Appearances - of - Power - Part 1

Appearances of power: Service user involvement in health
research
Elizabeth Mary Smith
Thesis submitted for the degree of

Doctor of Philosophy
University of London
1
Declaration
I declare that the work presented in this thesis is my own.
Signed ……………………………………… Date ………………….
2
Abstract
Service user (patients, carers and the public) involvement is an important issue for
health and social care services, professional education and research in the United
Kingdom and other Westernized nations. New policy requirements create challenges
and opportunities for the development of professional practice in all of these contexts.
In relation to health research, previous theoretical and empirical work on service user
involvement is scant and current conceptualizations do little to explain the changing
relationships, successes, problems, failures, experiences and values that shape thinking
and practice within this field.
This study aims to contribute theoretically informed understandings about the

significance and relationship between particular types of issues and perspectives in
order to suggest how future work in this area might be targeted. The method draws on
interpretive, discursive and reflective perspectives to analyse documentary sources and
empirical data (face-to-face interviews, conferences, workshops, discussions and
training events) from health researchers and members of consumer health organisations
in the UK and Australia. An overall conclusion is that understandings about user
involvement in health research have been influenced by established notions of
knowledge, needs and rights, representation and responsibility; encouraging particular
forms of service user/researcher relationships to develop. Although tensions about
professional power over user involvement in health research can not be eradicated they
can be better understood and negated. Where there is uncertainty, there is space for
manoeuvre, for example though language development, emergent epistemological
positions and the diversity offered by subjective experiences and views. Ultimately,
user involvement in health research has the potential to empower or disempower
according to each person's views about their own ability, potential and sense of being.
3
Contents
Acknowledgements..................................................................................................................................9
A note to service users and carers........................................................................................................11
Chapter 1: Introduction.............................................................................................................12
1.1 Context...........................................................................................................................................12
1.2 Problem statement.........................................................................................................................14
1.3 Aims and objectives.......................................................................................................................15
1.4 Overview of the method................................................................................................................15
1.5 What this study contributes..........................................................................................................20
Chapter 2: Background..............................................................................................................21
2.1 Introduction...................................................................................................................................21
2.2 Socio-political trends towards user involvement........................................................................21
2.3 Public involvement within the knowledge economy...................................................................37
2.4 Background to the study...............................................................................................................40
2.5 Finding focus..................................................................................................................................43
Summary................................................................................................................................................48
Chapter 3: Current knowledge about user involvement in health research.........................49

3.1 Introduction...................................................................................................................................49
3.2 Scope and analysis.........................................................................................................................49
3.3 The quantity of research activity.................................................................................................53
3.4 Themes of the research literature: definition, role, hierarchy, ideology..................................55
3.5 Methods studies have used............................................................................................................74
3.6 Gaps in the research evidence......................................................................................................78
3.7 Personal reflections on the literature...........................................................................................79
Summary................................................................................................................................................93
Chapter 4: A method for researching user involvement.........................................................95

4.1 Introduction...................................................................................................................................95
4.2 Approach........................................................................................................................................96
4.3 Reflexive methodology................................................................................................................106
4.4 Principles for a reflexive methodology......................................................................................113
4.5 Study data....................................................................................................................................118
4.6 Analysis.........................................................................................................................................135
Summary..............................................................................................................................................147
Chapter 5: How knowledge of user involvement has developed..........................................149

5.1 Introduction.................................................................................................................................149
5.2 What knowledge is......................................................................................................................149
5.3 A stratigraphy of knowledge......................................................................................................156
4
5.4 Concept themes............................................................................................................................159
5.5 Strategies and modes of enunciation..........................................................................................176
5.6 Hidden carer- and user-led activity...........................................................................................180
Summary..............................................................................................................................................185
Chapter 6: Tensions..................................................................................................................187
6.1 Introduction.................................................................................................................................187
6.2 More professional commitment ↔ Less professional control.................................................188
6.3 Increased transparency ↔ Uncertain intentions......................................................................196
6.4 ‘Equal say’ ↔ ‘Know-how’........................................................................................................201
6.5 The interplay of power................................................................................................................206
6.6 Individual’s perspectives of power.............................................................................................209
Summary..............................................................................................................................................214
Chapter 7: Ways forward........................................................................................................216

7.1 Introduction.................................................................................................................................216
7.2 Framework for an evidence base...............................................................................................216
7.3 Reasons for resistance.................................................................................................................223
7.4 Spaces for manoeuvre.................................................................................................................229
7.5 Experiences of power..................................................................................................................235
7.6 Judgement making......................................................................................................................244
7.7 Thinking styles.............................................................................................................................247
Summary..............................................................................................................................................252
Chapter 8: Discussion...............................................................................................................254
8.1 Introduction.................................................................................................................................254
8.2 Crux of the interpretation...........................................................................................................254
8.3 Summary of contribution............................................................................................................255
8.4 Value and validity........................................................................................................................256
8.5 Discussion of the approach.........................................................................................................261
8.6 Returning to the principles of a reflexive methodology...........................................................267
Summary..............................................................................................................................................278
Chapter 9: Conclusion..............................................................................................................280
9.1 Introduction.................................................................................................................................280
9.2 Summary of the main substantive findings...............................................................................280
9.3 Wider implications of the study.................................................................................................284
9.4 A model for future research.......................................................................................................288
9.5 Summary conclusions..................................................................................................................292
Bibliography..............................................................................................................................298
Appendices.................................................................................................................................340
Appendix 1: Terms and definitions....................................................................................................341
Appendix 2: Literature Review..........................................................................................................343
5
Appendix 3: Study Data Log..............................................................................................................378
Appendix 4: An introduction to Michel Foucault’s life and works................................................394
Appendix 5: Extract from the Bristol Inquiry Report Final Summary Report (BRI 2001)........399
Appendix 6: Search terms for the multi-method review.................................................................400
Appendix 7: Making reflexivity visible.............................................................................................403
Appendix 8: Researcher recruitment letter......................................................................................404
Appendix 9: Interview schedule.........................................................................................................406
Appendix 10: Summary notes of MPhil/PhD upgrade examination..............................................408
Appendix 11: Published papers..........................................................................................................409
6
Figures
Figure 1: Structure of the thesis.......................................................................................16

Figure 2: Historical context.............................................................................................23
Figure 3: International Association for Public Participation...........................................34
Figure 4: Study time line.................................................................................................41
Box 1: Objectives of the study at proposal stage (January 2004)...................................44
Box 2: Objectives at MPhil/PhD transfer (September 2005)..........................................45
Box 3: Objectives (October 2006)...................................................................................46
Box 4: Objectives (November 2007)...............................................................................47
Box 5: Search terms.........................................................................................................51
Figure 5: Previous research studies.................................................................................54
Figure 6: Research literatures relating to user involvement in health research...............56
Figure 7: Classification of consumers in health research (Boote et al 2002)..................57
Box 6: Classification of ‘users' (Barnes and Wistow 1994)...........................................59
Figure 8: Involving consumers in research processes (Hanley et al. 2000)....................62
Figure 9: Typology of user involvement in research (Smith et al. 2005b)......................63
Figure 10: A Ladder of Citizen Participation (Arnstein 1969)........................................65
Figure 11: Levels of control in user involvement (after Hanley et al. 2000)..................66
Figure 12: Hierarchical relationship between concepts of participation (Cahill 1996)...67
Figure 13: Participation continuum (Hickey and Kipping 1999)....................................68
Figure 14: Framework service user input and empowerment (Holosko et al. 2001)......68
Figure 15: Action research cycle.....................................................................................77
Figure 16: The approach to knowledge construction.......................................................97
Figure 17: Interaction between different levels of interpretation..................................116
Figure 18: Elements of reflective methodology............................................................117
Figure 19: Example of field notes.................................................................................129
Figure 20: Reflective themes.........................................................................................131
Figure 21: Systems of formation (discursive formations).............................................143
Figure 22: Power theory map........................................................................................146
Figure 23: Issues framework for user involvement in research (Smith et al. 2005b)....155
Figure 24: A stratigraphy of user involvement in health research.................................158
Figure 25: Constellation of user involvement in health research..................................160
Figure 26: System of formation of user involvement in health research.......................179
Figure 27: Lukes' three-dimensional framework of power (Clegg 1989 p.90).............199
Figure 28: Circuits of power (Clegg 1989, p.214)........................................................208
Figure 29: Power in user involvement in health research..............................................213
Figure 30: Framework for an evidence base..................................................................219
Figure 31: Outcome issues associated with user involvement in health research.........222
Figure 32: Opportunities for knowledge development..................................................234
Figure 33: Ability to deliberate......................................................................................236
Figure 34: Orientations of intent....................................................................................247
Figure 35: Time perspectives.........................................................................................265
Figure 36: Framing knowledges....................................................................................291
7
Tables
Table 1: Principles of consumer involvement (Telford et al. 2004)................................91

Table 2: Sources of empirical data used in the study....................................................123
Table 3: Overview of empirical data.............................................................................124
Table 4: Interview participants in Australia...................................................................128
Table 5: Summary of analyses.......................................................................................135
Table 6: Components of the analysis.............................................................................137
Table 7: Concepts of representation...............................................................................174
8
Acknowledgements
This study received funding from the Engineering and Physical Sciences Research
Council Doctoral Training Award programme. The National Nursing Research Unit and
King's College London provided support for registration fees in the first two years.
Travel and research costs for work in Australia were supported by a Research Travel
Grant from King's College London & Monash University.
Professor Fiona Ross, Dean of the Faculty of Health and Social Care Sciences,
Kingston University and St George's, University of London and by Dr Patricia Grocott,
Reader at the Florence Nightingale School of Nursing and Midwifery, King's College
London supervised the study. Professor Margaret O'Connor at the School of Nursing
and Midwifery, Monash University was Associate Supervisor. Professor Charlotte
Humphrey and Dr Angus Forbes, both at King's College London, kindly undertook the
MPhil/PhD transfer viva.
I would like to acknowledge the aforementioned for their support and advice. I would
also like to acknowledge the colleagues who I worked with on the National Institute for
Health Research Service Delivery and Organisation PIN project (Patient and public
involvement in nursing, midwifery and health visiting research): Professor Fiona Ross,
Professor Peter Beresford, Professor Jill Manthorpe, John Sitzia, Sheila Donovan and
Sally Brearley; and members of service user organisations involved in the study.
Dr Janette Bennett, National Nursing Research Unit, provided insight and support on
the use of Foucauldian theory. Dr Sara Shaw, University College London, provided
advice about policy discourse analysis. Professor Sally Redfern, National Nursing
Research Unit provided advice about research ethics. Thanks to Dr Natasha Campling
for discussions about user involvement in medical device development and for advice
about preparing for the MPhil to PhD transfer viva, to Angela Furne for suggestions
about finding a track and staying on it and to Theresa Nash for showing me a practical
way to build the thesis. Thanks to Professor Celia Davies for reading and reviewing an
9
article informed by this study and to Professor Peter Beresford for providing comments
on a final draft.
Thanks to present and former colleagues at the National Nursing Research Unit
(particularly Professor Peter Griffiths, Gian Brown, Dr Janette Bennett, Barbara Davey,
Sally Brearley, Professor Sally Redfern, Trevor Murrells and Dr Sarah Robinson);
colleagues at INVOLVE (particularly Helen Hayes), and to fellow postgraduate
students in the Florence Nightingale School of Nursing and Midwifery, King's College
London and at staff in the School of Nursing at Monash University Melbourne
(particularly Sharni, Sue and Louise).
Thanks to my friends: Janette and Michael Collins (especially for introducing me to the
wells and wheelbarrows of Killyleagh), Gian (especially for her constant
thoughtfulness), Nick, Molly & Luke Brown and Cheryl Kipping, Barbara Davey (in
particular for taking me to Carlucci's when I needed it), Sally Brearley (especially for
her kind words from Coventry to London), Ursula Hawgood (for her humour and
support); Sheila Donovan, Dr Catherine Evans and Louise Jolly (for tête à tête at the
Tate); my university friends: Emma, Rhian, Ceri, Liza, Bex, Janet and Jo; and to my
good friends Romy, Lizzy and Lise.
Thanks to my fiancé Bob and to my brothers David and Stephen. Special thanks to
Mam and Dad (Mrs Jacqueline Smith SRN SRM and Mr Stuart Brian Smith O.B.E MSc
F.M.A.).
10
A note to service users and carers
The study presented here explored the meaning of service user involvement in health
research. I approached the topic as being about relationships within social contexts.
Patients, carers and members of the public are part of those relationships. I hope that
anyone with an interest in this topic will find this thesis useful as it draws together a
wide range of sources and perspectives to discuss the issues.
As the study was undertaken as part of a doctoral training programme this thesis is
presented in a certain way. I hope to make the findings more accessible to a wider
audience by publishing in different formats, teaching and talking about the research at
conferences. It is a lengthy document and service users may wish to begin by reading
the introduction and summaries at the end of each chapter.
Overall, I have found that the development of relationships between service users and
researchers has been limited by classifying such relationships in certain ways and
working within constraints of existing research processes and structures. The
significance of external influencing factors has sometimes been overlooked, such as the
positioning power of established notions of needs, rights, knowledge, representation and
responsibility. I summarize these and other conclusions in the final section of the thesis
(Section 9.5).
To better understand how service users and researchers can work together we need to
focus more on examining the reasons behind resistance and discomfort. I have not
produced recommendations specifically for service users or for researchers. Instead,
anyone with an interest in this topic may wish to read about the methodological
challenges I encountered (Chapter 4), my views on what an evidence base should look
like (Chapter 7), and how it might be achieved (Chapter 8).
I would like to hear your views on any of the issues I have been interested in here.
11
Chapter 1: Introduction
1.1 Context
The issue of public involvement is important to many Westernized nations and in

relation to a wide range of public sector services such as housing, health, policing,
education and social care. Political moves to encourage public involvement create
challenges and opportunities for the development of professional practice. The present
study is concerned with service user (patients, public and carers) involvement in United
Kingdom health research. In this context government suggests that involving service
users directly and actively in all aspects of decision making can have benefits for the
National Health Service (NHS) (DH 2005, DH 2002a), for patient care (DH 2001a,
2001b, 2002a) and for service users themselves (NHSE 1998).
Research policy recommends researchers “involve patients and the public wherever
possible as active participants in the research process” (DH 2001c) and to develop
closer relationships with patients and the public as a matter of priority (DH 1990d, DH
2003a, DH 2006a). These trends in policymaking are reinforced by research
commissioning bodies, such as the National Institute for Health Research (NIHR 2008)
and the Medical Research Council (MRC 2008), which increasingly request to see plans
for service user involvement as part of proposals for funding.
Health researchers and service users have offered mixed support for these new
requirements. Some have reported opportunities and difficulties (Cayton 2002). Others
have raised questions about the potential impact of user involvement in health research
(Entwistle et al. 1998; Faulkner and Thomas 2002; Minogue et al. 2005, Fudge et al.
2007). Previous authors have stressed that translating the rhetoric of service user
involvement into reality is challenging (Reed et al. 2004), that it is still very much in the
preliminary stages (Baxter et al. 2001), and that it will be a costly process demanding a
strong willingness to re-examine attitudes around sharing of control (Couldrick 2000).
Debates about user involvement in health research are influenced by related agendas
across the health and social care sector. These include awarding greater say to the
12
‘consumers' of health services, establishing mechanisms for Patient and Public

Involvement (PPI) in health service development, and the drive to involve service users
in the recruitment, training and education of healthcare professionals. Some of the
broad social and political trends that underpin such moves are driven by concerns about
the safety and quality of health research. At the same time there can be considerable
differences between the language and concepts that are employed in relation to issues of
involvement in these different contexts. In health policy and research literatures the
term ‘service user' is generally used to mean patients, carers and members of the public,
however, alternative terms such as ‘patient' ‘public' ‘consumer' ‘citizen' ‘lay
representative' and ‘user representative' are also used. Here I will use the term service
user to reflect current terminology in the research literature, but when discussing policy
documents or the work of previous authors I will maintain the terms they use to show
different underpinning concepts and changes in language over time. A summary of
terms is provided in appendix 1.
A particular point of confusion in relation to user involvement in health research is the

distinction between activities associated with the techniques of doing user involvement
(for example planning, preparing, undertaking research with service users) and
researching user involvement. These types of activities are not mutually exclusive but
neither are they the same thing. To help make this distinction here, the practice of user
involvement in health research is referred to as ‘service user/researcher relationships',
while the field of study is described as ‘user involvement in health research'.
Previous authors have developed definitions and classifications for different types of
user involvement in health research. Although user involvement is often perceived as
being part of a research process it is often inconsistent with scientific values, such as
statistical representation, controlled intervention and randomization. Despite user
involvement in research being increasingly venerated in research policy and through
commissioners' requirements researchers have expressed concerns that service users
might affect the rigour of research and the quality of evidence produced. The conceptual
complexity of decisions about which service users should be involved can mean that
user involvement appears unrepresentative and biased.
13
One argument sometimes put forward is that user involvement in health research is
moral or ethically right because the overall aim should be to provide services that
service users need and want. Indeed, it has been suggested that service users involved
in research may experience therapeutic outcomes, access to health information, social
networking and opportunities (Boote et al. 2002; Baxter et al. 2001). In particular there
are concerns that research traditions and cultures stand in the way of ‘democratic
empowerment' and that ‘top-down' political agendas encourage tokenistic approaches
to public involvement (Beresford 2003). Values associated with involvement such as
transparency, equal say and active participation can be in tension with the values of a
competitive knowledge economy. Yet, in practice service user/researcher relationships
do not always accord with values of enhanced well-being and there is no guarantee that
individuals will not suffer harm.
The interrelated and sensitive nature of the issues sets a challenging context for
developing knowledge about user involvement in health research.
1.2 Problem statement
In the United Kingdom and across other Westernized nations research policy and
commissioning processes increasingly emphasise that service users should be actively
involved as partners in the research process. New requirements for service user
involvement raise many challenges for establishing the best ways of involving service
users in different contexts of health research. Current conceptualizations of user
involvement do not adequately reflect the changing relationships, intermittent successes,
emergent problems, unaccounted for failures, life changing experiences and moral
values that shape thinking and practice within this field. There is a need to develop
understandings about service user involvement in the full range of health research
contexts. There is also a need to establish the best ways of developing understandings
about user involvement in health research. Establishing the significance and relationship
between particular types of issues and perspectives is important for assessing the
14
implications of change in health research policy and practice and for targeting future
work in this area.
1.3 Aims and objectives
The overall aim of the thesis is to inform thinking within the field of health research
about user involvement and to suggest where, how and why future knowledge might be
developed. The objectives of the study have changed over time and this is explained
more in chapter 2.
The objectives were:

1. To review what is currently known about user involvement in health research.
2. To identify ways in which knowledge has developed and factors that have shaped
present understandings.
3. To examine points of tension and uncertainty about user involvement in health
research.
4. To identify how future knowledge may be developed.
1.4 Overview of the method
Research about user involvement in research could involve the study of issues about
service user/researcher interactions, or indeed situations where such relationships do not
occur. Different ontological, epistemological and methodological perspectives could be
drawn upon for the purpose of such a study. Hence, the act of researching user
involvement in health research is itself a social practice with its own developmental
rules and concepts.
The following figure depicts the four major components of the study– the topic of ‘user
involvement in health research, the approach, the method and the theoretical
perspectives which inform the analysis.
15
Figure 1: Structure of the thesis
Theory
Knowledge
Power
User involvement in
health research
Subjectivity
Current knowledge
Knowledge development
Tensions
Ways forward
Approach
Interpretation
Method
Discourse Reflectivity
Reflexive
methodology
In this study building knowledge of user involvement in health research began by

looking at what the issues in the field were and trying to make sense of them. My
interest has continued in this vein. Although my aim has not been to undertake a study
that is classifiable as user involvement in health research I did aim to contribute to
current knowledge about user involvement in health research. Nor has my aim been to
judge the evidence for or against user involvement in health research, although I am
interested in how judgements about user involvement are made. Instead I perceive user
involvement in health research as a shifting formation of ideas and practices that are
open to interpretation at the individual level. An interpretive component of the approach
16
enabled me to explore the possibility of different views about what knowledge is and
how it can be developed.
A discursive component of the approach helped me to consider health research as

comprising real and theoretical social interactions that span political, academic, health
service, voluntary and organisational settings as well as concerning individual people
engaged (or not) in these social systems and in making sense of their own experiences
and views. From this stance ‘health research' was approached in a broad sense as
including research on the topic of health and illness, research about health services
including the NHS, professions engaged in health research (medicine, nursing, allied
health professionals and so on), different disciplines within health research (clinical,
management, design, economics, education), and different processes and practices
associated with undertaking research about health or health services. The term ‘health
researchers' is used to mean professional researchers from a range of backgrounds, not
exclusively researchers who are healthcare professionals. While being a researcher
undertaking the research, it was not only research actions which lead to research
insights. In accordance with a reflective component of the approach I became an active
interpreter of discourse.
Reflective methodology provided a way to explore user involvement in health research

and make links with existing theory literatures about knowledge, power and
subjectivity. The methods developed in this study incorporated different forms of
reflexivity and applied these to the empirical material, the construction of data, the
interpretation and the study context. I developed the method by drawing on the work of
previous authors (discussed in chapter 4), in particular Mats Alvesson and Kaj
Sköldberg's (2000) Reflective Interpretation. The incorporation of reflective notes such
as the one here allowed me to present a close representation of the issues I have
encountered and the insights I have gained.
{Trigger (App3/Date:13.03.07): Supervision meeting, discussion of use of

reflexivity}
This study has been an opportunity and a challenge. It has led me to
experiences, activities and arenas that I might not otherwise have
17
considered or taken up. I have felt the excitement of the outset, new sights,
sounds and expectations; the thrill of discovery and new understanding;
and the relentless stages where passion turned to monotony. These feelings
were instrumental to my thinking at the time and to the development of the
thesis. Where they are relevant they are included. I aim to show that
including personal reflections encourages rigour and greater awareness of
value judgements.
A further part of the method has been to identify triggers (events and ideas) that have
informed the interpretations I have made. The header of reflective notes indicates the
trigger for reflection which can be cross-referenced to the study data log (Appendix 3).
These tables detail all of the unpublished documents and materials (conference
proceedings, meeting minutes, interview transcripts and so on) that I have collected and
used. The bibliography at the end of the thesis cites all of the policy documents, journal
articles and papers, documents published by research and related organisations and
literature published by consumer health organisations that I have made use of.
Recording and reflecting on triggers is particularly helpful for representing interaction

with other people over the course of the study, for example, authors from the literature,
participants in the study, my supervisors, and so on. The empirical data include the
views of researchers working in the academic sector, members of consumer and
voluntary health organisations, staff employed within the National Health Service and
an international perspective from academic and consumer research organisations in
Australia. These contributions are made evident though citation and quotation but they
can not be delimited either in relation to the fabric of the study or to my thinking.
Finding focus along side piecing together information has been a constant challenge.
The use of reflective methodology helps to reveal how the nature of my interactions
with different participants and sources of data changed as the research progressed.
Given that the study approach and methods were developmental it was necessary to
gain ethical approval in stages. Approval for data collection as part of the multi-method
review (explained in section 2.4) was granted by King's College London Research
Ethics Committee in March 2003. Approval for research undertaken in Australia was
18
granted by King's College London Research Ethics Committee in September 2005 and
was supported by Monash University, Melbourne. Plans for a case study were submitted
to King's College London Research Ethics Committee in July 2006. This application
was unsuccessful because of concerns about participant confidentiality, discussed in
section 2.5.
The study aimed to explore historical, conceptual and experiential perspectives of user
involvement in health research. I have drawn heavily, though not exclusively, upon
Michel Foucault's work to consider what user involvement in health research is and
how knowledge of it can be developed. This is not to say that other theory bases, for
example economics, psychology or physical sciences can not be usefully applied to this
topic. However, I was concerned with questions that were practical ‘how to do'
questions and I found that they were often about who had control. Bringing new
understanding to these questions has required an exploration of social theory in relation
to research practice, and vice versa.
From a Foucauldian perspective user involvement in research creates rules and

opportunities which can divide those who construct knowledge from those who
knowledge is about. The study is partly about the power of user involvement in health
research. It asks, what is happening within health research and development at a
national and local level - how is the agenda being used and why? The study is also
concerned with new notions of expertise, for example ‘expert patients' and ‘experienced
user-researchers'. User involvement in health research appears to reinforce existing
values of researcher status, such as objectivity and credibility. It can also create new
status frameworks that concern both researchers and service users. The analysis
(summarized in section 4.6) asked questions about such issues can be better understood.
In particular, how power is perceived and how it can be experienced. Overall, the study
connects service user involvement with existing theories of power to examine tensions
about professional control. ‘Appearances of power' in the title of this study could
therefore mean both, how power is manifest through user involvement in health
research and how power is experienced from the perspective of researchers and service
users.
19
1.5 What this study contributes
The contribution this study makes is described in detail in the discussion chapter
(Chapter 8). The crux of the interpretation is presented in Section 8.2. In summary, the
work contributes to the field of user involvement in health research in four main ways.
The study:
 Summarizes and critiques current knowledge of user involvement in health research

to show that relatively little research on user involvement in health research exists
and that current conceptualizations tend to underplay the significance of external
influencing factors.
 Uses theoretically informed and empirically based analyses to identify ways in

which understandings of user involvement in health research have and are being
developed. It shows that understandings about user involvement in health research
have been influenced by established notions of what constitutes research,
encouraging particular forms of service user/researcher relationships to develop.
 Employs social theories of power to identify and examine points of tension about
user involvement in health research. It shows that although concerns about
professional commitment and control over user involvement in health research can
not be escaped they can be better understood.
 Takes a critical perspective to examine the basis for developing an evidence base. It
suggests that knowledge and the agenda for user involvement in health research
could be strengthened by exploring four areas in particular: reasons for resistance,
experiences of power, judgement making and thinking styles.
The main substantive findings and implications of the study are discussed in the
concluding chapter (Chapter 9), where I develop a model with which judgements about
user involvement in health research can be perceived and explored.
20
Chapter 2: Background
2.1 Introduction
This chapter continues to set the scene by describing the historical social and political
conditions that have contributed to the development of user involvement in health
research. The discussion explores several distinctive trends that occurred in the decades
before user involvement became incorporated into research policy and research
commissioner's requirements. Using a Foucauldian inspired historical approach to the
issues, understandings of user involvement in health research are revealed to be
phenomena constructed and understood by the social formations surrounding it.
According to this perspective, user involvement in health research was not created as a
discrete idea. It began as a collection of propositions, beliefs and ideas that were borne
out of observation and experience. The focus here is to look at the influence of five
broad socio-political trends, including recognition of public interests in public services,
efficiency and responsiveness to users, rationing and priority setting, quality and safety,
and pluralism. Key events in the history of formation are identified to show that user
involvement in health research is not entirely driven by research evidence,
policymakers or social groups. Later, in chapter 5 I will build on this analysis by
looking at how understandings of user involvement have been interdependently
constructed by all of these forces. The closing section of the chapter sets the scene for
the study by explaining the background and steps towards finding focus.
2.2 Socio-political trends towards user involvement
Trends towards user involvement are not confined to health research but neither did
they occur in parallel or develop in the same way in other areas of social practice. As
Baxter argues:
“The development of lay involvement in research and development in other

fields started at different times and for different reasons. In community
development, both in the UK and overseas, it began in the early post-war
decades because of the realization that poverty was not being eradicated
by top-down policy directives. In social care, strong voices from the ‘user
21
movement’ influenced lay involvement particularly during the 1980’s and

1990’s”. (Baxter et al. 2001, p.viii)
It is important to consider what makes it possible for us to talk and think about user
involvement at this particular historical moment. This could help to explain how it is
possible for people “to have contrary opinions, and to make contradictory choices”
(Foucault 2006a, p.221) about user involvement in health research, including the
theoretical consistency between understandings and what it is that makes it possible for
them to co-exist. In this section I argue that social and political concerns about
recognition of public interests in public services, efficiency and responsiveness to users,
rationing and priority setting, quality and safety, and pluralism have influenced moves
towards user involvement in health research. To support the discussion, these trends and
corresponding key events are illustrated by the following figure (Figure 2).
Recognition of public interests in public services
Long before there was talk of user involvement in health research there were concerns
about whether public services work to public interests. In the 1960s stories of suffering,
exclusion and helplessness that some members of society endured were unacceptable to
modern British citizens (Higgins 1993). Although responsibility for the provision of
health care had increasingly been overtaken by the state, concerns about whether this
was in the best interests of all members of the public were being raised (Ham 2004).
Groups of patients, particularly those with experience of institutions for people with
disabilities or mental health problems began to lobby for a say in issues that affected
them. The sharing and connection of personal stories about the unacceptability of
institutions and professional practices gained support in the context of a wider public
disquiet with political and professional control. In the 1960s groups of the public took
direct action on political issues: epitomized by the student sit-ins in the United States.
Consumer advocates took up the notion of the ‘healthcare consumer' (Tomes 2006).
The Research Institute for Consumer Affairs (1963), part of the Consumers'
Association, drew attention to the silent voice of the public in general practice.
22
Figure 2: Historical context
Recognition of public interests in public

services
Arnstein (1969) publishes ‘A ladder of Legal requirement for public participation embodied in
citizen Participation in USA 1960's UK land use planning legislation
Movements for social welfare and inclusion including;

disabled people, people with learning difficulties, older
people, mental health service users/survivors, people with
HIV/AIDS, looked after young people. 1974 Establishment of Community Health
(Baxter et al 2001) 1970's Councils to represent the public interest
The Griffiths report encouraged market research and

patient satisfaction surveys (NHS Management
Inquiry 1983)
Working for Patients (DH 1989)
1980's
Focus on efficiency and responsiveness to users
NHS and Community Care Act requires local authorities to consult over community
care plans. Internal market and choice introduced (DH 1990)
NHS Management Executive (1992) publish Local Voices

encouraging health authorities to engage with the public
1992 Patient's Charter introduced to NHS (DH 1995)
Increased rationing and priority setting
Barnes & Wistow (1992) publish ‘Understanding
Establishment of Consumers in NHS Research (1996) user involvement'
Changes in professional education (DH 1994)
Cahill (1996) ‘Patient participation a concept analysis'
Blaxter (1995) presents discussion paper on consumer
NHS Executive (1998a) Patient Partnership strategy 1990's issues
Chalmers (1995) calls for patient priorities to be heard
Standing Advisory Group conference (1998a)
Attention to quality and safety
DH publishes ‘Partnerships in Action' Lord Hunt speaks at major conference on user involvement (DH 2000c) DH
(DH 1998a) and ‘A strategy for PPI in publishes ‘NHS Plan' (DH 2000a)
the NHS' (DH 1998b) DH publishes ‘Research and Development for a First Class
Evaluation of the National User service' (DH 2000b)
Involvement project (NHSE 1999)
DH publishes ‘Research Governance Framework for
Health & Social Care' (DH 2001c)
Publication of ‘Small voices big noises' (Baxter et al. 2001)
Review of Health and Social Care Act (DH
Study timeframe 2001d) (duty to involve becomes law in 2003)
DH publishes ‘Shifting the balance of
INVOLVE (2004b) Conference power' (DH 2002c)
Beresford (2005) discussion paper on
empowerment INVOLVE (2002b) Conference
2000's
Smith et al. (2005b) review of user Review by Crawford et al. on user involvement in health
involvement in nursing research (2002) and by Boote et al. on ‘Consumer
Turner & Beresford (2005) review of ‘User involvement in health research (2002)
controlled research' Beresford (2002) discussion of ideologies
Plurality and diversity Strategic review of NHS R&D NHS highlights patient involvement
INVOLVE (2006a) Conference (2003)
Launch of InvoNET (2006)
Launch of NIHR Research for Patient benefit
programme (DH 2006a)
‘Reward and Recognition' (DH 2006b)
Launch of National Centre for Involvement
(2006)
23
Also in the 1960s, a growing number of consumer lobby groups gave individuals a
position from which to campaign for particular health and service issues (Wood 2000).
‘Public participation' appears as a legal requirement in UK land use planning legislation
in the 1960s. Community Health Councils were established by Act of Parliament
considerably later in 1974 to monitor and review the National Health Service and to
recommend improvements. The Association of Community Health Councils for
England and Wales (ACHCEW) was set up in 1977 to provide advice and guidance to
CHCs and to represent CHCs at national level. Community Health Councils (CHCs) in
England and the Association of Community Health Councils for England and Wales
(ACHCEW) were abolished on 1 December 2003. In England they have been replaced
by a number of bodies but mainly Patient Advice and Liaison Services (PALS). Twenty
CHCs continue to exist in Wales.
These social and political trends towards recognition of public interests influenced and
supported more community focused and politically aware approaches to health research.
By the end of the 1970s a body of literature on community-focused and socially
orientated research methods (Bryman 2001), including Action Research (Reason and
Bradbury 2001) and co-operative enquiry (Heron 1996), was available internationally.
In the UK the women's lobby in maternity services and health consumer groups in
general fed into new social movements which influenced research agendas and
professional thinking about the power within research processes (Jones et al. 2004). The
disability movement used research findings to draw attention to every day issues of
discrimination and stigma in the health domain (Kelleher 2001), and these ideas were
progressed through emancipatory disability research (Mercer 2002). Voluntary
organisations for people with disabilities continue to be a strong influence in reframing
the way disability is perceived and have had a strong influence on the policy narrative.
This has had major implications for research because:
“The social model of disability reverses this causal chain to explore how
socially constructed barriers … have disabled people with a perceived
impairment”. (Barnes and Mercer 1997, p.1)
24
Efficiency and responsiveness to users
In the late 1970s cost containment at the macro level across the NHS involved setting
prospective global budgets for hospitals, controls over hospital building and the
acquisition of medical equipment, limits on doctor's fees and incomes, and restrictions
on the numbers undertaking education and training. In the late 1980s/early 1990s micro
efficiency and responsiveness to users introduced market like mechanisms in the health
service and management reforms as well as budgetary incentives (Ham 1997). The
Thatcher government were concerned about the urgent need for public interests to be
incorporated in the NHS (DH 1989). Public unrest about political and professional
decision making had escalated into a ‘crisis of confidence' in trust, transparency and
legitimacy and there was a need to show the public that their concerns were being heard
(Haywood and Hunter 1982).
Political interests on consumerism focus on the possibilities of introducing new types of

internal market forces to the NHS. Consumerism offers one way of reframing public
and professional relationships, by applying principles of access, choice, information,
redress and representation to public services (Potter 1988). Segal's (1998) analysis of
health systems in Australia, across Europe and North America suggests that both
demand side and supply side reform are essential for managing ever increasing health
service demands. Patient satisfaction is important to policymakers, healthcare
managers, practitioners and service users, despite problems with establishing a tangible
definition of satisfaction and its measurement (Crow et al. 2002) (discussed in Smith
and Ross 2006, Appendix 11). Concerns to improve the responsiveness of professional
practice to service users can also be detected in changes to professional education. The
literature on social work has always afforded users a prominent position, but in the
1980's users were becoming involved in education and selection of candidates to the
profession, influenced by the development of the Children Act the National Health
Service and Community Care Act (Parsloe and Swift 1997). These moves may derive
from the philosophical tenants of the profession to ensure individuals rights through
participation. Strategies to involve users in teaching, assessment and curriculum
planning are now prominent in medical and health care professional education (DH
25
1989, 1994). In relation to health research, public concerns about efficiency and
responsiveness of professional working feed into longer standing concerns about the
relevance of science to society. For example:
“Incipient and actual attacks on the integrity of science have led scientists to
recognize their dependence on particular types of social structure. Manifestos
and pronouncements by associations of scientists are devoted to the relations
of science and society. An institution under attack must re-examine its
foundations, restate its objectives, seek out its rationale” (Merton 1942, p.
267)
One sociological view of science suggests:
“Science stands in the region where the intellectual, the psychological and the
sociological interact. It is knowledge, therefore intellectual, conceptual and
abstract. It is inevitably created by individual men and women, and therefore
has a strong psychological aspect. It is public, and therefore moulded and
determined by the social relations between individuals”. (Ziman 1968, p.15)
When perceived as social phenomena the entity and institutional structure of research
can be deconstructed and examined. Foucault's work on science and knowledge
provide important contributions to these debates. In the Order of Things, Foucault
(2006b) unearthed codes that determined what could or could not be defined as
scientific at different points in history. He looked at what constituted science in the
Renaissance, the classical era and the modern era and showed that each historical
period has its own episteme which defines the conceptual limits of experience,
knowledge and truth. According to this view science can be thought of as a set of
theories or narratives that are constantly evolving with society.
“Science is a deceptively inclusive word which refers to a variety of distinct

though interrelated items. It is commonly used to denote (1) a set of
characteristic methods by means of which knowledge is certified; (2) a stock of
accumulated knowledge stemming from the application of these methods; (3) a
set of cultural values and mores governing the activities termed scientific; or
(4) any combination of the foregoing” (Merton, 1942 p.268).
Perceiving health research as being socially connected supports the view that research
should be relevant and responsive to the users of such knowledge. Yet, as I will go on to
discuss in chapter 3 who ‘users' are is a point of debate, and there can be difficulties of
reconciling competing demands for relevance (Rappert 1999).
26
The 1991 Research for Health Strategy included a commitment to involving consumers
at all stages of the research process (DH 1991c). However, there was no clear statement
about how this would be put into place. The following year Baroness Jay, a member of
the Central Research and Development Committee, was asked to chair an ad-hoc group
to address ‘consumer issues' within the NHS Research and Development Strategy. The
group met only once and agreed to commission three reports; consumer issues outside
of the NHS (unpublished), involving consumers in research (unpublished), and a
discussion paper summarizing the literature and methods of research on consumer
issues within the NHS (Blaxter 1995).
A further indication of a growing interest in users occurred in 1991. The regional

director of NHS Research and Development at Trent Regional Office was tasked with
the responsibility of managing a national programme of research on consumer issues for
the NHS Research and Development Programme (DH 1991a). Work was initiated to
ascertain what was happening in practice across the regions but this was halted because
of a national workshop to address issues raised by Professor Blaxter's report. It was not
until nearly two years later that a report of the workshop was published together with
the Blaxter report in August 1995. Blaxter's (1995) review of consumer issues in the
NHS urged funders and sponsors of research to support public participation in research.
It highlighted a perceived need amongst policy makers, researchers and members of the
public for research questions to relate to issues that are important to patients and the
public interest. The reason for the delay in publication is unclear, but there is an
apparent gap in activity at a national level between 1991 and 1994 possibly due to
changes taking place in the NHS as a whole. It has also been suggested that a lack of
organisational infrastructure and leadership stifled development (Standing Advisory
Group 1998a).
This is not to say that initiatives were not going on between voluntary health
organisations and research institutions. For example, “The People Factor” research
project was initiated in 1993 between West Lancashire Association for Disabled People,
a voluntary charitable organisation, run by a management committee consisting mainly
of disabled people and Edge Hill University College (Standing Advisory Group 1998a).
27
In October 1994 the Central Research and Development Committee agreed to set up a
group to monitor and advise on consumer involvement in the NHS Research and
Development Programme. It was also agreed that a Department of Health R&D
Standing Advisory Group on Consumer Involvement would be launched in 1996. These
developments indicate that there was genuine commitment in some quarters to
involving users in research, as well as mainstream interest in the rights of patients in
research (Goodare and Smith 1995).
Rationing and priority setting
In the mid 1990's policymakers across Westernized nations were attempting to shift the
focus of governance towards new forms of co-production with other agencies and with
citizens themselves through partnerships and community involvement (Newman et al.
2004). In 1997 the incoming Labour government's agenda to reform and modernize the
provision of centrally funded services, including the health service (DH 1997) and
social services (DH 1998d), reinforced the decentralisation of decision making
responsibility to local governance structures. Greater spending and commissioning
authority was awarded to local health bodies and public and private partnerships were
encouraged (DETR 1998, DH 2001e).
In the health service the launch of the NHS Plan (DH 2000a) aimed to achieve a
healthier nation (DH 1998e) by creating a service focused on the needs of patients.
Involvement of patients and the public is described as being central to achieving this
vision (DH 1999b). These plans took forward priorities identified by the Conservative
government to increase effectiveness of the NHS and clinical practice (DH 1996a,
1996b). Policies to encourage individual citizens to become more involved in decisions
and choices about their own health (Needham 2003) have influenced thinking about
participation in other contexts including management, education and research.
In health research the Central Research & Development Committee set up a unique
group to advise them on how best to involve consumers in the R&D process (Oliver
1996). At its first conference in 1998 entitled “Research: What’s in it for me?” four
28
hundred consumer advocates, researchers, clinicians, information specialists and service

managers were in attendance, indicating the strength of interest in this agenda. A key
note speech from Professor John Swales, Director of Research and Development at the
Department of Health at that time drew attention to the need to prioritize research that is
relevant to patient needs and values:
“Every decision which affects patients is influenced by evidence which reflects

research. That evidence may often be flawed, incomplete or misleading, but no
decision can be made in isolation. And there is more to evidence than
scientific observation and experiment. Evidence has to be relevant, reflect the
needs and values of the individuals at risk through illness. If it does not, it is
just as flawed as if it were based on invalid science or defective
statistics.”(Standing Advisory Group 1998b, p.9)
The Standing Advisory Group's first annual report “Research what’s in it for
consumers?” offered recommendations to the Central Research and Development
Committee including involvement of user perspectives in the development of the
revised NHS Research and Development Strategy and new Standing Groups for Health
Technology Assessment Service Delivery and Organisation and New and Emerging
Technologies (Standing Advisory Group 1998a, p.12). The report also emphasized
barriers to user involvement, including concerns about dissonance between
professionals and consumer priorities for research. Elsewhere, it has been argued that
attempts to gain public agreement and involvement regarding the principles
underpinning decisions about resource allocation may actually replicate underlying
power relationships because of the way they are set up and communicated (Higgs and
Jones 2001).
Despite these trends some researchers were arguing that user involvement should be a
higher priority in the design, undertaking and assessment of health research, as
illustrated by the following quotation.
“Wherever possible, consumer peer review should be sought. Where there has
been no consumer input into the original design (and obviously it will take
some time for this recommendation to be implemented), journals should seek
consumers' commentaries on published papers. These are tough requirements,
but they are likely to ensure greater rigour and relevance for future research”
(Goodare and Lockwood 1999, p.725)
29
In this climate of patient focused policy making, pockets of work were going on
through the Department of Health regional offices to involve consumers in activities to
prioritise and commission research and in peer review (Buckland and Gorin 1999).
Elsewhere, staff at the Centre of Health Economics, University of York produced a
database report on consumer involvement in research (Dixon et al. 1999); the Toronto
Group (Research as Empowerment) was set up in 1997 to enhance the role of service
users in research and to promote research as an empowering process (Hanley 2005).
The National User Involvement Project involved people with learning disabilities and
mental health service users in identifying ways of involving a wider range of service
users in commissioning decisions (Joseph Rowntree Foundation 1999). The Standing
Advisory Group's second annual report, Involvement Works (Standing Advisory Group
1999) focused on promoting such initiatives. It identified three studies in particular:
involvement of consumers in identifying research priorities at Mount Vernon Hospital
in Middlesex, a pilot project to involve consumers in the Health Technologies
Programme and the Shaping Our Lives Project.
“The Shaping Our Lives project is a consumer-led project and involves

consumers fully at all levels. It has found that services need to develop an
approach to provision based on user-defined objectives and outcomes. It now
supports the users of community care services in defining and working towards
the outcomes they want from these services”. (Standing Advisory Group 1999,
p.6)
Organisational priorities for the forthcoming year included classifying the different
types of consumer involvement in health research and promoting awareness of this
classification system. The Standing Advisory Group's long term goal had become more
ambitious. It was now:
“To ensure that consumers are involved as a matter of course in research &
development in the NHS at national, regional and local levels” (Standing
Advisory Group 1999, p.4)
The ‘Consumers in NHS Research Support Unit' was funded to encourage strategic
alliances and the development of regional collaborations with consumers (DH 1999a).
In the following year key publications were produced aimed at supporting consumers to
get involved and informing researchers about best practice (Consumers in NHS
30
Research 1999, 2000). At the second national conference, ‘Research: Who’s learning’
Lord Hunt of King's Heath, the Parliamentary Under Secretary of State for Health, gave
the keynote speech and offered strong support for the involvement of consumers in
NHS research and development. The topics of debate broadened to include
consideration of different techniques for involvement, user-led research, involvement in
randomised controlled trials and ethical review, and research involving people from
Black and Ethnic communities (Consumers in NHS Research 2000).
At the present time Labour's vision for the NHS sets new types of rationing and
prioritizing tasks for the informed consumer-citizen. Patients are encouraged to
prioritize their own health, through initiatives such as the Expert Patient Programme
(DH 2001f) and health checks. Since the beginning of 2006, patients in England have
had the choice of four or more providers when referred by a general practitioner for
hospital care. Choice is the mechanism which is proffered to assure service
organisations became more responsive to the needs and wishes of the consumer (Barnes
and Walker 1996). Funding will flow directly to those services that are required (Spiers
2003) and patients will receive a comprehensive range of high-quality services (Greener
2005).
Safety and quality
Despite trends towards greater recognition of public interests in public services and
greater emphasis on efficiency and responsiveness to users, arguably the most
significant drivers for user involvement have been concerns about safety in health
services and health research. From the late 1960s a wave of scandals surfaced about
specific treatments, such as the prescription of thalidomide to pregnant women. Medical
scandals drew attention to the lack of restrictions around medical research on patients. A
string of public enquiries including the child deaths associated with heart surgery in
Bristol (BRI 2001) and the retention of children's body parts for research at Alder Hey
Hospital in Liverpool (DH 2001g) were the ultimate trigger for political action. Writing
about the impact of the events at Bristol, Coulter (2002) argued that putting patients at
the centre of health services has benefits for the quality and safety of healthcare.
Government were being advised that:
31
“Direct dialogue with the public should move from being an optional add-on
to science-based policy-making and to the activities of research organisations
and learned institutions, and should become a normal and integral part of the
process” (House of Lords Select Committee on Science and Technology 2000,
para. 5.48).
National frameworks for the governance of health (DH 1999b) and research (DH
2001b) were established. In women's health, voluntary health organisations argued that
research should be undertaken in partnership with women, not on women and that
special consideration should be given by researchers and ethics committees to research
on pregnant, labouring and newly delivered women and their babies. The Association
for Improvements in the Maternity Services (AIMS) and The National Childbirth Trust
created their own charter for ethical research (NCT 2000), emphasising that research
should involve women in the planning of studies and should include user representatives
early in this process; echoing developments in health and social services (DH 2001h,
DH 2003b) and reinforcing a further recommendation made by the Select Committee:
“That the scientific merit of particular research grant proposals should

continue to be assessed by peer review; but that the Research Councils should
do more to involve stakeholders and the public in the wider task of setting the
priorities against which particular grants are made, and should seek greater
publicity for the process. We suggest that they might seek the considered
involvement of members of Parliament and local authorities, and of other
people active in their communities; and that they might hold occasional open
forum meetings in different locations” (House of Lords Select Committee on
Science and Technology 2000, paragraph 5.78).
Current research policy and research commissioning guidelines suggest that the quality
of research studies can be improved by involving service users (DH 2006a). The debate
about what quality means in this context is complicated by debates about whether a
range of factors, including the safety, acceptability and relevance of research, can be
improved through participation.
32
Plurality and diversity
An associated trend is the notion that quality within health services can be improved by
encouraging plurality. For example, policy for foundation trusts (DH 2002e) has freed
hospitals from central control and gave them greater range to deliver the type of
services they see fit. Another example is the independent sector treatment centre
programme, which will provide elective surgery and other services to the NHS. This
favouring of plurality supports the idea that there are a diverse range of roles for
patients and the public in decisions about health services and research.
In current health policy user involvement is described as offering multiple advantages

including: patients and the public being part of building quality services (DH 1999b,
2000c), overcoming social exclusion through active citizenship (DH 1998b, 2001a,
2002a) and promoting public participation in decision-making (DH 1998b, 1998c,
1999c, 2001a, 2002a). Health policy constructs multiple public identities through its use
of particular terms, for example, active citizenship (Higgins 1993), confident consumers
(Irwin 2006), and expert patients (DH 2001f), who are more responsible, more
informed and more actively involved in choices about their own health (Bournes 2000,
Dowding 1992). The personal choice of the individual service user becomes a strategic
controlling point for social change.
Conceptualizations of public participation, such as Arnstein's (1969) ladder of citizen

participation (illustrated in the next chapter), support the idea that different levels of
participation are appropriate in different circumstances of public decision making and
health care contexts (Brooks 2006). Rhodes and Nocon (1998) suggest that to overcome
potential difficulties of user involvement there is a need for varying approaches that are
receptive to user views and experience rather than a single model. The idea of
‘participation for purpose' is reconfirmed by concepts of deliberative participation
internationally, for example, in Canada Abelson et al. (2003) and Abelson et al. (2007),
and by the adoption of a spectrum of levels of participation developed by the
International Association for Public participation (Involve 2005) (Figure 3).
33
Growing interest in user-led and user-controlled research bring an interesting

dimensions to discussions about user involvement in health research. There are
differences in the way these terms are used in practice; user-controlled suggesting that
service users have overall control, whilst user-led research can mean that research is led
issues that are important to service users (Turner and Beresford 2005). Differences
between user involvement in ‘mainstream' health research and user-controlled research
relate to how the research process is perceived (Beresford 2003). Traditionally, research
had been perceived as a way of attaining and transforming lay knowledge (discussed
later in chapter 5). Interpretive and constructive approaches to research encourage the
exploration of subjective experience, for example using patients' experiences of health
services to inform service development (Nicolson et al. 2000).
Figure 3: International Association for Public Participation

(Involve 2005, p.18)
Inform
Public participation goal:
To provide the public with balanced and objective information to
assist them in understanding the problem, alternatives,
opportunities and/or structures
Consult
To obtain public feedback on analysis, alternatives and/or
decisions
Involve
To work directly with the public throughout the process to
ensure that public concerns and aspirations are consistently
understood and considered
Collaborate
To partner with the public in each aspect of the decision including
the development of alternatives and the identification of the
preferred solution
Empower
To place final decision-making in the hands of the public
34
An illustration of these trends is that in 2001 The Sainsbury Centre for Mental Health
launched a report, “Users’ Voices” (Rose D. 2001) of research carried out using a User
Focused Monitoring (UFM) model on the views of mental health service users about
their experiences of services. UFM had been pioneered by the Sainsbury Centre in 1997
with a pilot user-run project in the London Borough of Kensington, Chelsea and
Westminster. The report was ground breaking because people with severe and enduring
mental health problems created, carried out and analysed the research. This work has
been taken forward into the Service User Research Enterprise (SURE) at the Institute of
Psychiatry. SURE conducts primary research and reviews and provides advice to
service organisations and clinical academics.
In health research commissioning it is more widely accepted that research can be carried
out for different purposes, in different ways, with different outcomes. At the same time
there is a growing expectation that health research should have explicit utility to clinical
practice or health services leading to improved patient outcomes. These trends are
exemplified by translational research which is becoming increasingly important in the
shaping of basic scientific research (Wainwright et al. 2006). Whilst at face value this
expectation is very reasonable, translation of research into clinical practice may not
always be feasible or desirable. For example, in some situations research ‘at the bench'
may inform subsequent lab based research, or the development of technology, services
or professional education. Hence, health research may have no direct connection or
measurable impact on what happens ‘at the bedside'.
By 2002 many more health organisations and consumer organisations were engaging
service users in new ways. For example, Marie Curie's Palliative Care Research and
Development Unit set up a consumer group along with the Royal Free Hospital, London
and the Cancer Research UK Targeting Group (Consumers in NHS Research 2002).
The Alzheimer's Society Quality in Dementia (QRD) Consumer Network programme
had a 140-strong UK network of carers and users who were contributing to strategy
setting, commissioning, grant application review, grant awards and project monitoring
(Consumers in NHS Research 2002). Attention to engagement issues was also occurring
within the health service, for example attempts were being made to remove
35
stigmatization and discrimination from the NHS and provide more support for people
with mental health problems to take up employment within the health service (DH
2002d). More resources were also being made available nationally to support
researchers and the public to work together. For example the BUPA Foundation offered
funding of up to £500,000 for one or more projects to advance thinking and practice to
‘address consumer involvement in healthcare'(Consumers in NHS Research 2002).
Consumers in NHS Research produced a guide to paying consumers actively involved
in research, now replaced by the document Reward and Recognition (DH 2006b). Some
organisations were offering training and consultancy based on their experiences of
working with users, for example the Leonard Cheshire Disabled People's Forum on user
participation and empowerment and the “Voices in Action Project”, funded by the
Department of Health and based at the College of Health, offered a sustainable training
and support scheme for lay representatives working in the NHS.
In 2003 Consumers in NHS Research changed its name; to ‘INVOLVE'. Its overall aim
was now:
“To promote public involvement in research in order to improve the way that
research is prioritised, commissioned, undertaken, communicated and used.
INVOLVE believes that the active involvement of the public in the research
process leads to research that is more relevant to people and is more likely to
be used. Research which reflects the needs and views of the public, is more
likely to produce results that can be used to improve practice in health and
social care”. (INVOLVE 2003)
In recent years the idea of ‘research as empowerment' has also been applied in the
commercial sector. The market and public opinion research agency MORI formed a
Participation Unit with the objective of engaging people in a dialogue and to
“empower the researched to become the researcher” (Christopoulos and Rose 2005).
The organisation began to employ new more sophisticated techniques such as
pyramid interviewing (training members of the community to interview others), and
bifocal groups (discussion groups involving people with contrasting experiences e.g.
providers and users).
36
The national Health Act (DH 2006c) consolidates much of the current legislation
concerning the health service, which may award increased support for user involvement
in health research. It came into force on 1 March 2006. Section 11 of the Health and
Social Care Act (DH 2001d) became section 242 of the consolidated Act 2006. This
stipulates that:
“Each relevant English body must make arrangements, as respects health

services for which it is responsible, which secure that users of those services
are, whether directly or through representatives, are involved (whether by
being consulted or provided with information, or in other ways) in – the
planning of the provision of those services, the development and consideration
of proposals for changes in the way those services are provided, and decisions
to be made by that body affecting the operation of those services” (DH 2006c,
Section 242)
The establishment of the National Centre for Involvement (2006) aims to support NHS
staff and organisations to create services that are ‘directly' shaped by the views and
experiences of patients and the public.
2.3 Public involvement within the knowledge economy
In this section I look at user involvement in health research in the context of moves
towards participative governance across public services. I argue that whilst government
has pushed for greater direct public involvement in the governance of health services at
organisational and strategic levels, this trend has been less prominent in the context of
health research. Health service policy making has certainly been influenced by public
concerns about the efficiency and effectiveness of health services. From 1 January 2003
Section 11 of the Health and Social Care Act placed a statutory duty on NHS trusts,
Primary Care Trusts and Strategic Health Authorities to make arrangements to involve
and consult patients and the public in service planning and operation, and in the
development of proposals for changes. The establishment of new institutional
mechanisms to support public participation within and alongside the NHS, for example
Patient Advice and Liaison Services and Patient Forums (DH 2000b), reinforces this
policy. Patient and Public Involvement (PPI) has developed as a stream of work in the
37
NHS. NHS organisations have invested heavily in consultations such as patient surveys
and patient satisfaction questionnaires (Crow et al. 2002), often managed by external
consultancies and adopting approaches from market research (Jones et al. 1990). As a
consequence of these developments patient views and experiences are now more
commonly being sought and used to inform service development activities and to
improve the quality of health services.
The picture of public involvement in relation to health research is somewhat different.

The formulation of contemporary UK health research policy is characterised by
managed innovation and technology transfer, regulatory intervention, incentives,
assessment, governance and the influence of experts promote the standardization of
activities and encourage researchers and the organisations in which they are based to
take on particular types of research production and practices to the exclusion of others
(Shaw 2006). In this climate user involvement is a detectable rather than a substantial
feature of research policy. Where policies for user involvement exist they are
ambiguous and open to interpretation, for example, the ‘whenever appropriate' wording
on user involvement in the Research Governance Framework for Health and Social Care
(DH 2001c). There are associated tensions about professional control over research and
decisions about user involvement in research (discussed in chapter 6).
This is not to say that issues of public involvement have not been raised at different
stages of development of UK health research policy. For example a needs and feasibility
survey on the involvement of consumers in the NHS R&D Health Technology Research
Programme found that:
“The most fruitful way of identifying research questions that are important to
consumers is to discuss it with them face to face and offer follow up
support”. (Oliver et al. 2001 p.113)
Another example is the public consultation that preceded the government strategy for
NHS R&D in England, Best Research for Best Health (DH 2006a). As part of this
strategy, the National Institute for Health Research (NIHR) was launched on 1 April
2006. The strategy covers research involving patients, samples or data taken from
38
patients, people who are not patients, populations, health technology assessment, and
health services research. Under this programme a new funding strand, Research for
Patient Benefit, was launched. Also in April 2006 the UK Research Councils, working
together through Research Councils UK developed a Science in Society Strategy
(RCUK 2006). The UK Research Councils are the biggest public funders of research in
the UK, spending around £3 billion every year. The strategy set out to contribute to
raising pubic awareness of science and innovation, to encourage public involvement in
these subjects, and to encourage researchers to engage with the public.
Patient and public perspectives have not always been directly present in research policy
development, perhaps indicating an assumption amongst policymakers that professional
judgement serves the best interests of such groups. Professional representation of public
interests is suitably illustrated by the Cooksey review to build agreement on the best
institutional arrangements for the new single fund for health research. The review
concluded that, although good progress had been made in some areas further work was
needed to ensure that publicly funded health research is carried out in the most effective
and efficient way (Cooksey 2006). It is of note that the review focuses on health,
science and economic objectives, but that no mention is made of objectives from the
perspective of patients and the public. The UK is not unique in emphasising the capital
benefits of a knowledge economy, such values are upheld as being central to the success
of the European Union (Darr and Warhurst 2008).
Whilst health services and health research present different contexts for involvement it
is interesting to observe how this agenda has been made use of politically and
strategically in each of these domains. What I am also interested in here is what user
involvement means for individuals working or participating within one or more of these
domains.
39
2.4 Background to the study
My interests in user involvement are influenced by the social and political trends I have
previously discussed, in particular the issues of quality in research and who research is
for. I will explain more here.
The first event I have depicted on the study time line is a conversation between
Professor Fiona Ross and myself about what the study might comprise (Figure 4). Fiona
and I had recently completed a national scoping study to identify research priorities for
nursing and midwifery service delivery and organisation (Ross et al. 2004). This
exercise had involved consultations with members of Community Health Councils
(CHCs) across the UK. It was Fiona's idea to include a ‘user' perspective in the priority
setting exercise. I still don't know where this suggestion came from but it seemed like
an interesting thing to do. Sally Brearley, a member of a local CHC provided advice
and links to CHCs nationally. Meeting and talking with members of 38 CHCs during
the scoping exercise had a significant influence on the depth of information we gained
(Smith et al. 2005a). The information that had been provided to us seemed far clearer
and more pertinent to future development of nursing research than what was written in
policy documents or the published literature. This experience fuelled my desire to show
that involving service users in processes like this was achievable and that it added value
to research.
In 2004 the scoping exercise (Ross et al. 2004) led to calls for proposals in areas that
had been identified as research priorities. One of these areas was user involvement in
the design and undertaking of nursing, midwifery and health visiting research. I was
keen to apply for this funding and to review the evidence in this area. Fiona, Sally and I
developed a proposal that aimed to involve service users more actively in the review
process itself and we secured interest from Professor Peter Beresford at the Centre for
Citizen Participation, Professor Jill Manthorpe at the Social Care Workforce Research
Unit and John Sitzia at Worthing and Southlands NHS Trust (now at the UK Clinical
Research Collaborative).
40
Figure 4: Study time line
Discussions with Ross about topic area (Jun)

2003
Draw up initial ideas for PhD (Sep)
Begin making reflective field notes (Oct)

Initial literature review (Oct-Nov)
Submit proposal for PhD study (Nov)
Grocott agrees to supervise study (Dec) Registered for MPhil/PhD (Jan)
Initial contact s to experts/leads in field (Feb) Funded
start of Multi-method review (Mar)
Evidence-based decision making course (Mar) Ethical

approval for multi-method review (Apr)
Systematic searches of literature (Jun) Proposal 2004 Workshop in Glasgow (Aug)
for GSTT funding (Sep)
Reading about phenomenology (Oct)
Analyse current concepts of user

involvement in health research (Nov)
Presentation in Belfast (May)
Proposal for Australia research (Jun)

2005 Set up Australia research (Jul)
Conversion from MPhil to PhD (Sep)
Presentation of Work in Tokyo (Sep)
Australia research (Oct/Nov)

Initial analysis of Australia data (Feb)
Begin to read Foucault (Apr)

Seminar at Kings’ Fund, London (May)
Proposal for case study (Jun)
Ethics for case study rejected (Aug) 2006
Explore theories of discourse (Sep)
Growing interest in knowledge/power (Oct)
Reframe objectives & begin to write thesis (Oct)
Reading about ethnography & reflexivity (Nov)

Reading about Reflexive Interpretation (Feb)
Presentation in Hong Kong (April)

2007 Presentation in Dundee (May)
Analysis of findings (Aug/Sep)
Discuss potential examiners with supervisors (Oct) Submission of

examination forms (Dec)
Presentation of work in London (Dec)
End of Doctoral Training Award (Mar)

2008 Full first draft (Apr)
Submission of thesis
41
At that time Sally Brearley was providing formal advice to another project that Fiona
was leading on about the prevention of falls in older people (PROP) (Ross et al. 2005).
Between us our collective experiences gave us confidence to aim for more substantial
user involvement as part of a multi-method approach. Participation in research was an
issue I had encountered from a slightly different perspective through my work as a
Research Assistant at Kingston University and St George's Hospital Medical School
(now St George's, University of London) on Action Research studies with secondary
schools in South London. The work concerned issues about young people and sexual
health. I learnt about the practical aspects of research relationships, such as the need to
meet people on their own terms, how to build relationships through personalized
communication, understanding what is important to different people and the importance
of buffet lunches. I also learnt that different people respond to research in different
ways. Some were excited by the potential, others were concerned about the extra work
it might entail, or that they might not understand it, or relate to it. Research with young
people helped me to overcome some of my own apprehensions about research. In this
context quality research was not about strictly controlling situations it was about finding
the best ways to understand.
In July 2003 I took up a post as Research Associate in the Nursing Research Unit (now
the National Nursing Research Unit) at King's College London. I soon became involved
in a review project funded by the Commission for Health Improvement (superseded by
the Healthcare Commission) to identify how patient experiences have been attained and
used in three clinical areas: cataract care, hip replacement and knee arthroscopy. The
review reported issues about the diversity of patients, different perspectives of
continuity, and the need for patient information and support generally (Smith and Ross
2006) (Appendix 11). The findings suggested that to be effective patient views and
experiences need be incorporated within organisational structures, service development
initiatives, health service research, as well as in professional education and training.
In early 2004 funding was awarded for the multi-method review of user involvement in
nursing, midwifery and health visiting research (Smith et al. 2005b). The project
involved a service user reference group with membership drawn from nationally
42
focused voluntary organisations and patient interest groups. Working with service users
brought quality to the review process in terms of making informed decisions about the
scope of the review, deciding which issues to focus upon and increasing clarity of the
issues through discussion and debate (Smith et al. 2008) (see Appendix 11). Once the
multi-method review was underway I registered for a part-time MPhil/PhD. My interest
at that stage was in understanding how service users' knowledge could be used in
research. I was dubious about how much control service users should or could have over
research. For me it was important to show researchers what they stood to gain by
involving service users in their research.
2.5 Finding focus
The objectives of this study have changed over time. I began with a desire to provide
‘evidence' to researchers about ‘how to do' user involvement in health research. My
thinking along these lines was influenced by attending national debates:
- In February 2004, at the Better Information Better Communication in Healthcare
conference, organised by the Centre for Health Information Quality, Harry Cayton
(Director of Patient Experience and Public Involvement Department of Health)
raised issues about the importance of finding ways to involve patients in research
about communication and information.
- In March 2004, at the third annual NHS Service Delivery and Organisation
Conference the need for a better understanding of the impact of user involvement
in health and research was the focus of a heated debate.
I felt that using this study to identify factors that support user involvement in health
research would be beneficial for establishing ways for professionals and organisations
to prepare and to target future development activities and resources. My intention was to
develop some form of principles or a tool for researchers about ‘how to do user
involvement'. I would seek to explore researchers' views and perceptions of the
purpose, benefits and challenges for involving service users in research, and then,
43
drawing on existing concepts of readiness I would identify sequences of activities

associated with individuals and organisations becoming ready to involve service users
in research.
If this study had produced a tool or a technique about ‘how to do user involvement' it
would possibly help the advancement of the practice of user involvement in health
research. A technique could also be tested and validated in practice. It may even be of
use and application. However, it would not be theoretically grounded because the
underlying concepts of user involvement are uncertain. I would be assisting people to
practice user involvement, without knowing what user involvement in research is.
Furthermore, a plan to test the tool for validity (appropriateness) and reliability (test-
retest) in three different research contexts: nursing practice development,
medical/clinical trial, policy would be of limited use. I could not be sure that the tool
would measure or advocate the right thing for the right purpose.
Box 1: Objectives of the study at proposal stage (January 2004)
The purpose of the proposed research is to better understand the mechanisms that
enable service users to be involved in research about health services in order to develop
a tool for assessing readiness for user involvement.
The objectives of the research are:

1. To use the findings of a systematic literature and policy review on user involvement
in nursing, midwifery and health visiting research to identify barriers and facilitators
of user involvement in health research.
2. To consult with (i) service users (semi structured interviews) and (ii) researchers
(semi structured interviews) about their experiences of user involvement in health
research.
3. To use the findings of the literature review and interview data to develop a tool to
assess readiness for user involvement. (Phase 2)
4. To test the tool for validity (appropriateness) and reliability (test-retest) in three
different research contexts: nursing practice development, medical/clinical trial,
policy. (Phase 3)
The original focus of the study on organisational and professional ‘readiness' to involve
service users (see Box 2) was a poor fit. The concepts of readiness I encountered in the
literature narrowed an exploration of meaning because readiness generally has an
outcome associated with it, and precursors to it, as shown for example in relation to
44
self-care (Baker and Stern 1993), health promotion (Procheska et al. 1994), and
discharge planning (Schaefer et al. 1990, Titler and Matsen Pettit 1995, Williams et al.
1995). Other conceptualizations of readiness perceive it as a process of becoming ready
or adjusting to a change; as has been shown in the education of health care professionals
(Kagan 1992; Redman 1993; May and Campbell 1981); or in coming to terms with a
diagnosis (Dalton and Gottlieb 2003). Concepts of readiness suggested an intended end
point, and a method, or series of steps that can be reproduced for the same outcome.
This did not fit with my early understandings of user involvement. Its inscrutable and
unpredictable nature could not be meaningfully reduced to a series of action points.
Furthermore, readiness was suggestive of fulfilling the will of an external authority,
which did not reflect the processes of explanation, negotiation and relationship building
that I had experienced.
Box 2: Objectives at MPhil/PhD transfer (September 2005)
The study aims to identify the factors that support user involvement in health research in
order to establish ways for professionals and organisations to prepare, and to target
future development activities and resources. Although the focus of the study is user
involvement in nursing research, it crosses boundaries between nursing and health
services research and is informed by current health policy as well as social research
methods.
1. To identify influencing factors in the development of user involvement in policy and

research practice.
2. To build understanding about the influence of research cultures and research policy
on user involvement.
3. To explore views and perceptions of the purpose, benefits and challenges for
involving service users in research.
4. To identify the challenges and support needs for researchers involving service users
in research.
5. To use concepts of readiness to identify sequences of activities associated with
individuals and organisations becoming ready to involve service users in research.
Following the MPhil/PhD transfer in September 2005 (see Appendix for notes of the
discussion) I focused the study on user involvement in health research, drawing a
distinction between the political agenda for Patient and Public Involvement in the NHS.
Within this scope I could focus on alternative perspectives of the issues (see Box 3)
including a plan to undertake an international phase of the research in Australia. I was
45
becoming more aware of the fact that different types of knowledges exist and user
involvement can be used for different purposes in different contexts.
Box 3: Objectives (October 2006)
The thesis is that understandings of user involvement in health research are personal
and particular to individual researchers and that this has implications for practice, future
research and policy development.
The purpose of the research is therefore to interpret and seek to understand multiple
views of user involvement in health research, in terms of actions and philosophies.
1. To define current knowledge of user involvement in health research.

2. To use observation and inductive qualitative research methods to collect data
directly from the field of user involvement in health research.
3. To construct a theory of user involvement in health research using constant
comparative methods.
4. To use reflexivity to show the limits of the theory.
5. To test the theory for resonance.
Reading the research literature on user involvement continued to influence my views

about the focus of the study (explained later in Section 3.7). A rejected ethics
committee application in August 2006 was a key turning point because it raised
questions for me about the force of power I was feeling. I was not simply about the
outcome of the decision. I felt the rejection undermined my ability to research user
involvement and hampered my potential to complete the study. More significantly it
quashed my sense of being involved in a decision making system. These feelings
stimulated a dialogue between my supervisors and I that has been of utmost importance
to this study. I firstly had to come to terms with the rejection and then to understand my
unrest about the way the decision had been made. Once I had shared these views with
my supervisors we moved on to discussions about alternative lines of enquiry. My
interests had moved away from professional and organisational readiness, in the sense
of providing structure, technique or procedure – to exploring in a much broader sense,
how relationships between researchers and service users are constructed and
experienced.
46
Box 4: Objectives (November 2007)
The aim of this study is to construct a critical interpretation of user involvement in health
research.
The objectives are:

1. To use a systematic review of the evidence, Foucauldian Archaeology and my
subjective experience of reading the literature to show how the object of user
involvement in research can be approached from different epistemological stances.
2. To use Foucauldian genealogy and theories of power to explore the ways in which
user involvement is played out.
3. To use empirical data from UK academic research, NHS patient and public involvement
and an international comparison to explore subjective experiences of power in user
involvement in research.
4. To identify implications of the findings for NHS R&D, paying specific attention to
possibilities for professional education, policy and practices.
I became more interested in key concerns in this field about showing evidence of user
involvement. For example, leading researchers in the field suggest:
“Given the resource implications of undertaking user involvement in service

development there is a need for critical debate on the purpose of such
involvement as well as better evidence of the benefits claimed for it”. (Fudge et
al. 2008, p.313)
One of the issues I have spent time exploring is the value of attempts to measure or
evaluate user involvement in health research, and the problems of accessing the
intentions behind actions. Part of the contribution of this thesis is the identification of
possible ways forward for developing knowledge in this topic area. But first, it is
necessary to take stock of current knowledge in this field.
47
Summary
 The historical social and political conditions from which user involvement in health
research have emerged are characterised by concerns about recognition of public
interests in public services, efficiency and responsiveness to users, rationing and
priority setting, quality and safety, and pluralism.
 The motivation for this study came from previous work including a national scoping
study to identify research priorities for nursing and midwifery service delivery and
organisation, participative action research studies about young people's sexual
health and a review of patient experiences of care pathways. These studies indicated
that working with service users in the research could add quality and bring about
change.
 The focus of this study has changed over time. It began with the intention of
assessing organisational and professional readiness and providing ‘evidence' to
researchers about ‘how to do' user involvement in health research. The focus then
became to explore what user involvement is and what it does.
48
Chapter 3: Current knowledge about user involvement
in health research
3.1 Introduction
This chapter presents an analysis of the published research literature on user

involvement in health research. The aim is to identify and critique features of the field
and to seek out theoretical and empirical weaknesses. The vantage point for this
endeavour is important for where one stands determines what the landscape looks like.
Issues about how the field and the knowledge base are defined are discussed in more
depth in chapter 5. The position taken in this chapter is to use a structured review of the
published research literature on user involvement in health research. This is not the only
vantage point that this study makes use of. For example, at the end of this chapter I
present a personal experiential account of the literature. Previously, in chapter 2 I
identified social and political trends which have contributed to the construction of
understandings about user involvement in health research. The analysis presented in the
present chapter explores a different view: that knowledge about user involvement in
health research should be of sufficient quality and quantity to inform research practice
and, that researchers must be able to access evidence about approaches to identify
effective strategies for involving service users. This view is upheld by current
government research strategy (DH 2006a) and researchers currently working in this
topic area (for example, Beresford 2004; Boote et al. 2002; Fudge et al. 2008; Hanley
2005; Oliver et al. 2001; Simpson and House 2002; Staniszewska et al. 2007).
3.2 Scope and analysis
The definition of user involvement in health research applied for the purposes of this
literature review located user involvement according to a particular form of knowledge
generation (research), a form of practice (involvement) and a population group (users,
carers, patients, the public). The definition and selection of publications (summarized in
tables in Appendix 2) did not rest in any way upon the research being undertaken by
professional researchers, which allowed for the possibility that publications may be
49
Chapter 3: Current knowledge
written by service users. However, the parameters of the searches are important and
these are outlined below. Later, I will look at issues of defining the knowledge base
(Chapter 5) and issues to do with developing an evidence base (Chapter 7).
The searches were made in September 2007 and online sources were updated in June
2008. Searches were limited to electronic sources including: Ovid Medline, Embase,
CINAHL and the British Nursing Index (BNI) for the years 1960-September 2007. This
information was supplemented by publications identified in reference lists and
publications I was aware of through the multi-method review searches (see Appendix
6). These additional references are indicated and distinguished (*) in Appendix 2.
The core search strategy was based upon terms linked to user involvement in health
research and participative research in addition to alternative terms for involvement (Box
5). Where appropriate, terms were truncated to maximise the sensitivity of the search
terms. On all databases, the subject search was linked to alternate terms for users
(patients, carers, public, lay) and index terms relating to public involvement and
participation. The Boolean term (AND) was used to increase the specificity of the
returns. The searches were limited to editorial board or peer reviewed journals and
publications in the English language.
The UK is not unique in requiring researchers to show commitment to user involvement

in health research. For example, applicants for European research funding are required
to explain plans for involving a range of stakeholders (Wickham and Collins 2006).
However, user involvement is referred to and thought of in different ways
internationally. Differences in terminology complicate possibilities for international
comparison but it does not make them impossible (indeed, as I will go on to show, my
experiences in Australia have shaped and informed the thesis in many ways).
50
Box 5: Search terms
Key words
user
patient
client
carer
public
lay
stakeholder
Next to involvement Or participation
And or Next to
research
Index terms:
Patient involvement
Participation
User involvement
For the purpose of this particular analysis the aim was to make a broad assessment of
research publications in this field. It included:
 Papers concerning the involvement of service users within research studies in any
area of health research (including medicine, nursing, mental health and so on).
‘Research' was perceived as any form of systematic enquiry, including descriptive
evaluations. ‘User involvement' was defined in terms of the widely recognised core
criteria stipulated by INVOLVE (2007) as ‘active involvement' in designing,
undertaking or evaluating research.
 Papers concerning user involvement in the commissioning or review of research.
 Papers that examined the impact or experience of research by people identified as
research participants or people defined as being service user representatives.
 Papers about the education training or experiences of researchers working with
service users.
Excluded papers were:

• Papers primarily concerning user involvement in research about public services
other than health services, such as housing or crime prevention.
51
• Papers reporting on user involvement in health service development or health care

decision making and those concerned solely with user involvement in professional
education (however selective papers from these fields are used later in the findings
and discussion chapters).
• Papers concerning involvement or decision making about a person's own health
care.
• Papers primarily concerning competency or willingness to consent to participate in
clinical research trials. (Papers reporting potential participant's views of
participation are included)
• Papers relating to the involvement of clinical professionals in research or the
willingness of clinical professionals to support recruitment to research studies.
• Letters to journal editors or comments on articles.
Analysis
Selection against the inclusion criteria was tested by independently (a supervisor) rating
a sample of 20 papers that included a variety of research types and topics; and equal
numbers of studies. Agreement between the reviewers on the basis of titles was high (80
%+) areas of disagreement tended to be where studies were not explicitly about research
or user involvement. Abstracts were used to screen borderline citations.
Although the exercise aimed to draw conclusions about the methods studies have used it
was problematic to judge quality on this basis. Few if any studies complied with
consensus standards for the reporting of research, as emphasised by the evidence based-
practice movement since the publication of the CONSORT guidelines (Altman 1996),
even though these exist for many of the designs reported including non-randomised
evaluations (Des Jarlais et al. 2004) and qualitative research (Elliott et al. 1999). A more
detailed assessment of methodological aspects of quality would be difficult.
The core data extracted from each study was the locality of the research, the topic and
nature of research, the groups studied and sample sizes, key findings/issues raised
52
(Appendix 2). The papers were scrutinized to identify themes and trends in the
literature. Judgements about themes were discussed and refined with supervisors.
3.3 The quantity of research activity
Initial searches yielded 230 unique references. Early years (from 1991 to 1999)
rendered considerably fewer studies than the later years (2000 onward), with peaks of
publications in 2005 (20), 2006 (20) and 2007 (25). Figure 5 illustrates these trends. The
volume of publications has increased over the period the peak year of publication 2007
yielded 25 studies. Indexing of 2008 is unlikely to have been complete at the time of the
searches (September 2007) even though these were updated in June 2008. Hence figures
for 2008 are not included on the chart.
The relatively small amount of research specifically about user involvement in health
research may be because user involvement in health research occupies an ambiguous
space between methodological research and policy research and patient experience.
These issues are discussed in the following section. The UK appears to make a
disproportionately large contribution to the international literature on this topic which
could be attributable to the distinct policy agenda in this country. The papers identified
here indicate there is a small but significant level of activity within Europe, Canada and
Australia, although cultural differences in terminology are likely to mean the level of
activity is under represented in these searches.
Within the papers returned the types of terminology and focus fall into four broad
categories: community focused research (early 1990s and earlier), participation (from
mid 1990s onwards), consumer involvement (end of 1990s-early 2000s), and patient
and public involvement in research (early 2000s onwards). These trends correspond
with broader changes in health and research policy discussed in chapter 2. Papers relate
to disciplines across the health fields (medicine, nursing, midwifery, professional allied
to medicine, psychiatry and so on); as well as particular types of research contexts
(clinical trials, practice development, service evaluation). The context of particular
53
studies is explained in the text and summarized in the literature tables shown in
appendix 2.
Figure 5: Previous research studies
Patient participation in research
25
20
Consumer involvement
15
Community research
Participation
10
0
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
Notes:
Searches of online electronic databases including: Ovid Medline (11208), EMBASE (473),
British Nursing Index (51) were undertaken for the years 1960-2008. Initial searches yielded
230 unique references. Early years (from 1991 to 1999) rendered considerably fewer studies
than the later years (2000 onward), with peaks of publications in 2005 (20), 2006 (20) and
2007 (25). Total included: n=170
Excluded papers (n=15)
Additional included papers (n=43): Papers primarily relevant to the topic but not returned by
the above searches. (Not included in the above figure). Indicated (*) in Appendix 2.
54
3.4 Themes of the research literature: definition, role, hierarchy,

ideology
Thematic analysis of the research literature reveals that issues of definition, role,
hierarchy and ideology have a strong presence. These themes are discussed in this
section.
Some additional overarching points can be made about the research literature in this
field. First, most of the included papers make links to literatures associated with health
care or other disciplines, for example about lay-professional relationships in other
health fields (e.g. Baxter et al. 2001), the delivery of health care (e.g. Hickey and
Kipping 1998), health service design and development (e.g. Oliver et al. 2001; Holosko
et al.
2001; Paterson 2004), training and education (e.g. Faulkner 2004), and social capital
and social enterprise (e.g. Evans et al. 2004). Second, there are few papers that are
explicitly about user involvement in health research. There are overlaps with literatures
on research methods and techniques (depicted in the following figure 6) such as
information giving (e.g. Thornton 2001), consultation (e.g. Donovan et al. 2002),
deliberation (e.g. Davies et al. 2006), collaboration (e.g. Evans and Fischer 1999), and
participative action research (e.g. Kuo et al. 2006); often reflecting the types of methods
studies have used.
A third point is that the papers make links to theory literatures: for example, on
concepts of the patient or public (e.g. Dixon et al. 1999), public engagement (e.g. Shea
et al.
2005), decision making, consumerism (e.g. Barnes and Walker 1996), end-user
involvement (e.g. Browne et al. 2004), patient participation (e.g. Cahill 1996),
empowerment (e.g. Rodwell 1996), and health inequalities (e.g. Beresford 2005).
Hence, the high degree of overlap with a wide range of literatures from different
disciplinary fields, research methods and theory, together with alternative terminology,
an absence of indexing on online publication search engines, makes the field difficult to
define.
55
Figure 6: Research literatures relating to user involvement in health research
Research literature
on user involvement
in health research
Definition
Research literatures Role Theory/concept
associated with Hierarchy literatures
disciplines Ideology Patient, public, pubic
Health care engagement, decision making,
Sociology consumerism, end- user
Management involvement, participation,
Design/architecture empowerment, well-being
Education
Community development
Research literatures on
methods and techniques
Consultation, deliberation,
co-construction,
collaboration,
Action Research
(i) Definition
Issues of definition are a prominent theme of the literature and they are important for
understanding what user involvement in health research is and how it is understood.
Early articulations of user involvement in health research appeared in research literature
on community focused research in the late 1980's (for example Bang and Bang 1991,
Seeley et al. 1992). Researchers using methods of co-construction and participation
began to describe those involved in research as ‘participants' (e.g. Thurber et al. 1992,
Robertson 1994, Smith Rogers et al. 1994), rather than using the traditional term
‘research subject'. The term ‘subject' was criticised for representing people as passive
rather than active participants in the construction of knowledge. Hence, from the early
1990s onwards the literature more commonly uses the term ‘participant', even if the
nature of underlying power-relations remained the same.
56
From the mid 1990s onwards more papers address issues to do with the meaning
and mechanisms of patient participation in relation to health service development
(for example, Beresford and Croft 1996, Cahill 1996). Towards the end of the 1990s
until the early 2000s the literature focuses on ‘consumer involvement' in research
(for example, Heller et al. 1996, Liberati 1997 and Salcido 1997). Whilst from the
early 2000s onwards studies using the term ‘public involvement' begin to appear
internationally (as discussed by Beresford 2007), as do the terms ‘carer-led' and
‘survivor-led' research (Turner and Beresford 2005, Campbell et al. 2007).
Because the terms ‘consumer' ‘lay' ‘service user' ‘patient' and ‘public' have all been
used to describe ‘user' involvement in research this complicates how it can be identified
in practice. Boote et al's (2002) review of ‘consumer' involvement in health research
developed a classification based on individual, local and national groupings (figure 7).
Figure 7: Classification of consumers in health research (Boote et al 2002)
Consumers in health research
Individuals
• Patients and carers
• Patient representatives
• Patient advocates
• Citizens
• Potential users of services
(Inter)national
Local groups
consumer
• Population groups
organisations
• Support groups
• Statutory bodies
• Convened groups
• Charities
• (Inter)national
support groups
57
Although still widely used in policy and research publications, the term ‘consumer' has
been perceived as being a politically derived term at odds with values of partnership and
inclusion (Beresford 2005b). Movement in language away from the use of the phrase
‘consumer involvement' indicates an attempt to diminish links to consumerism. For
example INVOLVE was formerly known as Consumers in NHS Research (Chapter 2)
and the “Medical Research Council Consumer Liaison Group” is now called the
“Medical Research Council Advisory Group for Patient Involvement”.
Problems of definition are linked to issues of representation: defining who can, or

should represent the views of service users and under what circumstances (Telford et al.
2001). The issue of alternate concepts of representation is considered in greater depth
later in this thesis (Chapter 5). The literature also shows that differences in terminology
have emerged in relation to the contexts in which user involvement is thought of and
used. For example Beresford suggests:
“There is little consensus over terminology or categorisation here. While the

government has begun to talk in terms of ‘patient and public involvement’
(PPI), the term most often employed is service user involvement. This tends to
mean people who use, have used or are eligible to use health and social care
services, particularly on a long-term basis” (Beresford 2007, p.307)
A related issue is that the term ‘user' has been applied indiscriminately in policy and
research publications to mean those who may use research, those whom research is
about, and those it might affect (as well as having additional meanings in the context of
health and service use). For example, a review of social science research funded by the
European Union (Wickham and Collins 2006) found that of 276 projects only 88
explicitly defined who the users were (these included policy makers, commissioners,
trade union representatives, members of professional associations, members of the
academic community, industry or business representatives, and local
citizens/community organisations). In these studies there were very few examples of
service users having any direct involvement (n=35) either in defining research
questions, retrieving data or providing an interpretation of the results. An important
question is therefore how researchers themselves define who users are in relation to
particular studies and how this affects opportunities for involvement. Being a recipient
58
of health care and being a participant in research involve different standards for how the
patient/service user is to be treated. Confusion of the two can lead to profound
misunderstandings about what is appropriate (Lidz and Applebaum 2002). The issue of
how service users perceive themselves has been discussed in the context of vulnerable
and marginalized groups (Steel 2005), in relation to mental health service use (Happell
and Roper 2007) and illicit drug use (Fry and Dwyer 2001). Taken together these
studies point towards the positioning power of the term ‘service user' and the overriding
view that service users must be protected from the risks of health research.
Barnes and Wistow (1994) suggest that the concept of the service user is a relational
one and it is useful to conceptualize ‘users' in ways that indicate some of the differences
in the way people ‘use' services. The classification they have developed is as follows.
Box 6: Classification of ‘users’ (Barnes and Wistow 1994)
(1) Voluntary and involuntary users – indicating the degree of willingness with which
individuals engage with services.
(2) Short-term and long-term users – is the relationship concluded when a specific
outcome has been achieved, or is the service offered one which has to be sustained
throughout a lengthy period?
(3) Individuals as allies or competitors – do users of services identify themselves as
having common interests, do they regard themselves as competitors for scarce
resources?
(4) Individuals and group interests – is the service provided and received on an
individual or collective basis?
(5) Actual and potential users – service user should not be equated with the existence
of a need. There are many reasons why someone may not use services, although
he or she could be considered to have needs which services are intended to meet.
(6) Users and carers – carers may have needs in their own right and they may be
direct users of services. However, their needs should be distinguished from the
needs of those to whom they provide a service.
(7) Users and citizens – all citizens are potential users of services, and accountability
to citizens is an essential characteristic of public services. However, the interests
of current users of services may be seen to conflict with those of citizens who are
not users.
The problematic terminology of ‘involvement' has also been debated in the research
literature. The term implies a situation where service users are brought into professional
worlds yet service users may lead, control or undertake research (Turner and Beresford
59
2005). It implies a process of professional invitation and service user acceptance –

underplaying situations where involvement is seen as the fulfilment of a legal
requirement, a moral right or a process of equal partnership (Beresford 2005c). A
further point, and a central argument of the thesis, is that the term defines relationships
between researchers and service users according to existing practices, yet such
relationships can extend beyond, challenge or contest established practice.
(ii) Role
The second prominent theme of the published literature is the notion of service user
roles. The literature provides examples of studies which have awarded service users a
role in research design, including:
 defining research questions (O'Donnell and Entwistle 2004a),
 conceptualizing research issues (Ross et al. 2005),
 designing research methods, such as research trials (Hanley et al. 2001),
 designing or contributing knowledge to systematic reviews (Oakley et al. 1996;
Rose et al. 2004; Stewart et al. 2005),
 defining the purpose of user involvement in the research (Fudge et al. 2007; Rose
et al. 2004),
 designing how user involvement might be undertaken (Thornton et al. 2003).
Service users have been involved in undertaking research activities, for example in:
 recruitment of research participants (Elliot et al. 2002; France 2000);
 validation of research instruments or scales (Browne et al. 2004);
 undertaking data collection (Elliott et al. 2002; Simpson and House 2002). Some
studies have evaluated the impact of users as data gathers (peer-interviews,
administration of a structured questionnaire) but the evidence is inconclusive
because of individual characteristics and contextual issues.
 analysing or interpreting research data (Ross et al. 2005)
 creating research outputs (McClimens 2004) and in
 disseminating research findings (Flaskerud and Anderson 1999).
60
The literature also shows that a range of research activities had been formally evaluated
from a user perspective, including:
 quality of research projects (Hudson 2003; Ramon 2000),
 quality of user involvement in research (Telford et al. 2004),
 experiences of being involved in research (Faulkner 2004; Thornton 2002).
There is a substantial and related literature on lay perceptions of health research which
has been used to inform the recruitment from under represented groups to take part as
subjects in health research (for example, Bates et al. 2004; Alvarez et al. 2006,
Buchwald et al. 2006; Glogowska et al. 2001; Halbert et al. 2006; Molyneux et al. 2005;
Morgan et al. 2005; Plummer et al. 2002) or to plan health interventions (Olson et al.
1997; Ma et al. 2004) or the research intervention received (Daugherty et al. 1998).
This type of study has been essential for ensuring service user perspectives can help to
improve research quality. However, such studies are not generally classifiable as
‘active' user involvement in health research according to the models being discussed
here. There are accounts of participants in research trials being asked to evaluate their
experiences to inform development processes, or to design future clinical trials
(Donovan et al. 2002; Cox 2000; Thornton et al. 2003). Service users have also been
more widely involved in evaluations of service development projects, including user
involvement elements (e.g. Barnes and Wistow 1994) or evaluating the value of
research that has been done, for example, Saunders et al. 2007. The idea of participant
and user evaluation suggests that service users evaluate in ways that differ from
professionals, that is, from a subjective experiential position. Current classifications of
‘active' user involvement in research tend to position service users according to their
possible contributions to a process of research (Dixon et al. 1999); and in relation to
different stages of the research process (Hanley et al. 2000, see figure 8).
61
Figure 8: Involving consumers in research processes (Hanley et al. 2000)
62
A typology of user involvement was developed as an output of the multi-method review

(depicted in figure 9).
Figure 9: Typology of user involvement in research (Smith et al. 2005b)
Designing Doing Evaluating
Questions & Proposal/bids Research

topics outcomes
Peer recruitment
Research Data collection Data Research

methods analysis Developing process
outputs
Dissemination
Objectives of Impact of user
involvement involvement
Approaches to Providing support Quality

involvement involvement
Education/training
It has been argued that classifying user involvement according to activities within
research processes has created an image of user involvement as being static and
premeditated, underplaying its dynamic and emergent qualities (Beresford 2005b).
Current classifications may also serve to under-represent user's contributions to
research by approaching as corresponding with defined roles. Whilst this may indeed
be the case, user involvement may also occur in associated activities such as the
prioritization, commissioning, review, evaluation and regulation of research; or through
the contributions of employed user-researchers within organisations (Griffiths et al.
2003).
63
New roles for service users necessitate the development of training to equip users for
the role they are taking up. The London Primary Care Studies Programme was founded
and evaluated on this basis (Barnard et al. 2005). It tested a model of providing
incentives and enabling support for user involvement. An external support agency
(EQUIP, see www.essexequip.nhs.uk) was tasked with providing tailored support to
eleven projects across a broad range of topic areas in primary care. This study and other
research in this area (Lockey et al. 2004) tends to suggest that training for service users
is beneficial at the personal level and for engaging service users in research. Training in
Research for Service User Evaluation (TRUE) reconfirmed that training can have
enormous value to participants' personal development and confidence and lead to actual
involvement in research. Whilst nationally only a handful of organisations provide open
access training on a regular basis (Faulkner 2004), training for service users has been
identified as a priority.
A related issue, to emerge from community focused research, is under what

circumstances service users should be supported to formulate and initiate their own
research. In one study from the USA the Californian Breast Cancer Programme was
evaluated from the perspective of effective collaboration (Plumb et al. 2004).
‘Successes' of the programme were identified as: empowering women to formulate and
initiate research, involving underserved and hard-to-reach populations, addressing
important and useful research questions, increasing communities' skills and expertise,
and enabling lasting collaborations. Whilst the main ‘weakness' was identified as
involving the broader community in these projects. Barriers were perceived as
including: power imbalances due to the community's lack of experience in the dominant
research funding culture, funding limits and award delays, and the increased service
demand beyond the funding limits of the award that is created. These perspectives
reflect an alternative ideology of user involvement which I will go onto discuss later in
this section.
64
(iii) Hierarchy
Classifications of roles and contributions that users can make to research studies have
paved the way for the development of conceptualizations. The most frequently cited
model in the research literature is Arnstein's (1969) ladder of citizen participation
(Figure 10). Arnstein's model was developed in the United States to represent different
levels of control that the poor had within decision-making yet it has been applied in
relation to understanding user involvement in a wide range of contexts, for example, the
International Association for Public Participation (see figure 3), and UNICEF (Hart
1992).
Figure 10: A Ladder of Citizen Participation (Arnstein 1969)
65
A simplified and commonly applied version is the notion of consultation, collaboration

and control (Hanley et al. 2000), depicted in the following figure.
Figure 11: Levels of control in user involvement (after Hanley et al. 2000)
Levels of control
Power, initiative and decision

User control making is with users
Active, on-going
Collaboration partnership
Professionals use views

Consultation in their decision- making
Within such structural conceptualizations as the one presented here, consultation has
tended to be perceived as a low level of involvement as the researcher/professional
maintains control of the overall agenda (Rodgers 1994). Correspondingly user-led or
consumer-controlled research has been perceived as occupying the opposite end of an
involvement hierarchy/continuum, illustrated by the following conceptualizations.
In nursing practice (Cahill 1996) uses concept analysis to show a hierarchical

relationship between concepts of involvement, participation and partnership (Figure 12).
66
Figure 12: Hierarchical relationship between concepts of participation (Cahill

1996)
Patient
partnership
Patient participation
Patient involvement/
collaboration
Hickey and Kipping (1998) relate the concepts of mental health service user
involvement in decision-making to consumerist/democratic concepts of involvement
(Figure 13). The 'participation continuum' they devise challenges a hierarchical view of
user involvement but the suggestion that information/explanation is closer to
consumerism than to democratization is unfounded. As is the claim that user-control is
democratic or a form of participation.
An ‘empowerment' model, as put forward by Holosko et al. (2001), aims to provide a

structure for service user input to be integrated within human service organisations.
Whilst this might not necessarily relate to user involvement in research organisations or
studies, models such as this reinforce the view that involvement can be classified
according to levels of involvement and related to degrees of empowerment. However,
empowerment is a multi-dimensional concept that has attributes that relate to the service
user, the provider and context specific interactions (Gibson 1991).
67
Figure 13: Participation continuum (Hickey and Kipping 1999)
Figure 14: Framework service user input and empowerment (Holosko et al. 2001)
These models of user involvement have influenced thinking about who has overall
control of research and what is ‘meaningful' and what is ‘tokenistic' involvement
(discussed in chapter 6). Hierarchical conceptualizations perceive consultation to accord
with a low level of control which could mean they are viewed as being less useful than
activities classified as collaboration or user-controlled research.
68
There are compelling accounts of how involvement of users contributes to effective,

successful and meaningful research. For example, Lindenmeyer et al's (2007) account
in the context of diabetes research and Psillidis et al's (1997) account of the role of
participant advisory in the Brest Cancer Prevention Trial. There are also useful
explanations of the circumstances which enabled participation to occur, such as Tozer
and Thornton's (1995) account of older people as research advisors; or Wray's (2003)
account of creating effective user groups. However, it is more problematic to attempt to
explain who exactly contributed what and what impact this had. The danger is that in
attempting to do so the total value of users' contributions is lost through the process of
conceptual de-construction.
There is also the issue of showing what service users can gain from research. Again,
there are good and compelling accounts for the benefits of participation, such as that
presented by Peddie et al (2006) on women's participation in research on in vitro
fertilization treatment. Studies like this help to reveal reasons for participating and
explain the types of therapeutic qualities research can have. Similarly, little has been
done to draw links to other research on why people participate in research (Treloar et al.
2007), including the concept of altruism (Rosenbaum and Wells 2005). If user
involvement in health research is simply judged according to control over research
decisions, or service user impact on the research, the significance of what service users
stand to gain can be overlooked.
Another danger is that conceptualizations fail to account for the potential blurring of
lay-professional roles and responsibilities within research (Beresford 2005b). Tritter and
McCallum (2006) propose that a mosaic is a better metaphor for user involvement as it
“illustrates the relationship between horizontal and vertical accountability and enables
user involvement to be mapped and monitored”. While Tritter and McCallum recognise
that processes of user involvement relate to established social constructions, the focus
remains on the relationships between users and researchers overlooking the contexts
within which these may or may not occur.
69
As it is presently defined, user involvement in research tends to exclude learning from

ongoing collaborative relationships between research organisations and service user
groups, particularly international studies. Overlooking, for example:
 community-university partnerships (such as the study by Buckeridge et al. 2002),
 patient partnership models (for example studies by Raberharisoa 2003; Radda et
al. 2003; Schulz et al. 2001),
 large-scale human/environmental health programmes (for example, Goldstein et
al. 1998; Savitz 2007; Shell and Tarbell 1998), and
 inter-sectoral research partnerships (for example, Fielden et al. 2007; Schensul et
al. 2006; Spoth 2007).
More sophisticated models and descriptions of user involvement are beginning to

emerge within specific areas of health research. For example, in a study by Grocott et
al. (2007) on medical device development the relationship between service users,
researchers and industry partners is made more explicit. Or, for example the work of
Wilkinson et al. (2003) including people with dementia and their carers to promote
partnership and prevent inappropriate hospital admissions. In another study from the
USA (MacQueen and Cates 2005) a framework for prevention science clinical research
has been devised (drawing on advocacy and policy, community participation,
prevention research, acceptability research, and operations and programme
development) to explicitly link service user need to the implementation of results in
public health programmes.
(iv) Ideology
The notion of ideologies is a fourth prominent theme of the research literature.

Beresford and Croft (1996) show that political intentions are not the only reasons for the
emergence of user involvement in research and that social movements have influenced
user-led organisations to become more involved in health research. User-led or user-
controlled research is based on the idea that research can be a way for individuals to
transform their own lives. The conceptual foundations of user-led research view the
70
creation of knowledge as a way of transforming social relations (Beresford 2005a). An

associated form of user-led research is survivor research; developed by people who have
personal experience of mental health problems or cancer to improve their lives and
people like them (Faulkner and Nicholls 2000).
Beresford argues that user involvement in research can be classified according to two
different underlying ideologies which relate to consumerist and democratic models of
participation in policy and practice.
“Ideological differences can be detected between service user research and

mainstream user involvement in research. This reflects the broader ideological
differences underpinning different approaches to participation more generally.
Service user research, with its commitment to equality and change, reflects the
democratic model of participation that has developed from grass roots
movements, while mainstream support for service user involvement in
research has many parallels with managerialist/consumerist approaches to
participation, with their emphasis on seeking the views of service users as
customers for the purposes of market research and information gathering”.
(Beresford 2007, p.306)
As Beresford also argues it is helpful to be explicit about such different approaches to

involvement, not in order to privilege one or the other, but to be clear about their
different origins, intentions and value base (Beresford 2003). This perspective of
ideology is implicitly historical; it connects what might happen in research practice with
particular political and social movements. An overall trend is noticeable: user
involvement has tended to emerge from situations where the emphasis is on change
through research or where the aim is to include marginalized groups in research. In this
context ideology is less about “representation of the imaginary relationship of
individuals to their real conditions of existence” (Althusser 1972) and more about why
and how an individual chooses to create knowledge. More recently authors have drawn
links between research approach and health inequalities (Scambler 2001; Beresford
2007).
“Given that health inequalities is a subject that is centrally concerned with

exclusions and inequalities, it seems helpful to adopt research approaches that
prioritise addressing and overcoming such inequalities and exclusions in their
own philosophy, process and objectives”. (Beresford 2007, p.310)
71
Overall, the research literature indicates that increasingly the research process itself is
being perceived as a way of improving health or social well-being, or gaining
empowerment (Rodwell 1996) through the sharing of experiential knowledge (Kelleher
2001; Popay et al. 1998); and for offering users a way to influence systems that affect
their lives (Beresford 2003). An example from mental health research (Simpson and
House 2002) is provided in the next section and issues relating to these types of
outcomes are discussed later in chapter 7 (Section 7.2).
An ideal sometimes expressed in publications and debates about user involvement in

research is that the ‘voice' of service users should be heard (Baxter et al. 2001, Rose D.
2001). The notion of voice has been influential in conceptualizing interactions between
patients and health care professionals and health service organisations. For example,
Lohan and Coleman (2005) show that in relation to sexual health service development,
professional codes of practice (confidentiality, consent, contact tracing) contribute to the
‘absence of the lay voice'. Voice is often approached as an ethical issue or a question of
overcoming barriers to involvement. Researchers have asked exactly what the ‘user
voice' signifies and to what purpose the ‘voice' will be put (Seymour and Skilbeck
2002). There are also particular practical and methodological issues of research with
seldom heard groups, including:
- the very ill (Abma 2005; Small and Rhodes 2000; Stevens et al. 2003),
- the very old (Hayley et al. 2006),
- children (Simpson 2003),
- vulnerable and marginalized groups (Steel 2005),
- family caregivers for people receiving palliative care (Hudson 2003) or treatment
for advance HIV infection (Yates et al. 1997),
- those who suffer from serious physical or cognitive disability (Hubbard et al.
2003; Kaminsky et al. 2003; Kim et al. 2002),
- those who have suffered physical or psychological trauma (Crighton et al. 2002;
Dyregrov et al. 2000; Griffin et al. 2003; Newman and Kaloupek 2004),
- those who may suffer stigmatization (Lesser et al. 2005), and
- members of disadvantaged or under researched Black and ethnic minority
communities (El Ansari 2005; Berg 1999; Johnson 2002; Vasquez et al. 2006).
72
Ethical issues of involving and paying children (as subjects) in research has received
particular attention (AAPCNAC 2004; Bagley et al. 2007; Dixon-Woods et al. 2006;
Joffe et al. 2006; Kassam-Adams and Newman 2005; Scherer et al. 2007; Wendler et al.
2001). More generally, in relation to volunteering to participate in research it has been
argued that potential influences should be assessed specifically according to the
likelihood that they will control or dominate a potential participant's decision to
participate (Nelson and Merz 2002). I will go on to discuss the notion of the right to
decline to participate later in chapter 6.
Returning to the issue of active and direct user involvement, concerns about the
possible dominance of the strongest voices have been used as an argument against user
involvement and linked to concerns about representativeness (a concept theme explored
in chapter 5). The notion of voice links to discussions about the lay/professional ratio of
research advisory groups and establishing mechanisms to support quality deliberation,
for example using a ‘hot seat' for individuals to speak from or recording comments
outside of a group situation (Davies et al. 2006). This indicates that tensions about
‘equal say' are also important and these are examined later using social theories of
power (Chapter 6).
Debates about opposing ideologies of user involvement and concerns about voice
indicate a need to further understand the values which underpin research practice and
influence how user involvement is played out. Previous research on medial ethics
suggests that there are three possible value positions: rights-based, duty-based and goal-
based (Foster 2001). These positions have been applied to user involvement in the
context of research with people with cancer and those receiving palliative care
(Seymour and Skilbeck 2002). I will summarize these positions briefly here and revisit
the implications of such an understanding later in relation to findings from the present
study (discussed in chapter 8):
 Rights-based - People who hold values that align with a rights-based model will
consider service users as active decision makers who are able to make their own
choices about participation and their own assessment of risks.
73
 Duty-based - People who hold values that align with a duty-based model will
consider duty of care to participants in the research process to override any duty to
extend knowledge and consider the research process as therapeutic to the
participant.
 Goal-based - People who hold values that align with a goal-based model will try to
produce the maximum benefit in terms of the greatest applicability to the greatest
number.
In the research literature notions of ideology and values inform what constitutes best-
practice within service user/researcher relationships. For example the service user-led
organisation Shaping our Lives have produced guidance for access based on values of
inclusion (Beresford and Bransfield 2006). The government have provided guidance on
payment of service users, on the basis of reward and recognition for service users who
are involved (DH 2006b). Work has been done to establish principles and indicators of
successful user involvement in health research (Boote et al. 2006), discussed later in this
chapter. Guidance provided by the support organisation INVOLVE has also influenced
how user involvement in health research is thought of and this is discussed in chapter 5.
3.5 Methods studies have used
It was not an aim of this review to identify whether studies were led by professional
researchers or service users, and it would have been very difficult to establish this
without contacting authors directly. It is possible to say that the majority of studies were
descriptive of the process of working with a small number of service users for the
purposes of a particular research study.
It was often the case that a rationale for who and how service users were recruited was
not provided by authors, although recruitment difficulties were frequently discussed
(Thornton et al. 2003; Browne et al. 2004; Happell and Roper 2007). In both qualitative
and quantitative studies individuals were generally invited to participate on the basis of
74
existing skills or knowledge (Elliot et al. 2002; France 2000; Oakley et al. 1996; Rose et
al. 2004).
None of the research studies identified used randomization or controlled methods (for
the user involvement element of the research). The research literature is characterised
by its predominantly descriptive nature and small scale studies. Nevertheless descriptive
evaluations and studies using qualitative methods (such as those described by Denzin
and Lincoln 2003; and Silverman 2000) provide useful lessons and inspiration for future
research. Furthermore, detailed insights into the experiences of users involved in
research (for example, McClimens 2004) as well as the experiences of clinical
researchers (Abma 2005) can be informative and stimulate changes in research practice
(see Flaskerud and Anderson 1999). Some studies have used reflective evaluative
methods to reveal insights about working with service users (Oliver et al. 2001).
Community focused research studies (for example, Israel et al. 2001; Metzler et al.
2003; Noe et al. 2007; Nolan et al. 2003; Norris et al. 2007; Minkler et al. 2006; Parker
et al. 1998) have tended to use action orientated research methods, as depicted in figure
15. Although action research offers a developed approach to participation (Reason and
Bradbury 2001), the emphasis on community ownership can be problematic in multi-
site or national research studies or where research topics do not relate to any group of
present service users, such as in health promotion (Eakin and Maclean 1992).
Community-wide consultation can be a more appropriate approach for gaining user
perspectives on broader health issues (Contant et al. 2006).
Some more recent studies have evaluated user involvement at the time of occurrence
(Abma 2005, Faulkner 2004) or service users have written reflective accounts whist
being part of a research team (Reeve et al. 2002). In other cases ‘follow up' evaluation
studies of user experiences have been undertaken shortly after involvement (2-4 weeks)
(for example, Hanley et al. 2001, Elliott et al. 2002; Simpson and House 2002).
75
Where attempts have been made to assess the impact of user involvement in health
research, the methodological approaches chosen and specificity of research contexts
have rendered the generalisability of findings very low (see for example, Ross et al.'s
2005 discussion of methodological issues of involvement the context of falls
prevention). However some studies do draw comparative conclusions about user
involvement with different types of service users. For example, Fudge et al. (2007)
show that barriers to engaging older people in stroke research were the same as reported
factors hindering involvement of younger people, suggesting that age, per se, is not a
barrier.
Although there are few studies to compare, there appear to be significant differences in
how user involvement has developed in the context of experimental research, systematic
review and action orientated projects; when perceived in terms of service user roles,
levels of control and ideologies. However, there are inherent difficulties trying to draw
general conclusions about research practices or relationships in any area of health
research. There is a danger of over simplifying the range of professional groups, clinical
activities, research methods, client groups and their diversity of needs and so on that
research relates to (Steel 2005). Using methods of randomization or controlling user
involvement in order to research it carries risks of neglecting the value of dynamic and
context specific contributions.
76
Figure 15: Action research cycle
77
3.6 Gaps in the research evidence
Although involving service users in research is widely thought to be positive and

beneficial, there is limited research evidence for this. There could be several reasons for
this, including how the scope of the field is defined, how knowledge of user
involvement is defined and methodological issues associated with evaluating user
involvement. These issues are discussed in more depth in chapter 7.
With such a limited volume of research studies specifically about user involvement in
health research there is limited opportunity for comparison or meta-analysis between
studies. The development of programmes of research has been hindered because
relatively few focal areas have emerged. This could lead to a lack of sustained research
programmes and failure to follow-up on recommendations for future research from the
limited number of studies that have been undertaken. Where there is a ‘cluster' of
activity, for example on peer interviewing (Alvarez et al. 2006; Elliott et al. 2002, Kuo
et al. 2006), more could be done to apply learning from previous studies to address
recurrent practical and methodological issues.
The relative strength of research that reflects the role of user representatives in bringing
user perspectives into the research process (for example, Oliver et al. 2001; O'Donnell
and Entwistle 2004b) highlights a corresponding deficit in the testing of alternative
models of representation in different research settings with different types of service
users. It may be that the focus on gaining statistically representative views of patients
reflects the traditional subject role of service users in research.
A striking finding was the relative lack of studies on research specifically led by service
users or carers. While research commissioners are now more open to funding user-led
research and consumer and voluntary organisations have funded research led by patient-
researchers, it is surprising that the leadership role of users is not more prominent in the
published research literature. Instead the impression could be gained that user-led
research is somewhat peripheral, indeed absent, from health research. This issue is
explored later in chapter 5 (see section 5.6).
78
Some areas of health research have benefited from knowledge spread through strong
networks such as mental health, disability research and some areas of cancer research.
Clearly research from outside the specific field of research practice and from other
research disciplines, such as research on participation in health care, service
development and social care, informs and influences thinking about user involvement in
health research.
An important conclusion to be drawn is that current definitions and conceptualizations

tend to direct research attention on what is happening within research relationships and
within research studies, rather than on the external situations and underlying concepts
that shape and control such interactions. Three main areas of conceptual tension emerge.
Firstly the view that user involvement in research must be supported by policy makers,
research commissioners and the research community, but at the same time it should not
be ‘top-down' or controlled by professionals. Second that there should be more
transparency in research, particularly around processes of decision making, but at the
same time there are uncertainties about what the underlying intentions and interests are.
Third that service users should have ‘equal' say, but at the same time professional
‘know-how' about research systems means that service users can never gain an equal
footing.
3.7 Personal reflections on the literature
The purpose of including a personal account of the literature here is to show that
particular documents have shaped my understanding and outlook on this topic. The
literature discussed here extends beyond the scope of the structured review presented in
this chapter. The aim here is to identify sources of motivation, reflection and
transformation. The discussion covers 12 documents, discussed in the order which I
became aware of them:
1. Changing the relations of research production (Oliver 1992)
2. Small voices big noises (Baxter et al. 2001)
3. The Bristol Inquiry Report (BRI 2001)
79
4. The Alder Hey Inquiry Report (DH 2001g)

5. Research Governance Framework for Health and Social Care (DH 2001c)
6. Consumers in health research: A review and research agenda (Boote et al.
2002)
7. User involvement: Liberation or regulation? (Beresford 2002)
8. Public participation and collaborative governance (Newman et al. 2004)
9. Patient and public involvement in health: The evidence for policy
implementation (DH 2004)
10. The people's health service? (Barnes 1997)
11. Involving users in mental health services (Simpson and House 2002)
12. What does it mean to involve consumers successfully in NHS research?
(Telford et al. 2004)
1 - Changing the relations of research production (Oliver 1992)
The first paper I will discuss is by Sandy Oliver (Oliver 1992) published in the journal
Disability, Handicap and Society. I read this paper in 2002. The phrase used in the title
of the paper has had a great impact on my thinking. The paper struck me because it
argued that decisions about user involvement in research were not always in the hands
of researchers or service users. It drew my attention to the fact that some of the most
important decisions about relationships between researchers and service users are made
prior to and outside of the boundaries of those relationships. For example decisions
made at a commissioning level could support or prevent relationships from occurring or
characterise relationships in particular ways. Oliver's subsequent review of the progress
of lay involvement in the NHS Research and Development Programme drew attention
to the barriers associated with involving service users in commissioning processes
(Oliver 1996). Mullen (1999) went on to show that a number of approaches had been
used to involve service users in commissioning but there was little research that had
attempted to link influence on commissioning to quality of research outputs. In the
same year, a government report stated that ‘the involvement of service user
representatives in prioritizing research questions can ensure that the questions research
seeks to address are more appropriate to the needs of patients and communities' (DH
1999a). Through
80
subsequent studies, for example Oliver et al. (2001) and O'Donnell and Entwistle
(2004), researchers have continued to ask questions about where decisions on user
involvement are made.
2 - Small voices big noises (Baxter et al. 2001)
In the first year of the Millennium researchers and service users were pushing
government to make change. Across health and social care research people were
arguing that involving service users was a way of overcoming professional bias and
secrecy (Coats 2000) and that it was important to locate research questions in issues that
are important to service users (Tallon et al. 2000a). The organisation INVOLVE (at that
time called Consumers in NHS Research) had commissioned the organisation Folk.US
(www.folkus.org.uk) to undertake a review of the research literature. The review
provided evidence to back the calls for procedures and mechanisms to be put in place by
funders and sponsors of research to ensure that commitment to user involvement was
turned into a reality (Baxter et al. 2001). The review pulled together lessons from other
fields (social care, education, public health/health promotion, community development,
housing/regeneration, agriculture/environment and development overseas). The
resulting document, ‘Small voices big noises' is successful because it captures the main
arguments in an accessible way. It presents a number of lessons, including:
- The importance of the political philosophy and context of research.
- The importance of shared values, mutual respect and trust, and common
language in negotiations between partners about problems and solutions, as
well as the need for clarity in the aims and expectations of research and in
the roles and responsibilities for partners.
- The need for flexibility in research, including the methods used and the
direction of the research.
- The need for flexibility of funding and the time allowed to undertake
participatory projects.
The publication of this document placed increased pressure on research commissioners
to make user involvement a requirement and provide extra finances for it.
81
3 & 4 - The Bristol Inquiry and The Alder Hey Inquiry reports (2001)
‘Learning from Bristol: the report of the Public Inquiry into children's' heart surgery at
the Bristol Royal Infirmary, 1984-95' was published by the Department of Health in
2001 (BRI 2001). Experts on the Inquiry advised that Bristol had a significantly higher
mortality rate for open-heart surgery on children under 1 than that of other centres in
England. Between 1988 and 1994 the mortality rate at Bristol was roughly double than
elsewhere in five out of seven years (p.4).
“Sadly, a system of separate and virtually independent clinical directorates,

combined with a message that problems were not to be brought to the Chief
Executive for discussion and resolution, meant that there was power but no
leadership. The environment was one in which problems were neither
adequately identified nor addressed”. (BRI 2001, p.5)
In the same year the inquiry into the retention of children's body parts for research at
Alder Hey Hospital in Liverpool was published (DH 2001g). Both of these documents
are commonly quoted in relation to the Patient and Public Involvement agenda in the
NHS and policy for user involvement in health and social care research and this is why
I have selected them as the third and fourth documents that have been significant to my
thinking. I became aware of these reports during a project meeting in 2003. I listened to
John Sitzia's description of these incidents with disbelief. In particular I found it
difficult to comprehend how professionals could allow a situation where children's
body parts were routinely retained without consent. These events continue to have
significance in the light of current public debates about whether it is acceptable to have
an opt-out system of organ donation. Events at Alder Hey and Bristol were critical for
opening up debates that led to the involvement of patients in defining and monitoring
the quality and safety of healthcare (Coulter 2002).
“The public are entitled to expect that means exist for them to become
involved in the planning, organisation and delivery of healthcare”. (BRI
2001, p.18)
The Bristol Inquiry outlined nine principles to inform future policy about involving the
public and patients in the NHS (BRI 2001, p.19) (included as Appendix 5). The Alder
82
Hey Inquiry contributed to changes in research policy, including the next document I
wish to discuss.
5 - Research Governance Framework for Health and Social Care (DH 2001c)
The fifth document selected is the Department of Health Research Governance

Framework for Health and Social Care (DH 2001c). Research governance is part of
government strategy to re-establish public confidence and regulate health and health
research. Although several policy documents in health and social care recommend that
researchers involve users as active participants in the research process (NHSE 1998b;
NHSE 1998c) the Research Governance Framework for Health and Social Care
consolidated policy across health and social care research. The framework states that:
“All research undertaken under the Secretary of State for Health … is

required to involve participants or their representatives wherever possible in
the design, conduct, analysis and reporting of research” (DH 2001c).
The significance of this requirement should not be underestimated and neither should
the wording. The framework has provided many UK researchers with rationale to
involve service users in research and to encourage other researchers to fulfil their legal
duty to consult patients, carers and members of the public. It has also placed a
responsibility on the research community to judge what the terms ‘participants or their
representatives' and involvement ‘wherever possible' mean (this clause was added in
the final report it did not appear in the first draft of the framework in 2000). Whether or
not I agree or disagree with the precise wording of this phrase or the practices it intends
to encourage, the existence of this requirement in policy has been important to my
thinking. I have cited it in nearly every paper, report and presentation I have made about
user involvement in research and so have many other researchers and service users. It is
key to the formation of user involvement in health research and to the existence of this
study.
83
6 - Consumers in health research: A review and research agenda (Boote et al.

2002)
Boote et al's (2002) review paper was significant to my thinking for several reasons and
it is the sixth document selected but possibly the most relevant publication to the topic
of this study. The authors review current research in the topic area and outline
directions for future research. Later in the discussion chapter I make use of their work
to assess the contribution of this study (see chapter 8, section 8.4). Here I will outline
the influence Boote et al's paper had on my views about the topic. Prior to the study I
was aware that developments in participative research had long preceded policymakers'
and commissioners' interests in user involvement. There was also a literature on Action
Research that I had some knowledge of because of my earlier work on Action Research
projects. Boote et al. (2002) drew together and presented these trends to show that the
use of social research methods had led to greater recognition of difference and the
importance of personal experience in understanding social processes. The paper helped
me to perceive the complexities of the terminology in this field. There were terms such
as customer, consumer, client, patient, public, carer, as well as others. The authors
explained some of the conceptual problems of attaching meaning to terms that did not
account for the complexities of personal identity. I felt ignorant for not realizing that I
was using the term ‘service user' in a very uncritical way. I had not thought about
people who had never used health services or that people might not want to perceive
themselves as having any connection to a system of health. I thought about being
classified as a service user – this term seemed to position service use as the most
important aspect of a person's character. The term does not duly emphasise the diverse
range of people, needs, and experiences of the people that it includes and there is a
danger that these differences are overlooked or simplified.
7 - User involvement: Liberation or regulation? (Beresford 2002)
My seventh selection, Peter Beresford's paper ‘User involvement in health research and
evaluation: liberation or regulation?' (Beresford 2002) was published in the journal
Social Policy & Society. Beresford argued that there were two distinct ideologies of
84
involvement at play in contemporary health research: consumerist and democratic. In a

subsequent paper “User involvement in research: Exploring the challenges” Beresford
(2003) expanded this idea and argued that top-down politically/managerially controlled
user involvement had subsumed a bottom-up lay ideology of democratic empowerment.
“The two approaches do not necessarily sit comfortably together. One is

managerialist and instrumental in purpose, without any commitment to the
redistribution of power or control; the other is liberatory, with a commitment
to personal and political empowerment”. (Beresford 2003, p.41)
I felt a sense of historicism in user involvement that I had not appreciated. My interest
and research in this area was located at the end of a series of long and arduous struggles
by lobby groups and consumer health organisations to gain some say over health
services and research. What right or insight did I, a ‘mainstream researcher', have to be
studying user involvement? This question is revisited in the discussion chapter. Whilst I
have great respect for Peter Beresford's work, I began to feel positioned by it: my
research did not flow from an ideology of democratic empowerment so it must therefore
be part of an agenda for political control. I remain uncomfortable with this duality. In
practice, approaches are not underpinned by purely consumerist or democratic values as
Beresford's (2002) work suggests: a combination of values is enacted through
judgements made through the research, some of which were made before a study
begins. Beresford's views did not resonate with my experiences and observations of
user-research relationships and the multiple contingencies, corresponding concepts,
theories and values that shaped what did or didn't occur at project level. I do not deny
the different historical trends that have constructed ideas about user involvement in
health research, and I have identified such trends in chapter 2, but I do refute the idea
that individual studies should be classified in this way. A few years later I read Laclau
and Mouffe (1985) assertion that ideology would be better regarded as a set of practices
rather than a state of mind. Although I agree with this assertion, it was Beresford's work
that provoked my interest in exploring how notions of user involvement in research are
constructed and reinforced.
85
8 - Public participation and collaborative governance (Newman et al. 2004)
My eighth selection is somewhat outside the literature on user involvement. I found this
paper through reading about consumerism in the NHS. It was informative and it led me
towards seeing user involvement as a set of contingencies rather than ideology. The
review by Newman et al. (2004) provided a substantial account of trends towards
collaborative governance and some of the mechanisms which had been developed; such
as deliberative forums, user panels, youth forums and area based committees. A key
influence on this agenda was the decentralisation of power to local governance
structures, including awarding greater spending and commissioning authority to local
health bodies, and strengthening public and private partnerships (DETR 1998, DH
2002c). Newman et al. (2004) talked about attempts to shift the focus of governance
towards new forms of co-production with other agencies and with citizens themselves
through partnerships and community involvement. The paper highlighted the constraints
on the political opportunity structures created by the enhanced policy focus on public
participation, and the consequent limits to collaborative governance. Authors have
traced themes of participation across the breadth of public services (Croft and Beresford
1996). A paper by Barnes and Walker (1996) explains how expectations for consumers
to become more involved in making decisions and choices about their care reinforced
rights established by the Patients Charter (DH 1995). There are substantial literatures on
participation in political decision-making process which given more time I could have
made greater use of in this study. ‘People & Participation' a publication by Involve
(2005) provides a good overview of methods of participation, some of which have been
used to look at health issues:
 Appreciative Inquiry
 Citizen's Juries
 Citizens' Panels
 Community Empowerment Networks
 Consensus Building/Dialogue
 Consensus Conference
 Deliberative Mapping
 Deliberative Polling
86
 Democs ‘Deliberative Meetings of Citizens'

 Electronic processes
 Future Search Conference
 Participatory Appraisal
 Participatory Strategic Planning
 Planning for Real
 Open Space Technology
 User Panels
 Youth Empowerment Initiatives
9 - Patient and public involvement in health: The evidence for policy

implementation (DH 2004)
Despite the increase in consultation and involvement activities in the NHS few reports
have looked at the evidence to underpin these developments. The report entitled ‘Patient
and Public Involvement in Health: The evidence for Policy Implementation' (DH 2004)
summarised the results of the Health Partnership Research Programme. The report
looked at 12 research projects about user involvement in health across the UK. Of the
twelve projects, six focussed on patient and carer involvement in decisions about their
own treatment and care, four addressed public involvement in service planning and
delivery and two were principally concerned with education and training issues. One of
the funding conditions was that projects should involve patients and the public in their
design and execution. In one project service users were involved in the review of a
protocol and draft report, in others users were invited to participate in steering or
advisory groups. Although the report finds strong evidence that patient involvement in
health care decisions improves patient satisfaction and is rewarding for professionals,
the impact of involving users in the process of the 12 projects was not scrutinized to the
same degree. The report concluded that initiatives to involve patients and the public in
health need to be a shared corporate responsibility and that planning for public
involvement should encompass planning specific initiatives and planning the broader
cycle of learning, change and review. The findings of the report influenced my thinking
because it showed that NHS services could involve service users and that research could
87
involve service users, but that such initiatives could be disparate and difficult to draw
generic conclusions. Involvement took on different characteristics in the different
contexts it was employed within.
10 – The people’s health service? (Barnes 1997)
The issue of representation emerges in nearly every paper about user involvement in
research, and so it should. The issues of defining who can, or should, represent the
views of service users and under what circumstances were a concern of researchers,
policy makers, service users; everyone who had something to say about the topic. Boote
et al. (2002) provided me with a good outline of some of the issues. It seemed to me
that representation in user involvement was a complex bundle of issues and it was to do
with the way that researchers thought about representation. My discussions with
colleagues outside of research contexts told me that representation was not such a
concern where service users were involved in education or service development
initiatives. The paper by Marian Barnes was useful because it highlights the relationship
between different identities of health service users and mechanisms to enhance
participation. It expanded my thinking about what representation meant: it could refer to
someone who was elected by means of a democratic process; it could mean a
representative sample or someone who is typical of others who share similar
experiences. Barnes argued:
“The model of the statistically representative sample is not always the

appropriate one to adopt when developing opportunities for participation”.
(Barnes 1997, p.6)
I have purposely pursued issues of representation in this study (see chapter 5).
11 - Involving users in mental health services (Simpson and House 2002)
The eleventh document selected is a review of the evidence from comparative studies
on the effects of involving users in the delivery and evaluation of mental health
services. This review synthesized five randomised controlled trials and seven other
comparative studies. Half of the studies considered involving users in managing cases.
88
Simpson and House (2002) found that involving users as employees of mental health
services led to clients having greater satisfaction with personal circumstances and less
hospitalisation; and that providers of services who had been trained by users had more
positive attitudes toward users. Clients reported being less satisfied with services when
interviewed by users. The review concludes that users can be involved as employees,
trainers, or researchers without detrimental effect. Involving users with severe mental
disorders in the delivery and evaluation of services is feasible.
Although the findings are significant the review influenced my thinking because of the
way that the authors positioned user involvement. Simpson and House (2002) argued
that mechanisms for involving users or carers could be thought of as health
technologies. In a later paper Simpson and House (2003) suggested that user
involvement could be thought of as interventions. They argued that this perspective
encourages a rigorous approach to identifying aims for user involvement, then choosing
and implementing an approach designed to achieve those aims, and monitoring to see if
the aims have been achieved. I felt uncomfortable with this positioning. My
experiences of working with service users on the multi-method review had shown me
that there were problems with applying understandings of research methods and
concepts to user involvement in health research. For example, views about rigour and
generalisability of findings seemed to set goals for user involvement that were
unattainable. User involvement does not lend itself to randomised or comparative
research methods and the impact of service user views and opinions on the research
process can not be easily identified. Evaluations of participative research studies (from
the 1970's onwards) had shown the difficulty of trying to attribute any outcomes of the
research, or for healthcare practice, directly to the involvement of different members of
the community within the research. I felt that approaching user involvement in health
research as a complex intervention or as a research method was incapable of explaining
why user involvement had become a site for the accumulation of knowledge and power.
In their review of user involvement in the health field, Crawford et al. (2002) showed
that relatively little attention has been paid to impact and outcomes compared with
process. Although attention is beginning to turn to ways of evaluating user involvement
89
(discussed later in Section 7.2) it could be that some members of the research
community are waiting to see the evidence of impact before investing time and
resources in establishing such relationships.
12 – What does it mean to involve consumers successfully in NHS research?

(Telford et al. 2004)
Meanwhile in the field of health research, opposing arguments were being put forward
about the challenges researchers face to involve members of the public. These focused
on the idea that different researchers had different abilities to involve service users
(Beresford 2003). I had questions about why some researchers were adamant about user
involvement whilst others were not. Was it perhaps related to an individual researcher's
perception of research or their role as a researcher? It was a question that had occupied
Telford et al. (2001) and their survey of health researchers revealed that many
researchers were unsure if they had involved service users or not in their research:
suggesting that underreporting is an issue of this topic area. In their assessment of
working with user-researchers Trivedi and Wykes (2002) had shown that little was
known about the personal values or understandings that people might hold about user
involvement.
My twelfth and final selection, a paper by Telford et al. (2004), reported a study which
used consensus methods to establish quality criteria for the processes of user
involvement in health research. The work asserted the importance of evaluating research
processes and not simply the outcomes of user involvement. An expert workshop was
used to generate potential principles and indicators and these were tested using a
consensus building technique with 96 participants drawn from health, social care,
universities and consumer organisations. Eight principles were developed and consensus
was reached on at least one clear and valid indicator by which to measure each principle
(see Table 1). Telford et al's (2004) principles of quality in user involvement are useful
but they do not relate specifically to different research contexts or approaches to
involvement. These authors acknowledge this limitation and suggest further research is
90
needed to establish how transferable, appropriate and useful the principles are (Boote et
al. 2006).
Table 1: Principles of consumer involvement (Telford et al. 2004)
1 The roles of consumers are agreed between the researchers and consumers involved
in the research
2 Researchers budget appropriately for the costs of consumer involvement in research
3 Researchers respect the differing skills, knowledge and experience of consumers
4 Consumers are offered training and personal support, to enable them to be involved
in research
5 Researchers ensure that they have the necessary skills to involve consumers in the
research process
6 Consumers are involved in decisions about how participants are both recruited and
kept informed about the progress of the research
7 Consumer involvement is described in research reports
8 Research findings are available to consumers, in formats and in language they can
easily understand
It is also interesting to examine principles (A,B,C,E,G) where consensus could not be

reached:
- A: The research will lead to benefits for consumers in terms identified by the consumers
themselves;
- B: Consumers are involved in every stage of the research, from identifying the research agenda
through to sharing the research findings;
- C: Consumers' expectations of being involved in the research are made clear to the researcher;
- E: Consumers have the opportunity to engage in research in the manner and at the level they wish,
opting out of being involved in research at any time;
- G: Consumers are from sections of society and walks of life that are appropriate to the research.
The rejection of principle A relates to the concept of ‘end-users' and could be

attributable to uncertainties over who research is for. Research commissioners and
research organisations have tended to view members of service organisations (managers
and the executive rather than NHS staff) or policy makers to be the end users of
research. Attempts to close the research-practice gap (Rafferty and Traynor 1999) have
focused on making research more relevant and timely for these groups, failing to meet
the generally broader agendas of service users (Tallon et al. 2000a). However the
literature and policy on user involvement claim that it could improve the relevance and
91
appropriateness of research (DH 2001c; DH 2004). This is a rather sweeping view of

what user involvement can achieve; different groups of the public, service providers and
policy makers often hold different views about what is relevant or appropriate (Rappert
1999). Furthermore I felt that within each of these perspectives there would be
differences of opinion about the purpose and outcomes of different types of research.
Principle B is likely to have been rejected because of the practical issues of involving
people in every stage of the research and because of the lack of consensus about
whether involvement is always appropriate across different types of health research
(particularly secondary research). Principle E is likely to have been rejected because
this appears to be covered by research ethics. Yet this masks a major ethical issue that
needs addressing: whether service user involvement should be subject to or excluded
from the same ethical principles that are normally awarded to participants in research:
informed consent, confidentiality, right to withdrawal and so on. Rejection of principle
G is a particular curiosity considering the prevalence of issues about appropriate
representation in this field. The resulting principles for involvement emphasise quality
of process over quality outcomes, which is not necessarily a bad thing, but it does
reflect the conceptual difficulty of making general statements about the diverse and
unpredictable outcomes of user involvement (see section 7.2). For me this paper
emphasises both the strengths and masking ability of a consensus approach to the
development of knowledge. (This is an issue that I will look at in more depth in the first
part of the discussion chapter).
Another paper, by Nolan et al. (2003) in the journal Quality in Ageing entitled ‘The
Aldre Vast Sjuharad model revisited' also emphasises the importance of the process of
user involvement. It states that researchers using participative methods have to negotiate
a balance between developing valid generalisable knowledge and benefiting the
community that is being researched. I read these papers in close succession and they
raised questions for me about how best to research user involvement in health research.
This issue is the focus of the next chapter.
92
Summary
 This chapter presents a structured review of the published research literature on

user involvement in health research. Overall, relatively little research on
involvement in health research exists – in so far as it is represented by published
accounts of research in peer reviewed journals. The volume of research activity
seems to be disproportionately low given the degree of political interest in this
topic and commissioners requirements for user involvement in health research. In
addition to low volume, the scale and some of the quality of the research base is
also limited in terms of the possibility of drawing confirmed conclusions. With
such a limited volume of research studies there is limited opportunity for
comparison.
 The UK makes a disproportionately large contribution to the international

literature on ‘user involvement in health research' and it seems likely that this is
because of its existence as a distinct agenda in research policy in this country and
cultural differences in terminology. The small but significant level of activity
within Europe, Canada and Australia suggests that stronger alliances could be
made.
 The paucity of robust evidence about user involvement in health research could
indicate that practice in this area has been developed through researcher initiative
and networking for information about user involvement. Links to research from
other fields and research disciplines has usefully informed user involvement in
health research. Interest in user involvement has predominantly focused on issues
of practice and technique which may reflect the importance researchers have
placed on the value of this type of knowledge.
 The literature is characterised by a focus on particular themes and issues. The field
is notable for its problematic terminology; making definition and classification
difficult. The relative strength of research that reflects the role of user
representatives in bringing user perspectives into the research process highlights a
corresponding deficit in research on the representation of diversity.
93
 Role, hierarchy and ideology are also important themes of the research literature.
Current conceptualizations of user involvement tend to underplay the significance
of contextual factors in influencing whether user involvement does or does not
take place.
 While research commissioners are now more open to funding user-led research
and consumer and voluntary organisations have funded research led by patient-
researchers, it is surprising that the leadership role of users is not accounted for in
the published research literature. Authorship of research publications suggest that
user involvement tends to developed and discussed by professional researchers.
 Three main conceptual tensions emerge. Firstly the view that user involvement in
research must be supported by policy makers, research commissioners and the
research community, but at the same time it should not be ‘top-down' or
controlled by professionals. Second that there should be more transparency in
research, particularly around processes of decision making, but at the same time
there are uncertainties about what the underlying intentions and interests are.
Third that service users should have ‘equal' say, but at the same time professional
‘know-how' about research systems means that service users can never gain an
equal footing.
 A personal account of the literature reveals how my views towards user

involvement in health research changed over time as a consequence of reading
what others had written. I became more accepting of different epistemological
stances as I became more knowledgeable of them. It also shows that changes in
my understanding were rarely attributable to reading individual documents and
some documents affected my views even though I had not read them. Experience
taught me to go back to documents and revisit concepts and ideas that researchers
had raised but not felt to be significant in their own conclusions.
94
Chapter 4: Methods
Chapter 4: A method for researching user involvement
4.1 Introduction
The previous chapter showed that identifying the scope and strength of the knowledge
base for user involvement in health research is complicated by uncertainties about
terminology and definition. There are also questions about what type of knowledge can
and should be produced about user involvement in health research. This chapter
describes the position taken to construct knowledge in this particular study.
Developing an epistemological stance and approach was not a straightforward process.

It took time and went on along side gaining information about user involvement in
health research, as the following reflective notes show.
{Trigger (App3/Date:19.10.05): Interview with a senior nurse academic in

Australia (Participant 7)}
The struggle to frame my own thinking about user involvement and how it
might be researched is incredibly frustrating. I feel I am using a logical and
developmental way of knowing: working from one experience to the next,
using what I find as a platform for establishing my next line of enquiry. I
am reluctant to hinder my exploration by placing methodological limits
around the study. Writing down some of my assumptions about what it is
that I am trying to do with the research might help. Am I trying to motivate
change? - Perhaps, but not necessarily. What if I come to the conclusion
that user involvement is a wholly negative thing? – I am open to that
possibility but I don’t think it is likely because user involvement defines
such a complexity of ideas and interactions; some will be positive, some
negative. The value (impact) of involvement must be linked to the context
and methods of research.
{Trigger (App3/Date:21.10.05): Email from a supervisor about an interview with a

senior nurse academic in Australia}
Stretching my own thinking and undertaking research go hand in hand. It
is not my intention to construct a grand theory of user involvement or to
approach ‘it’ as a unified whole, which can be proved or disproved. The
field and my view of it are filled with developmental ideas that change
daily. Perhaps user involvement should be part of all research about
humans; but in practice user involvement can be contrived and
unnecessary, costly or a political game. My framework for thinking is this -
95
Chapter 4: Methods
that how I am thinking will itself develop, so I am trying to flow with it and
try to reflect on it as I go. I feel all research will invariably involve elements
of observation, theorization, interpretation and change, irrespective of the
framework within it is undertaken. These elements are easier for me to talk
about as part of what I am doing.
4.2 Approach
The aim of this study was to explore understandings of user involvement in health
research, to add depth and richness to current knowledge. I do not aim to present a
grand unifying theory of user involvement in health research or to discover underlying
elementary components of it. The approach to knowledge construction in this study is
based on the philosophical idea that any knowledge can only be partial, because:
(i) The meaning of knowledge is not fixed or agreed;
(ii) All knowledge is contingent and historically located;
(iii) Particular approaches to knowledge creation emphasise different dimensions of
objects and therefore no knowledge is complete;
(iv) The creation of knowledge is embedded in subjective experience.
In view of these philosophical issues, the position taken in this study is as follows:
(i) The interpretation/claims made are based in a view that knowledges are
transitional and multiple.
(ii) The interpretation/claims made are recognised as being contingent and historically
located.
(iii) An approach to knowledge creation has been taken (interpretation) that
emphasises that the object of user involvement in health research has multiple
dimensions.
(iv) A reflective stance is used in the study to surface subjectivity: experience,
observation, thinking, communication and so on.
The position is informed by philosophical perspectives of interpretation, discourse and

reflection. The use of these perspectives is illustrated by figure 16. The remainder of
96
Chapter 4: Methods
this section explains the specific perspectives drawn from each of these domains to
inform the approach.
{Trigger (App3/Date:15.10.06): Supervision discussion about what knowledge is}

This study is not based within the theoretical limitations of the practice I
seek to understand - user involvement in health research. It does not
involve service users in a way that might be recognized as corresponding
with a conceptualization of service user involvement. This study uses an
approach that is constructed from philosophical perspectives of
interpretation, discourse and reflectivity. Adopting such a stance allows me
to gain a position from which to explore user involvement in health
research from the perspective of what knowledge is, how it is being created
and the effects it can have.
Figure 16: The approach to knowledge construction
Knowledges are partial

Multiple interpretations coexist
Interpretation
User
involvement
in health
research
Reflection
Discourse
Knowledges are
Knowledges are contingent subjectively
and historically located created
Knowledge is linked to
power
Notes:
(i) The central octagon represents the multifaceted ‘object' of user involvement in health research.
97
Chapter 4: Methods
(ii) The shaded triangles represent the scope of the theories applied in ‘shedding light' in to the
object of user involvement in health research and on to the social context which constructs it as
an object of knowledge.
98
Chapter 4: Methods
Interpretation
User involvement in health research is characterised by conflicting perspectives about

how user involvement in health research should be undertaken and the outcomes it
should aim to achieve (discussed further in Section 7.2). Building shared understandings
of it may help to overcome disconnection of views. A key element of an interpretive
approach is that meaning is considered to be constructed. As Seedman suggests:
“Nothing means anything on its own. Meaning comes not from seeing or even
observation alone, for there is no ‘alone’ of this sort. Neither is meaning lying
around in nature waiting to be scooped up by the senses; rather it is
constructed. ‘Constructed’ in this context, means produced in acts of
interpretations” (Seedman 1991)
As human beings we can unconsciously form ideas about the world around us. David
Novitz calls such day-to-day personal interpretations of the world elaborative
interpretation (Novitz 2000, p.5). This form of interpretation is limited because it does
not allow re-identification of the object across time and conversational connections to
be made. In the context of user involvement elaborative interpretation can lead to
incompatible understandings and multiplicity of meanings. An alternative form of
interpretation, elucidatory interpretation, is a self conscious quest for understanding
(Novitz 2000, p.5). It requires self-awareness and an understanding of a lack of
comprehension and a desire to reach an understanding.
“Elucidatory interpretations have a very definite epistemic structure since they

require that interpreters be aware of their own ignorance, which they then
wittingly seek to dispel by imaginatively formulating hypotheses that function
as tentative solutions to some question”. (Novitz, 2000 p.7)
This study is based in conscious intent to construct understandings about user

involvement in research. According to Novitz successful interpretations enable us,
either as individuals or collectively, to negotiate the world. A confirmed interpretation
brings intelligibility and promotes reliable or adequate understanding. Who ‘us' is and
which types of worlds (social, geographical, professional and so on) we aim to
negotiate are equally important questions. On an individual level each of us constructs
beliefs based on elaborative and elucidatory interpretations of the world; this stock of
99
Chapter 4: Methods
interpretations then influences the way in which a person goes on to interpret the world.
This philosophical argument presents a problem for, although knowledge creation can
be removed from interpretive acts it can not be dissociated from subjectivity. Any
possibility of one true reality is beyond our awareness because it is filtered by our
individual understandings of the world (Madill et al. 2000).
The question of knowledge creation falls to a question of the utility of agreed

knowledges, as opposed to confirmations of truth (Hammersley 1992). That is, although
we do not have direct access to a true reality, interpretations of personal and alternate
realities, as well as shared and socially constructed realities are useful in society (Crotty
1988). The degree to which knowledges are agreed and consciously upheld is variable
between different groups of individuals and changes over time, for example under the
influence of paradigms and programmes of research. Shared realities can become a site
for the accumulation of knowledge and power as particular values become reinforced by
practices and infrastructures.
I did not aim to position participant's interpretations of the world as reality or to assume
that ‘reality consists of the meanings in a person's lived experience' (Omery and Mack
1995). This is described as a dauntingly radical claim as “it entails that there is no
distinction between ‘how it seems' and ‘how it really is”. From this position there can
be no illusions, no mistakes, no fallacies, no misconceptions, and no errors (Paley
2005). Neither did I aim to devise a description or concept of user involvement in
health research based on commonalities (an aim of Husserlian phenomenology). Early
on in the study I aimed to develop understandings of a full range of meanings from
which understanding might be drawn from, or constructed on: an aim more aligned
with Heideggarian phenomenology (Heidegger 1962). Yet, the effect of aggregating a
number of accounts would be merely to collate narratives and provide ‘a general
framework of categories in which all uniqueness, all lived experience, completely
disappears' (Paley 2005).
A grounded theory approach, as advocated by Glaser would have centred the study on
explaining the actor's problem (Guba and Lincoln 2005). If it had been possible to
100
Chapter 4: Methods
identify a group of actors the analysis would be steeped in a very radical way right from
the start (Alvesson and Sköldberg 2000 p.35). In this study rather than representing
what user involvement in health research means, participant's differing accounts were
approached as being generated from available social, cultural and political discourses,
discussed in the next section.
Discourse
Perspectives of discourse provide the second set of framing knowledges I am using to

develop understandings of user involvement in health research. I have principally drawn
on perspectives put forward by Michel Foucault. My introduction to Foucault's work
came through a colleague at the National Nursing Research Unit (Dr Janette Bennett) in
early 2004. Until this point I had little knowledge of Foucault's work and I perceived it
to be anti-establishment and anti-science. I also thought that Foucauldian Discourse
Analysis involved studying the minutia of interview transcripts; not something that I
thought would be useful for exploring the political and social dimensions of user
involvement in research. I read some introductory texts about Foucault's life and work
(summarized in Appendix 4). Janette helped me to explore how Foucauldian
perspectives might be applied to develop understanding of user involvement in health
research.
{Trigger (App3/Date:12.09.07): Supervision discussion of the method I am using}

Early on in the study my approach was informed by phenomenology and
ethnography. I perceived myself to be situated within the field of practice
that I was interested in studying and I hoped my situatedness would offer
me a deeper and more connected experience of user involvement in health
research. However my intentions for the approach presented me with
problems because it was not possible to easily identify when I was in the
field and when I was not. If I followed either method (phenomenology or
ethnography) the majority of what I was observing, experiencing and
generating would not necessarily fit. Where I was collecting data from
could not easily be defined according to concepts, such as a particular
community or according to particular practices or activities. The
phenomenological approach that I thought of using in the exploratory
phase only enabled me to access participants’ perspectives of the issues. I
felt that the contingencies that had played a part in the origin of user
101
Chapter 4: Methods
involvement extended the object of user involvement beyond the

perception or practice of health research. The concept of discourse
permitted an alternative route to explore user involvement in health
research.
As work on the multi-method review (Smith et al. 2005b) was nearing completion and I
had gained an insight into the types of issues within and surrounding user involvement
in research. I was better able to see how concepts of discourse might inform my work.
The first thing that made sense was Foucault's view that objects of knowledge are not
fixed, they are formed over time and:
“between institutions, economic and social processes, behavioural patterns,

systems of norms, techniques, types of classifications, modes of
characterization” (Foucault 2006a, p.49).
Foucault's words not only appealed to me because his approach would allow for
mobility and multiplicity of user involvement in health research, but they also allowed
me to explore the ‘ingredients of thought' that create user involvement in research. In
one respect, discourse can be perceived as being a regular set of linguistics or according
to Foucault ‘that which is sayable' (Foucault 2006a). In this study this could be the
language that has enabled user involvement in health research to be brought into being.
To Foucault, discourses are not simply a set of words or linguistic systems that bring
meaning to social practices, they are:
“A collection of statements, which provide a language for communicating

about a particular phenomenon and enable it to be brought into being in
the social world” (Foucault 2006a, p.31)
According to Foucault discursive formations are made of statements (propositions,

beliefs, theories and doctrines) and have institutional, philosophical and scientific
components. For example, health research has an institutional component in the form of
research departments, units and centres; philosophical elements, including paradigms
and methods; and scientific components, such as methodologies and measures. Each
component holds information – discourses – which facilitate meaning to user
involvement.
102
Chapter 4: Methods
{Trigger (App3/Date:09.09.06): Discussion with a Supervisor about what a PhD

requires}
Until now I have lacked the confidence to suggest that my attendance at
meetings, conferences and discussions are a valuable source of data about
user involvement in research. Yet, these sources are providing me with
relevant information and influencing my thinking. These ‘sites of human
interaction’ are the places, beyond the literature, where meaning of user
involvement in health research is being actively constructed.
Heartfield (1996) argues that the apparent unity of discursive formations is due to the
imposition of rules (concepts) that govern the formation of statements in a particular
time. Particular discursive formations can act as strategic controlling points for judging
the authority of statements. According to Foucault, we should not simply accept the
apparent unities of what is “already-said” (p.27), we must:
“tear away from them their virtual self-evidence, and to free the problems that
they pose; to recognise that they are not the tranquil locus on the basis of
which other questions (concerning their structure, coherence, systematicity,
transformations) may be posed, but that they themselves pose a whole cluster
of questions (What are they? How can they be defined or limited? What
distinct types of laws can they obey? What articulation are they capable of?
What sub- groups can they give rise to? What specific phenomena do they
reveal in the field of discourse?)”(Foucault 2006a, p.28)
In The Archaeology of Knowledge Foucault (2006a) outlines the constituent elements of

discursive formations (object, concept, enunciative modality, and strategy) and the
interplay of statements, divergence and dependencies. In the present study the notion of
a system of formation provides a useful framework for locating core components of the
analysis, explained later in this chapter (section 4.6). The analysis I have undertaken
draws on techniques that correspond with Foucauldian Discourse Analysis (Powers
2007). Discourse analysis has been widely applied to explore contemporary social and
political phenomena (Bacchi 2000, Ball 1993, Fairclough 1995, Gavey 1997, Parker
1992, Wood and Kroger 2000). The term has been used to mean linguistic units
(conversations, arguments, speeches and so on) or texts (documents, transcripts,
policies and articles), as well as the wider social constructs that shape attitudes and
behaviour (Potter and Wetherell 1987). In the social sciences approaches including
Frame Analysis (cognitive structures guiding perception and interpretation), Discursive
103
Chapter 4: Methods
Psychology (shared patterns of meaning), Foucauldian Discourse Analysis,

Poststructural Discourse Theory (political discourse analysis), and Critical Discourse
Analysis (challenging dominant discourse) have been previously described (Suurmond
2005). In some approaches to discourse analysis the emphasis is on understanding how
individuals recreate social practices, hence discourses are seen as:
“a diverse potentiality of acts which can be realised in different ways through

participants’ production of different interpretations in different social
contexts” (Gilbert & Mulkay, 1984 p.9).
Foucault took the apparent unity of discursive formations (such as psychopathology,

medicine or political economy) as a starting point for his inquiries. The study makes
use of perspectives drawn from Foucauldian archaeology to explore the manifestation
and formation of user involvement in health research and to trace the shifting historical
frameworks within which the topic emerged. I also draw on the idea of Foucauldian
genealogy, which can be thought of as the engagement and critical committed probing
of the roots of societal practices. I explain how I make use of these perspectives later in
this chapter. Elements of the study could be perceived as aligning with a critical
approach discourse analysis (Fairclough 1995) in that through this study there is a
possibility of creating knowledge to advance the way that user involvement is
perceived. There is potential for the research to affect perceptions and actions through
connecting with participants (Power 2007). At the same time there are four important
points that distinguish the work from an explicitly action orientated stance. These are as
follows:
1. The study does not aim to be ‘user involvement' in the sense that the approach I
have taken does not correspond with current conceptualizations of user involvement
in health research that are established in the research literature.
2. The study does not aim to champion user involvement in health research. However,
I do acknowledge that through building understandings of user involvement in
health research there is potential for the work to influence the way participants and
readers think about the issues and to reassert existing notions of user involvement.
104
Chapter 4: Methods
3. The study does not aim to convey a view of user involvement from the perspective
of any particular group of people. Thus, the study does not put forward the case of
researchers or service users or any other group, though it may well provide leverage
for change.
4. The study does not aim to inform those who may have participated within it. Indeed
it may be inaccessible to policy makers, research commissioners, health researchers
and service users because of the format, language, perspectives and examples it
uses.
This is not to say that service users' and researchers personal experiences of
involvement do not form an essential part of the understandings that are presented here.
It is an issue that has been at the forefront of decisions about the approach I have taken.
{Trigger (App3/Date:19.09.05): MPhil/PhD transfer viva}

During the study various colleagues have recommended placing greater
emphasis on the contributions of service users to this study. I have listened
to their advice and understood their concerns about showing ‘a service user
perspective’. Yet I have made a conscious decision not to do this. It has
certainly not meant that I have not engaged with different groups
throughout the study, including members of voluntary organisations and
consumer groups who provided extensive views and experiences about
research.
Reflectivity
The third set of framing knowledges that are applied in this study are perspectives of
reflectivity. According to the philosophy of interpretation theoretical assumptions,
language and pre-understanding constitute major determinants of interpretations, and
my role as the constructing subject can not be disconnected from the research. It has
been argued that rigour in research demands that researchers must raise their awareness
of the evaluative and political judgements and choices at every staged of the research
process (Gewirtz and Cribb 2006). Specific forms of reflexivity applied in this study are
discussed in the next section. A reflective element of the approach enabled an
immediate critical consciousness of the complex relationship between processes of
105
Chapter 4: Methods
knowledge production and the various contexts of such processes (Finlay 2002).
Reflexive methodology has meant directing attention to the way different kinds of
linguistic, social, political and theoretical elements are woven together in the
development of the interpretation, particularly the way in which the study data are
constructed, analysed and reported.
4.3 Reflexive methodology
Writings on reflectivity have not been transparent in terms of the difficulties,

practicalities and methods of the reflexive process (Mauthner and Doucet 2003). Here, I
explain how my opinions towards reflectivity changed over the study from perceiving it
as a fairly insubstantial and supplementary activity to using reflexive methodology to
build the interpretation. My starting point for reflectivity in this study began with
putting into practice personal ideas about reflectivity. Later I drew on the work of
previous authors, and more specifically the work of Mats Alvesson, Professor of
Business Administration at the University of Lund and Kaj Sköldberg and Professor of
Business Administration at Stockholm University.
Personal reflexivity
The early stages of this study were characterised by the application of a series of
methods; testing and trying to establish fit. The exploratory phase made use of
phenomenology, applying a technique devised by Lindseth and Norberg in their study
of the lay voice (2004). I was uncomfortable with this approach because user
involvement involves more than people's perceptions of it. I sought an approach which
would permit analysis of the broader historical and contextual issues. My supervisors
suggested that I should also maintain a reflective diary - to keep note of my reflections
on these types of issues. Keeping a diary seemed like a fairly straightforward and simple
activity to do alongside my research work. I felt that I should try not to be biased by my
preconceived views of research and reflection would help to surface these biases. I had
read about ‘bracketing' in Husserlian phenomenology and the literature on reflexivity
106
Chapter 4: Methods
presented similar ideas. For example Cutliffe (2000) argues that researchers need to
acknowledge their prior knowledge to bring this knowledge into the open in order to
discuss how it affected the theory development. I had also read that ‘putting aside one's
assumptions' was not an easy task (Paley 2005). Like Wilson (2002) I felt that it would
be possible to identify some of my assumptions but many more would remain beyond
my consciousness and it would be these assumptions that are likely to exert the greatest
influence. I was uncertain how I might ‘switch off' my preconceptions and, perhaps
more importantly, whether I would want to.
Whilst in search of a method, I began reading Alfred Schutz's views on the formation of
knowledge. Schutz argued that:
“Part of understanding the formulation of knowledge is acknowledging that the

world and the researcher and the researched are connected: the researcher
influences the research design and the questions that are asked; the researcher
contributes to co-constructed meanings because they are devised through
discussion and deliberation between the researcher and the participants; each
aspect of the research will be perceived and interpreted by the researcher and
this will influence the research process itself in different ways” (Schutz 1972).
This made sense to me. I began to perceive my position within the research as being
more than part of the research community that I was interested in understanding. A year
into the study my approach to reflexivity expanded to an attempt to explain how my
previous experiences as a researcher would influence the questions I was posing and the
way I might analyse the findings (material now incorporated into chapter 2 of this
thesis). I had been influenced by Finlay's (2002) views about reflexivity being a form of
self-reflection: “an analytical approach towards capturing more subjective and inter-
subjective dimensions of research”. In principle I agreed that I should record my
assumptions, expectations, behaviour/reactions and responses, however even at this
stage in the study I perceived reflectivity to be something that was predominantly about
my interaction with research participants and what I might unintentionally place an
emphasis on in the analysis.
107
Chapter 4: Methods
Critical reflexivity
Despite my limited view of reflexivity being personal reflexivity, other authors had
provided me with more connected definitions. For example, Calas and Smircich (1992)
describe reflexivity a process “that constantly assesses the relationship between
knowledge and ways of doing knowledge”. To Steier the core of reflection (reflexivity)
consists of an interest in the way we conduct ourselves socially while also constructing
objects (‘out there') in our research.
“Reflection means thinking about the conditions for what one is doing,
investigating the way in which the theoretical, cultural and political context of
individual and intellectual involvement affects interaction with whatever is
being researched, often in ways difficult to become conscious of. When we
reflect, we try to ponder upon the premises for our thoughts, our observations
and our use of language”. (Steier 1991)
These descriptions of reflexivity suggested to me that I should look more towards other
factors (other than me) that were influencing the study. As the study developed my
approach to reflexivity grew to include an attempt to make clear the decisions taken at
each stage of the research. I kept notes about lines of enquiry that were not pursued and
the reasons for this. I tried to bear in mind the factors, barriers and opportunities that
had led me to certain decisions. This form of reflexivity is described as critical
reflexivity by Knock and Harrington (1998). It aims to understand the political and
social constructions that shape the research study. In particular it involves making
explicit any ethical or political questions encountered in the research process and the
impact this may have had. It acknowledges that knowledge is active and entrenched in
socio-political context.
I perceived critical reflexivity to be a way of explaining the limitations placed on the

study – what stood in my way of doing what I wanted to do. As I read more I
discovered that other researchers have used reflexivity to explore researcher-participant
relationships. Critical reflexivity could also help to reveal the influence of the
observations or actions of the researcher within the social system they are observing. It
might include critically reflecting on how any theory being formulated affects the
108
Chapter 4: Methods
behaviour of the individuals or systems the theory aims to explain. For example,
feminist and emancipatory approaches emphasize the power differentials within the
various stages of the research process (Mauthner and Doucet 2003). According to this
view of reflexivity my presence within any particular research community affects the
community. Hence, observations are not independent of what is being observed. This
corresponded with my experiences of interviewing participants in this study and the
different responses this provoked, for example, the traditional role of interviewer-
interviewee was often purposely deconstructed or circumvented by researcher
participants (see discussion chapter).
Contextual reflexivity
According to (Bola 1998) I should also critically reflect on the theoretical framework
being applied. This was harder for me to perceive. Cutliffe (2003) suggests that
reflexivity is about making the research process more transparent – by drawing attention
to the researcher's a priori knowledge, values, beliefs; empathy related
phenomena/processes; the researcher's tacit knowledge; and the ‘magic' in method – to
enable the reader to assess the way the findings were generated from the data. Maranhão
(1991) suggests reflexivity involves “paying attention to how one thinks about
thinking”. What these ideas meant was not very clear to me until I began reading
literature on the philosophy of interpretation (in February 2006). In philosophical
hermeneutics, guided by the writings of Gadamer (1989), this form of reflectivity has
been called epistemological reflexivity. Until this point I intended to explain the
limitations of the study according to the particular research methodology I was using. It
had not occurred to me to delve deeper into the epistemological limitations, to ask
questions such as, how has the research question defined and limited what can be found,
and how could the research question be investigated differently? It was not until I had
drawn together a basic table of different epistemological stances (shown in Appendix 1)
that I was able to reflect upon the assumptions I had already made and the implications
of such assumptions for the research and its findings.
109
Chapter 4: Methods
It is difficult for me to comprehend the significance of the concepts not being used, but
researching the concepts that I have applied helps to explain the way the study has been
shaped. For example, I took it for granted that previous authors would provide me with
‘confirmed' definitions of power. As Clegg (1989) points out, major texts about power
are themselves major forms of power in the academic market place.
“Definitions and conceptualizations of terms so central to our experience of

social reality, as are those of ‘power’, are themselves enormously powerful”.
(Clegg 1989, p.22)
Contextual reflectivity helps me to at acknowledge that I have awarded power to

particular theories of power, most notably Clegg's model of power and to the
perspectives that underpin this model (e.g. Hobbes, Machiavelli and Foucault),
explained later in this chapter.
Self-critical reflection
A significant change in my use of reflectivity came as a result the phase of research I

undertook in Australia (September-October 2005, see Appendix 8 for further details).
During this time I maintained day-by-day field notes, which I typed directly into a
laptop computer. I chose to do this so that I could provide my supervisors with an
account of my activities. The record contained details of who I had met, what they had
told me, some of my immediate thoughts, and points of interest about each encounter.
Some months later (April 2006), I was transcribing the interview tapes. I returned to the
day-by-day field notes to trigger my memory and clarify details of the interviews. I
noticed that some of the notes I had made were now slightly out of kilter with my
current thinking. I was faced with a dilemma, should I change the entries or leave them
as they were. I decided I should leave the field notes as they were because they captured
my immediate thinking at the time. I created some new notes about my new thoughts.
I felt it was important to show how and why my views had changed so I created a list of
‘trigger events' that had led me to where I was now. My view of reflexivity again began
110
Chapter 4: Methods
to change. It was now about writing reflexive notes, leaving them for a while, and
coming back to reflect on them a few weeks or months later.
{Trigger (App3/Date:21.07.06): A conversation with a Supervisor}

My views have changed they will again. Where will this lead? What is
there after this? Have I initiated a way of thinking that is just over-
stimulated reflexivity? Half of what I think I write, maybe less. It seems as
though I have thought it, before I know it. Is introspection a way of
researching? Is my agenda overly personal – emancipatory, therapeutic?
Am I researching me? My thinking and reflections are themselves subject to
change. For example, an observation I may make and think little of on any
one day, may take on an entirely new significance at a later date. Placing
trust in this method is incredibly personal because it means putting trust in
one’s own personal cognitive processes and allowing time for current
uncertainties to become clearer. I am more aware of how I am shaping the
study at every move: it is tied to me.
However my use of reflexivity was sliding into self-conscious cynicism.
{Trigger (App3/Date:28.08.06): Transcribing interview data}

I have such a personal view of my relationship with this study. I feel like I
am one of two mirrors. I am pointed at another mirror, which is the study.
From my position all I can see is me, reflected in the study. And, all that is
reflected in me is the study. I can see glimpses of the background world.
Behind me extend reflections of previous images of the study. Behind the
study are previous reflections of me. I can see where the study has come
from and how my understandings have changed. But, I am in a reflection
trap. I am frustrated with looking at the study and at me. It feels overly
reflective and out of control. I am mentally going around in circles,
scrutinizing every decision that I have made and what I will do next.
I perceived reflexivity to be a threat to the study. I felt it prevented me from seeing

direction. Giddens (1976) described this form of reflexivity as personal reflexivity or
self-awareness. Colbourne and Sque (2004) have called it self-inspection. I did not
recognise that I had allowed this form of reflection to dominate the study. It was
limiting me but I could not perceive a release from this situation. I thought this was how
the remainder of the study would be. I resented reflexivity. I perceived it as a personal
burden to me. I felt that my struggle to manage reflexivity was a personal weakness. I
did not talk about it because I did not want to appear self indulgent, or egocentric. The
release came through dialogue and understanding that an alternative form of reflexivity
111
Chapter 4: Methods
would be far more helpful to both me and the study. At this point I had not fully
appreciated the possibilities of reflexivity as a method. I had perceived it firstly in terms
of validity (making my biases explicit), and then as a way of accounting for my actions
(showing what I had done), then as a process of self reflection.
Later, I would discover that several writers have expressed concerns about the risks of
self-reflective isolationism, self-absorption and impotent texts (as discussed by
Alvesson and Sköldberg 2000, p.246). A concern expressed by Baker (1990)
particularly struck a chord with my own experiences; namely, that excessive
textual/rhetorical awareness “could lead researchers into a house of mirrors, a self-
sufficient metalanguage which will isolate them from the rest of the world”. To counter
this tendency, according to Baker, it is necessary to widen our rhetorical and textual
focus to include the researcher's recognition of her- or himself as a rhetorician who is
part of a social and political context. By writing ourselves into this and recognizing that
we are engaged in persuasion within a social and consequently a political context, we
can avoid narcissistic self-centredness (Baker 1990). This offered me a broader
reflective stance: a position that enabled me to recognize that although I was part of the
study, I was not the only person or thing influencing it.
Interpersonal reflexivity
As time went on I became more conscious of the relationships that were central to the
history of the study and my research work. For example, I was able to initiate informal
and formal discussions about particular aspects of the interpretation to make a more
balanced use of reflexivity. Regular dialogue with my supervisors and with three
colleagues (Dr Catherine Evans, Sheila Donovan, Louise Jolly) who had formed an
informal PhD support group supported by Dr Janette Bennett, were major influences.
Through these relationships I considered ethical and political questions encountered
about the study. The comments on the reflective note in Appendix 7 illustrate one way
in which these peers became involved in reflexivity in this study. For example, how
participants in the study had responded to being interviewed and issues about
confidentiality when collecting observational data. Discussing these types of issues
112
Chapter 4: Methods
involved talking about tensions and uncertainties. I was conscious that this could be
perceived as ‘giving advice' or helping to ‘off load' and potentially making the study
look flawed or weak. The supervisory relationship and peer-support therefore became
important interpersonal places for exploring other interpersonal contexts of the research.
I found it reassuring to hear supervisor's and peer's experiences about how practical and
ethical issues had been resolved in other studies.
Providing information and support about reflexivity, and in particular interpersonal

reflexivity, through supervision and peer-support would appear to be a logical strategy
for assuring research governance in postgraduate studies.
4.4 Principles for a reflexive methodology
Mats Alvesson and Kaj Sköldberg's Reflective Interpretation emphasises interactions

between different forms of reflexivity. Although I had been working with a range of
types of reflexive data this method provided me with a more structured understanding
of different types of processes involved in the construction of data and the
interpretations I was making.
“Systematic reflection on several different levels can endow the interpretation

with a quality that makes empirical research of value. Reflections can in the
context of empirical research be defined as the interpretation of interpretation
and the launching of critical self-exploration of one’s own interpretations of
empirical material (including its construction)”. (Alvesson and Sköldberg
2000, p.6)
Alvesson and Sköldberg's views echoed the philosophical arguments that I had
grappled with in the early stages of the research, both in relation to my interest in how
power relates to knowledge operates in research and in how I could build knowledge in
this topic area. They suggested that the focus on data collection and processing in most
qualitative methodological theories is unreflective and that researchers should
recognise that:
“there is no such thing as unmediated data or facts; these are always the
results of interpretation. Yet the interpretation does not take place in a neutral,
113
Chapter 4: Methods
apolitical, ideology-free space. Nor is an autonomous, value-free researcher

responsible for it” (Alvesson and Sköldberg 2000, p.9)
Making use of Alvesson and Sköldberg's work I derived the following principles for a
reflective methodology.
Box 1: Principles for reflexive method (after Alvesson and Sköldberg 2000).
1. Systematics and techniques in research procedures: The study should follow some
well-reasoned logic in interacting with the empirical material, and use rigorous
techniques (for example grounded theory, ethnomethodology and inductive
ethnography) for processing the data.
2. Clarification of the primacy of interpretation: The study is fundamentally an

interpretive activity. It is driven by an interpreter – who interacts with and
contemplates other interpreters. Method cannot be disengaged from theory and other
elements of pre-understanding, since assumptions and notions in some sense
determine interpretations and representations of the object of study. Paying attention
to symbolic and embedded meaning is therefore an important form of reflection.
3. Awareness of political-ideological character of research: The study is a social

phenomena embedded in a political and ethical context. What is explored, and how it
is explored, can hardly avoid either supporting (reproducing) or challenging existing
social conditions. Different social interests will be favoured or disfavoured depending
on the questions that are asked (or not asked), and on how reality is represented and
interpreted. The interpretations and the theoretical assumptions on which these are
based are not neutral but are part of, and help to construct, political and ideological
conditions.
4. Reflection in relation to the problem of representation and authority. Texts affect one
another, they become split or fragmented. In this way both the author's claim to
authority and the texts' claim to reproduce some extrinsic reality are equally
undermined: the researching subject and the researched object are both called into
question.
Alvesson and Sköldberg emphasise culture, language, selective perception, and

subjective forms of cognition and ideology that permeate research activity. Reflective
Interpretation utilizes the researcher's assumptions, interpretations and interactions with
empirical material as data. It thereby offers a framework to ‘put research in perspective'
and confront favoured lines of interpretation, rather than pretending the ‘biases' of
subjectivity do not exist or that they can be ‘put aside' in the process of the research.
114
Chapter 4: Methods
Alvesson and Sköldberg identify four currents of methodology and philosophy of

science (2000, p.7), which they regard as important sources of inspiration for the
method:
- empirically orientated currents (in particular, grounded theory),
- hermeneutics,
- critical theory, and
- postmodernism.
These four orientations indicate the reflective areas in which the practice of reflective
research can be positioned. Alvesson and Sköldberg recommend not getting stuck in a
particular fixed methodological position. Instead, they advocate maintaining a certain
distance from the various methodological positions that inform a reflexive stance.
“Thus in reflective empirical research the centre of gravity is shifted from the
handling of empirical material towards, as far as possible, a consideration of
the perceptual, cognitive, theoretical, linguistic, (inter)textual, political and
cultural circumstances that form the backdrop to – as well as impregnate –
the interpretations”. (Alvesson and Sköldberg 2000, p.6)
Reflective Interpretation is built around a multi-layered, flexible structure of

interpretation and reflection in which the systematic interplay of reflective areas is
central. These authors illustrate this using the following diagram (Figure 17).
Reflexive methodology allowed me to work independently of any single methodology

and to question weaknesses inherent in the developmental mode of thought that I was
embracing. Alvesson and Sköldberg's approach provided an overall frame of reference
as well as more concrete ideas for incorporating reflexive elements in the research.
115
Chapter 4: Methods
Figure 17: Interaction between different levels of interpretation

(Alvesson and Sköldberg 2000, p.255)
Level of interpretation Possible reflective themes
Empirical material/
construction of data Multiplicity in interpretation and
consideration of pluralism in looking
at different aspects.
Surprise potential in the empirical
material. Favouring of certain
interpretations. Interpretive
repertoire.
Interpretation
Negation of data-confirming
interpretation (radically different
view of reality); consideration of why
certain interpretations dominate; the
presentation of counter-images; the
discussion of winners and losers as
the result of a particular
interpretation.
Critical interpretation
Self-reflection on elements of
dominance in the researcher's line(s) of
interpretation; the identification of
critical reflection on potentially
problematic forms of authority;
openness to other representations,
interpretations and conclusions than
Self-critical and those favoured.
linguistic reflection
{Trigger (App3/Date:27.02.07): Discussion with a colleague about reflective

methodology}
Reflective Interpretation offers me possibilities to develop understanding of
the challenges and possible implications for user involvement in health
research policy and practice at an individual, group (e.g. gender, ethnicity),
organisational and social level. Reflective Interpretation is also consistent
with a critical approach as it assumes the possibility of creating knowledge
to advance the way that user involvement is perceived and undertaken and
116
Chapter 4: Methods
it acknowledges that there is potential for the research to affect perceptions

and actions through connecting with participants.
The following figure draws together elements of reflective methodology applied in this
study (Figure 18). The figure was developed from the literature and by drawing on the
‘levels' of reflection identified by Alvesson and Sköldberg, illustrated by the previous
figure. Reflective themes that emerged through this study are described in the following
section.
Figure 18: Elements of reflective methodology
Self-critical
reflexivity
Interpersonal
reflexivity
Contextual
reflexivity
Critical
reflexivity
Reflective themes
117
Chapter 4: Methods
4.5 Study data
The study data was collected to support an analysis of user involvement in health
research in terms of its meanings, concepts, expressions and experiences associated with
it. Given that I was approaching user involvement in health research as being
discursively constructed a wide range of information could inform such a perspective.
For example documents that relate to historical events, such as policy decisions or
strategic changes in research organisations, systems or structures and so on. Potential
data could also include information about conceptual events, such as the emergence of
definitions or concepts; or experiences and perceptions of different groups of people in
different geographical and social contexts.
 Documentary sources
 Empirical data: collected from cites of discussion and human interaction in
relation to user involvement in health research (January 2004-December
2007)
 Reflective data: collected throughout the study (January 2003-March 2008).
In view of the exploratory nature of the study it has been appropriate to take up
opportunities to collect data as they occur and to focus attention on research practices.
Data collection has not been restricted to the places where user involvement is being put
into practice, for example within research studies thought of as ‘user involvement
projects'. It has been important to collect data from places where user involvement in
research is resisted or contested, for example to explore criticisms made explicit by the
research literature and to surface more subtle forms of resistance within the regularities
of human action (systems), and the rules and resources that human agents can draw on
to produce and reproduce (structures) (Giddens 1977).
{Trigger (App3/Date:09.09.06): Discussion with a supervisor about what a PhD

requires}
In an exploration of an object such as this there is little need for induction (I
use the term ‘induction’ here to mean initiating situations of data creation,
rather than an inductive research approach). A wealth of material is
available through observation. However if, the social world is the source of
data, there are questions to be asked about how I have selected, recorded
118
Chapter 4: Methods
and processed the information. It is important to be clear about which

information I have chosen to inform the study.
A discursive approach opens up extensive possibilities to collect data in the social world
which can not always be predicted or planned.
{Trigger (App3/Date:23.09.06): A conversation with a supervisor about research

methods}
I perceive the everyday research world to be my source of field data.
Generally, I am interested in the formation, use and consolidation of
research practices within this world. I am myself part of this world. Much
of the study data comes from my individual reflections of engaging with
this world. In this approach it is acknowledged that insights can be
serendipitous, found in conversation or discovered through observations.
Neither is it possible to observe everything that is relevant to the study questions, or to

record everything that can be observed. As Heidegger (1962) said, “the world might be
full but only a partial account can be made”. I argue that the data in any research study
are historical and partial and, that in any research endeavour it is not possible to fully
uncover the partiality of the data or to provide full account of what is absent. These
issues are revisited in more detail in the discussion chapter (Chapter 8). Three main
types of data inform the thesis: documents, empirical, reflective. These data are
interdependent and co-existent. For example, some authors of academic papers have
also participated in interviews, and the empirical data has provided stimulus for the
creation of reflexive data, and so on. The interconnectivity of information was a point
that Foucault raised in his critique of the material individualization of the book and the
œuvre (authorship) (Foucault 2006a, p.25). However, research governance requirements
and the need to show how the arguments of the thesis are supported by the data mean
that it is necessary to approach the data as being comprised of discrete pieces of
information.
Documentary sources
A full list of documents which are used to inform the thesis is provided in the
bibliography at the end of this document. These include policy documents, journal
119
Chapter 4: Methods
papers and articles, research reports, organisational/consumer literature and ‘grey'

literature such as PhD theses and conference proceedings identified through structured
searches of electronic databases, citation in other publications and through the study
through research networks, conferences, participants and colleagues. In assessing the
documentary data that informs this study, it is important to consider the ‘archive' that
has been available to me. Previous reviews undertaken by Baxter et al. (2001) and
Boote et al. (2002), for example, have contributed to constructing histories of user
involvement in health research. Such documents are themselves archives to other
references and commentaries on the formulation of knowledge in this field.
On one level these sources can be critiqued on the basis of authenticity, bias, distance,
and dependence (Alvesson and Sköldberg 2000, p79). From a Foucauldian perspective
it is also important to question what does not and what can not appear in the archive.
There are omissions from the documentary record, which could be thought of as
‘history's dark matter', for example, events that were not recorded or information that
was lost or eradicated. Griffin defines an event as a ‘historically singular happening that
takes place in a particular time and place and sequentially unfolds or develops through
time' (Griffin 1992, p.414). In this way history can be perceived as a multiple
overlapping series of legitimate versus excluded histories; and it is therefore important
to question the systems which establish statements as events, and the way in which
statements are “institutionalized, received, used, re-used, combined together” (Foucault
2006a, p.129) and so on. Furthermore, the archive can not include information where
language was not available for things to be said. Foucault suggests that history should
not be perceived simply as what has been said:
“It is based on the principle that everything is never said; in relation to what
might have been stated in a natural language (langue), in relation to the
unlimited combination of linguistic elements, statements (however numerous
they might be) are always in deficit; on the basis of grammar and of the wealth
of vocabulary available at a given period, there are, in total, relatively few
things that are said” (Foucault 2006a, p 134).
Another issue, as Foucault argues, is that modes of thinking have covert and implicit
rules governing the material restriction of knowledge. These could be thought of as the
120
Chapter 4: Methods
underlying structures which form the context of the individual thinker. We can never be
fully aware of how they shape or constrain the way that we think, and as Abraham
points out, this applies not only to those who have created history but to those who
judge events as being significant (Abrams 1982). In summary, there are practical and
physical limitations to the documentary data but there are also conceptual limitations to
how the topic can be represented by documents including what has and can be
documented.
Empirical data
There are close links between the topic of user involvement in health research and
Patient and Public Involvement (PPI) in service development and user/client
involvement in professional education. In light of these close links, in the early stages
of the study I began to explore the relationship between user involvement in health
research and the political agenda for Patient and Public Involvement in the NHS. To
inform my thinking I accessed UK web sites covering issues to do with user
involvement in health research and public involvement in health services, these were:
INVOLVE (www.invo.org.uk), Health Voice Network (www.healthvoice.org.uk). This
type of information, about issues and trends, what was being said, provided an
important starting point for how I went on to perceive and collect the empirical data.
The majority of messages being posted at that time were from health service researchers
and Patient and Public Involvement Leads (staff with a specific remit to support and
promote health service development through the involvement of patients and the
public). At this developmental stage in the study I was interested in finding out whether
different professional's (e.g. NHS researchers, education research, clinical research)
understandings about the meaning and purpose of user involvement might result in
different interpretations and actions. This is an interesting question and I spent some
time exploring the links (Smith et al. 2005c), however I chose to focus the study on
exploring user involvement in health research. I directed my attention to sites where
knowledges of user involvement in health research are being formed. These sites are not
defined as physical localities but they can have a geographical component in the form of
places (e.g. conference halls, seminar rooms) or institutions (e.g. research society
121
Chapter 4: Methods
meetings). Nor are these sites defined according to the views or actions of particular
people (a participant sample) but they do have a human component in the form of
speech acts, behaviour and social interactions.
Elements of data collection have been ethnographic. In part, data collection involved
‘living' among the people who the study is about, for example learning the language of
user involvement and participating in everyday life (Hammersley 1992). These
activities involved cultivating close relationships with individuals who can provide
specific information on aspects of user involvement in health research (Boyle 1994). It
was possible to gain access to informants, using them to gain access to yet more
informants in a chaining process. The use of ethnographic data gathering techniques
(field notes and interview recordings) has been supported by the use of mixed media
and information technology to log and record the data. However this study displays
notable points of departure from an ethnographic method. Namely:
- In ethnography it is usual to study a ‘community' (Boyle 1994). However,
delimiting a definable community in relation to user involvement in health
research is problematic because the boundaries of the topic area are diffuse and
intertwined with different areas of professional practice (such as health care,
management and education). It would also be misleading to suggest that user
involvement in health research belongs to a community that is defined by any type
of professional practice. It would also overlook the influence of policy makers,
commissioners, groups of society (patients, community groups, public).
Furthermore, to treat all of these different players as members of a community
fails to account for identities, views and experiences of the individual people
involved.
- The view of objective detachment in some forms of ethnography is not possible or
desirable in a complete sense. It is unclear what exactly the researcher should
attempt to remain detached from – the community members, the data that is being
collected, the outcomes or impact of the research.
- Data analysis and theory development generally come at the end of an
ethnographic research study. In the present study methods of reflexivity and
122
Chapter 4: Methods
analysis were applied as continuous iterative processes of data assimilation and

interpretation.
- Ethnographers may seek validation of induced theories by going back to members
of the culture for their reaction. In the present study, validation has been ongoing
and through talking about emergent findings in a range of ways; including through
the supervisory relationship, discussions with colleagues and peers, conference
presentations and publication.
The forms of empirical data used to inform the arguments I am making are outlined
in Table 2 and overall numbers are provided in Table 3.
Table 2: Sources of empirical data used in the study
An organised programme of presentations and talks relating to a

particular theme. Either a one-day event or run over several days.
Conference/ event
Data could include: conference programme, points covered by key
note speakers or during parallel sessions etc.
A formally arranged gathering structured around a presentation

followed by a group discussion of a specified topic. Data could
Seminar
include: information from an agenda, topics or key points,
questions from the seminar participants etc.
A facilitated interactive session focusing on sharing information

and learning, usually as part of a larger conference or event. Data
Workshop
could include: workshop materials, learning outcomes, points or
issues raised etc.
A meeting of two or more people at an agreed time with an

Scheduled discussion agenda. Data could include: views on topics raised, membership,
attendance, group dynamics etc.
An agreed one-to-one discussion structured around a predefined

topic guide. Data could include: views, perceptions, issues,
Interview
challenges as perceived by the interviewee, the nature of the
conversational interaction etc.
An accredited course or training programme requiring

Training event formal registration. Data could include: curricula, course
materials, methods of teaching, assignments or assessments
etc.
Announcements or events that are not generally recorded in the
published literature, for example a commissioning call or launch of
Other significant event
a database or website. Data could include: email notification,
newsletters or letter etc.
123
Chapter 4: Methods
Table 3: Overview of empirical data
2004 2005 2006 2007
Initiated - Initiated - Initiated - Initiated -

Conference/ Participated 2 Participated 3 Participated 4 Participated 3
event Observed 3 Observed 2 Observed 2 Observed -
Total 5 Total 5 Total 6 Total 3
Initiated - Initiated - Initiated 1 Initiated 1

Participated 2 Participated 3 Participated 2 Participated 4
Seminar Observed - Observed - Observed - Observed -
Initiated 1 Initiated - Initiated - Initiated 1

Participated - Participated 2 Participated 2 Participated 4
Workshop Observed - Observed - Observed - Observed -
Initiated - Initiated 10 Initiated 20 Initiated 15

Scheduled Participated 12 Participated 15 Participated 42 Participated 12
discussion Observed 11 Observed 13 Observed 7 Observed 12
Initiated - Initiated 16 Initiated - Initiated -

Participated - Participated - Participated - Participated -
Interview Observed - Observed - Observed - Observed -
Initiated - Initiated - Initiated - Initiated -

Participated 2 Participated - Participated - Participated 1
Training Observed - Observed - Observed - Observed -
Notes:
 Detail information about the data is shown in Appendix 3: Study data log.
 The informality of telephone conversations and email communication means that
although this type of information exchange was important, it is not included as data in
the study or in the study data log (with the exception of telephone interviews which are
classified here as interviews).
Participants
The perspective of participation that I am using in this study differs from formally
recruiting a statistically representative sample of subjects. I chose to interact with
people through the study in different ways (or not at all), which can be perceived as
124
Chapter 4: Methods
corresponding with different forms of participation (indirect/direct) (one-off/ongoing)

(formal/informal) or levels of participation (including non-participation).
My judgements about the forms of data that could best inform the study were based on
the questions driving the study and this affected the types of participation and
participants I engaged with. I felt strongly that the personal characteristics and
subjective experiences of the people directly involved and interested in user
involvement could not be overlooked. At the same time my decision to take a broad
approach to the issues, had practical and methodological implications for the forms and
levels of participation I could make use of.
{Trigger (App3/Date:13.09.06): Discussion with a supervisor about service users

in the study}
Initially I thought that my intention not to interview a sample of service
users was because of the theoretical issues of defining who a group of
service users are and on what basis to draw a representative sample. Two
years after the study began I have a better understanding of the reasons
behind my resistance. Again the literature on philosophy of interpretation
has been a great help to me. It has helped me to see that an interpretation is
only one small part of forming a wider understanding. This interpretation
focuses on how the object of ‘user involvement in research’ is discursively
constructed and used. This is not the same as constructing an interpretation
of participants’ subjective experiences of user involvement in health
research. However the use of a Foucauldian approach and reflexive
methodologies permits me to explore subjectivity and subjective experience
as one dimension of user involvement in health research. An interpretation
of user involvement in health research that was based in the subjective
experiences of service users would not be any less or more valid
interpretation. It would be created from a unique set of subjectivities and
would result in an alternative interpretation to the one I present here.
The approach did not suit or require the use of statistical representation to create
generalised findings about the views of all researchers, nor any other group of people
(Hammersley 1992). The approach did not take participants to be representative of all
health researchers or service users. I have not used the views of participants in a
phenomenological sense to create consensus about the meaning of user involvement in
research. Nor did I aim to give any particular group of participants ‘voice' (a theme of
125
Chapter 4: Methods
the research literature discussed in chapter 3) or to offer a sense of empowerment (see

them of ideology, section 3.4). Rather, I have engaged with individuals and groups of
people in different ways to explore particular perspectives and issues.
According to the reflective element of the approach, the way I made use of myself in
the research work had a significant influence on the data that was collected from
participants. In this particular study where I was located physically, what I was able to
observe, and who I was able to interact with were the main limitations. As a
consequence the empirical data has largely been drawn from the public domain,
including information that is available through open debates, publications, meetings and
conferences. Some data was collected using arranged methods (interviews, scheduled
meetings and debates). The vast majority of data was gathered by observation of
‘everyday' social events (Becker 1968) within the world of health research. As
previously stated in the introductory chapter, the extent of this world was taken in a
broad sense to include theory, debate, policy, practice, and perception, as well as
historical perspectives.
I kept notes about the data I was collecting. This included: date/time, locality, who was
speaking; and other types of information where it seemed relevant, such as observations
about group dynamics, the tone of discussions, any tensions or points of agreement.
This type of information was important for connecting what was being said with
context, discussed later in this chapter.
In circumstances where I formally approached individuals to give their views on the

topic it was necessary for me to gain written consent. In more ambiguous contexts of
participation there were ethical issues of instrumentality and manipulation. For example,
on several occasions I overheard conversations on a tube train that provided information
to me about the way research is perceived in society. Private information such as this is
not classified as data in this study, however it is not possible to say that hearing such
conversations has not influenced my thoughts or those of participants in the study.
126
Chapter 4: Methods
Data collection in Australia
Given that the study sought to gain an understanding of the political and social
influences on user involvement in health research an international phase of research was
undertaken in Melbourne Australia. The multi-method review (Smith et al. 2005b) had
revealed little information about how user involvement has been played out in different
countries and only a few research papers originating from northern European countries,
Canada, the United States of America, Australia and New Zealand could be identified
(see section 3.3). An existing collaboration between Kings College London and Monash
University was a strategic opportunity to explore the issues in relation to a different
health and research system.
Prior to the visit I spent time pursuing possible links by returning to the literature I had
collected during the multi-method review and searching it for contact names or
organisations in Australia (Griffiths et al. 2003; Clinton and Hazelton 2002). I
undertook internet searches to identify potential people to meet or interview and
sources of information. Professor Margaret O'Connor at Monash University supported
communication with academic researchers and consumer research organisations in
Victoria. I invited all researchers in the departments of nursing, midwifery and health
service research at Monash University to participate in the study. Twelve members of
staff agreed to be interviewed and other staff participated in discussion groups (28
participants, some of whom had given interviews). I devised a topic guide for the
interviews, which focused on participants' experiences of undertaking healthcare
research in Australia, their views of the challenges and priorities for healthcare research
and opinions about the role of patients and the public in research (see Appendix 9).
Interview participants were provided with written information about the study prior to
agreeing to participate. They were informed that their participation was voluntary and
that they could withdraw at any time. Written consent was obtained from everyone
participating in the interviews. The interviews were tape recorded and transcribed.
Participant names and job roles were removed to provide a degree of confidentiality.
127
Chapter 4: Methods
Table 4: Interview participants in Australia
Reference Organisation Role

Participant 1 Senior lecturer
Participant 2 Senior lecturer
Participant 3 Postgraduate student
Participant 4 Senior researcher
Participant 5 Senior academic
Participant 8
Academic research Senior researcher
Participant 12 Senior lecturer/course co-ordinator
Participant 14 Former senior academic
Participant 15 Senior lecturer midwifery
Participant 6 Carers' organisation Policy development officer
Participant 13 Consumer health organisation Policy coordinator
Participant 10 National nursing institution Research and development manager
Reflective data
In the first year of this study I collected reflective data in the form of handwritten notes
but I was uncertain about how I might use the notes to inform the thesis. I did not see
my early findings and thoughts as a position from which I would be able to show a
process of knowledge development. I was more interested in finding answers than in
ways of finding out. I tried to keep a reflective journal but it felt awkward and an
inefficient use of my time. It worked better for me to make reflective field notes on the
materials I had at hand at the time they were made – in the back pages of conference
brochures, on the blank overleaf of agendas, penned in the margins of papers, or made
on post-it notes inserted into books (as illustrated by the figure 19). I have kept all of
this data together to support recall of what has been said, by whom and to identify
triggers for reflectivity.
128
Chapter 4: Methods
Figure 19: Example of field notes
At the beginning of the second year of the study I also began typing passages of text
directly into Microsoft Word and dating these entries. This allowed me to comment on
the data and my previous interpretations of it. I could also trace different issues,
emergent findings, levels of interpretation and forms of reflexivity, in a way that would
be far more cumbersome using handwritten reflective notes or ‘in margin notes'. I could
also use my ‘e-notes' to collate electronic texts from other sources - email
correspondence, minutes of supervision meetings, diary entries and so on - that recorded
emergent thoughts and reflections issues at particular points in time.
Alvesson and Sköldberg's reflexive methodology provided suggestions for reflective

elements and themes. However, I would have liked more information about how to
structure, generate and limit reflexive data. As Gerwitz and Cribb suggest:
“There is no obvious limit to how far researchers can unpack, reflect upon,
analyse and defend the value judgements implicit in their work; and if this
process of reflection were taken to extremes, it seems that all research would
129
Chapter 4: Methods
collapse into a process of value analysis at the expense of the substantive

issues that are the focus of the research” (Gerwitz and Cribb 2006, p.).
Towards the end of the study it was easier for me to perceive themes (below) within the
reflective data. I took stock of what I had tended to reflect on by reading the reflective
notes I had made and sketching out the relationship between these themes
diagrammatically, reproduced in figure 20.
1. Knowledge: the meaning of knowledge, accepting multiple knowledges, the
relationship of evidence to knowledge (discussed in chapter 4).
2. Data: the data theory, exploring types of data and sources, using multiple
methodologies compared to a purist approach, delimiting the object of study
(discussed in this chapter).
3. Possibilities/alternatives: finding ways of expanding beyond my pre-
understandings, relationships of knowledge construction, creating conditions to
find alternatives (discussed in this chapter and chapter 7).
4. Awareness-rethinking: finding ways to keep track of new understandings and my
changing views over time (discussed in this chapter).
5. Influence: understanding positionality, interpretive repertoire and contextual
influences on the study (discussed in this chapter).
6. Thought styles: perceiving and comprehending concepts, applying concepts in an
appropriate way (discussed in chapter 7).
7. Research task: finding focus and clear research questions e.g. changing the
objectives of the study (discussed in chapter 2).
8. Personal abilities: overcoming problems, resilience building, dealing with thinking
in isolation (discussed in chapter 8).
9. Self (sense of subjectivity): capacity to accept and use subjective experience as a
position for knowledge creation (discussed in chapter 8).
10. Unknown limitations: dealing with uncertainty about what I don't know and what
can't be known (discussed in chapter 8).
11. Reflexivity: using reflexivity and reflective data (discussed in this chapter and
chapter 8).
130
Chapter 4: Methods
Figure 20: Reflective themes
Generalising from
experience
Integrating
reflective data in
the study – (with
empirical data/
Using reflexivity Nature of subjective and
documentary
to understand the sources) personal reflexivity used
topic (sources/focus/positionality)
Power in the
Decision making approach
trail
Understanding the breadth of the

topic area and creating boundaries Subjective opinions
about making a claim
to knowledge
Study
focus
Multiple
interpretations Subjective Validity of different
and issues in understanding of knowledge products
the topic area personal interests in
Utility of different
knowledge
tensions between products
practical versus theory

questions
Unknown Ability to gain a critical purchase

limitations on what is controlling the study
Other possibilities/ ways of seeing/

ability to access data
Keeping reflectivity in balance

with the topic
Placing limitations on reflectivity
131
Chapter 4: Methods
Relevance and validity of the data
The data used to support the thesis is highly relevant for a number of reasons. Firstly,
this study is about different knowledges and the manifestations of power that surround
and infuse user involvement in health research; not only the episodic relationships that
might occur within research projects, but the systems and structures of dispositional
power and domination that encourage, hinder or shape certain types of relationships.
This requires data from within and beyond situations that could be classified as user
involvement in health research. If data had only been collected about service
user/researcher relationships within research studies the study would overlook the
influence of policy, theory, social forces and so on.
Second, as Foucault asserts:
“We must be ready to receive every moment of discourse in its sudden

interruption; in that punctuality in which it appears, and in that temporal
dispersion that enables it to be repeated, known, forgotten, transformed, utterly
erased, and hidden, far from all view, in the dust of books. Discourse must not
be referred to the distant presence of the origin, but treated as and when it
occurs”. (Foucault 2006a, p.28)
In accordance with this view the study has included empirical data collected in situ at
the time it was created. The empirical data provides this study with the ability to
analyse the located and contingent nature of user involvement in health research in great
detail. The way I made use of myself in the research work had a significant influence on
the empirical data that was collected. The data are biased towards the events that I was
able to attend (conferences, seminars, workshops etc.), the particular sessions that I
chose to attend and the degree of my participation in these events. However, I argue that
any attempt to gain data about user involvement in health research will be partial as the
object of enquiry extends beyond any observable phenomena or definable community.
Third, the study is concerned with knowledges that construct and guide user
involvement in health research. Part of this is to explore the available evidence but it
also involves scrutinizing alternative ways that knowledge about user involvement is
132
Chapter 4: Methods
manifest and becomes accepted as ‘common sense' knowledge, for example through
shared learning and experience. This requires data about how knowledge of user
involvement in health research has been constituted, propagated and ordered in ‘real
life' contexts. Importantly, the data theory does not restrict the study of knowledge
development to the analysis of information that appears in the published literature or to
evidence created through research. The data theory accommodates the possibility that
knowledge is created through other mechanisms, such as consensus, debate and shared
learning.
Fourth, the study aims to explore subjective experiences of power associated with being
a researcher and the knowledges that construct and guide citation practices (the things
that a person does that make them identify with and be identifiable as a researcher). It is
therefore important that the data includes subjective experiences of individuals who
align themselves with these positions.
Fifth, the study aims to be a critical interpretation. It is therefore appropriate that the
study used reflective data to identify different types of influences on the study, such as:
theoretical and epistemological influences, political and social constructions that shape
the research, decisions taken at each stage, ethical issues encountered, interpersonal
contexts of the research, and the influence of the researcher in the research process.
The volume of information which informs this analysis meant that maintaining
accessible databases of published documents, empirical data, and reflective data became
a key concern. Data management was also perceived as part of ensuring the validity of
the claims being made, as expressed in the following reflective note:
{Trigger (App3/Date:21.08.06): Conversation with a supervisor about the study

data}
Convincing others of the accuracy and reliability of my observations and
cognitive processing is challenging; as is explaining how I have made
decisions about the data I have selected to inform the interpretation. Trisha
suggested providing examples and links to data sources (using a data
study log) and to make more explicit the systems I am using to code and
process different types of information.
133
Chapter 4: Methods
Hard copies of published chapters, papers and articles used in the study were maintained
in a box file library arranged in alphabetical order according to surname of the first
author. Documents not explicitly relating to user involvement in health research, for
example health service policy documents and discussion documents on related topics
were stored separately in date order. Documents collected in Australia (reports and
papers) were stored separately. A bibliography of all published documents was
maintained and supported using Endnote reference management software.
Hard copies of empirical data were stored in box files according to publication date
order. The nature of the material (often unpublished or not available electronically)
meant it was necessary to attain hardcopies of documents. Some of the empirical data
was more sensitive, for example interview transcripts and it was necessary to store this
material securely according to ethical requirements. All of the empirical material was
referenced using tables in Microsoft Word (Appendix 3: Data study log) where it was
classified according to the type of source in preparation for analysis.
134
Chapter 4: Methods
4.6 Analysis
The methods of the analysis were guided by the study questions, the approach
(interpretation, discourse, reflection) and the use of reflexive methodology. The
analyses are also informed by theoretical perspectives discussed later in this section.
The following table summarizes the analyses presented in this thesis.
Table 5: Summary of analyses
Questions Focus Analytic approach
Historical perspectives Socio political conditions Guided by Foucauldian

influencing user involvement in archaeology
(Chapter 2) health research
Review of empirical research, Structured review of research
Current knowledge
conceptualizations, themes of literature
(Chapter 3) the literature
Key mechanisms by which Guided by Foucauldian notion

understandings have developed of a system of formation
Knowledge development
(Chapter 5)
Key concepts influencing how user Foucauldian constellation of
involvement is thought of concept themes
Issues about control, intentions, Guided by social theories of
interests power
Tensions (Chapter
(Power theory map)
6)
Experiences of power Guided by Foucauldian
perspectives of subjectivity
Identification of places where Guided by Foucauldian
Ways forward new knowledge may be concepts of divergences and
(Chapter 7) developed dependencies within a system
of formation
I will now explain the core conceptual components which provide the building
blocks for these analyses.
135
Chapter 4: Methods
Core components
The approach to analysis of data in this study has been to treat the data as pieces of
information – fragments of discourse – which act as markers for the interpretation being
made. In the tradition of Foucauldian Discourse Analysis, the term ‘statement' is used
here to refer to these fragments. Foucault perceived statements to be the building blocks
of discourse. He suggested we must:
“Grasp the statement in the exact specificity of its occurrence; determine its
conditions of existence, fix at least its limits, establish its correlations with
other statements that may be connected with it, and show what other forms of
statement it excludes”. (Foucault 2006a, p. 31)
The core components of the analysis are summarized here.
(i) Statements
The first component of the analysis concerns the ways in which user involvement in
health research is constructed. It asks:
 What statements have and are being made about user involvement in health
research?
 By whom are statements being made? Is there agreement or tension between
different groups?
“Who is speaking? Who is accorded the right to use this sort of language?
Who is qualified to do so? Who derives from it his own special quality, his
prestige, and from whom, in return, does he receive if not the assurance,
at least the presumption that what he says is true?” (Foucault 1989, p.55)
 Where are these statements being made?
“We must also describe the institutional sites from which discourse is made …
and from which this discourse derives its legitimate source and point of
application” (Foucault 1989, p.56)
136
Chapter 4: Methods
Table 6: Components of the analysis
Stage Focus Aim
To identify statements (propositions, beliefs, doctrines) which

are relevant to the study questions: what is user involvement in
1 Statements health research and what does it do?
To identify themes of statements within the data
2 Divergences To identify points of diffraction between co-existing themes
3 Dependencies To identify factors which permit various themes to co-exist
To gain a clearer understanding of strategies for controlling

4 Strategies alternative themes (e.g. rules of consistency, points of
systematization)
To gain a clearer understanding of rules governing the

Enunciative
5 articulation and authority of statements – who can speak from
modalities
where and how
To explore the ways in which rules and opportunities affect what

6 Practice
can be said and done.
To explore the issues from the perspective of subjective

7 Subjectivity
experience.
The identification of statements was an ongoing process through the study and did not
occur in clearly defined stages as the previous table might suggest. The longitudinal
nature of the process of identifying statements is illustrated by the following reflective
note triggered by an interview with an academic researcher. The note illustrates that the
process of drawing links and identifying statements occurred throughout the study
including during data collection:
{Trigger (App3/Date:25.10.05): Interview with senior researcher (Participant

12)} I had another interview this afternoon. Some interesting points about
working with energy industry partners in research about local community
health concerns. It reconfirmed some of the views about stakeholders in
research that were raised at the discussion yesterday. It also reinforced in
my mind the way that people make use of the research process as a way to
support an agenda – in this case putting in a new high voltage electricity
137
Chapter 4: Methods
cable through a village that don’t want it because it might upset the
environment and the community’s health. I thought back about what [the
participant] at the carers’ organisation had been saying to me earlier on
that week about carers utilizing knowledge to lobby for change. This is one
purpose of research - a motivator of change- and change is invariably
linked to an agenda of some kind.
As is often the case in qualitative data analysis (Miles and Humberman 1994), the
processes of assimilating information influenced my judgements about the data and
hence what information I went on to collect. The notion of statements provided a way of
managing the developing interpretation by grouping issues that were emergent in the
data, as well as ensuring that the data remained ‘connected' to the contexts in which
statements had emerged. The emphasis was on perceiving statements that would explain
what was being said through the data rather than defining and applying categories to the
data, as is sometimes undertaken in a content or thematic analysis of qualitative data.
{Trigger (App3/Date:23.10.05): Interview with senior academic (Participant 9)}

As the interviews are progressing I am becoming more able to take the line
of questioning away from the original schedule. I am aware of moderating
my interview technique to probe particular issues in greater depth or to
explore a new perspective. I am also more actively seeking information
that contrasts with my own experiences of undertaking research. I am
finding myself taking the interviews off now further away from the
interview schedule. Each interview seems to add a new piece of
information or perspective to what I am trying to understand. I can make
more of this by probing further about certain aspects. For example with the
participant today I asked more about how the research structures work at
the university, to try to get a feel for the way researchers may or may not
be working together, and how that might compare with my own
experiences of research institutions in the UK.
The approach was to identify different statements across the data rather than to
summarize or synthesize the data in its entirety. Again, this was an ongoing process that
was supported by making reflective notes and maintaining the study data log (Appendix
3).
{Trigger (App3/Date:18.10.05): Reflections on how the interviews are progressing}

Collection of data has shifted from a situation where I was exploring broad
ideas and concepts to a situation where I am seeking to reconfirm ideas
from different perspectives and to draw links between different pieces of
information. The reflective notes are helping to identify issues or concepts
138
Chapter 4: Methods
that I have overlooked earlier on in the study. For example, stakeholding

was not a concept that I pursued with intent in my analyses of the
literature yet it has appeared in three reflective notes made on three
separate occasions.
This approach differs from some qualitative methods of analysis. For example
advocates of a grounded theory approach would constantly code and compare the data
until a point of theoretical saturation is reached (Strauss 1987, p.21). However a
grounded theory approach would have entailed spending too much energy on detailed
coding operations (Alvesson and Sköldberg 2000, p27). To illustrate this point, taking
the example of a seminar held at the Kings Fund on evidence for user involvement, at
least 20 issues were raised and discussed (see figure 19) but these could be synthesized
as three key statements about user involvement in health research: definitions of
evidence, what it is important to know about user involvement in research, and the
implications of measuring particular aspects of user involvement. In relation to the
questions this study seeks to address it would be a cumbersome and lengthy process to
record, code and categorise everything that I had observed.
(ii) Divergences
The second component of analysis aimed to explore the various discursive constructions
of user involvement in health research according to wider discourses which govern what
is acceptable to say or think about health research at this particular historical moment.
The analysis focused on the difference between various discursive constructions of user
involvement in health research and locating these within wider discourses. It asked:
What is the nature of the relationship between these discursive constructions? How do
they interrelate? What is their theoretical consistency? What theories and concepts are
being drawn upon?
(iii) Dependencies
According to Foucault a discourse can ‘rule in' certain ways of talking about a topic,
defining an acceptable and intelligible way to talk, write, or conduct oneself but it can
also ‘rule out', limit or restrict other ways of talking, of conducting ourselves in relation
to the topic or constructing knowledge about it (Hall 1997). The third component of the
139
Chapter 4: Methods
analysis focused on the factors which permit various discursive constructions of user
involvement in health research to co-exist. It asked:
 What is it that makes particular statements sayable? And what is not
sayable?
(iv) Strategies
The fourth component of involved gaining a clearer understanding of what various
discursive constructions of user involvement in health research are capable of achieving.
It asked:
 How do different statements relate to different forms of power?
(v) Positionings
Having identified various constructions of user involvement in health research, and
having located them within wider discourses, the fifth component was to look at the
subject positions which they offer. A subject position identifies “a location for persons
within the structure of rights and duties for those who use that repertoire” (Davies and
Harre 1999, p.35). Discourses therefore construct subjects as well as objects and make
available positions within networks of meaning that speakers can take up and place
others within. Subject positions are different from roles in that they offer discursive
locations from which to speak and act rather than prescribing a particular part to be
acted out. It asked:
 What is the nature of the subject positions created?
(vi) Practice
This component of the analysis was concerned with the relationship between discourse
and practice. It involved an exploration of the ways in which discursive constructions of
user involvement in health research open up or close down opportunities for action. It
asked:
 What is the relationship between discursive constructions of user
involvement and practice?
140
Chapter 4: Methods
(vii) Subjectivity
The final component of the analysis explored the relationship between user involvement
in health research and subjectivity, by posing the question:
 What can be felt, thought and experienced from within various subject
positions?
Analysing the data according to Foucauldian perspectives (statements, discourse,

subject positions for example) meant that the analysis could extend beyond existing
conceptualizations of user involvement. The main disadvantage of analysing the data in
this way was that separating statements from the context in which they were made
removes a degree of meaning from the data as a whole. In part, this is addressed in this
study by showing an ‘audit trail' using the data study log (Appendix 3) and reflective
notes. There is also a danger that the grouping of statements made at different times in
different settings could be perceived as failing to show due regard for the historical and
social context. The supervisory relationship was an important way of reflecting on the
analysis and interpretations being made. This particular issue is discussed in relation to
validity in chapter 8.
Analytic tools
I. System of formation
In this study the analysis of core components was guided by the idea of a system of
formation, in particular the idea of enunciative modalities which underpin the
formulation of statements. As Foucault explains we should ask:
“What is it that links them [statements] together? What necessity binds them
together? Why these and not others? Before attempting an answer to such
questions, we must first discover the law operating behind all these diverse
statements, and the place from which they come” (Foucault 2006a, p.53)
A feature of the approach taken here is the recognition of the temporally contingent
nature of user involvement in health research. Within the analysis temporality is
approached as including historical issues about where user involvement came from, why
it emerged when it did, and how it has changed over time. Foucault called such an
141
Chapter 4: Methods
exploration of the factors surrounding the construction and emergence of a new body of
knowledge an Archaeology. Foucault argued that it was important to return in history to
a “zero point” when an object of perception emerged in society and became
conceptually isolated from other concepts (Foucault 2006a). This zero point would mark
a social space where an object of perception was produced by social practices rather
than simply being an object of thought. Foucault's own use of archaeology has had
major effects on the study of subjectivity, power, knowledge, discourse, history,
sexuality, madness, the penal system and much more. Foucault employed multiple
concepts in his own archaeologies. I have made use of some of these ideas, in particular:
1) The surface of emergence: the theoretical contexts which accord objects status
“making an object manifest, nameable and describable” (Foucault 2006a, p.45).
2) Authorities of delimination: the institutions, groups, bodies of knowledge and
practice, authorities recognised by public opinion, the law and government. (p.46)
3) Grids of specification: systems according to which different kinds of object are
divided, contrasted, related, classified (p.46).
The following figure (Figure 21) is a schematic representation of the main features
Foucault describes in The Archaeology of Knowledge. This was developed by reading
Foucault's descriptions of elements of systems and visually illustrating the relationships
between them. The figure is constructed around the notion of ‘statements'. Statements
are borne out of observation, experience, data, practice, and so on and they relate to an
object. The theoretical contexts that provide statements with the possibility of
emergence are, according to Foucault, the surface of emergence. For example,
statements might emerge in the form of language, and they are influenced by ideas
about what can be said (authorities of delimination) and how statements can be
understood (grids of specification).
These perspectives are applied throughout the thesis but they are drawn together in
chapter 5 to examine how knowledge of user involvement has developed.
142
Chapter 4: Methods
Figure 21: Systems of formation (discursive formations)
Dispersion of elements
Strategies: Strategies for

controlling a range of
alternate themes in a
discursive formation; such
as rules of consistency,
points of systematization
Divergences:
Points of diffraction
between coexisting Enunciative modalities:
statements Rules governing the
articulation and authority of
Statements: Dependencies: statements - who can speak
Propositions, System of from where and how
beliefs, theories, dependencies
doctrines borne between levels of
out of observation, component
experience, data, elements that
practice allows Concepts: Rules governing
coexistence of the formation of statements;
statements succession, coexistence and
intervention
Objects: What a statement is

about
Surface of emergence: the theoretical contexts which accord objects status making an object
manifest, nameable and describable
Authorities of delimitation: institutions, groups, bodies of knowledge and practice, authorities

recognised by public opinion, the law and government
Grids of specification: systems according to which different kinds of object are divided,
contrasted, related, classified
143
Chapter 4: Methods
II. Power theory map
A second feature of the study is the focus on what user involvement does. This part of
the analysis is informed by social theories of power, drawn together in a power theory
map (Figure 22).
Although social theories of power have been applied in research about health care
practice (as discussed by Porter 1998 and Potter and Wetherall 1987 for example), my
introduction to this literature came through reading about Foucauldian perspectives. In
particular the notion of genealogy, which can be thought of as “the critically committed
probing of the roots of societal practices” (Alvesson and Sköldberg 2000 p.224).
Foucault showed that by analyzing discourse we can see their constraints and where
they situate different people. Here, I am interested in sites of tension and resistance,
which is a slightly different take on identifying the challenges or barriers to user
involvement. I am interested in the positive effects tensions and resistance can have for
how knowledge is developed.
The power theory map I have devised utilizes Foucauldian perspectives alongside a
range of other theories that might usefully inform an analysis of user involvement in
health research. Foucault calls so many things power and I required a more tenable
definition. Mapping theories in this way helps to examine the possibility of different
interpretations of what user involvement in health research is and what it does. The map
I have constructed is an early starting point for exploring alternative notions of power in
this context.
The basis of the map is derived from Clegg's (1989) Circuits of Power (presented in
chapter 6), who acknowledges the influence of Foucauldian and Habermasian
perspectives. It has previously been argued that there are advantages in combining the
work of these two writers, both provide stimulus for exploring science's relationship
with democracy (Jones 2001b). I have also incorporated perspectives put forward by
Mann (1993) which suggest power can be diffuse, authoritative, collective, distributive,
extensive and intensive.
144
Chapter 4: Methods
The map also incorporates ideas from Dahl (1961), Giddens (1984) and Jones (2003) to
highlight that issues of temporality are a theme in the theorization of power. In the
words of Bhavnani:
“Historical work is very important in providing a means of understanding

how the present, and possible future change, is influenced by the ways in
which the past is remembered and understood.”(Bhavnani 1990, p.143)
The theory map depicts the interface between social structures and the individual person
is the place where power is either recreated or resisted through the intentions and
actions of individuals. These ideas are developed further in this thesis in relation to
subjective experiences of research and orientation of intent. This corresponds with a
view of the relationship between structural approaches to power and human agency as
“a two-way transformative process” the nature of which changes over time in the
presence of “exogenous environmental contingencies” and “alternative/counter powers”
(Giddens 1984).
I will go on to explain the work of these and other authors cited on the map in chapter 6,
and examine power in relation to user involvement in health research. In the discussion
chapter (Chapter 8) I look more closely at what the social and subjective construction of
user involvement in health research might mean for policy, research and practice. I also
suggest some further authors that I could have usefully drawn upon.
145
Figure 22: Power theory map
Exogenous environmental contingencies (Giddens 1984)
(after Clegg 1989) Social structures
Episodic Power Relations Diffusion

Subjectivity
(Mann 1986)
Social relations (Foucault 1979)
Agencies: Condign: force, Compensatory: resources, Conditioned: ‘Ability to’
persuasion (Galbraith 1983) Access to means/resources (Dahl 1968)
Mechanisms: Violence, Wealth, Knowledge (Toffler 1990), Achieve intentions (Weber 1947) Achieve
Information, Reward (French & Raven 1960) Collective and goals (Parsons 1967)
Incentive structures: Cost, Choice (Dowding 1992) Standing distributive Make decisions (Bachrach and Baratz 1962) Deliberate
Agency Causal (Mann 1993)
conditions (Dahl 1961) (Habermas 1972, Bohman 1996)
Conditions for communication (Habermas 1978) Outcomes/effects of Adapt (Tilly 1999)
power
Rules of practice (Foucault 1979) ‘Potential to’

Social Integration Dispositional Obligatory passage points (Clegg 1989)
Citation practices (Foucault 1982) Reference:
Rules of meaning and membership
loyalty (French & Raven 1960) Expertise,
Subject positions (Foucault 1982) Legitimate (French & Raven 1960) Credibility (French & Raven 1960) Intention
Authoritative (Mann 1993) (Wrong 1979)
System Integration Facilitative

Domination ‘Sense of being’ Subordination/
“Will to power” intention and desire (Nietzsche 1967) Power superordination or reciprocity
by order (Galbraith 1983) (Simmel 1971)
Innovation in techniques of discipline and production and Constrained/Enabled (Machiavelli 1958)
utilization of technologies (Foucault 1977) Exploited/Empowered (Tilly 1984)
Cumulative structures (Archer 2000): Production, Coercion Consenting/Coerced (Gramsci 1971)
Persuasion (Runciman 1989)
Instrumental efficiency (Foucault 2000) Extensive and intensive (Mann 1993)
Cultural hegemony (Marx 1976, Laclau and Mouffe 1985)

Consciousness to power
Social dominance/normalized truths (Sidanius & Pratto 1993)
Alternative/counter powers (Giddens 1984)
Time Pre-existing Timeframes of structuration: Transitions, micro-

conditions (Dahl 1961) lifespan, generations (Giddens histories, development,
1984) models of power relations 146
(Jones 2003)
Chapter 4: Methods
Summary
 In this study understanding user involvement in health research means asking how
this phrase came to have social meaning and how it is used and experienced.
Issues in and around user involvement in health research have guided decisions
taken about what user involvement means and how it can be understood.
 The approach to knowledge development applied in this study is informed by

philosophical perspectives of interpretation, Foucauldian perspectives and
reflectivity. The approach is that:
(i) The interpretation/claims made are based in a view that knowledges are
transitional and multiple.
(ii) The interpretation/claims made are recognised as being contingent and
historically located.
(iii) An approach to knowledge creation has been taken (interpretation) that
emphasises that the object of user involvement in health research has
multiple dimensions.
(iv) A reflective stance is used in the study to surface subjectivity: experience,
observation, thinking, communication and so on.
 The study uses reflexive methods which support an exploration of issues,

challenges and possible implications at an individual, group, organisational and
the social level. The method of Reflexive Interpretation provides a multi-layered,
flexible structure of interpretation and reflection in which the systematic interplay
of reflective areas is central. It involves (i) Levels of interpretation: Empirical
material/Construction of data, Interpretation, Critical interpretation, Self-critical
and linguistic reflection (ii) Forms of reflexivity: Contextual, Critical,
Interpersonal, Personal, and (iii) Reflective themes: such as theoretical and
epistemological influences, political and social constructions that shape the
research study, decisions taken at each stage of the research, ethical and political
questions encountered in the research, interpersonal contexts of the research, and
the influence of the researcher in the research process.
147
Chapter 4: Methods
 The study makes use of three sources of data: documents, empirical data and
reflective data. The empirical data in this study includes data collected from UK
health service research, data collected in the UK NHS from Patient and Public
Involvement Leads, and data collected in southern Australia. The data are relevant
and appropriate to this study because (i) they are drawn from multiple sources
supporting a view that user involvement in health research is multidimensional:
including conceptual, practice and experiential dimensions (ii) the data relate to
the questions that are driving the study.
 In this study reflexivity changed overtime from self-critical and linguistic

reflection towards critical interpretation about what was shaping the study and the
construction of the data towards reflection about the interpretation and the
implications of the interpretation being made. Reflective themes included:
questions about the meaning of knowledge, the meaning of ‘data', finding ways of
exploring and delimiting the issues, and personal skills and abilities. It was not
possible to classify these themes as being purely relating to reflection about
contextual, critical, interpersonal, or personal factors.
 The analysis aimed to identify statement themes (propositions, beliefs, doctrines)

about user involvement in health research within the data, to identify points of
diffraction between co-existing themes, to identify factors which permit various
themes to co-exist, to gain a clearer understanding of strategies for controlling
alternative themes and rules governing the articulation and authority of statements,
to explore ways in which rules and opportunities affect what can be said and done,
and to explore these issues from the perspective of subjectivity (a person's own
experience).
 Two tools have been developed to inform the analysis. A ‘system of formation',
which draws together Foucauldian perspectives to highlight different factors that
how knowledge of user involvement in health research is developed. A ‘power
theory map' has been developed to inform the analysis in this study. It draws
together selected theoretical perspectives to examine points of tension.
148
Chapter 5: How knowledge of user involvement has
developed
5.1 Introduction
In progressing an understanding of user involvement in health research this study now

looks at how knowledge has been developed in this context. It begins by looking at how
knowledge has been perceived. Philosophical perspectives suggest that knowledge is
not a fixed entity and that some forms of knowledge are valued more than others. These
perspectives are explored in this chapter firstly from a historical perspective and then
using the Foucauldian notion of a constellation of concept themes.
The approach taken here focuses on the role of concepts in knowledge development and
the overall aim of the chapter is to build towards an understanding of how knowledge of
user involvement in health research has been developed. The Foucauldian notion of a
system of formation is used to illustrate the connectivity of these different factors.
In the final section of the chapter I use an interview with a member of a carer
organisation to illustrate how particular knowledges about user involvement in health
research may be perceived and valued for different purposes.
5.2 What knowledge is
To a certain extent, what knowledge is in relation to user involvement in research is

depends on the way a person wishes to think about it and what it is they want to
understand about it. There are many types of people with interests in this topic area,
with a range of professional backgrounds – mental health, social care, community
health, clinical sciences/lab research, knowledge transfer, commissioning, heads of
consumer health organisations (who might also commission research), social scientists,
epidemiologists, Action Researchers, nurse researchers, and so on. There are also
people who identify themselves as service users, survivors, carers or patients that are
interested in understanding more about user involvement in research.
149
Chapter 5: Knowledge development
{Trigger (App3/Date:07.03.06): Launch meeting of InvoNET}

The people in this audience are not the same. Some agree with user
involvement in health research, some do not and some do not know; some
have given it little thought, or are very new to the topic. Some people aim
to involve service users in research or to ‘make involvement happen’.
These people include professional researchers, but there are also
commissioners of research, members of research support organisations or
other individuals. People who identify themselves as service users are also
interested in finding out what user involvement is and how to put it into
action. There can be tensions between individuals or groups, and the
dynamics of meetings can be challenging. Some people may feel unease
about sharing their views about developing knowledge about user
involvement or how inclusive it can be. What ‘we’, as a group of people,
are interested in knowing is diverse in its scope and nature.
The question of what knowledge is in relation to user involvement in health research is

not simply a matter of personal preference for types of information. How knowledge is
thought of has implications for the development of future knowledge in this field. The
question contains ontological and epistemological dimensions that can be highlighted by
drawing on theoretical perspectives of knowledge. For example, knowledge of user
involvement could be perceived as lived consciousness: according to the work of
Ludwig Wittgenstein (1889-1951) it is important to identify relations between cognition
and language in the formation of knowledge (Warburton 2006). Jean-Paul Sartre's work
(1905-80) on experience, meaning and consciousness (Sartre 1956) would approach
knowledge in a similar way. In accordance with Martin Heidegger's (1889-1976) work
knowledge of user involvement might be perceived as how it appears to us in our
consciousness (Heidegger 1962). Such views of knowledge in relation to service user
involvement would tend to emphasise the experiences of those who are directly engaged
in this form of social practice.
According to the views put forward by other philosophers, such as Gaston Bachelard
(1884-1962) and Alexander Koyre (1892-1964), knowledge of user involvement would
comprise concepts, systems and structures. These authors approach science and
knowledge not as objective or constant truths but more as discontinuous activities of
communities that construct truth (Chimisso 2003). Their work questions the rationalist
150
and objective claims of science, in particular how concepts enter science and become
normal. This type of view of knowledge in relation to service user involvement would
tend to privilege historical and sociological perspectives of research systems and
structures.
From a Foucauldian perspective scientific knowledge is perceived as being organised

and operationalized knowledge. Much of Foucault's work is very relevant to the issues I
am exploring here because it examined how different theories of knowledge define what
is knowable and certain through history. According to Foucault what we think we know
about a topic in a particular period has a bearing on how we regulate and control that
topic. Hence, knowledge takes on different meanings in specific situations, historical
contexts and institutional regimes. In the Order of Things, (Foucault 2006b) Foucault
looked at what constituted science in the Renaissance, the classical era and the modern
era. He unearthed contemporary codes that determined what could or could not be
defined as scientific. Foucault found that each historical period has it own episteme. It
limits the totality of experience, knowledge and truth, and governs each science in one
period. A science becomes normal when scientists agree that their work has identified
and solved scientific problems. For example in the Renaissance episteme words and
things were united by their resemblance: convenientia (connected things near to one
another), sympathy (universal attraction), aemuation (similitude within distance) and
analogy (similar relations) (Foucault 2006b, p.74). In the Classical episteme
resemblance was replaced by discrimination. This way of knowing allowed the
classification of stable and separate identities and the possibility of representation. This
understanding led Foucault to suggest that knowledge could act to sustain a particular
regime of truth.
A more familiar expression of these ideas is the notion of a research paradigm. Guba
and Lincoln define a research paradigm as:
"a set of basic beliefs (or metaphysics) that deals with ultimates or first
principles. It represents a worldview that defines, for its holder, the nature of
the world, the individual’s place in it, and the range of possible relationships
to that world and its parts” (Guba and Lincoln, 1994, p. 105).
151
Guba and Lincoln made a significant contribution in articulating four differing

worldviews of research - positivist, post positivist, critical, and constructivist- based on
their ontological, epistemological and methodological assumptions (Guba and Lincoln
1994, 2005). Within a particular paradigm, particular types of knowledge may be
explicitly privileged or valued. For example, as shown in chapter 3, research interest in
user involvement have predominantly focused on issues of practice and technique,
which may reflect the importance researchers have placed on changing research
practice.
{Trigger (App3/Date:30.10.07): Discussion with a colleague about epistemology}

Any knowledge of user involvement in health research is constrained by
what information is available to us; what information we can access; and
the limitations of the approaches we use to build knowledge. We can not
access information about the present because all knowledge becomes
historical at its point of origin. Research methods and theoretical
approaches to knowledge construction offer tools to raise our awareness of
what is omitted or missing from the archive. However, at a much more
fundamental level any knowledge of user involvement in health research is
constrained by what we think we know about the topic – it has a bearing on
how we regulate and control that topic.
As I have previously stated knowledge has been a reflective theme of this study, and
this is again illustrated by the following reflective note.
{Trigger (App3/Date:30.10.07): Discussion with a colleague about epistemology}

Each time user involvement is scrutinized an alternative position is taken
and an alternative view is gained. For example, when presented as a body
of scientific knowledge the evidence on methods and outcomes appears to
be poor and it is difficult to draw any generalisations. When approached
from the perspective of a socially constructed object, user involvement in
health research appears to be influenced by social and political forces.
When approached from the perspective of the experience of the subject,
user involvement can be seen to provoke reflection about the meaning of
research and the conduct of research practice. The knowledge base in this
topic area is developmental and it is beneficial to scrutinize the topic from
alternative perspectives of what knowledge is – but we should also think of
the ‘knowledge base for knowledge’ as being developmental.
152
These perspectives suggest that knowledge is not a fixed entity and different types of
knowledge about user involvement in health research are valued more than others by
individuals and by society. The question of what user involvement in health research is
complicates attempts to build knowledge of it. On one level, definitions have been
important. For example the definition developed by INVOLVE:
“By involvement in research we mean that people who use services are active
partners in the research process rather than subjects of research. We also
include in these terms people who use services doing their own research.
Many people describe public involvement in research as doing research with
or by the public rather than to, about, or for the public” (INVOLVE 2004a, ii)
What user involvement in health research is can be further understood by scrutinizing

the linguistic and symbolic meaning of the components it refers to: ‘user' ‘involvement'
‘research'. It is also possible to scrutinize what these terms represent when they are
brought together in a phrase, for example, how this phrase is perceived and articulated
differently in a particular social and cultural context. However, this thesis argues that
understanding user involvement in health research is not only a matter of semantics or
definition: it involves a critical exploration of power.
“Sufficient evidence is available to show that the involvement of consumers in

all aspects of research benefits both researchers and consumers and that such
endeavours are achievable. We believe that for widespread adoption of
consumer involvement to occur, pressure will have to be brought to bear by
journal editors and research councils” (Tallon et al. 2000b, p.380)
Despite the strength of the claim presented above, few authors have attempted to define
the limits and nature of existing knowledge on user involvement in health research.
Some researchers have made inferences about where boundaries of user involvement in
health research might be drawn. For example the review by Boote et al. (2002) included
and excluded publications on the basis of their relevance to health research. Other
authors have defined knowledge according to a service context such as user
involvement in primary care research (Barnard et al. 2005), or according to a particular
type of research method such as user involvement in randomized controlled trials
(Hanley et al. 2001). Such issues of scope and definition of knowledge were a concern
that faced us during the multi-method review (Smith et al. 2005b). Although the
commissioning brief required us to approach the topic from the perspective of nursing,
153
midwifery and health visiting, working with the service user reference group was
important for exploring the scope of the concepts, the type of literature we drew upon,
and our views about what we were aiming to find.
Although the scope we had defined in the multi-method review was useful for including
a broad range and type of documents beyond the research literature, it was difficult to
conceptualize what might count as knowledge, evidence or theory. As the multi-method
review progressed we were talking and thinking more in terms of issues than evidence
and this is reflected in the framework that we developed from this work (illustrated by
figure 23), and paying specific attention to explaining how practical issues of
involvement can be overcome (Smith et al. 2005d). User involvement was approached
as comprising different categories of issues to do with context, researcher role,
methods/approaches, and rationale/outcomes of user involvement. The present study
provided an opportunity to develop this understanding by focusing on the structures of
meaning making that contribute and shape user involvement in health research.
154
Figure 23: Issues framework for user involvement in research (Smith et al. 2005b)
Issues Framework Trends in

knowledge & Research structures and
evidence systems (Commissioning,
Changing needs
ethics, governance,
& rights
institutions
Trends in practice
Research cultures
(theories, concepts,
Political trends Contextual strategies)
issues
Policy/managerial
requirements
Social credibility Rationale/ Researcher

Outcomes role Researcher
Research quality identities &
practices
Empowerment/
Methods/
therapy Researcher skills &
approaches
training
Iterative change
Forms of
representation Access issues (to users, to
Techniques and information, to research)
approaches
And so the question of how we might define evidence in this field was familiar to me at
the outset of the present study and I became more interested in concepts of knowledge.
{Trigger (App3/Date:09.09.06): Discussion with a supervisor about what a PhD

requires}
When I began this study I asked my supervisors what was required from a
PhD thesis. We discussed how a PhD should make a unique contribution
to knowledge. But when one is doing a study, which asks questions about
what knowledge is, one quickly begins to see that knowledge is by no
means a fixed entity. Reading debates about lay knowledge (particularly
the working paper on knowledge produced by Ray Pawson and colleagues
(Pawson et al. 2003) at the Social Care Institute for Excellence has
challenged me to question what knowledge is.
An overall conclusion that can be drawn is that different approaches and methods of
analysing user involvement in health research affect what can be said about the
155
knowledge that exists. The following section looks at the characteristic types of
knowledge about user involvement in health research that have been created.
5.3 A stratigraphy of knowledge
{Trigger (App3/Date:22.06.07): Postgraduate research symposium}

Research has been taken along particular courses and lines of thinking.
New studies are devised on the back of previous findings that are
themselves the consequence of research that is based on the findings of
research which preceded it. An evolutionary record of knowledge strains
exists. It is generally research outcomes that are critically reviewed rather
than the questions that are being asked. It is not until the knowledge base is
seen as a historical and socially connected entity, at a point when much
research has been undertaken, that the lines of thinking might be perceived.
Over time knowledge of user involvement in health research appears to have been laid
down according to certain lines of enquiry. To illustrate this I have depicted knowledge
of user involvement in health research as a stratigraphy (Figure 24). As an object of
knowledge user involvement in health research appears to be a new entity, yet its
conceptual roots dig deep into other research bases. It is hard to get a purchase on where
knowledge of user involvement in health research became differentiated from other
knowledges.
Attempts to build knowledge of user involvement in health research began with the
identification and definition of ‘user involvement in health research'. The creation of
particular phrases and terms in research and policy documents made it possible to talk
about ‘co-construction' of knowledge through ‘participation' and ‘involvement'
(Chapter 3). The existence of research terms in phenomenology describing processes of
‘deliberation' and ‘meaning making' suggested alternative ways of knowing to
experimentation and reasoning (Husserl 1980). The existence of concepts and terms
associated with community development, for example: ‘experience' ‘emancipation'
‘empowerment', made it possible to suggest alternative bases and rationale for
156
knowledge development and to contest the legitimacy and value of existing structures of
expertise and credibility (Irwin 2006).
Hence research in this field became more historically aware. Attention turned to
explanation of the origins of user involvement, fuelled by a concern to identify whose
agenda this is and whose motives drive it. This is typified by Beresford's seminal
writings on ‘top-down' managerial regulation versus ‘bottom-up' democratic
empowerment (discussed in chapter 3).
Interest then turned to finding ways to conceptualize practices and approaches to user
involvement in research. Lessons and models from other fields were purposely sought
and drawn upon (Baxter et al. 2001). The most pervasive example is Arnstein's (1969)
ladder of citizen participation (Figure 10, Chapter 3). New conceptualizations were
devised, and ‘consultation, collaboration, control' became the main frame of reference
for understanding. Hierarchical levels of control fuelled aspirations for more equal
forms of service user/researcher relationships and partnership approaches to decision-
making. They also paved the way for classifying particular forms of user involvement-
rather than particular practices- as being tokenistic. Hence, role development was the
logical next step, and this is reflected in the literature in the form of accounts of the
creation and role of user-researchers within projects and within established research
institutions.
Towards the end of the 1990s, as policy change for user involvement in health research
received a higher profile knowledge development expanded and entered a phase of
replication and testing. However, this form of enquiry has tended to be secondary and
subsidiary, in that knowledge has been developed through reflective accounts of
practice rather than prospective funded research studies. This phase of knowledge
development is characterised by researchers' accounts of attempts to involve particular
groups of the population in localized research contexts: typified by Ross et al's study of
involving older people in research about falls (Ross et al. 2005).
157
The next layer marks a concern to develop explanatory generalisations. Some larger
scale funded research studies aimed to draw conclusions about ‘what works' and how
user involvement can best be put into practice: for example the TRUE project (Faulkner
2004).
Figure 24: A stratigraphy of user involvement in health research
Evaluation of accountability structures Prediction based on current

statements
Adaptation and differentiation of practice
8. Application of theory and testing
7. Epistemological enquiry
Knowledge to date
6. Explanatory generalisations
5. Replication and testing
4 Role development
3. Conceptualization of practices/approaches
2. Explanation of historical origins
1: Identification and definition
Pre-conditions (surface of emergence)
This was followed by a greater level of general interest in epistemological issues – what
is the nature of evidence? This is typified by the focus of discussions supported by
InvoNET (see for example, Appendix 3: Ref: 07.03.06 and Ref: 21.02.08) and
INVOLVE conferences (INVOLVE 2004, 2006). More recently studies have attempted
to apply and test theory in this context. This is exemplified by Davies et al's (2006)
158
testing out of Habermasian perspectives in relation to the setting up of a consumer panel

at NICE.
Trends favouring plurality across the health domain (discussed in chapter 2) suggest that
knowledge development will move towards adaptation and differentiation of user
involvement theory and practice within different areas of health research. These issues
are discussed further in chapter 7.
5.4 Concept themes
Another way of looking at user involvement in health research is from the Foucauldian
perspective of a constellation (Foucault 2006a p.74) of concept themes. In chapter 2 I
identified five socio-political trends which have contextualized the development of
understanding about user involvement in health research. These trends connect service
user involvement with a number of major discursive formations, including governance:
law/policy making and implementation; economics: commissioning, consumerism;
research: science, user-controlled research, quality assessment, and; health:
medicine/healthcare practice, health service development. Within and between these
formations there are ‘themes', a Foucauldian idea used to mean associated types of
concepts. This section takes forward the analysis presented so far to look at four
constituent themes which have particular relevance to user involvement in health
research:
 Knowledge
 Needs and rights
 Responsibility
 Representation
On the following figure user involvement is depicted as being interrelated with the four
major discursive formations (double headed arrows). User involvement in health
research is positioned at the centre of this constellation simply because it is the object of
this study, and not because it is constructed equally by these formations, nor is user
159
involvement central to the formations upon which it is contingent. Discursive

formations and themes are unbounded, they are transient and interdependent. This
location illustrates that user involvement in research is not wholly located within
research, policy, consumerism or any other conceptual theme associated with these
major discursive formations. It is a product of all of these shifting ideas.
Figure 25: Constellation of user involvement in health research
Discursive Constellation
Knowledge Representation
Governance
User
involvement
Economics in health Research
research
Health
Needs & Rights
Responsibility
Notes
1. The inner circle represents the major discursive formations that contribute to the
construction of user involvement in health research
2. The outer circle represents categories of themes within the constellation
3. Discursive formations and themes are unbounded, they are transient and interdependent.
160
(i) Concepts of knowledge
An important issue in relation to knowledge development about user involvement in

health research is that knowledge is thought of differently across different contexts of
health research. Research language, epistemological stances and understandings of the
researcher role are not consistent features across different domains of health research.
Research context is an important controlling factor for how knowledge about service
user involvement is perceived and developed.
{Trigger (App3/Date:09.05.07): Conference on Service Delivery and Organisation,

hearing a clinical scientist present their views about what is important in health
research}
When I began to study user involvement in health research I already had
some insight into the social practices of research through my job as a
researcher. It could be that this helped me to appreciate the meaning of
research practices to other researchers. However research practices are not
consistent across all research and individual researchers align themselves
more with particular research methods and practices than with others.
There are big differences between what it means to be a clinical health
researcher and an applied/health service researcher.
The literature shows that there are differences in how user involvement has developed
in the context of experimental research, systematic review and action orientated
projects. Furthermore, where reviews have been done of user involvement in particular
research contexts these report variation according to client groups involved and the
types of activities they are involved in. For example, in 2000 the Medical Research
Council in collaboration with Consumers in NHS Research reported the results of a
survey on consumer involvement in the design, conduct and interpretation of
randomised controlled trials (INVOLVE 2000). The survey found that:
“19 of the 62 centres identified trials which had actively involved consumers.
Of these 19 centres, 10 reported only one such trial”.
“Most of the trials identified in the survey which involved consumers took
place in the perinatal, HIV/AIDS and cancer fields. Consumers were most
often involved in drafting or reviewing patient information leaflets for the
trial, developing the trial protocol, and in promoting recruitment to the trial”.
(Consumers in NHS Research, 2000 p.9)
161
Differences in the types of service user/researcher relationships that occur within

different contexts of health research can not simply be attributable to alternative
methodological approaches being applied; the subjectivities of researchers in these
fields must also be an influencing factor. Later in this section I will go on to reinforce
this point by discussing differences within the context of systematic review. But before
then I will look at a more significant influence across health research concerning the
issue of truth.
User involvement in health research sits at the centre of a number of ongoing

epistemological debates. These can be summarized as follows:
(i) Debates about the usefulness of lay perspectives (for example, Entwistle et al,
1998; Glasby and Littlechild 2001; Marshall et al. 2007).
(ii) Debates about the standing of lay knowledge compared to other types of
knowledge (Pawson et al. 2003; Popay and Williams 1998).
(iii) Debates about the appropriateness of particular research methods for accessing
such information (for example, Edwards 2000; Edwards and Titchen 2003;
Entwistle et al. 2002; Murphy 1986; Ryan et al. 2001).
(iv) Debates about ownership of lay knowledge, which traditionally has been with
those who have skills and ability to produce research evidence.
User involvement is also caught in the cross fire of a more fundamental philosophical
debate about what we can know about reality: what we can say is true. Consideration
for what user involvement in health research is and what it should do has been
undertaken in the presence of very strong views about truth, and in particular the
relationship between research and reality (Bhaskar 1978).
“Truth, Foucault says, is ‘a thing of this world’ – meaning that truth exists or
is given and recognised only in worldly forms, through actual experiences and
modes of verification; and meaning also that truth is a serious matter and a
serious force in our world, and that there is work for us to do in investigating
the presence and effects of truth in the histories of our societies.” (Foucault
1982)
These debates manifest themselves through user involvement in health research in a

number of ways, but generally using the language of ‘evidence'. The concept shapes
162
and influences the way knowledge of thought of in this field and this issue of an
evidence base is discussed in Section 7.2.
From the 1980's onwards notions of evidence were being more widely taken up through
ideas about evidence-based practice and evidence-informed policy although attention
was being drawn to differences with patient-centred medicine (Bensing 2000). Concepts
of quality in evidence were conceived according to scientific values of objectivity,
statistical representation, controlled intervention, replication and randomization. This
point links to notions of hierarchies of evidence as being a form of power (discussed in
chapter 3). Conceptualizations of evidence define a hierarchical relationship between
scientific knowledge and subjective experience (Lewis 2001), suggesting the former is a
more rational basis for decisions about human lives.
{Trigger (App3/Date:22.06.07): Postgraduate research symposium, a presentation

about a study looking at compliance to health promotion advice in alcohol misuse}
The questions and outcomes of research provide leverage for policy and
practice development. The professions are built on, and gain power
because of their ability to define what evidence is and where it needs to be
built. The professions say who is in and who is not in the profession – who
has a right to create the evidence, to create the policies and the practices
that shape society. Health promotion research and practice is an example
of professionals creating a role for themselves in legitimizing techniques of
control and compliance models. The concepts of risk, safety, personal
responsibility and self efficacy all serve to reinforce a social compliance
model. Tools and scales for screening individual’s readiness to change are
another professional tool for controlling individual behaviour. They
suggest a way of behaving and provide a knowledge/power framework for
professionals to draw upon. Through research knowledge becomes a tool
for controlling behaviour.
Historians and philosophers of science have looked at how ideas become normalized
within certain epochs and social settings. For example, Habermas criticised science for
being deluded by its normalized truths:
“Because science must secure the objectivity of its statements against the
pressure and seduction of particular interests, it deludes itself about the
fundamental interests to which it owes not only its impetus but the conditions of
possible objectivity themselves” (Habermas 1978, p.)
163
And so did Foucault. For Foucault, because the concept of knowledge is not generally
perceived as a natural entity that is embedded in human instincts there is a break
between knowledge and things (Foucault 2000). Views about truth have real
implications for the types of research that are valued and undertaken in any research
system. For instance, in relation to emancipatory research it has been argued that:
“One of the key barriers to the development of emancipatory research is what

is currently seen as ‘evidence’. Most research commissioners seek to fund
research that is perceived to be unbiased, objective and neutral. This has
meant that existing research structures have prioritized and valued more
traditional types of research over research that is seen to be emancipatory.
Emancipatory research may not be seen as ‘real’ research because it focuses
on people’s experiences and making changes. This means that, to date, very
few emancipatory research projects have been funded and even fewer have
received large amounts of funding …There is often a glass ceiling for many
service user researchers” (Hanley, 2005)
This significance of concepts of evidence as a controlling point for user involvement in

health research is also exemplified in the context of systematic review. Although
systematic reviews are increasingly being commissioned to inform policy development
and provide recommendations for practice and research, when compared with other
research contexts user involvement appears scarce and highly variable (Braye and
Preston-Shoot 2005). In the context of mental health systematic review, Salvi et al.
(2005) are able to identify 35 studies that have involved service users. Of these, three
studies were user-led, in three other studies the users were simply consulted but did not
have any active role in the research. The remaining 29 studies were based on
collaboration between service users and professional researchers.
A review of the work of the Cochrane Collaboration (Gheresi 2002) revealed different
approaches to involvement have been tried and different levels of success have been
achieved. Setting up stakeholder groups and involving users in reviewing have been
lobbied by some members of the Collaboration (Appendix 3: Ref: 22.10.05: XIII
Cochrane Colloquium). Whereas some Cochrane Review Groups have embraced the
concept and been fortunate in their ability to identify willing and able contributors, such
as the Consumer and Communication Review group (CCCRG 2007), others have not
been as successful (Kelson 1999). It is likely that a combination of factors has inhibited
164
the development of user involvement in this context (Smith et al. in press, Appendix
11). Whist it is reasonable to presume that health researchers what to help people make
well-informed decisions about health care this does not necessarily mean they all agree
with directly involving service users in the work they do. Some researchers have
expressed concerns about “contaminating the research” or introducing bias (Rhodes et
al. 2001); others have raised issue about the status of subjective views in the process of
generating evidence (Glasby and Beresford 2006); there are also issues about when and
how to incorporate lay knowledge into a systematic process.
Another research context where similar conceptual issues arise is in debates about
mixed-methods research. In this context academic debate tends to be about whether
qualitative and quantitative data can be combined to get closer to truths – rather than
whether knowledges can be categorically different and therefore incommensurate.
{Trigger (App3/Date:11.06.05): Attending and presented at a conference on Mixed-

methods in Health and Social care Research in Cambridge}
Professor David Morgan (Portland State University, USA) argued that a
paradigm-orientated approach to combining research data is based on
several seriously flawed interpretations of metaphysical concerns about the
nature of truth and reality and the concept of incommensurability. He
instead suggested that Pragmatism is the primary alternative to move
beyond simply justifying the utility of using multiple methods to address
the broader issues that interest society – such as the possibilities of and
limits of meaningful comparisons between qualitative and quantitative
research.
Such a view may result in the perception that user involvement is more compatible with
forms of research that are classified as belonging to a qualitative research paradigm
(defined by Guba and Lincoln 1994).
(ii) Concepts of needs and rights
User involvement in research has been approached as a question of moral rights (Boote
et al. 2002). This has sometimes led to the view that user involvement is ‘politically
165
correct' rather than reasoned or proven. Wider social discourses of needs and rights are
important to how user involvement in health research is thought of and enacted. For
example, ideas about social inclusion through research processes are made prominent
through attention to health inequalities (DHSS 1980) and the cost savings that could be
achieved by investing in health (Wanless 2004).
The presence of a socialised health system and a welfare state in the UK, based on
principles of needs and rights, has provided leverage for the idea that service users have
a right to say how public services are run. Suggestions that the public should have a
right to more say in decisions about health systems gained authority under the Thatcher
government's agenda to improve the quality and accountability of public services
(Barnes 1997): epitomised by the development (rather than the limited success of) ‘The
Patient's Charter' (DH 1995). Under the Blair government the view that patients have
consumer rights (and responsibilities) has been emphasised through the choice agenda
in the NHS (DH 2005) and through the wider context of scientific governance
(discussed in chapter 2). However, critics of health policy making in the UK have
argued that the definition of need is associated with concepts of professional expertise
(Spiers 2003).
In this study the notion of need was most often expressed as a perceived need for
research to better address the needs of the ‘end-users'. For example, one Australian
researcher said:
(PARTICIPANT 2 - AUS): I think it is vital because the nursing role is all based around the
patient. We really need to know and understand their concerns and their needs and expectations.
How the end-user is perceived and constructed appeared to be defined by the purpose of
a particular organisation (for example an academic department or health service) or
according to what it was that was being supplied (such as research findings or a health
intervention). Furthermore, the terms end-user and provider suggest a situation of
dominance and subordination: the end-user is being provided with something they can
not do without. The situation is more complex than it appears. An end-user might not
wish to receive or use whatever they being provided with, or a person may not perceive
166
themselves in this way. According to a consumerist view disconnection between

providers and end-users can be harmful to either party, for example, providers may not
be able to show they are effective or profitable and end-users may have unmet needs. It
also been argued that the end-users may in reality be the commissioners of research
(Wickham and Collins 2006).
In relation to user-led research, research processes have been used as a platform for
lobbying for change and the mobilization of social movements. From this perspective
research is perceived as offering social opportunity and empowerment to disadvantaged
or disempowered groups (Beresford 2005). It has been argued that user-controlled
activities tend to be undertaken by groups of service users who have a clearly articulated
collective understanding of needs (Taylor 2006). A good example of this is a recent
formal investigation by the Disability Rights Commission (2006). In this case research
has bee used to draw attention to the unmet physical health needs of people with
learning disabilities and/or mental health problems and to highlight the widespread
paucity of professional and public understanding about disability. Recently Beresford
(2007) has drawn links between concepts of health inequalities and user involvement in
health research, suggesting:
“Unless attention is paid to addressing diversity in user involvement in

research, then it may have the effect of reinforcing rather than casting light on
existing exclusions and marginalisation – to the potential detriment of any
attempt to understand and address health inequalities”. (Beresford 2007)
From this perspective research is perceived as a form of agency and the rights of
individuals are in some sense immanent in the act of participation. I will go on to look
at some of these embedded forms of power in more depth in the next chapter. The final
point I will make here is that researchers in this study who were in support of more
active relationships with service users talked about the basic right to decline to
participate. This view was reinforced by framing participation as an issue of social
inclusion and rights. This view has been expressed by other researchers working with
people near the end of life (Agrawal 2003).
(PARTICIPANT 4 - AUS): When I see that patients, in caring for them, particularly in in- patient
units, we are very good at doing the dying thing, you know helping people to have
167
a better death. I don't think we are very good at focusing on having a better life. They don't do
much living in palliative care units. Perhaps being involved in research is one way of being
engaged still in the land of the living…If we say that a person has a right to decline to be
involved in research, they also have a right to accept.
In relation to this point, a key development in the UK is that a draft NHS Constitution
(DH 2008a) sets out a pledge to inform patients about research, thereby shifting
participation to an opt-out right. The pledge states:
“ the NHS will do all it can to give patients, from every part of England, with
any illness or disease, a right to know about research that is of particular
relevance to them and, if they choose, to take part in approved medical
research that is appropriate for them. Patients can therefore expect that a
health professional or a research professional who owes the same duty of
confidentiality as a health professional may use care records, in confidence, to
identify whether they are suitable to participate in approved clinical trials.
Appropriate patients will be notified of opportunities to join in, and will be free
to choose whether they wish to do so, after a full explanation”. (DH 2008b,
p.24)
In the case of the NHS Constitution, user involvement in research is perceived of in

terms of informed choice. In the next section I look at how concepts of responsibility
have also been applied to award meaning to user involvement in health research.
(iii) Concepts of responsibility
Responsibility is the third concept theme that significantly shapes current thinking about
user involvement in health research. Notions of responsibility are expressed though
research policy as a concern for who is responsible for what or for whom and how
different responsibilities can be enacted and assured (DH 2001c). For example, concepts
of professional responsibility such as duty of care and informed consent guide
understandings of user involvement in health research. At the same time new forms of
responsibility have emerged as a consequence of user involvement in research which
may not currently be recognized by existing accountability structures.
Related developments in patient and public involvement in the NHS have been
influenced by moves away from paternalistic practice (Coulter 1999) and towards
168
patient-centred care (Stewart 2001). However, paternalism is not simply about making
decisions about a person's best interests, it is enacted through culturally normalized
practices. Decision making power has tended to be awarded to health professions but
shifts towards patient-centred approaches at the service level are increasingly
prominent, including joint clinical decision-making (DH 1998a), multi-professional
working (DH 2001e) and patient-focused benchmarking (DH 2001i). These moves
increasingly emphasise individual responsibility for health and active citizenship.
{Trigger (App3/Date:13.05.04): Debate as part of King’s College 175th

Anniversary Seminars on Does the NHS treats patients as citizens?}
This debate, held at St Thomas’ Hospital was led by Professor Sir Ian
Kennedy, Shadow Chairman of the newly formed Commission for
Healthcare Audit and Inspection (the Healthcare Commission). A part of
the debate that I found particularly interesting was about the growing
emphasis being placed on personal responsibility for health. Sir Ian
Kennedy posed a question to the audience about whether people who
choose to smoke who develop smoking related diseases should be required
to pay for their care. This provoked a debate about responsibilities of the
state and citizens.
In the context of health research governance requirements reconfirm and support

researcher responsibility for both service users and for participants in the research. For
example responsibilities for the conduct of research and storage of research data
emphasise that researchers are responsible for attaining informed consent and assuring
data protection (DH 2001c). These requirements protect participants but at the same
time they reinforce notions of professional ownership of research and professional
power (Jones 2001a). As shown in chapter 3, the ethical issues of involving and paying
children (as subjects) in research has received particular attention (for example,
AAPCNAC 2004; Bagley et al. 2007; Dixon-Woods et al. 2006; Smith Rogers et al.
1994), these developments could both inform and complicate the involvement of
children as active participants in future research.
This is not to say that users have not entered into and changed the practices of
established research cultures by taking up professional user-researcher posts and
responsibilities for the conduct of research (Griffiths et al. 2003). However, where user
involvement is occurring in other fields, such as health service development and
169
professional education, it is not generally subject to the same processes of scientific or

ethical scrutiny. Yet other issues come into play. For example in clinical practice, such
as sexual health services, professional codes of practice (confidentiality, consent,
contact tracing) have been identified as contributing to the ‘absence of the lay voice'
(Lohan and Coleman 2005).
In summary concepts of responsibility influence the development of knowledge about

user involvement in health research but notions of responsibility differ according to the
different fields in which user involvement occurs. This could mean that knowledge
develops along particular lines according to what is deemed permissible in a particular
field.
(iv) Concepts of representation
The fourth concept them to influence user involvement is representation. Although

classifications of service users have been devised in terms of patient characteristics and
patient groups (see chapter 3) issues of representation and representativeness remain an
area of conceptual confusion and tension (Harrison and Mort 1998; Williamson 2007).
As Beresford and Campbell (1994) have shown in the context of disability “the issue of
unrepresentativeness has been used to devalue, exclude and disempower disabled
people and service users” (p. 324). This section looks more closely at the ways in
which concepts of representation can affect user involvement.
The views or opinions of a representative sample of service users can be achieved using
research sampling and research methods, for example according to population
demographics or practice populations (Involve 2005). However, it is difficult to attain
and maintain statistically generalisable samples in the context of aiming to achieve more
active forms of participation throughout all stages of the research process, most
significantly the aforementioned practical and moral issues of defining a ‘community'
from which to sample from (discussed in chapter 4). The difficulty of applying
principles of statistical generalisability to service user involvement is illustrated by a
170
comment from a professional researcher made during a discussion organised by

InvoNET:
“Public involvement could refer to the whole population – that’s a big sample
to select from”. (Appendix 3: Comment from a delegate, Ref:21.02.08)
Furthermore, the views of statistically representative samples of people may not

always provide the most appropriate form of information about health services or
health issues. For example, in community and service development projects that aim
to improve user access to health services, particular groups have been selectively
targeted to be involved in research because of their age, ethnicity or religion (for
example, Ammerman et al.2003; Corbie-Smith et al. 2003). Similarly, Jones et al.
(2004) describe the role of consumer groups in representing the collective interests
of patients, users and carers. Whilst these are familiar arguments in relation to
qualitative health research, what I am arguing here is that alternative concepts of
representation have been applied to user involvement in health research with
differing consequences.
Characterising service users as representatives has been criticised for the inference that
such individuals are representative of a group of patients or service users or that the
same people are involved more often than others. Following this line of argument user
involvement has been problematized as a question of fairness and hence upholding
representativeness has been approached as a professional duty. Williamson (2007)
argues that ‘unrepresentativeness safeguards patients collectively, since it prevents
undue weight being given to the views of just a few, perhaps isolated individuals”. Yet
approaches associated with achieving objectivity over self interest: randomization,
controlled intervention, statistical representation for example; are not easily upheld
when individual service users are recruited on the basis of their personal experiences of
a service or health condition. It is also important to recognise that the decision to
become involved is a two way process and there will be personal reasons for why
individuals may choose to become involved.
171
Others have argued that although service users cannot be representative a service user
perspective can be represented (Russell et al., 2002). Service users have suggested that
other stakeholders are not expected to be representative and that generally they have no
special mandate, so “why raise this particular argument in relation to the constituency
that is likely to be most liable to exclusion” (Beresford 2007). Through working on the
multi-method review to recruit service users from national consumer and voluntary
organisations I learnt that representation issues could be mediated by having a clear idea
about which views individuals are being asked to represent. Whilst recruiting the
service user reference group it became clear that voluntary organisations make decisions
about which studies to become involved in on the basis of the time it will take, the topic
of the research and the stage at which they are approached (Smith et al. in press)
(Appendix 11).
“It makes sense to spend more of our energy participating in our own
initiatives rather than being represented in service systems. That is where the
real promise of participation, empowerment and our representation in society
lies”. (Beresford and Campbell 1994, p.324)
This supports the view that representation issues involve the interplay of ‘self-selection'
decisions, rather than simply being an issue of whether those individuals who are
present represent ideal types.
Concerns about representation link to the concern that some service users are involved
more than others and this can mean individuals become ‘professionalized' to research
environments (Rhodes et al. 2001, Buckland and Gorin 2001). Such concerns are
reported in an account of the setting up of a consumer panel to inform the work of
National Institute for Clinical Excellence (Davies et al. 2006). Management concerns
about representation led to a situation of periodic ‘refreshing' of panel members, even
though members themselves reported feeling more able to contribute to decisions when
they were more knowledgeable of the workings of the Institute. Further problems arose
when members from Black and Ethnic minority groups were spared from de-selection
because of a concern to maintain representation of these groups. These issues illustrate
how awarding value to particular forms of representation can create discriminatory
172
practices at the individual level which risk undermining real service user/researcher
relationships.
When taken to an extreme end representation of the de-contextualized subject could

have dehumanizing effects. Attempting to identify ‘ideal' types of people runs the risk
of losing coherence of meaning, particularly about diversity issues. Similarly,
combining multiple and simultaneous perspectives could result in an unrecognisable
image of the issues.
Another important issue is that processes of representation – making copies of what

service users have said, extracting issues, quotation and so on – may have a
disintegrating effect on the originality of self. That is, involvement in research could
affect a person's view of themselves, such as their view of their personal abilities. This
point is made in a slightly different way by Beresford when he suggests that direct
forms of involvement are likely to be more meaningful and create more authentic forms
of knowledge (Beresford 2005c). Indirect, arms length forms of user involvement such
as questionnaire surveys of patient satisfaction generally comprise greater degrees of
abstraction and this may be the reason why they have been perceived as being tokenistic
(Powell et al. 2004).
It is helpful to apply Foucauldian thinking to consider what is being represented and the
signifying systems that might be applied. Foucault argued (2006b) that representation is
a socially constructed set of concepts (p.394). This is a helpful approach to
understanding representation issues as it helps to explore tensions of applying particular
concepts of representation to user involvement in health research. Using the research
literature it is possible to draw together different concepts of representation that may be
applied to user involvement (Table 7).
173
Table 7: Concepts of representation
Domains Examples Discussion papers
Forms
One person one vote, equally weighted Beresford and Campbell

Democratic representation voting
(1994)
The use of shared or weighted votes to
Proportional representation represent different groups
Could include randomization or
Statistical representation controlled samples Barnes (1997)
Representational Nominated or elected individuals
membership Barnes (1997)
represent the views of their candidacy
Representation by someone An individual is nominated or identified
who is ‘typical' of others because they are thought to share Barnes (1997)
similar experiences or characteristics as
others
Dispositional representation Individuals become or adopt Smith et al. (2008)
representative roles by virtue of their
Representation of shared job/organisational membership
interests (radicalism, Beresford and Campbell
lobbying) Self nominated representatives form (1994)
groups on the basis of shared interests
Representation of self Personal image, interests, decisions
Items/objects
Events, interventions, actions, practices,

Phenomena relationships, trends
Beliefs, expectations, views,
Perceptions experiences, opinions, desires, wants
Concepts Need, inequality, risk
Options, suggestions, recommendations,

Choices issues
Signifiers
Direct representation Individual people represent themselves

Individual people are represented by
Indirect representation
others (consciously or unconsciously)
Texts A document, text, letter or report
A speech, a verbal decision or an

Language undocumented act
174
There is also the important question of those things that are conceivable but not
representable. A modernist perspective accepts that the unrepresentable exists and this is
discussed further in the next chapter in relation to the transparency of research. This line
of thinking suggests that it is not always possible or desirable to rely on representation.
Sometimes the only way to present what is conceivable but not representable is
abstraction. In these circumstances researchers must find alternative ways of
understanding the issues at hand; for example through the use of poetry or picture
making (Schwartz 2003).
An alternative option to representation is non-presentation. In the context of health

research this could be perceived as the abandoning of service user involvement in health
research in favour of individuals making their own meaning. In this situation the rules
for what research is are deconstructed: the aura of the researcher, the environment or
institution where knowledge products are created; the techniques of process are judged
insufficient to determine what research is. Such a situation could provoke a shackle
down to defend established research methods. Although this might seem a radical
perspective, all of these elements are detectable in the history of service user
involvement and through the rise of user-controlled research. However, as new forms of
research are devised issues of representation are not escaped they are manifest in new
ways. Concepts of representation in user-controlled research emphasise representation
of self and collective interests, raising new types of issues about how individuals can
best represent themselves and identify what their interests or needs are.
A related theme that has been important to the development of user involvement in
health research is knowledge of the individual person. Foucault observed that scientific
knowledge of the person as an individual is a relatively recent modern project. In
Madness and Civilization (Foucault 2006c) and The Birth of the Clinic (Foucault 1979)
Foucault argued that the emergence of new types of knowledge, such as psychology and
psychiatry, was enmeshed in the problems and practices of power, the social
government and the management of individuals. The issue links back to concepts of
responsibility through current trends in health policy and practice which emphasise self
175
knowledge and self care. It also links to the issue of whether, following from Foucault's
assertion, the involved user is:
“the individual caught in relations of power, that creature who is to be

trained to know themselves and to willingly contribute their knowledge to the
system” (Foucault 1982).
5.5 Strategies and modes of enunciation
Strategic attempts to develop understandings of user involvement in health research

have included the creation of requirements, resourcing, guidance, principles and
education. Modes of enunciation through which knowledge has been expressed include
policy, evidence, consensus, experience and rhetoric. Policy requirements for user
involvement in health research and trends in research commissioning are significant
(chapter 2), but in this section I argue that other mechanisms have also been important.
The examples I focus on are the commissioning of research specifically about user
involvement, guidance development and the use of information communication
technology.
One way in which knowledge has strategically been developed is through resourcing.
The organisation INVOLVE has been instrumental in commissioning research about
public involvement in health and social research. The overall aim has been to show that
patient and public involvement adds quality to health research. Commissioned studies
include an evaluation of the public involvement in the London Primary Care Studies
Programme (The PC11 Project), and evaluation of the work of Folk.US, and user-led
work to identify definitions of user-controlled research (discussed in chapter 3).
The strategic development of user involvement through network resourcing provides

new opportunities for knowledge development. One example is the UK Co-ordinating
Committee for Cancer Research (UKCCCR) Consumer Liaison Group. This group
includes cancer patients, carers and representatives from cancer charities to advise the
UKCCCR on involving consumers in research; develop policy and guidance; advise on
176
recruitment of consumers to other committees/groups; and contribute to evaluation of

involvement activities.
Guidance is another strategic mechanism through which knowledge about service user
involvement is shaped and reproduced. Guidance produced by INVOLVE encourages
members of the public and researchers to work together in particular ways:
 The Public Information Pack (PIP) (INVOLVE 2007b) is made up of four booklets
and is aimed at members of the public who are interested in getting involved in
NHS, public health and social care research.
 Commissioning guidelines: A series of seven guidelines to help commissioners,
researchers and the public to think about public involvement in research
commissioning (INVOLVE 2006b).
 Guide to paying and reimbursing members of the public actively involved in
research (INVOLVE 2006c). This guide is a revision of an earlier guide published
in February 2002.
 A guide to actively involving young people in research: for researchers, research
commissioners and managers (INVOLVE 2004b).
 Involving the public in NHS, public health and social care research: briefing notes
for researchers (INVOLVE 2002a).
In nursing research the RCN Research Society have encouraged nurse researchers to
consider how they work with service users. Guidance (RCN 2007) and a supporting
webpage (www.rcn.org.uk) provide answers to ten key questions about the meaning and
practical aspects of user involvement.
There are also less formalized mechanisms which shape the development of knowledge
about service user involvement. Understandings about user involvement in health
research have been supported by advances in information technology and greater levels
of IT literacy within the general population. The Internet has provided a place for lay
knowledge creation and dissemination and a place where any capable individual can
debate the validity of professional knowledge and authority. For example, the charity
DIPEx (www.dipex.org) was founded in July 2001 by Dr Ann McPherson and Dr
177
Andrew Herxheimer following their personal experiences of illness. The organisation

publishes interviews with everyday people about their own experiences of serious
illness, health problems or health related matters to help patients, professionals and
students to learn about conditions. The DIPEx charity now funds the DIPEx Research
Group based in the Department of Primary Care, University of Oxford. Patient Opinion
is another not for profit organisation and website (www.patientopinion.co.uk). The
organisation was founded by Paul Hodgkin, a GP who wanted to find a way to make the
wisdom of patients available to NHS staff.
I will go on to show in the next chapter that to a certain extent knowledge in this field
has developed through metaphors and stereotyping of roles. Thus, knowledge has not
purely been created through empirical research or according to scientific values. For
example the prevalence of notions of ‘being heard' or ‘giving voice' (see chapter 3,
section entitled Ideology) indicate the power of metaphor. The stereotyping of service
users as individuals who contribute ‘life experience' to research also shapes thinking in
terms of the management of user knowledge through research expertise. All of these
modes of enunciation have a tendency to reinforce the notion that service user and
researcher knowledges are categorically different.
The following figure (Figure 26) uses the Foucauldian idea of a system of formation to
draw together some of the main findings from this chapter about the factors that relate
to knowledge development about user involvement in health research. It presents user
involvement in health research as an open system of formation, connected to other
forms of social practice.
178
Figure 26: System of formation of user involvement in health research

Policy-making Service management Clinical practice Education Social groups
Dispersion of elements
between formations
Strategies:
 Requirements
 Resourcing
 Guidance
 Principles
Statement Dependencies
themes Divergences (Factors allowing  Education
(Propositions, (Points of diffraction co-existence of themes)
beliefs, doctrines) between themes)
Modes of enunciation:
Alternative
language/terms  Policy
Purposes Rationales  Evidence
 Consensus
Epistemological  Experience
Processes Approaches/ stances
methods of  Rhetoric
involvement
Outcomes
Subjective
Reporting Concept themes:
experiences and
views  Knowledge
 Needs & Rights
 Responsibility
 Representation
User involvement in health

research
Conditions of emergence:
 Health care reform e.g. cost containment, micro efficiency and responsiveness to users, rationing
and priority-setting, quality standards and outcomes, plurality and diversity
 Social trends e.g. public dissatisfaction with professional practice, social movements and
charitable organisations
 Information technology e.g. World Wide Web and online discussion boards
Authorities:
 International and European law e.g. Human Rights, rights of the child
 UK statute e.g. Health and Social care Act, Research Governance Framework for Health and Social
Care
 Regulators: research funding bodies e.g. National Institute for Health Research, Research councils
and charities; research ethics committees; organisations with a remit for reporting and publication
of research e.g. research registers, research quality frameworks, journal instructions for authors
 Support organisations e.g. INVOLVE, NHS Centre for Involvement and professional bodies (e.g.
RCN, RCGP) which provide guidance
 Research teams and individuals contributing to the field
Specification:
 Definitions of health research according to governance and regulation
 Scope of health research commissioning and assessment
179
An important conclusion to be drawn from these findings is that user involvement in

health research is influenced by contemporary epistemological thinking across health
research. Yet attempting to make the study of service user involvement more ‘scientific'
may not be a fruitful way forward. Discussions about the validity of particular concepts
of knowledge, needs and rights, responsibility and representation are clearly essential
for informing how future knowledge is developed. The final section of this chapter
reinforces this argument by looking at how carer- and user-led research can become
hidden with the current episteme.
5.6 Hidden carer- and user-led activity
This section gives specific attention to the issue of carer-led research. It draws from an
interview with a policy lead at a state-wide carers' organisation (held on 12th November
2005). The interview is included here because it illustrates many of the issues identified
in this chapter about how particular knowledges about user involvement in health
research may be construed in different ways and valued for different purposes. In
particular the interview explains some of the reasons why an apparent absence of carer-
led research activity may in fact be an inaccurate picture.
I began the interview by stating that I had found relatively few publications specifically
about carers being involved in research. The participant suggested this was true in the
case of academic research and that carers have tended to be considered to be a subgroup
of consumers, but that this situation was changing. In his experience academic research
interests were beginning to focus on understanding the experiences and views of carers
as a group of people in their own right.
(PARTICIPANT 6 - AUS): Academics are only recently discovering carers in a significant scale,
and most of the research to-date has been about carer burden, the burden of care, that has been the
dominant focus…There has been a tendency to subsume those [carers] under the heading of
consumers. The consumer movement has matured a bit and if you frame your arguments well,
people can understand how carers are not actually a subset of consumers.
180
The participant provided a strong and convincing argument for the views and
experiences of carers to be included in health research based on the contribution of
carers to health and social care systems.
(PARTICIPANT 6 - AUS): The strongest part of that argument is to emphasise that they are the
major community care workforce, which is a much different thing than being a consumer. It puts
a different context on involvement in research.
Although it may be economically and politically advantageous for carers to be

perceived as providers of care rather than consumers, carer organisations are now more
engaged in political decision making and research.
(PARTICIPANT 6 - AUS): There are carers driving research and participating in research… This
organisation has done an enormous amount of research over the years, some of it academic, a lot
of it is more practical. There are different agendas here. The agenda here is not about getting
peer-reviewed papers published. It is about generating change. So a lot of the research is of a
different nature, different scale, it doesn't have kind of an academic focus.
The participant further emphasises the volume and significance of research activity
going on within carer organisations. He describes carer-led research as being hidden
because of the way research is often conceptualized.
(PARTICIPANT 6 - AUS): Academic research, and drawing on academic sources of information

about this [user involvement]… It is very limiting. It's not even half. It is a particular lens or map
of the world and it is such a small part of the world. There is so much else going on around today.
Some academics are interested but for the most part it is not on the map.
Conceptualizing user involvement in health research solely in terms of publication

criteria or scientific values runs the risk of excluding carers' contribution to the
development of knowledge.
(PARTICIPANT 6 - AUS): … what I am keenly aware of is if you conceptualize it differently

you actually see a lot of practice in which carers are driving research agendas and engaging in a
lot of doing their own focused research...
The participant provides an example of the type of knowledge that might be overlooked
from a scientific view of evidence. He describes this form of knowledge as ‘hands on'
and pragmatic. The discussion reveals different views about who research is for and that
the needs and interests of carers are district from other groups of service users or health
researchers.
181
(PARTICIPANT 6 - AUS): A lot of it is hands-on pragmatic stuff that is never going to be

published in an academic journal, so it is off the academic radar, but all that sort of stuff is driven
by knowledge of what carers want and need, and that is research. So there is a tremendous
amount of stuff around but it is probably more what the academics would call the grey literature
… And it is not particularly of interest to academics because they haven't been involved in it,
they haven't driven it.
In another respect the organisation itself can be perceived to be the end-user of research.
Examples provided included using research as a way of informing organisational
development, carer education and development of career services. Organisation-focused
research is valued by the participant for having benefits for volunteer members in terms
of skills development and motivating interest in particular issues.
(PARTICIPANT 6 - AUS): Sometimes it is just about agenda setting and keeping people
interested, skilling up researchers, you know there is a whole bunch of other reasons that you do
research other than to create a product, like process outcomes.
The participant then explains the nature of the relationship that the organisation has
with the academic and independent research community. He argues that carer-driven
research includes research that is commissioned by carer organisations and provides
examples of where independent researchers have been funded to undertake work on
behalf of the organisation. This corresponds with the previous point that some research
is carer-focused and some research is used to influence decisions that impact on carers
(the participant describes this as the ‘academic side'). He feels that direct carer
involvement is less likely to occur in the latter. In his view, it is the organisation's role
to act on behalf of carers through prioritizing and commissioning processes. The
participant argues that because the organisation is carer-driven the research it
commissions is carer-driven. This indicates a complex pattern of involvement, which
incorporates decisions about direct involvement, indirect involvement (conveying or
presenting the views and perspectives of carers), and advocacy on behalf of carers.
(PARTICIPANT 6 - AUS): Some sort of carer-participation or carer-control it is probably pretty

limited in the academic side. There are sort of two different avenues. But what happens, on the
academic side is we often contract academics to do the work and we give them an idea of some of
the carer-driven research… But nobody says – this is carer driven research. But it was
commissioned by carer advocates and we have members of the organisation who are carers. So it
is very much carer-driven research. But there is nothing that says here is a list of carer-driven
research, it is subtler than that.
182
The following extract indicates that the notion of an entirely carer-driven research
agenda at an organisational level is idealistic. It is accepted that most organisations will
undertake research because of funding opportunities or partnerships.
(PARTICIPANT 6 - AUS): It is like most organisations, it is a mixed approach. Some of it is

opportunistic, opportunities arise, or funding arises, so you just do it. In other instances people
come to us and they want to do something on carers and they need access and typically we can
provide access. So there is a lot of stuff that is done in partnership that might not be purely carer-
driven, it might be done in partnership.
This view suggests that within research partnerships this organisation can fulfil an
intermediary role in providing direct access to carers. However, the participant goes on
to explain how the organisation itself is a source of information about the issues and
problems that carers are experiencing, built up through direct communication with
carers.
(PARTICIPANT 6 - AUS): Also, because of the services we run we have got extensive data of
what is going on and we have got a helpline, we can bring up a list – what have carers rung about
in the last 12 months that they need help with, which will tell you the current issues and trends.
Research is perceived as one way of substantiating ground level issues in policy

debates. Particular examples of issues were to do with young carers and older carers of
disabled children.
(PARTICIPANT 6 - AUS): Obviously we want to achieve things at a policy level and although
research isn't the be-all it can be really useful for leverage. So we will seek out research that has
been done or maybe seek out opportunities to get some done.
Another way of connecting carer concerns with decision making was through member's
own personal experiences of caring. Indeed, personal experience was seen as a
characteristic and strength of the organisation that might not always be recognised.
(PARTICIPANT 6 - AUS): Keep in mind, in an organisation like this, probably more than fifty
percent of the staff here are carers. Even though they might be professionals, may of them are
here because of personal commitments to caring. A lot of people have ended up here because of
personal experiences and an interest in doing better for carers.
In his opinion, to be credible at a political level consumer organisations need to show

they are closely connected to consumer issues and that their membership reflects those
they represent. The participant enforced this point by recounting concerns that had
encountered whilst working for another consumer organisation in the district.
183
(PARTICIPANT 6 - AUS): I went to work there… that was an interesting organisation because
they didn't claim to represent consumers they claimed to advocate on behalf of consumers. The
issues of having a constituency were practical and different people put different views forward
about this, but the arguments were that they are less connected to the consumer movement if they
advocate for consumers or on behalf of consumers. To claim that they had constituency through
the membership is a bit slippery at a political level.
In the case of carer-research, the participant suggests personal perspectives can be

limited and research evidence is important for validating what people already know and
for gaining greater leverage in decisions or for securing funding.
(PARTICIPANT 6 - AUS): There is so much that people might actually know, just in terms of
general experience, that they find it is actually quite difficult to generate change without some
research to sort of back up the claims. You know another part of the picture is the evidence-based
movement. That is driving people's view that you need research because you need evidence. It is
pretty recent but more and more people are saying – what is your evidence for that intervention
works better than that one. Without that in the system then everyone is just making assumptions.
We are here in the middle of putting together our pre-budget submission for the State and they
want to see some evidence. You know you can't just make claims about this, that and the other.
You can, but you won't get any mileage. It has to be based on evidence.
Taken together these statements indicate that this particular consumer-led organisation
engages with, uses and undertakes research in multiple ways for different purposes.
Different research activities might be focused on the needs of carers or on the
organisation supporting carers. Alternatively research might be used to inform or
influence research funders or policy makers. Carer-driven research might be perceived
as research that is undertaken by carers but it may also describe the process of
connecting carer concerns with decision-making. The views expressed by this one
participant here illustrate that user involvement in health research is, in part, based in
the collective actions of individual people (including a diversity of carer, patient,
service user and public groups), consumer and voluntary organisations and lay
networks.
Understanding the complex function of research in relation to any one organisation can
provide a basis for understanding how carers are or can be involved in health research
organisations and systems.
184
Summary
 This chapter looks at how knowledge of user involvement in health research has
developed, focusing on the concepts that have been applied.
 Concept themes infuse and shape notions of user involvement in health research.
These include concepts of knowledge, needs and rights, responsibility, and
representation. These themes are variable and manifest interdependently of any
specified field of practice or knowledge. User involvement in health research has
been influenced by concepts that are present within political and scientific
discourses and tensions have arisen because of problems with applying scientific
values (such as methodological rigour) and political values (such as democratic
representation) to it.
 Scientific discourse influences thinking about knowledge as evidence in user

involvement in health research and there are concerns to develop an evidence-
base. At the same time the impact of views and opinions of service users cannot be
easily predicted or evaluated according to scientific principles. Furthermore it is
difficult to attribute any outcomes of the research, or for healthcare practice,
directly to the involvement of service users.
 Concepts of professional responsibility, for example duty of care and consent, are
present in understandings of user involvement in health research. At the same time
new forms of responsibility have emerged as a consequence of user involvement
in research which may not be recognized as being part of research practice. For
example, ensuring service users receive necessary information and support to
participate in research.
 The theme of representation has been applied to user involvement in health

research. The application of particular notions of representation, for example
statistical representation and democratic representation, can create situations
where user involvement is criticised for being unrepresentative or unfair.
185
 Knowledge in this field has developed through strategic mechanisms such as

policy making, resourcing, guidance development, formal and informal
networking, as well as through metaphor and stereotyping of roles. Thus,
knowledge has not purely been created through empirical research or according to
scientific values. For example the prevalence of notions of ‘being heard' or
‘giving voice' indicate the power of metaphor. The stereotyping of service users
as individuals who contribute ‘life experience' to research also shapes thinking in
terms of the management of user knowledge through professional expertise and
the notion that service user and researcher knowledges are categorically different.
 Discussions about the validity of particular concepts of knowledge, needs and

rights, responsibility and representation are essential for informing how future
knowledge is developed otherwise there is a risk that knowledge from carer- and
user-led research may go unrecognised within the current episteme.
186
Chapter 6: Tensions
6.1 Introduction
The emphasis of this section is on broadly identifying how tensions associated with user
involvement in health research might be better understood. The argument put forward
here is that theories of power can help to bridge tensions between ideals and the realities
of developing service user/researcher relationships. The tensions explored here were
identified in the analysis of the research literature (Chapter 3). These were:
1. That user involvement in research must be supported by policy makers, research

commissioners and the research community, but at the same time it should not be
‘top-down' or controlled by professionals.
2. That there should be more transparency in research, particularly around processes of

decision making, but at the same time there are uncertainties about what the
underlying intentions and interests are.
3. That service users should have ‘equal' say, but at the same time professional ‘know-
how' and the interrelated nature of power mean that service users can never gain an
equal footing in research systems.
These tensions are examined and illustrated by drawing on the study data. Preliminary
analysis of the data (see chapter 4 for details) revealed ten key themes relating to service
user/researcher relationships. These themes could be broadly classified as corresponding
with three types of statements about:
 Purpose - the reasons given for user involvement in health research, including
themes about: motives, incentives, purpose, expectation and resistance, non-
engagement.
 Process - how user views should be gained and used, including themes about:
engagement, contribution and communication.
187
Chapter 6: Tensions
 Outcomes - statements about the impact of user involvement, including themes

about: gain/benefits, drawbacks/losses, evaluation/measures, history/memory,
and change over time.
Rather than discussing the data according to these themes, I have chosen to draw
selectively from the literature on social theories of power (making use of the power
theory map developed in chapter 4) to consider how theoretical perspectives might help
to explain the reasons behind the three tensions identified. Issues of how these tensions
might be experienced are also important and this is given specific attention in the final
section of the chapter.
6.2 More professional commitment ↔ Less professional control
There are political advantages to encouraging closer relationships between researchers,

clinicians and the public. A plethora of policy documents support public involvement in
health and social care service development, education and research. Because policy
documents express multiple rationales for public involvement researchers have
expressed scepticism about what the underlying intentions for user involvement are. It
may be that professional and public trust has more readily been lost than gained because
of this apparent lack of clarity. An associated point of tension is the presumption that
greater public consultation can achieve social consensus (Irwin 2006). Politically driven
public consultations have been criticised as a way of moderating public expectations
and using selected public views to support predetermined policy decisions, all at great
public expense (Davies et al. 2006).
Despite these tensions research commissioners have been encouraged to align with
policies for involvement and to show greater commitment to user involvement within
their own structures and systems as well as in the research they commission. The work
of the INVOLVE Support Unit has included producing guidelines for commissioners,
researchers and the public on involvement in research grant applications, peer reviewing
applications and research commissioning (INVOLVE 2006b) and supporting an online
188
Chapter 6: Tensions
discussion forum (www.invo.org.uk/forum). A new web based resource supported by

INVOLVE and the UK Clinical Research Collaboration (UKCRC), ‘People in
Research' was set up to help members of the public make contact with organisations
that want to actively involve them in clinical research (www.peopleinresearch.org).
These activities indicate a strategic attempt to strengthen commitment to user
involvement at a research commissioning level. Hence major UK health research
commissioners, including those under the umbrella of the National Institute for Health
Research (NIHR 2008), are now more actively supporting and encouraging user
involvement in the research they commission. For example, since 2000 the Medical
Research Council (MRC 2008) has used a number of mechanisms for promoting public
involvement in its research and policy-making activities, including an Advisory Group
for Public Involvement which provides advice on ways in which the MRC can involve
the general public and patients in its activities.
INVOLVE's targeting of research commissioners was carried forward in July 2008 in a

one day workshop for invited consumers, commissioners and researchers with
experience of consumer involvement in the commissioning process. Common themes
raised during the workshop were: the language of commissioning, payments and the
financial implications of involving members of the public, support for everyone
involved in the process, and the need to ‘close the loop' by informing members of the
public about the outcomes of the research (INVOLVE 2008).
It has been argued that in the main health researchers have engaged with ideas about
service user involvement because not doing so will harm their chances of being awarded
research funding (Beresford 2005b). This argument is substantiated by the fact that
where commissioners request details of user involvement in research proposals it
appears more likely to occur (Wickham and Collins 2006). Knowing-how to make
selective use of user involvement is an example of what Foucault refers to as savoir
(Foucault 2000, p.266) (discussed later in this chapter). This type of knowledge falls
short of rigorous scientific process but commands ratification within a social group, for
example through knowing what to write about user involvement when making grant
applications. Know-how confers recognized instrumental benefit and researchers will
189
Chapter 6: Tensions
sometimes seek to utilize models or service users who have participated in other studies.
Whilst this is not necessarily a bad thing the danger is that user involvement becomes a
perfunctory part of the research process, closing down opportunities for knowledge
development.
{Trigger (App3/Date:19.06.07): Supervision, discussion of user involvement}

When thinking how user involvement is constructed, a sceptical, but
perhaps accurate view is that researchers involve who they want, how they
want, when they want, and as long as these people are not paid employees
of a research organisation this can be called user involvement.
Research on user involvement in the planning and delivery of health services seems to
support this view. For example in the context of mental health service provider Trusts in
London, Rutter et al. (2004) show that user involvement remains in the gift of
professionals: managers retained control over decision making and expected users to
address Trust agendas and conform to Trust management practices; in contrast users
wanted to achieve concrete changes to policies and services but had broader aspirations
to improve the status and condition of people with mental health problems. Taken
together the aforementioned issues point toward an underlying problem with diverse
expectations about what service user/researcher relationships should comprise.
More recent accounts of user-researchers being employed in academic settings who

develop and lead projects (Griffiths et al. 2003; Rose D. 2001; Simpson and House
2002) challenge the view that in practice user involvement is professionally-led, but it
may still be the case that it is led by the professions. Some views expressed through the
study data support suggestions that policies for user involvement are a threat to
professional and service user's confidence in the process of involvement. The speed of
policy change may be the cause of unrest as both service users and researchers said they
need time to negotiate issues and devise new ways of working.
{Trigger (App3/Date:23.08.04): “Learning from the patient experience”

Conference: Open-floor discussion}
During the discussion several researchers employed within the NHS
articulated a belief that policies for public involvement were devised and
enforced ‘top-down’. Individuals appeared to be generally sceptical about
policy implementation in health service organisations. Although the
190
Chapter 6: Tensions
political rhetoric is that public involvement is a good thing some people

clearly think that the policy is to the detriment of quality research and
health services.
Another view put forward was that there is a political idealism about user involvement,
but that service users do not expect or want a role in many aspects of health research, or
indeed health service decision-making.

The current healthcare climate is politicized and performance driven. The
aim is to raise standards but there is a growing concern that the rush to
involve patients and the public is raising false expectation, eroding
professional confidence and devaluing the impact of individual patient-
professional relationships … The challenge is to achieve an effective
relationship that uses lay perspectives appropriately and maintains
professional confidence. The debate finished with consensus that user
involvement should not be used to achieve political objectives.
However, other participants suggested that further policy was required to encourage
professional commitment, suggesting that existing performance structures
(commissioning and target setting) could be used to encourage greater commitment
from researchers, NHS managers and health educators.

There was a lengthy debate about how public involvement could be
enforced. Some individuals thought that until commissioners state that
patient and public involvement must be part of all aspects of health service
development there will be no impact. Their view was that the new agenda
for PPI needs to be part of strategic health authority performance
indicators. There was some concern that it takes specialist training to
involve members of the public, particularly in project planning and that
staff often did not recognise practical or ethical issues before starting work.
A consensus point was that more support was required in the form of
professional training for public involvement.
The notion of a policy push in the UK was also expressed by participations from
Australian academic and health service provider organisations. In Australia the
requirement for acute sector providers to set up a Consumer Advisor Committee as part
191
Chapter 6: Tensions
of Clinical Governance legislation has been influential. The Consumer Focus

Collaboration was established in 1997 as a national organisation for consumer
involvement. This led the way for the National Resource Centre on Consumer
Participation and the Health Issues Centre.
{Trigger (App3/Date:21.10.05): Discussion group with Australian health care

researchers}
Participants from Australia debated the difference between the concepts of
‘patient’, ‘service user’ and ‘stakeholder’. In the two group discussions that
I conducted there was a strong view that different stakeholders should be
involved in decisions about health research. Ownership of research
agendas and who contributes to the research process were raised as
important issues. Individual participants in the interviews had said that
research can be a powerful way of motivating change. One discussion
group described the challenges of deciding who stakeholders are and
experiences of engaging with community members, commerce or industry,
private health care providers and charitable organisations. Not all of the
participants in the discussion felt that patients or community
representatives should always be involved in decisions about health
services or research. Both groups agreed there was a need for research
policy at a national level across Australia to provide guidance about
involving different stakeholders in different types of research.
The notion of top-down control over user involvement tends to corroborate a view of
power as being something that can be possessed and used at will. Such a view is
founded in the new science of mechanics pioneered in the seventeenth century. It is a
view that has been associated with Thomas Hobbes' 1968 conception of sovereign
power (Warburton 2006). A top-down perspective of user involvement contains
positivist themes which construct power as phenomena that can be measured and
observed. From this perspective to be powerful in research means negation of the power
of others by a predetermined ‘locus of will'. This particular view of power reflects
current concerns with measuring the impact of user involvement on research. Power is
perceived as being ‘zero-sum', that is it is finite and can be possessed (for example in
the form of decision making power) and passed between individuals. The central
problem with a top-down causal theory of power is that it rests on what can be admitted
or attributed to be a cause and how it could be said to be so. This was highlighted by
192
Chapter 6: Tensions
Dahl's (1961) work on behaviourism in his study of New Haven: Who Governs? When
applied to user involvement in health research such a view does not wholly explain how
individuals exist amongst each other, what people think and what they intend to happen
(Weber 1978).
Another way of understanding tensions about control in user involvement to look at the
positions which powerful people occupy. Wrong (1979) suggests that power can be
dispositional and that it can therefore be equated with a set of capacities. For example
the credentials and experience required to become a researcher. Power could be an
ability to achieve goals (Parsons 1967). The view that power can be dispositional
(Parsons 1951) challenges the idea that user involvement in health research has a top-
down origin or that control is always expressed through predetermined acts. It also
draws attention to the existence of less obvious forms of power, such as the pre-
established social structures that frame new forms of relationships between service users
and researchers. Indeed a strong message coming through the study data is the notion
that professional's expectations that patients should fit into existing professional
frameworks leads to tokenism.

Several participants in the discussion expressed concerns about tokenism in
relation to their own practice. This was described as functioning in at least
two ways: the expectation that individuals should be responsible for an
entire organisation’s involvement activities; and concerns that commitment
to user involvement in research may be short-lived if political priorities
shift.
{Trigger (App3/Date:07.10.06): INVOLVE Conference: Group discussion}

The issue of tokenism was raised at least three times during the afternoon
by presenters and members of the audience. Individuals used the term to
mean a negative professional trait (being tokenistic), an unacceptable
approach to involvement (being tokenistic), and a phenomena related to
professional outnumbering at meetings (token service user).
{Trigger (App3/Date:21.09.06): ‘An informal conversation about user

involvement in health research’}
One of the participants in this conversation suggested that on the
conference circuit (in wound care) patients are largely absent. Those who
193
Chapter 6: Tensions
do participate seem to be involved by professionals in a tokenistic way, for

example presenting at the closing session of a conference rather than being
invited to attend the whole event.
Tokenism in user involvement has also been thought of as meeting political targets by
constructing a token role for service users or restricting numbers of service users
involved, thereby restricting user involvement to ‘tick box exercise'. This definition of
tokenism is illustrated by the following quote from Professor Bob Sang, a senior
academic researcher in this field:
“Tokenism is the dark side of pluralism. If someone says ‘they’re responsible

for our work with consumers’ – then it’s in a box, and that’s tokenism. If you
hear anyone say ‘My committee has two consumer members’ then tokenism is
operating” (Standing Advisory Group on Consumer Involvement 1998a,
p.46)
Although all researcher participants interviewed in this study had experience of

undertaking research with patients or carers as the subjects of research, few had
experience of involving service users at a development stage, supporting previous
reports in the research literature (Staniszewska et al. 2007). Judgements about the
purpose of user involvement were therefore generally made by professional researchers.
Some participants did not perceive a distinction between defining purpose and the
process of involvement or what implications this might have for the way they worked
with service users. Others perceived patient's experiences to be data within the research,
seeing themselves as an intermediary between patients and the research process. Other
participants felt that it was more important that researchers and service users develop
relationships and understanding through the research process.
Participants approached professional commitment to user involvement as an issue of

getting the practical aspects of engagement right. Particular issues raised by participants
were ethical processes and payments. Most of the researchers in Australia that took part
in the interviews were undertaking research with patients near the end of life, people
with mental health problems or learning disabilities. Participants expressed strong
concerns that because these groups are often perceived to be vulnerable they may not be
given the opportunity to contribute to research. In relation to end of life care,
194
Chapter 6: Tensions
researchers felt that it was difficult to gain ethical approval because members of ethics
committees were concerned about burdening patients or that research might affect the
nurse-patient relationship. In the account of one project given below the participant
explains that some ethics committees support research with patients receiving palliative
care whilst others appear less understanding and paternalistic.
(PARTICIPANT 4 - AUS): A number of ethics committees, one in particular, were

extraordinarily paternalistic about me bothering dying people with research. Their attitude seemed
to be that I was insensitive and presumptuous to think that dying people would want to be
involved in research… Another project that I did about depression in palliative care, which is a
very prevalent problem… I got one ethics committee that was so outranged that dying people
would be asked if they were suicidal, which is a standard question about depression...They are so
paternalistic that they almost treat the person as if they are already dead.
Other researchers confirmed the view that it was particularly difficult to initiate
research about particular social or health issues because individual members of ethics
committees may hold strong moral beliefs.
(PARTICIPANT 4 - AUS): I have a colleague here who is in research with mad pregnant women.
One ethics committee suggested to her that it was totally inappropriate that mad women were
allowed to get pregnant... It was a dangerous statement to make in a public forum like that … It
was one strong voice but nobody disagreed. You know I think that when a statement is made and
no one disagrees they are complicit in a statement.
Other challenges were perceived as including concerns about burdening patients,

communication between the researcher and participants, and the different priorities of
researchers and health care staff.
(PARTICIPANT 2 - AUS): There is a reluctance to give researchers access to patients in any of

the health care organisations, you know the public health organisations. And there are very good
reasons for that. First of all they don't want them inundated with researchers because it's
exhausting. They are in hospital because they are acutely ill, that would impose a burden, I think,
on them. I think the other thing is that there is competitiveness in terms of access between
research teams. I found that on the stroke study. We were all sort of targeting the same
population group.
A more general issue participants raised in relation to barriers to involvement was the
issue of payments. Payments have been shown to be important for enabling
participation (Trivedi and Wykes 2002; Baxter et al. 2001), and this is now part of
research policy (DH 2006). Some participants from Australia felt strongly that being
able to provide payments was desirable but unrealistic because few research
commissioners recognise this to be a necessary part of research funding. The cost of
195
Chapter 6: Tensions
funding language translators was also seen as a barrier to enabling certain groups of the
community to participate in research. Receiving payment and being treated hospitably
can be a way of feeling valued and acknowledged however service users also say that
they have gained from their involvement in research in other ways, including improved
confidence, skills, knowledge, networking and friendship, indicating that these benefits
may be important for keeping service users involved and should be perceived as
positive outcomes of user involvement (see section 7.2 for discussion of the issues).
Career opportunity also appears to be an important deciding factor for whether service
users remain involved. Recent increases in the number of employed user researchers
within research organisations has contributed to the idea that service users are
embarking on a career rather than making a time-limited contribution to a particular
project or department (Griffiths et al. 2003). Through job roles service users are
acknowledged as contributing knowledge, skills, links to networks, insights, view points
and ways of thinking that might otherwise not be gained.
Surfacing and exploring these issues helps to show that professional commitment in the
form of fulfilling ethical requirements, finding resources for involvement,
communicating ideas and encouraging service users to become involved is important,
but these activities could also be perceived as reinforcing professional control over
research or leading to tokenism. The next section explores tensions associated with the
transparency of research and understanding issues about what the underlying intentions
and interests are.
6.3 Increased transparency ↔ Uncertain intentions
An issue closely associated with user involvement is the perceived need to improve the
transparency of research. User involvement has emerged in the context of concerns
about safety and quality in health and research (see chapter 2). This concern is
reinforced by a general yet significant view that public trust and public participation in
research demand transparent and accountable oversight of research (Knoppers and Joly
196
Chapter 6: Tensions
2007), as well as better understandings about public understandings of science (Michael

2002).
More specific concerns have been raised about the reporting of health research,
particularly that some research is never published or is poorly reported. Guidelines have
been developed with the aim of improving the reporting of randomized controlled trials,
for example CONSORT (Consolidated Standards of Reporting Trials) (Altman 1996)
and the publication of qualitative research (Elliott et al. 1999). The establishment of the
EQUATOR (Enhancing the QUAlity and Transparency Of health Research) Network
(http://www.equator-network.org) seeks “to improve the quality of scientific
publications by promoting transparent and accurate reporting of health research”. The
use of guidelines by authors and journal editors has important implications for the
reporting of user involvement in health research, an issue discussed later in chapter 7.
On another level, the issue of transparency has been approached as a matter of
providing more information about decisions that have been made through the research
and the processes that have been used to reach decisions.
“it is the researcher, who, within certain confines of funding (which are
clearly not important), by defining the research, has power in relation to
‘those researched’. This means, for example, that decisions not only about the
conduct of the study, but also the write-up, the analysis, and what is
determined to be the peripheral or irrelevant lie, in the main, with the
researcher. The power of the researcher is, however, not always noted or
analysed – precisely because the power is transparent” (Bhavnani 1990,
p.141)
Hence, in user involvement, transparency of research is not only about reporting what
has been done or found, but questioning what decisions have been made (or not made)
about health research and who has made them.
Some researchers have suggested that user involvement in health research is

unachievable because the public are apathetic towards research and those who do
become involved are driven by a personal agenda, payments or bribery (as discussed by
Boote et al. 2002). However these views were not generally expressed by participants in
this study and there was a strong feeling that members of the public are enthusiastic
about wanting to help. Participants did talk about the challenge of understanding the
197
Chapter 6: Tensions
motivation for different people to becoming involved and how a better understanding
this could be helpful in raising awareness, providing different people with information
about involvement initiatives and addressing issues to enable them to be involved.
Not attempting to understand these motives could be perceived as one way of resisting
user involvement through non-decision making. Bachrach and Baratz (1962) suggest
that, first the powerful may not attend to, may not listen to, or may not ‘hear' demands
articulated by the less powerful. An example of this could include situations where
topics for research studies are selected and defined by professional groups without any
possibility for public involvement. Or, if demands do gain admission to the decision
making agenda they may be sequestered or quashed. It was not possible to find any
examples of this happening in the study data. It may be that service users choose not to
engage in studies where they have little say over the topic of the research or that such
events are ever documented.
The second way in which non-decision making may operate is through anticipated
reaction: issues are not raised because it is anticipated that they will be opposed or
ignored (Bachrach and Baratz 1962). In the context of health research this could explain
why large scale public consultations about health do not often receive good response
rates, provoking questions about the value of such processes (Davies et al. 2006).
A third way in which non-decision making may occur is by means of the mobilization
of bias. In this situation dominant interests exert such a degree of control over the
values, beliefs and opinions of less powerful groups that they determine which demands
come to be expressed or indeed perceived. A current example in the context of health
research is political debate about laws on embryonic stem cell research. The debate is
framed by the expectation that scientific understanding of stem cells can be translated
into regenerative medicine, yet research with biomedical scientists suggest that
institutional factors and stem cell science itself are major barrier to therapies
(Wainwright et al. 2006). Bachrach and Baratz (1962) describe these phenomena as the
‘hidden face of power'. Power can produce a situation where there is little or no
198
Chapter 6: Tensions
behaviourally admissible evidence of power being exercised, but none the less, power is
pervasively present.
Lukes' (1974) ‘three dimensions of power' depicted in figure 27, provides a useful
framework for understanding how decisions about user involvement in health research
can be made (or not made) consciously or unconsciously.
Figure 27: Lukes’ three-dimensional framework of power (Clegg 1989 p.90)
Three-dimensional view
incorporates the first, second and a third dimension
Two-dimensional view incorporates the

first and second dimension
One-dimensional
view incorporates
only the first
dimension
Key elements First dimension Second Third dimension

dimension
Behaviour Interpretive Evaluative

Objects of analysis understanding of theorization of
intentional action interests in action
Concrete decisions Non decisions Political agenda
Issues Potential issues Issues and

potential issues
Indicators Overt conflict Covert conflict Latent conflict
Field of analysis Express policy Express policy Relations between

preferences revealed preferences express policy
in political embodied in sub- preferences and
participation political grievances ‘real interests'
199
Chapter 6: Tensions
Lukes' third dimension of power is perceived as something that is unknowably lodged

in one's subjectivity and consciousness. This conceptualisation has been used in
Marxian debates (Marx 1976) about social control through hegemony. The concept of
hegemony has been raised in relation to user involvement (see chapter 7, section 7.2).
This view of power suggests that at its extreme power operating through user
involvement in health research can:
“prevent people, to whatever degree, from having grievances by shaping their

perceptions, cognitions and preferences in such a way that they accept their
role in the existing order of things, either because they can see or imagine no
alternative to it, or because they see it as natural and unchangeable…”
(Lukes 1974 p.24)
However, Laclau and Mouffe, influenced by Foucault (2000, p.119), suggest that rather
than thinking of either ideology or hegemony as a state of mind:
“one would better regard it as a set of practices, primarily of a discursive

provenance which seeks to foreclose the indefinite possibilities of signifying
elements and their relations, in determinate ways” (Laclau and Mouffe 1985)
This understanding of power suggests the practice of user involvement rests in the
thoughts and acts of individuals. Hence, attention may need to turn to understanding
how people make judgements about user involvement.
If power in user involvement and health research can be hidden within non-
decisions and normalized social practices this raises the question of how to access
the intentions behind actions. The issue of intention has particular resonance for
debates about what the underlying intentions for user involvement in health research
are (an issue explored in the next chapter). The work of Bertrand Russell (Russell
2004) draws a link between power and intention by suggesting that power is based
on the number of intended effects that a person achieves. Like Russell, Max Weber
(1978) defined power in terms of intention, as:
‘the probability that an actor within a social relationship will be in a position

to carry out his own will despite resistance, regardless of the basis on which
this probability rests” (Weber 1947 p. 152).
200
Chapter 6: Tensions
Although Wrong (1979) widened this definition to “the capacity to produce intended
and foreseen effects on others” this overlooks any unintended effects of social
interaction. In the context of user involvement, this could be a particular weakness as
often the most significant outcomes are unforeseen or unintended (Baxter et al. 2001;
Boote et al. 2002; Smith et al. 2005b). Explanations and outcomes of user involvement
should not simply be taken at face value or without consideration of the context in
which they were achieved. The implications for measuring the cause or effects of user
involvement are discussed in the next chapter (see section 7.2).
A further problem that stands in the way of conceptualizing or measuring user

involvement in health research is the notion that interests are formed on the basis of
agreed moral principles (Clegg 1989, p.10). Put another way, intentions are not fixed
and there are moral dilemmas about separating interests from wants and preferences,
and who defines what is in a persons best interests (Benton 1981). In both Lukes'
model, and user involvement hierarchies, power is positioned as a function of moral
intent. That is, individuals will strive to make decisions on the basis of established
moral codes. However, as Benton points out it is extremely unlikely that any one
decision will suit everyone involved. An understanding of how power operates in user
involvement in health research should therefore seek to encompass the messier
negotiations and interceptions between individuals and the types of interest-positions
(personal, relational and social) that underpin different types of decisions. These issues
are explored later in chapter 7 in relation to judgement making (Section 7.6).
6.4 ‘Equal say’ ↔ ‘Know-how’
Issues about processes of user involvement in health research often centre on the notion
of voice (see chapter 3) and the principle of equal say. However, in their evaluation of
the Citizens Council for the National Institute of Clinical Excellence Davies et al.
(2006) show that alternative views about what equal means can be a source of tension.
In the present study participants suggested that hearing service user's ‘stories' is a
valuable way of enabling professionals to understand the impact of illness and the
201
Chapter 6: Tensions
patient's experience of health services. Involving service users on research committees

and advisory groups was perceived as a way to bring real life perspectives into
professional worlds.
(PARTICIPANT 1 - AUS): Sitting on committees, where you have consumer advocates and
consumer advocates, people giving their perspective. It is fundamental that they be there because
you don't realise as a professional how you can be stuck in a linear sort of groove where you don't
actually see it from someone else's perspective… Often we get, particularly in medical research,
and nursing research too, we get caught up in what's important to us and it may have nothing to
do with patient care … Patient advocates are really the people, who represent, and are articulate to
represent, another point of view; they are always going to be relevant. Particularly, they keep the
profession from being narrow minded and not too linear in the way they approach things.
The work of Jürgen Habermas is useful for understanding this view of user
involvement. Habermas (1978) was concerned with distinctions between ‘life' and
‘system' worlds in society. He argued that in order for society to function to best effect
there needs to be a balance between the erratic and creative life world and the rational
system world. One example of how this mechanism my operate through user
involvement, is that it is now fairly common for service users who sit on boards or
committees for national research organisations to be invited to present a session at
health conferences. The intention is to bring the ‘lived experience' of health or illness
into professional arenas.

Conference: presentation by a service user on being involved in cancer research
network}
Service user presenters tend to justify their presence to the audience with
phrases such as ‘I am here to share my lived experience’, ‘My world is
different to your world’. The presenter of the afternoon session began by
saying “I am presenting the patients’ perspective”.
Habermasian perspectives position communication as the only way of reaching mutual,

shared and uncoerced understandings. These perspectives have been used to examine
lay health knowledge (Williams and Popay 2001). In the context of user involvement in
health research, attention to the conditions under which people are able to communicate
their views and knowledges – an ‘ideal speech situation'- could transcend problems of
basing definitions of power on interests (Clegg 1989). That is, focusing on improving
the conditions of deliberation may be more advantageous than debating who defines
what is in the best interests of service users.
202
Chapter 6: Tensions
{Trigger (App3/Date:12.08.07): Supervision, discussion about use of literature in

the study}
The idea that researcher power simply comprises access to resources
provides inadequate explanation of what power is and how it is
experienced – unless a definition of resources includes knowledge,
language, finances, education, experience, age, social class, gender, ethnic
background, beliefs, outlook/personality and so on. Habermas’ ideas about
deliberative inequalities suggested to me that power is more than having
access to resources and it is to do with a person’s stock of knowledge.
From this perspective it is important to recognise differentials in the cultural resources

that service users and researchers are able to access. Jones (2001a) has argued that these
resources are themselves an interpretation of power. This perspective has been
developed further to identify deliberative inequalities (Bohman 1996) which are thought
to include differences in understanding of language, knowledge of systems, access to
knowledge, degrees of accountability and social environments. In the context of mental
health policy making Hodge (2005) suggests that Habermasian perspectives are limited
because they are based on an idealistic view that as individual beings we have an innate
understanding of and orientation to mutual understanding and a desire to reach
consensus.
An important issue is that the use of ‘story' terminology may serve to underplay the
importance participants awarded to this type of knowledge in terms of gaining insight
and awareness of possibilities for change. The role of ‘storyteller' implies that service
users bring particular types of existing knowledge to research, which could mean there
is less emphasis on the possibility of service users creating knowledge through their
involvement.
More recent concepts of power suggest that power is manifest through the interaction of
people within social structures. In Giddens' (1984) structuration theory social structure
is produced by, and acts back on, the ‘knowledgeable agents' who are the subjects of
that structure. Social agents instantiate social structure through their constitution of it
and structure is both rule and resource. According to Giddens (1977) “system concerns
the regularities of human action”, while “structure concerns the rules and resources that
203
Chapter 6: Tensions
human agents can draw on to produce and reproduce systems”. So, structures of user
involvement in health research could be thought of as the prevailing knowledge
resources that people can draw upon to produce and reproduce systems for involving
service users.
In light of the fact that a top-down concept of control in user involvement in health
research does not fully explain the workings of a research world comprised of
international research programmes, organisations working in partnership,
interdisciplinary networks and multi-professional research teams, it is unsurprising that
such a view also fails to explain how power operates between research systems and
society. These phenomena are better explained by a notion of power as being diffuse.
It is perhaps more useful to take Niccolò Machiavelli's much earlier work on Roman
history as a foundation for understanding the politics of public involvement. According
to a Machiavellian view (1958) power is imprecise, contingent and strategic. It is found
in situ and operates through discourses and civic virtue. When the state is in danger, the
good citizen may have to stoop from the highest morality (Machiavelli 1970).
Foucault's views on power are borne out of a similar problematic to Machiavelli.

Foucault argued that power can be established through codes of practice, discipline and
supervision rather than through supremacy. Knowledge constituted, not just in texts but
in definite institutional and organisational practices can be thought of as ‘discursive
practices' (Foucault 2000c). That is, knowledge is reproduced through practices made
possible by the framing assumptions of that knowledge. Foucault's dissection of
discursive practices, particularly those surrounding mental health, crime and sexual
health, revealed the techniques that “allowed the effects of power to circulate in a
manner at once continuous, uninterrupted, adapted, and individualized throughout the
entire social body” (Foucault 2000, p.120). Power and the use of knowledge are not
guaranteed to be safe, legitimate, or salutary because the knowledge that guides or
instrumentalizes the exercise of power is valid and scientific. Hence, Foucault
approached power as a system of thought that can become controlling socially
legitimated and institutional.
204
Chapter 6: Tensions
From this perspective of power, knowledge in user involvement in health research does
not only relate to the creation of evidence, it also concerns intersubjective
communication. It includes knowing how to gain credentials, esteem, expertise, work to
best practice guidelines, and meet the responsibilities of regulation and accountability.
Foucault used the term savoir (Foucault 2000, p.266) to describe this “middle sort of
knowledges” which may not be recognised as rigorous scientific process but
nevertheless have instrumental benefit.
Tensions between ‘equal say' and ‘know-how' surface around the use of research
language. Participants from Australia gave several examples of circumstances where
research or medical terminology had caused tensions in forming working relationships
with service users. For example, the term ‘palliative care' was described as being a
sensitive term because patients may not perceive their care in this way or may not wish
to be thought of as a person who is near end of life.
(PARTICIPANT 2 - AUS): I had to do quite a lot of background work with the site managers and
the nurses and other people to make sure that there was nothing in there [the project information
sheet] that would be worrying. Because if you talk about palliative care it had sort of negative
connotations and also people going into residential aged care don't necessarily know at all about
palliative care. So if you want to talk and engage with them about that is quite tricky... I think it is
good to pilot anything that you want to use either with the community or with residents, with a
non-professional person because language is extremely important there.
A related point is that participants from academic nursing often felt strongly that all
nurses should be able to understand and use research terminology in order to make use
of research findings and to provide a role in making research more accessible to service
users. The utility of research to nurses themselves was an issue that several participants
felt strongly about, as well as involving nurses in research. Again, these issues resonate
with arguments about user involvement in research.
(PARTICIPANT 5 - AUS): I think involving nurses in research at the clinical level is a good way
to go. That is important because we are doing work here [within an academic setting] but if it is
not getting out there then – it is important to educate nurses about research… The barriers are that
they didn't really understand the research; issues about the inaccessibility of the research; where
the findings are published, which journals they read.
205
Chapter 6: Tensions
This point reminds us that ‘equal say' and ‘equal know-how' are by no means equal
between or within professional groups.
6.5 The interplay of power
So far in this chapter I have suggested that a range of theories of power can help to
understand tensions about user involvement in health research. Following this view it
seems important that an approach to understanding user involvement in health research
provides a way of explaining how different types of power relate to one another.
Clegg's model is useful in this respect because it approaches power as moving though
circuits in a transformative way, including:
(i) Agency: Episodic power relations, Social relations, Agencies, Standing conditions,
Outcomes
(ii) Social integration: Rules of practice, Disposition, Obligatory passage points,
Rules fixing relationships of meaning and membership
(iii) System integration: Domination, Facilitative, Innovation in techniques of
discipline and production.
Clegg's model is a useful basis for looking at service user/researcher relationships

within social contexts, principally because it emphasises that power functions in
different ways that can have overlapping, cumulative or related effects. One example of
how power can interplay is the issue of access. It has been argued that in an ideal
situation:
“Everyone should be involved as early as possible in the research process, so

that both lay people and professionals feel that they ‘own’ the project and
will be committed to it.” (Baxter et al. 2001, p.116)
At the level of agency, access has been approached as a question of support for service
users to become involved – that is who has access to means and resources and the nature
of the prevailing standing conditions. However, user involvement in health research
involves the interplay of less recognised more socially embedded forms of power. At
206
Chapter 6: Tensions
the level of social integration, user involvement in health research can be seen to be a
consequence of dispositional power which concerns rules of meaning and membership.
At the level of social integration, regulation of research through ethics and governance
procedures can affect the timing of when service users can be engaged in research
studies. Furthermore, established citation practices involved with being a researcher (the
things that a person does that make them identify with and be identifiable as a
researcher), mean that responsibility for assuring these requirements lies with the
researcher. According to Clegg:
“System integration, premised on technical competition and environmental

efficiency, will tend to be a potent source of transformation and strain. It will
open up opportunities both for resistance under existing rules and for changes
in the rules, changes which can create new agencies, new handicaps and
advantages, and new pathways through existing fields of force. Before any
such changes can take place, however, there has to be effective organisation
on the part of any agencies which aspire to strategy”. (Clegg 1989, p.19)
The notion that organisation is an important aspect of how power operates (Clegg 1989,
p.17) is a useful perspective because it broadens the question of control beyond research
organisations, to include the important issues of how service users and researchers
organise themselves. For one user-controlled organisation, Shaping Our Lives, the
emphasis is on encouraging professionals to adopt a broader approach to access in terms
of environmental access, cultural access, communication access, and access through
resources to ensure that involvement is a zero cost activity (Beresford and Bransfield
2006). This broader conceptualization of access resonates with Dahl's notion of the ever
present uneven playing field and Bohman's notion of communicative inequalities, and
corresponds with the Foucauldian view that power comprises potential or actual forces
that can affect one another and the people that are caught within them.
207
Figure 28: Circuits of power (Clegg 1989, p.214)
Level of circuit Type of power

Circuits of Power
Standing conditions
Episodic Social Outcomes
Agencies
Relations Means Resources
Power Focus
Agency Causal Control
Relations
reproduce/transform
control/contest
Rules of
Social Obligatory
Practice Dispositional fix/re-fix
Integration Passage Rules fixing relationship of
Points meaning and membership
Exogenous
environmental
facilitate/restrict
contingencies
System Innovation in techniques of

Domination Integration Facilitative discipline and production
empower/disempower
208
Chapter 6: Tensions
6.6 Individual’s perspectives of power
This section continues to look at tensions in relation to user involvement but it takes a
slightly different perspective, asking how user involvement might be experienced by
individual people. I again draw on the work of key social theorists to discuss how user
involvement in health research might be better understood.
The notion of subjective experience positions power within the perception and
experience of the individual person. Put another way, this perspective asks why in
particular social situations we may feel powerful or we may feel powerless. A sense of
power not only relates to social context but to who we are as a person. Our experiences
of power can be heightened when a sudden change in social context is felt, for example
if a person becomes a patient. Such an experience is highlighted by the following
quotation take from a recent consumer review:
“Staff are in uniform and in groups wielding power. I was in pyjamas, alone
and too timid to question things. I was having major surgery and relied on
them for my survival” (Which? 2007, p.17)
In other circumstances individuals may feel empowered by a positive health experience

or being involved in a research study. Understanding the meaning of empowerment is
important because it is a recurrent theme of the user involvement literature and such
concepts shape ideas about user involvement in practice. In order to analyse the
meaning of empowerment in relation to user involvement it is useful to consider
Foucault's views on this concept. Foucault suggests that:
“This form of power applies itself to immediate everyday life which categorises
the individual, marks him by his own individuality, attaches him to his own
identity, imposes a law of truth on him which he must recognise and which
others have to recognise in him. It is a form of power which makes individuals
subject”. (Foucault 1983, p.211)
Current conceptualizations of user involvement (see chapter 3) suggest that power can
be gained through the practice of research. Indeed an empowerment model of user
involvement would position it as an explicit aim (Boote et al. 2002). However, notions
of empowerment through research are complicated by the fact that the philosophical
209
Chapter 6: Tensions
basis for ‘being empowered' and ‘emancipating oneself' are not the same. Benton's
(1981) “paradox of emancipation” raises a further conceptual problem, that is:
“how to reconcile a conception of socialist practice as a form of collective

self-emancipation with a critique of the established order which holds that
the consciousness of those from whom collective self-emancipation is to be
expected is systematically manipulated, distorted and falsified by essential
features of that order. (Benton 1981, p.162)
What is in a persons best interests and what a person might want are not necessarily the
same thing – raising the question of who can judge whether a person has been
empowered or not. Similar issues are inherent in attempts to award voice:
“The problems with the general criterion of giving a voice are of two sorts.
First, although it is acknowledged that some voices have been silenced, the
lack of adequate attention to the reasons for this silencing means that the
voices of the silenced come to be understood as being of equal value to those
which have preceded them. The difficulty of this equal weighting is that the
heretofore silenced voices cannot then be seen as forms of resistance of
challenge to domination- their inclusion is merely seen as providing a more
accurate picture…The second problem is related to this lack of interrogation.
For not only is it then necessary to interrogate the ways in which the silenced
voices are inscribed with unequal power relationships, but further, it is
necessary to make explicit the political framework which underpins such an
interrogation.” (Bhavnani 1990, p.143)
The relationship between power and identity is also relevant to questions of service
user/researcher relationships. The tendency to see service user/researcher as an
oppositional relationship places the service user outside of the process of self-
constitution. The actions of individual service users can only be seen within the context
of professional structuring of social relations (Mauthner and Doucet 2003). From this
perspective actual service user/researcher relationships are framed by social meaning
over and above whatever interpretations particular individuals might put on them. A
particular issue that emerges is whether it is better for service users to engage with
existing research systems and seek to establish rights for service users who are involved,
or alternatively to develop research practices and roles according to ‘non-mainstream'
ideals through user-controlled research.
210
Chapter 6: Tensions
From a Foucauldian perspective of identity there are no fixed categories of service user
or researcher and therefore people chose to align themselves with particular subject
positions by making citation practices. Thus, being a researcher or a service user in
research involves citing already established formations of knowledge (Butler 1993).
“What individual subjects do, say, ’act out’ is subsumed within, and must
always be connected to, performativity, to the citational practices which
reproduce and subvert discourse, and which at the same time enable and
discipline subjects and their performances” (Gregson and Rose 2000, p.441)
Identities can be seen to be “points of temporary attachment to the subject positions

which discursive practices construct for us” (Hall 2000, p.19). At the same time socially
available discourses about what ‘being a researcher' or ‘being a service user involved in
research' constitutes change over time and between places. In this context judgements
about user involvement take on personal meaning at the individual level because they
are connected to an individual's perspectives of their own identity.
{Trigger (App3/Date:18.10.05):Reflections on how the interviews are progressing}

Findings from the interviews illustrate that understandings of user
involvement in health research are personal and particular to the
individual and this has implications for what is being done (or not done) to
involve service users in research practice. An area that has provoked my
interest is how user involvement in health research is being influenced by
the interplay of political/social forces and personal preferences.
There is no one clear or agreed message about what a good researcher (or a good service
user) should do, indeed some political, organisational, ethical and social messages can
be in tension or contradictory. For example the view of researcher as efficient
‘knowledge worker' (Darr and Warhurst 2008) could be inconsistent with the additional
time and investment that working with service users entails. Identities can be further
confused if individuals take up multiple roles: clinician, researcher, author for example.
Furthermore, according to Weedon (1987) we are capable of critical reflection and we
may resist, challenge or reject the culturally available discourses within our current
epoch.
Foucault's work is helpful for clarifying these complex issues because it suggests codes
of conduct are a form of ethics, rather than approaching morality as obedience to
211
Chapter 6: Tensions
commandments and prohibitions (Foucault 2000). In other words researcher power can
be approached as a set of commitments and adherences that can be transformed and
remade through work, encounters and engagements: a position that I will examine more
closely in the rest of this thesis.
To draw a close to this chapter, figure 29 applies some of the issues discussed here to
propagate the power theory map from chapter 4. The figure presents user involvement
in health research as an open system of power, connected to other social systems and
acknowledging that dispersion of power can occur between systems because individuals
carry out different social practices. The figure depicts the interface between social
structures and the individual person is the place where ideas about user involvement are
either recreated or resisted through the intentions and actions of individuals. On the
basis of these findings the following chapter explores some ways forward for
knowledge development.
212
Figure 29: Power in user involvement in health research
Environmental influences
Service user/researcher relationships Personal experiences

• Requirements and incentives for user involvement ‘Ability to’
• Funding for research and user involvement
• Access research resources
• Information about user involvement • Achieve goals e.g. to improve health or health
• Prevailing expectations and conditions services
• Communication structures • Make decisions about how to research
• Types of outcomes that are privileged • Express views/deliberate about health and research
issues
• Adapt to change in research systems
Rules of research
• Roles that are available to be taken up e.g. ‘Potential to’

researcher, user-researcher, participant • Take up/resist particular roles
• Criteria and responsibilities associated with roles • Loyalty to ideas or ways of working
• Rules of practice and ‘know-how' for being a health • Gain status, expertise, credibility
researcher • Identify and organise interests
‘Sense of being’
• Valued as a partner rather than controlled
Prevailing research structures • Enabled rather than constrained
• Empowered rather than exploited
• Research organisations, programmes and units • Consenting not coerced
• Research ethics and governance systems • Conscious that power operates in different ways
• Methods and techniques of knowledge production
• Use of research technologies, monitoring and
reporting systems
Support or challenge from other types of social structures
Change over time
213
Chapter 6: Tensions
Summary
 This chapter has explored how tensions associated with user involvement in
health research might be better understood by making use of social theories of
power. It may not be possible to overcome tensions about professional control
but it may be possible to achieve a closer match between ideals and the realities
of service user/researcher relationships.
 Three main conceptual tensions have been explored. Firstly the view that user
involvement in research must be supported by policy makers, research
commissioners and the research community, but at the same time it should not
be ‘top-down' or controlled by professionals. Second that there should be more
transparency in research, particularly around processes of decision making, but
at the same time there are uncertainties about what the underlying intentions and
interests are. Third that service users should have ‘equal' say, but at the same
time professional ‘know-how' about research systems means that service users
can never gain an equal footing.
 To understand tensions surrounding service user/researcher relationships (and

non-relationships) it is useful to move from a linear notion of control towards a
more encompassing understanding of the interplay of multiple underlying forms
of power. Notions of power in service user/researcher relationships have
generally been limited to an ‘agency' model – that is who has control over
means and resources. However, user involvement in health research involves the
interplay of less recognised more socially embedded forms of power. At the
level of social integration, user involvement in health research can be seen to be
a consequence of dispositional power which concerns rules of meaning and
membership; most significantly citation practices involved with being a
researcher. This role has traditionally been viewed as ‘generating knowledge',
but user involvement in health research brings into question which people can
align with this subject position.
214
Chapter 6: Tensions
 Any attempt to measure the impact of user involvement in health research faces
the problem of understanding intent. Any purpose, process or outcome is
constructed by a multitude of conscious and unconscious decisions, made by a
multitude of actors, both within and outside of user/researcher relationships.
 Debates about service user involvement have focused on the idea of political
authority rather than the ‘micro' subjective judgements and the ‘macro' social
contexts that shape research practice. When perceived in this way, tensions are
not simply a matter of the difficulties of fulfilling policy requirements there are
issues about less accessible dimensions of power such as non-decision making
and the intentions behind actions.
215
Appendices
216

Appearances - of - Power - Part 1

Uploaded by

Copyright:

Available Formats

Appearances - of - Power - Part 1

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

Appearances - of - Power - Part 1

Uploaded by

Copyright:

Available Formats

Appearances of power: Service user involvement in health

Elizabeth Mary Smith

Thesis submitted for the degree of

I declare that the work presented in this thesis is my own.

Signed ……………………………………… Date ………………….

This study aims to contribute theoretically informed understandings about the

Chapter 3: Current knowledge about user involvement in health research.........................49

Chapter 4: A method for researching user involvement.........................................................95

Chapter 5: How knowledge of user involvement has developed..........................................149

Chapter 7: Ways forward........................................................................................................216

Figure 1: Structure of the thesis.......................................................................................16

Table 1: Principles of consumer involvement (Telford et al. 2004)................................91

The issue of public involvement is important to many Westernized nations and in

‘consumers' of health services, establishing mechanisms for Patient and Public

A particular point of confusion in relation to user involvement in health research is the

1.2 Problem statement

1.3 Aims and objectives

The objectives were:

1.4 Overview of the method

Figure 1: Structure of the thesis

In this study building knowledge of user involvement in health research began by

A discursive component of the approach helped me to consider health research as

Reflective methodology provided a way to explore user involvement in health research

{Trigger (App3/Date:13.03.07): Supervision meeting, discussion of use of

Recording and reflecting on triggers is particularly helpful for representing interaction

From a Foucauldian perspective user involvement in research creates rules and

1.5 What this study contributes

 Summarizes and critiques current knowledge of user involvement in health research

 Uses theoretically informed and empirically based analyses to identify ways in

2.2 Socio-political trends towards user involvement

“The development of lay involvement in research and development in other

movement’ influenced lay involvement particularly during the 1980’s and

Recognition of public interests in public services

Figure 2: Historical context

Recognition of public interests in public

Movements for social welfare and inclusion including;

The Griffiths report encouraged market research and

NHS Management Executive (1992) publish Local Voices

Efficiency and responsiveness to users

Political interests on consumerism focus on the possibilities of introducing new types of

One sociological view of science suggests:

“Science is a deceptively inclusive word which refers to a variety of distinct

A further indication of a growing interest in users occurred in 1991. The regional

Rationing and priority setting

hundred consumer advocates, researchers, clinicians, information specialists and service

“Every decision which affects patients is influenced by evidence which reflects

“The Shaping Our Lives project is a consumer-led project and involves

Safety and quality

“That the scientific merit of particular research grant proposals should

Plurality and diversity

In current health policy user involvement is described as offering multiple advantages

Conceptualizations of public participation, such as Arnstein's (1969) ladder of citizen

Growing interest in user-led and user-controlled research bring an interesting

Figure 3: International Association for Public Participation

“Each relevant English body must make arrangements, as respects health

2.3 Public involvement within the knowledge economy

The picture of public involvement in relation to health research is somewhat different.

2.4 Background to the study

Figure 4: Study time line