symptoms and disability is not uncommon (Henschke et al. 2008; Pengel
et al. 2003). Besides the unpredictable course over time, it is generally assumed that the longer the
pain and disability persists, the higher the chances are to develop a chronic stage of pain and dis-
ability with little prospect of a successful recovery (Frank et al. 1996; Pincus et al. 2002, 2006).
Effective strategies are required to prevent such a deteriorated health condition from developing
and to improve the coping skills and resources available to LBP sufferers.
In the literature, LBP is regularly classified according to the duration of symptoms: acute and
subacute LBP constitutes an episode with a duration of less than 12 weeks, and chronic LBP describes
an episode of pain with a duration longer than 12 weeks (Airaksinen et al. 2006; Koes et al. 2001; van
Tulder et al. 2006). The subacute stage has been referred alternately to symptoms within 3–12 weeks
from pain onset or 6–12 weeks from pain onset (Frank et al. 1996; Hayden et al. 2005a). Only in a
minority of cases (approximately 15%) can the pathoanatomical source of LBP be identified. This
leaves the majority of cases in the category of non-specific or uncomplicated LBP (Deyo et al. 1992).
The type of LBP dealt with in this chapter is non-specific, unless stated otherwise.
Many treatment alternatives have been proposed for LBP throughout the last decades even
though the evidence base for many treatments is still weak (Spitzer 1987; Airaksinen et al. 2006).
One of the most promising intervention alternatives is physical exercise. In acute LBP, physical
exercise is generally considered to have no beneficial value compared to no treatment or other
conservative treatments (Hayden et al. 2005a). In cases of subacute or chronic LBP, however,
physical exercises are viewed as a valuable treatment alternative for LBP and exercise has been
recommended in a number of clinical guidelines for subacute and chronic LBP (Chou et al. 2007;
Koes et al. 2001; Airaksinen et al. 2006). Nevertheless, there are still doubts with regard to the
clinical importance of its effect size (van Tulder et al. 2007). Physical exercise treatments in clini-
cal studies may, however, vary with regard to its underlying concepts, the types of exercises
included, the frequency, intensity, and length of exercise sessions, and whether or not exercise is
combined with other treatment components such as patient education or cognitive-behavioural
approaches. This heterogeneity in the content of exercise treatments complicates our understand-
ing of the real effects of physical exercise for LBP. Other aspects that might also influence these
effects are the skills and experience of the caregiver and the exercise adherence and level of effort
by the patient (Helmhout et al. 2008). The interpretation of studies on physical exercises, and LBP
treatment in general, may further be hampered by the variation in the type of chronic LBP patients
who participate in such studies. These patients can vary from those who have episodic LBP that
is merely an annoyance, to those whose LBP prevents work and social participation and is
accompanied by psychological distress (Dufour et al. 2010; Mayer 2010). There is some evidence
that the profile of the patient can influence their response to exercise-based treatment (Haldorsen
et al. 2002). Clear characterization of study populations in intervention studies is therefore very
important.
In this chapter, we review the effects of physical exercise in subacute and chronic LBP. Although
the Cochrane systematic literature review by Hayden et al. (2005) has provided a reasonable
summary of the effects of physical exercise interventions, the present chapter will place more
emphasis on the context of daily clinical practice, and how that may influence exercise outcomes.
To accomplish this, we explored the PubMed and Embase databases for publications in the last
5 years with particular interest in relevant systematic reviews and additional randomized control-
led trials. Based on this literature search, the most common types of exercise therapy will be
described according to concept and content of treatment. For the purpose, we subdivide physical
exercise interventions into the following types of exercise interventions:
◆ Exercises aimed at improving aspects of physical fitness (e.g. strength, muscle endurance,
movement coordination).
◆ Specific spinal stabilization exercises.
◆ McKenzie exercises.
◆ Physical exercises from a cognitive-behavioural perspective (aimed at reducing fear or altering
cognitions).
◆ Physical exercise interventions for disabled workers with LBP (i.e. the occupational healthcare
perspective).
Although it should be noted that there may be some overlap, we believe this classification
corresponds with the current literature and with clinical practice as common in many countries
around the world. In addition, the effects of these subgroups of exercise therapies as reported in
the literature will be summarized and their clinical implications will be discussed.
28.2 Treatment aimed at increasing physical fitness

and/or activity
Many exercise programmes are aimed at increasing aspects of physical fitness such as muscle
strength, muscle cross-sectional area, muscle endurance, and coordination of movements. In gen-
eral, it is believed that the improvement of physical fitness and, consequently, an increase in
habitual physical activity is beneficial for the LBP patient. This may also imply that reduced physi-
cal fitness and physical activity levels play a role in the aetiology of LBP (i.e. the de-conditioning
syndrome). The evidence for this relationship is, however, still weak. In a Norwegian case–control
study, which compared groups of chronic and subacute LBP patients with a group of healthy
controls, it was found that abdominal and back muscle endurance were reduced when comparing
chronic LBP patients to subacute LBP patients, and to healthy controls (Brox et al. 2005). A more
recent systematic review on the longitudinal relationship between physical capacity and the risk
of musculoskeletal disorders concluded that due to inconsistency in multiple studies there is
inconclusive evidence for a relationship between trunk muscle strength, or mobility of the lumbar
spine and the occurrence of LBP (Hamberg-van Reenen et al. 2007). Another systematic review
on the association between physical de-conditioning and chronic LBP concluded that there was
conflicting and limited evidence for the presence of cardiovascular de-conditioning and multi-
fidus disuse in chronic LBP patients (Smeets et al. 2006).
The Cochrane systematic review on the effects of exercise therapy identified five trials in
subacute populations with data available to calculate pooled estimates for pain and four trials for
PHYSICAL EXERCISE INTERVENTIONS 533
self-reported functioning (Hayden et al. 2005a). All trials used non-exercise comparison groups.
The content of the exercise treatments of the included trials varied from extension, flexion and/
or more general exercises (Davies et al. 1979; Moffett et al. 1999; Storheim et al. 2003) to McKenzie
exercises (Cherkin et al. 1998) and to graded activity interventions (Lindstrom et al. 1992; Staal
et al. 2004), which are exercise interventions that follow a cognitive-behavioural approach.
The pooled effects estimates for pain (i.e. 1.9 points at a 100-point scale) and functional status
(i.e. 1.1 point at a 100-point scale) at short- and intermediate-term follow-up were small and
non-significant leading to the conclusion that there was insufficient evidence to support or refute
the effectiveness of exercise therapy in subacute LBP (Hayden et al. 2005a). In a more recent large
(n=259) randomized controlled trial (RCT) by Pengel et al. physiotherapy-directed exercise was
compared with advice and placebo treatment in a population of subacute LBP patients. This
study showed that exercises were significantly, but only slightly, more effective for pain and func-
tioning at 6 weeks when compared to a placebo treatment (Pengel et al. 2007).
The results of exercise treatments in chronic LBP populations are more promising than the
results in subacute populations. The meta-analysis by Hayden et al. found statistically significant
effects of exercise therapy in chronic LBP at short-, intermediate-, and long-term follow-up peri-
ods based on 46 studies. The pooled mean differences in improvement was 7.3 points (of
100 points) for pain and 2.5 points (of 100 points) for function at the earliest follow-up point
and similar effects were found for the intermediate and long-term follow-up (Hayden et al.
2005a). The studies included were mostly conducted in healthcare settings, commonly used indi-
vidually designed and delivered exercise programmes, and often included strengthening or trunk-
stabilizing exercises. Conservative care was frequently added to the exercise programme, including
treatment components such as advice to stay active and education, manual therapy, or even
behavioural therapy (Hayden et al. 2005a). The authors of this review study concluded that exer-
cise therapy seemed to be slightly effective in decreasing pain and improving function in adults
with chronic LBP, particularly in healthcare populations (Hayden et al. 2005a). Similar conclu-
sions were reported in the evidence review which accompanied the European guidelines for
chronic LBP (Airaksinen et al. 2006). The authors found, based on a best-evidence rating system
applied to 45 studies, strong evidence that exercise is more effective than general practitioners’
(GPs’) care for pain and disability reduction, and return-to-work at 3 to 6 months. Moderate
evidence was found when the comparison group included passive treatments. Besides the com-
parison between exercise and GPs’ care or other treatments, the authors also compared different
types of exercise therapies to each other (Airaksinen et al. 2006). In one of these comparisons
muscle reconditioning and strengthening exercises were compared to other types of exercises.
Based on the results of eight randomized studies it was concluded that there was strong evidence
for muscle reconditioning and strengthening exercises being not superior to other types of exer-
cises. In addition, one study provided limited evidence that there are no differences between
aerobic and either reconditioning or conventional physiotherapy exercises (Mannion et al. 1999;
Mannion et al. 2001; Airaksinen et al. 2006).
Based on studies with available data in roughly the same set of studies Hayden et al performed
a meta-regression analysis alongside their meta-analysis to investigate which types of exercise
therapies are most beneficial for chronic LBP patients (Hayden et al. 2005b). The conclusions of
this literature study were slightly different when compared to the European guideline evidence
review. Of the different types of exercise treatments, stretching exercises demonstrated the largest
improvements in function outcomes and muscle strengthening exercises the largest improvement
in pain outcomes (Hayden et al. 2005b). Further, the meta-regression model revealed that the
effects of exercise on pain and function outcomes were greater if the exercise programme was
individually designed and supervised (including home exercises), had a high dose or intensity
(t20 hours), and encompassed an additional conservative care intervention component (Hayden
et al. 2005b). Although these results are interesting and seem to provide tools for daily clinical
practice it should be noted that the effects found for pain and function outcomes were still not
considered large enough to represent a clinically important effect (i.e. 10 and 20 points for pain
and function scores respectively on a 100-point scale) (Hayden et al. 2005a, 2005b).
If exercise programmes are aimed at improving aspects of physical fitness, changes in physical
fitness or performance are expected to be associated with changes in clinical symptoms or disabil-
ity (Airaksinen et al. 2006). Of the 46 studies included in the European evidence review only seven
studies investigated the relationships between pain and disability on the one hand, and spinal
mobility/range of motion, trunk strength or trunk endurance on the other hand (Airaksinen et al.
2006). Among those studies only two studies reported weak associations between pain/disability
and spinal mobility (correlation coefficient r= 0.2–0.4) (Elnaggar et al. 1991; Mannion et al. 1999)
and trunk strength (correlation coefficient r=0.48) (Martin et al. 1986). Although these and other
results (Airaksinen et al. 2006; Smeets et al. 2006) point out there is hardly any relationship
between changes in pain and disability and direct changes in aspects of physical fitness or
performance after engaging in exercise treatments, the research literature suggests that physical
fitness and activity programmes are to some extent effective in chronic LBP patients. There are,
however, still many questions with regards to optimal content of treatment and mechanisms of
action of physical fitness and activity programmes that remain to be answered in future studies.
28.3 Specific spinal stabilization exercises

Specific spinal stabilization exercises are aimed at improving coordination, strength, and endur-
ance of muscles involved in the active stabilization of the spine and trunk. Other terms that are
used for this specific type of exercise therapy are segmental stabilization or core stabilization.
According to Panjabi, spinal stability can be provided by the articular, muscular, and neural
systems which together control intervertebral movement (Panjabi, 1992a, 1992b). The muscles
involved in spinal stability can be subdivided in global and local muscles (Bergmark, 1989;
Richardson and Jull, 1995; Standaert et al. 2008). The global muscles are the large torque produc-
ing muscles which link the pelvis to the thoracic cage. They provide general trunk stabilization
and balance external loads. The local muscles are directly linked to the vertebrae and are assumed
to be responsible for spinal stability (Richardson and Jull 1995). The local muscles targeted in
spinal stabilization programmes are primarily deep spinal muscles, such as the lumbar multifidi
and the transversus abdominis (Richardson and Jull 1995). The recruitment of these muscles is
assumed to be altered in LBP patients and this needs to be enhanced and restored in order to
reach pain reduction and functional recovery (Ferreira et al. 2006; Standaert et al. 2008). According
to an experimental study by Hodges and Richardson (1996) the contraction of the transversus
abdominis was delayed in LBP patients compared to healthy controls when responding to distur-
bances of the spine produced by arm movements. In another study it was shown that lumbar
multifidus wasting was present in patients with acute and subacute LBP (Hides et al. 1994). The
recovery of the multifidus muscle after acute LBP, which did not occur spontaneously, could
be improved by means of specific localized exercise therapy (Hides et al. 1996). The results of
these studies together with the findings of other studies (MacDonald et al. 2006; Standaert et al.
2008) gave support to the hypothesis that inefficient muscular stabilization of the spine is associ-
ated with sustained LBP. Following this model it seems obvious that exercise therapy should
promote coordination and stabilization training of the lumbar spine musculature.
During spinal stabilization programmes patients are taught to recruit their deep spinal muscles.
They are for example asked to gently draw in their abdominal wall in order to achieve isometric
co-contraction of the transversus abdominis and multifidus (Richardson and Jull 1995 ).
Progression can be achieved by increasing the holding time and the number of repetitions, and by
gradually increasing from low loads with minimal body weight to more functional body positions
with higher external loads. Furthermore, the exercise should not only be performed with a static
neutral lumbar spine but also in more extreme static body positions. Finally, co-contraction of
the deep spinal muscles have to be held in dynamic functional movements of the trunk (Richardson
and Jull 1995).
Two systematic reviews were published in 2006 on the effects of spinal stabilization exer-
cises for LBP. Rackwitz et al. (2006) conducted a systematic review on the effects of so-called
segmental spinal stabilization exercises for acute, subacute and chronic LBP and Ferreira et al.
(2006) summarized the effects of spinal stabilization exercises for spinal and pelvic pain and
followed, if possible, a meta-analytical approach. Both systematic reviews were more or less based
on the same set of studies and, not surprisingly, similar conclusions were found. In chronic LBP
spinal stabilization exercises were found to be superior when compared to GPs’ care but the addi-
tional value to general physiotherapeutic management was not clear (Ferreira et al. 2006; Rackwitz
et al. 2006).
The systematic review by Ferreira et al. has recently been updated and now provides the most
up-to-date information regarding the effectiveness of spinal stabilization exercises (Macedo et al.
2009). In this systematic review only trials were included which included participants with per-
sistent non-specific LBP. Persistent LBP was defined as non-acute LBP of at least 6 weeks duration
or recurrent LBP. The follow-up time points were subdivided in short-term follow-up (less than
3 months after randomization), intermediate term follow-up (at least 3 months and less than
12 months after randomization) and long-term follow-up (12 months follow-up or more).
Relevant subgroups were created and pooled analyses were performed with pain, disability, and
quality of life scales converted to 0–100 scales. The authors of the review distinguished four com-
parisons: (1) spinal stabilization versus minimal intervention (i.e. no intervention, general prac-
titioners’ care or education) or versus another type of intervention with spinal stabilization
exercises added as a supplement; (2) spinal stabilization versus spinal manipulative therapy;
(3) spinal stabilization versus other types of exercise therapy; and (4) spinal stabilization versus
surgery (Macedo et al. 2009).
Four trials were identified that compared spinal stabilization exercises with either general prac-
titioners’ care or no intervention (Moseley, 2002; Niemisto et al. 2003; O'Sullivan et al. 1997;
Shaughnessy and Caulfield 2004), and three trials compared spinal stabilization exercises added
as a supplement to another intervention to this intervention alone (Goldby et al. 2006; Koumantakis
et al. 2005; Stuge et al. 2004). The pooled results of these seven trials together showed statistically
significant effects in favour of the spinal stabilization exercises for pain at short-term follow-up
(effect = −14.3 points, 95% confidence interval [CI]: −20.4 to −8.1), intermediate-term follow-up
(effect = 13.6, 95% CI: −22.1 to −4.1), and long-term follow-up (effect = −14.4, 95% CI: −23.1
to −5.7). Significant effects were also found for disability at long-term follow-up (effect = −10.8,
95% CI: −18.7 to −2.8) but no effects for quality of life. In four trials, spinal stabilization exer-
cises were compared with spinal manipulative therapy (Critchley et al. 2007; Ferreira et al.
2007; Goldby et al. 2006; Rasmussen-Barr et al. 2003). In most comparisons the pooled effects
were in favour of spinal stabilization although the effects were small and only reached significance
in a few estimates (Macedo et al. 2009). Significant effects at intermediate term follow-up were
found for pain (effect = −5.7, 95% CI: −10.7 to −0.8), disability (effect = −4.0, 95% CI: −7.6
to −0.4) and quality of life (effect = −6.0, 95% CI: −11.2 to −0.8). Five trials were identified
which compared spinal stabilization exercises with other kinds of exercises (Critchley et al. 2007;
Ferreira et al. 2007; Kladny et al. 2003; Miller et al. 2005; Stevens et al. 2007). A significant pooled
effect, based on the results of three trials, was only found for disability at short-term follow-up
(effect =−5.1, 95% CI: −8.7 to 1.4). No significant differences were found for the other effect
estimates. Finally, one trial compared spinal stabilization exercises with surgery and found no
significant differences between these two treatments (Brox et al. 2003).
The results of the systematic reviews described above indicate that spinal stabilization exercises
may be effective in comparison to standard GPs’ care, no intervention, or as a supplement to
a more general physiotherapeutic management. There is no convincing evidence that spinal
stabilization exercises are more effective than other kinds of exercise interventions, spinal manip-
ulative therapy, or surgery.
28.4 McKenzie therapy

The McKenzie method is a popular treatment for the management of LBP. It includes both an
assessment and a management component (May and Donelson 2008). While commonly consid-
ered as an exercise approach, the method emphasizes education and self-management and for
some patients a lumbar roll is provided. Manual therapy is a less frequently used part of the treat-
ment package. The method was developed by a New Zealand physiotherapist, Robin McKenzie,
based upon his clinical observations of patients with LBP. The method has evolved over time and
in this chapter we will briefly describe the method as described in McKenzie and May’s 2003 texts
(McKenzie and May, 2003a, 2003b).
In common with the approach advocated in clinical practice guidelines, the initial step in the
McKenzie method is to exclude patients with confirmed or suspected serious pathology. Findings
from the history-taking and physical examination are then used to determine the provisional
classification of non-serious LBP into one of three syndromes: the derangement syndrome; the
dysfunction syndrome; or the postural syndrome. LBP presentations that after five visits do not
fit any of the above-mentioned syndromes are classified as ‘other’ or ‘non-mechanical syndrome’
and are considered not suitable for treatment with the method. Unlike most clinical practice
guidelines the method does not distinguish between patients with non-specific LBP and radicu-
lopathy; or advocate a different approach based upon the duration of the episode.
Most patients fit the derangement syndrome and for these patients the guiding treatment prin-
ciple is to encourage directions of movement and postures that produced a proximal shift in pain
(centralization) and that prevented distal migration of pain (peripheralization). It is important to
note that this is not equivalent to generic prescription of extension exercises for all patients. In the
dysfunction syndrome symptoms are thought due to adhesions, contracture, and adaptive short-
ening of soft tissues. Treatment of the dysfunction syndrome therefore emphasizes movements
that encourage the process of remodelling of these tissues. A small percentage of patients seeking
care for LBP fit the postural syndrome and these patients receive advice on the link between
posture and pain, and posture correction. As mentioned previously, self-management is a major
element in all three syndromes.
Most clinical practice guidelines do not provide separate recommendations for various types of
exercise such as the McKenzie method. There are, however, two systematic reviews that have
aimed to determine the effectiveness of McKenzie therapy. The 2004 Clare review (Clare et al.
2004) concluded that McKenzie therapy does result in a greater decrease in pain and disability in
the short term than do other standard therapies. The effect of treatment was, however, quite small
with the weighted mean difference and 95% CI being −8.6 (−13.7 to −3.5) for pain and −5.4 (−8.4
to −2.4) for disability where both outcomes were measured on a 0-100 scale. At intermediate
follow-up the effect for disability was smaller still and not significant and there was no pooled
estimate for pain.
The 2006 Machado review (Machado et al. 2005), similarly to the earlier review, noted that the
effects observed for McKenzie method exercises in trials seemed quite small. A key finding of the
Machado review was that five of the 11 located trials applied treatment in a generic fashion, with
patients not receiving treatment matched to their clinical classification. These authors noted that
this approach potentially underestimates the effect of treatment with the McKenzie method.
Accordingly at the present point the efficacy of McKenzie therapy is unclear.
28.5 Physical exercises from a cognitive-behavioural perspective

In addition to physical exercise, cognitive-behavioural interventions have been highlighted as a
possible treatment alternative for chronic LBP. Recommendations for cognitive-behavioural
treatment are not uncommon (Airaksinen et al. 2006; Chou and Huffman, 2007), and there is
evidence in support of such treatment programmes (Ostelo et al. 2005). Cognitive-behavioural
treatment can be defined as a treatment that focuses on reducing disability through the modifica-
tion of environmental contingencies and cognitive processes (Ostelo et al. 2005). Three behav-
ioural approaches can be distinguished. Operant treatments aim at the modification of behaviour
by positive reinforcement of healthy behaviours and withdrawal of attention towards pain behav-
iours, and time-contingent instead of pain-contingent management of pain patients (Fordyce
1976). Cognitive treatments aim to modify maladaptive cognitions, beliefs and expectations with
regards to pain and disability, and respondent treatment aims to modify physiological responses
(e.g. reducing muscle tension by means of relaxation exercises) (Ostelo et al. 2005).
Although cognitive-behavioural treatment seems, at least conceptually, very different in com-
parison to physical exercise interventions an overlap exists between these two approaches. First,
in many multimodal treatment packages, elements of cognitive-behavioural treatments are com-
bined with physical exercises and sometimes other elements to obtain an optimal content of
intervention (Guzman et al. 2001; Karjalainen et al. 2001). Second, physical exercises are some-
times used as a mean to modify behaviour, cognitions or physiological response and can therefore
be considered in these cases as a form of cognitive-behavioural treatment. For the purpose of this
chapter we will limit the discussion to this combined form of treatment; in particular, operant
and graded exposure treatment that incorporates physical exercise. Graded activity interventions
are examples of operant treatments. According to the operant conditioning theory, as applied to
LBP, the future occurrence of healthy behaviour tends to increase as a result of graded activity (i.e.
gradually increased exercise intensity) and positive reinforcement, in contrast to inactive pain
behaviour (Fordyce 1976; Lindstrom et al. 1992). The graded activity intervention can be described
as an intervention that consists of individual counselling, education, and physical exercises,
applied according to operant conditioning principles. Following this approach, Swedish investi-
gators conducted a RCT, which compared a graded activity intervention to usual care in Swedish
automobile industry (Lindström et al. 1992). The results of the study by Lindström and colleagues
showed that the participants of the graded activity intervention returned to work significantly
earlier than the control group (Lindström et al. 1992).
A similar RCT to Lindström et al (1992) was conducted by Staal et al in a population of Dutch
airline workers (Hlobil et al. 2005; Staal et al. 2004). In this study the graded activity intervention
consisted of two sessions of physical exercises a week, which were supervised by skilled physio-
therapists. The exercises were based on the functional demands of specific work tasks and on
individual functional capacity. Functional capacity was determined for each exercise by a number
of baseline assessments in which the worker exercised to the limit of tolerance. The worker
then set a quota of exercises to be performed at each exercise session. During the course of the
intervention the load of the exercises was gradually increased towards a preset exercise goal,
following a time-contingent exercise scheme. This means that the exercise quota were preset and
independent of variation in the severity of pain. The exercise goals were further connected with
return-to-work (RTW) goals, which were also set by the worker eventually after consulting the
physiotherapist, and the intervention stopped when full return to regular work was achieved
(Hlobil et al. 2005; Staal et al. 2004). The results of the study by Staal et al. were similar to those
reported in the Swedish trial: graded activity resulted in significantly earlier return-to-work com-
pared to the control group. No effects were found for pain and self-reported disability (Hlobil
et al. 2005; Lindstrom et al. 1992; Staal et al. 2004). The study with Dutch airline workers was
followed by a more or less similar study which compared graded activity to a participatory ergo-
nomics intervention and to usual care. In this study, carried out in a heterogeneous group of
workers from different companies, no effects were found for graded activity (Anema et al. 2007;
Steenstra et al. 2006). The study authors attributed differences to the level of implementation and
the large number of participating physiotherapy clinics. A total number of 47 physiotherapists
had been involved in their study who all received additional training on treating pain patients
according to an operant conditioning behavioural approach (Steenstra et al. 2006). However, it
had not been checked to what extent the physiotherapists had the appropriate skills to follow this
approach. In the study by Staal et al. this was regularly done which was in fact also easier due to
the limited number of physiotherapist involved in this trial (Staal et al. 2004).
Two out of the three trials described above found favourable effects on return-to-work caused
by a graded activity intervention. When we look at the evidence in a non-occupational healthcare
setting the picture becomes less promising. Some RCTs, conducted in patients with chronic LBP,
have compared graded activity-like interventions to active physical therapy (Smeets et al. 2008)
and to guideline-based physiotherapy in primary care (van der Roer et al. 2008). Another RCT in
acute and subacute patients found that augmenting physical therapy with either graded activity or
graded exposure did not result in different effects at 6-month follow-up (George et al. 2008). The
effects of graded activity versus usual care have also been investigated in patients recovering from
first-time lumbar disc surgery but no effects were found in this study either (Ostelo et al. 2003).
Graded exposure in vivo is a cognitive-behavioural treatment approach that aims at restoring
function by means of systematically reducing pain-related fear. Patients with LBP have to estab-
lish a personal graded hierarchy of fear-eliciting activities by using a series of photographs of daily
activities (Leeuw et al. 2008). After that, they are gradually and systematically exposed to person-
ally tailored and fear-provoking activities which are derived from the individual fear hierarchy
based on the series of photographs (Leeuw et al. 2008). In an RCT, Woods and Asmundson com-
pared graded exposure to graded activity and to a waiting-list group in a population of chronic
LBP patients (Woods and Asmundson 2008). Significant effects were found in favour of the
graded exposure group for several measures of pain-related fear compared to the other treatment
conditions although not for pain-related disability (Woods and Asmundson 2008). More or less
similar results were found in another RCT which compared graded exposure to graded activity
over 6 months follow-up (Leeuw et al. 2008). No effects were found for symptoms and self-
reported disability but graded exposure appeared to be superior in decreasing pain catastrophiz-
ing and perceived harmfulness (Leeuw et al. 2008).
Although some beneficial effects have been found in occupational settings, the overall evidence
base for graded activity interventions is still not very strong. The same holds for graded exposure
when compared with waiting-list controls and graded activity. Maybe these types of interventions
are especially effective in particular subgroups of patients which need to be selected prior to the
start of the intervention. Though this might seem an attractive hypothesis, it still needs to be
confirmed in future studies.
28.6 Exercise interventions to improve work endurance

A number of exercise intervention trials for LBP have been specifically targeted to affected workers
with the goal of facilitating a return to usual work or improving work endurance. These exercise
programmes have been variously described as work conditioning, work hardening, functional
restoration, physical conditioning, or function-centred rehabilitation. Although the content of
work conditioning programmes has a substantial overlap with other types of exercise programmes,
work conditioning programmes are usually tailored to meet the physical work demands of a spe-
cific job, and the primary goal of intervention is to reduce or prevent occupational disability. This
can be accomplished by analysing job tasks, simulating physical job demands in exercise sessions,
conducting exercise sessions in the workplace environment, advising to reduce awkward pos-
tures, and coaching to deal effectively with lingering or recurrent job discomfort. The impetus for
work conditioning programmes has been to reduce the cost to the employer of sickness absence
and disability associated with LBP, especially when pain is the result of a work injury. In recent
years, however, the line between work conditioning and other exercise programmes has begun to
blur as RTW has become a more prominent outcome in LBP treatment.
An extensive literature review of RCTs for work conditioning programmes among injured
workers was conducted and published by the Cochrane Collaboration Group in 2003 (Schonstein
et al. 2003), and this review captured most of the available trial results published before 2001.
Eighteen RCTs involving work conditioning programmes were identified, and the authors con-
cluded that work conditioning programmes were effective to reduce sick days for workers with
chronic LBP (but not acute LBP) compared with usual care. However, this conclusion only
applied to exercise programmes that: (1) integrated cognitive-behavioural approaches with inten-
sive physical training (aerobic capacity, muscle strength and endurance, coordination); (2) were
in some way work-related; and (3) were given and supervised by either a physiotherapist or a
multidisciplinary team. Thus, there is a reasonable evidence base supporting the use of work con-
ditioning programmes for working adults with chronic LBP, provided that the programme
includes these characteristics. Other reviews published prior to 2003 (Guzman et al. 2001;
Karjalainen et al. 2001; Teasell and Harth, 1996), presented more mixed conclusions, but a meta-
analysis published in 2004 (Kool et al. 2004) also concluded that exercise programmes reduced
sick leave in patients with non-acute, non-specific LBP.
One important result of the 2003 Cochrane Review (Schonstein et al. 2003) was to highlight
differences and similarities among work conditioning programmes described in intervention
trials. Some of the more significant contrasts between exercise programmes included: inpatient
versus outpatient delivery, differences in programme duration (from 1 to 18 months), and differ-
ences in duration of follow-up (from 12 weeks to 12 months). Nearly all work conditioning
programmes included a cognitive-behavioural component (to redirect patient goals toward func-
tional improvement and away from pain relief), and most programmes involved direct patient
supervision by a physiotherapist.
A number of RCTs of work conditioning programmes have been conducted since 2000, and
many of these studies have sought to determine whether such programmes are cost-effective
(Steenstra et al. 2003), how work conditioning programmes compare with other treatments
(Jousset et al. 2004), and which patients are most likely to benefit (Hebert et al. 2008). These
types of research questions have been driven by rapidly rising healthcare costs in many countries,
which has led to an interest in making work conditioning programmes more streamlined or cost-
effective. Other questions raised by researchers are whether benefits of work conditioning
programmes might vary by the timing of intervention, by treatment confidence or other patient
expectations, and by local contextual factors that relate to treatment settings and disability
compensation systems (Staal et al. 2005).
One conclusion that can be drawn from studies of work conditioning programmes is that
sickness absence can be further reduced if exercise is paired with a worksite visit, job modifica-
tion, or other employer-based intervention (Karjalainen et al. 2001). For example, Loisel and
colleagues (Loisel et al. 1997) found that a multidisciplinary work rehabilitation programme for
workers with chronic LBP was more effective when combined with a workplace intervention
employing participatory ergonomic approaches. This study was conducted with workers from
participating employers who had expressed interest in improving disability outcomes. In a graded
activity exercise trial that reduced lost days among airline workers, Staal and colleagues (Staal
et al. 2004) also incorporated a number of workplace elements (return-to- work goal-setting,
therapeutic modified duty, overcoming work avoidant beliefs) to improve outcomes. However,
when workers are randomized from the general population, where employers may vary in their
level of involvement and cooperation, an exercise programme may actually delay recovery. In a
trial for subacute LBP (4–6 weeks absent from work), one study found that randomization to a
graded activity exercise programme actually delayed RTW, while those randomized to a com-
parison workplace intervention had improved RTW outcomes (Anema et al. 2007). In summary,
there is emerging evidence that a worksite visit and the cooperation of a supportive employer
might be necessary to provide an effective bridge between supervised exercise and the realities of
physical job requirements.
Recent studies have also highlighted the need for physiotherapists and other providers to con-
sider psychological and workplace factors that might hinder recovery and the importance of
incorporating appropriate counselling and advice as part of a restorative exercise protocol. Key
workplace factors include heavy physical demands, the inability to modify work, job stress, an
unsupportive workplace, job dissatisfaction, poor expectation of RTW, and fear of re-injury
(Shaw et al. 2009). Key psychological factors include depressed mood, social isolation, pain cata-
strophizing, fear avoidant beliefs, and low self-efficacy for managing pain (Kendall and Burton
2009). While a work conditioning programme may not be sufficient to overcome all of these
potential barriers to RTW, this information may be helpful for physiotherapists to set appropriate
goals and expectations for an individualized work conditioning programme. Research is needed
to improve screening methods of prognostic factors and relate these to specific strategies that
might be employed in a work conditioning programme.
Efforts to identify subsets of patients most likely to benefit from work conditioning programmes
have not been conclusive (van der Hulst et al. 2005). However, there is some evidence that a
worker who believes work will inevitably lead to a more serious re-injury may be a good candidate
for a work conditioning programme (George et al. 2008). This approach allows workers to experi-
ence gradual success in improved functional capacity, and this counters a pre-existing belief
that increased activity will inevitably lead to increased pain. This has been offered as the explana-
tion for work conditioning programmes to show greater improvements in self-reported work
ability and lost days compared with conventional physiotherapy (Kool et al. 2005, 2007; Roche
et al. 2007).
In summary, randomized trials have provided evidence that work conditioning programmes
are effective to reduce sickness absences related to LBP, but it is unclear whether these benefits are
mediated by increases in muscle strength or endurance, by improved coping strategies, or by
global improvements in perceived health and self-efficacy. Work conditioning programmes may
be more effective when they incorporate customized patient education and counselling and
worksite visits.
28.7 Conclusion
Exercise is generally considered a valuable treatment strategy for subacute and chronic LBP. The
study results summarized in this chapter confirm this common thought. Nevertheless, the
presented results need to be interpreted with caution. In general, the effects which have been
found and reported for physical exercises are rather small. The pooled effect estimates as described
in the Cochrane review by Hayden and colleagues for subacute LBP were very small and not sig-
nificant (i.e. 1.9 points at a 100-point scale for pain and 1.1 points at a 100-point scale for func-
tion) and somewhat larger (i.e. 7.3 points at a 100-point scale for pain and 2.5 points at a 100-point
scale for function) but significant for chronic LBP (Hayden et al. 2005a). Although reaching sta-
tistical significance is important, the clinical importance of effect sizes should not be disregarded:
are the effects important from the perspective of the back pain sufferer? Research has been done,
also in the field of LBP, to determine which effect size could be considered clinically important
(van Tulder et al. 2007). Based on earlier clinimetric studies between-group differences were
considered clinically important if the magnitude was 20% or more for pain and 10% or more for
functioning (van Tulder et al. 2007). Van Tulder and colleagues investigated to what extent the
significant effects as reported in the Cochrane review on exercise therapy and LBP reached an
appropriate level of clinical importance. Of the 18 trials in the Cochrane review which reported
significant effects in favour of the exercise group only six reported effect sizes that met these
standards of clinical importance (van Tulder et al. 2007). This implies that it is questionable
whether the effects reported in at least two-third of the trials affect the pain and disability as
perceived by the patient.
We can not conclude that one type of exercise is clearly more beneficial than other types of
exercise since the differences between exercise therapies are either small or non-significant. If
this is true, less intense exercise therapies are to be preferred especially when viewed from a cost-
effectiveness perspective (Heymans et al. 2006). In clinical practice, patient preferences regarding
exercise treatments should not be ignored since expectations, confidence, and satisfaction regard-
ing the chosen treatment strategy can influence treatment success (Helmhout et al. 2008). Another
challenge for back pain researchers is to find out if particular patient groups are more or less likely
to respond to specific exercise programmes (Helmhout et al. 2008; Staal et al. 2008). Subgroup
analyses in RCTs are therefore needed as well as clear characterization of study populations and
exercise treatments which are matched to individual patient characteristics. It has been reported
that a classification approach in order to match treatments to patients resulted in improved dis-
ability when compared with treatment unmatched to this classification (Brennan et al. 2006).
Other studies indicated that patients with a directional preference to movement, had better
results when they exercised in the preferred direction compared to exercises in a random direc-
tion or opposite direction (Long et al. 2004) and that patients with high fear-avoidance beliefs
improve more with exercise treatment addressing fear-avoidance than with standard care (George
et al. 2003). It has also been suggested that patients with relatively low levels of disability and fear-
avoidance beliefs have higher chances of RTW as a result of a graded activity programme (Staal
et al. 2008). Research on clinical prediction rules to predict who is likely to respond to an exercise
programme is however still scarce and inconclusive (Hicks et al. 2005; Teyhen et al. 2007).
Although the evidence is not very strong, physical exercise can still be considered a beneficial
and also safe strategy for LBP (Staal et al. 2005). The exact working mechanisms are not clear yet
and may be related to improved aspects of physical fitness, altered mood, fears and behaviour,
decreased sensitization, or combinations of these different mechanisms. Physical exercise matches
well with an active self-management approach in which patients become more and more
responsible for their back health. For a large majority of patients this could be followed by a
physically active lifestyle with little support of care providers despite eventual still existing pain.
References
Airaksinen, O., Brox, J. I., Cedraschi, C., et al. (2006). Chapter 4. European guidelines for the management
of chronic nonspecific low back pain. Eur Spine J, 15(Suppl 2), S192–300.
Anema, J. R., Steenstra, I. A., Bongers, P. M. et al. (2007). Multidisciplinary rehabilitation for subacute low
back pain: graded activity or workplace intervention or both? A randomized controlled trial. Spine, 32,
291–8.
Bergmark, A. (1989). Stability of the lumbar spine. A study in mechanical engineering. Acta Orthop Scand
Suppl, 230, 1–54.
Brennan, G. P., Fritz, J. M., Hunter, S. J., Thackeray, A., Delitto, A. & Erhard, R. E. (2006). Identifying
subgroups of patients with acute/subacute “non-specific” low back pain: results of a randomized
clinical trial. Spine, 31, 623–31.
Brox, J. I., Sorensen, R., Friis, A. et al. (2003). Randomized clinical trial of lumbar instrumented fusion and
cognitive intervention and exercises in patients with chronic low back pain and disc degeneration.
Spine, 28, 1913–21.
Brox, J. I., Storheim, K., Holm, I., Friis, A. & Reikeras, O. (2005). Disability, pain, psychological factors
and physical performance in healthy controls, patients with sub-acute and chronic low back pain: a
case-control study. J Rehabil Med, 37, 95–9.
Cherkin, D. C., Deyo, R. A., Battie, M., Street, J. & Barlow, W. (1998). A comparison of physical therapy,
chiropractic manipulation, and provision of an educational booklet for the treatment of patients with
low back pain. N Engl J Med, 339, 1021–9.
Chou, R. & Huffman, L. H. (2007). Nonpharmacologic therapies for acute and chronic low back pain: a
review of the evidence for an American Pain Society/American College of Physicians clinical practice
guideline. Ann Intern Med, 147, 492–504.
Chou, R., Qaseem, A., Snow, V., et al. (2007). Diagnosis and treatment of low back pain: a joint clinical
practice guideline from the American College of Physicians and the American Pain Society. Ann Intern
Med, 147, 478–91.
Clare, H. A., Adams, R. & Maher, C. G. (2004). A systematic review of efficacy of McKenzie therapy for
spinal pain. Aust J Physiother, 50, 209–16.
Critchley, D. J., Ratcliffe, J., Noonan, S., Jones, R. H. & Hurley, M. V. (2007). Effectiveness and
cost-effectiveness of three types of physiotherapy used to reduce chronic low back pain disability:
a pragmatic randomized trial with economic evaluation. Spine, 32, 1474–81.
Davies, J. E., Gibson, T. & Tester, L. (1979). The value of exercises in the treatment of low back pain.
Rheumatol Rehabil, 18, 243–7.
Deyo, R. A., Rainville, J. & Kent, D. L. (1992). What can the history and physical examination tell us about
low back pain? JAMA, 268, 760–5.
Dufour, N., Thamsborg, G., Oefeldt, A., Lundsgaard, C., Stender, S. (2010). Treatment of chronic low back
pain: a randomized, clinical trial comparing group-based multidisciplinary biopsychosocial
rehabilitation and intensive individual therapist-assisted back muscle strengthening exercise. Spine,
35, 469–76.
Elnaggar, I. M., Nordin, M., Sheikhzadeh, A., Parnianpour, M. & Kahanovitz, N. (1991). Effects of spinal
flexion and extension exercises on low-back pain and spinal mobility in chronic mechanical low-back
pain patients. Spine, 16, 967–72.
Ferreira, P. H., Ferreira, M. L., Maher, C. G., Herbert, R. D. & Refshauge, K. (2006). Specific stabilisation
exercise for spinal and pelvic pain: a systematic review. Aust J Physiother, 52, 79–88.
Ferreira, M. L., Ferreira, P. H., Latimer, J., et al. (2007). Comparison of general exercise, motor control
exercise and spinal manipulative therapy for chronic low back pain: A randomized trial. Pain, 131, 31–7.
Fordyce, W. (1976). Behavioral methods for chronic pain and illness. St. Louis, MO: CV Mosby.
Frank, J. W., Brooker, A. S., Demaio, S. E., et al. (1996). Disability resulting from occupational low back
pain. Part II: What do we know about secondary prevention? A review of the scientific evidence on
prevention after disability begins. Spine, 21, 2918–29.
George, S. Z., Fritz, J. M., Bialosky, J. E. & Donald, D. A. (2003). The effect of a fear-avoidance-based
physical therapy intervention for patients with acute low back pain: results of a randomized clinical
trial. Spine, 28, 2551–60.
George, S. Z., Fritz, J. M. & Childs, J. D. (2008). Investigation of elevated fear-avoidance beliefs for patients
with low back pain: a secondary analysis involving patients enrolled in physical therapy clinical trials.
J Orthop Sports Phys Ther, 38, 50–8.
Goldby, L. J., Moore, A. P., Doust, J. & Trew, M. E. (2006). A randomized controlled trial investigating the
efficiency of musculoskeletal physiotherapy on chronic low back disorder. Spine, 31, 1083–93.
Guzman, J., Esmail, R., Karjalainen, K., et al (2001). Multidisciplinary rehabilitation for chronic low back
pain: systematic review. BMJ, 322, 1511–16.
Haldorsen, E. M., Grasdal, A. L., Skouen, J. S., Risa, A. E., Kronholm, K. & Ursin, H. (2002). Is there a
right treatment for a particular patient group? Comparison of ordinary treatment, light
multidisciplinary treatment, and extensive multidisciplinary treatment for long-term sick-listed
employees with musculoskeletal pain. Pain, 95, 49–63.
Hamberg-Van Reenen, H. H., Ariens, G. A., Blatter, B. M., Van Mechelen, W. & Bongers, P. M. (2007).
A systematic review of the relation between physical capacity and future low back and neck/shoulder
pain. Pain, 130, 93–107.
Hayden, J. A., Van Tulder, M. W., Malmivaara, A. V. & Koes, B. W. (2005a). Meta-analysis: exercise
therapy for nonspecific low back pain. Ann Intern Med, 142, 765–75.
Hayden, J. A., Van Tulder, M. W. & Tomlinson, G. (2005b). Systematic review: strategies for using exercise
therapy to improve outcomes in chronic low back pain. Ann Intern Med, 142, 776–85.
Hebert, J., Koppenhaver, S., Fritz, J. & Parent, E. (2008). Clinical prediction for success of interventions for
managing low back pain. Clin Sports Med, 27, 463–79, ix–x.
Helmhout, P. H., Staal, J. B., Maher, C. G., Petersen, T., Rainville, J. & Shaw, W. S. (2008). Exercise
therapy and low back pain: insights and proposals to improve the design, conduct, and reporting of
clinical trials. Spine, 33, 1782–8.
Henschke, N., Maher, C. G., Refshauge, K. M., et al. (2008). Prognosis in patients with recent onset low
back pain in Australian primary care: inception cohort study. BMJ, 337, a171.
Heymans, M. W., De Vet, H. C., Bongers, P. M., Knol, D. L., Koes, B. W. & Van Mechelen, W. (2006). The
effectiveness of high-intensity versus low-intensity back schools in an occupational setting: a pragmatic
randomized controlled trial. Spine, 31, 1075–82.
Hicks, G. E., Fritz, J. M., Delitto, A. & McGill, S. M. (2005). Preliminary development of a clinical
prediction rule for determining which patients with low back pain will respond to a stabilization
exercise program. Arch Phys Med Rehabil, 86, 1753–62.
Hides, J. A., Stokes, M. J., Saide, M., Jull, G. A. & Cooper, D. H. (1994). Evidence of lumbar multifidus
muscle wasting ipsilateral to symptoms in patients with acute/subacute low back pain. Spine, 19, 165–72.
Hides, J. A., Richardson, C. A. & Jull, G. A. (1996). Multifidus muscle recovery is not automatic after
resolution of acute, first-episode low back pain. Spine, 21, 2763–9.
Hlobil, H., Staal, J. B., Twisk, J., et al. (2005). The effects of a graded activity intervention for low back pain
in occupational health on sick leave, functional status and pain: 12-month results of a randomized
controlled trial. J Occup Rehabil, 15, 569–80.
Jousset, N., Fanello, S., Bontoux, L., et al. (2004). Effects of functional restoration versus 3 hours per week
physical therapy: a randomized controlled study. Spine, 29, 487–93; discussion 494.
Karjalainen, K., Malmivaara, A., Van Tulder, M., et al. (2001). Multidisciplinary biopsychosocial
rehabilitation for subacute low back pain in working-age adults: a systematic review within the
framework of the Cochrane Collaboration Back Review Group. Spine, 26, 262–9.
Kendall, N. A., Burton, A.K. (2009). Tackling musculoskeletal problems: a guide for clinic and workplace.
Identifying obstacles using the psychosocial flags framework. Norwich: TSO Information and
Publishing Solutions (The Stationery Office).
Kladny, B., Fischer, F. C. & Haase, I. (2003). [Evaluation of specific stabilizing exercise in the treatment of
low back pain and lumbar disk disease in outpatient rehabilitation]. Z Orthop Ihre Grenzgeb, 141, 401–5.
Koes, B. W., Van Tulder, M. W., Ostelo, R., Burton, A. & Waddell, G. (2001). Clinical guidelines for the
management of low back pain in primary care: an international comparison. Spine, 26, 2504–13;
discussion 2513–14.
Kool, J., De Bie, R., Oesch, P., Knusel, O., Van den Brandt, P. & Bachmann, S. (2004). Exercise reduces
sick leave in patients with non-acute non-specific low back pain: a meta-analysis. J Rehabil Med, 36,
49–62.
Kool, J. P., Oesch, P. R., Bachmann, S., Knuesel, O., et al. (2005). Increasing days at work using
function-centered rehabilitation in nonacute nonspecific low back pain: a randomized controlled trial.
Arch Phys Med Rehabil, 86, 857–64.
Kool, J., Bachmann, S., Oesch, P., et al. (2007). Function-centered rehabilitation increases work days in
patients with nonacute nonspecific low back pain: 1-year results from a randomized controlled trial.
Arch Phys Med Rehabil, 88, 1089–94.
Koumantakis, G. A., Watson, P. J. & Oldham, J. A. (2005). Trunk muscle stabilization training plus general
exercise versus general exercise only: randomized controlled trial of patients with recurrent low back
pain. Phys Ther, 85, 209–25.
Leeuw, M., Goossens, M. E., Van Breukelen, G. J., et al. (2008). Exposure in vivo versus operant graded
activity in chronic low back pain patients: results of a randomized controlled trial. Pain, 138, 192–207.
Lindström, I., Ohlund, C., Eek, C., Wallin, L., et al. (1992). The effect of graded activity on patients with
subacute low back pain: a randomized prospective clinical study with an operant-conditioning
behavioral approach. Phys Ther, 72, 279–90.
Loisel, P., Abenhaim, L., Durand, P., et al. (1997). A population-based, randomized clinical trial on back
pain management. Spine, 22, 2911–18.
Long, A., Donelson, R. & Fung, T. (2004). Does it matter which exercise? A randomized control trial of
exercise for low back pain. Spine, 29, 2593–602.
Macdonald, D. A., Moseley, G. L. & Hodges, P. W. (2006). The lumbar multifidus: does the evidence
support clinical beliefs? Man Ther, 11, 254–63.
Machado, L. A., Maher, C. G., Herbert, R. D., Clare, H. & Mcauley, J. (2005). The McKenzie Method for
the management of acute non-specific low back pain: design of a randomised controlled trial
[ACTRN012605000032651]. BMC Musculoskelet Disord, 6, 50.
Macedo, L. G., Maher, C. G., Latimer, J. & Mcauley, J. H. (2009). Motor control exercise for persistent,
nonspecific low back pain: a systematic review. Phys Ther, 89, 9–25.
Mannion, A. F., Muntener, M., Taimela, S. & Dvorak, J. (1999). A randomized clinical trial of three active
therapies for chronic low back pain. Spine, 24, 2435–48.
Mannion, A. F., Muntener, M., Taimela, S. & Dvorak, J. (2001). Comparison of three active therapies for
chronic low back pain: results of a randomized clinical trial with one-year follow-up. Rheumatology
(Oxford), 40, 772–8.
Martin, P. R., Rose, M. J., Nichols, P. J., Russell, P. L. & Hughes, I. G. (1986). Physiotherapy exercises for
low back pain: process and clinical outcome. Int Rehabil Med, 8, 34–8.
May, S. & Donelson, R. (2008). Evidence-informed management of chronic low back pain with the
McKenzie method. Spine J, 8, 134–41.
Mayer, T.G. (2010). Point of view. Spine, 35, 477.
Mckenzie, R. & May, S. (2003a). The lumbar spine: Mechanical Diagnosis & Therapy, vol. 1. Waikanae:
New Zealand Spinal Publications.
Mckenzie, R. & May, S. (2003b). The lumbar spine: Mechanical Diagnosis & Therapy, vol. 2. Waikanae:
New Zealand, Spinal Publications.
Miller, E., Schenk, R., Karnes, J. & Rousselle, J. (2005). A comparison of the McKenzie approach to a
specific spine stabilization program for chronic low back pain. The Journal of Manipulative and Manual
Therapy, 13, 103–12.
Moffett, J. K., Torgerson, D., Bell-Syer, et al. (1999). Randomised controlled trial of exercise for low back
pain: clinical outcomes, costs, and preferences. BMJ, 319, 279–83.
Moseley, L. (2002). Combined physiotherapy and education is efficacious for chronic low back pain. Aust
J Physiother, 48, 297–302.
Niemisto, L., Lahtinen-Suopanki, T., Rissanen, P., Lindgren, K. A., Sarna, S. & Hurri, H. (2003).
A randomized trial of combined manipulation, stabilizing exercises, and physician consultation
compared to physician consultation alone for chronic low back pain. Spine, 28, 2185–91.
O'Sullivan, P. B., Phyty, G. D., Twomey, L. T. & Allison, G. T. (1997). Evaluation of specific stabilizing
exercise in the treatment of chronic low back pain with radiologic diagnosis of spondylolysis or
spondylolisthesis. Spine, 22, 2959–67.
Ostelo, R. W., De Vet, H. C., Vlaeyen, J. W., et al. (2003). Behavioral graded activity following first-time
lumbar disc surgery: 1-year results of a randomized clinical trial. Spine, 28, 1757–65.
Ostelo, R. W., Van Tulder, M. W., Vlaeyen, J. W., Linton, S. J., Morley, S. J. & Assendelft, W. J.
(2005). Behavioural treatment for chronic low-back pain. Cochrane Database Syst Rev, 1,
CD002014.
Panjabi, M. M. (1992a). The stabilizing system of the spine. Part I. Function, dysfunction, adaptation, and
enhancement. J Spinal Disord, 5, 383–9.
Panjabi, M. M. (1992b). The stabilizing system of the spine. Part II. Neutral zone and instability hypothesis.
J Spinal Disord, 5, 390–6.
Pengel, L. H., Herbert, R. D., Maher, C. G. & Refshauge, K. M. (2003). Acute low back pain: systematic
review of its prognosis. BMJ, 327, 323.
Pengel, L. H., Refshauge, K. M., Maher, C. G., Nicholas, M. K., Herbert, R. D. & McNair, P. (2007).
Physiotherapist-directed exercise, advice, or both for subacute low back pain: a randomized trial.
Ann Intern Med, 146, 787–96.
Pincus, T., Burton, A. K., Vogel, S. & Field, A. P. (2002). A systematic review of psychological factors as
Pincus, T., Vogel, S., Burton, A. K., Santos, R. & Field, A. P. (2006). Fear avoidance and prognosis in back
pain: a systematic review and synthesis of current evidence. Arthritis Rheum, 54, 3999–4010.
Rackwitz, B., De Bie, R., Limm, H., Von Garnier, K., Ewert, T. & Stucki, G. (2006). Segmental stabilizing
exercises and low back pain. What is the evidence? A systematic review of randomized controlled trials.
Clin Rehabil, 20, 553–67.
Rasmussen-Barr, E., Nilsson-Wikmar, L. & Arvidsson, I. (2003). Stabilizing training compared with
manual treatment in sub-acute and chronic low-back pain. Man Ther, 8, 233–41.
Richardson, C. A. & Jull, G. A. (1995). Muscle control-pain control. What exercises would you prescribe?
Man Ther, 1, 2–10.
Roche, G., Ponthieux, A., Parot-Shinkel, E., et al. (2007). Comparison of a functional restoration program
with active individual physical therapy for patients with chronic low back pain: a randomized
controlled trial. Arch Phys Med Rehabil, 88, 1229–35.
Schonstein, E., Kenny, D. T., Keating, J. & Koes, B. W. (2003). Work conditioning, work hardening
and functional restoration for workers with back and neck pain. Cochrane Database Syst Rev, 1,
CD001822.
Shaughnessy, M. & Caulfield, B. (2004). A pilot study to investigate the effect of lumbar stabilisation
exercise training on functional ability and quality of life in patients with chronic low back pain. Int
J Rehabil Res, 27, 297–301.
Shaw, W., Van der Windt, D., Main, C., Loisel, P., Linton, S. & the “Decade of the Flags” Working Group.
(2009). Early patient screening and intervention to address individual-level occupational factors (“blue
flags”) in back disability. J Occup Rehabil, 19, 64–80.
Smeets, R. J., Wade, D., Hidding, A., Van Leeuwen, P. J., Vlaeyen, J. W. & Knottnerus, J. A. (2006). The
association of physical deconditioning and chronic low back pain: a hypothesis-oriented systematic
review. Disabil Rehabil, 28, 673–93.
Spitzer, W. O., Leblanc, F.E., Dupuis, M. (1987). Scientific approach to the assessment and management of
activity-related spinal disorders. A monograph for clinicians. Report of the Quebec Task force on
Spinal disorders. Spine, 12, 1–59.
Staal, J. B., Hlobil, H., Twisk, J. W., Smid, T., Koke, A. J. & Van Mechelen, W. (2004). Graded activity for
low back pain in occupational health care: a randomized, controlled trial. Ann Intern Med, 140, 77–84.
Staal, J. B., Rainville, J., Fritz, J., Van Mechelen, W. & Pransky, G. (2005). Physical exercise interventions
to improve disability and return to work in low back pain: current insights and opportunities for
improvement. J Occup Rehabil, 15, 491–505.
Staal, J. B., Hlobil, H., Koke, A. J., Twisk, J. W., Smid, T. & Van Mechelen, W. (2008). Graded activity for
workers with low back pain: who benefits most and how does it work? Arthritis Rheum, 59, 642–9.
Standaert, C. J., Weinstein, S. M. & Rumpeltes, J. (2008). Evidence-informed management of chronic low
back pain with lumbar stabilization exercises. Spine J, 8, 114–20.
Steenstra, I. A., Anema, J. R., Bongers, P. M., De Vet, H. C. & Van Mechelen, W. (2003). Cost effectiveness
of a multi-stage return to work program for workers on sick leave due to low back pain, design of a
population based controlled trial [ISRCTN60233560]. BMC Musculoskelet Disord, 4, 26.
Steenstra, I. A., Anema, J. R., Bongers, P. M., De Vet, H. C., Knol, D. L. & Van Mechelen, W. (2006).
The effectiveness of graded activity for low back pain in occupational healthcare. Occup Environ Med,
63, 718–25.
Stevens, V., Grombez, G. & Parleviet, T. (2007). The effectiveness of specific exercise therapy versus device
exercise therapy in the treatment of chronic low back pain patients. 6th Interdisciplinary World
Congress of Low Back and Pelvic Pain. Barcelona.
Storheim, K., Brox, J. I., Holm, I., Koller, A. K. & Bo, K. (2003). Intensive group training versus cognitive
intervention in sub-acute low back pain: short-term results of a single-blind randomized controlled
trial. J Rehabil Med, 35, 132–40.
Stuge, B., Laerum, E., Kirkesola, G. & Vollestad, N. (2004). The efficacy of a treatment program focusing
on specific stabilizing exercises for pelvic girdle pain after pregnancy: a randomized controlled trial.
Spine, 29, 351–9.
Teasell, R. W. & Harth, M. (1996). Functional restoration. Returning patients with chronic low back pain
to work – revolution or fad? Spine, 21, 844–7.
Teyhen, D. S., Flynn, T. W., Childs, J. D. & Abraham, L. D. (2007). Arthrokinematics in a subgroup of
patients likely to benefit from a lumbar stabilization exercise program. Phys Ther, 87, 313–25.
Van der Hulst, M., Vollenbroek-Hutten, M. M. & IJzerman, M. J. (2005). A systematic review of
sociodemographic, physical, and psychological predictors of multidisciplinary rehabilitation-or, back
school treatment outcome in patients with chronic low back pain. Spine, 30, 813–25.
Van der Roer, N., Van Tulder, M., Barendse, J., Knol, D., Van Mechelen, W. & De Vet, H. (2008).
Intensive group training protocol versus guideline physiotherapy for patients with chronic low back
pain: a randomised controlled trial. Eur Spine J, 17, 1193–200.
Van Tulder, M., Becker, A., Bekkering, T., et al. (2006). Chapter 3. European guidelines for the
management of acute nonspecific low back pain in primary care. Eur Spine J, 15(Suppl 2), S169–91.
Van Tulder, M., Malmivaara, A., Hayden, J. & Koes, B. (2007). Statistical significance versus clinical
importance: trials on exercise therapy for chronic low back pain as example. Spine, 32, 1785–90.
Woods, M. P. & Asmundson, G. J. (2008). Evaluating the efficacy of graded in vivo exposure for the
treatment of fear in patients with chronic back pain: a randomized controlled clinical trial. Pain,
136, 271–80.
Chapter 29
Contextual Cognitive-Behavioural
Therapy for Chronic Pain
(Including Back Pain)
Lance M. McCracken
29.1 Introduction
Although many of the distressing and unwanted experiences of chronic pain are controllable to
some extent or other, this is in most cases incomplete, and some attempts to control these experi-
ences can themselves directly cause problems. It is possibly a cruel irony that many ways to
attempt to reduce pain, in conditions of chronic pain, do not improve daily living but curtail it.
This problem is illustrated in effects of prolonged rest, withdrawal from activity, or pain-related
retirement. It is also illustrated in expense, time, and hope spent in the search for new treatments,
and in adverse effects from medications. The choice to try to avoid pain typically includes costs.
These often include reduced participation in what would otherwise be important activities and
experiences for people with pain. Recent theoretical developments help us to understand how
behaviour patterns can become dominated by experiences that ostensibly ‘need’ to be avoided,
modified, or controlled, rather than guided by experiences of personal goals or values, and these
theoretical developments are yielding new directions for treatment.
29.2 The history of behavioural and cognitive-behavioural

approaches to chronic pain
There has been a natural and healthy evolution within the cognitive and behavioural therapies
both for chronic pain and outside of work on chronic pain. It is worth reviewing this evolution.
If we better understand the course this is taking, the work we do may better align with what is
progressive in this process.
More than 40 years ago it was suggested that we could do better than to look at chronic pain as
either a physical problem or a psychogenic problem and that we could take a direct interest in
patient functioning and behaviour. It was further suggested that we could do this without requir-
ing that we understand specific sources of the pain inside the skin, in the structures or functions
of tissues or organs. This approach specifically called for a focus on patient behaviour, learning
principles, and social circumstances as a means for understanding and for reducing the suffering
and disability of those with chronic pain. This was called the ‘operant behavioural’ approach
(Fordyce 1976). In turn, roughly 25 years ago, an expansion of this focus was called for, an expan-
sion including patients’ interpretations, beliefs, other cognitive processes, and pain coping strate-
gies. This approach was referred to as ‘cognitive-behavioural’ or as cognitive-behavioural therapy
(CBT; Turk et al. 1983).
During the four decades of developments in treatment for chronic pain a series of other devel-
opments also have occurred. These either were simply little discussed in relation to chronic pain,
perhaps because they did not register as relevant during earlier times, or had not occurred until
very recently. CBT for chronic pain had considerable early success and there may have been little
reason to look elsewhere to improve upon the results obtained (e.g. Flor et al. 1992). These devel-
opments that were happening but not intersecting with the mainstream of CBT for chronic pain
include, for example, mindfulness (Kabat-Zinn 1990 ), functional analytic psychotherapy
(Kohlenberg and Tsai 1991), dialectical behaviour therapy (Linehan 1993), acceptance and com-
mitment therapy (ACT, Hayes et al. 1999a), and, more recently, updated versions of traditional
methods such as behavioural activation (e.g. Martell et al. 2004; Dimidjian et al. 2006), and an
application of ACT for chronic pain called contextual cognitive-behavioural therapy (CCBT;
McCracken et al. 2005). The chapter focuses on some of these developments, their implications
for treatment of chronic pain, and research results that have accumulated in their favour.
29.3 What is CCBT?

CCBT is a version of Acceptance and Commitment Therapy (ACT) that developed as an interdis-
ciplinary, primarily group-based, treatment format for chronic pain. It is an approach that arises
from within the wider family of CBT for chronic pain (McCracken 2005a). It includes a broad,
deep, and theoretically integrated conceptual model and adds a number of distinctive therapeutic
processes to those that are currently available within CBT. Its theoretical foundation is what is called
functional contextualism (Hayes et al. 1999a), an approach that is pragmatic, non-mechanistic,
behaviourally-focused, and cognitive. CCBT integrates the differing emphases from the operant
and cognitive behavioural approaches so that its analyses wholly consider the environment, overt
behaviour, cognition, and emotion. The ‘environment’ within CCBT, however, is defined not
merely as the ‘outside world’ around behaviour. It is defined with two dimensions, historical and
situational, and is referred to as ‘context’. Behaviour arises in situations due to the influences
exerted by events present in these situations by virtue of the individual’s unique history. Another
way to say this is that a situation cannot be understood in its influence over behaviour except in
relation to the history that gives the situation that influence or, loosely, its ‘meaning’. This context
is also ‘transdermal,’ it crosses the barrier of the skin, as elements of context invariably involve
sensations in the body or content of cognitive responses, such as thoughts or images.
From a functional contextual standpoint cognitive and emotional experiences are considered
in a particular fashion, not based on whether they have a maladaptive form or structure, but
based on how they are coordinated with individual responses and patterns of behaviour. Here it
is the influence exerted by thoughts and feelings that is of primary concern and not whether
these feel uncomfortable or specify frightening or unwanted experiences. Thus, behaviour is
considered contextually determined through relations with events in the environment, both
inside and outside the skin, relations determined by the individual’s history, most particularly
in the form or their learning history, but also including their cultural and genetic history. This
functional contextual philosophy entails a certain model of truth, such as the truth of a model
of cause and effect. Hypotheses are not considered true or false in an ultimate sense, as in the
type of truth that might exist as a reality waiting to be uncovered in nature. The model of truth
here considers ‘workability’ as the criteria. An analysis is true when it effectively meets the goals
laid out for the analyses. This is called a ‘pragmatic’ approach to truth as distinguishable from
an objectivist or rationalist approach. This may seem an esoteric or unnecessary distinction but
the spirit of it infuses both contextual behavioural science and therapy methods that emerge
from it.
CONTEXTUAL COGNITIVE-BEHAVIOURAL THERAPY 549
Within the CCBT framework people suffer largely from normal processes of thinking and lan-
guage, from how these interact with direct experience, and from the ways these interactions
engender psychological inflexibility (Hayes et al. 1999a; McCracken 2005). The processes under-
lying psychological inflexibility include experiential avoidance, cognitive fusion, values-failures,
loss of contact with present moment, preoccupation with self-as-content, and failures of commit-
ted action (Hayes et al. 2006). The parallel therapeutic processes in relation to these include
acceptance, cognitive defusion, values clarification and values-based action, moment-to-moment
awareness, contact with self-as-context, and committed action. These processes are not wholly
independent from each other, and they substantially overlap in the qualities they add to behav-
iour patterns. The purpose of this chapter is to review these processes underlying CCBT and
summarize data supporting their applicability to chronic pain management.
29.4 Acceptance of chronic pain

We provided a number of functionally similar definitions of acceptance of chronic pain in the
past (e.g. McCracken 1998; McCracken et al. 2004). In essence it is a willingness to actively con-
tact pain. It is a quality of behaviour that is realistic, flexible, practical, and free from unnecessary
restrictions from pain. It is the engagement in activity with pain present plus an absence of
attempts to control or avoid pain (McCracken et al. 2004). The most commonly used measure of
acceptance of pain is the Chronic Pain Acceptance Questionnaire (CPAQ; Geiser 1992; McCracken
et al. 2004; Volwes et al. 2008).
Currently there are more than 35 published studies of acceptance of pain including experi-
ments (e.g. Hayes et al. 1999b; Gutierrez et al. 2004; Masedo and Esteve 2007), clinical studies
(e.g. McCracken 1998; McCracken and Eccleston 2003; Viane et al. 2003; McCracken et al. 2004,
McCracken and Eccleston 2005; Cheung et al. 2008; Mason et al. 2008), and treatment outcome
studies (Dahl et al. 2004; McCracken et al. 2005, 2007b; Volwes and McCracken 2008; Wicksell
et al. 2008). All of these studies support the role of acceptance as a process that appears to lessen
the impact of pain on behaviour. In the clinical studies employing both retrospective and pro-
spective designs, acceptance of chronic pain has been shown to significantly correlate with patient
physical, social, and emotional functioning, work status, and medication use. These relations
hold independent of pain and relevant patient background variables, such as gender, age, educa-
tion, and pain duration. It has been demonstrated that, in relation to key aspects of patient
functioning, such as anxiety, depression, and physical and psychosocial disability, a measure
of acceptance of chronic pain is a significantly better predictor than a commonly used measure of
coping with pain (McCracken and Eccleston 2003; McCracken and Eccleston 2006), a measure of
attention and vigilance to pain (McCracken 2007), and about equal to a measure of catastrophiz-
ing (Vowles et al. 2007).
It is possible to summarize the magnitude of the relations between acceptance of pain as meas-
ured by the CPAQ and measures of patient functioning. This can be done, for example, by select-
ing a number of studies that used similar measure of patient functioning and combining their
findings. Four studies based on entirely independent samples were readily identified meeting
these criteria (McCracken and Eccleston 2003; Nicholas and Asghari 2006; McCracken et al. 2007;
Cheung et al. 2008). Table 29.1 shows the results from combining the correlation analyses of these
studies. The mean correlations from these analyses (combined N=811) between the CPAQ total
score and measures of anxiety, depression, physical disability, and psychosocial disability, were
r=0.59, r=0.60, r=0.39, and r=0.52, respectively.
The constituent parts of acceptance as measured by the CPAQ, activity engagement and pain
willingness, have been questioned with the suggestion that the pain willingness component is
Table 29.1 Summary of correlation studies of acceptance of pain as measured with the CPAQ in
relation to measure of patient, emotional, physical, and psychosocial functioning
Domain of Functioning Range r Mean r

Anxiety −0.44 to −0.72 −0.60
Depression −0.56 to −0.66 −0.59
Physical Disability −0.26 to −0.47 −0.39
Psychosocial disability −0.42 to −0.63 −0.52
Note: these data based on four independent studies, total N=811.
possibly unnecessary (Nicholas and Asghari 2006). Nonetheless, there are conceptual, empirical,
and clinical bases for continued inclusion of both components (McCracken et al. 2007). In a
recent study the dual factor structure of the CPAQ was replicated in a large sample (N=333) and
confirmed in a second independent sample (N=308; Vowles et al. 2008). It was also discovered
based on cluster analyses that most patients seeking treatment for chronic pain have relatively
similar scores on both subscales of the CPAQ. However, in these data there was a large group
(45% of the study sample) who were relatively high on one scale, activity engagement, and low on
the other, pain willingness. These patients represent a group who essentially report they are active
but they also report they must eliminate their pain. This group is interesting because this presen-
tation is somewhat discordant. One might speculate that they are presenting their life as intact
even though it is not, such as to portray a positive appearance. Possibly their life is intact in many
ways but their pain experience is weighing heavily upon them mentally and emotionally. This
latter group may be important to identify as their treatment needs may be very specific. Perhaps
they will not require behavioural activation strategies or physical rehabilitation but some of
psychological work do address a different range of mental or emotional impacts of pain. Other
scenarios are possible as well. Perhaps these patients are the proverbial ‘activity cyclers’ or those
who chronically overdo activity (e.g. McCracken and Samuel 2007) (reviewed in this volume in
Chapters 15 and 27).
Acceptance can be open to a number of varying definitions and can be overall a difficult process
to understand. For example, it is easy to assume that acceptance is a form of belief or way of think-
ing, such as a belief that pain is not controllable or the belief that it will be a permanent experi-
ence. It also has been likened to a form of positive thinking or belief, such as the belief that one
can function with pain (Nicholas and Asghari 2006). From the patient’s perspective, acceptance
may have a negative quality or may be equated with giving up or quitting. In our applications of
the term, it is not intended to mean any of these things. According to a particularly functional
contextual definition it is a process that includes thinking and believing but is not determined
solely by the content of thinking and believing. It is not an exclusively mental process but a qual-
ity of whatever action is taken in relation to pain. It is the quality of not struggling with it, of
allowing room in experience for it, and being willing to have it present. Whether the person
experiencing pain thinks or believes in these processes is another matter. It is not a global act of
quitting but selective actions of quitting responses to pain that take pain as a barrier to function-
ing or as an experience that must change for healthy functioning to proceed. When acceptance is
present in behaviour contact with pain coordinates awareness, observing, or flexible responding;
not avoiding, wrestling, bracing or restricted responding. Positive thinking or optimistic beliefs
may achieve an ostensibly similar behaviour pattern, but this may be a more limited version of the
behaviour pattern, one that depends on the content of thoughts and beliefs and is hostage to the
capriciousness of these.
29.5 Cognitive de-fusion

As humans we have at least one ability that other animals do not have: we can experience events
located in other places and times as if they are present. When we do this we then respond accord-
ingly, with the emotional experiences and actions occasioned by the influences these events hold,
again, even though the events themselves are not present. It is our ability to speak and think and
‘listen’ to the content of speaking and thinking that makes this so. As a result much of the time we
do not have direct contact with experiences around us in life but only with verbally constructed
or modified versions of these experiences. Our day-to-day contact with the world around us is
constantly filtered and our psychological experiences altered through our interpretations, evalu-
ations, thoughts, and beliefs. ‘Cognitive fusion’ is the description of this process by which our
thoughts become merged and undifferentiated from the events they describe or the person who
has them (Hayes et al. 1999a).
Most versions of CBT include work with thoughts and beliefs in some way, whether that be, for
example, a focus on identifying and restructuring irrational beliefs (Ellis 1994), or on disputing or
altering automatic thoughts and schemas (Beck 1976). There are, however, many forms of CBT,
and these focus on differing levels, forms, or processes of cognition and can apply differing meth-
ods for addressing these. This has led to the criticism that CBT lacks a unifying theory of change
(David and Szentagotai 2006). In fact, studies of CBT in patients with anxiety and depression
have questioned whether changes in dysfunctional attitudes are primarily responsible for benefits
of CBT (Burns and Spangler 2001) and whether methods directed at change in automatic thoughts
or core schemas are necessary for positive treatment outcomes (Jacobson et al. 1996; Dimidjian
et al. 2006). A comprehensive review of component analysis and mediation studies of CBT found
little evidence for the role of cognitive change as causal in the symptom improvements achieved
in CBT (Longmore and Worrell 2007).
The model of cognition and cognitive processes underlying CCBT is based directly on the
approach from ACT (Hayes et al. 1999a). Within this approach cognitive processes are consid-
ered at two levels, the level of content, including the form or frequency of thoughts, and the level
of function or context, including the situation that determines the impact thoughts exert on other
responses. While treatment methods may include a focus on either level, a relative emphasis in
CCBT and ACT is placed on contextual change. Notice, however, that when treatment methods
imply that cognitive content determines behaviour and that cognitive change is necessary for
behaviour change, this increases the behaviour regulatory influences of thoughts, which, ironi-
cally, holds the potential to create greater difficulties.
A core therapeutic process in ACT is what is called ‘cognitive defusion’. This process entails a
loosening of the influence of thought content on behaviour and increasing contact between
behaviour and potential influences beyond thought content. Through methods that promote
cognitive defusion, the chronic pain sufferer becomes more aware of the process of thinking, and
less entangled in the content of thinking. From this process they can respond more flexibly while
in contact with otherwise distressing, discouraging, or restricting thoughts. Cognitive defusion
means a weakening of the role of thoughts as the sole or overwhelming basis for action without
necessarily reformulating the content of thoughts. This does not mean blocking thoughts or even
reducing ‘attention’ to thoughts, but does include contacting thoughts in a particular way. By
bringing behaviour in contact with thoughts in a defused and less exclusive fashion, and altering
the way they are experienced, in a broader context, cognitive defusion allows free and healthy
action without the necessity of positive thoughts and beliefs. This is particularly helpful when
positive thoughts or beliefs can be impossible to achieve or to maintain.
Cognitive defusion methods generally approach thought content indirectly. These meth-
ods include the use of metaphor, paradox, humour, confusion, and experiential exercises
(Hayes et al. 1999a). They typically do not include logical confrontation, direct verbal persuasion,
or empirical verification (Hayes et al. 1999a). Again, methods that seek to directly reformulate the
content of thoughts tend to emphasize the importance of this content and of changing it and
therefore tend to solidify the controlling role of thoughts over behaviour rather than dismantle it.
Responses that attempt to change thought content are themselves responses controlled by that
thought content.
Confrontation of dysfunctional beliefs and cognitions, as is done in cognitive therapy can
achieve a degree of cognitive defusion. These methods typically include methods to identify
thoughts as thoughts and then include recording this on paper and looking at the reality around
these thoughts. Each of these methods, which on the surface form part of a process of changing
thoughts, can as a side effect alter the influence of the thought even before the stage at which a
more rational or accurate thought is concocted. Again, the potential drawback of methods that
direct themselves at change in content is that they can emphasize the necessity of rational think-
ing and reinforce the behavioural imperative of doing, or feeling emotionally, what the literal
content of thought demands one to do or feel.
Methods for targeting cognitive fusion include exercises that raise awareness of: (1) the experi-
ence of having a thought rather than being stuck in the content of the thought; (2) the experience
of difficulties presented by trying to control thoughts; (3) the un-workability of following
some thought content; and (4) the ability to act according to what the person feels is important
while in contact with thoughts that ‘say’ otherwise. These forms of ‘awareness’ so to speak add
certain qualities to subsequent behaviour creating flexibility in the relationship between thoughts
and action.
29.6 Values clarification and values-based action

‘Values’ can be defined as chosen life directions. They are qualities of behaviour that reflect what
the individual cares most about in their life. Values are an individual’s answers to the question,
‘What do you want your life to stand for?’ (Hayes et al. 1999a). Values-based processes are
particularly useful in chronic pain when the pain sufferer’s behaviour has become stuck in a focus
on pain, and other unwanted experiences, and has lost contact with what they would otherwise
hold as most important. For many chronic pain sufferers daily life can be about struggling with
pain, trying to control it, and seeking relief from, it. As this dominates their daily activity, their
interests in family, friends, intimate relations, work, and health, or other concerns become subor-
dinated. A focus on clarifying values, on making them present when choices appear, and on
enhancing values-based action, can counter a nearly exclusive focus on pain as the primary guide
for action.
Values-related processes have been discussed in relation to a number of differing psychological
treatment approaches in the past, including client-centred therapy (Rogers 1964) and motiva-
tional interviewing (Wagner and Sanchez 2002). However, values do not appear to have been
explicitly emphasized in CBT in the past as they have been in ACT and CCBT. Notice, values are
not the same as goals. Goals have a primarily practical, concrete, and achievable quality and values
do not. Values are chosen ongoing directions. Goals tend to be future-focused and outcome ori-
ented. Values are present focused and process oriented. A number of recent attempts to measure
values-related processes have arisen from this new emphasis, both in general samples (Wilson
and Murrell 2004) and in individuals with chronic pain (McCracken and Yang 2006; McCracken
and Vowles 2008).
We developed a brief measure of values and values-based action in a study of 140 consecutive
patients with chronic pain seen for an assessment in a specialty pain centre in the UK. This measure
is called the Chronic Pain Values Inventory (CPVI; McCracken and Yang 2006). The CPVI asks
patients to consider the values they hold in domains of family, intimate relations, friends, work,
health, and growth or learning. They are first asked to rate the importance with which they hold
their values and then the success they have had in living according to them, both on a 0–5 scale.
Results from this study demonstrated that patients’ importance ratings in each domain are
significantly higher than their success ratings. As predicted, average success ratings positively
correlated with acceptance of pain and negatively correlated with a self-report measure of pain-
related avoidance. These average success ratings also significantly correlated with measures of
disability, depression, and pain-related anxiety. In regression analyses, these ratings accounted for
significant variance in these same variables independent of acceptance of pain, suggesting that
these two processes combined can account for significant unique variance in patient functioning
(McCracken and Yang 2006). A later study of chronic pain sufferers conducted prospectively,
over an 18.5-week interval, showed similar results, that acceptance of pain and values-based
action in combination measured at one time account for significant variance (6.5% to 27.0%)
in pain-related distress, depression, and disability at the later point in time (McCracken and
Vowles 2008).
Values clarification is often a challenging process, sometimes fraught with emotional or practi-
cal barriers (Hayes et al. 1999a). It requires in most cases that patients consider circumstances
that may be very distressing. When people honestly identify what they care most about and at the
same time realize they are not acting in accord with that, they may experience painful feelings of
loss, embarrassment, shame, or guilt. This distress from clarifying values may lead patients to
avoid doing so, such as during treatment. Further, many people find it difficult to identify what is
important to them in a way that is not significantly influenced, or even completely obscured, by
what significant others or society in general regard as important. This can produce actions that
are superficial or lack deep meaning for a person. Taking a course of action to avoid disapproval
from others is negatively controlled and is not the same behaviour pattern as taking it under
positive influences, in the service of what one about cares about most. Finally, patients may
immediately dismiss particular directions their lives may take and label them as ‘impossible.’ This
dismissal and automatic judgement, however, may be misleading and may unnecessarily exclude
a real possibility, even if the possible movements in the direction of the value in question are small
or slow (Hayes et al. 1999a; McCracken 2005a).
29.7 Mindfulness
Chronic pain sufferers, like any one of us, can find themselves excessively ‘caught up’ with their
thoughts and their own reactions to sensations in their body, and their emotional experiences.
This entanglement can lead to disorganized, impulsive, or ineffective behaviour. They can have
the natural pain, stress, and unwanted experiences in life magnified in their effects on their behav-
iour by the ways they struggle, defend against, and harden their stance toward these things. They
may multiply their distress by feeling and acting distressed about their distress, and so on (e.g.
McCracken and Keogh, 2009). These processes entail a loss of contact with the present moment
and a tendency to react to experiences automatically in ways that hinder healthy functioning.
A therapeutic response to counteract these processes of suffering is training in a skill set including
direct, present-focused, and accepting awareness, or what is called mindfulness (Kabat-Zinn
1990). Said in a more technical fashion, mindfulness is a process of contact with experiences, such
as thoughts, mental images, urges, and emotional feelings that alters some of the otherwise auto-
matic and misleading functions of these experiences. The predominant attitudes of mindfulness
include openness, curiosity, and compassion.
Mindfulness differs from better-known methods such as relaxation, for example. Relaxation is
often done for the specific purpose of controlling what is experienced, usually to reduce distress-
ing sensations or emotions. Mindfulness, on the other hand, is done for the purpose of simply
being aware of sensations and emotions without doing anything else about them. Guided imagery
exercises are another example. Imagery exercises are often done to change the content of what is
being experienced in order to gain the emotional, physical, or behavioural effects that come with
that. Mindfulness is done to observe and learn about whatever content spontaneously occurs in
experience while also maintaining a connection with the present moment, an attitude of interest,
but a neutral attitude with respect to needing to take any action. There are many types of mindful-
ness exercises. Some of these include a focus on the body, on sensations of breathing, or on sensa-
tions during movement (Kabat-Zinn 1990), and some include the use of an imagined scene as a
framework for contacting one’s experiences mindfully (Hayes et al. 1999a).
Although mindfulness has been of formal interest within pain management for more than 20
years, there is still a relative lack of evidence regarding its usefulness. There are uncontrolled trials
supporting the role of mindfulness-based methods in treatment for chronic pain (Kabat-Zinn
et al. 1985; Kaplan et al. 1993). Meta-analyses of mindfulness-based treatments for a range of
conditions including pain also conclude that these treatments are of potential benefit for those
with chronic pain (Baer 2003; Grossman et al. 2004). There are three recent randomized trials of
mindfulness that have produced significant results in relation to symptoms of depression in
women with fibromyalgia (Sephton et al. 2007), in relation to acceptance of pain and for physical
functioning in older adults (Morone et al. 2008), and in relation to coping, emotional function-
ing, and joint tenderness in patients with rheumatoid arthritis, particularly those with a history of
depression (Zautra et al. 2008).
Relatively few studies of mindfulness have attempted to measure the process or processes
involved and examine them in relation to measures of patient functioning. We completed a study
of mindfulness in patients with chronic pain (McCracken et al. 2007) in which we administered
the Mindful Attention Awareness Scale (MAAS; Brown and Ryan 2003) to a sample of 105 con-
secutive patients seeking specialty treatment in the UK. The MAAS is a 15-item measure of mind-
fulness in which all of the item content is negatively-keyed (i.e. ‘I could be experiencing some
emotion and not be conscious of it until some time later,’ ‘I find it difficult to stay focused on
what’s happening in the present,’ ‘I find myself preoccupied with the future or the past’). Items
are rated from 1 (‘almost always’) to 6 (‘almost never’) and are summed to produce a single sum-
mary score. Correlation analyses of the MAAS showed that mindfulness was positively correlated
with acceptance of pain (r=0.28), as predicted. It was also negatively correlated with measures of
pain-related distress, anxiety, depression, interference with cognitive functioning, and pain
medication use. In multiple regression analyses in which age, gender, education, duration of pain,
pain intensity, and acceptance of pain, were entered first, mindfulness accounted for significant
increments in explained variance for five of eight measures of patient functioning, including
depression, pain-related anxiety, as well as physical, psychosocial, and ‘other’ disability. The
strength of these results with regard to mindfulness is highlighted by the fact that acceptance of
pain, entered before mindfulness in each equation, is a reliably strong predictor of patient func-
tioning. Mindfulness was a particularly good predictor of depression where it accounted for
11.0% of the variance after the series of other potential predictors were taken into account. The
variance in these equations accounted for by acceptance of pain and mindfulness combined
averaged 28.0% across the eight equations (McCracken et al. 2007).
Acceptance, ACT, and mindfulness are often discussed together, particularly in relation to
‘expanding the cognitive-behavioural tradition’ (e.g. Hayes et al. 2004). There is clearly a shared
sensibility between ACT and mindfulness but these approaches come from different traditions: ACT
from the behaviour analytic side of the cognitive and behavioural therapies, a family of therapy
approaches that are no more than 50 years old; and mindfulness from eastern spiritual practices,
practices that are about 3000 years old. ACT includes six partly overlapping therapeutic processes:
acceptance, contact with the present, values-based action, committed action, self-as-context, and
cognitive defusion (Hayes et al. 2006). Although this does not appear in formal technical analyses
or in popular definitions of mindfulness, mindfulness appears to include four of these ACT proc-
esses: acceptance, contact with the present, self-as-context, and cognitive defusion.
There is only one study we have identified in which processes underlying mindfulness were
investigated in a sample of people with chronic pain. This is a study that again used the MAAS
(McCracken and Thompson 2009). This study of mindfulness included 150 persons seeking
treatment for chronic pain. There data from the MAAS were factor analysed and yielded four fac-
tors: Acting with Awareness (e.g. ‘I could be experiencing some emotion and not be conscious of
it until some time later’), Present Focus (e.g. I find myself preoccupied with the future or the
past’), Responsiveness (e.g. ‘I tend not to notice feelings of physical tension or discomfort until
they really grab my attention’), and Social Awareness (e.g. ‘I forget a person’s name almost as
soon as I’ve been told it for the first time’). Subsequent correlation analyses showed that just two
of these factors: Acting with Awareness and Present Focus appeared to account for the relations
of the overall mindfulness process with measures of emotional, physical, and social functioning
in these patients. In fact, the Present Focus factor emerged as the strongest predictor of patient
functioning, achieving particularly strong correlations with psychosocial disability, r =−0.60, and
depression, r =−0.60, both p <0.001 (See Figure 29.1).
29.8 Chronic pain in a social context

While social influence is widely acknowledged in chronic pain, current models of social influ-
ence appear limited. These models are often univariate, mechanistic, static, or unidirectional
–0.6
–0.5
Correlations
–0.4
–0.3 Act w/aware

Present focus
–0.2
Responsiveness
–0.1 Social awareness
0
ss
n
lit
lit
io
re
bi
bi
ss
st
sa
sa
re
di
di
di
ep
in
oc
D
Pa
Ph
hs
yc
Ps
Fig. 29.1 Correlations of mindfulness factors from the Mindful Attention Awareness Scale with
measures of patient functioning (N=150). Note: correlations marked by bars above the dashed line
significant at p <0.05. Pain distress measured with a 0–10 rating from ‘not at all distressing’ to
‘extremely distress;’ psychosocial and physical disability measures with the Sickness Impact Profile
(Bergner et al. 1981); and depression measured with the British Columbia Major Depression
Inventory (Iverson and Remick 2004). Data from McCracken and Thompson (2009).
(e.g. Peat et al. 2004; Schwartz et al. 2005). A functional contextual approach to chronic pain has
several implications for the role of social influence in chronic pain and suffering. These include a
potentially more incisive and complex analysis of the role of significant others in chronic pain
(McCracken, 2005b), models of ‘self’ and identity (Crombez et al. 2003) and the role of social
influences in the context of treatment (McCracken 2005a).
Social influence on chronic pain has been of interest to researcher and clinicians for at least
40 years. For most of that time, however, the role of social responses in chronic pain has been
considered in relation to processes of reinforcement or punishment for overt displays of pain and
disability, so-called ‘pain behaviour’ (Fordyce 1976). Alternately, social influence has also been
considered in relation to buffering or stressing effects on emotional experience of chronic pain
(Cano et al. 2004; reviewed in this volume in Chapter 17). A complete social analysis of chronic
pain from a functional contextual approach is well beyond the scope of this chapter. However, a
few illustrative examples of how social influence might be analysable from a more contextual
approach are possible (McCracken 2005b).
One important point is to observe that people engage in behaviour strongly determined
by social experience everyday of their life, even when there are no other people present, and
obviously this will be true for chronic pain sufferers. This is a key distinction, from a contextual
standpoint, what makes behaviour socially determined is not the immediate presence of others
but a history of social influence that coordinates the current behaviour. When patients have expe-
riences of guilt, shame, humiliation, self-criticism, and even anger, these experiences, the behav-
iour that attends them, and the relationship between the two is certainly social in origin.
Surely social and non-social influences will be woven together in complex ways and social
influence may be present where it is not expected. For example, a person with pain who responds
to experiences of that pain with activity avoidance may have that pattern so tightly controlled by
present experiences of thoughts and emotions that other immediate social influences play little
role. In contrast, consider social aspects of learning by instruction or observing the behaviour
of others, or consider the social roots of guilt and how strong an influence guilt can play over
behaviour patterns.
The apparent solicitousness or aversiveness of a social response may have little to do with its
effect on behaviour and again there may be unexpected results. For example, a patient who acts
with increasing avoidance and passivity may have this behaviour pattern strengthened (not weak-
ened) if it meets angry or frustrated reactions from their environment, particularly if they feel
invalidated or like their pain experience is not being treated as legitimate. Most of us will rigidly
defend against being disbelieved. On the other hand a patient whose difficulties in functioning
meet offers of help may respond by hiding these difficulties in the future or by struggling to carry
on without help, such as when receiving help includes feeling dependent on others, like a burden,
or like an ‘invalid.’ Either of these ostensibly different patterns could strengthen a pattern of
behaviour that limits functioning in the longer term (McCracken 2005a, 2005b). From the per-
spective presented here it is not the appearance of the response as either ‘solicitous’ or ‘punishing’
that is critical but it is the psychological context in which it occurs, including the pain sufferer’s
history, that gives those responses their meaning and function.
There are many different conceptualizations of ‘self.’ For example, the self can be understood
as an object or subject, ‘self-as known’ or ‘self-as-knower’ (James 1950). There is general agree-
ment that our sense of self is essentially social in origin (James 1950; Skinner 1953). We increas-
ingly learn who we are, in a sense, as this becomes important for others. In turn, we experience an
inclination to act consistently with this sense of our self (Hayes et al. 1999a). To act unpredictably
or ‘not like our self’ is often experienced as aversive by other and is discouraged in their responses.
From another viewpoint, however, this sense of self of social origin is basically a story, a constructed
or conceptualized sense of self that is not the same as the actual person as a human being or bio-
logical organism.
A number of different problems can arise from this situation of the conceptualized self for
those with chronic pain. Social pressure for behavioural consistency will inevitably come to bear
on chronic pain sufferers. This is particularly pertinent in situations where such consistency is
either impossible or unworkable. Chronic pain is associated with changes in roles and behaviour,
changes that occur in relation to private experiences that cannot be observed and are not fully
understood by others. This can be distressing, such as when expectations from others are not met,
and this lead them to criticize or express disappointment. It can also be restricting on functioning,
such as when essentially arbitrary social influences press for behaviour patterns that are consistent
to the outside observer but essentially ineffective.
It is a natural process of thinking and speaking in a social context that we can come to hold too
tightly onto a certain sense or our self, a self made up of our thoughts and beliefs about who we
are. This self we can refer to as ‘self-as-content’ (Hayes et al. 1999a). For chronic pain sufferers
strongly held beliefs such as ‘I never need help from others,’ ‘I never quit,’ ‘I am always happy,’
‘I never talk about my feelings,’ or ‘I am always the victim,’ can rigidly determine behaviour
patterns and prevent change in behaviour that would otherwise entail better functioning. If the
belief ‘I never need help from others’ is defended in action as a matter of personal identity, seek-
ing help, even when this is the most effective available action, may not occur. Likewise, if the belief
‘I never quit’ is followed in the face of a behaviour pattern that consistently meets unwanted
consequences, whether this be in seeking medical treatments that consistently fail, staying in an
unhealthy relationship, or following an unrealistic program of physical exercise, then this behav-
iour pattern is unworkable. From this type of analysis treatment approaches for chronic pain can
advocate the development in treatment of a different sense of self, a self that is not determined
in the content of thought and belief, but emerges as an awareness of this content, a sense of
self as the location where this content occurs, a sense of ‘self as context’ (Hayes et al. 1999a;
McCracken 2005a).
Many individuals with complex problems of chronic pain will have experienced the feeling of
being disbelieved, may be highly attuned to this possibility in subsequent healthcare visits, and
may want to avoid it. This is understandable. To suffer and to have the legitimacy of that suffering
doubted can be to feel accused of lying, or that one will be abandoned with uncontrollable pain.
This is a situation any person would fight to avoid. And perhaps the most natural response to this
situation is to try to ‘prove’ there is something wrong. When the pain sufferer’s behaviour is
under this kind of influence it is a serious dilemma for treatment. It might be difficult or impos-
sible for a person to prove they are suffering and to recover lost functioning at the same time. As
the consequences associated with being disbelieved are often painful or frightening, even actions
that hint at disbelief may be experienced as such. For example, to imply that certain activities may
be done without pain relief may be to send a message that sounds illogical or expresses disbelief.
Requests for actions that appear ‘impossible’ can feel de-legitimizing, and can be experienced as
discounting a person’s pain, even if these are expressed with the best of intentions.
Treatments like CCBT are intensive and highly emotionally evocative. They require great care
in the development of the relationships between treatment participants and treatment providers.
A primary focus of treatment includes creating occasions when patients will contact painful expe-
riences, specifically as these are a necessary part of making the changes they desire. It is typically
on these occasions in daily life that the patient’s behaviour is failing to serve their goals and values.
These occasions provide the opportunity for developing different and more effective actions,
actions that engender flexibility rather than struggling or avoidance. In order to effectively deliver
treatment of this type, treatment providers will need a well developed analysis or conceptualization
of the behaviour problems being targeted. They will also require repeated patient consent, and
adequate rapport to create occasions that are clinically relevant and unlikely to punish the very
behaviour patterns that are trying to be developed (Kohlenberg and Tsai 1991).
It is presumed that effective therapeutic behaviour will require psychological flexibility on the
part of the treatment provider, including acceptance of painful psychological experiences, contact
with the present moment, values-based action, and cognitive defusion. In our work we have
shown that these qualities in healthcare worker behaviour are correlated with greater health and
well-being in healthcare provider themselves and in less hindrance from patients’ pain in the
performance of work-related activities (McCracken and Yang 2008), however, we have not yet
demonstrated fully that these qualities produce better treatment results, again, as we presume
they will. This will require further study.
29.9 Treatment outcome from contextual approaches to

chronic pain
Despite the relatively recent development of contextual approaches, there are now many
published randomized controlled trials of these approaches for a range of behaviour problems.
This includes at least 20 published trials of ACT in a wide range of clinical areas (Hayes et al. 2006)
and more than six trials of mindfulness-based methods in relation to chronic pain alone (Baer
et al. 2003). There are two published, small-scale, randomized trials of ACT related to chronic
pain (Dahl et al. 2004; Wicksell et al. 2008) and at least three other effectiveness trails of ACT-
based treatment that did not include randomization to separate treatment conditions (McCracken
et al. 2005, 2007; Vowles and McCracken 2008).
Of the two randomized trials of ACT related to chronic pain one applied this approach to
persons at risk for long-term disability due to pain and stress in a work setting (Dahl et al. 2004).
This trial showed dramatically reduced missed workdays and healthcare use from four hours of
treatment. The other ACT trial in this area included persons with chronic pain associated with
whiplash in a specialty pain centre (Wicksell et al. 2008). This trial showed significant improve-
ments in disability, life satisfaction, fear, and depression from 10 sessions. Both of these studies
were conducted in Sweden. Both included ‘treatment as usual’ as the control condition.
All three of our treatment trials were conducted as analyses of an actual ongoing tertiary care
treatment service operating within the National Health Service in the UK. This service is designed
for people with highly complex cases of chronic pain that have repeatedly failed to benefit from
other primary and secondary care services. As mentioned, in this respect our analyses of treat-
ment outcome results are a test of the application of a treatment model to actual clinical practice
and are therefore more akin to effectiveness studies than efficacy trials. Further, as a publicly
funded national specialty service, providing services to pain sufferers who have few other options,
our patient selection criteria for services are particularly inclusive.
Our first study of treatment outcome included 108 chronic pain sufferers, about half of
whom reported back pain as their primary problem (McCracken et al. 2005). They participated
in 3- or 4-week, group-based, treatment delivered by an interdisciplinary team of physiothera-
pists, occupational therapists, nurses, physicians, and clinical psychologists. This treatment
included graded physical conditioning, exposure-based methods, skills training, education,
mindfulness training, and experiential psychological methods based on the ACT and CCBT
model (McCracken 2005a). The comparison condition for this study consisted of data from a
waiting interval priory to treatment that was an average of 3.9 months in duration. Hence there
was no randomization to conditions.
The analyses in our first treatment study were based on outcome measures collected at initial
assessment, the start of treatment, immediately post treatment, and at a three-month follow-up
visit for a consecutive sample. Results showed that patient functioning remained unchanged
during the waiting phase prior to treatment but significantly improved in nine key domains
following treatment and remained significantly improved on all measured domains at follow-up.
These domains included pain, depression, pain-related anxiety, physical disability, psychosocial
disability, daily rest, walking speed, activity tolerance, and medication use (all p <0.001). Other
results showed that acceptance of pain significantly increased during treatment and changes in
acceptance were significantly correlated with changes in other key outcome variables (McCracken
et al. 2005).
In our second outcome study we examined clinically significant change in a subset of highly
disabled patients (n=53), about 60% of whom reported back or whole body pain (McCracken
et al. 2007). These are patients we treat in a hospital-based course because they have extremely
limited mobility and self-care and thus are unable to participate in any pain management course
without nursing support for transfers, mobility, bathing, or dressing. These patients significantly
improved at post treatment in pain-related distress, physical and psychosocial disability, depres-
sion, anxiety, directly assessed activity tolerance, and hours spent resting during the day due to
pain. The average effect size across outcome variables was d =0.75. Based on analyses that take
into account degree of temporal stability in outcome measures (Jacobson et al. 1999), 47.6% of
patients demonstrated reliable improvement in total disability and 61.9% demonstrated reliable
improvement in depression scores. Applying criteria for ‘recovery’ based on level of post treat-
ment functioning of successfully treated standard cases from our service, 52.8% of the highly
disabled patients met criteria for clinical recovery in a least one primary outcome domain, includ-
ing physical or psychosocial disability, or depression (McCracken et al. 2007). This is equivalent
to saying that for every 1.9 highly disabled patients treated with this approach one achieves recov-
ery in at least one key domain of functioning. It will be noticed that once again there was no
randomization to an active or control condition. All patients were assigned to active treatment.
The design allowed for pre to post comparisons and comparison between two groups who
differed in their initial level of disability.
In our most recent investigation we examined treatment outcome and process in 171 consecu-
tive patients that were an entirely independent sample from the previous trials (Vowles and
McCracken 2008). As with the previous trials the effects of treatment were broad, as reflected in
multiple measures of pain, emotional, physical, and social functioning, healthcare use, and directly
assessed physical performances. Effect sizes at post treatment averaged d =1.07 and at three month
follow-up averaged d =0.89. Based on reliable change analyses 75.4% of patients were reliably
improved in at least one domain, including anxiety, depression, or disability. Change in the com-
bined processes of acceptance of pain and values-based action accounted for significant variance
in improvements during treatment or at follow-up in each of seven key measures of functioning.
Average variance accounted for at post treatment was 15% and at follow-up was 17%.
29.10 Summary
In many ways our main cognitive-behavioural treatment methods for chronic pain have stayed
very much the same for the past 25 years. Perhaps this is a product of early success (Flor et al.
1992). There could be other factors: a lack of a truly well-integrated and progressive theory of
human behaviour and suffering. Maybe there were other battles, such as to hold onto some piece
of territory in a field where psychological methods are clearly the underdog. Psychological meth-
ods for pain are set in a culture where solutions for pain that look like a fix, including medications
or even high technology devices, have much greater immediate appeal. These apparent solutions
it seems are much more enticing than approaches that require learning, facing discomfort and
uncertainty, approaches that say the answer to your pain requires change in your behaviour.
At the same time that CBT for pain was enjoying its successes clinical psychology outside of
pain management has been growing and changing faster than inside pain management. This is an
impression, said with no data. If it is true, what does it mean?
Increasing research findings outside of pain management and, increasingly, inside of pain
management, demonstrate that there is more than one way to live a full life when one has been
suffering with chronic pain. Methods available from CBT as it has been practised for 25 years are
one way (Morley et al. 1999). These methods are varied but tend to include training in coping and
methods designed predominantly to change, control, or decrease unwanted physical and emo-
tional feelings, or dysfunctional or misleading thoughts and beliefs (e.g. Williams et al. 1996;
DeJong et al. 2005). Another way, rather than coping with or attempting to control feelings,
thoughts, or beliefs, is to a focus on contextual change around all of the painful and discouraging
experiences that arise with chronic pain.
Contextually-based therapies, like ACT and CCBT or mindfulness-based approaches, aim to
manipulate the contextual elements that afford psychological experiences their influence over
behaviour patterns. They do not primarily aim to manipulate the content of psychological experi-
ences themselves. They aim not to decrease sensations, emotions and thoughts but to modify how
they are experienced. Contextual change is not superficial. It attempts to get to the roots from
which suffering and behaviour disruption emerge, largely in the interaction of processes of lan-
guage and cognition with direct experience. Contextual processes are perhaps less naturally
apparent in our everyday experience. Contextual methods are perhaps subtle, sometimes illogical,
metaphorical or paradoxical, not dominantly change-oriented, and typically flexible in their
balance between change and acceptance.
There are now more than ten recent trials investigating contextual approaches for chronic pain,
including acceptance or mindfulness-based methods. Six of these are fully randomized and con-
trolled and some of the remaining ones are non-randomized or partially controlled effectiveness
studies. Most of controlled studies are small in scale. Findings from these studies are encouraging.
They are encouraging because they show impacts on variables of high importance, such as work
status and healthcare use. They are also encouraging because they include process analyses. There
is additional work to do: larger randomized trials, and development work to produce versions of
treatment that allow greater access, outside of specialty care, and versions that deliver greater
impact more efficiently.
References
Bach, P. and Hayes, S.C. (2002). The use of Acceptance and Commitment Therapy to prevent the
rehospitalization of psychotic patients: a randomized controlled trial. Journal of Consulting and Clinical
Baer, R.A. (2003). Mindfulness training as a clinical intervention: a conceptual and empirical review.
Clinical Psychology Research and Practice, 10, 125–43.
Beck, A.T. (1976). Cognitive therapy and the emotional disorders. New York: Meridian.
Bond, F.W. and Bunce, D. (2000). Mediators of change in emotion-focused and problem-focused worksite
stress management interventions. Journal of Occupational Health Psychology, 5, 156–63.
Brown, K.W. and Ryan, R.M. (2003). The benefits of being present: mindfulness and its role in
psychological well-being. Journal of Personality and Social Psychology, 84, 822–48.
Burns, D.D. and Spangler, D.L. (2001). Do changes in dysfunctional attitudes mediate changes in
depression and anxiety in cognitive behavioral therapy? Behavior Therapy, 32, 337–69.
Cano, A., Gillis, M., Heinz, W., Geisser, M., and Foran, H. (2004). Marital functioning, chronic pain, and
psychological distress. Pain, 107, 99–106.
Cheung, N.M., Wong, T.C.M., Yap, J.C.M., and Chen, P.P. (2008). Validation of the Chronic Pain
Acceptance Questionnaire (CPAQ) in Cantonese-speaking Chinese patients. Journal of Pain, 9, 823–832.
Crombez, G., Morley, S., McCracken, L.M., Sensky. T., and Pincus, T. (2003). Self, identity and acceptance
in chronic pain. In J.O. Dostrovsky, D.B. Carr, M. Koltzenburg (Eds.) Proc. 10th World Congress on
Pain, Vol. 24, pp. 651–9. Seattle: IASP Press.
Dahl, J., Wilson, K.G. and Nilsson, A. (2004). Acceptance and Commitment Therapy and the treatment of
persons at risk for long-term disability resulting from stress and pain symptoms: a preliminary
randomized trial. Behavior Therapy, 35, 785–801.
David, D. and Szentagotai, A. (2006). Cognitions in cognitive-behavioral psychotherapies: toward an
integrative model. Clinical Psychology Review, 26, 284–98.
De Jong, J.R., Vlaeyen, J.W.S., Onghena, P., Cuypers, C., den Hollander, M. and Ruijgrok, J. (2005).
Reduction of pain-related fear in complex regional pain syndrome type I: the application of graded
exposure in vivo. Pain, 116, 264–275.
Dimidjian, A., Hollon, S.D., Dobson, K.S., et al. (2006). Randomized trial of behavioral activation,
cognitive therapy, and antidepressant medication in the acute treatment of adults with major
depression. Journal of Consulting and Clinical Psychology, 74: 658–70.
Ellis, A. (1994). Reason and emotion in psychotherapy (Rev. edn.). Secaucus, NJ: Birch Lane.
Flor, H., Frydrich, T., Turk, D.C. (1992). Efficacy of multidisciplinary pain treatment centers: a
meta-analytic review. Pain, 49, 221–230.
Fordyce, W.E. (1976). Behavioral methods for chronic pain and illness. Saint Louis, MO: Mosby.
Geiser, D.S. (1992). A Comparison of acceptance-focused and control-focused psychological treatments in
a chronic pain treatment center. Unpublished Doctoral Dissertation, Reno: University of Nevada.
Grossman, P., Niemann, L., Schmidt, S. and Walach, H. (2004). Mindfulness-based stress reduction and
health benefits: a meta-analysis. Journal of Psychosomatic Research, 57, 35–43.
Gutierrez, O., Luciano, C., Rodriguez, M. and Fink, B.C. (2004). Comparison between an
acceptance-based and a cognitive-control-based protocol for coping with pain. Behavior Therapy,
35, 767–83.
Hayes, S.C., Strosahl, K. and Wilson, K.G. (1999a). Acceptance and Commitment Therapy: An experiential
approach to behavior change. New York: Guildford Press.
Hayes, S.C., Bissett, R.T., Korn, A., et al. (1999b).The impact of acceptance versus control rationales on
pain tolerance. Psychological Record, 49, 33–47.
Hayes, S.C., Follette, V.M. and Linehan, M.M. (2004). Mindfulness and acceptance: expanding the
cognitive-behavioral tradition. New York: The Guilford Press.
Hayes, S.C., Luoma, J.B., Bond, F.W., Masuda, A. and Lillis, J. (2006). Acceptance and Commitment
Therapy; model, processes and outcomes. Behaviour Research and Therapy, 44, 1–25.
Iverson, G.L. and Remick, R. (2004). Diagnostic accuracy of the British Columbia Major Depression
Inventory. Psychological Reports, 95, 1241–1247.
Jacobson, N.S., Dobson, K.S., Truax, P.A., et al. (1996). A component analysis of cognitive-behavioral
treatment for depression. Journal of Consulting and Clinical Psychology, 64, 295–304.
Jacobson, N.S., Roberts, L.J., Berns, S.B. and McGlinchey, J.B. (1999). Methods from defining and
determining the clinical significance of treatment effects: description, application, and alternatives.
Journal of Consulting and Clinical Psychology, 67, 300–7.
Jacobson, N.S., Christensen, A., Prince, S.E., Codova, J. and Eldridge, K. (2000). Integrative behavioral
couple therapy: an acceptance-based, promising new treatment for couple discord. Journal of
Consulting and Clinical Psychology, 68, 351–5.
James, W. (1950). Principles of psychology (reprint edition). New York: Dover Publications.
Kabat-Zinn, J. (1990). Full catastrophe living: using the wisdom of your body and mind to face stress, pain, and
illness. New York: Dell Publishing.
Kabat-Zinn, J., Lipworth, L. and Burney, R. (1985). The clinical use of mindfulness meditation for
self-regulation of chronic pain. Journal of Behavioral Medicine, 8, 163–90.
Kaplan, K.H., Goldenberg, D.L. and Galvin-Nadeau, M. (1993). Impact of a meditation-based stress
reduction program on fibromyalgia. General Hospital Psychiatry, 15, 284–9.
Kohlenberg, R.J. and Tsai, M. (1991). Functional analytic psychotherapy: creating intense and curative
therapeutic relationships. New York: Springer.
Longmore, R.J. and Worrell, M. (2007). Do we need to challenge thoughts in cognitive behavior therapy.
Clinical Psychology Review, 27,173–87.
Ma, S.H. and Teasdale, J.D. (2004). Mindfulness-based cognitive therapy for depression: replication and
exploration of differential relapse prevention effects. Journal of Consulting and Clinical Psychology, 72, 31–40.
Martell, C., Addis, M. and Dimidgian, S. (2004). Finding the action in behavioral activation, in S.C. Hayes,
V.M. Follette, M.M. Linehan (ed), Mindfulness and acceptance: expanding the cognitive-behavioral
tradition, pp. 152–67. New York: The Guilford Press.
Masedo, A.I. and Esteve, M.R. (2007). Effects of suppression, acceptance and spontaneous coping on pain
tolerance, pain intensity and distress. Behaviour Research and Therapy, 45, 199–209.
Mason, V.L., Mathias, B. and Skevington, S.M. (2007). Accepting low back pain: is it related to good
quality of life? Clinical Journal of Pain, 24, 22–9.
McCracken, L.M. (1998). Learning to live with the pain: acceptance of pain predicts adjustment in persons
with chronic pain. Pain, 74, 21–7.
McCracken, L.M. (2005a). Contextual Cognitive-Behavioral Therapy for chronic pain, Progress in pain
research and management, vol 33. Seattle, WA: IASP Press.
McCracken, L.M. (2005b). Social context and acceptance of chronic pain: the role of solicitous and
punishing responses. Pain, 113, 155–9.
McCracken, L.M. (2007). A contextual analysis of attention to chronic pain: what the patient does with their
pain may be more important than their awareness or vigilance alone. The Journal of Pain, 8, 230–6.
McCracken, L.M. and Eccleston, C. (2003). Coping or acceptance: what to do with chronic pain? Pain,
105, 197–204.
McCracken, L.M. and Eccleston, C. (2005). A prospective study of acceptance of pain and patient
functioning with chronic pain. Pain, 118, 164–9.
McCracken, L.M. and Eccleston, C. (2006). A comparison of the relative utility of coping and
acceptance-based measures in a sample of chronic pain sufferers. European Journal of Pain, 10, 23–9.
McCracken, L.M. and Keogh, E. (2009). Acceptance, mindfulness, and values-based action may counteract
fear and avoidance of emotions in chronic pain: An analysis of anxiety sensitivity. The Journal of Pain,
10, 408–15.
McCracken, L.M. and Samuel, V.M. (2007). The role of avoidance, pacing, and other activity patterns in
McCracken, L.M. and Thompson, M. (2009). Components of mindfulness in patients with chronic pain.
Journal of Psychopathology and Behavioral Assessment, 31, 75–82.
McCracken, L.M. and Vowles, K.E. (2008). A prospective analysis of acceptance of pain and values-based
action in patients with chronic pain. Health Psychology, 27, 215–20.
McCracken, L.M. and Yang, S-Y. (2006). The role of values in a contextual cognitive-behavioral analysis of
McCracken, L.M. and Yang, S-Y. (2008). A contextual cognitive-behavioral analysis of rehabilitation
workers’ health and well-being: influences of acceptance, mindfulness, and values-based action.
Rehabilitation Psychology, 53, 479–85.
McCracken, L.M., Vowles, K.E. and Eccleston, C. (2004). Acceptance of chronic pain: component analysis
and a revised assessment method. Pain, 107, 159–66.
McCracken, L.M., Vowles, K.E. and Eccleston, C. (2005). Acceptance-based treatment for persons with
complex, longstanding chronic pain: a preliminary analysis of treatment outcome in comparison to a
waiting phase. Behaviour Research and Therapy, 43, 1335–46.
McCracken, L.M., Vowles, K. and Thompson, M. (2007a). Comment on Nicholas and Asghari. Pain,
128, 283–4.
McCracken, L.M., MacKichan, F. and Eccleston, C. (2007b). Contextual Cognitive-Behavioral Therapy for
severely disabled chronic pain sufferers: effectiveness and clinically significant change. European
McCracken, L.M., Gauntlett-Gilbert, J. and Vowles K. (2007c). The role of mindfulness in a contextual
cognitive-behavioral analysis of chronic pain-related suffering and disability. Pain, 131, 63–69.
Morley, S., Eccleston, C. and Williams, A.C. (1999). Systematic review and meta-analysis of randomized
Morone, N.E., Greco, C.M. and Weiner, D.K. (2008). Mindfulness meditation for treatment of chronic low
back pain in older adults: a randomized controlled pilot study. Pain, 134, 310–19.
Nicholas, M.K. and Asghari, A. (2006). Investigating acceptance in adjustment to chronic pain: is
acceptance broader than we thought? Pain, 124, 269–79.
Peat, G., Thomas, E., Handy, J. and Croft, P. (2004). Social networks and pain interference with daily
activities in middle and old age. Pain, 112, 397–405.
Rogers, C.R. (1964). Toward a modern approach to values: the valuing process in the mature person.
Journal of Abnormal and Social Psychology, 68, 160–7.
Sephton, S.E., Salmon, P., Weissbecker, I., et al. (2007). Mindfulness meditation alleviates depressive
symptoms in women with fibromyalgia: results of a randomized clinical trial. Arthritis & Rheumatism
(Arthritis Care and Research), 57, 77–85.
Skinner, B.F. (1953). Science and human behavior. New York: The Free Press.
Schwartz, L., Jensen, M.P., Romano, J.M. (2005). The development and psychometric evaluation of an
instrument to assess spouse responses to pain and well behavior in patients with chronic pain: the
Spouse Response Inventory. Journal of Pain, 6, 243–52.
Turk, D.C., Meichenbaum, D. and Genest, M. (1983). Pain and behavioral medicine: a cognitive-behavioral
perspective. New York: The Guilford Press.
Viane, I., Crombez, G., Eccleston, C., et al. (2003). Acceptance of pain is an independent predictor of
mental well-being in patients with chronic pain: empirical evidence and reappraisal. Pain, 106, 65–72.
Vowles, K.E. and McCracken, L.M. (2008). Acceptance and values-based action in chronic pain: a study of
treatment effectiveness and process. Journal of Consulting and Clinical Psychology, 76, 397–407.
Vowles, K.E., McCracken, L.M., McLeod, C. and Eccleston, C. (2008). The Chronic Pain Acceptance
Questionnaire: confirmatory factor analysis and identification of patient subgroups. Pain, 140, 284–91.
Wagner, C.C. and Sanchez F.P. (2002). The role of values in motivational interviewing. In W.R. Miller,
S. Rollnick (Eds.) Motivational interviewing: preparing people for change (2nd edn.), pp. 284–98.
Wicksell, R.K., Ahlqvist, J., Bring, A., Melin, L. and Olson, G.L. (2008). Can exposure and acceptance
strategies improve functioning and life satisfaction in people with chronic pain and
whiplash-associated disorders (WAD)? A randomized controlled trial. Cognitive Behavior Therapy,
37, 169–82.
Williams, A.C. de C., Richardson, P.H., Nicholas, M.K., et al. (1996). Inpatient vs. outpatient pain
management: results of a randomized controlled trial. Pain, 66, 13–22.
Wilson, K.G. and Murrell, A.R. (2004). Values work in Acceptance and Commitment Therapy. In
S.C. Hayes, V.M. Follette, and M.M. Linehan (Eds.) Mindfulness and acceptance: expanding the
cognitive-behavioral tradition, pp. 121–51. New York: Guilford Press.
Zautra, A.J., Davis, M.C., Reich, J.W., et al. (2008). Comparison of cognitive behavioral and mindfulness
meditation interventions on adaptation to rheumatoid arthritis for patients with and without history
or recurrent depression. Journal of Consulting and Clinical Psychology, 76, 408–21.
Zettle, R.D., Rains, J.C. (1989). Group cognitive and contextual therapies in treatment of depression.
Journal of Clinical Psychology, 45, 436–45.
Chapter 30
Rehabilitation Programmes to Prevent

Severely Disabling Chronic Back Pain
Michael K. Nicholas and Rob J.E.M. Smeets
30.1 Introduction
Once pain has persisted for several months and is associated with marked interference in daily
activities and depressed mood, low self-efficacy, catastrophizing thoughts, and fear-avoidance
beliefs, the risk is very high for continuing deterioration and reduced chances of returning to
work and pre-injury lifestyles. Previous chapters in this book have clearly described the impor-
tance of identifying and helping those considered to be at risk of this secondary disability as soon
as possible. But despite the best efforts of primary healthcare providers, a proportion of cases will
still end up with chronic pain and many will continue to seek help. In their pursuit of help they
often run the added risk of iatrogenic complications—complications caused by the treatments
and attention they receive rather than the original injury. In many countries injured workers will
be at increasing risk of losing their jobs the longer they are away from the workplace. They may
have to deal not just with their pain and the associated activity limitations, but also with financial
pressures, relationship difficulties, and the unwanted effects of multiple treatments. This chapter
is directed at this group of people and their management and rehabilitation options.
Typically, the aims of rehabilitation at this stage are multiple. While pain relief is often the
immediate focus of the patient, it is often unlikely to be achieved and some accommodation will
have to be made to that. But other goals may well be feasible. These may be described broadly as
quality of life goals, and may include return to desired work, home or social activities, as well as
improved sleep, mood, and health. Some of these may initially be quite basic activities of daily
living or independent functioning, like sitting long enough to achieve a task on a computer or
drive to the shops. Others may be more complex, but whatever the task, the fundamental princi-
ple in all rehabilitation is that success will require the active participation of the patient suffering
from pain. Rehabilitation cannot be done to someone; it can only be done with him or her. This
has important implications for healthcare providers and agencies responsible for facilitating reha-
bilitation. It means we need to consider not just the nature of the pain but also the person in pain
and his perspectives, abilities and goals (or desires), as well as the context in which he lives—
the workplace, the home, even the broader social environment via the available social support
systems.
In previous chapters the early identification of those at risk as well as potential contributing
personal and environmental factors were described, so they will not be repeated here. Instead, we
will explore the tasks likely to be involved in helping those who have slipped through the early
management ‘net’ and the evidence for the interventions employed as pain persists beyond the
first few months.
30.2 A conceptual framework for assessment and

multidisciplinary rehabilitation
Biopsychosocial models of pain have provided the most widely-accepted conceptual framework
for pain rehabilitation for over 40 years and different versions of this model have been extensively
described throughout this book. A version of a biopsychosocial approach has also been described
for general use in the World Health Organization (WHO) International Classification of
Functioning, Disability, and Health (2001). This classification has attempted to capture the key
features of the biopsychosocial models and to provide a basis for operational definitions of these
key elements. Specifically, it identifies three core elements: body systems (body functions
and structures), the individual (activity) and society (participation), all of which are expected to
interact (see Figure 30.1).
As can be seen in Figure 30.1, bodily functions and structures are defined as physiological func-
tions of body systems or anatomical structures. Activity is defined as the capacity of the person to
execute a task or action. This action is an inherent or intrinsic feature of the person himself, with-
out the influence of varying environmental factors (e.g. bending forward). Participation is the
actual performance of activities in the person’s current environment aimed at engaging in a life
situation (e.g. work, preparing a meal, picking up a child from school, driving the car to the
supermarket, despite limited ability to bend forward, sit, stand, or walk). This account of a
biopsychosocial model encourages the eliciting of information about the effects the environment
(physical and social) as well as personal factors (such as age, education, fear of injury, motivation)
may have on the person’s actions. As such, it is expected to set the scene for possible interven-
tions. Depending upon the assessed features of a particular individual patient, these interventions
may vary from one person to another. Thus, if environmental factors (e.g. excessively rigid
and unrealistic expectations for recovery by the workplace supervisor) appear to be acting as a
barrier to an injured worker’s rehabilitation then that would need to be addressed by the reha-
bilitation team as part of their intervention, possibly even through specific training for the work-
place supervisor (e.g. Shaw et al. 2006, 2008). Equally, if the injured worker displayed strong
Health condition
(disorder or disease)
Bodily functions Activities Participation

and structures (activity limitations) (participation restrictions)
(impairments)
Sitting Activities of daily living
Pain, strength, VO2max Standing work/school
Walking Leisure time activities
Stairs family/social life
External factors Personal factors

e.g. solicitous spouse, e.g. age, sex, education
employer attitudes coping style, catastrophizing, fear,
self-efficacy, motivation
Fig. 30.1 WHO framework.

PREVENTION REHABILITATION PROGRAMMES 567
catastrophic responses to his or her pain then they too would need to be targeted as part of the
intervention plan (e.g. Sullivan et al. 2006).
An important feature of the WHO model is that it indicates specific types of data that might be
collected and that, in turn, suggests the involvement of different disciplines in the assessment
process, as few individuals would be expected to have the skills required to assess all elements.
This provides overt encouragement for a multidisciplinary approach which would be expected to
include a formulation of the individual case, ideally at a team meeting, followed by preparation of
a treatment plan. This case formulation approach has a long history in behavioural therapies and
was recently outlined for application to pain patients by Linton and Nicholas (2008).
30.3 Importance of treatment settings and contexts

The studies describing early interventions for low back pain (LBP) are usually conducted in
primary care settings, such as general medical practitioner or private physiotherapy clinics, as well
as hospital outpatient physiotherapy departments (e.g. Hay et al. 2005; Jellema et al. 2005; Pengel
et al. 2007; Underwood et al. 2007). Typically, the focus of treatments in these settings is symptom
relief and return to normal functioning at work or home as soon as possible. But as the pain
becomes identified as more chronic and early interventions are not succeeding patients tend to be
seen in a wider range of settings and by a wider range of healthcare providers, with some variation
between countries and healthcare systems. Thus, different hospital departments, different spe-
cialists, and rehabilitation services staffed by a range of healthcare disciplines and even commer-
cial healthcare institutes specialized in early return to work are often introduced. These might
include orthopaedic surgeons, neurologists, anaesthesiologists, occupational or rehabilitation
medicine specialists, clinical psychologists, social workers, occupational therapists, nurses spe-
cialized in pain management, as well as physiotherapists and providers of numerous complemen-
tary therapies (eg. Blyth et al. 2005; Loisel et al. 2002; Skouen et al. 2006; Smeets et al. 2006).
The focus of interventions can also start to shift. Instead of symptom relief and rapid return to
normal lifestyle, the focus often moves towards symptom management and gradual return to
normal lifestyle and in very severe cases adapting another lifestyle with new activities and roles. In
many countries, not only may the treatment providers and the focus of treatment start to change
as pain becomes more chronic, but other interested parties, such as third-party (e.g. workers
compensation) insurers and the workers’ employer, may begin to enter the picture. As pain
becomes more chronic, while immediate return to work may become less of an expectation for
both the healthcare providers and the patients, the third-party funder and the employer may well
have quite different expectations. Franche et al. (2005a) tried to tease out a number of possible
expectations, or paradigms, which might be held by the key stakeholders (the patient, the health-
care providers, the employer, the insurance company, the union, etc.) and to explore how these
might influence outcomes. Potentially, where these paradigms differed or were in conflict, the
effectiveness of even treatments with good supporting evidence might be reduced.
The presence of a compensation claim, for instance, has been widely recognized as a potential
inhibitor of good treatment outcomes. This isn’t just in rehabilitation settings. A meta-analysis by
Harris (2005) found that the outcomes of surgery differed according to whether or not the patients
had a compensation claim, regardless of country or type of surgery. Where the intervention takes
place might also influence the outcomes achieved even when seemingly similar approaches are
being taken. For example, Loisel et al. (2002) found that treatment that included being at work for
workers with persisting back pain achieved fewer lost work days at long-term follow-up than
treatments that did not include the workplace. A systematic review by Franche et al. (2005b)
also found that when the treatment provider actively engaged with the workplace as part of the
treatment process, better return to work outcomes and reduced costs were achieved (see
Chapter 19 by Shaw et al. for further discussion of this topic).
The cultural and societal contexts can also play an important role in the way people with LBP
behave. The results of a large population-based intervention (via advertisements on television
and on bill boards by major roads) showed that changing the societal beliefs and expectations
regarding the management of LBP, especially the activity restrictions due to pain, reduced the
overall reporting of LBP, associated healthcare consumption and economic costs due to LBP
(Buchbinder and Jolley 2004). Nevertheless, as the positive results tended to decline in the space
of a few years once the advertising campaign ceased, reinforcement of the newly acquired beliefs
seems to be important.
In sum, when considering the management of patients who do not respond to early interven-
tion methods the tasks facing the treatment provider are likely to become more complex and
to involve more stakeholders. In fact, if this is not recognized by the treatment provider who
continues to treat the pain as if it were acute it is quite possible they run the risk of unintentionally
making things worse and prolonging their patient’s disability. These iatrogenic risks have
been well described by others (e.g. Loeser 1996), but the point when the pursuit of pain relief
should cease to be the focus and rehabilitation goals should become dominant will remain hotly
debated.
30.4 Timing of rehabilitation and decision rules

To date, while there is general agreement that rehabilitation approaches should be commenced as
early as possible, consensus is lacking on when that should be. In many countries there may be
systemic obstacles to early intervention that may effectively restrict access to early rehabilitation.
Nevertheless, even in the acute phase, it is argued that the aim of treatment providers should not
only be to reduce the pain but also check for potential risk factors for the development of chronic-
ity and associated disability (van der Windt et al. 2008). However, recent research has found that
when primary healthcare providers (general practitioners or physiotherapists) are specifically
trained to identify and address psychosocial risk factors for chronic disability, the results have
been rather disappointing (Hay et al. 2005; Jellema et al 2005). In part, this has been attributed to
the general practitioners having difficulty in identifying the psychosocial risk factors (Jellema
et al. 2005). In the study by Hay and colleagues (2005) no treatment integrity was reported, but
the therapists were provided training for only 2 days which might indicate that these rather inex-
perienced care-providers were not sufficiently trained to diagnose and treat psychosocial risk
factors. In these two studies also, the number of patients with psychosocial risk factors in patients
with (sub)acute complaints seemed rather low, diluting the potential effects of this intervention.
Timing is also strongly dependent on the financial resources available to the healthcare system
as well as the accessibility and availability of well equipped and experienced healthcare providers.
For example, in the Netherlands the access to a clinical psychologist in the primary care setting
has recently been improved due to a change in the healthcare system. Nevertheless, the number
of permitted sessions is limited and most patients have to pay additional fees and last but not
least, the number of psychologists sufficiently trained in secondary prevention of chronic pain
syndromes is too low to meet the demand. As a result many of these patients are being treated
longer than necessary by other primary healthcare providers, mostly using a pain-focussed rather
than a rehabilitation-focussed approach.
Given the evidence that untreated psychosocial factors are likely to lead to more patients with
chronic pain and associated disability, the net result is that pain management and rehabilitation
facilities become the last resort for an increasing number of people. But as there are relatively few
of these facilities, their capacity to meet the demand is rapidly overtaken and long waiting times
for assessment and treatment are commonplace. The delays involved can also pose a serious risk
for further deterioration in the well-being of these patients. As most compensation systems for
injured workers have built in time limits for return to work, so that those not back at work within
a set time (e.g. 6 months) may lose their jobs, the additional risk posed by long waiting times is
that these patients may lose their jobs while waiting for help. Once an injured worker loses his/her
job, the chances of return to work following treatment decline rapidly. Thus, outcomes for reha-
bilitation are usually much worse when dealing with unemployed workers. Ideally, timing for
rehabilitation in these systems should intervening before the worker has lost his/her job.
Identifying and targeting specific contributors to disability has long been argued as a logical
development in pain rehabilitation. Turk (1990), and more recently Vlaeyen and Morley (2005),
have argued that if we could select the right treatment for the right patient it would be both more
effective and potentially more cost-effective. This approach requires assessing patients with spe-
cific features, such as depression or fear-avoidance responses and offering them interventions
targeted more specifically at these features.
The use of decision rules that might guide the selection of suitable individuals for specific inter-
ventions has recently gained some momentum with the publication of a number of studies that
have tested different selection methods. This topic has been well-covered already in an earlier
chapter (Chapter 11), but some examples of applications are relevant here. These have ranged
from workplace characteristics (e.g. Shaw et al. 2006) to specific psychological risk factors, espe-
cially heightened catastrophizing (see Chapter 13). Some studies have also described successful
interventions in cases identified according to their overall psychosocial risk profile rather than
any particular characteristic. For example, Gatchel et al. (2003 used a statistical algorithm to select
acute LBP patients as high risk for developing chronic disability. These patients were randomly
assigned to either a cognitive-behavioural functional restoration group or a treatment-as-usual
group. At follow-up the functional restoration group displayed significantly fewer indices of
chronic pain disability on a wide range of work, healthcare utilization, medication use, and self-
reported pain variables. A recent Canadian study (Schultz et al. 2008) with subacute LBP workers
also found that selection of patients based on their overall risk profile was more effective than
usual care. Specifically, Schultz et al. showed that only those at high risk of protracted disability
required a more comprehensive biopsychosocial intervention. A similar finding with more
chronic cases was reported from a large Norwegian study by Haldorsen et al. (2002); in that study
it was only the cases with the poorest prognosis (identified by operational criteria) who required
more comprehensive pain rehabilitation.
Sullivan and Stanish (2003) also reported encouraging results with their community-based
programme (the Pain-Disability Prevention [PDP] Programme) with patients who had a mean
time off work of 7 months due to back pain. In that study the treatment by psychologists targeted
identified risk factors, such as catastrophizing responses to pain, heightened fear of movement or
re-injury, and depressed mood. Although not a randomized trial, a majority (60%) of PDP-
treated clients returned to work, compared to an 18% base rate of return (achieved previously).
Sullivan et al. (2005) showed that, of 215 injured workers who completed the PDP Programme,
treatment-related reductions in pain catastrophizing significantly predicted return-to-work
(RTW). A similar programme (the Progressive Goal Attainment Programme, PGAP) for primary
care professionals (mainly physiotherapists and occupational therapists) with individuals who
had been work disabled (on average just over 7 months) due to whiplash symptoms, and were
selected on the basis of the same heightened psychological risk factors as those employed in the
PDP Programme, achieved a 75% return to work rate compared to 50% in an earlier cohort with
usual treatment (Sullivan et al. 2006).
A number of controlled studies have reported similar positive findings for various psychosocial
interventions for selected patients with subacute LBP (Anema et al. 2007; Damush et al. 2003;
Hagen et al. 2000; Hasenbring et al. 1999; Hlobil et al. 2005; Karjalainen et al. 2005; Linton et al.
2005; Schiltenwolf et al. 2006; Staal et al. 2002; van den Hout et al. 2003) and while these are not
specifically trials of matching patients’ characteristics to interventions, they are consistent with
this idea.
In relation to persisting or chronic pain conditions, a number of reviews and meta-analyses of
randomized controlled trials (RCTs) have reported consistent findings that for patients with
disabling chronic pain conditions coordinated, multidisciplinary interventions based on cogni-
tive-behavioural methods provide a generally effective option for rehabilitation (Airaksinen et al.
2006; Guzman et al. 2001; McCracken and Turk 2002; Morley et al. 1999; Nielsen and Weir 2001).
In the main, these studies have demonstrated that quality of life indicators tend to improve, and
be maintained for up to a year following these interventions. However, the achievement of goals
like RTW were rarely addressed by many of these programmes (e.g. Williams et al. 1999). This
prompted one reviewer to write that the results in this domain were inconsistent (Guzman et al.
2001). Even so, some studies have demonstrated that reasonable RTW outcomes can be achieved,
but there are caveats.
One of the largest and best-designed RTW studies with a more chronic pain population was
described by Haldorsen et al. (2002). Importantly, the Haldorsen et al. study pointed to one of the
emerging trends in this area: the role of selection of suitable patients and targeting of identified
risk factors for disability. In that study, the patients (with LBP duration at least 2 months and up
to 2 years) were divided into three subgroups according to an operationally-defined algorithm for
prognosis (good, medium and poor prognoses). The individuals in these groups were then
randomly assigned to one of three levels of intervention: a light, advice-based approach by the
general practitioner, a brief, structured advice and activity programme, and a comprehensive,
multidisciplinary, cognitive-behavioural pain rehabilitation programme. The RTW results over
the following year clearly demonstrated a dose-problem interaction, with the poor prognostic
group having much stronger RTW outcomes after the comprehensive intervention relative to the
two lighter ‘doses’. On the other hand, the medium prognostic group did just as well with the
medium-level intervention as in the comprehensive intervention, and the good prognostic group
did just as well in the light, advice intervention. This indicated that it would be false economy to
direct all patients to the comprehensive programme, but if selected well enough good results could
be obtained from all three groups if they were assigned interventions according to assessed need.
Close examination of several other randomized controlled studies reveals a similar picture and
reinforces the idea of selecting the right dose for the right patient. For example, Williams et al. (1999)
showed that moderately disabled patients with mixed chronic pain conditions responded better to
a 4-week comprehensive inpatient programme and maintained these gains at 1-year follow-up over
a similarly disabled group who received a lighter dose in the form of 8 × 3-hour weekly sessions as
outpatients, who still improved relative to baseline and a group receiving standard care from their
general practitioner. Also, Marhold and Linton (2001) found that more disabled chronic back pain
patients responded less well to a brief (12 × 2 hour) outpatient cognitive-behavioural programme
compared to a group of less chronic and less disabled back pain patients.
Although the contingencies and philosophies operating in many clinical settings will tend to
mean that clinicians will try to do what they can with what they have available, the evidence is
building to suggest that some patients will have a much better chance of successful rehabilitation
if they receive a dose of treatment proportionate to their need. This means that vigorous repre-
sentations may need to be made to ensure this happens. Recent experiences in the US health
maintenance environment provides testimony to this imperative (Robbins et al. 2003).
There is certainly growing evidence that patients can be clustered according to various dimen-
sions and that these can have different implications for outcomes (e.g. Boersma and Linton 2005),
but studies in which patients are assigned to interventions according to their sub-group charac-
teristics remain limited. Perhaps the closest example is provided by the Haldorsen et al. (2002)
study. Clearly there is an urgent need to replicate and extend these findings in other countries.
Overall, it is still the case that while the logic of matching patients to treatments seems compel-
ling and the available evidence is supportive, the lack of RCTs evaluating the approach is a hand-
icap and must be a priority in the next few years. Fortunately, some studies of this type are
underway and some answers should be available shortly. But more will be needed.
30.5 Integrating/combining with other treatment modalities

As noted in the earlier section on the implications of a biopsychosocial perspective for managing
persisting pain, a multidimensional assessment will often indicate that multiple interventions
might be appropriate in individual cases. While much of the research literature on interventions
seems to focus on the delivery of single modality interventions (e.g. a trial of an analgesic or an
exercise programme), the reality in clinical practice is that multiple interventions will be consid-
ered and a key question is whether they should be delivered in series (with the following interven-
tions depending upon the outcome of the earlier ones) or in parallel (i.e. simultaneous delivery).
Typically, these would be directed against different targets, but before commencing such a multi-
dimensional intervention process adequate preparation of the patient, in addition to the develop-
ment of a coordinated plan, would seem appropriate. This is especially since different interventions
can easily appear at cross purposes. For example, a patient could be offered a drug to relieve pain
while also being encouraged to accept the presence of the pain and to exercise despite the pain.
Traditionally, many have taken the approach of waiting for adequate analgesia to be established
before commencing rehabilitation. Collett (2001) described this as ‘capitalising on improved
analgesia’ (p. 140). Although this expectation may hold with acute pain, like postsurgical cases
(Stein 2000) and with cancer pain (Bruera et al. 1999), the evidence for patients with chronic
non-cancer pain still seeking help for their pain despite repeated trials of medication is not so
supportive of this position (Nicholas et al. 2006).
Unfortunately, there are relatively few good examples of combined interventions in the pain
literature and the vast majority of treatment studies report outcomes from one intervention per-
formed in isolation (Turk 2001). In some of these cases the researchers report significant improve-
ments across a range of outcome domains, but generalizing from these examples requires caution.
For example, Dreyfuss et al. (2000) reported improvements in disability with radiofrequency
neurotomy for patients with chronic lumbar zygapophysial joint pain (a procedure aimed at pain
relief), but examination of the disability level of their sample on the Roland and Morris Disability
Questionnaire (Roland and Morris 1983), revealed that the mean score at pretreatment was 7.
This reflects only mild disability and is well below the levels typically reported in those attending
multidisciplinary pain rehabilitation settings (Nicholas et al. 2008). De Jong et al. (2005), for
example, reported a mean score of 15 on the Roland and Morris Questionnaire at pretreatment.
This means we should be cautious about generalizing from the Dreyfuss study as the radiofre-
quency procedure has not been demonstrated in more disabled chronic pain populations.
Nevertheless, it would be interesting to see if a combination of radiofrequency neurotomy and a
pain management programme would yield better results than either alone in the more disabled
populations. To date, this has not been reported.
Despite the inherent difficulties in providing combined treatment modalities, there is emerging
evidence that a strategy of combining treatments aimed at different targets in chronic pain patients
can lead to better results than a single treatment on its own. In an uncontrolled trial Molloy et al.
(2004) reported significant gains in mood and disability for patients diagnosed with failed back
surgery syndrome when they were treated with a combination of an implanted spinal cord stimu-
lator or a morphine pump plus a comprehensive 3-week, multidisciplinary cognitive-behavioural
pain management programme rather than either an implant or a programme alone. This suggests
that such combined approaches are possible, but no RCTs have yet been reported.
There are studies that have combined pharmacotherapy and rehabilitation interventions. For
example, Holroyd et al. (1995, 2001) found that the combination of propranolol hydrochloride
with relaxation plus biofeedback, as well as tricyclic antidepressants plus stress management
training was more effective for reducing migraine headache activity than either of these interven-
tions when used alone. Similarly, Kishino et al. (2000) reported that a combination of a psycho-
logical intervention that targeted patient motivation and exercise was superior to a graded exercise
programme in patients with chronic back pain seen following spinal surgery. In patients with
rheumatoid arthritis, Leibing et al. (1999) found that the combination of pharmacological and
cognitive-behavioural treatment (CBT) achieved greater improvements on measures of depres-
sion, helplessness, affective pain scores and improved coping or adjustment compared to stand-
ard rheumatological (pharmacological) treatment. As a result the authors recommended the use
of CBT as an effective adjunct to standard (rheumatological) treatment of rheumatoid arthritis
outpatients. Similar findings were reported by Sharpe et al. (2001) following a randomized
controlled trial with patients who had recent onset (less than 2 years) seropositive rheumatoid
arthritis. To date, however, despite repeated calls (e.g. Haythornthwaite 2005; Turk 2001) there is
a lack of trials of combined pharmacotherapy and multidisciplinary pain management in patients
with chronic LBP.
It can be argued that many multidisciplinary CBT programmes are a form of combined inter-
vention (e.g. van Tulder et al. 2000), but while they combine members of different disciplines in
a coordinated way, typically they are driven by a self-management and rehabilitation philosophy
and all elements are focused on equipping the participants to function as independently as pos-
sible of healthcare providers. This usually means medication is withdrawn (e.g. Nicholas 2008)
rather than introduced, and that physical therapy involvement is more like that of a trainer than
a ‘hands-on’ therapist (like manipulation). In this section we have defined combined interven-
tions as those that target different aspects of a case simultaneously, based on a biopsychosocial
assessment and case-formulation. These may include pharmacological or surgical interventions
(i.e. dependent upon the healthcare provider) being provided alongside self-management
approaches. While there is broad evidence in support of this approach, some of which was
described earlier, there is clearly a need for more applied studies. In a review of outcomes from
multidisciplinary pain clinics Clark (2000) found that multidimensional interventions were more
effective across a range of dimensions compared to single modality (mostly medical) treatments
alone. These findings should not be that surprising, but they do not overcome the need for more
studies designed to test the combined hypothesis specifically. Of course, at a practical level the
obstacles facing those trying to integrate different treatment modalities should not be underesti-
mated. In addition to the challenges of grappling with the content of treatment packages, the
treatment providers still face the task of integrating the treatment processes with the patient’s
social environment. This will be addressed in the next section.
30.6 Integrating with the social environment

Although most pain rehabilitation is conducted under the banner of a biopsychosocial model,
recent reviews have questioned the degree to which the social (or environmental component) is
included in intervention studies (Blyth et al. 2007). However, there is growing evidence for the
value of including this often neglected dimension in pain rehabilitation. Much of the stimulus for
this seems to have come from occupational rehabilitation, but some have been exploring the util-
ity of including patients’ families in the treatment process as well (Newton-John, 2002). The same
applies for healthcare professionals involved in the care of patients with chronic pain related dis-
ability, especially since it is evident that their attitudes and beliefs influence treatment content and
outcome. For example, clinicians with more biomedical orientations tend to give advice which
results in a less active lifestyle (Bishop et al. 2008; Houben et al. 2005).
One of the early studies of occupational rehabilitation noted that better RTW outcomes were
achieved when employers had what they termed ‘proactive RTW policies’ (Hunt and Harbeck
1993). That is, they had policies that tried to accommodate to the needs of injured workers and to
facilitate their return to work rather than waiting for a medical interventions to do it for them.
More recently, in a systematic review of this literature Franche et al. (2005a) concluded that there
was moderately strong evidence from seven studies to support the value of liaison between
healthcare providers and their patients’ workplaces in facilitating RTW. The importance of
including the workplace in rehabilitation for injured workers was further reinforced by the expe-
rience of McCluskey et al. (2006) who found their guideline-based return to work programme for
injured workers was effectively undermined by the workplace’s policies. Similarly, Shaw et al.
(2006) have demonstrated that providing specific training for workplace supervisors was very
effective in reducing workplace disability with injured workers.
Despite this evidence there is surprisingly little information in the pain rehabilitation litera-
ture of efforts to incorporate linkages to the workplace as an element of this work. In part, this
may be due to health system limitations within many countries, especially those where return to
work is not seen as a goal of healthcare (e.g. Waddell et al. 2002). To the extent that this means
direct efforts to help injured workers return to work may be discouraged then it should not be
surprising that links to workplaces are rarely included in many public hospital pain rehabilita-
tion services (e.g. McCracken et al. 2007; Williams et al. 1996). However, given the known
benefits of work on health and wellbeing it would seem an important area for expansion in
these types of programmes in the future (Waddell et al. 2003). In some countries the public
authority responsible for injured workers has directly funded pain rehabilitation services. In
Canada, for example Tschernetzki-Neilson et al. (2007) described improved RTW outcomes
achieved by the services in one state when the focus of rehabilitation was shifted from processes
and functional capacity to RTW. This study underscored the point that system-level changes
may be needed to bring about some changes in rehabilitation outcomes (see Chapter 19 for a
fuller discussion).
While the logic of this extension of pain rehabilitation from the clinic to the workplace (and
home) is not difficult to grasp, reliably delivering it in practice is a major problem in many health
systems. This is especially the case where health has traditionally been closely defined within a
biomedical perspective or paradigm. As Franche et al. (2005b) pointed out, the paradigm for
healthcare providers has typically focussed on the individual patient, his or her medical status,
and treatments accepted under local funding arrangements. RTW in this paradigm is left to the
patient to undertake once they are medically ‘better’. There is usually no direct benefit for the
healthcare funders if the patient returns to work. In the UK, for example, the National Health
Service is responsible for funding health services, but whether a patient returns to work or not
makes no difference to the health service funding, though it may to the Department of Work and
Pensions which is responsible for pensions if the person does not RTW (Waddell et al. 2002).
However, the Department of Work and Pensions makes no contribution to the funding of health
services, so they cannot differentially support treatments that might help someone RTW.
There is clearly an imperative for considering major social outcomes, such as RTW by injured
workers, beyond the specific biomedical, diagnostic framework, regardless of the site of pain. This
calls for approaches that can accommodate a real biopsychosocial perspective, not just a bio-
psycho version that is so often presented (e.g. Blyth et al. 2007). To date, there are no examples in
the pain literature of outcomes from comprehensive initiatives that target the full range of factors
that may contribute to ongoing pain-related disability. However, there are examples of national and
state-based rehabilitation or injury management systems that have reported attempts to engage in
these processes (e.g. Kendall et al. 2004; Tschernetzki-Neilson et al. 2007; WorkCoverNSW 2008).
30.7 Research agenda: the future

Multidisciplinary pain management programmes have been proven effective, but effect sizes are
typically only moderate. Possible reasons for this have been advanced but definitive evidence for
interventions that reliably achieve better outcomes remains scarce. The case for more compre-
hensive interventions that might include multidisciplinary programmes, but other facets as well,
has been argued here as a logical response to the shortcomings of the current literature. However,
good evidence for the superiority of more comprehensive approaches over more narrowly
focussed approaches is still limited. Despite the inherent difficulties of large scale research on
comprehensive approaches it is hard to see how this field can make further real progress without
such projects. This calls for researchers in this area to move out of the clinic and to negotiate with
those responsible for broader aspects of the injury management and rehabilitation systems.
Fortunately, there are precedents upon which we can draw at least inspiration (e.g. Loisel et al.
2002), but the difficulties should not be underestimated—as some have found to their cost (e.g.
McClusky et al. 2006).
Despite the clear need for evaluation of more comprehensive interventions over single modal-
ity approaches, at the individual level, results vary greatly and for a proportion of individuals this
treatment is not successful at all. The question of ‘What works for whom?’ remains only partially
answered. Some individually targeted interventions for certain risk factors have been described
and these do seem promising but more of these studies are urgently needed. Many studies have
suffered from insufficient power to explore these questions, but it is also likely that interventions
that are too narrowly focussed may miss important individual targets (e.g. just focussing on fear
of movement through graded activity when over-reliance of medication may also need attention).
In the 9 years since Haldorsen et al. (2002) published their study comparing level of intervention
with level of prognostic signs, no other study has come near the degree of ‘real world’ testing of
multidisciplinary interventions that group achieved. More of these sorts of studies must be a high
priority. This review has also found remarkably few studies that investigate the integration of
multiple, simultaneous interventions, particularly combining psychological-behavioural and
medical-surgical regimes, still appear rarely in the research literature even though they are com-
mon in clinical practice. There are good examples in the literature which ought to be amenable to
replication or adaptation (e.g. Leibing et al. 1999; also see Haythornthwaite et al. 2005). Innovative
designs, such as N = 1 methods, might represent a means for further exploration of this topic in
clinically meaningful ways (e.g. Linton and Nicholas 2008).
This review also raises the question of should we expect there to be a treatment to help every-
one? As pointed out earlier (and in Chapters 19 and 24), work-related issues and motivational
issues are likely to remain significant barriers to successful rehabilitation programmes, largely
because they cannot be reliably controlled by the providers of these interventions. At a pragmatic
level, if providers cannot change the system in which they work then identifying those individuals
who are likely to respond to current interventions, given their limitations, remains a worthwhile
focus for research, despite limited results to date (Underwood et al. 2007). The corollary in this
regard is to explore ways of helping those who are unlikely to benefit from available options. This
might mean shifting the focus from pain management or pain rehabilitation per se, to dealing
with interpersonal or family issues, as exemplified in work on communications and anger
management with pain populations (e.g. Greenwood et al. 2003).
While consensus on the content, intensity and duration of treatment modalities necessary for
successfully preventing disability remains elusive, there is growing evidence for specific features
of these interventions for particular patient subgroups (Guzman et al. 2001; Haldorsen et al. 2002;
Smeets et al. 2006). Long-term effectiveness and adherence to behavioural change remains a
challenge that is likely to be beyond the resources of most pain rehabilitation programmes in
isolation. The analysis presented here (and elsewhere in this book) suggests that answers to these
questions are likely to come from work on patient selection for programmes, tailored interven-
tions, and the peri-programme environment, such as the workplace, other treatment providers,
and the broader context of the rehabilitation system, rather than a healthcare facility alone.
References
Airaksinen, O., Brox, J.I., Cedraschi, C. et al. (2006). European guidelines for the management of chronic
nonspecific low back pain. European Spine Journal, 15 (Suppl. 2), S192–S300.
Anema, J.R., Steenstra, I.A., and Bongers, P.M. (2007). Multidisciplinary rehabilitation for subacute low
back pain: graded activity or workplace intervention or both? A randomized controlled trial. Spine,
32, 291–8.
Bishop, A., Foster, N., Thomas, E. et al. (2008). How does the self-reported clinical management of patients
with low back pain relate to the attitudes and beliefs of health care practitioners? A survey of UK
general practitioners and physiotherapists. Pain 135, 187–95.
Blyth, F.M., March, L., Nicholas, M.K. and Cousins, M.J. (2005). Self-management of chronic pain:
a population-based study. Pain, 113, 285–92.
Blyth, F.M., Macfarlane, G.J., and Nicholas, M.K. (2007). Understanding the role of psychosocial factors in
the development of chronic pain: the key to better outcomes for patients? Pain, 129, 8–11.
Boersma, K., and Linton, S.J. (2005). Screening to identify patients at risk profiles of psychological risk
factors for early intervention. Clinical Journal of Pain, 21, 38–43.
Boersma, K. and Linton, S.J. (2006) Psychological processes underlying the development of a chronic pain
problem: a prospective study of the relationship between profiles in the fear-avoidance model and
disability. Clinical Journal of Pain 22, 160–6.
Bruera, E., Walker, P., and Lawlor, P. (1999). Opioids in cancer pain. In Stein, C. (Ed.). Opioids in Pain
Control: Basic and Clinical Aspects, pp. 309–24. Cambridge University Press, Cambridge.
Buchbinder. R., and Jolley, D. (2004). Population based intervention to change back pain beliefs: three year
follow up study. British Medical Journal, 328, 321.
Clark, T.S. (2000). Interdisciplinary treatment for chronic pain: is it worth the money? Baylor University
Medical Center Procedings, 13, 240–3.
Collett, B.J. (2001). Chronic opioid therapy for noncancer pain. British Journal of Anaesthesia, 87, 133–44.
Damush, T.M., Weinberger, W., Perkins, S.M. et al. (2003). The long term effects of a self-management
program for inner city primary care patients with acute low back pain. Archives of Internal Medicine,
163, 2632–8.
de Jong, J.R., Vlaeyen, J.W.S., Onghena, P., Goosens, M.E.J.B., Geilen, M. and Mulder, H. (2005). H. Fear
of movement/reinjury in chronic low back pain. Education or exposure in vivo as mediator to fear
reduction. Clinical Journal of Pain, 21, 9–17.
Dreyfuss, P., Halbrook, B., Pauze, K., Joshi, A., McLarty, J. and Bogduk, N. (2000). Efficacy and validity of
radiofrequency neurotomy for chronic lumbar zygapophysial joint pain. Spine, 25, 1270–7.
Franche, R.-L., Cullen,K., Clarke, J., Irvin, E., Sinclair, S. and Frank, J. (2005a). Workplace-based return to
work interventions: A systematic review of the quantitative literature. Journal of Occupational
Rehabilitation, 15(4), 607–32.
Franche, R-L., Baril, R., Shaw, W., Nicholas, M.K. and Loisel, P. (2005b). Workplace-based return-to-work
interventions: Optimizing the role of stakeholders in implementation and research. Journal of
Occupational Rehabilitation 15(4), 525–42.
Gatchel, R.J., Polatin, B.P., Noe, C. et al. (2003). Treatment- and cost-effectiveness of early intervention
for acute low-back pain patients: A one-year prospective study. Journal of Occupational Rehabilitation,
13, 1–9.
Greenwood, K.A., Thurston, R., Rumble, M.,Waters, S.J. and Keefe, F.J.(2003). Anger and persistent pain:
current status and future directions. Pain, 103, 1–5
Guzman, J., Esmail, R., Karjalainen, K., et al. (2001). Multidisciplinary rehabilitation for chronic low back
pain: systematic review. British Medical Journal, 322, 1511–16.
Hagen, E.M., Eriksen, H.R., and Ursin, H. (2000). Does early intervention with a light mobilization
program reduce long-term sick leave for low back pain? Spine, 25, 1973–76.
Haldorsen, E.M.H., Grasdal, A.L., Skouen, J.S. et al.(2002). Is there a right treatment for a particular
patient group? Comparison of ordinary treatment, light multidisciplinary treatment, and extensive
multidisciplinary treatment for long-term sick-listed employees with musculoskeletal pain. Pain,
95, 49–63.
Harris, I.A. (2005). Association between compensation status and outcome after surgery: a meta-analysis.
Journal of the American Medical Association, 293, 1644–52.
Hasenbring, M., Ulrich, H.W., Hartmann, M. et al. (1999). The efficacy of a risk factor based cognitive
behavioral intervention and EMG-biofeedback in patients with acute sciatic pain: an attempt to
prevent chronicity. Spine 24 (23), 2525–35.
Hay, E.M., Mullis, R., Lewis, M. et al. (2005). Comparison of physical treatments versus a brief pain
Haythornthwaite, J.A. (2005). Clinical trials studying pharmacotherapy and psychological treatments and
together. Neurology, 65(suppl 4), S20–S31.
Hlobil, H., Staal, B.J., Twisk, J. et al. (2005). The effects of a graded activity intervention for low back pain
in occupational health on sick leave, functional status and pain: 12-month results of a randomized
controlled trial. Journal of Occupational Rehabilitation, 15, 569–80.
Holroyd, K.A., O’Donnell, F.J., Stenland, M. et al. (2001). Management of chronic tension-type headache
with tricyclic antidepressant medication, stress management therapy, and their combination - a
randomized controlled trial. Journal of the American Medical Association, 285, 2208–17.
Holroyd, K.A., France, J.L., Cordingley, G.E., et al. (1995). Enhancing the effectiveness of
relaxation-thermal biofeedback with propranolol hydrochloride. Journal of Consulting and Clinical
Houben, R.M.A., Ostelo, R.W.J.G., Vlaeyen, J.W.S. et al. (2005) Health care providers’ orientations
towards common low back pain predict perceived harmfulness of physical activities and
recommendations regarding return to normal activity. European Journal of Pain, 9, 173–83.
Hunt, A. and Harbeck, R. (1993). The Michigan Disability Prevention Study. WE Upjohn Institute for
Employment Research, Kalamazoo, MI.
Jellema, P., van der Windt, D.A.M.W., van der Horst, H.E., Twisk, J.W.R., Stalman, W.A.B. and Bouter,
L.M. (2005). Should treatment of (sub)acute low back pain be aimed at psychosocial prognostic
factors? Results of a cluster-randomised clinical trial in general practice. British Medical Journal,
331, 84–90.
Karjalainen, K., Malmivaara, A., Mutanen, P., Roine, R., Hurri, H. and Pohjolainen, T. (2004).
Mini-intervention for subacute low back pain two-year follow-up and modifiers of effectiveness.
Spine, 29, 1069–76.
Kendall, N.A.S., Linton, S.J., and Main, C.J. (2004). Guide to Assessing Psych-social Yellow Glagsin Acute
Low Back Pain: Risk Factors for Long-Term Disability and Work Loss. Accident Compensation
Corporation and the New Zealand Guidelines Group, Wellington, New Zealand
Kishino, N.D., Polatin, P.B., Brewer, S. and Hoffman, K. (2000). Long-term effectiveness of combined
spinal surgery and functional restoration: a prospective study. Journal of Occupational Rehabilitation,
10, 235–9.
Leibing, E., Pfingsten, M., Bartman, U., Leibing, U.R. and Gerhard, S. (1999). Cognitive-behavioral
treatment in unselected rheumatoid arthritis outpatients. Clinical Journal of Pain, 15, 58–66.
Linton, S.J., Boersma, K., Jansson, M., Svärd, L., and Botvalde, M. (2005). The Effects of
Cognitive-Behavioral and Physical Therapy Preventive Interventions on Pain-Related Sick Leave:
A Randomized Controlled Trial. Clinical Journal of Pain, 21, 109–19.
Linton, S.J. and Nicholas, M.K. (2008). After assessment, then what? Integrating findings for successful case
formulation and treatment tailoring. In Breivik, H., Campbell, W.J., Nicholas, M.K. (Eds.) Practice and
Procedures, Textbook of Clinical Pain Management 2nd edition, Hodder Arnold, New York, pp. 95–106.
Loeser, J.D. (1996). Mitigating the dangers of pursuing cure. In: Cohen, M.J.M., Campbell, J.N. (eds.) Pain
Treatment Centers at a Crossroads: A Practical and Conceptual Reappraisal. IASP Press, Seattle, WA,
pp. 101–8.
Loisel, P., Lemaire, J., Poitras, S. et al. (2002). Cost-benefit and cost-effectiveness analysis of a disability
prevention model for back pain management: a six year follow up study. Occupational and
Marhold, C., Linton, S.J. and Melin, L. (2001).A cognitive-behavioral return-to-work program: effects on
pain patients with a history of long-term versus short-term sick leave. Pain, 91, 155–63
McCluskey, S., Burton, A.K. and Main, C.J. (2006). The implementation of occupational health guidelines
principles for reducing sickness absence due to musculoskeletal disorders. Occupational Medicine, 56, 237–42.
McCracken, L.M. and Turk, D.C. (2002). Behavioral and cognitive-behavioral treatment for chronic pain:
outcome, predictors of outcome, and treatment process. Spine, 27(22), 2564–73.
McCracken, L. M., MacKichan, F. and Eccleston, C. (2007). Contextual cognitive-behavioral therapy for
severely disabled chronic pain sufferers: Effectiveness and clinically significant change. European
Molloy, A.R., Nicholas, M.K., Asghari, A. et al. (2006). Does a combination of intensive
cognitive-behavioural pain management and a spinal implantable device confer any advantage?
A preliminary examination. Pain Practice, 6, 96–103.
Morley, S., Eccleston, C. and Williams, A. (1999). Systematic review and meta-analysis of randomized
excluding headache. Pain, 80(1–2), 1–13.
Newton-John, T.R.O. (2002) Solicitousness and chronic pain: a critical review. Pain Rev, 9(1), 7–27.
Nicholas, M.K. (2008). Pain management in musculoskeletal conditions. Best Practice in Research and
Clinincal Rheumatology, 22(3), 451–70.
Nicholas, M.K., Molloy, A.R. and Brooker, C. (2006). Using opioids with persisting, noncancer pain:
A biopsychosocial perspective. Clinical Journal of Pain, 22(2), 137–46.
Nicholas, M.K., Asghari, A. and Blyth, F.M. (2008). What do the numbers mean? Normative data in
chronic pain measures. Pain, 134, 158–73.
Nielson, W. and Weir, R. (2001). Biopsychosocial approaches to the treatment of chronic pain. Clinical
Journal of Pain, 17, S114–127.
Pengel, L.H.M., Refshauge, K.M., Maher, C.G., Nicholas, M., Herbert, R.D. and McNair, P. (2007).
Physiotherapist-directed exercise, advice, or both for sub-acute low back pain: a randomized controlled
trial. Annals of Internal Medicine, 146(11), 787–96.
Robbins, H., Gatchel, R.J., Noe, C. et al. (2003). A prospective one-year outcome study of interdisciplinary
chronic pain management: Compromising its efficacy by managed care policies. Anesthesia and
Analgesia, 97, 156–62.
Roland, M. and Morris, R. (1983). A study of the natural history of back pain. Part 1.Development of a
reliable and sensitive measure of disability in low-back pain. Spine, 8, 141–4.
Schiltenwolf, M., Buchner, M., Heindl, B., von Reumont, J., Müller, A., and Eich, W. (2006). Comparison
of a biopsychosocial therapy (BT) with a conventional biomedical therapy (MT) of subacute low back
pain in the first episode of sick leave: a randomized controlled trial. European Spine Journal, 15,
1083–92.
Schultz, I.Z., Crook, J., Berkowitz, J., Milner, R., Meloche, G.R. and Lewis, M.L. (2008). A prospective
study of the effectiveness of early intervention with high-risk back-injured workers – a pilot study.
Sharpe, L., Sensky, T., Timberlake, N., et al. (2001). A blind, randomized, controlled trial of
cognitive-behavioural intervention for patients with recent onset rheumatoid arthritis: preventing
psychological and physical morbidity. Pain, 89, 275–83.
Shaw, W.S., Robertson, M.M., McLellan, R.K., Verma, S. and Pransky, G. (2006). A controlled case study
of supervisor training to optimize response to injury in the food processing industry. Work, 26,107–14.
Shaw, W.S., Quan-nha Hong, Q., Pransky, G. and Loisel, P. (2008). A Literature Review Describing the
Role of Return-to-Work Coordinators in Trial Programs and Interventions Designed to Prevent
Workplace Disability. Journal of Occupational Rehabilitation, 18, 2–15
Skouen, J.S., Grasdal, A. and Haldorsen, E.M.H. (2006). Return to work after comparing outpatient
multidisciplinary treatment programs versus treatment in general practice for patients with chronic
widespread pain. European Journal of Pain, 10, 145–52.
Smeets, R.J.E.M., Vlaeyen, J.W.S., Kester, A.D.M. and Knottnerus, J.A. (2006). Reduction of Pain
Catastrophizing Mediates the Outcome of Both Physical and Cognitive-Behavioral Treatment in
Chronic Low Back Pain. Journal of Pain, 7(4), 261–71.
Staal, J.B., Hlobil, H., van Tulder, M.W., Koke, A.J.A., Smid, T., and van Mechelen, W. (2002). Return-
to-work interventions for low back pain. A descriptive review of contents and concepts of working
mechanisms. Sports Medicine, 32, 251–67.
Stein, C. (2000). What’s wrong with opioids in chronic pain? Current Opinion in Anesthesiology, 13, 557–9.
Sullivan, M.J. and Stanish, W.D. (2003). Psychologically based occupational rehabilitation: The
Pain-Disability Prevention Program. Clinical Journal of Pain, 19, 97–104.
Sullivan, M.J., Ward, L.C., Tripp, D., Adams, H. and Stanish, W.D. (2005). Secondary prevention of work
disability: Community-based psychosocial intervention for musculoskeletal disorders. Journal of
Occupational Rehabilitation, 15(3), 377–92.
Sullivan, M.J., Adams, H., Rhodenizer, T. and Stanish, W.D. (2006). A psychosocial risk factor-targeted
intervention for the prevention of chronic pain and disability following whiplash injury. Physical
Therapy, 86(1), 8–18.
Tschernetzki-Neilson, P.J., Brintnell, S., Haws, C. and Graham, K. (2007). Changing to an
outcome-focused program improves return to work outcomes. Journal of Occupational Rehabilitation,
17, 473–86.
Turk, D.C. (1990). Customizing treatment for chronic pain patients: who, what, and why. Clinical Journal
of Pain, 6(4), 255–70.
Turk, D.C. (2001). Combining somatic and psychosocial treatment for chronic pain patients: perhaps 1 + 1
does = 3. Clinical Journal of Pain, 17(4), 281–3.
Turk, D.C. and Okifuji, A. (1998). Treatment of chronic pain patients: clinical outcomes, cost-effectiveness,
and cost-benefits of multidisciplinary pain centers. Critical Reviews in Physical Rehabilitation Medicine,
10, 181–208.
Underwood, M.R., Morton, V. and Farrin, A. (2007). Do baseline characteristics predict response to treatment
for low back pain? Secondary analysis of the UK BEAM dataset. Rheumatology, 46(8), 1297–302.
van den Hout, J.H., Vlaeyen, J.W., Heuts, P.H., Zijlema, J.H. and Wijnen, J.A.(2003). Secondary
prevention of work-related disability in nonspecific low back pain: Does problem-solving therapy help?
A randomized clinical trial. Clinical Journal of Pain, 19, 87–96.
van der Windt, D., Hay, E., Jellema, P. and Main, C.J.(2008). Psychosocial interventions for low back pain
in primary care: Lessons learned from recent trials. Spine, 33, 81–89.
van Tulder, M.W., Ostelo, R.W.J.G., Vlaeyen, J.W.S., et al. (2000). Behavioural treatment for chronic
low-back pain. The Cochrane Database of Systematic Reviews, 2, CD002014.
Vlaeyen, J.W.S., de Jong, J., Geilen, M., Heuts, P.H.T.G. and van Breukelen, G. (2002). The treatment of
fear of movement/(re)injury in chronic low back pain: further evidence on the effectiveness of
exposure in vivo. Clinical Journal of Pain, 18, 251–61.
Vlaeyen, J.W.S. and Morley, S. (2005). Cognitive-Behavioral Treatments for Chronic Pain: What Works
for Whom? Clinical Journal of Pain, 21, 1–8.
Waddell, G., Aylward, M. and Sawney, P. (2002). Back pain, incapacity for work and social security benefits:
An international literature review and analysis. The Royal Society of Medicine Press, London.
Waddell, G., Burton, A. K. and Main, C. J. (2003). Screening of DWP clients for risk of long-term incapacity:
a conceptual and scientific review. Royal Society of Medicine Monograph; Royal Society of Medicine
Press, London.
WHO. (2001). The International Classification of Functioning, Disability and Health (ICF). 2nd ed. WHO,
Geneva.
Williams, A.C. Nicholas, M.K., Richardson, P.H., Pither, C.E. and Fernandes, J. (1999). Generalizing from
a controlled trial: the effects of patient preference versus randomization on the outcome of inpatient
versus outpatient chronic pain management. Pain, 83(1), 57–65.
Williams, A.C., Richardson, P.H., Nicholas, M.K. et al. (1996). Inpatient versus outpatient pain
management: results of a randomised controlled trial. Pain, 66, 13–22.
WorkCover NSW. (2006).Management of Soft Tissue Injuries (General Guide, Insurer Guide and version for
Treatment Providers), Locked Bag 2906, Lisarow, NSW 2252, Australia.
Index
acceptance and commitment therapy (ACT) 234–5, Avoidance of Physical Activities 298
328, 548, 554–5 Avoidance of Social Activities 298
see also cognitive-behavioural therapy, contextual
aceclofenac 475 Back Beliefs 173
acetaminophen 475 Back Disability Risk Questionnaire (BDRQ) 380
activities of daily living 355, 366, 517, 565 back pain myths 445
actuarial screening methods 212–13 Back Schools 434
adaptive coping strategies 81, 256, 259, 323, 435, baroreceptor sensitivity 138–9
458, 465 CNS input 139–40
adaptive response pattern 297, 300–1, 303 baroreflex mechanism 138
adrenal cortex 71 Beck Depression Inventory 79, 224, 303, 523
adrenocorticotrophic hormone (ACTH) 71 bed rest, lack of efficacy 188, 189, 193, 238, 398, 405–6
aerobic capacity 190–3 behaviour change 445–6
affect, dynamic model 297 Behavioural Endurance Scale (BES) 224, 302
Agency for Health Care Policy and Research 419, 422 behavioural experiments 285
aggrecan 46 behavioural response measurement 302
AGREE instrument 420–2 beliefs regarding back pain
allodynia 24 patients 441
allostasis 70 physicians 405–15, 442–3
allostatic load 70–1 see also back pain myths
American College of Physicians, American Pain benzodiazepines 475, 478
Society Low Back Pain Guidelines Panel 420, 422 beta-endorphin 10
anandamide 57 Biering-Sorensen test 157, 158, 161, 172, 176
anger 118, 133, 136 biofeedback 115, 523, 572
ankylosing spondylitis 48 biomechanical models 116–17
anterior cingulate cortex 89, 90–1, 109 biopsychosocial pain model 270, 520–1
anterior pituitary gland 71 black flags 206, 381, 447
anticholinergics 475 see also flags framework
anticipatory postural adjustments 125 blood pressure, and pain sensitivity 137
antidepressants 473, 478–9 blue flags 207, 380, 381, 447
anxiety sensitivity 231, 235–6 see also flags framework
and pain-related anxiety 236–7 body composition 187–8
Anxiety Sensitivity Index (ASI) 235 body mass index 173
assessment 283 BOLD responses 62–3
exposure in vivo 283 bone strength 187–8
pain-related fear/anxiety 234, 277–82 brain
psychosocial risk 517–25 grey matter changes 110
attention 275, 296 structural changes 94–5, 105–14
and pain perception 99, 254–5 cause vs. consequence 110–11
attention bias 276, 281, 316–17 chronic pain 106
attentional load hypothesis 307 cortical reorganization 105
attentional network 254–5 specificity of 108–10
Attitudes to Back Pain Scale for Musculoskeletal spontaneous pain 106
Practitioners (ABPS-MP) 408 structural imaging 94–5, 106–8
attorneys, involvement in workers’ compensation chronic back pain 107–8
cases 369 white matter connectivity 110
see also disability see also different regions
autopoietic symptom perpetuation 69, 70 brain-computer interface 115
avoidance behaviour 271, 274–5 brain-derived neurotrophic factor (BDNF) 61
avoidance-endurance model (AEM) 224, 295–314, 498 buprenorphine 478
basic assumptions 297–8
endurance-related components 295–7, 301–3 C-fibres, mechanically insensitive 56
subgroups 303–4 calcitonin-gene related peptide 56
see also fear-avoidance cannabinoids, endogenous 45
Avoidance-Endurance Questionnaire (AEQ) cannabis 45
224, 298, 302 capsaicin 57
582 INDEX
cardiac output 190 prognostic approach 25–7

cardiovascular response to pain 137–40 social context 555–8
Care and Research Institute Italy 420, 422 static vs. dynamic 28–9
cartilage intermediate layer protein 47 Chronic Pain Acceptance Questionnaire (CPAQ)
catastrophizing 81, 97, 117, 118, 122, 251–67, 274, 497 549–50
brain mechanisms 99 Chronic Pain Coping Inventory (CPCI) 296, 459
Communal Coping Model 252, 256 chronic pain syndromes 489
construct of 251–3 see also chronic low back pain; CLBP; fibromyalgia;
and pain experience 253–8 temporomandibular joint disorder
attention 254–5 Chronic Pain Values Inventory (CPVI) 553
coping 256–7 chronic regional pain syndrome 105, 108
emotion 255–6 chronic widespread pain 10
endogenous pain modulation 257–8 cingulate cortex 109
expectancies 253–4 classical conditioning 117
and pain outcome 258–9 Classification of Chronic Pain 21
treatment implications 259–62 CLBP 3–4, 269–70
see also fear-avoidance causality 81–2
celecoxib 475, 476, 477 as chronic disease 504
Center of Excellence for Orthopaedic Pain deconditioning syndrome 157, 185–200
Management Austria 420, 422 epidemiology 4–5
central augmentation 486–7 European guidelines 426
central disinhibition 61–2 explanatory models 270
central nervous system fear-avoidance 269–94
altered processing 60–2 health-related fitness 187–93
baroreceptor sensitivity 139–40 muscle atrophy 188–9
structural brain changes 105–14 muscle composition 189
structural brain imaging 94–5, 106–8 natural history 207–8
CGRP see calcitonin-gene related peptide neuropathic components 59–60
chronic back pain 208 outcome definition 208–10
brain processes in 91–4 PAL in 186–7
cognitive factors 97–9 progression to 77, 78–81
learning and memory 95–7 self-rating of risk 516–19
lower back see CLBP stress system functioning in 77–8
psychophysiology 115–52 supraspinal mechanisms 62–3
structural brain changes 94–5, 105–14 functional plasticity 62–3
imaging 107–8 structural plasticity 63
chronic fatigue syndrome 48, 194 treatment see treatment
chronic low back pain see CLBP Clinic on Low-Back Pain in Interdisciplinary Practice
chronic pain 21–40 (Clip) Guidelines 420, 422
acceptance of 549–50 clinical decision-making 213–14
approach to 29–33 clinical screening methods 212–14
generalizability of 31–3 clinicians see physicians
prognostic risk score 30–1 CNR1/2 polymorphism 45
back see chronic back pain; CLBP Cochrane Back Review Group 419
clinical implications 33–5 cognition 122–3, 315
cognitive bias 316–19 cognitive avoidance 137
attention bias 316–17 cognitive defusion 551–2
interpretation bias 317 cognitive factors 97–9
memory bias 317–19 cognitive fusion 551
cognitive models 319–26 cognitive models 319–26
enmeshment 320–6 pain processing 319–20
pain processing 319–20 self-pain enmeshment 320–6
cognitive-behavioural therapy 547–63 cognitive processing 315–37
cognitive defusion 551–2 biases in chronic pain 316–19
history 547–8 attention bias 316–17
mindfulness 553–5 interpretation bias 317
treatment outcome 558–9 memory bias 317–19
values clarification and values-based action 552–3 clinical practice implications 326–9
definitions future directions 329–31
conceptual bases 24–5 cognitive responses 302–3
duration-based 21–7, 33 cognitive-behavioural pain models 341–3
outcome probability 29, 33, 34 cognitive-behavioural therapy 8, 118, 327, 435, 436
as multidimensional phenomenon 27 chronic pain 547–63
INDEX 583
cognitive defusion 551–2 deep brain stimulation 115

history 547–8 default mode network 94
mindfulness 553–5 degenerative disorders 116
treatment outcome 558–9 dehydroepiandrosterone-sulphate (DHEA-s) 10
values clarification and values-based action 552–3 depression 275–6, 497
contextual 547–63 diathesis–stress models 118–19, 231
definition 548–9 diathesis–stress–environment model 488–9
with exercise 537–8 diazepam 478
patient selection 514–15 diclofenac 475, 476
risk factor-based 500, 513–28 overdose 476
psychosocial risk assessment 517–25 diffuse noxious inhibitory control (DNIC) 25
self-rating of chronicity risk 516–17 diffusion tensor imaging 107
stages of 437 dipyrone 476
time point for treatment 513–14 disability
treatment selection 515 factors affecting 396
cognitive-evaluative pain component 89 pain-related 306–7
collagen 46 patients’ perception of 395–6
gene polymorphisms 46 workplace management 380
Communal Coping Model 252, 256, 257, 343–4 disability syndrome 356
see also coping strategies distraction 99, 296
community-based interventions 7–8, 398 Distress Risk Assessment Method (DRAM) 220–1
comorbidity 48, 120, 124, 126, 128, 130, 134, 141, 207 distress-endurance response (DER) 224, 297, 299,
compensation neurosis 363–4 301, 304, 305, 521, 523
COMT polymorphism 44 distrust, injured workers’ sense of 366
concept of general activation 119 disuse syndrome 272, 276
concept of response specificity 119 and fear-avoidance 298
conditioned stimuli 271 doctor shopping 392
confrontation 136, 272–3, 342 dopamine 10
congenital insensitivity to pain with anhidrosis 58 dorsal root ganglia 55
convergent validity 23–4 dorsolateral prefrontal cortex 108
coping strategies 81, 256–7, 296 dorsomedial nucleus 75
adaptive 256 Drug Committee of the German Medical Society
Communal Coping Model 252, 256 420, 422
endurance coping 79, 523 dual-diagnostic approach 502
spouse involvement 346–7 duration-based approach to chronic pain
Coping Strategies Questionnaire (CSQ) 296, 303 21–7, 33
cortical plasticity 109 Dutch College of General Practitioners
cortical reorganization 105 419, 422
corticotrophin releasing hormone 71 Dutch Physiotherapy Guidelines 420, 422
cortisol 71 Dutch Quality Institute for Healthcare 420, 422
psychosocial factors affecting 80 dynamic model of affect (DMA) 297
COST B13 Working Group on Guidelines for Chronic dynorphin A 10
Low Back Pain in Primary Care 420, 422, 423–4 dyslipidaemia 193
COST B13 Working Group on Guidelines for the
Management of Acute Low Back Pain in Primary early identification 22–3
Care 420, 422, 423–4 educational interventions see patient education
Council on Technology Assessment in Health efficacy expectancies 253–4
Care 420, 422 electroencephalography (EEG) 62, 89, 96
counselling 521 electromyography (EMG) 155–84
couple’s therapy 347–8 drawbacks 161
COX inhibitors 472, 474, 476–7 loading for 161
pharmacokinetics 477 mean power frequency 159–60
see also individual drugs median frequency 159–60
cyclo-oxygenase inhibitors see COX inhibitors reliability 162
cytokines 10, 58–9 surface 190
and HPG function 76 EMG see electromyography
emotional distress 255–6, 495–502
Danish Institute for Health Technology empathy models of pain 344–6
Assessment 420, 422 employer fraud 367
decision to consult 441 endocrine risk factors 10–11
deconditioning syndrome 157, 185–200 endogenous pain modulation 257–8
muscle changes 188 endorphins 177
see also health-related fitness endurance coping 79, 523
584 INDEX
endurance-related pain responses 188–9, 295–314 of pain see pain-related fear/anxiety

adaptive 297, 300–1, 303 of pain-causing activities 279
affective 303, 308–9 Fear Avoidance Beliefs Questionnaire for Health Care
pain-related fear/anxiety 309 Practitioners (FABQ) 407
positive mood 308–9 Fear of Pain Questionnaire-III (FPQ-III) 234, 278
avoidance-endurance model (AEM) 224, 295–314 Fear of Pain Questionnaire-Short Form
behavioural responses 302 (FPQ-SF) 234, 278
clinical practice consequences 310 Fear Survey Schedule-II (FSS-II) 238
cognitive responses 302–3 fear-anxiety-avoidance models 232, 233
distress-endurance response (DER) 224, 297, 299, Fear-Avoidance Beliefs Questionnaire (FABQ) 279, 298
301, 304, 305 fear-avoidance learning 271–2
eustress-endurance pattern (EER) 297, 300, 303–4, fear-avoidance physical therapy 438
500, 523 fear-avoidance response (FAR) 79, 97, 122, 173, 224,
fear-avoidance model (FAM) 79, 97, 122, 173, 224, 269–94, 295, 297, 342, 498, 515, 523
269–94, 295, 297 attention 99, 254–5, 275
fear-avoidance response (FAR) 79, 97, 122, 173, catastrophizing see catastrophizing
224, 269–94, 295, 297, 342, 498, 515, 523 depression 275–6
feelings of self-control/self-efficacy 308 disuse syndrome 272, 276
and neurocognitive functions 307–8 and disuse syndrome 298
occurrence of 205–7 future directions 286–7
outcomes 304–7 hypervigilance 240–1, 275
pain intensity 304–6 models of 272–3
pain-related disability 306–7 see also pain-related fear/anxiety
overt physical activity 306–7 female sex hormones in CLBP 78
patient subgrouping on 498–502 fentanyl 478
and physical overuse/overload 299–300 fibromyalgia 10, 48, 91, 194, 485, 489
stability vs. variability 309–10 Finnish Medical Association 420, 422
environmental factors 487–8 flags framework 207
epidemiology 3–20 black flags 206, 381, 447
goals of 3–5 blue flags 207, 380, 381, 447
success of 5 orange flags 207
etoricoxib 475, 476, 477 red flags 27, 207
European guidelines 423–4 yellow flags 27, 203–29, 381, 447
acute low back pain 425 in risk identification 205–6, 521
CLBP 426 screening for 220–5
eustress-endurance pattern (EER) 297, 300, 303, flexion-relaxation 189–90
500, 523 Florida Agency for Health Care Administration &
exercise interventions 176–8, 531–46 Department of Health 420, 422
cognitive-behavioural perspective 537–8 flupirtine 473, 475, 478
increased physical fitness/activity 532–4 fMRI 9, 62, 89, 91–3, 106
McKenzie therapy 536–7 follicle stimulating hormone 75
spinal stabilization 534–6 fractional anisotropy 107
work endurance 539–40 functional contextualism 548
expectancies for treatment response 253–4 functional magnetic resonance imaging see fMRI
exposure in vivo 282–6 functional plasticity 62–3
assessment 283 functional somatic syndromes 6
behavioural experiments 285
efficacy of 285–6 G protein-coupled receptors 57
goal setting 283–4 gabapentin 474
graded exposure (GivE) 234, 284–5 GCH1 polymorphism 43–4
hierarchies 284 generalized anxiety disorder (GAD) 237
patient education 284 genetic factors 43–54
extended pain 319–20 intervertebral disc disease 45–7
extensor muscles 155 loss of pain perception 48–9
musculoskeletal pain syndromes 48
FAAH polymorphism 44 pain perception 485–6
false negatives 217, 218 pain sensitivity 43–5
false positives 215, 217, 218 pro-inflammatory genes 47–8
family loss 366–7 risk of back pain 11–13
fatty acid amide hydrolase inhibitors 45 ghrelin 76
fear 118, 136–7, 270–1 glial cells 61
of movement/re-injury 279 glucocorticoids 475
of negative evaluation 231, 239–40 goal setting 283–4, 521–2
INDEX 585
gonadotropin inhibitory hormone 75 Illness Attitudes Scale (IAS) 239

gonadotropin-releasing hormone (GnRH) 74 illness behaviour 6
graded balancing 522–3 see also pain behaviours
Graded Chronic Pain Scale 501 illness/injury sensitivity 231, 238–9
graded in vivo exposure (GivE) 234, 284–5 imaging studies 89–104
see also exposure in vivo biochemical changes 94–5
grey matter changes 110 brain processes 91–4
growth hormone 76 structural brain imaging 94–5, 106–8
growth hormone releasing hormone 76 therapeutic interventions 99–100
guidelines 397, 419–29 see also individual imaging modalities
diagnostics and treatment 423 immediate pain 319–20
Europe 423–4, 425, 426 immunological risk factors 10–11
implementation 424–6 incapacitation, need to prove 369–70
quality of 420–3 individual-specific response 119, 120, 133–7
as vehicle of education 446 injured workers 355
compensation systems 355–75
health anxiety 6–7 malingering 360–1
health care providers see physicians perceptions of 365–9
Health Care Providers Pain and Impairment anger with compensation scheme 367–9
Relationship Scale (HC-PAIRS) 407 conflict with compensation system 366
health-related fitness 185 sense of being distrusted 366
aerobic capacity 190–3 stress and personal/family loss 366–7
body composition and bone strength 187–8 recovery curve 356–7
in CLBP 187–93 secondary gain 361–5
components of 185 vs. other patient groups 358–9
future research 194 Institute for Clinical Systems Improvement 420, 422
metabolic factors 193 insula 89, 90, 109
muscle strength and endurance 188–9 insulin 193
postural control 189–90 insulin resistance 193
Healthcare Providers’ Pain and Impairment insulin-like growth factor-1 76
Relationship Scale 394 insurance fraud 367
heart rate 190 integrative behavioural couple therapy 348
helplessness 97, 117 International Association for the Study of Pain
hereditary sensory and autonomic neuropathy 49 (IASP) 21
holistic treatment 409 international guidelines see guidelines
Hospital Anxiety and Depression Scale 32 International Paris Task Force, France/Canada 420, 422
HPA see hypothalamic–pituitary–adrenal axis interoceptors 286
HPG see hypothalamic–pituitary–gonadal axis interpretation bias 317
HPGH see hypothalamic–pituitary–growth hormone axis intervertebral disc disease 45–7
hyperalgesia 24, 58–9, 233 interviews 280
hypervigilance 240–1, 275 motivational 464
hypochondriasis 23 Intolerance of Ambiguity Scale 237
hypocortisolaemia 77 intolerance of uncertainty 232, 237–8
hypothalamic–pituitary–adrenal (HPA) axis 10, 27, irritable bowel syndrome 108, 110
71–4 isolectin B4 56
acute stress 71–3 Israeli Low Back Pain Guideline Group 419, 422
chronic stress 73–4
in CLBP 77 ketoprofen 477
normal function 71 Kiel Pain Inventory 79, 296, 302, 303, 523
hypothalamic–pituitary–gonadal (HPG) axis 69, 74–6 kinesiophobia 136, 239, 271, 279, 282, 407, 499
acute stress 75 see also fear-anxiety-avoidance model
chronic stress 75–6
normal function 74–5 lack of access to care 369
regulation by body weight 76 Lasègue sign 493
hypothalamic–pituitary–growth hormone (HPGH) latent transition regression analysis 29
axis 69, 76–7 lead time of screening 215–17
acute stress 76–7 learned pain response 95–7, 140
chronic stress 76–7 ligamentous injuries 116
normal function 76 lornoxicam 477
hypothalamus 71 low back pain 4, 269–70
chronic see CLBP
ibuprofen 476, 477 definitions 5, 6, 26
IL1RN polymorphism 47 European guidelines 425
586 INDEX
low back pain (cont.) perceived harmfulness 279–80

patient education 434–5 MPI see West Haven-Yale Multidimensional Pain
prevalence 5 Inventory
primary care management 7–8 MPRCQ 456, 458–9
progression to CLBP 77, 78–81 initial development 459
secondary prevention 503 rationale 458–9
work-related risk factors 377–88 MPRCQ2 459–61
risk factors brief versions 461
hypothetical 9–13 MRI 107
peripheral/central sensitization 55–67 MRS 95
population-based research 5–7 MSPQ 220
treatment see pain management Multiaxial Assessment of Pain (Turk and Rudy’s
lumiracoxib 476, 477 Dysfunctional Chronic Pain taxonomy) 28
luteinizing hormone 75 Multidimensional Fatigue Inventory 79
Multidimensional Pain Inventory (MPI) 492–5
McKenzie therapy 536–7 Multidimensional Pain Readiness to Change
magnetic resonance imaging see MRI Questionnaire see MPRCQ
magnetic resonance spectroscopy see MRS multidimensionality of pain 22
magnetoencephalography (MEG) 62, 89 muscle atrophy 188–9
maladaptive coping strategies see catastrophizing muscle composition 189
male sex hormones, protective role 78 muscle fibre types 170–1
malingering 360–1 muscle re-conditioning 176
marital partners see significant others muscle relaxants 204, 475, 478
maximum voluntary contraction (MVC) 157, 176 muscle strength 188–9
MC1R polymorphism 44 muscular dysfunction 116, 141
mechanically insensitive afferents 56 muscular fatigue 155–84
media campaigns 436 and back pain 155–7
medial basal hypothalamus 74 as consequence of back pain 173–6
median eminence 75 deconditioning syndrome 157
mediators 205 effects of exercise training 176–8
meloxicam 476, 477 measurement 157–61
membrane channels 57 physical/anatomical factors 162, 170–1
memory bias 317–19 as predictor of back pain 171–3
metabolic equivalents 191 reliability of testing 162, 163–9
metalloproteinases 47 muscular strength 155–6
metamizol 477 musculoskeletal pain syndromes 48
methocarbamol 478
3-methoxy-4-hydroxyphenylglycol (MHPG) 10 naproxen 475, 476–7
Million Visual Analog Scale 79 National Advisory Committee on Health and
Mindful Attention Awareness Scale (MAAS) 554 Disability; New Zealand 419, 422
mindfulness 234, 548, 553–5, 558 National Health and Medical Research Council
Minnesota Multiphasic Personality Inventory (MMPI) Australia 420, 422
491, 518 negative predictive value 215
moderators of treatment 205 nerve growth factor 10, 56
Modified Somatic Perception Questionnaire 172 neurocognitive functions 307–8
Modified Zung Depression Inventory 220 neurodegeneration 108
mood-as-input model 194 neurological risk factors 9–10
moral hazards 357–8, 370–1 neuropathic pain 10
morphine 478 glial cells in 61
motivation 456–63 neuropeptides 10
future research 461–3 neurophysiological changes 23
MPRCQ 458–9 neurostimulation 115
MPRCQ2 459–61 neurotrophin receptors 57–8
brief versions 461 neurotrophins 58
Pain Stages of Change Questionnaire nociception 55–60
(PSOCQ) 457–8 genetic transmission of 11–12
motivation-specific response pattern 119 perception of 111
motivational interviewing 464 transduction/transformation in nociceptor
motivational issues 453–69 terminals 57–8
motivational model of pain self-management 454–6 sensitization of nociceptor terminals 58–9
motivational-affective pain component 89 nociceptors 55–6
movement polymodal 56
fear of 136, 239, 271, 279 sleeping 56
INDEX 587
non-steroidal anti-inflammatory drugs see NSAIDs exercise 531–46

non-verbal pain behaviours 79, 340 exposure in vivo 282–6
noradrenaline 10 implications of catastrophizing on 259–62
NSAIDs 204, 473, 475, 490 low back pain 204
nucleus raphne magnus 138 motivational issues 453–69
nucleus tractus solitarius 138 pain-related fear/anxiety 234–5, 282–6
patient education see patient education
observational measures of pain-related fear 281 pharmacotherapy 471–82
obsessive–compulsive disorder (OCD) 237 rehabilitation 565–79
Obstacles to Return-to-work Questionnaire settings/contexts 567–8
(ORQ) 380 pain matrix 63, 89–91, 105
occupational outcomes 206 pain memory 95–7
operant conditioning 118, 140 pain perception 90, 99
operant model of pain 339–41 pain personality 121–2
opioids 473, 475, 478 pain processing 319–20
OPRM1 polymorphism 44–5 pain sensitivity 43–5
orange flags 207 and cardiovascular response 137–40
Örebro Musculoskeletal Pain Screening Questionnaire loss of 48–9
(ÖMPSQ) 222–3, 380 Pain Stages of Change Questionnaire (PSOCQ) 457–8
orphenadrine 478 pain-causing activities, fear of 279
outcome probability 29 Pain-Coping Inventory (PCI) 303
overgeneralization 504 Pain-Disability Prevention (PDP) Programme 569
oxycodon 475, 478 pain-related disability 306
oxymorphone 475, 478 pain-related fear/anxiety 232–4, 271, 274
and anxiety sensitivity 236–7
PABS-PT 407, 409 assessment 234, 277–82
pain 485 interview 280
chronic see chronic pain observational measures 281
classification of 491–2 psychological measures 281
diathesis–stress–environment concept 488–9 self-assessment 280–1
immediate 319–20 self-report measures 277–80
individual differences 485–8 endurance effects 309
environmental contributions 487–8 patient subgrouping on 495–8
genotype 485–6 treatment 234–5, 282–6
prior learning histories and psychological assessment 283
variations 486–7 behavioural experiments 285
multidimensionality of 22 exposure in vivo 282–6
nature of 443–4 goal setting 283–4
patient subgroups 491 graded exposure 284–5
endurance response/emotional distress 498–502 hierarchies 284
see Multidimensional Pain Inventory (MPI) patient education 284
pain-related fear/emotional distress 495–8 vulnerability to 276–7
patient uniformity myth 489–91 PAL see physical activity in daily life
reflex-spasm model 116 paracetamol 204, 476
Pain Anxiety Symptoms Scale (PASS) 234, 278, 499 paraventricular nucleus 71
Pain Anxiety Symptoms Scale-20 (PASS-20) 234, 278 Paris Task Force 423
pain avoidance behaviour 136–7 PARIS-CBA 519–25
pain behaviours 118, 123–4, 491, 515, 556 empirical evidence 523–5
and avoidance learning 272 principles of 519–20
non-verbal 79, 340 procedure and steps of intervention 520–3
punishing responses 343 Participatory Ergonomics (PE) approach 380
pain catastrophizing see catastrophizing pathological pain 58–9
Pain Catastrophizing Scale (PCS) 407 patients 22
pain facilitation, descending pathways 138 beliefs/expectations 441
pain generator 490 characteristics 441–2
Pain and Impairment Relationship Scale (PAIRS) classification of pain 491–2
278, 407 perceptions of disability 395–6
pain intensity 304–6 patient education 284, 433–52
pain management 204 content 443–5
cognitive-behavioural therapy see cognitive- context 445
behavioural therapy effectiveness 434, 439–40
customization of 502–3 evaluation of 436–7
dropouts 494–5 flags framework 447
588 INDEX
patient education (cont.) Pictorial Fear of Activity Scale-Cervical

general principles 434 (PFActS-C) 280
implementation 446 piroxicam 476, 477
low back pain 434–5 polymyalgia rheumatica 10
psychologically informed re-activation 437–9 population studies 5–7, 13–14
public health initiatives and media campaigns 436 risk identification 208–9
spouse involvement 346–7 positive mood 308–9
strategy 445–7 Positive Mood Scale (PMS) 303
sustainability 439–40 positive predictive value 215
types of intervention 435–6 positron emission tomography see PET
patient subgroups 491 postherpetic neuralgia 106
endurance response/emotional distress 498–502 postural adjustments to loading 156
see Multidimensional Pain Inventory (MPI) postural control 125–32, 189–90
pain-related fear/emotional distress 495–8 predictive factors 204–5
patient uniformity 489–91 prefrontal cortex 89
patient-mediated strategies 446–7 pregabalin 474
periaqueductal grey 61, 91, 138 prevention 209
peripheral/central sensitization 55–67 prior learning history 486–7
altered CNS processing 60–2 pro-inflammatory genes 47–8
nociception 55–60 processes of change 456
supraspinal mechanisms 62–3 prognostic approach to chronic pain 29, 33, 34–5
personal loss 366–7 prognostic factors 205
personality 121–2 prognostic risk score 30–1
PET 89, 93–4 generalizability of 31–3
phantom limb pain 105, 108 prognostic studies 214
pharmacotherapy 471–82 Progressive Goal Attainment Programme (PGAP) 569
antidepressants 473, 478–9 proopiomelanocortin (POMC) 71
combined 475 propranolol 572
COX inhibitors 472, 474, 476–7 proprioceptors 286
muscle relaxants 204, 475, 478 propyphenazone 476, 477
opioids 473, 475, 478 psoriatic arthritis 48
strategies for 472–5 psychological factors 486–7
see also individual drugs psychological measures of pain-related fear 274
Philadelphia Panel 420, 422, 423 psychological pain models 117–19
Photograph Series of Daily Activities (PHODA) 279 classical conditioning 117
physical activity in daily life (PAL) 185, 186, 306 cognitive-behavioural approach 118
future research 193–4 diathesis–stress model 118–19
patients with CLBP 186–7 operant conditioning 118
physical deconditioning see deconditioning syndrome sensitization 117
physical fitness 185 psychophysiological models 119–20
increasing 532–4 psychophysiological studies 115–52
physical overuse 299–300 biomechanical models 116–17
physical work demands 378–9 psychological pain models 117–19
physician as disability advisor 391–404 stress-related factors 120–37
educational interventions 396–7 psychophysiological variables 125–37
guidelines 397 individual-specific response 133–7
legislative changes 398–9 static and dynamic postures 125, 131–2
patients’ perceptions of disability 395–6 symptom-specific response 132–3
recommendations psychosocial adaptation 123
content of 392 psychosocial interventions 515–16
physicians’ characteristics influencing 393–4 see also cognitive-behavioural therapy
scope of role 392 psychosocial risk assessment 517–19
systems approach to improvement 397 PARIS-CBA 519–25
physician education 396–7 psychosocial risk factors 203–29, 514
physician-patient relationship 394–5 Psychosocial Risk for Occupational Disability
physicians’ attitudes/beliefs 405–15, 442–3 Instrument (PRODI) 380
changing of 410–11 psychosocial stressors 7, 118, 121
current knowledge 411–13 public health initiatives 436
differences between professional groups 411 punishing responses to pain behaviours 343
effect on clinical decisions 408–10, 442–3
effect on treatment orientations 442 quality-adjusted life years (QALYs) 211–12
measurement of 407–8 Quebec Task Force on Spinal Disorders 419,
and risk of long-term problems 406–7 422, 423
INDEX 589
range of motion 190 Distress Risk Assessment Method (DRAM) 220–1

readiness to change 457 Örebro Musculoskeletal Pain Screening
recovery, effect of workers’ compensation schemes Questionnaire (ÖMPSQ) 222–3
on 356–9 Risk Screening of Back Pain (RISC-BP) 223–5
recovery curve 356–7 STarT Back Decision Tool (SBDT) 221–2
red flags 27 secondary gain 361–5
see also flags framework features of 363
reflective listening 464 terms related to 364
reflex-spasm model 116 types of 362
rehabilitation 565–79 secondary loss 365
combination with other treatment modalities 571–2 secondary prevention 503, 514
conceptual framework 566–7 self-as-content 557
decision rules 568–71 self-as-knower 556
future research 574–5 self-assessment of pain-related fear 280–1
integration with social environment 573–4 self-control 308
timing of 568–71 self-efficacy 308, 444–5, 455
treatment settings/contexts 567–8 self-management
relaxation 554, 572 motivation 456–63
respiratory exchange ratio 191 motivational interviewing 464
response expectancies 253–4 motivational model 454–6
response stereotypy 120 self-pain enmeshment 320–6
Ret 56 clinical practice implications 326–9
rheumatoid arthritis 10, 48 pain and self 321–2
RIsc SCreening for Back Pain (RISC-BP) 519 Schema Enmeshment Model of Pain 322–6
risk factors 204 self-rating of chronicity risk 516–19
endocrine and immunological 10–11 self-reported fear of pain 277–80
fear-avoidance 79, 97, 122, 173, 224, 269–94 self-system therapy 328
genetic 11–13 sensitivity of screening tests 215
hypothetical 9–13 sensitization to pain 117
neurological 9–10 sensory-discriminative pain component 89
peripheral/central sensitization 55–67 Sentence Completion Test for Chronic Pain 324
population-based research 5–7 serotonin metabolites 10
psychosocial see yellow flags sick certification 406, 409
work-related 377–88 sick-listing 394, 399
risk identification significant others 339–54
clinical epidemiology 209–10 conceptualization
individual concepts 210 cognitive-behavioural models 341–3
nature and focus 206–7 Communal Coping Model 252, 256, 257, 343–4
population studies 208–9 empathy models 344–6
yellow flags in 205–6 operant model 339–41
Risk Screening for Back Pain (RISC-BP) 223–5, 500 pain-related interaction process 349
rofecoxib 477 recommendations 348–50
Roland–Morris Disability Questionnaire 173, 571 treatment research
Roman chair test 161 couple therapy 347–8
rostral ventromedial medulla 61, 138 spouse involvement in patient education 346–7
Royal College of General Practitioners 419, 422 sleeping nociceptors 56
Royal Dutch Association for Physiotherapy; Manual social climate at work 380
Therapy Guidelines 420, 422 Social Cognitive Theory 446
somatization 23
Schedule for Nonadaptive and Adaptive Personality somatosensory cortex 89, 105
(SNAP) 518 somatosensory mechanisms 23, 116–17
Schema Enmeshment Model of Pain 322–6 somatostatin 76
SCL-90-R depression score 29, 30 specificity of screening tests 215
screening 210–25 spinal cord dorsal horn, synaptic plasticity 60
accuracy 215 spinal stabilization exercises 534–6
actuarial vs. clinical 212–14 spouse-assisted coping skills training 346–7
combination of methods 218–19 stages of change 456
feasibility 219 STarT Back Decision Tool (SBDT) 221–2
lead time 215–17 static and dynamic postures 125–32
optimal cut-off points 217–18 stimulus-specific response pattern 119
principles 210, 211 stress
purposes of 210–12 and CLBP (chronic low back pain) 77–8
yellow flags 220–5 HPA axis in 71–4
590 INDEX
stress (cont.) values clarification 552–3

HPG axis in 75–6 values-based action 552–3
HPGH axis in 76–7 Veterans Health Administration 420, 422
injured workers 366–7 voltage gated sodium channels 58
stress-related factors 120–37 voxel-based morphometry 94–5, 107
chronification of pain 120–4
cognitions 122–3 Waddell Physical Activity scale 298
comorbidity 124 Waddell signs 98
pain behaviours 118, 123–4 well behaviours, positive reinforcement 343
personality 121–2 West Haven-Yale Multidimensional Pain Inventory
psychosocial adaptation 123 (MPI) 492–5
psychosocial stressors 7, 121 profile distribution 492
stroke volume 190 white matter connectivity 110
Stroop task, modified 281 WHO
structural abnormalities 116 International Classification of Functioning,
structural brain changes see brain, structural changes Disability, and Health 566
structural plasticity 63 screening criteria 211
Structured Clinical Interview 79 work avoidance 399
substance P 10, 56 work capacity 392
surface EMG 190 Work Disability Diagnosis Interview (WoDDI) 380
Survey of Pain Attitudes 459 work endurance exercises 539–40
Swedish Norrtälje Musculoskeletal Intervention Centre work-related risk factors 377–88
study 80 future research and practice improvement 382–4
Swiss Medical Society 420, 422 intervention strategies 381–2
Symptom Checklist-90R 491 modifiable 379
symptom specificity 120 perceptions on health and work 380
symptom-specific response 132–3 physical work demands 378–9
synaptic plasticity 60 screening for 380–1
systemic lupus erythematosus 48 social climate at work 380
workplace disability management 380
Tampa Scale for Kinesiophobia (TSK) 136, 239, 279, workers’ compensation systems 355–75
407, 497, 499 anger with 367–9
task persistence behaviour 304 conflict with 366
temporomandibular joint disorder 44, 81, 485 effect on injury recovery 356–9
tetrahydrobiopterin 44 implications for social policy 371–2
tetrahydrocannabinol 45 and injured worker behaviour 359
tetrazepam 478 moral hazards 357–8, 370–1
thalamus 89 negative effects 360–70
The Back Book 435, 436–7, 438, 439–40 attorney involvement 369
theory of emotion 119 lack of access to care 369
Thought Suppression Scale (TSS) 224, 302 malingering and secondary gain 360–5
thrombospondin-2 47 need to prove incapacitation 369–70
tissue Doppler imaging 125, 131 perceptions of injured workers 365–9
tizanidine 478 union involvement 369
TRAAK channels 57 Working Backs Scotland 398
tramadol 475, 478 workplace disability management 380
Transactional Model of Health 341–2 World Health Organization see WHO
transient receptor potential channels 57
Transtheoretical Model of Behaviour Change yellow flags 27, 203–29, 381, 447
(TTM) 456 in risk identification 205–6, 521
treatment effect modifiers 205 screening for 220–5
treatment strategies see pain management Distress Risk Assessment Method (DRAM) 220–1
TREK channels 57 Örebro Musculoskeletal Pain Screening
trkA receptors 56 Questionnaire (ÖMPSQ) 222–3
TRPA1 polymorphism 44 Risk Screening of Back Pain (RISC-BP) 223–5
tumour necrosis factor alpha 48, 59, 77 STarT Back Decision Tool (SBDT) 221–2
see also flags framework
UK BEAM Trial 438
unconditioned stimuli 271 zygapophysial joints 471–2
unions, involvement in workers’ compensation
cases 369

From Acute To Chronic Back Pain

Uploaded by

Copyright:

Available Formats

From Acute To Chronic Back Pain

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

From Acute To Chronic Back Pain

Uploaded by

Copyright:

Available Formats

What is the book about?

What is the book about?

Who edited the book?

Who edited the book?

From Acute to Chronic

Part 1 Current Developments in Epidemiology

Part 2 Risk Factors of Chronic Back Pain and Disability:

Part 3 Risk Factors of Chronic Back Pain and Disability:

Part 4 Risk Factors of Chronic Back Pain and Disability:

Part 5 Practitioner’s Role in the Process of Care

Part 6 Clinical Implications: New Approaches to

Part 7 Clinical Approaches for Patients with Acute

Part 8 Subgroup-Specific Approaches for Patients at

Part 9 Clinical Approaches for Patients with Established

A. Vania Apkarian Ida Flink

Kate M. Dunn Mark P. Jensen

Sandra Kamping Gary J. Macfarlane

Michael K. Nicholas Adina C. Rusu

David O’Riordan Rita Santos

Tamar Pincus William S. Shaw

Pradeep Suri Jeanine Verbunt

Dennis C. Turk Michael Von Korff

MMPI Minnesota Multiphasic Personality PPI proton-pump inhibitors

Epidemiology of Back Pain, from

Box 1.1 Beneficial outcomes of epidemiological study of

1.1.2Recent success of epidemiology in improving understanding

1.1.3 Defining LBP in epidemiology studies

1.2 Risk factors for LBP already established in population-based

Box 1.2 Proposed standardized definitions of low

1.3 Evidence on management for LBP in primary care

1.4 Potential biological risk factors hypothesized from

1.4.2 Endocrine and immunological

Box 1.3 Genes implicated in human pain conditions*

1.5 The need and potential benefit of population-based studies

1.5.1 Beyond the bedside

Defining Chronic Pain by Prognosis

2.1 Duration-based definitions of chronic pain

Box 2.1 Examples of typical back pain patients

2.2 Conceptual bases for defining chronic pain2

2.3 Duration-based and prognostic approaches

Box 2.2 Definitions of low back pain episodes

Episode of care for low back pain

Episode of work absence due to low back pain

2.4 Chronic pain as a multidimensional phenomenon

2.5 Chronic pain—static or dynamic?

2.6 Defining chronic pain by outcome probability

2.7 Development of the approach

Table 2.1 Scoring rules for estimating prognostic risk score

Item Item value Risk score value

Item Item value Risk score value

2.8 Generalizability of the prognostic approach to other samples

Baseline risk score group

2.9 Comparison of prognostic and duration approaches

2.10 Clinical implications

Risk Factors of Chronic

Genetic Factors Modulating Chronic

3.2 Polymorphisms associated with subtle modulations of

3.3 Genetic polymorphisms associated with intervertebral

3.4 Polymorphisms in pro-inflammatory genes