EVIDENCE-BASED

PROGRAMMING FOR
OLDER ADULTS
EDITED BY : Marcia G. Ory and Matthew Lee Smith
PUBLISHED IN : Frontiers in Public Health
 Frontiers Copyright Statement
About Frontiers
© Copyright 2007-2015 Frontiers
Media SA. All rights reserved. Frontiers is more than just an open-access publisher of scholarly articles: it is a pioneering
All content included on this site,
such as text, graphics, logos, button
approach to the world of academia, radically improving the way scholarly research
icons, images, video/audio clips, is managed. The grand vision of Frontiers is a world where all people have an equal
downloads, data compilations and
software, is the property of or is opportunity to seek, share and generate knowledge. Frontiers provides immediate and
licensed to Frontiers Media SA permanent online open access to all its publications, but this alone is not enough to
(“Frontiers”) or its licensees and/or
subcontractors. The copyright in the realize our grand goals.
text of individual articles is the property
of their respective authors, subject to
a license granted to Frontiers. Frontiers Journal Series
The compilation of articles constituting The Frontiers Journal Series is a multi-tier and interdisciplinary set of open-access, online
this e-book, wherever published,
as well as the compilation of all other journals, promising a paradigm shift from the current review, selection and dissemination
content on this site, is the exclusive
property of Frontiers. For the
processes in academic publishing. All Frontiers journals are driven by researchers for
conditions for downloading and researchers; therefore, they constitute a service to the scholarly community. At the same
copying of e-books from Frontiers’
website, please see the Terms for time, the Frontiers Journal Series operates on a revolutionary invention, the tiered publishing
Website Use. If purchasing Frontiers system, initially addressing specific communities of scholars, and gradually climbing up to
e-books from other websites
or sources, the conditions of the broader public understanding, thus serving the interests of the lay society, too.
website concerned apply.
Images and graphics not forming part
of user-contributed materials may
Dedication to quality
not be downloaded or copied
Each Frontiers article is a landmark of the highest quality, thanks to genuinely collaborative
without permission.
Individual articles may be downloaded interactions between authors and review editors, who include some of the world’s best
and reproduced in accordance academicians. Research must be certified by peers before entering a stream of knowledge
with the principles of the CC-BY
licence subject to any copyright or that may eventually reach the public - and shape society; therefore, Frontiers only applies
other notices. They may not be
re-sold as an e-book.
the most rigorous and unbiased reviews.
As author or other contributor you Frontiers revolutionizes research publishing by freely delivering the most outstanding
grant a CC-BY licence to others to research, evaluated with no bias from both the academic and social point of view.
reproduce your articles, including any
graphics and third-party materials By applying the most advanced information technologies, Frontiers is catapulting scholarly
supplied by you, in accordance with
the Conditions for Website Use and
publishing into a new generation.
subject to any copyright notices which
you include in connection with your What are Frontiers Research Topics?
articles and materials.
All copyright, and all rights therein, Frontiers Research Topics are very popular trademarks of the Frontiers Journals Series:
are protected by national and
international copyright laws. they are collections of at least ten articles, all centered on a particular subject. With their
The above represents a summary unique mix of varied contributions from Original Research to Review Articles, Frontiers
only. For the full conditions see the
Conditions for Authors and the
Research Topics unify the most influential researchers, the latest key findings and historical
Conditions for Website Use. advances in a hot research area! Find out more on how to host your own Frontiers
ISSN 1664-8714
Research Topic or contribute to one as an author by contacting the Frontiers Editorial
ISBN 978-2-88919-585-5
DOI 10.3389/978-2-88919-585-5 Office: [email protected]

Frontiers in Public Health 1 June 2015  |  Evidence-based programming for older adults
 EVIDENCE-BASED PROGRAMMING
FOR OLDER ADULTS
Topic Editors:
Marcia G. Ory, Texas A&M Health Science Center, College Station, USA
Matthew Lee Smith, The University of Georgia, Athens, USA

There is increased world-wide concern about

the impact of multiple chronic conditions,
especially among the rapidly aging population.
Simultaneously, over the past decade there
has been an emergence of state-wide and
national initiatives to reduce the burden
of chronic conditions that draw upon the
translation of evidence-based programs (EPB)
into community practice. Yet, little has been
written about the national and international
implementation, dissemination, and
Copyright: Texas A&M Health Science Center sustainability of such programs.

This Research Topic features articles about EBPs for older adults, including a range of articles
that focus on the infrastructure needed to widely disseminate EBP as well as individual
participant impacts on physical, mental, and social aspects of health and well-being. Using
a pragmatic research perspective, this Research Topic will advance knowledge that aims
to enhance practice, inform policy and build systems of support and delivery in regard to
the reach, effectiveness, adoption, implementation, and maintenance of evidence-based
interventions for older adults. The focus is on knowledge transfer rather than knowledge
generation but with a dual emphasis on the dissemination and sustainability of EBP that have
been tested and shown effective as well as the adaptation of practice-based interventions into
evidence-based programs. This Research Topic draws upon grand-scale efforts to deliver these
programs, and include both U.S. as well as international examples.

Commentaries discuss processes in the development and measurement of EBP and reflect
perspectives from program developers and major national and regional funders of EBP
as well as professionals and practitioners in the field. The full-length articles focus on
four major programmatic areas: (1) chronic disease self-management programs; (2) fall
prevention programs; (3) general wellness and physical activity programs; and (4) mental

Frontiers in Public Health 2 June 2015  |  Evidence-based programming for older adults
 health programs. Additionally, articles are included to discuss cross-cutting issues related to
building partnerships and the research infrastructure for the implementation, evaluation,
and dissemination of evidence-based programming.

The intent of this Research Topic is to enhance practice, inform policy, and build systems
of support and delivery for EBP. It is written for a diverse audience and contains practical
implications and recommendations for introducing, delivering, and sustaining EBP in a
multitude of settings.

Citation: Marcia G. Ory and Matthew Lee Smith, eds. (2015). Evidence-based programming for older
adults. Lausanne: Frontiers Media. doi: 10.3389/978-2-88919-585-5

Frontiers in Public Health 3 June 2015  |  Evidence-based programming for older adults
 Table of Contents

09 Research, practice, and policy perspectives on evidence-based programing

for older adults
Marcia G. Ory and Matthew Lee Smith

Section I: Perspectives from: National Stakeholders Guiding the

Evidence-Based Program Movement
18 Fostering healthy aging through evidence-based prevention programs:
Perspectives from the Administration for Community Living/Administration
on Aging
Michele L. Boutaugh and Laura J. Lawrence
21 Improving lives through evidence-based health promotion programs:
a national priority
Richard Birkel, Emily Dessem, Simona Eldridge, Kristie Kulinski, Sue Lachenmayr,
Michelle Spafford, Binod Suwal, Albert Terrillion, Mary Walsh and Wendy Zenker
23 An introduction to the Centers for Disease Control and Prevention’s efforts to
prevent older adult falls
Margaret Kaniewski, Judy A. Stevens, Erin M. Parker and Robin Lee
26 Community-based wellness and prevention programs: the role of Medicare
Erin Murphy Colligan, Naomi Tomoyasu and Benjamin Howell
28 Foundation engagement in healthy aging initiatives and evidence-based
programs for older adults
Mary Ellen Kullman

Section II: Perspectives from: Evidence-Based Program Developers

30 Chronic Disease Self-Management Program: insights from the eye of the storm
Kate Lorig
33 Corrigendum: Chronic Disease Self-Management Program: Insights from the
Eye of the Storm
Kate Lorig
34 A matter of balance: older adults taking control of falls by building confidence
Margaret Haynes, Patricia League and Gloria Neault
36 “Stepping On”: stepping over the chasm from research to practice
Jane E. Mahoney
38 Village or tribe? Expectations, roles, and responsibilities for effective fall
prevention efforts
Tiffany E. Shubert

Frontiers in Public Health 4 June 2015  |  Evidence-based programming for older adults
 40 EnhanceFitness: a 20-year dissemination history
Susan J. Snyder, MeghanThompson and Paige Denison
42 Translating Fit and Strong!: Lessons learned and next steps
Susan Hughes, Renae L. Smith-Ray, Amy Shah and Gail Huber
45 Texercise: the evolution of a health promotion program
Holly Riley
47 Translating PEARLS: lessons learned from providers and Participants
Mark B. Snowden, Lesley E. Steinman, Pamela Piering, Sluggo Rigor and Andrea Yip

Section III: Perspectives from: Evidence-Based Program Networkers

50 The CDC Healthy Aging Research Network: advancing science toward action
and policy for the evidence-based health promotion movement
Basia Belza, Mary Altpeter, Steven P. Hooker and Gwen Moni
53 Chronic disease self-management support: public health perspectives
Teresa J. Brady, Lynda A. Anderson and Rosemarie Kobau
59 Evidence-based leadership council – a national collaborative
Margaret Haynes, Susan Hughes, Kate Lorig, June Simmons, Susan J. Snyder,
Lesley Steinman, Nancy Wilson, Roseanne DiStefano, Jennifer Raymond,
Stephanie FallCreek, Martha B. Pelaez and Don Smith
62 Working toward a multi-program strategy in fall prevention
(Bonita) Lynn Beattie
64 Building the older adult fall prevention movement – steps and lessons learned
Ellen C. Schneider and (Bonita) Lynn Beattie
70 Public health system perspective on implementation of evidence-based
fall prevention strategies for older adults
Sallie R. Thoreson, Lisa M. Shields, David W. Dowler and Michael J. Bauer
73 Falling for a balance partner
Sara B. May and Cherie A. Rosemond

Section IV: CDSME Program Dissemination Through the ARRA

75 Setting the stage: measure selection, coordination, and data collection for a
national self-management initiative
Kristie P. Kulinski, Michele Boutaugh, Matthew Lee Smith, Marcia G. Ory and
Kate Lorig
81 National dissemination of chronic disease self-management education
programs: an incremental examination of delivery characteristics
Matthew Lee Smith, Marcia G. Ory, SangNam Ahn, Kristie P. Kulinski,
Luohua Jiang, Scott Horel and Kate Lorig
88 Workshop characteristics related to Chronic Disease Self-Management
Education program attendance
Matthew Lee Smith, Marcia G. Ory, Luohua Jiang, Kate Lorig, Kristie P. Kulinski and
SangNam Ahn
95 The role of Session Zero in successful completion of Chronic Disease
Self-Management Program workshops
Luohua Jiang, Matthew Lee Smith, Shuai Chen, SangNam Ahn, Kristie P. Kulinski,
Kate Lorig and Marcia G. Ory

Frontiers in Public Health 5 June 2015  |  Evidence-based programming for older adults
 101 Reaching diverse participants utilizing a diverse delivery infrastructure:
a replication study
Matthew Lee Smith, Marcia G. Ory, SangNam Ahn, Basia Belza, Chivon A. Mingo,
Samuel D. Towne Jr. and Mary Altpeter
110 The reach of chronic-disease self-management education programs to
rural populations
Samuel D. Towne Jr., Matthew Lee Smith, SangNam Ahn and Marcia G. Ory
117 Factors associated with successful completion of the Chronic Disease
Self-Management Program among middle-aged and older Asian-American
participants: a national study
SangNam Ahn, Matthew Lee Smith, Jinmyoung Cho, Luohua Jiang, Lindsey Post
and Marcia G. Ory
123 Chronic Disease Self-Management Education (CDSME) program delivery
and attendance among urban-dwelling African Americans
Chivon A. Mingo, Matthew Lee Smith, SangNam Ahn, Luohua Jiang,
Jinmyoung Cho, Samuel D. Towne Jr. and Marcia G. Ory
132 Factors associated with Hispanic adults attending Spanish-language disease
self-management program workshops and workshop completion
Matthew Lee Smith, SangNam Ahn, Luohua Jiang, Kristie P. Kulinski and
Marcia G. Ory

Section V: CDSME Program Implementation and Outcomes

139 Methods for streamlining intervention fidelity checklists: an example from the
Chronic Disease Self-Management Program
SangNam Ahn, Matthew Lee Smith, Mary Altpeter, Basia Belza, Lindsey Post and
Marcia G. Ory
147 Factors supporting implementation among CDSMP organizations
Deborah Paone
158 Examining sustainability factors for organizations that adopted Stanford’s
Chronic Disease Self-Management Program
Michiyo Tomioka and Kathryn L. Braun
166 Chronic Disease Self-Management Program in the workplace: opportunities
for health improvement
Matthew Lee Smith, Mark G. Wilson, David M. DeJoy, Heather Padilla, Heather
Zuercher, Phaedra Corso, Robert Vandenberg, Kate Lorig and Marcia G. Ory
172 Meeting the challenge of cancer survivorship in public health: results from
the evaluation of the chronic disease self-management program for cancer
survivors
Betsy C. Risendal, Andrea Dwyer, Richard W. Seidel, Kate Lorig, Letoynia Coombs
and Marcia G. Ory
180 Implementing a chronic disease self-management program into China:
the Happy Life Club™
Colette Joy Browning, Hui Yang, Tuohong Zhang, Anna Chapman, Shuo Liu,
Joanne Enticott and Shane Andrew Thomas
185 Implementing chronic disease self-management approaches in Australia and
the United Kingdom
Colette Joy Browning and Shane AndrewThomas

Frontiers in Public Health 6 June 2015  |  Evidence-based programming for older adults
 189 Cost-effectiveness of the chronic disease self-management program:
implications for community-based organizations
Rashmita Basu, Marcia G. Ory, Samuel D. Towne Jr., Matthew Lee Smith,
Angela K. Hochhalter and SangNam Ahn
197 Healthcare cost savings estimator tool for chronic disease self-management
program: a new tool for program administrators and decision makers
SangNam Ahn, Matthew Lee Smith, Mary Altpeter, Lindsey Post and Marcia G. Ory
209 Linking evidence-based program participant data with Medicare data: the
consenting process and correlates of retrospective participant consents
Philip Lloyd Ritter, Marcia G. Ory, Matthew Lee Smith, Luohua Jiang, Audrey Alonis,
Diana D. Laurent and Kate Lorig

Section VI: Evidence-Based Falls, Physical Activity, and Mental Health

Programs
216 Fall prevention in community settings: results from implementing Tai Chi:
Moving for Better Balance in three states
Marcia G. Ory, Matthew Lee Smith, Erin M. Parker, Luohua Jiang, Shuai Chen,
Ashley D. Wilson, Judy A. Stevens, Heidi Ehrenreich and Robin Lee
222 Fall prevention in community settings: results from implementing Stepping
On in three states
Marcia G. Ory, Matthew Lee Smith, Luohua Jiang, Robin Lee, Shuai Chen,
Ashley D. Wilson, Judy A. Stevens and Erin M. Parker
228 Translation of The Otago Exercise Program for adoption and implementation
in the United States
Tiffany E. Shubert, Matthew Lee Smith, Marcia G. Ory, Cristine B. Clarke,
Stephanie A. Bomberger, Ellen Roberts and Jan Busby-Whitehead
236 Developing an evidence-based fall prevention curriculum for community
health workers
Julie A. St. John, Tiffany E. Shubert, Matthew Lee Smith, Cherie A. Rosemond,
Doris A. Howell, Christopher E. Beaudoin and Marcia G. Ory
245 Effects of an evidence-based falls risk-reduction program on physical activity
and falls efficacy among oldest-old adults
Jinmyoung Cho, Matthew Lee Smith, SangNam Ahn, Keonyeop Kim, Bernard Appiah
and Marcia G. Ory
254 Gait speed among older participants enrolled in an evidence-based fall risk
reduction program: a subgroup analysis
Jinmyoung Cho, Matthew Lee Smith, Tiffany E. Shubert, Luohua Jiang,
SangNam Ahn and Marcia G. Ory
261 The conversion of a practice-based lifestyle enhancement program into a
formalized, testable program: from Texercise Classic toTexercise Select
Marcia G. Ory, Matthew Lee Smith, Doris Howell, Alyson Zollinger, Cindy Quinn,
Suzanne M. Swierc and Alan B. Stevens
270 Translation of Fit & Strong! for middle-aged and older adults: examining
implementation and effectiveness of a lay-led model in CentralTexas
Marcia G. Ory, Shinduk Lee, Alyson Zollinger, Kiran Bhurtyal, Luohua Jiang and
Matthew Lee Smith

Frontiers in Public Health 7 June 2015  |  Evidence-based programming for older adults
 280 Fit & Strong! promotes physical activity and well-being in older cancer
survivors
Jana Reynolds, Lorie Thibodeaux, Luohua Jiang, Kevin Francis and Angie Hochhalter
288 Adoption of evidence-based health promotion programs: perspectives of
early adopters of Enhance®Fitness in YMCA-affiliated sites
Basia Belza, Miruna Petrescu-Prahova, Marlana Kohn, Christina E. Miyawaki,
Laura Farren, Grace Kline and Ann-Hilary Heston
300 Development and evaluation of a fidelity instrument for PEARLS
Laura Farren, Mark Snowden, Lesley Steinman and Maria Monroe-DeVita

Section VII: Cross-Cutting Perspectives for Evidence-Based Programming

307 Perceived utility of the RE-AIM framework for health promotion/disease
prevention initiatives for older adults: a case study from the U.S.
evidence-based disease prevention initiative
Marcia G. Ory, Mary Altpeter, Basia Belza, Janet Helduser, Chen Zhang and
Matthew Lee Smith
325 National dissemination of multiple evidence-based disease prevention
programs: reach to vulnerable older adults
Samuel D. Towne Jr., Matthew Lee Smith, SangNam Ahn, Mary Altpeter, Basia
Belza, Kristie Patton Kulinski and Marcia G. Ory
335 Healthcare providers’ perceptions and self-reported fall prevention practices:
findings from a large New York health system
Matthew Lee Smith, Judy A. Stevens, Heidi Ehrenreich, Ashley D. Wilson,
Richard J. Schuster, Colleen O’Brien Cherry and Marcia G. Ory
340 CDC and YMCA: a promising partnership for delivering fall prevention
programing
Heidi Ehrenreich, Maureen Pike, Katherine Hohman, Margaret Kaniewski,
Matt Longjohn, Gaya Myers and Robin Lee
343 A missing piece in the infrastructure to promote healthy aging programs:
education and work force development
Janet Christine Frank
346 Effect of physical activity, social support, and skills training on late-life
emotional health: a systematic literature review and implications for
public health research
Mark B. Snowden, Lesley E. Steinman, Whitney L. Carlson, Kara N. Mochan,
Ana F. Abraido-Lanza, Lucinda L. Bryant, Michael Duffy, Bob G. Knight, Dilip V. Jeste,
Katherine H. Leith, Eric J. Lenze, Rebecca G. Logsdon, William A. Satariano,
Damita J. Zweiback and Lynda A. Anderson
360 EvidenceToPrograms.com: a toolkit to support evidence-based
programming for seniors
Alan B. Stevens, Shannon B. Coleman, Richard McGhee and Marcia G. Ory

Frontiers in Public Health 8 June 2015  |  Evidence-based programming for older adults
 EDITORIAL
published: 27 April 2015
doi: 10.3389/fpubh.2015.00136

Research, practice, and policy

perspectives on evidence-based
programing for older adults
Marcia G. Ory 1* and Matthew Lee Smith 2
1
Department of Health Promotion and Community Health Sciences, Texas A&M Health Science Center, School of Public
Health, College Station, TX, USA, 2 Department of Health Promotion and Behavior, College of Public Health, The University of
Georgia, Athens, GA, USA

Keywords: older adults, evidence-based programs, chronic disease, physical activity, fall prevention, behavioral
health, implementation and dissemination research

Framing Evidence-Base Programing for Older Adults:

Understanding the Interacting Influences of Research,
Practice, and Policy
Demographers warn us of the “gray tsunami” approaching our global doorstep (1). Researchers are
called upon to document the extent to which the growing burden of chronic conditions impacts
America’s aging population and examine the uptake and effectiveness of different intervention
approaches for improving the health and well-being of older adults across settings and populations
(2). Working in conjunction with researchers, practitioners are asked to develop, adopt, and adapt
innovative evidence-based health promotion and disease management programing that can be
broadly implemented, disseminated, and sustained as appropriate in community and clinical settings
(3, 4). Building on a growing research base and inventory of treatment options, policy makers are
charged with identifying and supporting needed care and services that can meet the Triple Aims of
Edited and reviewed by: health reform (i.e., better health, better health care, and better value) (5, 6).
Joav Merrick, This Research Topic on evidence-based programing for older adults reflects decades of progress
Ministry of Social Affairs, Israel by researchers, practitioners, aging service providers, and policy makers working together to under-
*Correspondence: stand how to help older adults achieve optimal health and well-being. Such efforts have transformed
Marcia G. Ory successful aging from a theoretical concept into an achievable goal (7, 8).
[email protected] The scientific roots of this Research Topic are many, but our (Ory and Smith) personal interest
began with the evaluation of the Administration on Aging (AoA)’s national disease prevention
Specialty section: initiatives introduced in the 2000s, which will be described in length later in this volume (9–11).
This article was submitted to Public With our colleagues in the Centers for Disease Control and Prevention (CDC)-funded Healthy
Health Education and Promotion, Aging Research Network (12, 13), we began documenting the national roll-out of evidence-based
a section of the journal Frontiers in programs for older adults. We were concerned with many issues: (1) who were the major stakeholders
Public Health in this national effort?; (2) what programs were being offered and who they were reaching?; (3) what
Received: 21 April 2015 could we say about the fidelity, dissemination, and sustainability of different programs?; (4) what
Accepted: 21 April 2015 was known about the impact of different programs in different populations and settings?; and (5)
Published: 27 April 2015
what were the best strategies for advancing the evidence-based movement?
Citation: As we explored these questions, we realized the need to look beyond single silos or perspectives
Ory MG and Smith ML (2015)
to understand how researchers, program developers, and policy makers could work together more
Research, practice, and policy
perspectives on evidence-based
closely. Such collaborations are essential to develop, promote, and support evidence-based program-
programing for older adults. ing that reflects stakeholders’ perspectives and increases the likelihood of being embedded into exist-
Front. Public Health 3:136. ing structures. Ideally, evidence-based programs reflect a translation of testable research theories
doi: 10.3389/fpubh.2015.00136 into key intervention elements that resonate with program adopters and intended participants.

Frontiers in Public Health | www.frontiersin.org 9 April 2015 | Volume 3 | Article 136

Ory and Smith Perspectives on evidence-based programing

However, it is critical that interventions are seen as desirable of specific attention to (or impacts on) different settings and
and feasible for both organizations and intended audiences if populations.
they are to be adopted. Thus, a dynamic interaction between
research and practice is desirable to ensure the appropriateness The Evidence: From Humble Beginnings
of program content and delivery, especially as they are dis-
seminated and evaluated in different populations and settings. Traditional stereotypes of aging viewed older adults as inap-
Similarly, it is important to examine the role the policy con- propriate targets for community-based health promotion pro-
text plays in sustainable program success. For example, health- grams because they were believed to be uninterested in such
care policies are theoretically designed to meet national health programs and/or unable to benefit from such preventive efforts
care goals. Researchers and practitioners can help document (21). However, research from the National Institute on Aging
the benefits and consequences of current policies facilitating or began documenting the value of a range of self-care and self-
impeding the growth and sustainability of evidence-based pro- management efforts targeted at older adults (22). From a practice
gramming. Research about program effectiveness can inform perspective, older adults are entitled to a variety of programs
new policy directions, and practitioners can provide real-world and services through the Older Americans Act (23), with Title
views about the practicality of different service and programing IIID providing community-based resources for health promo-
options. tion activities. In addition to providing support for congregate
In formulating this Research Topic, our collective objective meals, early AoA programs focused on providing education about
was to identify the most effective programs and to under- the importance of healthy eating and being physically active;
stand individual, social, community, and environmental fac- two key risk factors for older adult health identified by national
tors that influence program reach, adoption, implementation, experts (24). As described in the article by the Administration
dissemination, and sustainability. This perspective aligns with for Community Living (ACL) (9), starting in the early 2000s,
many emergent themes and frameworks in evidence-based pub- there was a growing impetus to develop and test best practices
lic health and medicine such as the RE-AIM planning and for health promotion/disease prevention programs. These activ-
evaluation framework (14, 15), the dissemination and imple- ities coincided with the broader movement toward evidence-
mentation framework (16), and the movement toward transla- based practice emerging in medicine, public health, behavioral
tional research in promoting population health (17–20). As we medicine (25), and complemented the recognition that education
framed this body of work, we created a heuristic framework alone seldom resulted in sustained behavior change (26). Also,
(see Figure 1) to reflect the three key interacting perspectives of during this early period, there was a growing body of research
research, practice, and policy. Secondarily, we wanted to repre- about “what works” to promote healthy aging, but most studies
sent key players such as program developers and national stake- had been conducted with limited populations and settings under
holders, the role of different program types, and the importance controlled situations by academics and were not designed for

FIGURE 1 | Evidence-based programs for older adults: interacting influences and areas of study.

Frontiers in Public Health | www.frontiersin.org 10 April 2015 | Volume 3 | Article 136

Ory and Smith Perspectives on evidence-based programing

widespread dissemination in real-world settings by practitioners programing movement for older adults, there are a multitude of
(27, 28). other players at the national and regional level. The CDC has
been a leader in the effort to promote public health solutions for
Guiding the Evidence-Based Movement: healthy aging and fall prevention (12, 34). From a policy per-
Past, Present, and Future spective, the Centers for Medicare and Medicaid Services are pro-
moting policy-based research on community-based wellness and
In the past, there were few researchers involved in developing promotion programs (35). In addition to the public sphere, private
evidence-based programs for older adults, few community pro- foundations such as the Archstone Foundation (36), which works
grams adopting these programs, few practitioners delivering these to prepare society for the needs of an aging population, recognize
programs, and even fewer policy makers focused on strategies evidence-based programming as an important tool for realizing
for guaranteeing sustainable funding streams. An initial step in their goals. Regionally, the Health Foundation of South Florida
promoting evidence-based programming was informing the aging has become a national leader in demonstrating the importance
services provider network about the definition of evidence-based of a collaborative approach to implementing multiple evidence-
health promotion and disease prevention and its value for practi- based programs (37). Two interrelated themes emerge from this
tioners and policy makers. section: (1) the importance of involving top stakeholders in the
Toward this end, the National Council on Aging (NCOA) (29) field; and (2) the need for partnerships across research-, practice-,
served as the Technical Resource Center for the AoA’s new initia- and policy-based agencies. Having champions well-positioned in
tives in this area. Under the leadership of Nancy Whitelaw, first national organizations from different aging and health sectors has
Director of the Center for Healthy Aging, a variety of resources helped accelerate the evidence-based movement.
was created. These resources included the now classic briefing
on “Using the Evidence-Base to Promote Healthy Aging” (30) Perspectives from Evidence-Based Program
and a series of online training modules on different aspects of
Developers
evidence-based programing (31).
This section focuses on the evolution of the evidence-based move-
The articles in this Research Topics provide an excellent
ment from the perspective of the program developers themselves
overview of the evolution from past to present activities, especially
Included is information regarding the processes involved in devel-
related to the dissemination and testing of evidence-based chronic
oping and taking some of the major evidence-based programs for
disease self-management programs, physical activity programs,
older adults to scale, including: (1) the Stanford suite of CDSME
fall prevention programs, and to a lesser extent, behavioral health
programs (38); (2) a Matter of Balance (39); (3) stepping On
programs. While great strides have been made over the past three
(40); (4) Otago Exercise Program (41); (5) enhance fitness (42);
decades, there is still considerable room for improvement related
(6) fit and strong! (43); (7) texercise (44); and (8) Program for
to program delivery, dissemination and sustainability.
Encouraging Active and Rewarding Lives (PEARLS) (45). Many of
Authors of this volume were asked to reflect about future
these programs have a long history, as exemplified by the Chronic
implications for research, practice, or policy. Solid groundwork
Disease Self-Management Program (CDSMP) that has its roots as
has been laid, suggesting that the evidence-based movement has
a doctoral dissertation in the 1980s (46).
the foundation for even greater dissemination among an aging
The program developers generously share the lessons they
population. Our early work focused on the first 100,000 partici-
learned including the importance of: (1) building programs with
pants in the suite of programs referred to generically as Chronic
the end user in mind; (2) defining roles and responsibilities
Disease Self-Management Education (CDSME) programs. Recent
of partners from diverse sectors to build a culture of preven-
statistics indicate the rapid proliferation of programs with over
tion; (3) setting up a training and certification infrastructure for
300,000 persons engaged in evidence-based programs deliv-
widespread dissemination with fidelity; and (4) acknowledging
ered through the aging services network since 2010, including
the necessity for policy changes to provide sustainable funding
more than 230,000 with CDSME alone (K. Kulinski, personal
streams. Additionally, the contributors express their belief in the
communication).
true value of having a national data repository for real-time and
Policy changes, such as the new mandate from ACL limiting
continued tracking of the reach and representativeness of older
Title IIID reimbursement to evidence-based programs, will serve
participants in evidence-based programs (47). As with any inter-
to increase the number of evidence-based programs disseminated
vention, a major challenge is balancing the need for program stan-
to older adults through the aging services network (32). Addi-
dardization (based on essential intervention elements) with adap-
tionally, efforts to embed evidence-based programs into existing
tations desired for broader applicability to different populations
health care systems and funding streams bode well for the long-
and settings consistent with the latest research (48).
term growth and sustainability of evidence-based programing
for older adults (9, 33). As an example, the 2015 White House
Conference on Aging includes policy briefs highlighting strategies Perspectives from Evidence-Based Program
for promoting health and preventing disease and injury (33). Networkers
The national stakeholders have helped spawn networks whose
Perspectives from National Stakeholders Guiding primary missions intersect with the goal of accelerating the
the Evidence-Based Program Movement implementation, evaluation, and dissemination of evidence-based
While the ACL (9), in partnership with the NCOA as its techni- programs for older adults. The CDC’s Healthy Aging Research
cal assistance partner (29), helped mobilize the evidence-based Network has been instrumental in advancing science toward

Frontiers in Public Health | www.frontiersin.org 11 April 2015 | Volume 3 | Article 136

Ory and Smith Perspectives on evidence-based programing

action and policy (12, 13). Additionally, the CDC’s National CDSME Program Dissemination through the
Center for Chronic Disease and Prevention has played a major ARRA
role in advancing the study and application of self-management The American Recovery and Reinvestment Act of 2009 (ARRA)
support (34), while the National Center for Injury Prevention provided funds to disseminate CDSME programs in 45 states,
and Control has provided a framework for identifying and inter- Puerto Rico, and the District of Columbia between 2010 and 2012
vening upon modifiable risk factors to prevent falls in later life (57). This initiative afforded the opportunity to address several
(49). The program developers of some of the most tested and questions about the evolution of these programs over time and
widely available evidence-based programs for older adults have their dissemination in different populations and settings. The
recently come together to establish an Evidence-Based Leader- introductory article helps set the stage by overviewing the ARRA
ship Council (50). Envisioning even greater numbers of partic- initiative and reviewing methodological details about measure
ipants benefiting from evidence-based programs in the future, selection and data collection (47). While the database is large,
this council is developing an infrastructure to offer technical containing the first 100,000 participants in the ARRA initiative,
assistance in implementation, dissemination, marketing, train- there is only limited data about participant demographics, work-
ing efficiencies, licensing, and evaluation. In fall prevention, a shop characteristics, and participant attendance. Nevertheless, we
national network of State Fall Prevention Coalitions has been were able to address several practice- and policy-based research
developed to mobilize further awareness about the need for fall questions.
prevention, assist in the implementation of evidence-based pro- Even in this brief funding period, we see an evolution in
grams, and help set priorities for and implement needed system the national roll-out, with participant recruitment accelerated
change (51, 52). At the state level, state departments of pub- over time (58). This was likely enabled by the establishment of
lic health are working collaboratively to implement a variety of an improved delivery infrastructure. Not only were subsequent
evidence-based fall prevention strategies, many of which require cohorts of participants reached more quickly, later participants
partnerships across public health, aging, and health care sectors tended to be more diverse in terms of socioeconomic and health
(53). At the local level, volunteer program facilitators and program factors (58). Our explorations of the relationships between work-
participants are forging partnerships that help care providers and shop characteristics and program attendance revealed the com-
recipients (54). plexity of these relationships, which differed by delivery site rural-
It is heartening to see a variety of networks working together ity and type and also signaled the need to consider broader issues
to promote evidence-based programming that can make a dif- of program costs when determining ideal class sizes (59). There
ference in the lives of older adults and their caregivers. Comple- was confirmation in the value of 0 or orientation classes as a way of
mented by national stakeholders, these networks are providing the boosting class attendance (60). As expected, there was a variety of
needed research and programmatic infrastructure to accelerate different delivery settings that enabled community practitioners
the evidence-based movement. They are also identifying existing to reach large numbers of participants. As expected, different
policies that can facilitate or impede the broader dissemination delivery sites were employed in different geographic areas and
and sustainability of evidence-based programs for older adults and attracted different types of participants, which confirmed the
addressing them accordingly. importance of implementing evidence-based programs through
multiple channels for maximum reach and diversity (61).
The Value of Research: Dissemination, This dataset also offered researchers with opportunities to
Implementation, and Outcomes examine similarities and differences in recruitment and atten-
dance based on participant characteristics based on geographical
In this section, we address what is being learned from national, location (i.e., rural and underserved areas) as well as racial/ethnic
state, and local studies about the program dissemination and minority groupings (i.e., Asian, African American, and Hispanic).
implementation processes and health-related outcomes. These are From these efforts, we see that participants living in rural areas are
best characterized as translational or pragmatic research studies less likely to have evidence-based programs. Additionally, though
conducted in real-world settings (55, 56). The major questions individuals from rural areas represented a relatively small propor-
are often descriptive: (1) what do we know at a given point in tion of participants (25%), they experienced higher program com-
time about who is participating in evidence-based programs?; (2) pletion rates (62). With this national dataset, we were also able to
what do we know about factors associated with successful pro- get a rare glimpse of Asian American participants and factors asso-
grammatic completion?; (3) what is the extent to which intended ciated with their relatively high program completion rates (63).
outcomes are achieved?; and (4) how do these translational efforts An examination of urban-dwelling African American participants
compare to the original randomized clinical trials or controlled showed unique patterns of delivery and attendance, which can
studies? There is emerging research interest in understanding the beneficially inform future policy and practice efforts (64). A final
spatial distribution of programs relative to need, mechanisms analysis of factors associated with workshop enrollment and reten-
associated with program success, who is most likely to bene- tion based on workshop language among Hispanic participants
fit, and the cost-effectiveness of individual and bundled pro- suggested the need for increased community capacity to deliver
gramming. This research has led to the creation of guidebooks, Spanish-led workshops (65). A common theme across all these
checklists, and other tools that can help practitioners and policy analyses was the need for tailored interventions and strategies to
makers plan strategically and evaluate different evidence-based attract and retain more participants from underserved areas and
programs. minority backgrounds.

Frontiers in Public Health | www.frontiersin.org 12 April 2015 | Volume 3 | Article 136

Ory and Smith Perspectives on evidence-based programing

CDSME Program Implementation and Outcomes compendium of evidence-based programs (84), Tai-Chi and Step-
Adding to the emerging literature about the effectiveness of ping On, demonstrates the power of such programs to improve
CDSMP (66, 67), several articles investigated factors influencing the health and quality of life among older adults at risk for falling
program implementation and outcomes associated with the suite (85, 86). Further implementation was needed to prepare the Otago
of CDSME programs. Maintaining program fidelity is a major Education Program, a home-based fall prevention program, for
issue in program implementation, which can be facilitated by widespread dissemination. This preparation included the devel-
introducing and using streamlined fidelity checklists that pro- opment of an online training module for physical therapists (87).
vide guidance about processes before, during, and after program Broad public health dissemination of fall prevention programs
implementation (68). A case study approach with “successful” requires greater appreciation of fall-related risks and the pre-
implementers was employed to examine organizational factors ventability of falls. An evidence-based fall prevention curriculum
associated with long-term implementation of CDSMP in two for community health workers has been developed to enable
states (69). Findings suggested the importance of utilizing strate- trusted members of the community to spread the word about
gies for addressing both internal and external factors for enhanc- fall prevention strategies and link underserved populations to
ing organizational capacity to support evidence-based programs. evidenced-based programs (88).
Specifically testing the Scheirer’s framework for sustainability In addition to programs listed in the CDC falls prevention com-
(70, 71), another study examined factors necessary for sustain- pendium, there are evidence-based fall risk reduction programs.
ing CDSMP delivery with a more localized perspective. Similar Analyses on A Matter of Balance, an evidence-based program
sustainability factors were found, suggesting the importance of originally designed to enhance confidence in preventing and man-
strategies such as enhancing organizational readiness, promoting aging falls (i.e., falls efficacy) and reduce fear of falling, reveal the
program champions, providing technical assistance, and having mediating effect of increased physical activity on falls efficacy (89).
access to participants and funding streams (72). A related study demonstrates significant impacts on gait speed, a
Several articles are focused on the adaptation of CDSMP to dif- major risk factor for falling and institutionalization (90). Expand-
ferent settings and populations. Greater attention needs to be paid ing our knowledge about the general benefits of this intervention
to strategies for successfully adopting CDSMP in the workplace to different demographic and health subgroups, this subgroup
to meet the needs of persons not typically enrolled in CDSMP analysis suggests the importance of targeting specific populations.
programs (73). It is seen that the original CDSMP can be success- It also recommends future research examining the relationship of
fully adapted to new populations, such as cancer survivors (74). functional performance to more distal fall outcomes (90).
Additionally, self-management programs have been successfully Adaptations to a variety of physical activity programs for older
delivered in other countries such as China (75), Australia, and the adults are being further evaluated. Processes involved in the con-
United Kingdom (76), although it is important to understand how version of a practice-based lifestyle program to a formalized,
the socio-political context impacts the delivery and success of such testable evidence-based program are described (91). Such trans-
strategies. lations require an understanding of the benefits and challenges
In response to practice and policy concerns, researchers are of both approaches as related to balancing program reach and
starting to examine the cost-effectiveness of different evidence- sustainability. Studies on two adaptations of Fit and Strong! (43)
based programs. Building on prior research documenting the have been conducted to examine: (1) program processes and out-
cost-effectiveness of CDSMP (77), a related study examines the comes involved in adapting Fit and Strong! to a lay-led model (92);
cost-effectiveness of CDSMP in terms of impact on quality- and (2) the adaptation to a new population of cancer survivors
adjusted life years, demonstrating the added value of CDSMP (78). (93). The translated interventions’ ability to achieve many of the
Knowing that practitioners and policy makers value information previously reported outcomes shows the potency of evidence-
about program costs and cost-related outcomes, a user-friendly based behavior principles to different settings and populations
tool has been developed to help stakeholders customize national (94, 95). A case study of factors associated with the early adop-
estimates to their local situation (79). In anticipating further tion of enhance fitness in new settings reveals that many of the
cost-effectiveness studies, it is important to understand how cur- same strategies that have been used to promote sustainability
rent data might be linked to administrative health claims and of CDSMP, including assessing organizational readiness, under-
challenges such linkages might present (80). standing adoption across all phases from early to late, and develop-
ing new revenue streams, are also relevant to physical activity pro-
grams (96). As with CDSMP and other physical activity or fall pre-
Evidence-Based Fall Prevention, Physical
vention programs, the development of a fidelity tool for behavioral
Activity, and Mental Health Programs
health programs, such as PEARLS, is important for monitoring
In addition to the suite of CDSME programs, we invited articles
program implementation across settings and populations (97).
about other evidence-based programs that address major public
health issues facing the growing older adult population. With
the magnitude and impact of falls on older Americans, it was Cross-Cutting Perspectives for
especially salient to include evidence-based fall prevention pro- Evidence-Based Programing
grams (81–83).The CDC has been a leader in the implementation
and evaluation of a comprehensive approach to fall prevention, This Research Topic identifies many cross-cutting issues essen-
including both community and clinical approaches. A state-wide tial for understanding and enhancing evidence-based program
evaluation of two community-based programs listed in the CDC delivery, including perceptions of key stakeholders and lessons

Frontiers in Public Health | www.frontiersin.org 13 April 2015 | Volume 3 | Article 136

Ory and Smith Perspectives on evidence-based programing

learned from the field. With a growing emphasis on translational on their websites, there is a new Evidence2programs Toolkit and
research to address public health problems, there is now a pro- website designed to help community-based organizations navi-
liferation of dissemination and implementation frameworks to gate through the abundance of information about evidence-based
guide research, practice, and policy related to program planning programing (108).
and implementation. As an example, the ACL has organized
many of its initiatives around the RE-AIM Framework (14). An Conclusion
unknown issue is the actual uptake of this framework in the field.
A case study of the perceived utility of the RE-AIM Framework by Evidence-based programing for older adults has come of age. Past
state agency service providers and public health partners revealed successes in identifying evidence-based programing have led to
primarily positive endorsement of the framework for planning, new emphases into translating research into practice and policy.
implementation, and evaluation (98). However, there was some There are now dedicated efforts being made to understand and
concern about adopting the framework as a whole, which suggests incorporate best practices in building and sustaining programs
areas for further technical assistance and support. over time. This includes identifying and employing strategies that
As evidence-based programs roll-out nationally, there are ques- will improve delivery system infrastructure for enhancing partic-
tions about the ability and value of states and counties to imple- ipant recruitment to, and retention in, evidence-based programs.
ment multiple evidence-based programs. An early study showed Additionally, a national system is developing to track the spread of
that the majority of older adults lived in regions with access to programs across geographical areas and monitor key factors such
only one evidence-based program (99). Since different programs as delivery sites, participant characteristics, program attendance,
attract different populations, there is benefit in having multiple and even limited outcome measurements.
programs offered in a given community and an infrastructure for This Research Topic identifies forces mobilizing the evidence-
cross-training to help spread programs to populations who can based movement: perspectives from program developers regard-
benefit the most. ing their successes and remaining challenges; the strength of
A major theme throughout this collection of articles is large and small networks in implementing and disseminating an
the importance of engaging end users and diverse partners evidence-based approach across aging, public health, and medical
in the design and implementation of evidence-based pro- care sectors; factors influencing the dissemination, implementa-
grams. Thus, before implementing the STEADI Tool Kit, tion, and outcomes associated with CDSME programs; the emerg-
a clinically based fall prevention program, it was impor- ing literature specifying what is known about community-based
tant to assess health provider’s perceptions about falls among falls, physical activity, and behavioral health interventions; and
their older patients and their current fall prevention prac- cross-cutting issues in the field.
tices (100). This information is critical for understanding the This collection of articles can be seen as a reflection of the
barriers and facilitators when trying to introduce the Tool evidence-based programing of the past, present, and future.
Kit as a clinical resource for fall risk assessment, treatment, Dramatic progress has been made over the past three decades.
and referral. Further, non-traditional partners, such as the Yet, more attention is needed to monitor and understand the
YMCA who have similar missions and delivery systems as dynamic interplay between specific intervention components
traditional aging service providers, offer promising opportu- (e.g., type, duration, and intensity) and various health, health
nities for collaborative efforts to disseminate evidence-based care, and cost-related outcomes across different settings and pop-
programs (101). ulations. Having a better grasp on such information can guide
The importance of building strong linkages across aging, public and drive efforts to better target and tailor interventions for
health, and medical care sectors is becoming well-recognized specific populations and settings. We recommend that future
and is now built into many national and state initiatives (57). actions should be driven by a greater appreciation of interacting
Less appreciated are the roles of other sectors such as the educa- research, practice, and policy influences on the development,
tional system, which can help build a vibrant workforce for the implementation, dissemination, and sustainability of evidence-
implementation, evaluation, and dissemination of evidence-based based programs. It is our greatest hope that this Research Topic
programing (102). provides guidance to practitioners, stimulates new and unan-
Whereas the literature about the effectiveness of evidence- swered research questions, and informs policy decisions that
based programs for CDSME and fall prevention has blossomed in can help support and strengthen evidence-based programing for
recent years with many meta-analyses (103–106), little is known older adults.
about effective interventions for emotional health. A recent sys-
tematic review indicates strong evidence for skills training inter- Acknowledgments
ventions, calling for additional evidence for other social support
or physical activity intervention strategies (107). This Research Topic reflects our work on evidence-based pro-
Finally, whereas in the early years there was a lack of infor- graming for older adults supported by many sources includ-
mation about evidence-based programs, in some areas, there ing grant support from the Administration on Aging through
is now a profusion of information, making it hard for practi- the American Recovery and Reinvestment Act funding for the
tioners and policy makers to know where to retrieve reliable Chronic Disease Self-Management Program and other evidence-
information for selecting and implementing evidence-based pro- based disease prevention initiatives. Additionally, the National
grams. While the national stakeholders have excellent materials Council on Aging (NCOA) provided support through contracts

Frontiers in Public Health | www.frontiersin.org 14 April 2015 | Volume 3 | Article 136

Ory and Smith Perspectives on evidence-based programing

to Texas A&M Health Science Center. The National Institute on Control and Prevention network activities (i.e., Healthy Aging
Aging provided support for our Community Research Center Research Network) and State Fall Prevention Program Grants. All
for Senior Health. We also recognize support for our healthy conclusions are those of the authors and do not reflect official
aging and fall prevention activity through the Centers for Disease positions of the funding agencies.

References Available from: http://cphss.wustl.edu/Products/Documents/Bookchapter_

Luke_DisseminationandImplementationResearch.pdf
1. National Institute on Aging and World Health Organization. Global Health and 19. Glasgow RE, Vinson C, Chambers D, Khoury MJ, Kaplan RM, Hunter C.
Aging. NIH Publication No. 11-7737 (2011). Available from: http://www.who. National institutes of health approaches to dissemination and implementation
int/ageing/publications/global_health.pdf?ua=1 science: current and future directions. Am J Public Health (2012) 102:1274–81.
2. U.S. Department of Health and Human Services. Multiple Chronic Conditions: doi:10.2105/AJPH.2012.300755
A Strategic Frame-Work [Internet] (2014). Available from: http://www.hhs.gov/ 20. Institute of Medicine. The CTSA Program at NIH: Opportunities for
ash/initiatives/mcc/ Advancing Clinical and Translational Research. Available from: http://
3. Ory MG, Ahn S, Towne SD Jr, Smith ML. Chronic disease self-management www.iom.edu/Reports/2013/The-CTSA-Program-at-NIH-Opportunities-
education: program success and future directions. In: Malone M, Capezuti E, for-Advancing-Clinical-and-Translational-Research.aspx
Palmer RM, editors. Geriatrics Models of Care: Bringing ‘Best Practice’ to an 21. Ory MG, Hoffman M, Hawkins M, Sanner B, Mockenhaupt R. Challenging
Aging America. Cham: Springer International (2015). aging stereotypes: designing and evaluating physical activity programs. Am J
4. Ory MG, Towne S, Park CH, Chodzko-Zajko W. Practical strategies for imple- Prev Med (2003) 25(3S2):164–71. doi:10.1016/S0749-3797(03)00181-8
menting group exercise programs for older adults. In: Sullivan G, Pomidor 22. Ory MG, DeFriese GH, editors. Self Care in Later Life. New York, NY: Springer
A, editors. Exercise in Aging Adults: A Guide for Practitioners. New York, NY: Publishing Co (1998).
Springer International (2015). 23. U.S. Department of Health and Human Services. Administration on Aging.
5. Berwick DM, Nolan TW, Whittington J. The triple aim: care, health, and cost. Older Americans Act (OAA) Reauthorization [Internet] (2014). Available
Health Aff (2008) 27:759–69. doi:10.1377/hlthaff.27.3.759 from: http://www.aoa.acl.gov/AoA_Programs/OAA/Reauthorization/Index.
6. Ory MG, Ahn S, Jiang L, Smith ML, Ritter P, Whitelaw N, et al. Successes of a aspx
national study of the chronic disease self-management program: meeting the 24. Wellman NS, Kamp B, Kirk-Sanchez NJ, Johnson PM. Eat better & move more:
triple aim of healthcare reform. Med Care (2013) 51(11):992–8. doi:10.1097/ a community-based program designed to improve diets and increase physical
MLR.0b013e3182a95dd1 activity among older Americans. Am J Public Health (2007) 97(4):710–7.
7. O’ Neill G, Pruchno R. Toward the 2015 white house conference on aging: doi:10.2105/AJPH.2006.090522
creating an aging policy vision for the decade ahead. Gerontologist (2015) 25. Evidence-Based Behavioral-Practice. Available from: http://www.ebbp.org/
55(2):179–82. doi:10.1093/geront/gnv013 26. Ory MG, Jordan P, Bazzarre T. Behavioral change consortium: setting the stage
8. Gonzales E, Matz-Costa C, Morrow-Howell N. Increasing opportunities for for a new century of health behavior change research. Health Educ Res (2002)
the productive engagement of older adults: a response to population aging. 17(5):500–11. doi:10.1093/her/17.5.500
Gerontologist (2015) 55(2):252–61. doi:10.1093/geront/gnu176 27. Ory MG, Evashwick CJ, Glasgow RB, Sharkey JR. Pushing the boundaries
9. Boutaugh M, Lawrence L. Fostering healthy aging through evidence-based of evidence-based research: enhancing the application and sustainability of
prevention programs: perspectives from the administration for community health promotion programs in diverse populations. In: Browning CJ, Thomas
living/administration on aging. Front Public Health (2014) 2:236. doi:10.3389/ SA, editors. Behavioral Change: An Evidence-Based Handbook for Social and
fpubh.2014.00236 Public Health. Edinburgh: Elsevier Churchill (2005). p. 267–93.
10. U.S. Department of Health and Human Services. Administration for Commu- 28. Wilcox S, Dowda M, Griffin SF, Rheaume C, Ory MG, Leviton L, et al. Results
nity Living. About ACL [Internet] (2014). Available from: http://www.acl.gov/
About_ACL/Index.aspx
®
of the first year of active for life : translation of two evidence-based physical
activity programs for older adults in community settings. Am J Public Health
11. Boutaugh ML, Jenkins SM, Kulinski KP, Lorig KL, Ory MG, (2006) 96(7):1201–9. doi:10.2105/AJPH.2005.074690
Smith ML. Closing the diversity gap: the work of the administration on 29. Birkel R, Dessem E, Eldridge S, Kulinski K, Lachenmayr S, Spafford M,
aging, administration for community living. Generations (2015) 38(4): et al. Improving lives through evidence-based health promotion programs: a
107–18. national priority. Front Public Health (2014) 2:255. doi:10.3389/fpubh.2014.
12. Belza B, Altpeter M, Hooker SP, Moni G. The CDC healthy aging research 00255
network: advancing science toward action and policy for the evidence-based 30. Center for Healthy Aging. Using the Evidence Base to Promote Healthy Aging.
health promotion movement. Front Public Health (2104) 2:261. doi:10.3389/ Issue Brief. Number 1. Revised Spring (2006). Available from: http://www.ncoa.
fpubh.2014.00261 org/news-ncoa-publications/publications/issuebrief_1-r_usingeb.pdf
13. Wilcox S, Altpeter M, Anderson L, Belza B, Bryant L, Jones DL, et al. 31. Available from: http://www.ncoa.org/improve-health/center-for-healthy-
The healthy aging research network: building capacity for public health aging/online-training-modules/
and aging practice. Am J Health Promot (2013) 28(1):206. doi:10.4278/ajhp. 32. U.S. Department of Health and Human Services. Administration on Aging.
121116-CIT-564 Disease Prevention and Health Promotion Services (OAA Title IIID) [Internet]
14. Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of health (2014). Available from: http://www.aoa.acl.gov/AoA_Programs/HPW/Title_
promotion interventions: the RE-AIM framework. Am J Public Health (1999) IIID/index.aspx
89(9):1322–7. doi:10.2105/AJPH.89.9.1322 33. Available from: http://www.whitehouseconferenceonaging.gov/blog/policy/
15. Belza B, Glasgow R, Toobert D. RE-AIM for Program Planning: Overview and post/healthy-aging-policy-brief
Applications. Washington, DC: National Council on Aging (2007). 34. Brady TJ, Anderson L, Kobau R. Chronic disease self-management support:
16. Damschroder LJ, Aron DC, Keith RE, Kirsh SR, Alexander JA, Lowery JC. public health perspectives. Front Public Health (2014) 2:234. doi:10.3389/
Fostering implementation of health services research findings into practice: fpubh.2014.00234
a consolidated framework for advancing implementation science. Implement 35. Colligan EM, Tomoyasu N, Howell B. Community-based wellness and pre-
Sci (2009) 4(1):50. doi:10.1186/1748-5908-4-50 vention programs: the role of medicare. Front Public Health (2014) 2:189.
17. Brownson RC, Colditz GA, Proctor EK. Dissemination and Implementation doi:10.3389/fpubh.2014.00189
Research in Health: Translating Science to Practice. 1st ed. New York, NY: 36. Ellen Kullman M. Foundation engagement in healthy aging initiatives and
Oxford University Press (2012). evidence-based health promotion programs for older adults. Front Public
18. Brownson RC, Colditz GA, Proctor EK. Dissemination and Implementa- Health (2014) 2:190. doi:10.3389/fpubh.2014.00190
tion Research in Health: Translating Science to Practice. Oxford Univer- 37. Health Foundation of South Florida. Healthy Aging Initiative (2015). Available
sity Press (2012). 560 p. doi:10.1093/acprof:oso/9780199751877.001.0001 from: http://hfsf.org/healthy_aging.aspx

Frontiers in Public Health | www.frontiersin.org 15 April 2015 | Volume 3 | Article 136

Ory and Smith Perspectives on evidence-based programing

38. Lorig K. Chronic disease self-management program: insights from the eye of (CDSMP) workshops. Front Public Health (2014) 2:205. doi:10.3389/fpubh.
the storm. Front Public Health (2014) 2:253. doi:10.3389/fpubh.2014.00253 2014.00205
39. Haynes M, League P, Neault G. A matter of balance: older adults taking control 61. Smith ML, Ory MG, Ahn S, Belza B, Mingo CA, Towne SD Jr., et al. Reaching
of falls by building confidence. Front Public Health (2014) 2:274. doi:10.3389/ diverse participants utilizing a diverse delivery infrastructure: a replication
fpubh.2014.00274 study. Front Public Health (2015) 3:77. doi:10.3389/fpubh.2015.00077
40. Ellen Mahoney J. Stepping on: stepping over the chasm from research to 62. Towne SD, Smith ML, Ahn S, Ory MG. The reach of chronic disease self-
practice. Front Public Health (2014) 2:148. doi:10.3389/fpubh.2014.00148 management education programs to rural populations. Front Public Health
41. Shubert TE. Village or tribe? Expectations, roles, and responsibilities for (2014) 2:172. doi:10.3389/fpubh.2014.00172
effective fall prevention efforts. Front Public Health (2014) 2:163. doi:10.3389/ 63. Ahn S, Smith ML, Cho J, Jiang L, Post L, Ory MG. Factors associated
fpubh.2014.00163 with successful completion of the chronic disease self-management program
42. Thompson M, Snyder SJ, Denison P. Enhance fitness: a twenty year dissem- (CDSMP) among middle-aged and older Asian American participants: a
ination history. Front Public Health (2014) 2:270. doi:10.3389/fpubh.2014. national study. Front Public Health (2014) 2:257. doi:10.3389/fpubh.2014.
00270 00257
43. Hughes S, Smith-Ray RL, Shah A, Huber G. Translating fit and strong! – lessons 64. Mingo CA, Smith ML, Ahn S, Jiang L, Cho J, Towne SD, et al. Chronic dis-
learned and next steps. Front Public Health (2014) 2:131. doi:10.3389/fpubh. ease self-management education (CDSME) program delivery and attendance
2014.00131 among Urban-dwelling African Americans. Front Public Health (2014) 2:174.
44. Riley H. Texercise: the evolution of a health promotions program. Front Public doi:10.3389/fpubh.2014.00174
Health (2014) 2:262. doi:10.3389/fpubh.2014.00262 65. Smith ML, Ahn S, Jiang L, Kulinski K, Ory MG. Factors associated with
45. Snowden MB, Steinman LE, Piering P, Rigor S, Yip A. Translating PEARLS: hispanic adults attending Spanish-language disease self-management pro-
lessons learned from providers and participants. Front Public Health (2014) gram workshops and workshop completion. Front Public Health (2014) 2:155.
2:256. doi:10.3389/fpubh.2014.00256 doi:10.3389/fpubh.2014.00155
46. Lorig K, Laurin J, Holman HR. Arthritis self-management: a study of the effec- 66. Ory MG, Ahn S, Jiang L, Lorig K, Ritter P, Laurent D, et al. National study of
tiveness of patient education for the elderly. Gerontologist (1984) 24(5):455–7. chronic disease self-management: six month outcome findings. J Aging Health
doi:10.1093/geront/24.5.455 (2013) 5(3):1258–74. doi:10.1177/0898264313502531
47. Kulinski K, Boutaugh M, Smith ML, Ory MG, Lorig K. Setting the stage: 67. Ory MG, Ahn S, Jiang L, Smith ML, Whitelaw N, Ritter P, et al. Successes of a
measure selection, coordination, and data collection for a national self- national study of the chronic disease self-management program: meeting the
management initiative. Front Public Health (2014) 2:206. doi:10.3389/fpubh. triple aim of health care reform. Med Care (2013) 51(11):992–8. doi:10.1097/
2014.00206 MLR.0b013e3182a95dd1
48. Ory MG, Resnick B, Smith ML. Changing behavior throughout the life-course: 68. Ahn S, Smith ML, Altpeter M, Belza B, Post L, Ory MG. Methods for streamlin-
translating the success of aging research. Transl Behav Med (2012) 2(2):159–62. ing intervention fidelity checklists: an example from the chronic disease self-
doi:10.1007/s13142-012-0129-4 management program. Front Public Health (2014) 2:294. doi:10.3389/fpubh.
49. Kaniewski M, Stevens J, Parker E, Lee R. An introduction to the centers for 2014.00294
disease control and prevention’s (CDC) efforts to prevent older adult falls. 69. Paone D. Factors supporting implementation among CDSMP organizations.
Front Public Health (2014) 2:119. doi:10.3389/fpubh.2014.00119 Front Public Health (2014) 2:237. doi:10.3389/fpubh.2014.00237
50. Haynes M, Hughes S, Lorig K, Simmons J, Snyder SJ, Steinman L, et al. 70. Scheirer MA. Is sustainability possible? A review and commentary on empir-
Evidence-based leadership council – a national collaborative. Front Public ical studies of program sustainability. Am J Eval (2005) 26:320. doi:10.1177/
Health (2014) 2:136. doi:10.3389/fpubh.2014.00136 1098214005278752
51. Beattie BL. Working toward a multi-program strategy in fall prevention. Front 71. Scheirer MA, Dearing JW. An agenda for research on the sustainability of
Public Health (2014) 2:254. doi:10.3389/fpubh.2014.00254 public health programs. Am J Public Health (2011) 101(11):2059–67. doi:10.
52. Schneider EC, Beattie BL. Building the older adult fall prevention movement- 2105/AJPH.2011.300193
steps and lessons learned. Front Public Health (2014) 2:194. doi:10.3389/fpubh. 72. Tomioka M, Braun KL. Examining sustainability factors for organizations that
2014.00194 adopted Stanford’s chronic disease self-management program. Front Public
53. Thoreson SR. Public health system perspective on implementation of Health (2014) 2:140. doi:10.3389/fpubh.2014.00140
evidence-based fall prevention strategies for older adults. Front Public Health 73. Smith ML, Ory MG. Chronic disease self-management program (CDSMP)
(2014) 2:191. doi:10.3389/fpubh.2014.00191 in the workplace: opportunities for health improvement. Front Public Health
54. Rosemond CA. Falling for a balance partner. Front Public Health (2014) 2:130. (2014) 2:179. doi:10.3389/fpubh.2014.00179
doi:10.3389/fpubh.2014.00130 74. Risendal BC, Dwyer A, Lorig K, Seidel RW, Coombs L, Ory MG. Meeting
55. Glasgow RE, Magid DJ, Beck A, Ritzwoller D, Estabrooks PA. Practi- the public health challenge of cancer surviorship: results of the evaluation of
cal clinical trials for translating research to practice: design and measure- the chronic disease self-management program (CDSMP) for cancer survivors.
ment recommendations. Med Care (2005) 43(6):551–7. doi:10.1097/01.mlr. Front Public Health (2014) 2:214. doi:10.3389/fpubh.2014.00214
0000163645.41407.09 75. Browning CJ, Yang H, Zhang T, Chapman A, Liu S, Enticott J, et al. Imple-
56. Tunis SR, Stryer DB, Clancey CM. Practical clinical trials: increasing the value menting a chronic disease self-management program into China: happy life
of clinical research for decision making in clinical and health policy. JAMA club. Front Public Health (2014) 2:181. doi:10.3389/fpubh.2014.00181
(2003) 290:1624–32. doi:10.1001/jama.290.12.1624 76. Browning CJ, Thomas SA. Implementing chronic disease self-management
57. U.S. Administration on Aging. American Recovery and Reinvestment Act Com- approaches in Australia and the United Kingdom. Front Public Health (2014)
munities Putting Prevention to Work Chronic Disease Self-Management Pro- 2:162. doi:10.3389/fpubh.2014.00162
gram Announcement (2010). Available from: http://www.cfda.gov/index?s= 77. Ahn S, Basu R, Smith ML, Jiang L, Lorig K, Whitelaw N, et al. The impact
program&mode=form&tab=step1&id=5469a61f2c5f25cf3984fc3b94051b5f of chronic disease self-management programs: healthcare savings through a
58. Smith ML, Ory MG, Ahn S, Kulinski K, Jiang L, Lorig K. National dissemi- community-based intervention. BMC Public Health (2013) 13:1141. doi:10.
nation of chronic disease self-management education (CDSME) programs: an 1186/1471-2458-13-1141
incremental examination of delivery characteristics. Front Public Health (2014) 78. Basu R, Ory MG, Towne SD, Smith ML, Hochhalter A, Ahn S. Cost effec-
2:227. doi:10.3389/fpubh.2014.00227 tiveness of the chronic disease self-management program: implications for
59. Smith ML, Ory MG, Jiang L, Lorig K, Kulinski K, Ahn S. Workshop char- allocating community resources. Front Public Health (2014) 2:27. doi:10.3389/
acteristics related to chronic disease self-management education (CDSME) fpubh.2015.00027
program attendance. Front Public Health (2014) 2:19. doi:10.3389/fpubh.2015. 79. Ahn S, Smith ML, Altpeter M, Post L, Ory MG. Healthcare cost savings
00019 estimator tool for chronic disease self-management program (CDSMP): a new
60. Jiang L, Smith ML, Chen S, Ahn S, Kulinski K, Lorig K, et al. The role of session tool for program administrators and decision makers. Front Public Health
zero in successful completion of chronic disease self-management program (2014) 2:42. doi:10.3389/fpubh.2015.00042

Frontiers in Public Health | www.frontiersin.org 16 April 2015 | Volume 3 | Article 136

Ory and Smith Perspectives on evidence-based programing

80. Ritter PL, Ory MG, Smith ML, Jiang L, Alonis A, Laurent D, et al. Linking prevention initiative. Front Public Health (2014) 2:143. doi:10.3389/fpubh.
evidence-based program participants data with medicare data: the consenting 2014.00143
process and correlates of retrospective participant consents. Front Public 98. Towne SD, Smith ML, Ahn S, Altpeter M, Belza B, Kulinski K, et al. National
Health (2014) 2:176. doi:10.3389/fpubh.2014.00176 dissemination of multiple evidence-based disease prevention programs: reach
81. Towne SD, Smith ML, Yoshikawa A, Ory MG. Geospatial distribution of to vulnerable older adults. Front Public Health (2014) 2:156. doi:10.3389/
fall-related hospitalization incidence in Texas. J Safety Res (2015) 53:11–16. fpubh.2014.00156
doi:10.1016/j.jsr.2015.01.002 99. Smith ML, Stevens JA, Ehrenreich H, Wilson A, Schuster RJ, O’Brien Cherry
82. Stevens JA, Corso PS, Finkelstein EA, Miller TR. The costs of fatal and non- C, et al. Healthcare providers’ perceptions and self-reported fall prevention
fatal falls among older adults. Inj Prev (2006) 12(5):290–5. doi:10.1136/ip.2005. practices: findings from a large New York health system. Front Public Health
011015 (2014) 2:17. doi:10.3389/fpubh.2015.00017
83. Centers for Disease Control and Prevention. Cost of Falls Among Older Adults 100. Ehrenreich H, Pike M, Hohman K, Kaniewski M, Longjohn M, Myers G,
(2014). Available from: http://www.cdc.gov/homeandrecreationalsafety/falls/ et al. CDC and YMCA: a promising partnership for delivering fall preven-
fallcost.html tion programming. Front Public Health (2014) 2:235. doi:10.3389/fpubh.2014.
84. Centers for Disease Control and Prevention. CDC Compendium of 00235
Effective Fall Prevention Interventions: What Works for Community- 101. Frank JC. A missing piece in the infrastructure to promote healthy aging
Dwelling Older Adults. 2nd ed (2012). Available from: http://www.cdc. programs: education and work force development. Front Public Health (2014)
gov/HomeandRecreationalSafety/Falls/compendium.html 2:287. doi:10.3389/fpubh.2014.00287
85. Ory MG, Smith ML, Parker EM, Jiang L, Chen S, Wilson AD, et al. Fall 102. Gillespie LD, Robertson MC, Gillespie WJ, Sherrington C, Gates S, Clemson
prevention in community settings: results from implementing Tai Chi: moving LM, et al. Interventions for preventing falls in older people living in the
for better balance in three states. Front Public Health (2014) 2:258. doi:10.3389/ community. Cochrane Database Syst Rev (2012) 9:CD007146. doi:10.1002/
fpubh.2014.00258 14651858.CD007146.pub3
86. Ory MG, Smith ML, Jiang L, Lee R, Chen S, Wilson AD, et al. Fall 103. American Geriatrics Society, British Geriatrics Society. AGS/BGS Clinical
prevention in community settings: results from implementing stepping Practice Guideline: Prevention of Falls in Older Persons. New York, NY: Amer-
on in three states. Front Public Health (2014) 2:232. doi:10.3389/fpubh.2014. ican Geriatrics Society (2010).
00232 104. Brady TJ, Murphy LM, O’Colmain B, Beauchesne D, Daniels B, Greenberg
87. Shubert TE, Smith ML, Ory MG, Clarke C, Bomberger S, Roberts E, et al. M, et al. A meta-analysis of health status, health behaviors and healthcare
Translation of the Otago exercise program for adoption and implementa- utilization outcomes of the chronic disease self-management program. Prev
tion in the United States. Front Public Health (2014) 2:152. doi:10.3389/fpubh. Chronic Dis (2013) 10:120112. doi:10.5888/pcd10.120112
2014.00152 105. Anderson G. Chronic Care: Making the Case for Ongoing Care. Princeton,
88. St. John JA, Shubert TE, Smith ML, Rosemond CA, Howell D, Beaudoin C, NJ: Robert Wood Johnson Foundation (2010). Available from: www.rwjf.org/
et al. Developing an evidence-based fall prevention curriculum for commu- content/dam/farm/reports/reports/2010/rwjf54583
nity health workers. Front Public Health (2014) 2:209. doi:10.3389/fpubh.2014. 106. Snowden MB, Steinman L, Carlson WL, Mochan KN, Abraido-Lanza AF,
00209 Bryant LL, et al. Effect of physical activity, social support and skills training on
89. Cho J, Smith ML, Ahn S, Kim K, Appiah B, Ory MG. Effects of an late-life emotional health: a systematic literature review and implications for
evidence-based falls risk-reduction program on physical activity and falls public health research. Front Public Health (2014) 2:213. doi:10.3389/fpubh.
efficacy among oldest-old adults. Front Public Health (2014) 2:182. doi:10. 2014.00213
3389/fpubh.2014.00182 107. Stevens AB, McGhee R, Ory MG. EvidenceToPrograms.com: a toolkit to
90. Cho J, Smith ML, Shubert TE, Jiang L, Ahn S, Ory MG. Gait speed among support evidenced-based programming for seniors. Front Public Health (2014)
older participants enrolled in an evidence-based fall risk reduction program: a 2:18. doi:10.3389/fpubh.2015.00018
subgroup analysis. Front Public Health (2014) 2:26. doi:10.3389/fpubh.2015. 108. Glasgow RE, Lichtenstein E, Marcus AC. Why don’t we see more transla-
00026 tion of health promotion research to practice? rethinking the efficacy-to-
91. Ory MG, Smith ML, Howell D. The conversion of a practice-based lifestyle effectiveness transition. Am J Public Health (2003) 93(8):1261–7. doi:10.2105/
enhancement program into a formalized, testable program: from texercise AJPH.93.8.1261
classic to texercise select. Front Public Health (2014) 2:291. doi:10.3389/fpubh.
Conflict of Interest Statement: The authors declare that the research was
2014.00291
conducted in the absence of any commercial or financial relationships that could
92. Lee S, Ory MG, Zollinger A, Bhurtyal K, Jiang L, Smith ML. Translation of fit
be construed as a potential conflict of interest.
& strong! for middle-aged and older adults: examining implementation and
effectiveness of a lay-led model in central Texas. Front Public Health (2014)
This paper is included in the Research Topic, “Evidence-Based Programming for Older
2:187. doi:10.3389/fpubh.2014.00187
Adults.” This Research Topic received partial funding from multiple government and
93. Reynolds J, Thibodeaux L, Jiang L, Francis K, Hochhalter A. Adaptation of an
private organizations/agencies; however, the views, findings, and conclusions in these
evidence-based physical activity intervention, fit & strong!, promotes physical
articles are those of the authors and do not necessarily represent the official position of
activity and well being in older cancer survivors. Front Public Health (2014)
these organizations/agencies. All papers published in the Research Topic received peer
2:171. doi:10.3389/fpubh.2014.00171
review from members of the Frontiers in Public Health (Public Health Education and
94. Ory MG, Smith ML, Mier N, Wernicke M. The science of sustaining health
Promotion section) panel of Review Editors. Because this Research Topic represents
behavior change: the health maintenance consortium. Am J Health Behav
work closely associated with a nationwide evidence-based movement in the US,
(2010) 34(6):647–59. doi:10.5993/AJHB.34.6.2
many of the authors and/or Review Editors may have worked together previously in
95. Belza B, Prahova MP, Kohn M, Miyawaki CE, Farren L, Kline G, et al. Adoption
some fashion. Review Editors were purposively selected based on their expertise with
of evidence-based health promotion programs: perspectives of early adopters
evaluation and/or evidence-based programming for older adults. Review Editors were
of enhance fitness in YMCA-affiliated sites. Front Public Health (2014) 2:164.
independent of named authors on any given article published in this volume.
doi:10.3389/fpubh.2014.00164
96. Farren L, Snowden M, Steinman L, Monroe-DeVita M. Development and Copyright © 2015 Ory and Smith. This is an open-access article distributed under the
evaluation of a fidelity instrument for PEARLS. Front Public Health (2014) terms of the Creative Commons Attribution License (CC BY). The use, distribution or
2:200. doi:10.3389/fpubh.2014.00200 reproduction in other forums is permitted, provided the original author(s) or licensor
97. Ory MG, Altpeter M, Belza B, Helduser J, Zhang C, Smith ML. Perceived are credited and that the original publication in this journal is cited, in accordance with
utility of the RE-AIM framework for health promotion/disease prevention accepted academic practice. No use, distribution or reproduction is permitted which
initiatives for older adults: a case study from the U.S. evidence-based disease does not comply with these terms.

Frontiers in Public Health | www.frontiersin.org 17 April 2015 | Volume 3 | Article 136

 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00236

Fostering healthy aging through evidence-based

prevention programs: Perspectives from the
Administration for Community Living/Administration
on Aging
Michele L. Boutaugh 1 * and Laura J. Lawrence 2
1
Administration on Aging, Administration for Community Living, U.S. Department of Health and Human Services, Atlanta, GA, USA
2
Administration on Aging, Administration for Community Living, U.S. Department of Health and Human Services, Washington, DC, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: healthy aging, prevention, evidence-based programs, chronic disease self-management education, older adults, community programs

In 2012, the Administration for Commu- networks and their partners at the state completing medical forms); and required
nity Living (ACL) emerged as a new operat- and local community level. We are espe- fewer emergency department visits and
ing division within the U.S. Department of cially proud of the impact of these grants, hospitalizations (7).
Health and Human Services (HHS), bring- which have helped to forge aging and pub- These kinds of data have helped our
ing together the Administration on Aging lic health partnerships and to build a pro- grantees and their partners participate in
(AoA); the Administration on Intellectual gram delivery infrastructure in 48 U.S. various health care reform efforts. We
and Developmental Disabilities (AIDD), states and territories. This national infra- are currently supporting 22 grants, all
(formerly known as the Administration on structure has enabled over 264,000 indi- financed by the Affordable Care Act Pre-
Developmental Disabilities); and the Office viduals throughout the country to partici- vention and Public Health Fund (PPHF).
on Disability. The ACL name reflects both pate in evidence-based chronic disease self- Each of these grants enables state units
the aspirations of the people we serve and management education (CDSME), dia- on aging and state departments of health
our new mission to maximize the inde- betes self-management training, physical across the nation to achieve two impor-
pendence, well-being, and health of older activity, falls prevention, nutrition edu- tant goals. They increase the number of
adults, people with disabilities across the cation, and depression management pro- older adults and adults with disabilities
lifespan, and their families and caregivers grams (4). who complete CDSME programs to main-
(1). Consistent with that mission is a long- We are even more excited about what is tain or improve their health. They also help
standing commitment to the translation happening now and the potential that lies agencies build sustainable systems for con-
of evidence-based prevention programs ahead. The Patient Protection and Afford- tinuing to deliver such programs after the
from the research setting into community able Care Act of 2010 (5) has created grant period ends. As a result, many states
practice. many new demonstration projects aimed are utilizing diverse strategies to sustain
The Administration on Aging contin- at achieving the triple aim of healthcare their programs including embedding pro-
ues to administer the Older Americans Act reform: “better health, better health care, grams within other Affordable Care Act
(OAA) (2), which authorizes a national and lower costs”(6). These projects provide initiatives such as patient-centered med-
aging network and formula grants to states. a tremendous opportunity to demonstrate ical homes and Accountable Care Orga-
These grants fund a wide array of services the value of evidence-based community nizations; partnering with Medicaid and
including congregate and home-delivered programs. Programs like the Chronic Dis- other health insurance providers; collabo-
meals; transportation; personal and respite ease Self-Management Program (CDSMP) rating with Federally Qualified Health Cen-
care; dementia care; caregiver support ser- not only improve individual lives, but can ters,Veterans Administration Medical Cen-
vices; and programs to protect elder rights be tools to help achieve the “triple aim.” ters, and other healthcare organizations;
(see Figure 1). A portion of OAA fund- A recent national study of CDSMP, which and teaming with non-traditional partners
ing also supports health prevention and was partially supported by ACL, demon- such as the Department of Corrections and
promotion activities. As of 2012, Con- strated this potential value. Participants in behavioral health agencies (8).
gressional appropriations require that this the study showed significant improvements The PPHF CDSME grants are part
funding be used only for evidence-based in health (e.g., self-reported health, pain, of ACL’s larger vision to use partner-
programs (3). fatigue, and depression); experienced mea- ships to help reshape healthcare and build
Since 2003, AoA has also provided surable improvements in quality of care a person-centered, comprehensive system
competitive grants to support collabora- (e.g., patient–physician communication, that coordinates acute care, long-term care,
tions between the aging and public health medication compliance, and confidence and community services. For instance,

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 236 | 18
Boutaugh and Lawrence Perspectives from ACL/AoA

Administraon for
Community Living
Administraon on Aging

56 State Units, 618 Area Agencies &

264 Tribal Organizaons

20,000 Service Providers &

Hundreds of Thousands of Volunteers

Provide Older Americans Act Services and Supports to Help One in Five (11M) Seniors Remain at Home

6.4 million
224 24.5 29 million 138,000 4 million case 867,000 420,000
hours of
million million hours of in- caregivers management caregivers ombudsman
respite
meals rides home care trained hours assisted consultaons
care

FIGURE 1 | Aging network structure and Older Americans Act services are shown. Data from 2012 State Program Reports (SPR) and National Ombudsman
Reporting System (NORS). Available from: http://www.agid.acl.gov/DataGlance/

ACL, the Centers for Medicare & Medicaid While proud of what we have achieved, http://www.aoa.acl.gov/AoA_Programs/HPW/
Services, and the U.S. Department of Vet- we are also mindful of the challenges that Title_IIID/index.aspx
4. Kulinski KP, Boutaugh M, Smith ML, Ory MG,
erans Affairs have partnered to invest in a lie ahead. Our goal is to make these pro-
Lorig K. Setting the stage: measure selection, coor-
national framework called Aging and Dis- grams universally accessible. We have made dination, and data collection for a national self-
ability Resource Centers (ADRCs). These great progress, but there are still gaps in management initiative. Front Public Health (2015)
centers provide a widely accessible “no our coverage. We cannot reach the millions 2:206. doi:10.3389/fpubh.2014.00206
wrong door” system, where older adults, that we still need to reach on our own. We 5. Public Law 111 – 148 – Patient Protection
and Affordable Care Act, 42 U.S.C. §18001
people with disabilities, and veterans of are continuing to work with researchers,
(2010). Available from: http://www.gpo.gov/fdsys/
all ages can learn about and access a full foundations, national organizations, and pkg/PLAW-111publ148/content-detail.html
range of long-term care services and sup- advocacy groups to strengthen our capac- 6. Berwick DM, Nolan TW, Whittington J. The triple
ports (9). Many ADRCs are working with ity to partner with health care entities and aim: care, health, and cost. Health Aff (2008)
hospitals and other partners on care transi- managed care plans. We are also contin- 27:759–69. doi:10.1377/hlthaff.27.3.759
7. Ory MG, Ahn S, Jiang L, Smith ML, Ritter P,
tion programs to better manage discharges ually exploring effective ways to integrate Whitelaw N, et al. Successes of a national study
from hospital to home or other care set- community-based organizations into new of the chronic disease self-management program:
tings, and are serving as centralized refer- delivery and financing models. ACL is com- meeting the triple aim of health care reform.
ral sources for the CDSMP workshops and mitted to pursuing every opportunity to Med Care (2013) 51(11):992–8. doi:10.1097/MLR.
0b013e3182a95dd1
other evidence-based programs. sustain and expand support for evidence- 8. U.S. Department of Health and Human Services.
We continue to collaborate with other based prevention programs to improve Administration on Aging. 2012 Prevention and Pub-
federal and private agencies to address the lives of older adults and people with lic Health Funds: Empowering Older Adults and
the HHS Strategic Framework on Multi- disabilities. Adults with Disabilities through Chronic Disease
ple Chronic Conditions in bringing to scale Self-Management Education Programs [Internet]
(2014). Available from: http://www.aoa.acl.gov/
and enhancing sustainability of evidence- REFERENCES AoA_Programs/HPW/ARRA/PPHF.aspx
based self-management programs (10). 1. U.S. Department of Health and Human Services. 9. U.S. Department of Health and Human Ser-
And in September 2014, we released 14 Administration for Community Living. About ACL vices. Administration for Community Living.
state and tribal falls prevention grants and [Internet] (2014). Available from: http://www.acl. Center for Disability and Aging Policy, Office
gov/About_ACL/Index.aspx of Integrated Programs, Aging and Disability
a new National Falls Prevention Resource 2. U.S. Department of Health and Human Ser- Resource Centers Program [Internet] (2014). Avail-
Center award financed by the Affordable vices. Administration on Aging. Older Amer- able from: http://www.acl.gov/Programs/CDAP/
Care Act PPHF. This new grant program icans Act (OAA) Reauthorization [Internet] OIP/ADRC/index.aspx
will increase access to evidence-based com- (2014). Available from: http://www.aoa.acl.gov/ 10. U.S. Department of Health and Human Services.
munity programs to reduce falls and falls AoA_Programs/OAA/Reauthorization/Index.aspx Multiple Chronic Conditions: A Strategic Frame-
3. U.S. Department of Health and Human work [Internet]. (2014). Available from: http://
risk while also increasing the sustainabil- Services. Administration on Aging. Disease www.hhs.gov/ash/initiatives/mcc/
ity of such programs through innovative Prevention and Health Promotion Services (OAA 11. U.S. Department of Health and Human Ser-
funding arrangements (11). Title IIID) [Internet] (2014). Available from: vices. Administration for Community Living.

www.frontiersin.org April 2015 | Volume 2 | Article 236 | 19

Boutaugh and Lawrence Perspectives from ACL/AoA

Announcement. ACL Funds Evidence-Based Falls in the Research Topic received peer review from members programs: Perspectives from the Administration for Com-
Prevention Grants [Internet] (2014). Avail- of the Frontiers in Public Health (Public Health Edu- munity Living/Administration on Aging. Front. Public
able from: http://www.acl.gov/NewsRoom/Press_ cation and Promotion section) panel of Review Editors. Health 2:236. doi: 10.3389/fpubh.2014.00236
Releases/archive_ACL/2014/2014_09_25a.aspx Because this Research Topic represents work closely asso- This article was submitted to Public Health Education
ciated with a nationwide evidence-based movement in and Promotion, a section of the journal Frontiers in
Conflict of Interest Statement: The authors declare
the US, many of the authors and/or Review Editors may Public Health.
that the research was conducted in the absence of any
have worked together previously in some fashion. Review Copyright © 2015 Boutaugh and Lawrence. This is
commercial or financial relationships that could be
Editors were purposively selected based on their expertise an open-access article distributed under the terms of
construed as a potential conflict of interest.
with evaluation and/or evidence-based programming for the Creative Commons Attribution License (CC BY).
This paper is included in the Research Topic, “Evidence- older adults. Review Editors were independent of named The use, distribution or reproduction in other forums
Based Programming for Older Adults.” This Research authors on any given article published in this volume. is permitted, provided the original author(s) or licen-
Topic received partial funding from multiple government sor are credited and that the original publication
and private organizations/agencies; however, the views, Received: 16 June 2014; accepted: 28 October 2014; in this journal is cited, in accordance with accepted
findings, and conclusions in these articles are those of the published online: 27 April 2015. academic practice. No use, distribution or reproduc-
authors and do not necessarily represent the official posi- Citation: Boutaugh ML and Lawrence LJ (2015) Fos- tion is permitted which does not comply with these
tion of these organizations/agencies. All papers published tering healthy aging through evidence-based prevention terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 236 | 20
 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00255

Improving lives through evidence-based health promotion

programs: a national priority
Richard Birkel , Emily Dessem, Simona Eldridge, Kristie Kulinski *, Sue Lachenmayr , Michelle Spafford ,
Binod Suwal , Albert Terrillion, Mary Walsh and Wendy Zenker
National Council on Aging, Washington, DC, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: evidence-based programs, healthy aging, self-management, program sustainability, older adults

The National Council on Aging (NCOA) the Patient Protection and Affordable Care As a resource center, we recognize how
has set forth on an ambitious course to Act (6) (ACA) in 2010 bolstered efforts essential it is to develop the acumen and
improve the health and economic secu- already underway to engage those pro- skills necessary to form meaningful and
rity of 10 million older adults by 2020. viding and paying for health care. In mutually beneficial relationships for pro-
Specific to older adult health, the promo- addition to increasing the quality and gram reimbursement, and are commit-
tion of proven, cost effective programs is affordability of health insurance and low- ted to working with our national net-
vital to our collective success. Through our ering the rate of uninsured individuals, work of partners to expand this knowledge
Center for Healthy Aging (CHA), we col- a number of ACA initiatives focus on base.
laborate with federal, state, and community improving patient outcomes and satisfac- To this end, CHA formed the
partners to further the impact and sus- tion. Evidence-based programs, particu- Community-Integrated Health Care
tainability of evidence-based health pro- larly those emphasizing self-management, Workgroup in early 2014 to assist our
motion programs. Our passion for this are well-positioned to serve as the carrot network in their efforts to obtain reim-
work, coupled with our reputation as a bridging community-based organizations, bursement to sustain implementation of
valued resource for organizations offering which have a history of successful and evidence-based programs. Participants
these programs, has afforded us the priv- efficient program delivery, with the newly include members of the aging services
ilege of serving as a technical assistance incentivized health care sector. Further network from area agencies on aging,
resource center for the U.S. Administration supporting the value of this collaboration senior resource centers, and other settings.
on Aging (AoA) since CHA’s inception (1). is a recent national study of Stanford Uni- The objectives for the group are to: (1)
In addition, CHA has been the leader of the versity’s Chronic Disease Self-Management develop specific definitions and parame-
Falls Free©Initiative (2), a national collab- Program (7), finding many significant ters of community-integrated health care;
orative effort to reduce falls among older improvements aligned with the Institute of (2) promote best practices in community-
adults. Healthcare Improvement’s Triple Aims of based organization/health care integration
With more than 10,000 baby boomers better health, better care, and lower cost occurring in the aging network; and (3)
turning 65 each day (3), leveraging pre- (8, 9). identify barriers to this integration and
cious resources and cultivating innova- Evidence-based programs are being potential steps address them.
tive partnerships is critical for making a implemented in nearly every state, with In addition to various activities that
population health impact. As a national hundreds of thousands of participants fall under the scope of our role
resource center, CHA identifies, develops, benefiting. In light of dwindling fed- as AoA’s national resource center on
and disseminates best practices and tools eral resources to support these programs chronic disease self-management educa-
for use by program implementers. We and recognizing that there is no single tion (CDSME) programs, NCOA’s Self-
have a rich history of successful collabora- “golden ticket” to program sustainabil- Management Alliance (SMA) is also a key
tion with aging services organizations, and ity, program implementers are looking conduit as we work toward our goal of
over the past few years have thoughtfully instead to a variety of blended funding improving the health of millions of older
expanded our network to include health streams. The rapid changes in health care adults. The SMA promotes strategic col-
care organizations and other private sector delivery under the ACA have afforded an laboration among government, business,
partners. opportunity to integrate evidence-based and non-profit organizations to achieve
Citing solid evidence, proponents of programs into developing health systems the goal of making evidence-based self-
evidence-based health promotion pro- and initiatives such as Accountable Care management an integral part of health
grams have long asserted that these inter- Organizations, Managed Care Organiza- care. It fosters information sharing, con-
ventions have a positive impact on health tions, Community-Based Care Transitions, sensus development, research and demon-
and wellness (4, 5), and the passage of and Patient-Centered Medical Homes. strations, communications, and public

www.frontiersin.org April 2015 | Volume 2 | Article 255 | 21

Birkel et al. Improving lives through evidence-based programs

policy in support of nationwide scaling to boost participant engagement require 9. Berwick DM, Nolan TW, Whittington J. The triple
of self-management and other evidence- additional research and experimentation. aim: care, health, and cost. Health Aff (2008)
27:759–69. doi:10.1377/hlthaff.27.3.759
based programs. The SMA is involved in Developing the capacity to offer programs
a series of efforts to better understand and on a consistent basis with broad geographic
delineate the value proposition of CDSME reach is critically important to the success Conflict of Interest Statement: The authors declare
that the research was conducted in the absence of any
implementation for health systems and to of our partnership with health care organi-
commercial or financial relationships that could be
identify “building blocks” for integrated zations; a statewide system with the capa- construed as a potential conflict of interest.
community health systems with the goal bility to deliver programs to their members
of sustainable reimbursement for CDSME within a reasonable period of time and This paper is included in the Research Topic, “Evidence-
Based Programming for Older Adults.” This Research
programs. within close proximity to where they live
Topic received partial funding from multiple government
We also recognize the value of taking a and work is expected. We are confident and private organizations/agencies; however, the views,
“two venue” approach to program imple- that within these challenges exist opportu- findings, and conclusions in these articles are those of the
mentation, with participants able to select nities for further innovation, collaboration, authors and do not necessarily represent the official posi-
in-person or online workshops. NCOA dis- and impact, and are excited about what lies tion of these organizations/agencies. All papers published
in the Research Topic received peer review from members
tributes the online suite of Stanford Univer- ahead. of the Frontiers in Public Health (Public Health Edu-
sity’s self-management programs, known cation and Promotion section) panel of Review Editors.
as Better Choices, Better Health® (includ- REFERENCES Because this Research Topic represents work closely asso-
ing variants specific to diabetes, arthri- 1. Center for Healthy Aging. (2014). Available from: ciated with a nationwide evidence-based movement in
tis, cancer survivors, and caregiving). The http://www.ncoa.org/improve-health/center-for- the US, many of the authors and/or Review Editors may
healthy-aging/ have worked together previously in some fashion. Review
opportunity to enroll in a workshop in- 2. Falls Free©Initiative. (2014). Available from: http:// Editors were purposively selected based on their expertise
person or online is a considerable value- www.ncoa.org/improve-health/center-for-healthy- with evaluation and/or evidence-based programming for
add for organizations, allowing them to aging/falls-prevention/falls-free-initiative.html older adults. Review Editors were independent of named
cast a wider net as they engage partners and 3. Pew Research Center. Baby Boomers Retire. (2010). authors on any given article published in this volume.
Available from: http://www.pewresearch.org/daily-
participants alike.
number/baby-boomers-retire/ Received: 16 June 2014; accepted: 09 November 2014;
At NCOA, we are committed to provid- 4. Beattie BL, Whitelaw N, Mettler M, Turner D. A published online: 27 April 2015.
ing national technical assistance, informed vision for older adults and health promotion. Am Citation: Birkel R, Dessem E, Eldridge S, Kulinski
leadership, and strategic resources to J Health Promot (2003) 18(2):200–4. doi:10.4278/ K, Lachenmayr S, Spafford M, Suwal B, Terrillion
advance the implementation and sustain- 0890-1171-18.2.200 A, Walsh M and Zenker W (2015) Improving lives
5. Bryant LL, Altpeter A, Whitelaw NA. Evaluation through evidence-based health promotion programs:
ability of evidence-based health promo-
of health promotion programs for older adults: an a national priority. Front. Public Health 2:255. doi:
tion programs. A number of challenges introduction. J Appl Gerontol (2006) 25(3):197–213. 10.3389/fpubh.2014.00255
remain, and working with our large net- doi:10.1177/0733464806288562 This article was submitted to Public Health Education
work of partners to identify feasible solu- 6. Patient Protection and Affordable Care Act, 42 U.S.C. and Promotion, a section of the journal Frontiers in
tions is at the top of our agenda. To §18001. (2010). Public Health.
7. Ory MG, Smith ML, Ahn S, Jiang L, Kulinski KP, Copyright © 2015 Birkel, Dessem, Eldridge, Kulinski,
meet the demands of an aging popula- Jiang N. National Study of the Chronic Disease Self- Lachenmayr, Spafford, Suwal, Terrillion, Walsh and
tion and ensure access to these proven Management Program: A Brief Overview. Washing- Zenker. This is an open-access article distributed under
programs, a robust workforce of program ton, DC: National Council on Aging (2013). the terms of the Creative Commons Attribution License
facilitators is necessary. Given the para- 8. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, (CC BY). The use, distribution or reproduction in other
mount challenge of health systems and Whitelaw N, et al. Success of a National Study of forums is permitted, provided the original author(s) or
the Chronic Disease Self-Management Program: licensor are credited and that the original publication in
societies globally to support positive behav- meeting the triple aim of health Care Reform. this journal is cited, in accordance with accepted aca-
ior change in an effort to tackle the pre- Med Care (2013) 51(11):992–8. doi:10.1097/MLR. demic practice. No use, distribution or reproduction is
ventable causes of chronic illness, strategies 0b013e3182a95dd1 permitted which does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 255 | 22
 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00119

An introduction to the Centers for Disease Control and

Prevention’s efforts to prevent older adult falls
Margaret Kaniewski *, Judy A. Stevens, Erin M. Parker and Robin Lee
Division of Unintentional Injury Prevention, Centers for Disease Control and Prevention, National Center for Injury Prevention and Control, Atlanta, GA, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: CDC, older adults, fall prevention, opinion, evidence-based programming

The Centers for Disease Control and Pre- developed a clinical practice guideline that Otago Exercise Program (7, 8), YMCA’s
vention’s National Center for Injury Pre- (1) encourages providers to conduct fall Moving for Better Balance program, and
vention and Control (CDC) envisions a risk assessments to identify patients who the Tai Chi for Arthritis program (9).
society where older adults (persons 65 and are at risk of falling and (2) describes These state grantees are also leverag-
older) can live long, safe, and healthy lives. evidence-based interventions that can be ing additional resources from the Area
Falls are a threat to older adults’ health and incorporated into a patient’s plan of care Agencies on Aging (AAA), senior services
can significantly limit their ability to live (4). Recommended interventions include network, the YMCA, and other commu-
independently. One in three older adults interventions delivered in clinical settings, nity programs developed for seniors.
falls each year, resulting in over $30 billion (e.g., medication review and modification, • Within the clinical setting, the CDC
in direct medical costs (1). gait and balance assessment with referral grantees and their partners are help-
For more than 20 years, CDC has been to physical therapy), as well as participa- ing healthcare providers implement the
conducting research to help prevent falls tion in community-based fall prevention AGS/BGS clinical practice guideline by
and resulting injuries among older adults. programs. Linking clinical medicine to providing the STEADI (Stopping Elderly
Research has identified important and community fall prevention programs can Accidents, Deaths and Injuries) tool kit
modifiable risk factors. These include mus- be an important step in improving uptake (10). Based on the AGS/BGS guideline,
cle weakness, gait and balance problems, of evidence-based practices to prevent the STEADI tool kit gives clinicians the
psychoactive medication use, poor vision, older adult falls. tools that they need to conduct standard-
and environmental hazards (2). Building Based on this information, CDC devel- ized fall risk assessments and recommend
on this knowledge, various falls interven- oped a fall prevention approach that inte- appropriate interventions. In addition to
tions have been developed and tested. A grates clinical practice and evidence-based addressing a patient’s specific fall risk fac-
recent Cochrane Review (3) identified community fall prevention programs. The tors, such as hypotension and underlying
159 randomized controlled trials (RCT) approach expands current health care prac- chronic conditions, suitable patients may
of falls interventions that included nearly tice by supporting providers in making fall be referred to community fall prevention
80,000 participants. This meta-analysis prevention a routine part of clinical care programs based on their level of fall risk,
found that group exercise programs (e.g., and encouraging providers to link clin- as shown in Figure 1.
Tai Chi), home-based exercise programs ical practice with community-based fall
(e.g., Otago), and home safety modifica- prevention programs. To this end, CDC’s Scaling up and sustaining this approach
tions (e.g., installing bathroom grab bars), Injury Center provides targeted techni- is challenging and requires bringing
combined with behavioral changes rec- cal and programmatic assistance to sev- health care and public health together.
ommended by an occupational therapist, eral state health departments and med- Indeed, healthcare management organi-
significantly reduced falls among older ical providers to help them implement fall zations, health care plans, health care
adults. Implementing these interventions prevention programs and measure impact. providers, state health departments, and
on a large scale and increasing older adults’ community organizations all have a role in
access to these interventions can prevent a • Within the community, CDC-funded this integrated approach to fall prevention.
substantial number of falls and fall-related grantees – the Oregon Health Authority, The combined contribution of all these sec-
injuries. New York State Department of Health, tors helps expand reach, reduce barriers
Medical providers can play an impor- and Colorado Health Department of to implementing clinical and community
tant role by identifying older adults who Public Health and Environment – are approaches, and maximize public health
are likely to fall and providing clinical pre- implementing evidence-based programs impact.
ventive services to help reduce fall risks. that reduce older adult falls. Supported To help make older adult fall preven-
To aid medical providers, the American programs include Tai Chi: Moving for tion a routine part of clinical care, CDC
and British Geriatrics Societies (AGS/BGS) Better Balance (5), Stepping On (6), the is supporting efforts to increase market

www.frontiersin.org April 2015 | Volume 2 | Article 119 | 23

Kaniewski et al. CDC’s fall prevention efforts

High Risk Otago

Exercise
Program

Moderate Risk
Stepping On
Program

Low Risk Tai Chi

Programs

FIGURE 1 | Linking clinical care with community programs based on an older adult’s risk level.

penetration of the STEADI tool kit and to As the U.S. population ages, fall injuries 7. Cambell AJ, Robertson MC, Gardner MM. Ran-
scale up its use by health care providers. will increase (11). The efforts of the domised controlled trial of a general practice
programme of home based exercise to prevent falls
For example, to accomplish this, CDC’s CDC, state health departments, AAAs,
in elderly women. Br Med J (1997) 315:1065–9.
Injury Center is creating electronic clini- researchers, advocacy organizations, pro- doi:10.1136/bmj.315.7115.1065
cal decision support modules that can be fessional organizations, health care pro- 8. Campbell AJ, Robertson MC, Gardner MM. Falls
adopted by most electronic health record fessionals, and many others are critical to prevention over 2 years: a randomized controlled
(EHR) systems. The goal is to integrate reducing older adult falls. CDC’s efforts trial in women 80 years and older. Age Ageing
(1999) 28(6):513–8. doi:10.1093/ageing/28.6.513
fall prevention activities into EHR sys- and the contributions described in this
9. Voukelatos A, Cumming RG, Lord SR, Rissel C.
tems so that users can efficiently man- journal issue will help further fall preven- A randomized, controlled trial of a Tai Chi for
age patient workflow, care, referrals, and tion research and practice. Policy makers the prevention of falls: the Central Sydney Tai
billing. and practitioners should find this issue Chi trial. J Am Geriatr Soc (2007) 55(8):1185–91.
The STEADI EHR modules will incor- helpful in improving and increasing their doi:10.1111/j.1532-5415.2007.01244.x
10. Stevens JA, Phelan EA. Development of STEADI: a
porate the Centers for Medicare and Med- efforts to prevent older adult falls. fall prevention resource for health care providers.
icaid Services (CMS) incentive programs, Health Promot Pract (2013) 14(5):706–14. doi:10.
which reward providers and health care REFERENCES 1177/1524839912463576
organizations for screening patients for 1. Stevens JA, Corso PS, Finkelstein EA, Miller TR. 11. Centers for Disease Control and Prevention,
fall risk and implementing fall preven- The costs of fatal and non-fatal falls among older National Center for Injury Prevention and Con-
adults. Injury Prevent (2006) 12(5):290–5. doi:10. trol. Web–based Injury Statistics Query and Report-
tion strategies for their high-risk patients. 1136/ip.2005.011015 ing System (WISQARS) [Online]. (2010).
For providers dedicated to promoting 2. Rubenstein LZ. Falls in older people: epidemiol-
the health and well-being of their older ogy, risk factors and strategies for prevention. Age Conflict of Interest Statement: The findings and con-
patients, this offers an opportunity to Ageing (2006) 35(Suppl 2):ii37–41. doi:10.1093/ clusions in this commentary are those of the authors
ageing/afl084 and do not necessarily represent the official position
receive incentive payments for their efforts
3. Gillespie LD, Robertson MC, Gillespie WJ, Sher- of the Centers for Disease Control and Prevention.
to deliver evidence-based health care. rington C, Gates S, Clemson LM, et al. Inter-
Additionally, EHR modules will incorpo- ventions for preventing falls in older people liv- This paper is included in the Research Topic, “Evidence-
rate the reimbursable ICD-10 diagnos- ing in the community. Cochrane Database Syst Based Programming for Older Adults.” This Research
tic codes that clinicians will be able to Rev (2012) 9:CD007146. doi:10.1002/14651858. Topic received partial funding from multiple government
CD007146.pub3 and private organizations/agencies; however, the views,
use when addressing fall risk with their
4. American Geriatrics Society, British Geriatrics findings, and conclusions in these articles are those of the
patients. Society. AGS/BGS Clinical Practice Guideline: Pre- authors and do not necessarily represent the official posi-
A STEADI online training course will vention of Falls in Older Persons. New York: Amer- tion of these organizations/agencies. All papers published
soon be available that will teach clinicians ican Geriatrics Society (2010). in the Research Topic received peer review from members
(physicians, physician assistants, and nurse 5. Li F, Harmer P, Mack KA, Sleet D, Fisher KJ, of the Frontiers in Public Health (Public Health Edu-
Millet LM, et al. Tai Chi: moving for better bal- cation and Promotion section) panel of Review Editors.
practitioners) to conduct fall risk assess- ance – development of a community-based falls Because this Research Topic represents work closely asso-
ments and recommend appropriate inter- prevention program. J Phys Act Health (2008) ciated with a nationwide evidence-based movement in
ventions using materials from the STEADI 5(3):445–55. the US, many of the authors and/or Review Editors may
tool kit. The online training will also pro- 6. Clemson L, Cumming RG, Kendig H, Swann have worked together previously in some fashion. Review
vide information on the EHR modules M, Heard R, Taylor K. The effectiveness of a Editors were purposively selected based on their expertise
community-based program for reducing the inci- with evaluation and/or evidence-based programming
and information on how medical staff can dence of falls in the elderly: a randomized trial. for older adults. Review Editors were independent of
operationalize the EHR modules in their J Am Geriatr Soc (2004) 52(9):1487–94. doi:10. named authors on any given article published in this
practice. 1111/j.1532-5415.2004.52411.x volume.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 119 | 24
Kaniewski et al. CDC’s fall prevention efforts

Received: 02 July 2014; accepted: 31 July 2014; published This article was submitted to Public Health Education The use, distribution or reproduction in other forums is
online: 27 April 2015. and Promotion, a section of the journal Frontiers in permitted, provided the original author(s) or licensor are
Citation: Kaniewski M, Stevens JA, Parker EM and Lee R Public Health. credited and that the original publication in this journal
(2015) An introduction to the Centers for Disease Control Copyright © 2015 Kaniewski, Stevens, Parker and Lee. is cited, in accordance with accepted academic practice.
and Prevention’s efforts to prevent older adult falls. Front. This is an open-access article distributed under the terms No use, distribution or reproduction is permitted which
Public Health 2:119. doi: 10.3389/fpubh.2014.00119 of the Creative Commons Attribution License (CC BY). does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 119 | 25

 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00189

Community-based wellness and prevention programs: the

role of Medicare
Erin Murphy Colligan 1 *, Naomi Tomoyasu 1 and Benjamin Howell 2
1
Centers for Medicare and Medicare Services, Baltimore, MD, USA
2
CVS Health, Woonsocket, RI, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: Medicare, wellness, community-based programs, Affordable Care Act, aging, fitness, falls prevention, self-management

Community-based wellness and preven- beneficiaries, Congress called upon the conditions. These findings suggest that
tion programs have long served to address Secretary of Health and Human Services physical fitness may be a critical mecha-
the needs of an aging population with mul- to evaluate these programs under Section nism through which to achieve benefits in
tiple chronic diseases. Title IIID of the 4202(b) of the Affordable Care Act. To health, utilization, and cost outcomes.
Older Americans Act, passed in 1987, called address this statute, CMMI first directed an The next phase of research under the
for the Administration on Aging (AoA) evidence review and environmental scan of 4202(b) legislation is currently underway:
to fund “education and implementation existing wellness programs, which Altarum a prospective evaluation of new partici-
activities that support healthy lifestyles and Institute completed in 2011. Altarum rated pants in wellness programs. Acumen LLC,
promote healthy behaviors (1).” AoA, now the strength of peer-reviewed literature in partnership with Westat, is conducting
a part of the Administration for Commu- surrounding a variety of wellness and pre- this study, with initial results expected in
nity Living (ACL), has continued to review vention programs and found several with 2016. While the retrospective study showed
the evidence base for wellness and preven- the potential to benefit older adults (4). some promising results, it was limited
tion programs and launched the Evidence- The results of this first phase of research by the inherent selection bias in benefi-
Based Prevention Program in 2003 to informed the selection of promising pro- ciaries who voluntarily enrolled in well-
increase access to such programs for older grams for the second phase of research, ness programs. Although Acumen matched
adults (1, 2). The National Council on a retrospective analysis of claims-based program participants to non-participants
Aging, in conjunction with AoA, operates a outcomes for Medicare recipients who based on clinical and sociodemographic
clearing house of evidence supporting well- participated in select wellness interven- factors as well as cost and utilization pat-
ness interventions and provides technical tions from 1999 to 2012. Acumen LLC terns, there still may be an unobserv-
assistance to organizations implementing completed the retrospective study in Jan- able difference in people who seek out
the interventions (3). uary 2013, and their findings along with community-based wellness and preven-
Unfortunately, community-based well- those of Altarum formed the basis for a tion programs. The prospective study will
ness and prevention programs have yet Report to Congress delivered In November attempt to address selection bias through
to be incorporated into the continuum of 2013 (5). a population survey to measure Medicare
care for Medicare recipients. At the Cen- As described in the Report to Con- beneficiaries’ readiness to engage in well-
ters for Medicare and Medicaid Innova- gress, Acumen found statistically signifi- ness interventions, which will allow for
tion (CMMI), we have embraced a broader cant total medical cost savings for four more precise matching between partici-
view of addressing prevention and wellness established wellness programs: Enhance- pants and controls based on personal moti-
in our beneficiaries. We now have several Fitness (EF), Arthritis Foundation Exer- vation factors. The survey will also gauge
efforts underway to bridge the gap between cise Program (AFEP), Arthritis Foundation knowledge of and interest in wellness pro-
clinical and community-based care. In this Tai Chi Program (AFTCP), and Matter of grams, which will inform efforts to scale
article, we discuss the results of research Balance (MOB). Two additional programs, interventions.
to date, describe current efforts to evaluate the widely disseminated Chronic Disease Along with the 4202(b) evaluation,
and engage with community-based well- Self-Management Program (CDSMP) and CMMI is considering proposing an innov-
ness and prevention programs, and outline the Arthritis Foundation Aquatic Program ative new community-based service deliv-
some challenges that we have recognized (AFAP), demonstrated reductions in inpa- ery model, the Accountable Health Com-
in fully integrating these interventions into tient hospital costs, which indicate a poten- munity (AHC), aimed at achieving bet-
the Medicare system. tial for future long-term savings. One ter care and lower health care costs for
Recognizing the potential of common element of the programs that beneficiaries with highly prevalent chronic
community-based wellness and preven- were associated with total cost savings is diseases within a defined geographic area.
tion programs to improve health and that they all included consistent physical AHCs would utilize funds from CMMI
reduce medical costs among Medicare activity to prevent and manage chronic as well as from other public and private

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 189 | 26
Colligan et al. Medicare and community-based wellness programs

funding sources to provide a full range of wellness intervention will require recon- 6. Other changes to simplify, modernize, and clarify
preventive, non-medical, and community- sideration of how CMS can compensate medicaid benchmark requirements and coverage
requirements, 78 Federal Register 135 (15 July 2013),
based health services. CMMI funds would non-traditional health providers. Med-
pp. 42226–7.
be used to strengthen the local infrastruc- icaid recently allowed for a broader
ture and provide the “glue” to coordi- range of providers to administer pre- Conflict of Interest Statement: The authors declare
nate and align services provided by clin- ventive services to “(1) prevent disease, that the research was conducted in the absence of any
ical and social service partners as well as disability, and other health conditions commercial or financial relationships that could be
private payers within communities. The or their progression; (2) prolong life; construed as a potential conflict of interest.
model is designed to create a founda- and (3) promote physical and mental
This paper is included in the Research Topic, “Evidence-
tion upon which these integrated commu- health and efficiency,” as long as the ser- Based Programming for Older Adults.” This Research
nity services are available and leveraged vices are recommended by a physician Topic received partial funding from multiple government
to achieve the greatest impact on CMS or licensed provider (6). Medicare, how- and private organizations/agencies; however, the views,
beneficiaries. ever, has yet to incorporate non-clinical findings, and conclusions in these articles are those of the
authors and do not necessarily represent the official posi-
These efforts demonstrate a commit- providers into the reimbursement system. tion of these organizations/agencies. All papers published
ment on the part of CMMI to incorporate We hope that our current projects will in the Research Topic received peer review from members
community-based wellness and prevention offer more perspective into these chal- of the Frontiers in Public Health (Public Health Edu-
programs in the continuum of care for lenges and provide insights on how to cation and Promotion section) panel of Review Editors.
Medicare beneficiaries. Nonetheless, there make community-based wellness and pre- Because this Research Topic represents work closely asso-
ciated with a nationwide evidence-based movement in
are challenges that remain in terms of vention programs accessible and available the US, many of the authors and/or Review Editors may
fully integrating community-based inter- to a broader population of Medicare bene- have worked together previously in some fashion. Review
ventions into the Medicare payment sys- ficiaries. Editors were purposively selected based on their expertise
tem. First, the cost/benefit ratio of imple- with evaluation and/or evidence-based programming
for older adults. Review Editors were independent of
mentation costs to Medicare savings needs
REFERENCES named authors on any given article published in this
to be made clear before wellness pro- 1. Disease Prevention and Health Promotion Ser- volume.
grams are widely offered to Medicare ben- vices (OAA Title IIID). Available from: http://www.
eficiaries. Current programs are primarily aoa.acl.gov/AoA_Programs/HPW/Title_IIID/index. Received: 16 June 2014; accepted: 28 September 2014;
funded through grants and do not take aspx published online: 27 April 2015.
2. Evidence-Based Prevention Program. Available from: Citation: Colligan EM, Tomoyasu N and Howell B
into consideration the costs of delivering
http://www.acl.gov/NewsRoom/Publications/docs/ (2015) Community-based wellness and prevention pro-
the interventions on a larger scale and Evidence-Based_Prevention_Program_1.pdf grams: the role of Medicare. Front. Public Health 2:189.
whether or not the payer would receive 3. Center for Healthy Aging. Available from: http:// doi: 10.3389/fpubh.2014.00189
returns on investment. The prospective www.ncoa.org/improve-health/center-for-healthy- This article was submitted to Public Health Education
study will try to address this issue by cal- aging/ and Promotion, a section of the journal Frontiers in
4. Smith B, Kloc M, Korda, H. Environmental Scan Public Health.
culating the cost to administer programs of Community-Based Prevention and Wellness Pro- Copyright © 2015 Colligan, Tomoyasu and Howell. This
in light of savings accrued to Medicare. grams in the United States: Evidence Review Report. is an open-access article distributed under the terms of
Furthermore, community-based wellness Ann Arbor, MI: Altarum Institute (2011). the Creative Commons Attribution License (CC BY). The
and prevention programs rely on a non- 5. Report to Congress: The Centers for Medicare & use, distribution or reproduction in other forums is per-
clinical workforce of lay instructors that Medicaid Services’ Evaluation of Community-based mitted, provided the original author(s) or licensor are
Wellness and Prevention Programs under Section credited and that the original publication in this journal
do not fit into current Medicare pay- 4202 (b) of the Affordable Care Act. (2013). Avail- is cited, in accordance with accepted academic practice.
ment structures. Thus creating a bene- able from: http://innovation.cms.gov/Files/reports/ No use, distribution or reproduction is permitted which
fit for community-based prevention and CommunityWellnessRTC.pdf does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 189 | 27

 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00190

Foundation engagement in healthy aging initiatives and

evidence-based programs for older adults
Mary Ellen Kullman*
Archstone Foundation, Long Beach, CA, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: evidence-based health programs, chronic disease self-management, older adults, public health and aging, Foundations, healthy aging initiatives,
community health partnerships, fall prevention programs

In trying to improve health, particularly and adoption of EBP by community- Reflecting upon the last few decades,
for the millions of older Americans with based organizations and health care sys- there has been a remarkable improve-
chronic conditions, many researchers focus tems. ment in the number, variety, and qual-
solely on improving the professional health Whether it is foundation funding, Older ity of evidence-based programs. A few
care system. Better medications, care pro- Americans Act funds, or resources of the examples of the programs include: the
tocols, and other clinical interventions are organizations offering the programs, funds Chronic Disease Self-Management Pro-
important, but much of chronic care takes are limited. So are participants’ and pro- gram, addressing several chronic condi-
place away from the clinic or hospital and gram providers’ time and energy. The tions; A Matter of Balance, Stepping On,
in one’s home or community (1). Evidence- development of EBP for older adults allows and Tai Chi for Better Balance, address-
based health promotion programs (EBP) limited resources to be directed to pro- ing fall prevention, PEARLS (Program
are an important way by which commu- grams with the greatest probability for pos- to Encourage Active, Rewarding Lives
nity agencies and health professionals can itive impact. As a funder looking at hun- for Seniors), and IMPACT (Improving
work together to offer older adults and dreds of requests each year, confidence in Mood-Promoting Access to Collaborative
their families proven ways to take con- the science is critically important. We rec- Treatment) addressing depression; Fit and
trol of their health and live the lives they ognize that when we support any given pro- Strong! and EnhanceFitness, addressing
want (2). gram, other programs may struggle. EBP exercise. There is now an alternative to the
Archstone Foundation is a private inde- give funders greater assurance that what we free-for-all of “do it yourself ” programs,
pendent grant making foundation with a fund will deliver meaningful results. developed without standards or proof of
mission of preparing society to meet the Archstone Foundation and many other outcomes, as was the case only a few years
needs of an aging population. During the funders have supported EBP out of a desire ago. We now have a system of programs
last two decades, the Foundation has sup- to improve health outcomes and quality of with varying levels of evidence targeting
ported the development, evaluation, and life for older adults. We value the programs’ a number of health and quality of life
dissemination of a range of evidence-based self-management strategies that empower concerns. The Centers for Disease Con-
programs in areas including fall preven- older adults, while effectively improving trol and Prevention, the SAMHSA National
tion, physical activity, chronic disease man- their health. EBP have been supported out Registry of Evidence-Based Programs and
agement, caregiver support, and mental of a desire to provide high quality, effec- Practices, the Administration of Commu-
health. It also funded the development of tive, and sustainable programs with a broad nity Living, and the Agency for Healthcare
Better Choices, Better Health, the online reach, and proven outcomes when done Research and Quality are a few of the agen-
version of the Stanford Chronic Disease with fidelity to the original model (4). cies that have established processes for eval-
Self-Management Course, to broaden its Healthy aging initiatives for older adults uating, and certifying or registering pro-
reach to underserved populations. Arch- require broad and effective community col- grams as evidence-based. We are also seeing
stone Foundation has supported the cul- laborations (5). Researchers, community- a growing number of programs addressing
tural adaptation of EBP to meet the based organizations, older adults, health the needs of our diverse older adults (6).
needs of increasingly diverse older adults. care systems, government, and funders, are This entire Frontiers in Public Health jour-
It has funded local and national coali- all important partners in the development nal issue is devoted to the study of EBP and
tions to support the dissemination of and effective use of EBP. Funders vary in their value.
EBP, including the Falls Free Initiative led their approach to grant making and where It is important to recognize that there
by the National Council on Aging (3). in the process they can engage in partner- has been resistance to the widespread use
Most recently, the Foundation has funded ships. Some will fund program develop- of evidence-based programs. This resis-
the Evidence-Based Leadership Council ment, while others may fund evaluation, tance offers lessons in how to improve
to ease the challenges of dissemination and/or replication. the field (7). Some have feared that EBP

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 190 | 28
Kullman Foundation engagement in evidence-based programs

will stifle the creativity of practitioners and new EBP. There is tremendous opportunity the Healthcare Workforce. Washington, DC: The
that the programs cannot respond to the to build partnerships and to continue to National Academies Press (2008).
unique attributes of a community, espe- grow this exciting movement for improv-
cially diverse communities. This should ing health and quality of life for older Conflict of Interest Statement: The author declares
encourage researchers to look at how pro- adults through evidence-based programs that the research was conducted in the absence of any
grams can be customized and better tar- and healthy aging initiatives. commercial or financial relationships that could be
geted to specific populations and their construed as a potential conflict of interest.
needs. A community-based participatory REFERENCES
1. Wagner EH, Austin BT, Davis C, Hindmarsh M, This paper is included in the Research Topic, “Evidence-
research approach that engages older adults Schaefer J, Bonomi A. Improving chronic illness Based Programming for Older Adults.” This Research
and practitioners along with researchers care: translating evidence into action. Health Aff Topic received partial funding from multiple government
in the development of new or adaptations (2001) 20(6):64–78. doi:10.1377/hlthaff.20.6.64 and private organizations/agencies; however, the views,
of existing EBP may hold promise (8). 2. Bodenheimer T, Lorig K, Holman H, Grumbach findings, and conclusions in these articles are those of the
K. Patient self-management of chronic disease authors and do not necessarily represent the official posi-
Further, perceived costs and administrative
in primary care. JAMA (2002) 288(19):2469–75. tion of these organizations/agencies. All papers published
barriers in offering EBP, key impediments doi:10.1001/jama.288.19.2469 in the Research Topic received peer review from members
to the spread of EBP, may suggest process 3. Beattie BL. The national fall free initiative, working of the Frontiers in Public Health (Public Health Edu-
improvements in these programs’ manage- collaboratively to affect change. J Safety Res (2011) cation and Promotion section) panel of Review Editors.
ment and delivery, such as those being 42(6):521–3. doi:10.1016/j.jsr.2010.11.009 Because this Research Topic represents work closely asso-
4. Frank JC, Coviak CP, Healy TC, Lelza B, ciated with a nationwide evidence-based movement in
explored by the Evidence-Based Leader-
Casado BL. Addressing fidelity in evidence-based the US, many of the authors and/or Review Editors may
ship Council. Examples of improvements health promotion programs for older adults. have worked together previously in some fashion. Review
to support organizations that wish to offer J Appl Gerontol (2008) 27(1):4–33. doi:10.1177/ Editors were purposively selected based on their expertise
multiple EBP could include shared data 1090198114543007 with evaluation and/or evidence-based programming
management systems, common evaluation 5. Beattie BL, Whitelaw N, Mettler M, Turner D. for older adults. Review Editors were independent of
Vision for older adults and health promotion. Am named authors on any given article published in this
tools, coordinated training and technical J Health Promot (2003) 18(2):200–4. doi:10.4278/ volume.
assistance, and common core curricula (9). 0890-1171-18.2.200
Looking to the future, our challenge 6. Bass D. Evidence-based practice: critical thinking Received: 18 June 2014; accepted: 28 September 2014;
is to expand the breadth of offerings, about translating research to practice. Generations published online: 27 April 2015.
(2010) 34(1). Citation: Kullman ME (2015) Foundation engagement
improve the quality, and ease the adop-
7. Bass DM, Judge KS. Challenges implement- in healthy aging initiatives and evidence-based pro-
tion of evidence-based programs. To realize ing evidence-based programs. Generations (2010) grams for older adults. Front. Public Health 2:190. doi:
the potential of EBP, we will need to pre- 34(1):51–8. 10.3389/fpubh.2014.00190
pare our workforce to understand, imple- 8. Wethington E, Breckman R, Meador R, Reid MC, This article was submitted to Public Health Education
ment, manage, and promote the programs Sabir M, Lachs M, et al. The pilot studies pro- and Promotion, a section of the journal Frontiers in
gram: mentoring translational research. Gerontol- Public Health.
(10). The use of EBP for older adults is ogist (2007) 27(6):845–50. doi:10.1093/geront/47. Copyright © 2015 Kullman. This is an open-access arti-
still a relatively new phenomenon, and 6.845 cle distributed under the terms of the Creative Commons
we are far from bringing even the most 9. Schneider EC, Altpeter M, Whitelaw N. An inno- Attribution License (CC BY). The use, distribution or
established programs to scale. Evidence vative approach for building health promotion reproduction in other forums is permitted, provided the
will change over time, and ongoing work program capacity: a geriatric volunteer training original author(s) or licensor are credited and that the
curriculum. Gerontologist (2007) 47(3):398–403. original publication in this journal is cited, in accordance
will be necessary. The growing diversity doi:10.1093/geront/47.3.398 with accepted academic practice. No use, distribution or
in the older adult population will compel 10. Institute of Medicine of the National Acade- reproduction is permitted which does not comply with
us to develop, evaluate, and disseminate mies. Retooling for an Aging America: Building these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 190 | 29

 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00253

Chronic Disease Self-Management Program: insights from

the eye of the storm
Kate Lorig*
Patient Education Research Center, School of Medicine, Stanford University, Stanford, CA, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: self-management, patient education, CDSMP, chronic disease, translational research

In the early 1990s, the Stanford Patient desired behavior predicts their level of suc- program. Insight: publish as soon as prac-
Education Research Center developed the cess. SE can be enhanced through skills tical using language understandable out-
Chronic Disease Self-Management Pro- mastery, modeling, reinterpretation, and side the scientific community. Publications
gram (CDSMP) to test the hypothesis that social persuasion (1). All of these are used should be aimed at scientific, practice, and
people with comorbid conditions could throughout the program. For example, par- policy communities.
benefit when placed in a common inter- ticipants made action plans (skills mas- Based on this interest, we started offer-
vention. At that time, the existing para- tery) and shared with other participants ing one or two yearly trainings in 1999.
digm consisted of having patients attend their confidence in achieving their plan Our aim was to give each organization the
only disease-specific education programs. each week. If a participant’s confidence was capacity to train its own leaders and to grow
In 2013 alone, 50–100,000 people in 36 low, then the leaders and other participants its own program. As developers, we saw our
countries attended the CDSMP. How did helped them problem-solve (2). Insight: role as offering training and technical assis-
this happen? We do not know the answer, theories are useful – but only if theories are tance. What began with 20–40 trainers per
but have some ideas. The following is a brief translated into programmatic elements. year has grown to 400 or more new trainers
history and some key insights. The original randomized trial had four per year. Each pair of new trainers has the
outcome categories that were of interest to capacity to offer programs and train local
DEVELOPMENT different communities (3). Behaviors such leaders. Insight: building organizational
In 1990, to determine patient-perceived as exercise were of interest to the behavioral capacity is an important translation ele-
problems, we held 11 focus groups with science community as was SE. Symptom- ment. To do this, one must devote resources
people with chronic conditions. Partici- based outcomes (pain, depression, fatigue) to training and supporting others.
pants talked predominately about symp- were of interest to patients and health-
toms, and thus the program was built care providers, while changes in utilization, LEGAL STUFF AND AGILITY
around breaking the symptom cycle and such as days in hospital and emergency By the early twenty-first century, requests
tools that participants could use to accom- department visits were of interest to health for training were rapid. There was a need to
plish this. By basing a program on end-user service researchers, government, and oth- put more structure around the translation
problems, we assured their interest. Insight: ers who pay for health care. Insight: choose process. There has never been a business
one cannot underestimate the importance outcomes that are of interest to communi- plan. Rather, the business of translation
of having happy and excited end-users. ties and policy makers you hope will use was and continues to respond to changing
This can only be accomplished by meeting and adopt your program. needs. Early on, Stanford administration
user needs. showed little interest in our activities. As
We developed the CDSMP for transla- EARLY REPLICATION we involved more organizations, the Uni-
tion into practice. It is taught by peers. At the end of the original randomized trial, versity became concerned about liability.
Every minute was and continues to be there were improvements in all four cat- To mitigate this issue, we worked with the
scripted for both content and process. egories. Hospitalization was reduced by Stanford Office of Technology Licensing to
Insight: the design process accounted for 8 days. Based on these data, Kaiser Perma- establish policies. Insights: program devel-
many of the things that enabled the nente, one of the original study partners, opers need to worry about liability and
CDSMP to be a success. Translation can- decided to trial the program nationally in licensure issues.
not occur without a set protocol that can 1998. This longitudinal study had similar There were five potentially competing
be followed by others. outcomes to the original trial (4). Insight: interests, the legal interests of the Uni-
The CDSMP was based on self-efficacy having a respected partner who is also an versity, the need to keep the workshops
(SE) theory. While many interventions are early adopter gives translation a head start. affordable for adopting agencies, the need
informed by theory, the CDSMP system- At about this same time, having read to sustain a training technical-assistance
atically incorporated SE theory. SE theory our original article, others from around the (TA) infrastructure, the financial interests
states that one’s confidence in achieving a country began to call inquiring about the of the program developers, and the need

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 253 | 30
Lorig Chronic Disease Self-Management Program insights

to minimize bureaucracy. At this stage approximately 50,000 people (6). Between CURRENT CHALLENGES
in translation, many program developers 2005 and 2010, organizations not funded In 2014, the CDSMP continues to gather
form their own companies or collaborate by AoA also began to offer the CDSMP. momentum. It has multiple funders among
with an existing company. However, the These included major health plans, a third- U.S. federal agencies as well as U.S. foun-
developers were not interested in becom- party insurer and local agencies. In 2010, as dations and health-care systems. As the
ing entrepreneurs. We decided to continue part of the American Recovery and Reinvest- program has grown, so have the challenges
working within the University. License ment Act, ARRA recovery funding, AoA in for its creators. (Please note that there are
price ($500) was set to allow an agency to collaboration with CDC, provided grants many other challenges for those offering
offer 30 workshops over 3 years for approx- to 45 states, Puerto Rico and the Dis- the programs.)
imately 300 participants. Insight: in trans- trict of Columbia for disseminating the
lating products to widespread use, there CDSMP. The goal of 50,000 completers ENCOURAGING AND DISCOURAGING
are many competing interests. It is best (those who had attended four or more ADAPTATIONS
to acknowledge these and work at a fair sessions out of six) was reached and sur- There is constant pressure to adapt and
compromise early. passed. Insight: even in bad times good modify the CDSMP. These requests usu-
Between 2000 and 2010, both the things can happen. Insight: when opportu- ally come from people who have not seen
licensing and training policies adapted to nity knocks it is important to have “shovel or participated in the program and usu-
changing times and became more cod- ready” projects. ally know “what is best for my popula-
ified. With the help of the Office for As part of the AARA funding, the tion.”These requests range from wanting to
Technology Licensing, we created and con- CDSMP was evaluated in a large study change content to changing length or for-
tinue to create different types of licenses. involving 22 organizationally and geo- mat. Insight: there is distrust of anything
See http://patienteducation.stanford.edu/ graphically diverse sites. The outcomes “Not Invented Here.”
licensing for current license policy and demonstrated that the program continued We usually tell the requesters to try
http://patienteducation.stanford.edu/train to meet the triple aims of health care, bet- the program, and then ask the partici-
ing/trnpolicies.html for training policy. ter care, better health, and lower costs (7). pants what they want to change. Requests
Insight: while personal preference and Following ARRA, more secure funding was for changes in format and length or large
knowledge can run early translation efforts, achieved in the AoA (now ACL) budget. amounts of content cannot be met without
true widespread translation requires “rules Authorizing legislation in the Older Amer- rewriting the CDSMP and re-evaluating
and regulations.” icans Act has long included Title IIID for the new format with a new population.
Disease Prevention and Health Promotion. This has been done successfully a few
POLICY Beginning in 2012, ACL required states to times and has resulted in the pain self-
In 2003, the U.S. Administration on Aging, use these funds ($21,000,000) for evidence- management program and the hepatitis-c
AoA (now a unit of the Administration based programs. Also in 2012, CDSMP self-management program, among others
for Community Living, ACL) in collab- became a small line item in the AoA (now (8, 9). Recently, we have encouraged groups
oration with CDC and other public and ACL) budget financed through the 2012 wanting to make changes to ask permis-
philanthropic organizations, funded 14 Affordable Care Act Prevention and Public sion for small, rapid-change cycle exper-
sites to embed evidence-based programs Health Fund. While the funding was much iments and to report the finding. How-
into community-based organizational net- reduced from that received from AARA, ever, few have conducted such experiments.
works. It was only after several of the appli- 22 states received grants. Some states that Insight: when given a process rather than
cants wanted to use the CDSMP that the had been funded under AARA were not permission for making change, there is little
head of the National Council on Aging refunded through these grants. However uptake.
TA Center for these grants called Stanford. they continue to offer the CDSMP utilizing
Until this time, no one at Stanford knew IIID and other monies coming from many FIDELITY
anything about this initiative. Because of sources. These include foundations, health Evidence-based programs always have
this collaboration, more than 3000 people care, and other local, state, and federal the challenge of standardization. Without
had participated in evidence-based pro- agencies. UniteHere, a union of mostly low standardization, the evidence base is lost.
grams including the CDSMP (5). Insight: paid service workers, recently completed its As the core of trainers has grown larger
sometimes adoption on a national level second year of offering the CDSMP, mostly (over 1000 master trainers and many thou-
comes from the grass roots up. in Spanish. They have reached several hun- sand leaders), maintaining quality pro-
In 2006, the U.S. Department of Health dred workers in the Los Angeles area and grams is more difficult. The use of webi-
and Human Services announced collab- are currently expanding the program to nars, administrative and fidelity manuals,
oration between AoA, NCOA, and the their members in many other cities. Insight: and email discussion groups helps with the
Atlantic Philanthropies to build CDSMP if a program meets local needs and is centralization of key training and techni-
capacity across the United States. AoA liked by both agencies and participants, cal assistance (10, 11). Insight: fidelity is a
awarded funding to 27 states. This fund- there is life even when funding is reduced. delicate balance that constantly has to be
ing mandated adoption of the CDSMP Insight: if grant funds can build capacity re-evaluated and maintained.
and encouraged the use of other evidence- and engagement, sometimes programs can It is unusual for a program creator
based programs. These programs served be sustained through other sources. to remain involved with widespread

www.frontiersin.org April 2015 | Volume 2 | Article 253 | 31

Lorig Chronic Disease Self-Management Program insights

translation. There have been several chal- Care (1999) 37(1):5–14. doi:10.1097/00005650- Stanford Patient Education Research Center
lenges. First of these is moving among 199901000-00003 (2012). 25 p. Available from: http://patient
2. Bandura A. Self-Efficacy: The Exercise of Control. education.stanford.edu/licensing/FidelityManual
academic, training, technical assistance,
New York: W. H. Freeman (1997). 604 p. 2012.pdf
promoting, and cheerleading roles. Insight: 3. Lorig K, Laurent DD, Plant K, Krishnan E, Rit-
if you do not like juggling, do not join the ter PL. The components of action planning and
circus. their associations with behavior and health out- Conflict of Interest Statement: The author receives
The second is how to finance core trans- comes. Chronic Illn (2014) 10(1):50–9. doi:10. royalties from Stanford University and Bull Publishing.
1177/1742395313495572
lations activities such as training, technical
4. Lorig KR, Sobel DS, Ritter PL, Laurent D, Hobbs This paper is included in the Research Topic, “Evidence-
assistance, and updating materials. Monies M. Effect of a self-management program on Based Programming for Older Adults.” This Research
from federal agencies and foundations, for patients with chronic disease. Eff Clin Pract (2001) Topic received partial funding from multiple government
the most part, go for program delivery 4:256–62. and private organizations/agencies; however, the views,
and are seldom earmarked for these core 5. Tilly J. The administration on aging’s experiences findings, and conclusions in these articles are those of the
with health, prevention, and wellness. Generations authors and do not necessarily represent the official posi-
activities. This means that the core func- (2010) 34(1):20–5. tion of these organizations/agencies. All papers published
tions must become self-sustaining through 6. Ory MG, Smith ML, Patton K, Lorig K, Zenker W, in the Research Topic received peer review from members
charging for such activities as training, Whitelaw N. Self-management at the tipping point: of the Frontiers in Public Health (Public Health Edu-
materials, and TA. Insight: the financing of reaching 100,000 Americans with evidence-based cation and Promotion section) panel of Review Editors.
core translation activities can help or hin- programs. J Am Geriatr Soc (2013) 61(5):821–3. Because this Research Topic represents work closely asso-
doi:10.1111/jgs.12239 ciated with a nationwide evidence-based movement in
der translation and must be planned and 7. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, the US, many of the authors and/or Review Editors may
flexible. Whitelaw N, et al. Successes of a National Study of have worked together previously in some fashion. Review
the Chronic Disease Self-Management Program: Editors were purposively selected based on their expertise
SUMMARY meeting the triple aim of health care reform. with evaluation and/or evidence-based programming
This is a personal 22-year retrospective Med Care (2013) 51(11):992–8. doi:10.1097/MLR. for older adults. Review Editors were independent of
0b013e3182a95dd1 named authors on any given article published in this
look at insights gained as the CDSMP has 8. LeFort S, Gray-Donald K, Rowat KM, Jeans ME. volume.
moved from concept to translation. This Randomized controlled trial of a community-
retrospective look has been both surprising based psychoeducation program for the self- Received: 16 June 2014; accepted: 09 November 2014;
and humbling. I look forward to learning management of chronic pain. Pain (1998) published online: 27 April 2015.
74:297–306. doi:10.1016/S0304-3959(97)00190-5 Citation: Lorig K (2015) Chronic Disease Self-
what comes next.
9. Groessl EJ, Weingart KR, Stepnowsky CJ, Gif- Management Program: insights from the eye of the storm.
ford AL, Asch SM, Ho SB. The hepatitis C Front. Public Health 2:253. doi: 10.3389/fpubh.2014.
ACKNOWLEDGMENTS self-management programme: a randomized con- 00253
I wish to acknowledge and thank The Stan- trolled trial. J Viral Hepat (2010) 18:358–68. doi: This article was submitted to Public Health Education
ford Patient Education Research Center 10.1111/j.1365-2893.2010.01328.x and Promotion, a section of the journal Frontiers in
10. Stanford Patient Education Research Cen- Public Health.
staff, our participants, trainers, and leaders
ter. Implementation Manual, Stanford Self- Copyright © 2015 Lorig . This is an open-access article
from around the world. Management Programs, 2008 [Internet]. Stanford, distributed under the terms of the Creative Commons
CA: Stanford Patient Education Research Cen- Attribution License (CC BY). The use, distribution or
REFERENCES ter (2008). 28 p. Available from: http://patient reproduction in other forums is permitted, provided the
1. Lorig KR, Sobel DS, Stewart AL, Brown BW, education.stanford.edu/licensing/Implementation original author(s) or licensor are credited and that the
Bandura A, Ritter P, et al. Evidence suggest- _Manual2008.pdf original publication in this journal is cited, in accordance
ing that a chronic disease self-management pro- 11. Stanford Patient Education Research Center. Pro- with accepted academic practice. No use, distribution or
gram can improve health status while reduc- gram Fidelity Manual, Stanford Self-Management reproduction is permitted which does not comply with
ing hospitalization: a randomized trial. Med Programs, 2012 Update [Internet]. Stanford, CA: these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 253 | 32
 GENERAL COMMENTARY
published: 01 June 2015
doi: 10.3389/fpubh.2015.00153

Corrigendum: “Chronic Disease

Self-Management Program: insights
from the eye of the storm”
Kate Lorig *

Patient Education Research Center, School of Medicine, Stanford University, Stanford, CA, USA

Keywords: corrigendum, self-management, patient education, CDSMP, chronic disease, translational research

A Corrigendum on

Chronic Disease Self-Management Program: insights from the eye of the storm
by Lorig K. Front Public Health (2014) 2:253. doi:10.3389/fpubh.2014.00253

There is an error in the original manuscript under the Section “Early Replication”. The number of
days of reduced hospitalization should be 0.8, not 8 as originally published.

Conflict of Interest Statement: The author receives royalties from Stanford University and Bull Publishing.

This paper is included in the Research Topic, “Evidence-Based Programming for Older Adults.” This Research Topic received partial
funding from multiple government and private organizations/agencies; however, the views, findings, and conclusions in these
articles are those of the authors and do not necessarily represent the official position of these organizations/agencies. All papers
published in the Research Topic received peer review from members of the Frontiers in Public Health (Public Health Education
and Promotion section) panel of Review Editors. Because this Research Topic represents work closely associated with a nationwide
evidence-based movement in the US, many of the authors and/or Review Editors may have worked together previously in some
fashion. Review Editors were purposively selected based on their expertise with evaluation and/or evidence-based programming
for older adults. Review Editors were independent of named authors on any given article published in this volume.

Edited and reviewed by: Copyright © 2015 Lorig. This is an open-access article distributed under the terms of the Creative Commons Attribution License
Matthew Lee Smith, (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited
and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or
The University of Georgia, USA
reproduction is permitted which does not comply with these terms.
*Correspondence:
Kate Lorig
[email protected]

Specialty section:
This article was submitted to Public
Health Education and Promotion,
a section of the journal
Frontiers in Public Health
Received: 19 May 2015
Accepted: 19 May 2015
Published: 01 June 2015
Citation:
Lorig K (2015) Corrigendum: “Chronic
Disease Self-Management Program:
insights from the eye of the storm”.
Front. Public Health 3:153.
doi: 10.3389/fpubh.2015.00153

Frontiers in Public Health | www.frontiersin.org 33 June 2015 | Volume 3 | Article 153

 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00274

A matter of balance: older adults taking control of falls

by building confidence
Margaret Haynes, Patricia League* and Gloria Neault
MaineHealth, Elder Care Services, Portland, ME, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: fall prevention program, cognitive-behavioral therapy, fear of falling, cost savings, lay leader

This commentary will present the chal- balance/volunteer lay leader (MOB/VLL) leader model has made MOB/VLL available
lenges and successes of implementing and model was developed with a translational to over 45,000 older adults across the U.S.
sustaining an evidence-based falls preven- research grant from the U.S. Administra- Participants report significant increases
tion program using a lay leader model. tion on Aging to increase adoption of in falls efficacy, falls management, and
The evolution from professional educa- the program and thereby reach signifi- falls control at 6 weeks, 6, and 12 months,
tor to lay leader will be described, as well cant numbers of older adults in the com- achieving comparable outcomes with those
as the benefits of this model for individ- munity. The core elements of A MOB of participants in the RCT (5). The suc-
ual participants, reducing falls and finan- include (a) cognitive restructuring and cess with MOB/VLL suggests that other
cial savings for CMS. Falls are the leading behavioral activation activities that pro- evidence-based programs currently requir-
cause of death from injury and the most mote the belief that falls and fear of falling ing professional staff can be adapted for
common cause of non-fatal injuries, result- are controllable; (b) enhancing falls self- facilitation by volunteers. Further, this suc-
ing in emergency department visits in the efficacy and falls management by helping cessful translation of a professionally led
older adult population with an estimated participants set realistic goals for increasing health promotion program into a volun-
cost of over $30 billion for direct medical activity; (c) promoting changes in mod- teer lay leader model promotes embedding
costs alone. Older adults who fall and are ifiable risk factors such as securing loose the program in community-based orga-
not injured may develop a fear of falling rugs in their home environment; and (d) nizations, thus, making it more broadly
and limit activities with subsequent fur- teaching exercises known to reduce risk of available to older adults in diverse settings.
ther loss in physical function, resulting in falling by increasing strength and balance Volunteer lay leaders who facilitate the
an increased risk of future fall (1). (4). MOB/VLL maintains these cognitive program report a sense of confidence about
A matter of balance (MOB) was devel- restructuring activities. Experts in exercise teaching, find it a rewarding experience and
oped and tested in the 1990s at Boston were consulted concerning adaptations to are enthusiastic about seeing older adults
University’s Roybal Center for enhance- ensure that exercises taught in the transla- gain more independence. In a follow-
ment of late-life function as a compre- tion promoted increased strength and bal- up survey, lay leaders indicated that they
hensive approach to maximizing activity ance needed to reduce risk of falling and gained a sense of accomplishment (80%),
engagement and function and reducing fall were safe for persons with osteoporosis found their purpose in life had increased
risks with funding from the National Insti- and/or joint replacements. (48%), felt they could make a positive
tute on Aging (2, 3). Professionally led, Utilizing a train-the-trainer model, the difference in another person’s life (76%),
utilizing physical therapists, occupational partnership for healthy aging (PFHA) and increased their own confidence about
therapists, registered nurses, and social adapted the program, remaining true to managing falls (84%) (5).
workers, this evidence-based, small group the original MOB model. Since 2006, over The Centers for Medicare & Medic-
health promotion program for older adults 900 Master Trainers have been educated aid Services’ Evaluation of Community-
used cognitive-behavioral techniques to in 38 states by the PFHA in all aspects of based Wellness and Prevention Programs
reduce the fear of falling (2, 3). Par- the program utilizing a Master Trainer cur- under Section 4202 (b) of the Afford-
ticipant outcomes from the randomized riculum. Master Trainers then teach VLL able Care Act documents the economic
clinical trial (RCT) included significant utilizing a VLL curriculum and manual. A value of MOB/VLL (6). Participation was
improvements in falls management, falls Guest Therapist handbook was developed associated with total medical cost savings,
self-efficacy, falls control, increased activity to include a professional visit to one class to reflecting cost savings in the unplanned in-
levels, and reductions in social isolation (2). address participant concerns, demonstrate patient, skilled nursing facility, and home
From a community perspective, uti- how to get up from a fall and other clinical health settings. For example, there was a
lization of health care professionals as issues. Each participant receives a partici- $938 decrease in total medical costs per year
class leads made the intervention expen- pant workbook for their use at home. In driven by a $517 reduction in unplanned
sive and difficult to sustain. A matter of the past 7 years, the translation to a lay hospitalization costs, a $234 reduction in

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 274 | 34
Haynes et al. Matter of balance: older adults

skilled nursing facility costs, and an $81 essential that programs are readily available This paper is included in the Research Topic, “Evidence-
reduction in home health costs (6). and accessible. Based Programming for Older Adults.” This Research
Topic received partial funding from multiple government
Matter of balance/volunteer lay leader
and private organizations/agencies; however, the views,
offers participating organizations the REFERENCES
findings, and conclusions in these articles are those of the
opportunity to bring an evidence-based 1. Centers for Disease Control and Prevention. Falls
authors and do not necessarily represent the official posi-
Among Older Adults: An Overview (2014). Avail-
fall prevention program to the community. able from: http://www.cdc.gov/homeandrecreation
tion of these organizations/agencies. All papers published
A host of delivery organizations are used, in the Research Topic received peer review from members
alsafety/falls/adultfalls.html
of the Frontiers in Public Health (Public Health Edu-
including aging service providers, health 2. Tennstedt S, Howland J, Lachman M, Petereson EW,
cation and Promotion section) panel of Review Editors.
departments, trauma departments, reha- Kasten L, Jette A. A randomized, controlled trial
Because this Research Topic represents work closely asso-
of a group intervention to reduce fear of falling
bilitation teams, universities, and hous- ciated with a nationwide evidence-based movement in
and associated activity restriction in older adults. J
ing. Benefits of offering an evidence- Gerontol B Psychol Sci Soc Sci (1998) 53B(6):384–92.
the US, many of the authors and/or Review Editors may
based program include new collaborations have worked together previously in some fashion. Review
doi:10.1093/geronb/53B.6.P384
Editors were purposively selected based on their expertise
and strengthening current partnerships. 3. Howland J, Lachman ME, Peterson EW, Cote J, Kas-
with evaluation and/or evidence-based programming for
It also serves as a link to support older ten L, Jette A. Covariates of fear of falling and
older adults. Review Editors were independent of named
associated activity curtailment. Gerontologist (1998)
adults living independently in the com- authors on any given article published in this volume.
38:549–55. doi:10.1093/geront/38.5.549
munity. Creating dynamic partnerships 4. Tennstedt S, Peterson E, Howland J, Lachman MA.
makes this program available to numerous Matter of Balance: Managing Concerns About Falls. Received: 16 June 2014; accepted: 25 November 2014;
older adults, resulting in decreased falls, Boston: Roybal Center Consortium, Trustees of published online: 27 April 2015.
Boston University (1998). 140 p. Citation: Haynes M, League P and Neault G (2015)
increased cost savings, and provision of
5. Healy TC, Peng C, Haynes P, McMahon E, Botler J, A matter of balance: older adults taking control of falls
continued involvement in life. Gross L. The feasibility and effectiveness of trans- by building confidence. Front. Public Health 2:274. doi:
It is imperative that a MOB contin- lating a matter of balance into a volunteer lay 10.3389/fpubh.2014.00274
ues to reach older adults in the commu- leader model. J Appl Gerontol (2008) 27(1):34–51. This article was submitted to Public Health Education
nity. Strength, balance, and decreased fear doi:10.1177/0733464807308620 and Promotion, a section of the journal Frontiers in
6. Centers for Medicare & Medicaid Services. Report to Public Health.
of falling improve older adults’ quality of Copyright © 2015 Haynes, League and Neault . This is
Congress: The Centers for Medicare & Medicaid Ser-
life and independence. To accomplish this, vices’ Evaluation of Community-Based Wellness and an open-access article distributed under the terms of the
we must engage health care providers to Prevention Programs Under Section 4202 (b) of the Creative Commons Attribution License (CC BY). The
increase referrals, enabling older adults to Affordable Care Act. Baltimore: CMS (2013). use, distribution or reproduction in other forums is per-
continue to live independently in their mitted, provided the original author(s) or licensor are
Conflict of Interest Statement: The authors declare credited and that the original publication in this journal
homes, senior housing, senior living, or that the research was conducted in the absence of any is cited, in accordance with accepted academic practice.
assisted living. A MOB is one step for commercial or financial relationships that could be No use, distribution or reproduction is permitted which
an older adult to stay engaged, but it is construed as a potential conflict of interest. does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 274 | 35

 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00148

“Stepping On”: stepping over the chasm from research to

practice
Jane E. Mahoney 1,2 *
1
University of Wisconsin School of Medicine and Public Health, Madison, WI, USA
2
Wisconsin Institute for Healthy Aging, Madison, WI, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: falls prevention, older adults, community-based programs, dissemination science, implementation science

TRANSFER OF STEPPING ON: FROM US national dissemination to answer that is involved in implementing Stepping On
AUSTRALIA TO US question. Following the replicating effec- before committing to its success. With the
Stepping On is a small-group, self-efficacy tive programs framework (2), we began CDC, we developed an implementation
based, 7-week community workshop by identifying the program’s key elements. guide (4) to help agencies understand what
designed to reduce falls. It addresses four These elements are not obvious because the program entails. Ultimately, root cause
major areas: strength and balance exer- Stepping On is a complex behavior change analysis changed how we select, train, and
cises, medication review, home modifica- intervention, with many activities and coach new leaders, identify and recruit
tion, and vision. Sessions are facilitated by objectives for each session. An interna- participants, and prepare organizations to
a trained leader and a peer co-leader. Physi- tional panel of experts in falls, adult learn- implement the program.
cal therapists teach participants to perform ing, and exercise identified key elements
and advance balance and strength exercises through the modified Delphi technique DEVELOPMENT OF TRAINING
during three sessions and a pharmacist, low (3). Once key elements were identified, we INFRASTRUCTURE
vision expert, and community safety expert prepared a US version of the Stepping On Once we had refined the program for
attend one session each. A randomized Leader’s Manual, trained one new leader, national dissemination, we needed a struc-
controlled trial, published in 2004, showed and implemented the program once, mon- ture to house it. We created the Wisconsin
Stepping On participants had a 31% reduc- itoring each weekly session to evaluate Institute for Healthy Aging (WIHA) to fos-
tion in falls compared to controls (1). fidelity. We observed substantial fidelity ter successful dissemination of evidence-
We brought Stepping On to Wisconsin lapses. For example, the leader taught (i.e., based health promotion programs in Wis-
from Australia in 2006, initially training lectured) rather than facilitated, did not consin, and national dissemination of Step-
nine leaders from eight counties. Training leave time for Q&A, and rarely encouraged ping On. WIHA now trains Stepping On
was informal; leaders read the Australian participants to share experiences. Partici- leaders and master trainers, licenses orga-
Stepping On manual and conversed with pants did not progress their exercises. nizations to deliver the program, and pro-
program developer, Dr. Clemson. In the Using root cause analysis, we identified vides technical assistance and updates.
original study, occupational therapists led underlying causes of the fidelity lapses and Master trainers observe one workshop ses-
the workshop. However, we did not require mapped solutions. First, we identified three sion for each new leader they have trained,
leaders to be health care professionals. Our prerequisites for being trained as a Step- providing coaching after the session to
initial results were mixed. While leaders, ping On leader: (1) background as a health ensure fidelity. Once a leader has success-
host organizations, and participants loved professional, allied health professional, or fully delivered two workshops and passed a
the program and it spread quickly, evalua- fitness expert; (2) experience facilitating fidelity check, he/she may become trained
tion of 151 participants showed no reduc- an adult self-management program; and as a master trainer.
tion in falls from 6 months before the work- (3) professional experience working with
shop to 6 months after. Was the program older adults. Second, we better defined the SUCCESSES IN REACH AND
not suitable for community settings in the target population for the program. Indi- EFFECTIVENESS
US, or did it need more development to viduals who use a standard walker for Stepping On has been implemented in
improve fidelity of implementation? indoor ambulation may be too frail to ben- Wisconsin and 19 other states with over
efit from Stepping On, and may require 7,000 older adults participating to date.
REPACKAGING STEPPING ON: a more individualized approach. In addi- Community-based organizations value the
DEFINING KEY ELEMENTS AND tion, older adults with impaired cogni- program, and WIHA’s training and coach-
ADDRESSING FIDELITY tion may not be able to participate fully. ing results in successful adoption and high-
The CDC provided 4 years of funding to Third, we learned that sponsoring orga- fidelity implementation. Older adults enjoy
develop and test a Stepping On package for nizations need to clearly understand what Stepping On and recruitment is relatively

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 148 | 36
Mahoney Stepping on falls prevention program

easy. Invited experts, once having partici- implement evidence-based health promo- 4. Wisconsin Institute for Healthy Aging. Stepping
pated, want to continue. Since we reconfig- tion programs. There is no reimbursement On an Implementation Guide: How to Prepare
for, Implement, and Evaluate Stepping On in
ured the implementation package based on (yet) through Medicare or Medicaid, and
Community Settings. WIHA (2012). Available
root cause analysis, the program has been little investment from insurance or health from: https://wihealthyaging.org/_data/files/SO_
highly effective. Evaluation of 2,018 partic- maintenance organizations. While increas- materials/Stepping-On-Manual_10-17-2013.pdf .
ipants from 2008 to 2011 showed a signifi- ing participant fees would help fund pro- 5. Gustafson DH, Hundt AS. Findings of innovation
cant 50% reduction in falls from 6 months gram implementation, it would hinder par- research applied to quality management principles
for health care. Health Care Manage Rev (1995)
before to 6 months after the program. ticipation by low-income older adults. We
20:16–33. doi:10.1097/00004010-199521000-00004
need policy changes that enable all at- 6. Wasi P. The Triangle that Moves the Mountain.
risk older Americans to benefit from this Bangkok: Health System Research Institute (2004).
CONTINUING CHALLENGES AND
effective program.
SOLUTIONS Conflict of Interest Statement: Dr. Jane E. Mahoney
A number of challenges hamper imple- is a Co-Author of Stepping On: building confidence
mentation and sustainability. For exam-
CONCLUSION and reducing falls in older adults. Leader Manual, 3rd
We have successfully translated Stepping North American Edition, Freiberg Press, Cedar Falls,
ple, some organizations struggle to identify
On from research to practice. This transla- IA, USA.
leaders and guest experts for the work-
tion has been possible only through united
shop. To overcome barriers to adoption, This paper is included in the Research Topic, “Evidence-
efforts of researchers, policy-makers, and
WIHA piloted a coaching intervention Based Programming for Older Adults.” This Research
community agencies. Such a combina-
to help organizations implement Stepping Topic received partial funding from multiple government
tion of stakeholders, dubbed the “trian- and private organizations/agencies; however, the views,
On. The intervention, based on a process
gle that moves the mountain” (6), cre- findings, and conclusions in these articles are those of the
improvement methodology called NIATx authors and do not necessarily represent the official posi-
ates success not only for the present but
(5), was effective. In a randomized trial, tion of these organizations/agencies. All papers published
also for the future. Expanding Stepping
counties receiving coaching had twice the in the Research Topic received peer review from members
On through continued partnerships across of the Frontiers in Public Health (Public Health Edu-
increase in number of workshops in 1 year,
public health, aging, health care, and injury cation and Promotion section) panel of Review Editors.
compared to counties on the wait list
prevention sectors is the necessary next Because this Research Topic represents work closely asso-
(p = 0.056). Currently, to help organiza- ciated with a nationwide evidence-based movement in
step to achieve the goal of population level
tions with start-up, WIHA provides coach- the US, many of the authors and/or Review Editors may
reduction in falls and related injuries.
ing, a pre-leader training webinar, and a have worked together previously in some fashion. Review
wide array of materials through its web- Editors were purposively selected based on their expertise
site (www.wihealthyaging.org). Addition- ACKNOWLEDGMENTS with evaluation and/or evidence-based programming

ally, WIHA hosts a leader listserv, quarterly The author gratefully acknowledges Betsy for older adults. Review Editors were independent of
Abramson, Valeree Lecey, and Lindy Clem- named authors on any given article published in this
newsletter, and an annual Healthy Aging volume.
Summit where leaders learn and exchange son, for their helpful suggestions.
ideas with researchers, community part- Received: 16 June 2014; accepted: 03 September 2014;
ners, and health care providers. REFERENCES published online: 27 April 2015.
1. Clemson L, Cumming RG, Kendig H, Swann Citation: Mahoney JE (2015) “Stepping On”: stepping
Program reach is also a challenge. M, Heard R, Taylor K. The effectiveness of a over the chasm from research to practice. Front. Public
Implementation is limited among African- community-based program for reducing the inci- Health 2:148. doi: 10.3389/fpubh.2014.00148
American, Hispanic, tribal, and other dence of falls in the elderly: a randomized trial. J Am This article was submitted to Public Health Education
minority cultures. In response to this Geriatr Soc (2004) 52:1487–94. doi:10.1111/j.1532- and Promotion, a section of the journal Frontiers in
5415.2004.52411.x Public Health.
need, we are working on an adaptation, 2. Kilbourne AM, Neumann MS, Pincus HA, Bauer Copyright © 2015 Mahoney. This is an open-access arti-
“Pisando Fuerte,” for Spanish-speaking MS, Stall R. Implementing evidence-based inter- cle distributed under the terms of the Creative Commons
seniors. Such adaptations are urgently ventions in health care: application of the replicat- Attribution License (CC BY). The use, distribution or
needed to extend benefits of this evidence- ing effective programs framework. Implement Sci reproduction in other forums is permitted, provided the
based program. Increased funding will help (2007) 2:42. doi:10.1186/1748-5908-2-42 original author(s) or licensor are credited and that the
3. Custer RL, Scarcella JA, Stewart BR. The modi- original publication in this journal is cited, in accordance
expand Stepping On’s reach. Title III- fied Delphi technique – a rotational modification. with accepted academic practice. No use, distribution or
D of the Older Americans Act provides J Vocat Tech Educ (1998) 15. Available from: http: reproduction is permitted which does not comply with
minimal funds for the aging network to //scholar.lib.vt.edu/ejournals/JVTE/ these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 148 | 37

 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00163

Village or tribe? Expectations, roles, and responsibilities

for effective fall prevention efforts
Tiffany E. Shubert *
Division of Geriatric Medicine, Center for Aging and Health, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: fall prevention, aging, evidence-based practice, physical therapy, systems-change, policy

Village: A group of houses and associ- the older can be embedded into the contin- if the patient is asked to be screened then
ated buildings, larger than a hamlet and uum of the community. For example, the she is at risk of falling, and is not going to be
smaller than a town. older adult could move from programs that safe in the community program. This is not
target the more frail and deconditioned, an atypical behavior; studies have shown
Tribe: A social division in a traditional
like Stepping On (7), to those that target that less than 30% of healthcare providers
society consisting of families or commu-
more robust individuals, like Tai Chi (8). who interact with older adults screen for
nities linked by social, economic, reli-
This proposed model supposes that falls on a routine basis (10).
gious, or blood ties, with a common
infrastructure is in place to build a con- Physical therapists are also uncertain
culture and dialect. Oxford Dictionary.
tinuum of care where none exists. To about their roles and responsibilities in
Effective fall prevention efforts bridge the achieve this model, stakeholders have called the fall prevention continuum. For exam-
silos between clinical and community prac- for multi-level, multi-component inter- ple, few physical therapists are aware of
tice. A fall experienced by an older adult ventions, with the goal of engaging pol- evidence-based programs that target pop-
is rarely a straightforward event. Typically, icy makers, healthcare providers, commu- ulations at risk of falling (11). They also
falls are due to complex inter-related med- nity providers, and older adults them- may not understand the role of State Fall
ical, behavioral, and environmental risk selves. Many have compared these efforts Prevention Coalitions, or perceive them
factors (1). For many older adults, medical to building a “village” of providers (9). as partners in creating a continuum. In
risk factors such as medication reconcil- The concept of “village” is appealing, a survey of PTs interested in disseminat-
iation, treatment of atrial fibrillation, or though may be inherently flawed. A village ing the Otago Exercise Program (OEP), the
physical therapy to address gait and balance is a group of buildings that simply share majority of PTs indicated that support of
impairments are primary in fall prevention the same physical location. These build- a program by State-Based Fall Prevention
(2). However, this is only the beginning of ings are not necessarily inter-related, inter- Coalitions was not a facilitator to program
the fall prevention story. dependent, or even connected by a com- implementation (11).
Once medical risk factors are managed, mon culture or value system. Besides being A similar story exists from the public
the focus of risk management should tran- in the same physical location, there is no health perspective. State-Based Fall Coali-
sition to behavioral and environmental fac- common commitment among members of tions identified working with healthcare
tors (3). This will ensure that the older a village. providers to disseminate evidence-based
adult has the ability to safely interact with This scenario of assumed but not con- fall prevention programs as a top priority
their environment to prevent a future fall. firmed alignment of priorities and goals (12). However, it is clear that a disconnect
One of the most robust interventions is often plays out in fall prevention. Many exists between the expectations and actions
strength and balance training to minimize public health providers mistakenly assume of healthcare providers by the Coalitions
fall risk (4). Two hours of strength and that healthcare providers integrate fall risk may be resulting in gaps in the continuum.
balance training done each week is the min- screening and management into their prac- A final example is the complex and mis-
imum dose required to effectively prevent tices. For example, an evidence-based fall understood role of older adults. Though
a fall or fall-related injury (5). prevention exercise program is offered in almost all Fall Prevention Coalitions have
To achieve this dose of exercise typically the community. An older adult is inter- the goal of education and public aware-
requires a behavior change (6). Established ested in attending the program, and must ness, few, if any, actually have older adults
protocols to transition from a clinically be cleared by their physician before par- as active members of their coalitions (12).
supervised rehabilitation program to an ticipating. The older adults request a falls Preliminary evidence from pilot studies
evidence-based community program will screen from their physician. The physician, supports a disturbing trend that even by
support this behavior change. Once the ini- however, does not understand her expected educating healthcare providers and offer-
tial clinical-community transition is com- role in fall prevention. She has not been ing innovative programing, many older
plete, to further support behavior change, trained in fall screening. She assumes that adults are likely to refuse when offered

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 163 | 38
Shubert Village or tribe?

an intervention to minimize their risk for ACKNOWLEDGMENTS on health care. Phys Ther (2009) 89:324–32. doi:
falling. This work was funded in part by the 10.2522/ptj.20070107
11. Shubert TE, Smith ML, Ory MG, Clarke CB,
A “tribe” differs from a “village” in that University of North Carolina at Chapel
Bomberger SA, Roberts E, et al. Translation of
there is a shared common culture and Hill Prevention Research Center, supported The Otago Exercise Program for adoption and
values. Everyone has a prescribed job to by Cooperative Agreement Number U48- implementation in the United States. Front Pub-
achieve the common goals. For effective DP001944, Centers for Disease Control lic Health (2015) 2:152. doi:10.3389/fpubh.2014.
multi-level fall prevention efforts to hap- and Prevention Special Interest Project 12- 00152
12. Schneider EC, Beattie BL. Building the older adult
pen, we may want to shift the paradigm 058, and the Bureau of Health Profes-
fall prevention movement – steps and lessons
from assuming each stakeholder under- sions (BHPr), Health Resources and Ser- learned. Front Public Health (2015) 2:194. doi:10.
stands their roles to describing and moti- vices Administration (HRSA), Department 3389/fpubh.2014.00194
vating stakeholders to be part of a shared of Health and Human Services (DHHS) 13. St. John JA, Shubert TE, Smith ML, Rose-
social movement. under grant #UB4HP19053, Carolina Geri- mond CA, Howell DA, Beaudoin CE, et al.
Developing an evidence-based fall prevention cur-
What would this look like for future atric Education Center. riculum for community health workers. Front Pub-
efforts? Current tribe building efforts have lic Health (2015) 2:209. doi:10.3389/fpubh.2014.
demonstrated success. For example, the REFERENCES 00209
Oregon State Department of Public Health 1. Tinetti ME, Kumar C. The patient who falls: “it’s
(DPH) engaging the state chapter of the always a trade-off ”. JAMA (2010) 303:258–66.
American Physical Therapy Association to doi:10.1001/jama.2009.2024 Conflict of Interest Statement: The author declares
2. Prevention of Falls in Older Persons AGS/BGS Clin- that the research was conducted in the absence of any
educate physical therapists about the role ical Practice Guidelines. (2010). Available from: commercial or financial relationships that could be
of public health in fall prevention and pro- http://www.medcats.com/FALLS/frameset.htm construed as a potential conflict of interest.
viding partnership models. This partner- 3. Stevens JA, Phelan EA. Development of STEADI: a
ship was designed to facilitate the imple- fall prevention resource for health care providers.
mentation of the OEP. It was discovered Health Promot Pract. (2013) 14(5):706–14. doi:10. This paper is included in the Research Topic, “Evidence-
1177/1524839912463576 Based Programming for Older Adults.” This Research
that physical therapists were not familiar 4. Gillespie LD, Robertson MC, Gillespie WJ, Lamb Topic received partial funding from multiple government
with the OEP. Once they were invited to SE, Gates S, Cumming RG, et al. Interven- and private organizations/agencies; however, the views,
engage with the program, and supported tions for preventing falls in older people liv- findings, and conclusions in these articles are those of the
by the Oregon State DPH, OEP adop- ing in the community. Cochrane Database Syst authors and do not necessarily represent the official posi-
Rev (2012) 9:CD007146. doi:10.1002/14651858. tion of these organizations/agencies. All papers published
tion and implementation rates increased.
CD007146.pub2 in the Research Topic received peer review from members
The Community Health Worker (CHW) 5. Tiedemann A, Sherrington C, Close JC, Lord SR. of the Frontiers in Public Health (Public Health Edu-
Training described by St. John et al. is Exercise and sports science Australia position state- cation and Promotion section) panel of Review Editors.
another example (13). The goal is to edu- ment on exercise and falls prevention in older peo- Because this Research Topic represents work closely asso-
cate CHWs about their role in fall preven- ple. J Sci Med Sport (2011) 14:489–95. doi:10.1016/ ciated with a nationwide evidence-based movement in
j.jsams.2011.04.001 the US, many of the authors and/or Review Editors may
tion, and in turn to help the CHWs educate
6. Buchner DM, Sheppard L, Senior J, Park CH, have worked together previously in some fashion. Review
their older adult clients. This will ensure Mockenhaupt R, Chodzko-Zajko W, et al. Physical Editors were purposively selected based on their expertise
that the CHW can contribute to the tribe activity to prevent or reverse disability in seden- with evaluation and/or evidence-based programming
by contributing to a knowledge base they tary older adults: the National Blueprint Consen- for older adults. Review Editors were independent of
can use to educate and engage with other sus Conference summary report: strategic priori- named authors on any given article published in this
ties for increasing physical activity among adults volume.
healthcare providers who care for their aged 50 or more. Am J Prev Med (2003) 25:214–5.
clients (13). doi:10.1016/S0749-3797(03)00188-0
Starting with a small group and cross- 7. Clemson L, Cumming RG, Kendig H, Swann Received: 16 June 2014; accepted: 11 September 2014;
ing between disciplines, educating and M, Heard R, Taylor K. The effectiveness of published online: 27 April 2015.
a community-based program for reducing the Citation: Shubert TE (2015) Village or tribe? Expec-
engaging all key players, and building a
incidence of falls in the elderly: a randomized trial. tations, roles, and responsibilities for effective fall
common culture of fall prevention will J Am Geriatr Soc (2004) 52:1487–94. doi:10.1111/ prevention efforts. Front. Public Health 2:163. doi:
be the key to creating an effective tribe. j.1532-5415.2004.52411.x 10.3389/fpubh.2014.00163
Every member, no matter how old or 8. Li F, Harmer P, Fisher KJ, McAuley E, Chaume- This article was submitted to Public Health Education
young, licensed professional or commu- ton N, Eckstrom E, et al. Tai Chi and fall reduc- and Promotion, a section of the journal Frontiers in
tions in older adults: a randomized controlled trial. Public Health.
nity provider, has a significant role to play, J Gerontol A Biol Sci Med Sci (2005) 60:187–94. Copyright © 2015 Shubert . This is an open-access arti-
they just may not know it yet. As we move doi:10.1093/gerona/60.2.187 cle distributed under the terms of the Creative Commons
forward in dissemination and implemen- 9. Ganz DA, Alkema GE, Wu S. It takes a village Attribution License (CC BY). The use, distribution or
tation of evidence-based fall prevention to prevent falls: reconceptualizing fall prevention reproduction in other forums is permitted, provided the
programs nationally, the more members and management for older adults. Inj Prev (2008) original author(s) or licensor are credited and that the
14:266–71. doi:10.1136/ip.2008.018549 original publication in this journal is cited, in accordance
we recruit to the tribe, the more success- 10. Shumway-Cook A, Ciol MA, Hoffman J, Dudgeon with accepted academic practice. No use, distribution or
ful we will be at addressing the problem BJ, Yorkston K, Chan L. Falls in the Medicare pop- reproduction is permitted which does not comply with
of falls. ulation: incidence, associated factors, and impact these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 163 | 39

 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00270

EnhanceFitness: a 20-year dissemination history

Susan J. Snyder *, Meghan Thompson and Paige Denison
Partners in Care Foundation, San Fernando, CA, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: EnhanceFitness, evidence-based, dissemination, older adults, healthcare providers

EnhanceFitness (EF) is a prime exam- the program and oversee its dissemina- addition, these agencies have helped to
ple of an evidence-based physical activ- tion. Senior Services’ dissemination strat- fund Senior Services’ development of a
ity program that has been disseminated egy has been to license, train, and sup- comprehensive online data management
far beyond its original study site. This port community-based delivery sites that platform that allows both local and cen-
commentary will provide an overview adopt EF. This strategy has been quite tralized reporting about participant- and
of the evidence, history, successes, chal- successful over the years and has bal- site-level participation and outcomes. The
lenges, and vision for increasing the anced the need to maintain fidelity to the same platform supports Senior Services’
availability and accessibility of EF. This program’s protocols with the mission to tracking of program licensing and funding,
overview is intended to be an example expand the program’s reach in a sustainable as well as site, staff, and training informa-
for evidence-based programs that want to way (2). tion. This internal data management infra-
move beyond their study sites to wider Since the years following the original structure has been invaluable in allowing
dissemination. study, from 1999 to mid-2014, EF has been Senior Services to monitor program reach
In 1994, researchers at the University offered in 34 states, at nearly 700 locations. and fidelity, while the user-facing element
of Washington Health Promotion Research Early on, expansion was largely due to allows EF’s licensees to provide meaningful
Center (UW HPRC) and Group Health Senior Services’ marketing of the program reports both to their funders and to their
Cooperative (a health maintenance organi- at the annual conference of the National participants, and to monitor the success of
zation) collaborated with Senior Services, Council on Aging and American Society of their own implementation efforts. At the
a non-profit community-based organiza- Aging. In 2006, the US Health and Human other end of the spectrum from federal
tion, to conduct a trial of a multicompo- Services’ Administration on Aging (AoA) agencies, program participants themselves,
nent disability prevention program. One included EF as one of the approved pro- not to mention their physicians, bolstered
hundred older adults were recruited for grams for the Choices for Independence by subjective (5) and objectively demon-
a 6-month study at a State of Washing- grants, placing it in the AoA’s highest tier of strable (6) changes in health and fitness,
ton senior center. Even before the pilot evidence-based programs (3). As a result, have been important champions for dri-
study ended, participants requested to keep program adoption increased significantly ving demand for program expansion at the
the exercise class component of the inter- in the grantee states. This growth contin- community level.
vention as a permanent activity at the ued in the following year (2007), when the EnhanceFitness has benefited from a
center. Not only were the center mem- Centers for Disease Control and Prevention strong continuing relationship with its
bers excited about participating, but study Arthritis Program (CDC-AP) reviewed and original academic and healthcare research
measures showed that the intervention classified EF as “arthritis-friendly” and it partners. This relationship has provided
group significantly improved in fitness and was adopted as a recommended inter- many opportunities to participate in
health: 13% improvement in social func- vention by the Arthritis Program (4). As research and evaluation efforts beyond the
tion; 52% improvement in depression; and of mid-2014, according to data collected initial study. This work and the resulting
35% improvement in physical function- and maintained by Senior Services from articles published in professional journals
ing. The control group (center members program implementation sites since 1999, (7) have ensured that the program’s pro-
who did not participate in the program EF has served over 45,000 unduplicated tocol is kept up-to-date with the latest
but who attended other senior center activ- participants. research in older adult fitness. Evaluation
ities) deteriorated in these measures over Throughout the program’s history, cru- of the program’s adaptation for various cul-
the same period (1). cial support from several directions has tural groups has demonstrated its ability
After completion of the original study, spurred and sustained its growth. National to achieve acceptance and positive out-
wishing to see that the successful program policymakers and funders have embraced comes in new settings (8). Increased pro-
move beyond the original study site, part- the program, prompting significant uptake gram dissemination to a variety of sites
ner agencies agreed that Senior Services far beyond the borders of the program’s and populations brings increased organiza-
was best positioned to hold the license for home region of western Washington. In tional complexity as Senior Services seeks

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 270 | 40
Snyder et al. EnhanceFitness dissemination history

to support EF licensees in their imple- will fill an ever-growing need. Expansion Available at: http://www.cdc.gov/pcd/issues/2005/
mentation and sustainability challenges. of locations through national partners, apr/04_0142j.htm
The technicalities of managing what may such as the Y of the USA and its 2,600
be multiple new class sites, as well as branches, will strengthen the EF network Conflict of Interest Statement: The authors declare
recruiting, training, and retaining certi- as a whole. Lastly, focus on increased refer- that the research was conducted in the absence of any
fied fitness professionals, are often diffi- rals of patients from healthcare providers commercial or financial relationships that could be
construed as a potential conflict of interest.
cult for EF licensees during implementa- will fortify the links between the healthcare
tion of the program. Most EF licensees are system and older adult wellness and self- This paper is included in the Research Topic, “Evidence-
non-profits and sustainability in funding efficacy. This is all possible based on the Based Programming for Older Adults.” This Research
is a common challenge as they must pay strong foundation built over the last two Topic received partial funding from multiple government
and private organizations/agencies; however, the views,
for the initial EF license, annual license decades. findings, and conclusions in these articles are those of the
renewal fees, and instructors’ wages. The authors and do not necessarily represent the official posi-
expertise of EF partners and researchers REFERENCES tion of these organizations/agencies. All papers published
has assisted in studying solutions to these 1. Wallace JI, Buchner DM, Grothaus L, Leveille S, Tyll in the Research Topic received peer review from members
challenges. L, LaCroix AZ, et al. Implementation and effective- of the Frontiers in Public Health (Public Health Edu-
ness of a community-based health promotion pro- cation and Promotion section) panel of Review Editors.
EnhanceFitness has a vision for the gram for older adults. J Gerontol A Biolo Sci Med Sci Because this Research Topic represents work closely asso-
future based on learning from additional (1998) 53a(4):M301–6. ciated with a nationwide evidence-based movement in
research, partnerships, and funding since 2. Belza B, Snyder S, Thompson M, LoGerfo J. From the US, many of the authors and/or Review Editors may
its original study. A new partnership with research to practice: EnhanceFitness, an innova- have worked together previously in some fashion. Review
tive community-based senior exercise program. Top Editors were purposively selected based on their expertise
the American Council on Exercise, experts
Geriatr Rehabil (2010) 26(4):299–309. doi:10.1097/ with evaluation and/or evidence-based programming for
in the field of physical activity curricu- TGR.0b013e3181fee69e older adults. Review Editors were independent of named
lum development and training, will ensure 3. Available from: http://www.aoa.gov/AoA_ authors on any given article published in this volume.
that EF instructors have the continuing Programs/HPW/Title_IIID/index.aspx
education that they need to serve older 4. Available from: http://www.cdc.gov/arthritis/ Received: 18 June 2014; accepted: 21 November 2014;
interventions.htm published online: 27 April 2015.
adults with coexisting chronic conditions.
5. Senior Services. EnhanceFitness Program Database – Citation: Snyder SJ, Thompson M and Denison P (2015)
Online delivery of training modules for Participant Self-Rating and Satisfaction Survey Data EnhanceFitness: a 20-year dissemination history. Front.
instructors will help bridge the accessibility [Electronic File]. Seattle, WA: Senior Services (2014). Public Health 2:270. doi: 10.3389/fpubh.2014.00270
gap in remote areas for continuing edu- 6. Ackermann RT, Williams B, Nguyen HQ, Berke EM, This article was submitted to Public Health Education
cation and support of that infrastructure. Maciejewski ML, LoGerfo JP. Healthcare cost dif- and Promotion, a section of the journal Frontiers in
ferences with participation in a community-based Public Health.
The program’s participation in Centers for group physical activity benefit for medicare man- Copyright © 2015 Snyder, Thompson and Denison. This
Medicaid and Medicare Services (CMS) aged care health plan members. J Am Geriatr Soc is an open-access article distributed under the terms of
study beginning in 2014 is another step (2008) 56:1459–65. doi:10.1111/j.1532-5415.2008. the Creative Commons Attribution License (CC BY). The
toward a long-time vision of EF becoming 01804.x use, distribution or reproduction in other forums is per-
a Medicare reimbursable benefit. Contin- 7. Available from: http://www.projectenhance.org/ mitted, provided the original author(s) or licensor are
AboutEvidenceBasedPrograms/Citations.aspx credited and that the original publication in this journal
ued work with the University of Washing- 8. Snyder S, Belza B. Eliminating disparities in is cited, in accordance with accepted academic practice.
ton on the adaptation of the EF program communities of color through the Lifetime Fit- No use, distribution or reproduction is permitted which
for participants with cognitive impairment ness Program [abstract]. Prev Chronic Dis (2005). does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 270 | 41

 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00131

Translating Fit and Strong!: Lessons learned and next steps

Susan Hughes 1,2 *, Renae L. Smith-Ray 2 , Amy Shah 2 and Gail Huber 3
1
University of Illinois at Chicago School of Public Health, Chicago, IL, USA
2
University of Illinois at Chicago Center for Research on Health and Aging, Chicago, IL, USA
3
Northwestern University School of Physical Therapy, Chicago, IL, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: older adults, Fit and Strong, physical activity, longitudinal studies, arthritis

Fit and Strong! began in 1998. It grew lower extremity joint pain, the natural ten- TRANSLATION STEPS
out of the Hughes doctoral dissertation dency is to stop moving around, which CHANGE IN INSTRUCTORS
many years ago that examined the impact is, of course, the worst thing that peo- The efficacy trial sought to demonstrate
of a model long-term home care program ple can do. A sedentary lifestyle leads to that a structured program of aerobic exer-
for older adults. We learned at that time further joint stiffening, pain, muscle weak- cise and resistance training would not
(1981) that arthritis was the most com- ness, aerobic de-conditioning, and weight harm persons with painful lower extrem-
mon chronic condition reported by home- gain; potentially setting people up for the ity joints. The program was delivered by
bound clients and the condition that was onset of co-morbid conditions like heart trained physical therapists who had expe-
most frequently cited by them as interfer- disease and diabetes (5–9). So we decided rience working with persons with OA, but
ing greatly with their function. To learn that our intervention must consist of a this was an expensive model for transla-
more about this story, we obtained fund- multiple component physical activity pro- tion. By this time, we had obtained funding
ing from the National Institutes of Health gram that included aerobic walking and to test different ways of bolstering main-
(NIH) to conduct a prospective, longitudi- strength training. We also wanted to design tenance of physical activity after Fit and
nal study in Chicago of 600 seniors who a short term (8 weeks) program that had Strong! ends. This effectiveness trial was
were unselected for presence of arthritis long-term results. Therefore, we talked conducted on the south side of Chicago,
at baseline. We found again that arthritis to experts in the field and learned that enabling us to expand the reach of the pro-
was the most common condition reported we needed to also include a health edu- gram into largely African American com-
and the number one cause of disability (1). cation/behavior change component. Like munities. We used this study as an oppor-
We also measured participant joint impair- the evidence-based Chronic Disease Self- tunity to conduct a natural experiment.
ment and conducted an analysis to try to Management Program (CDSMP), we bor- We used the physical therapist instructor
determine which joints were causing the rowed heavily from the self-efficacy litera- model with the first 200+ enrollees and
problem. Analyses clearly indicated that ture to design this piece that helps people then taught the remaining 300 enrollees
osteoarthritis (OA) in the lower extrem- gain mastery over their OA through an using certified exercise instructors. Out-
ity joints was the culprit, a scenario that active lifestyle (10). comes were very strong at 8 weeks and
makes sense when considering that peo- The resulting program, Fit and Strong!, 6 months with both types of instructors,
ple use these large weight bearing joints consists of three 90-min sessions per week attendance was high and participant evalu-
to perform most activities of daily living over 8 weeks. The first hour of each session ations glowing (12). Therefore, we decided
such as transferring, climbing stairs, and is devoted to exercise (flexibility, aerobic, to move forward with the certified exercise
toileting (2). and lower extremity strengthening) and the instructor model. Overall long-term effects
We conducted the longitudinal study last 30 min is devoted to a structured health from this trial were very strong, including
in order to understand the links between education/group problem solving curricu- significant gains in physical activity over
presence of OA and development of dis- lum. We tested the program in an effi- 18 months of follow up that were accom-
ability. Once we understood the causal cacy trial that found differential benefits in panied by improved lower extremity OA
chain, it was clear that our next step should the treatment group on physical activity, symptoms, observed performance gains
be the development of an intervention to self-efficacy for exercise, and lower extrem- in lower extremity strength, and mobil-
interrupt it. We examined the OA physi- ity stiffness at 8 weeks. At 6 months, those ity (risk factors for falls), and anxiety and
cal activity literature and found that peo- gains were maintained and we saw addi- depression out to 18 months (13).
ple with OA have two problems. They tional benefits on self-efficacy for adher-
are aerobically de-conditioned and have ence to physical activity over time and PARTNERSHIP WITH AAA’S
weaker muscles than age-matched con- lower extremity pain. Several of these gains We subsequently received funding from
trols (3, 4). People who have OA have a were maintained at 12 months with large the Centers for Disease Control and Pre-
lot of pain in their joints. For those with effect sizes (11). vention (CDC) to test the translation of

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 131 | 42
Hughes et al. Translating Fit and Strong!

Fit and Strong! in partnership with Area have already been trained in an evidence- pleased to see that the pilot of the Reynolds
Agencies in Aging in Illinois and North based program and layering the Fit and et al., adaptation of Fit and Strong! for can-
Carolina. This work with community part- Strong! training on top. Currently, we are cer survivors improved their engagement
ners enabled us to develop a license and implementing the lay leader model in IL, in physical activity, self-efficacy for aerobic
fee structure, fine tune Lead (i.e., T), Mas- TX, and MI, USA. Sites that are using exercise as well as symptoms of anxiety and
ter, and Instructor trainings, and develop the lay leader approach continue to use depression.
many other types of materials includ- our interactive website to enter pre- and To conclude, the enduring hallmark of
ing a program implementation guide and post-participant outcome assessments and an evidence-based program is the capac-
participant and instructor manuals. We attendance data. We will be analyzing the ity to produce the same, consistent results
also developed an interactive website that outcome data soon to learn whether par- across different populations, geographic
enables us to track participant attendance, ticipant outcomes with this new instruc- sites, and instructors. Fit and Strong!
conduct program evaluations, and collect tor model are as strong as outcomes seen has demonstrated the capacity to pro-
a reduced set of outcome measures for all with the physical therapist and certified duce nearly identical participant outcomes
participants at baseline and immediately exercise instructor models. Anecdotal feed- across six different evaluations with Cau-
post-program. The outcomes now tracked back from participating sites has been very casian, African American, and Hispanic
across sites include: body mass index positive. participants; across sites in IL, MI, NC,
(BMI), lower extremity joint pain and stiff- TX, and AZ; and with three different types
ness, self-efficacy for exercise, engagement NEXT STEPS of instructors – physical therapists, certi-
in physical activity, and energy/fatigue. We Finally, while offering the program on fied exercise instructors, and experienced
also learned along the way that some sites the south side of Chicago we were asked evidence-based program lay leaders. Our
find it more practical to offer the pro- by participants to include more informa- program that combines structured physical
gram two times per week. We now allow tion in the Fit and Strong! participant activity with health education for building
providers to make this adaptation to the manual about diet/weight management. self-efficacy and behavior change is start-
program when necessary as long as they We researched this issue, learned about ing to demonstrate similar positive out-
cover the full complement of 24 sessions, the strong relationship between over- comes with additional clinical populations
which extends the program to 12 weeks in weight/obesity and knee OA, and obtained like cancer survivors and is being tested
length. funding from the National Institute on with persons with symptoms of depres-
Aging to compare the effectiveness of cus- sion. We are also very excited about the
HISPANIC VERSION OF FIT AND STRONG! tomary Fit and Strong! to that of a new potential contribution of the new Fit and
More recently, we developed and tested a version, Fit and Strong! Plus, that includes Strong! Plus program to not only promote
new Hispanic version of the program, ¡en both physical activity and an explicit a physically active lifestyle but also simulta-
Forma y Fuerte!, in Chicago and Phoenix. dietary change/weight management com- neously promote healthy eating and weight
That pilot taught us that many older Lati- ponent. The new program also has 24 ses- management. As the foregoing demon-
nos who have immigrated to the U.S. have sions, but the health education curriculum strates, the chapter is still very much being
low levels of formal education. Our partic- has been adapted to include dietary behav- written on Fit and Strong! adaptations
ipant manuals are written for eighth grade ior change information intended to facili- and outcomes, so stay tuned for future
literacy levels. We are now revising the tate participant weight loss. Early returns developments!
Hispanic Manual to a fourth grade level from this study have been quite positive
and will work with instructors to use pic- (15). We are also currently working with REFERENCES
1. Hughes SL, Dunlop D. The prevalence and impact
tures and stories to get points across. We a colleague in the Department of Psychia-
of arthritis in older persons. Arthritis Care Res
obtained participant baseline, 8-week and try at University of Illinois at Chicago who (1995) 8(4):257–64. doi:10.1002/art.1790080409
6-month outcomes for this pilot. Prelimi- is testing an adapted version of Fit and 2. Dunlop DD, Hughes SL, Edelman P, Singer RM,
nary analyses show strong results at both Strong! for persons who exhibit symptoms Chang RW. Impact of joint impairment on disabil-
time points and we plan to publish the of depression. This pilot is currently under- ity and disability-specific domains at four years.
J Clin Epidemiol (1998) 51:1251–61. doi:10.1016/
findings very soon (14). way with older veterans who have been
S0895-4356(98)00128-0
seeking treatment for depressive symp- 3. Minor MA, Hewett JE, Webel RR, Anderson
LAY LEADER EFFORT toms. This effort to adapt Fit and Strong! SK, Kay DR. Efficacy of physical condition-
There is currently no process in place for use with a specific clinical population ing exercise in patients with rheumatoid arthri-
for providers to help graduates of an is very similar to the effort reported in tis and osteoarthritis. Arthritis Rheum (1989)
32:1396–405. doi:10.1002/anr.1780321108
evidence-based program move on to a this issue in the Reynolds et al. article to 4. Semble EL, Loeser RF, Wise CM. Therapeutic exer-
different, complementary evidence-based adapt and test the program with cancer sur- cise for rheumatoid arthritis and osteoarthritis.
program. For this reason, we obtained vivors (16). Similar to the Reynolds pilot, Semin Arthritis Rheum (1990) 20:32–40. doi:10.
foundation funding to examine initial steps the depression pilot team also removed 1016/0049-0172(90)90092-T
that could be taken to bundle Fit and the arthritis-specific material from manual 5. Lemontowski PW, Zelicof SB. Obesity and
osteoarthritis. Am J Orthop (2008) 37(3):
Strong! with other evidence-based pro- and replaced it with material on recogniz- 148–51.
grams like Matter of Balance and CDSMP. ing symptoms of depression and managing 6. Vignon E, Valat J, Rossignol M, Avouac B, Rozen-
This lay leader effort is training people who them with physical activity. We are very berg S, Thoumie P, et al. Osteoarthritis of the

www.frontiersin.org April 2015 | Volume 2 | Article 131 | 43

Hughes et al. Translating Fit and Strong!

knee and hip and activity: a systematic inter- 12. Seymour RB, Hughes SL, Campbell RT, Huber GM, and private organizations/agencies; however, the views,
national review and synthesis (OASIS). Joint Desai P. Comparison of two methods of conduct- findings, and conclusions in these articles are those of the
Bone Spine (2006) 73:442–55. doi:10.1016/j.jbspin. ing the Fit and Strong program. Arthritis Rheum authors and do not necessarily represent the official posi-
2006.03.001 (2009) 61(7):876–84. doi:10.1002/art.24517 tion of these organizations/agencies. All papers published
7. Alberti KGMM, Eckel RH, Grundy SM, Zim- 13. Hughes SL, Seymour RB, Campbell RT, Desai in the Research Topic received peer review from members
met PZ, Cleeman JI, Donato KA, et al. Harmo- P, Huber G, Chang HJ. Fit and Strong! Bolster- of the Frontiers in Public Health (Public Health Edu-
nizing the metabolic syndrome: a joint interim ing maintenance of physical activity among older cation and Promotion section) panel of Review Editors.
statement of the International Diabetes Federa- adults with lower extremity osteoarthritis. Am J Because this Research Topic represents work closely asso-
tion Task Force on Epidemiology and Preven- Health Behav (2010) 34(6):750–63. doi:10.5993/ ciated with a nationwide evidence-based movement in
tion; National Heart, Lung, and Blood Institute; AJHB.34.6.10 the US, many of the authors and/or Review Editors may
American Heart Association; World Heart Feder- 14. Der Ananian C, Hughes SL, Miller A, Shah A. Six- have worked together previously in some fashion. Review
ation; International Atherosclerosis Society; and Month Outcomes of ¡Fuerte y en Forma! in Latinos Editors were purposively selected based on their expertise
International Association for the Study of Obe- with Arthritis. Gerontological Society of America with evaluation and/or evidence-based programming
sity. Circulation (2009) 120:1640–5. doi:10.1161/ 2014 Conference (2014). for older adults. Review Editors were independent of
CIRCULATIONAHA.109.192644 15. Smith-Ray RL, Fitzgibbon ML, Tussing- named authors on any given article published in this
8. Healy GN, Matthews CE, Dunstan DW, Win- Humphreys L, Schiffer L, Shah A, Huber GM. volume.
kler EAH, Owen N. Sedentary time and cardio- Fit and Strong! Plus: design of a comparative
metabolic biomarkers in US adults: NHANES effectiveness evaluation of a weight management Received: 23 June 2014; accepted: 15 August 2014;
2003-06. Eur Heart J (2011) 32(5):590–7. doi:10. program for older adults with osteoarthri- published online: 27 April 2015.
1093/eurheartj/ehq451 tis. Contemp Clin Trials (2013) 37(2):178–88. Citation: Hughes S, Smith-Ray RL, Shah A and
9. Klein S, Allison DB, Heymsfield SB, Kelley DE, doi:10.1016/j.cct.2013.11.014 Huber G (2015) Translating Fit and Strong!: Lessons
Leibel RL, Nonas C, et al. Waist circumference and 16. Reynolds J, Thibodeaux L, Jiang L, Francis K, learned and next steps. Front. Public Health 2:131. doi:
cardiometabolic risk: a consensus statement from Hochhalter A. Fit & Strong! promotes physical 10.3389/fpubh.2014.00131
Shaping America’s Health: Association for Weight activity and well-being in older cancer survivors. This article was submitted to Public Health Education
Management and Obesity Prevention; NAASO, Front Public Health (2015) 2:171. doi:10.3389/ and Promotion, a section of the journal Frontiers in
The Obesity Society; the American Society for fpubh.2014.00171 Public Health.
Nutrition; and the American Diabetes Associa- Copyright © 2015 Hughes, Smith-Ray, Shah and Huber.
tion. Obesity (2007) 15(5):1061–7. doi:10.1038/ Conflict of Interest Statement: The authors declare This is an open-access article distributed under the terms
oby.2007.632 that the research was conducted in the absence of any of the Creative Commons Attribution License (CC BY).
10. Bandura A. Self Efficacy: The Exercise of Control. commercial or financial relationships that could be The use, distribution or reproduction in other forums is
New York: W.H. Freeman and Company (1997). construed as a potential conflict of interest. permitted, provided the original author(s) or licensor are
11. Hughes SL, Seymour RB, Campbell RT, Huber credited and that the original publication in this journal
G, Pollak N, Sharma L, et al. Long-term impact This paper is included in the Research Topic, “Evidence- is cited, in accordance with accepted academic practice.
of Fit and Strong program. Gerontologist (2006) Based Programming for Older Adults.” This Research No use, distribution or reproduction is permitted which
46:801–14. doi:10.1093/geront/46.6.801 Topic received partial funding from multiple government does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 131 | 44
 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00262

Texercise: the evolution of a health promotion program

Holly Riley *
Center for Consumer and External Affairs, Texas Department of Aging and Disbality Services, Austin, TX, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: health promotions, older adults, Texas, physical activity and wellbeing, partnership, history

Texercise, a program developed by the state work of Texas’ 28 Area Agencies on Aging In 2004, the Texas Legislature reorga-
of Texas, promotes healthy lifestyle choices (AAA) and the SUA’s activities. nized Texas’ Health and Human Services
to help residents age and live their best. The AAAs saw the potential to use Texer- agencies. The SUA, then known as the
This commentary provides the conceptual- cise in implementing Title IIID of the Older Texas Department of Aging, became the
ization of the program, its growth over the Americans Act, which focuses on disease Texas DADS (4). The creation of DADS
years and how it is disseminated. The cur- prevention and health promotion. And the expanded the Texercise program’s primary
rent federal direction on evidence-based SUA saw the potential to enhance its mes- population (60+) to include people 45
prevention program for seniors from the saging to reach beyond the 60-plus popu- and over as well as people with intellec-
Administration for Community Living and lation into worksites and the community. tual and developmental disabilities. It also
its potential influences on the Texercise Texercise also offered the SUA opportu- provided opportunities to further the pro-
program are also provided. nities to involve state-level programs and gram’s resources and delivery methods.
business in partnerships through the pro- New partnerships were developed, includ-
CONCEPTUALIZATION gram. ing partners that provide in-kind incentives
Texercise, a health promotion program of and event development.
the Texas Department of Aging and Dis- DISSEMINATION Texercise Classic, a 12-week program,
ability Services (DADS), has seen dramatic After the launch, a small Texercise book- was developed to accommodate requests
growth since its inception in 1998. Origi- let featuring physical activity exercises and from community organizations wanting
nally developed by the state unit on aging recommendations was developed with the a group exercise program. Texercise Clas-
(SUA) to support the Aging Texas Well guidance of Dr. Kenneth H. Cooper of sic includes motivational and recognition
initiative (1), Texercise is now a compre- the Cooper Aerobics Institute. The initial resources along with the Texercise hand-
hensive health program with outreach and reaction to the booklet was overwhelm- book. A 30-min exercise DVD, featuring
programing across the state. Texas’ obesity ing. The aging network, state partners, and balance, strength, endurance, and flexibil-
statistics and health indicators (2) high- older adults requested more, free wellness ity exercises, was created to illustrate how
light the need to help people improve resources featuring practical information. to perform the exercises. Immediately, Tex-
their health and quality of life. The Aging The positive feedback from these stake- ercise Classic was a hit with senior cen-
Texas Well initiative (3) focuses on prepar- holders helped to generate a more compre- ters, nursing homes, assisted living facili-
ing Texas for a growing aging population hensive Texercise handbook, website, and ties, faith based organizations, and work-
and helps Texans understand the impor- fitness fact sheets. sites.
tance of planning for their futures. Good At this point, the primary distribution The expansion of the target popula-
physical health is a key component of this methods for Texercise resources were the tion and these resources posed a welcomed
initiative. states 28 AAAs and a handful of state- challenge for the Texercise program. More
When first conceived, Texercise was an level partners. Partnerships were developed staff was needed to manage the demand
8-month long statewide exercise campaign with the Texas Governor’s Advisory Coun- and growth of the program and its in-kind
created to support the Aging Texas Well cil on Physical Fitness, major non-profit partnerships. In response, DADS dedicated
campaign. The primary Aging Texas Well associations and statewide media organi- more staff time to administer Texercise,
message was that individuals, local com- zations to increase awareness of Texercise. develop partnerships and meet the needs
munities and the state need to take the As organizations outside the aging network of Texas communities.
appropriate steps to prepare for an aging learned about the program, they began Another challenge was keeping the pro-
society. Texercise was developed to support requesting Texercise resources and tools. gram relevant and timely. Baby boomers
that message, focusing on physical health Simultaneously, non-traditional partner- were turning 60, and they wanted a pro-
and wellness action steps. After the initial ships (e.g., trade associations, civic groups, gram that represented their generation. A
Texercise campaign, agency leaders saw the faith-based entities, private industry, and major redesign of Texercise in 2009 resulted
value of a permanent state-level health pro- businesses) were developed to share Texer- in a fresh, engaging look, and updated
motions program that would support the cise resources. content of all the program’s resources.

www.frontiersin.org April 2015 | Volume 2 | Article 262 | 45

Riley The evolution of texercise

ALIGNING WITH EVIDENCE-BASE Department of Aging and Disability Texercise Select. Front Public Health (2015) 2:291.
MOVEMENT Services Texercise program had been rec- doi:10.3389/fpubh.2014.00291
7. Smith ML, Ory MG, Jiang L, Howell D, Chen S,
Even with heightened awareness of the ognized by the International Council on
Pulczinski JC, et al. Texercise select effectiveness: an
negative health outcomes associated with Active Aging and the Centers for Disease examination of physical activity and nutrition out-
obesity, the United States has not seen a Control and Prevention’s Reference Guide comes. Translational Behav Med: Pract, Policy and
decline in obesity and unhealthy lifestyles. of Physical Activity Programs for older Res (2014) (in-press).
More structured, comprehensive disease adults. In addition, the President’s Coun- Conflict of Interest Statement: The author declares
prevention and wellness programs are in cil on fitness, sports and nutrition and the that the research was conducted in the absence of any
demand. Funders want to see returns on Texas Cardiovascular Disease and Stroke commercial or financial relationships that could be
their investment, and many now require Council also have recognized the Texercise construed as a potential conflict of interest.
that their funding be spent on programs program for its community leadership. The
This paper is included in the Research Topic, “Evidence-
that have been proven to be effective and development of Texercise Select is expected Based Programming for Older Adults.” This Research
are evidence-based. to increase awareness and recognition of Topic received partial funding from multiple government
In 2012, the Administration on Aging Texercise as an evidence-based program. and private organizations/agencies; however, the views,
began requiring that Title IIID Older This alignment with the evidence-based findings, and conclusions in these articles are those of the
authors and do not necessarily represent the official posi-
Americans Act monies be spent on movement will help DADS reach more
tion of these organizations/agencies. All papers published
evidence-based programs (5). DADS lead- people with the message that through regu- in the Research Topic received peer review from members
ership was committed to ensuring the aging lar preventative habits, Texans can age and of the Frontiers in Public Health (Public Health Edu-
and disability networks could continue to live their best for many years to come. cation and Promotion section) panel of Review Editors.
use the Texercise program’s resources. In Because this Research Topic represents work closely asso-
ciated with a nationwide evidence-based movement in
2013, DADS contracted with Texas A&M REFERENCES the US, many of the authors and/or Review Editors may
School of Public Health to develop a Tex- 1. Governor of the State of Texas Executive Depart- have worked together previously in some fashion. Review
ercise component that promises to achieve ment. RP42 Relating to the Creation of the Editors were purposively selected based on their expertise
Aging Texas Well Advisory Committee and Plan.
evidence-based recognition. with evaluation and/or evidence-based programming
(2005). Available from: http://governor.state.tx.us/ for older adults. Review Editors were independent of
This new component (6, 7), Texer- news/executive-order/3687/ named authors on any given article published in this
cise Select, is the perfect complement to 2. Texas Department of State Health Services. Texas volume.
existing evidence-based health programs. Health Indicators. (2014). Available from: http://
While many of these programs focus on healthindicators.dshs.texas.gov/ Received: 16 June 2014; accepted: 12 November 2014;
3. Texas Department of Aging and Disability Ser- published online: 27 April 2015.
behavior modifications to address a spe-
vices. Aging Texas Well. (2014). Available from: Citation: Riley H (2015) Texercise: the evolution of a
cific need, Texercise Select emphasizes pre- http://www.dads.state.tx.us/services/agingtexaswell/ health promotion program. Front. Public Health 2:262.
vention through healthy behaviors. It fea- index.html doi: 10.3389/fpubh.2014.00262
tures structured, facilitator-led classes that 4. Texas Department of Aging and Disability Services. This article was submitted to Public Health Education
focus on nutrition and physical activ- DADS Organizational Structure. About the Agency. and Promotion, a section of the journal Frontiers in
(2014). Available from: http://www.dads.state.tx.us/ Public Health.
ity. Two classes for 10 weeks are adminis- news_info/about/index.html Copyright © 2015 Riley. This is an open-access article
tered in a group setting. The classes and 5. Administration for Community Living. Disease Pre- distributed under the terms of the Creative Commons
associated materials provide participants vention and Health Promotion Services (OAA Title Attribution License (CC BY). The use, distribution or
with a chance to develop healthy habits IIID). (2014). Available from: http://www.acl.gov/ reproduction in other forums is permitted, provided the
while also creating a social support group. NewsRoom/NewsInfo/2013/2013_07_16c.aspx original author(s) or licensor are credited and that the
6. Ory MG, Smith ML, Howell D, Zollinger A, Quinn original publication in this journal is cited, in accordance
Goals and barriers are discussed in this C, Swierc SM, et al. The conversion of a practice- with accepted academic practice. No use, distribution or
group setting, as well as opportunities to based lifestyle enhancement program into a for- reproduction is permitted which does not comply with
recognize positive changes. malized, testable program: from Texercise Classic to these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 262 | 46
 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00256

Translating PEARLS: lessons learned from providers and

participants
Mark B. Snowden 1 , Lesley E. Steinman 1 *, Pamela Piering 2 , Sluggo Rigor 3 and Andrea Yip 4
1
University of Washington Health Promotion Research Center, Seattle, WA, USA
2
Healthy Aging Consultant, Seattle, WA, USA
3
International Drop-In Center, Seattle, WA, USA
4
Seattle-King County Aging and Disability Services, Seattle, WA, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: evidence-based programs, RE-AIM, dissemination, implementation, depression, healthy aging, community-based

The Program to Encourage Active, Reward- includes several ingredients for program people that touch a potential participant’s
ing Lives (PEARLS) began 15 years ago success: it was designed with an adopting life are appropriate referral sources – from
when the director of our local area agency organization as a key partner, the model the Meals on Wheels delivery person to
on aging (AAA) approached the Uni- trains existing staff to deliver PEARLS so the resident services coordinator in a low-
versity of Washington Health Promotion new staff do not need to be hired, and the income housing facility. Participants have
Research Center (HPRC). She was look- program is successfully funded in some shared that having a trusted person –
ing for a way to serve older adults with locations through several diverse funding whether a familiar case worker or pastor –
depression, including those served by the streams. We have learned a lot from orga- make the referral is particularly important
agency’s home and community-based ser- nizations and staff that deliver PEARLS when discussing a sensitive subject such
vices (HCBS) program. Depression in this through our monthly technical assistance as depression (9). These gatekeepers can
population is high, when we analyzed calls and other dissemination research and be trained to administer a brief validated
data from 16,032 elders receiving HCBS activities. We have also learned a lot from depression screen such as the two-item
in Washington State in 2005, two-thirds program participants through PEARLS patient health questionnaire (PHQ-2) (11).
met criteria for clinical depression (1). sessions, focus groups and interviews. In addition to provider engagement, it
This partnership between the university A selection of key learnings is provided is essential to use culturally appropriate
and local aging service providers created below, organized by Glasgow’s RE-AIM materials and media for the target com-
PEARLS, a brief, home-based program to framework to help improve the success munity. Strategies include putting photos
teach people tools to effectively tackle the of evidence-based program delivery in of PEARLS counselors on recruitment fly-
things in their lives that overwhelm them, “real-world” settings (4–6). This frame- ers and publishing participant stories in
and to in turn, improve their depressive work consists of five elements – reach, newsletters, community papers, or in digi-
symptoms. These tools include a seven-step effectiveness, adoption, implementation, tal form (12, 13). Former PEARLS partici-
approach to problem-solving and action and maintenance – that present the over- pants agree that the “best way to reach peo-
planning to increase physical, social, and all public health impact of a program or ple is through people” as they can demon-
pleasant activities. PEARLS is a structured policy. It is important for programs to per- strate what PEARLS is through sharing
intervention delivered in 6 to 8 one-hour form well across each of these five elements their experiences about how the program
visits over the course of a 4- to 5-month in order to maximize overall impact (7). helped them. Motivational interviewing
period. Sessions are tapered from weekly techniques are also useful for engaging
to monthly to give participants an oppor- REACH participants who are reluctant to join the
tunity to practice and learn the skills. More Recruitment is an ongoing implementation program.
information about the program can be challenge for PEARLS. Previous data sug-
found at www.pearlsprogram.org. gest that 10% of eligible participants are EFFECTIVENESS
The Program to Encourage Active, referred to PEARLS and 50% of those are Since the original PEARLS study, PEARLS
Rewarding Lives (PEARLS) reduced enrolled in the program (8, 9). Barriers continues to show positive results in older
depression and improved quality of life exist for both those tasked with recruiting adults with major depression, with all-
in two randomized controlled trials (2, 3). participants and for those invited to partic- age adults, with veterans and vet’s spouses
Since then, UW HPRC continues to work ipate in the program. Successful PEARLS or widows, and with elders with low
with the local AAA and other sites to help programs engage a range of community- literacy and with limited English profi-
translate the evidence-based program into based providers to refer to PEARLS that is ciency. PEARLS has been implemented
everyday practice. The implementation similar to the gatekeeper approach used in with bicultural, bilingual counselors in
challenges are striking given that PEARLS other mental health programs (10). Many Hispanic, Chinese, Vietnamese, Korean,

www.frontiersin.org April 2015 | Volume 2 | Article 256 | 47

Snowden et al. PEARLS: provider and participant perspectives

and Filipino communities, and using identify appropriate and accessible longer- program needs including an online com-
trained medical interpreters with Somali term treatment options after the brief ponent and site-based trainings. Former
and Russian-speaking elders. More recent intervention ends. PEARLS program participants are being
studies demonstrate that the improve- engaged to spread the word about PEARLS
ment in depressive symptoms extended IMPLEMENTATION in their communities.
for 18 months following baseline (14). We One of the reasons that evidence-based There are many opportunities for con-
often hear stories on our technical assis- programs are adopted is because research tinuing to improve how PEARLS is deliv-
tance calls about how PEARLS benefits a shows them to be effective. Thus, there ered and spread across the country and
participant’s life, such as helping a client is a concern that fidelity to the research beyond. While PEARLS programs continue
change their blood pressure medication to model be maintained when implement- to spread across the country, this dissemi-
minimize side effects, submitting paper- ing the program. We developed a PEARLS nation pattern is more the result of pas-
work for subsidized housing, or getting fidelity instrument to assist in measuring sive diffusion and a by-product of the
a respite care worker to come in 1 day a fidelity and found that differences exist for ongoing training program and PEARLS
week. As one 95-year-old participant put clinical supervision, counselor assessment, inclusion in several evidence-based prac-
it, “PEARLS rocks,” as he now does 50 client eligibility, and some content and for- tice registries. We need future research
repetitions on his rocking chair for phys- mat of PEARLS sessions (16). We do not of more active dissemination approaches
ical activity. Immigrant elders that partici- necessarily view this as a negative since pro- (such as policy-level interventions) cou-
pate in PEARLS have overcome loneliness grams need to adapt PEARLS to fit their pled with ongoing dissemination and
and homesickness, feel more self-sufficient, local implementation environment. implementation research for overcoming
independent, an overall sense of dignity, Whether a person is appropriate for challenges. An economic evaluation of the
and at “peace-of-mind,” and acculturate PEARLS is one of the most common program through formal cost effective-
more quickly into their new community questions we get on our technical assis- ness or return on investment (ROI) analy-
through social contacts and physical fit- tance calls. In practice, organizations see ses might facilitate broader dissemination
ness. PEARLS participants have also iden- complex clients who often do not have activities. Future directions for PEARLS
tified how the PEARLS process and work- any other acceptable options for depres- also include building capacity through
sheet helped them to improve their focus sion treatment. Expanding eligibility cri- expanding online and regional training
on certain issues and their ability to prior- teria may require adaptations; for exam- options. With continued interest in fidelity
itize and plan, thus, feeling more control ple, focusing on increasing physical and assessment, a larger validation study of the
over things that had once seemed quite social activities rather than problem- PEARLS fidelity instrument is needed to
scattered (15). solving for participants with mild cogni- establish the validity of the items as well
tive impairment or for those for whom a as the innovative methodological approach
ADOPTION problem-solving approach is not a cultural of having a self-reported fidelity assess-
During the initial PEARLS research study, norm. Some adaptations occur naturally ment. Exploring the relationship of fidelity
master’s level social workers and nurses as PEARLS spreads across the country in to client outcomes could then follow and
were trained to deliver the intervention. diverse settings and communities. Strate- allow for refinement of the instrument
A geriatric psychiatrist provided clinical gies for working with low-literacy partic- and better elucidation of the key compo-
supervision. In practice, bachelor’s level ipants include reading worksheets aloud nents of PEARLS to guide program adapta-
case managers and social work interns have and having the counselor or caregiver help tion to best fit local implementation needs.
successfully implemented PEARLS. They fill out the worksheet, being mindful of Addressing these needs will help PEARLS
may not only require more supervision what is written when others will read the achieve its full potential in improving the
up front (such as with administering the worksheets. lives of older adults.
PHQ-9) but also come to PEARLS with
less ingrained therapeutic modalities that MAINTENANCE ACKNOWLEDGMENTS
may need to be put aside when deliver- The Program to Encourage Active, Reward- The authors thank PEARLS providers and
ing a structured, participant-driven pro- ing Lives is now active in 50 agencies across participants who have shared their expe-
tocol like PEARLS. A clinical psycholo- 18 states. Some agencies have only begun riences and expertise through PEARLS
gist or other clinician with experience in implementing the program in the past technical assistance and research activities.
late-life depression and problem-solving 6 months while others are over 10 years old.
treatment can provide clinical supervision, Sustainable funding for PEARLS remains
along with a medical provider who brings
REFERENCES
a challenge yet successes such as the Cal- 1. Carlson W, Snowden M. Community treatment of
expertise on co-occurring chronic con- ifornia “Millionaire’s Tax” supporting the older adults: principles and evidence supporting
ditions and medication use. Community Mental Health Services Act, prevention and mental health service interventions. Clin Geriatr
mental health agencies can offer PEARLS early intervention (PEI) funding in Los Med (2014) 30(3):655–61. doi:10.1016/j.cger.2014.
as part of their menu of options for per- Angeles, and a property tax levy and a 04.010
2. Ciechanowski P, Wagner E, Schmaling K, Schwartz
sons with mental illness. PEARLS may also Medicaid waiver in Washington State hold S, Williams B, Diehr P, et al. Community-
be a first step in a person’s depression promise. The PEARLS training program integrated home-based depression treatment in
treatment, using the PEARLS sessions to continues to support new and existing older adults: a randomized controlled trial. JAMA

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 256 | 48
Snowden et al. PEARLS: provider and participant perspectives

(2004) 291(13):1569–77. doi:10.1001/jama.291. Takes To The Street. New York, NY: Guilford (1990). This paper is included in the Research Topic, “Evidence-
13.1569 p. 195–209. Based Programming for Older Adults.” This Research
3. Ciechanowski P, Chaytor N, Miller J, Fraser R, 11. Kroenke K, Spitzer RL, Williams JB. The Topic received partial funding from multiple government
Russo J, Unutzer J, et al. PEARLS depression treat- patient health questionnaire-2: validity of and private organizations/agencies; however, the views,
ment for individuals with epilepsy: a random- a two-item depression screener. Med Care findings, and conclusions in these articles are those of the
ized controlled trial. Epilepsy Behav (2010) 19(3): (2003) 41(11):1284–92. doi:10.1097/01.MLR. authors and do not necessarily represent the official posi-
225–31. doi:10.1016/j.yebeh.2010.06.003 0000093487.78664.3C tion of these organizations/agencies. All papers published
4. Glasgow RE, Vogt TM, Boles SM. Evaluating the 12. PEARLS Helps End Depression for Seattle’s Filipino in the Research Topic received peer review from members
public health impact of health promotion inter- Elders, Filipino-American Bulletin/New America of the Frontiers in Public Health (Public Health Edu-
ventions: the RE-AIM framework. Am J Public Media, News Feature, Sluggo Rigor, Posted (2013). cation and Promotion section) panel of Review Editors.
Health (1999) 89(9):1322–7. doi:10.2105/AJPH. Available from: http://newamericamedia.org/ Because this Research Topic represents work closely asso-
89.9.1322 2013/03/pearls-helps-end-depression-for-seattles- ciated with a nationwide evidence-based movement in
5. Green LW, Glasgow RE. Evaluating the rel- filipino-elders.php the US, many of the authors and/or Review Editors may
evance, generalization, and applicability of 13. The Program to Encourage Rewarding Active Lives: have worked together previously in some fashion. Review
research: issues in translation methodology. A Digital Story, by Paul Snow and by Lori San- Editors were purposively selected based on their expertise
Eval Health Prof (2006) 29:126–53. doi:10.1177/ ford, Aging and Disability Services (2012). Available with evaluation and/or evidence-based programming
0163278705284445 from: http://www.agingkingcounty.org/PEARLS/ for older adults. Review Editors were independent of
6. Klesges LM, Estabrooks PA, Glasgow RE, Dze- 14. Chaytor N, Ciechanowski P, Miller JW, Fraser R, named authors on any given article published in this
waltowski D. Beginning with the application in Russo J, Unutzer J, et al. Long-term outcomes from volume.
mind: designing and planning health behavior the PEARLS randomized trial for the treatment
change interventions to enhance dissemination. of depression in patients with epilepsy. Epilepsy Received: 15 June 2014; accepted: 09 November 2014;
Ann Behav Med (2005) 29:66S–75S. doi:10.1207/ Behav (2011) 20(3):545–9. doi:10.1016/j.yebeh. published online: 27 April 2015.
s15324796abm2902s_10 2011.01.017 Citation: Snowden MB, Steinman LE, Piering P, Rigor
7. Belza B, Toobert DJ, Glasgow RE. RE-AIM for 15. Steinman L, Cristofalo M, Snowden M. Implemen- S and Yip A (2015) Translating PEARLS: lessons learned
Program Planning: Overview and Applications. tation of an evidence-based depression care man- from providers and participants. Front. Public Health
Center for Healthy Aging and National Council on agement program (PEARLS): perspectives from 2:256. doi: 10.3389/fpubh.2014.00256
Aging. (2014). Available from: http://www.ncoa. staff and former clients. Prev Chronic Dis (2012) This article was submitted to Public Health Education
org/improve-health/center-for-healthy-aging/ 9:110250. doi:10.5888/pcd9.110250 and Promotion, a section of the journal Frontiers in
content-library/IssueBrief_ReAim_Final-2.pdf 16. Farren L, Snowden M, Steinman L, Monroe-DeVita Public Health.
8. Unpublished Data, R-18 PEARLS Dissemination M. Development and evaluation of a fidelity Copyright © 2015 Snowden, Steinman, Piering , Rigor
Study. instrument for PEARLS. Front Public Health and Yip. This is an open-access article distributed under
9. Steinman L, Hammerback K, Snowden M. It could (2015) 2:200. doi:10.3389/fpubh.2014.00200 the terms of the Creative Commons Attribution License
be a pearl to you: exploring recruitment and reten- (CC BY). The use, distribution or reproduction in other
tion of the program to encourage active, rewarding forums is permitted, provided the original author(s) or
lives (PEARLS) with hard-to-reach populations. Conflict of Interest Statement: The authors declare licensor are credited and that the original publication in
Gerontologist (2013). doi:10.1093/geront/gnt137 that the research was conducted in the absence of any this journal is cited, in accordance with accepted aca-
10. Raschko R. The gatekeeper model for the isolated, commercial or financial relationships that could be demic practice. No use, distribution or reproduction is
at-risk elderly. In: Cohen NL, editor. Psychiatry construed as a potential conflict of interest. permitted which does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 256 | 49

 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00261

The CDC Healthy Aging Research Network: advancing

science toward action and policy for the evidence-based
health promotion movement
Basia Belza 1,2 *, Mary Altpeter 3 , Steven P. Hooker 4 and Gwen Moni 1
1
Department of Health Services, Health Promotion Research Center, University of Washington, Seattle, WA, USA
2
Department of Biobehavioral Nursing and Health Systems, Health Promotion Research Center, University of Washington, Seattle, WA, USA
3
Center for Health Promotion and Disease Prevention, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
4
School of Nutrition and Health Promotion, Arizona State University, Phoenix, AZ, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: evidence-based, dissemination, aging, networks, health promotion

Despite recent progress in the uptake of research in diverse communities to develop and design of the initiative (e.g., pro-
evidence-based health promotion (EBHP) research priorities (4–10) and to build a gram selection, evaluation components),
programs within communities, many fac- knowledge base for EBHP programs (11– helped launch workshops for prospective
tors contribute to the need to focus on 14). Through these partnerships, HAN grantees, incorporated RE-AIM into the
dissemination. These include the growth researchers have developed and tested prac- grant proposal structure, reviewed grant
in the aging population, health care tical tools and resources for the develop- applications, and provided training mate-
resource limitations, and interests in pre- ment, implementation and evaluation of rials for staff and grantees about RE-AIM.
serving community-based opportunities interventions and frameworks (e.g., RE- HAN also served on the leadership coun-
for maintaining independence and max- AIM) for their dissemination and sustain- cil and provided grantees with post-award
imizing quality of life. For these reasons, ability (15, 16). technical assistance.
The Prevention Research Centers’ Healthy For instance, HAN created and tested
Aging Research Network (HAN), funded many of the programs described in this WE ENHANCE CAPACITY
by the Centers for Disease Control and issue of Frontiers [i.e., EnhanceFitness, To support the translation of EBHP pro-
Prevention’s (CDC’s) Healthy Aging Pro- Chronic Disease Self-management Pro- grams into practice and policy, HAN
gram, has as its core mission, to translate gram (CDMSP), Fit and Strong! and Pro- has helped to enhance the capacity of
effective healthy aging interventions into gram to Encourage Active Rewarding Lives researchers and practitioners. At the local,
sustainable community-based programs. (PEARLS) (17–23)]. Nationally, we pro- state, and national level, HAN has men-
Researchers and community-based stake- vided technical assistance on EBHP pro- tored and provided leadership opportu-
holders collaborate across HAN’s seven gram implementation and evaluation for nities for graduate students, early career
member center and two affiliate univer- the Administration on Aging (within the investigators, and CDC Healthy Aging
sities (Figure 1) to develop and imple- Administration for Community Living) Program fellows by encouraging them
ment health promotion programs for older and grantee organizations. We have docu- to actively participate in HAN EBHP
adults at individual, organizational, envi- mented our methods of technical assistance initiatives. Working with practice part-
ronmental, and policy levels (1–3). This in numerous peer-reviewed publications ners and national stakeholders, HAN has
commentary highlights selected HAN con- (2, 13, 19, 24, 25). also built professional capacity by devel-
tributions to the EBHP movement from An example of our regional efforts oping and delivering accessible, state-
2001 to 2014. These contributions serve is reflected in the HAN’s EBHP part- of-the-science trainings and resources.
as examples of potential models for future nership with the Health Foundation of These include: conferences, online train-
partnership efforts to enhance implemen- South Florida (HFSF) and Florida Healthy ing modules (www.healthyagingprograms.
tation, dissemination, and sustainability of Aging Collaborative. HFSF is a not-for- org/content), monographs, and issue briefs
EBHP programs. profit grant-making organization with a about EBHP practice and various aspects
focus on expanding access to affordable, of program delivery and quality assurance,
WE BUILD THE FOUNDATION FOR EBHP quality healthcare for underserved popu- physical activity, mental health, environ-
PROGRAMS lations in Florida’s Broward, Miami-Dade, ment, and policy (26, 27).
The HAN has engaged researchers and and Monroe counties. HFSF launched a Healthy Aging Research Network
practitioners from multiple disciplines and tri-county, 5-year $7.6 million health pro- secured and leveraged a CDC conference
community organizations. We use the prin- motion and disease prevention initiative. grant to develop and deliver research-
ciples of community-based participatory HAN assisted with the initial planning to-practice symposia on physical activity,

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 261 | 50
Belza et al. Advancing science

FIGURE 1 | CDC Healthy Aging Research Network (HAN) member centers and affiliates (FY 2009-2014).

mental health, and environmental poli- level, HAN worked with the Washington promotion for older adults, and investigate
cies. This series brought together national State Unit on Aging to apply the rec- the effectiveness of EBHP programs.
research and community partners to strate- ommendations to the agency’s depres-
gize how best to disseminate and sustain sion screening policy to utilize a val- ACKNOWLEDGMENTS
effective community-based programs and idated depression screening measure in We acknowledge our HAN members who
practices. For this series, HAN engaged annual assessments of clients who receive have led and participated in the initia-
new partners (e.g., AARP, The Carter services. As a result, the Area Agencies tives described in this commentary. The
Presidential Center, and the Rosalynn on Aging in Washington have a better HAN is a Prevention Research Centers
Carter Georgia Mental Health Forum, understanding of what proportion of their program funded by the CDC Healthy
CDC Healthy Communities Program). clients are depressed. In addition, prac- Aging Program. Efforts were supported
HAN also secured additional funds from titioners can use this screening measure in part by cooperative agreements from
the Retirement Research Foundation to to determine client eligibility for PEARLS, CDC’s Prevention Research Centers Pro-
develop post-conference products and an evidence-based program for depres- gram: U48-DP-001911, 001908, 001921,
from the Agency for Healthcare Qual- sion. Consequently, evidence-based proce- 001924, 001936, 001938, and 001944. The
ity and Research to provide technical dures and programs are now integrated findings and conclusions in this report are
assistance. The ultimate result was the into this state’s existing aging and social those of the authors and do not necessarily
dissemination and uptake of a monograph services. represent the official position of the Cen-
(26), two coordinated and well-attended In summary, HAN is the go-to source ters for Disease Control and Prevention
series of online webinars, as well as pre- for technical assistance in large-scale EBHP or the Department of Health and Human
sentations and action briefs. HAN also program and policy design, implementa- Services.
contributed to the training of practition- tion, and evaluation efforts with regional,
ers through presentations to the National national, and academic partners. HAN has REFERENCES
1. Hunter RH, Anderson LA, Belza B, Bodiford K,
Association of Chronic Disease Directors harnessed the power and cost-effectiveness
Hooker SP, Kochtitzky CS, et al. Environments
and National Association of State Units on of multi-disciplinary, multi-site endeavors for healthy aging: linking prevention research and
Aging – Healthy Aging Initiative. and become a recognized leader, able to public health practice. Prev Chronic Dis (2013)
convene disparate groups of stakeholders 10:E55. doi:10.5888/pcd10.120244
WE AFFECT PRACTICE AND POLICY to build the science for EBHP. HAN inves- 2. Lang JE, Anderson L, LoGerfo J, Sharkey J, Belansky
E, Bryant L, et al. The prevention research centers
At the national level, the Task Force tigators will continue to serve as facili-
Healthy Aging Research Network. Prev Chronic Dis
on Community Preventive Services pub- tators and bridge builders to expand the (2006) 3(1):A17.
lished recommendations from a HAN overall public health and aging network 3. Wilcox S, Altpeter M, Anderson L, Belza B, Bryant
investigator-led review of community- within and outside of academia. Going for- L, Jones DL, et al. The Healthy Aging Research Net-
based depression interventions on The ward, HAN investigators will continue to work: building capacity for public health and aging
practice. Am J Health Promot (2013) 28(1):206.
Community Guide (28–32). This was the conduct EBHP research to improve capac- doi:10.4278/ajhp.121116-CIT-564
first time the Task Force accepted the find- ity building, determine optimal methods 4. Prohaska T, Belansky E, Belza B, Buchner D, Mar-
ings of an “external” review. At the state for facilitating systems change in health shall V, McTigue K, et al. Physical activity, public

www.frontiersin.org April 2015 | Volume 2 | Article 261 | 51

Belza et al. Advancing science

health, and aging: critical issues and research pri- community-based exercise program on function (2007) 33(3):222–49. doi:10.1016/j.amepre.2007.
orities. J Gerontol B Psychol Sci Soc Sci (2006) and health in older adults: the EnhanceFitness 04.035
61(5):S267–73. doi:10.1093/geronb/61.5.S267 program. J Appl Gerontol (2006) 25(4):291–306. 30. Steinman LE, Frederick JT, Prohaska T, Satariano
5. Hughes SL, Williams B, Molina LC, Bayles doi:10.1177/0733464806290934 WA, Dornberg-Lee S, Fisher R, et al. Recommenda-
C, Bryant LL, Harris JR, et al. Characteris- 18. Belza B, Snyder S, Thompson M, LoGerfo J. From tions for treating depression in community-based
tics of physical activity programs for older research to practice: EnhanceFitness ®, an innov- older adults. Am J Prev Med (2007) 33(3):175–81.
adults: results of a multisite survey. The ative community-based senior exercise program. doi:10.1016/j.amepre.2007.04.034
Gerontologist (2005) 45(5):667–75. doi:10.1093/ Top Geriatr Rehabil (2010) 26(4):299–309. doi:10. 31. Snowden M, Steinman L, Frederick J. Treating
geront/45.5.667 1097/TGR.0b013e3181fee69e depression in older adults: challenges to imple-
6. Anderson L, Logsdon RG, Hochhalter AK, Sharkey 19. Smith ML, Belza B, Altpeter M, Ahn S, Dick- menting the recommendations of an expert panel.
JR. Introduction to the special issue on promot- erson JB, Ory M. Disseminating an evidence- Prev Chronic Dis (2008) 5(1). Available from: http:
ing cognitive health in diverse populations of based disease self-management program for older //www.cdc.gov/pcd/issues/2008/jan/07_0154.htm
older adults. The Gerontologist (2009) 49(S1):S1–2. Americans: implications for diversifying partici- 32. Snowden M, Steinman L, Frederick J, Wilson
doi:10.1093/geront/gnp073 pant reach through delivery site adoption. In: Mad- N. Screening for depression in older adults:
7. Hughes SL, Leith KH, Marquez DX, Moni G, dock J, editor. Public Health – Social and Behav- recommended instruments and considerations for
Nguyen HQ, Desai P, et al. Physical activity and ioral Health. Rijeka: InTech – Open Access Pub- community-based practice. Clin Geriatr (2009).
older adults: expert consensus for a new research lisher (2012). p. 385–404. doi:10.5772/36728 Available from: http://www.consultant360.com/
agenda (an Editor’s Choice article). The Geron- 20. Hughes SL, Seymour RB, Campbell RT, Huber G, articles/screening-depression-older-adults-
tologist (2011) 51(6):822–32. doi:10.1093/geront/ Pollak N, Sharma L, et al. Long term impact of recommended-instruments-and-considerations-
gnr106 Fit and Strong! on older adults with osteoarthri- community
8. Satariano WA, Guralnik JM, Jackson RJ, Marottoli tis. The Gerontologist (2006) 46(6):801–14. doi:10.
RA, Phelan EA, Prohaska TR. Mobility and aging: 1093/geront/46.6.801
new directions for public health action. Am J Public 21. Etkin CD, Prohaska TR, Harris BA, Latham N, Conflict of Interest Statement: The authors declare
Health (2012) 102(8):1508–15. doi:10.2105/AJPH. Jette A. Feasibility of implementing the Strong for that the research was conducted in the absence of any
2011.300631 Life program in community settings. The Geron- commercial or financial relationships that could be
9. Sharkey JR, Locher J, Sahyoun N, Wilcox S. Nutri- tologist (2006) 46(2):284–92. doi:10.1093/geront/ construed as a potential conflict of interest.
tion and aging: nutritional health inequity (editor- 46.2.284
ial). J Aging Res (2012) 2012:164106. doi:10.1155/ 22. Steinman L, Cristofalo M, Snowden M. Implemen- This paper is included in the Research Topic, “Evidence-
2012/164106 tation of an evidence-based depression care man- Based Programming for Older Adults.” This Research
10. Anderson LA, Day KL, Vandenberg AE. Using a agement program (PEARLS): perspectives from Topic received partial funding from multiple government
concept map as a tool for strategic planning: the staff and former clients. Prev Chronic Dis (2012) and private organizations/agencies; however, the views,
healthy brain initiative. Prev Chronic Dis (2011) 9:110250. doi:10.5888/pcd9.110250 findings, and conclusions in these articles are those of the
8(5):A117. 23. Steinman L, Hammerback K, Snowden M. It could authors and do not necessarily represent the official posi-
11. Brownson RC, Ballew P, Dieffenderfer B, Haire- be a pearl to you: exploring recruitment and reten- tion of these organizations/agencies. All papers published
Joshu D, Heath GW, Kreuter MW, et al. Evidence- tion of the Program to Encourage Active, Reward- in the Research Topic received peer review from members
based interventions to promote physical activity: ing Lives (PEARLS) with hard-to-reach popu- of the Frontiers in Public Health (Public Health Edu-
what contributes to dissemination by state health lations. The Gerontologist (2013). doi:10.1093/ cation and Promotion section) panel of Review Editors.
departments. Am J Prev Med (2007) 33(1):S66–78. geront/gnt137 Because this Research Topic represents work closely asso-
doi:10.1016/j.amepre.2007.03.011 24. Bryant LL, Altpeter M, Whitelaw NA. Evalua- ciated with a nationwide evidence-based movement in
12. Frank J, Coviak C, Healy T, Belza B, Casado B. tion of health promotion programs for older the US, many of the authors and/or Review Editors may
Addressing fidelity in evidence-based health pro- adults: an introduction. J Appl Gerontol (2006) have worked together previously in some fashion. Review
motion programs for older adults. J Appl Geron- 25(3):197–213. doi:10.1177/0733464806288562 Editors were purposively selected based on their expertise
tol (2008) 27(4):4–34. doi:10.1177/109019811454 25. Smith ML, Ory MG, Belza B, Altpeter M. Per- with evaluation and/or evidence-based programming
3007 sonal and delivery site characteristics associated for older adults. Review Editors were independent of
13. Hughes SL, Seymour RB, Campbell RT, Whitelaw with intervention dosage in and evidence-based named authors on any given article published in this
N, Bazzarre T. Best-practice physical activity fall risk reduction program for older adults. Transl volume.
programs for older adults: findings from the Behav Med (2012) 2:188–98. doi:10.1007/s13142-
national impact study. Am J Public Health (2009) 012-0133-8 Received: 15 June 2014; accepted: 12 November 2014;
99(2):362–8. doi:10.2105/AJPH.2007.131466 26. Belza B. PRC-HAN Physical Activity Con- published online: 27 April 2015.
14. Logsdon RG, McCurry SM, Pike KC, Teri L. Mak- ference Planning Workgroup. Moving Ahead: Citation: Belza B, Altpeter M, Hooker SP and Moni
ing physical activity accessible to older adults with Strategies and Tools to Plan, Conduct, and G (2015) The CDC Healthy Aging Research Net-
memory loss: a feasibility study. The Gerontolo- Maintain Effective Community-Based Physical work: advancing science toward action and policy for
gist (2009) 49(Suppl 1):S94–9. doi:10.1093/geront/ Activity Programs for Older Adults. Seattle, WA: the evidence-based health promotion movement. Front.
gnp082 (2007). Available from: http://depts.washington. Public Health 2:261. doi: 10.3389/fpubh.2014.00261
15. Glasgow RE, Vogt TM, Boles SM. Evaluating the edu/hprc/older-adults-primer/monograph.pdf This article was submitted to Public Health Education
public health impact of health promotion inter- 27. Belza B, Glasgow R, Toobert D. RE-AIM for Pro- and Promotion, a section of the journal Frontiers in
ventions: the RE-AIM framework. Am J Public gram Planning: Overview and Applications. Wash- Public Health.
Health (1999) 89(9):1322–7. doi:10.2105/AJPH. ington, DC: National Council on Aging (2007). Copyright © 2015 Belza, Altpeter, Hooker and Moni.
89.9.1322 28. Guide to Community Preventive Services. Improv- This is an open-access article distributed under the terms
16. Harris JR, Cheadle A, Hannon PA, Forehand M, ing Mental Health and Addressing Mental Ill- of the Creative Commons Attribution License (CC BY).
Lichiello P, Mahoney E, et al. A framework for ness. (2014). Available from: http://www.the The use, distribution or reproduction in other forums is
disseminating evidence-based health promotion communityguide.org/mentalhealth/index.html permitted, provided the original author(s) or licensor are
practices. Prev Chronic Dis (2012) 9:110081. doi: 29. Frederick JT, Steinman LE, Prohaska T, Satari- credited and that the original publication in this journal
10.5888/pcd9.110081 ano WA, Bruce M, Bryant L, et al. Community- is cited, in accordance with accepted academic practice.
17. Belza B, Shumway-Cook A, Phelan E, Williams based treatment of late life depression an expert No use, distribution or reproduction is permitted which
B, Snyder S, LoGerfo J. The effects of a panel-informed literature review. Am J Prev Med does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 261 | 52
 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00234

Chronic disease self-management support: public health

perspectives
Teresa J. Brady 1 *, Lynda A. Anderson 2,3 and Rosemarie Kobau 4
1
Arthritis Program, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta, GA, USA
2
Healthy Aging Program, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta, GA, USA
3
Rollins School of Public Health, Emory University, Atlanta, GA, USA
4
Epilepsy Program, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta, GA, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: self-management support, evidence-based interventions public health, life course, prevention research

A PUBLIC HEALTH PRIORITY importance of self-management support community, environmental, and systems
The Centers for Disease Control and Pre- in its framework for addressing multiple levels resources (12). This definition
vention (CDC) has a longstanding com- chronic conditions (MCC). One of the also embraces a life course perspective
mitment to developing and promoting four goals of the framework is to “max- that attends to individual autonomy and
evidence-based strategies to prevent or imize the use of proven self-care manage- decision-making as well as role changes and
delay disease and disability (1, 2). Signif- ment and other services by individuals with other adaptations to life events (13).
icant among these strategies is support MCC” (10). Self-management support takes many
for self-management of chronic diseases. Chronic disease self-management sup- forms. It includes interventions such as
About one-half of all U.S. adults have at port occurs at the intersection of public the Chronic Disease Self-Management Pro-
least one chronic condition (3) and over health, clinical healthcare delivery, social gram (CDSMP) (14) and the falls pre-
two-thirds of Medicare beneficiaries aged services, aging services networks, and other vention programs featured in this special
65 years or older have two or more chronic community resources. In this commen- issue (15). It also includes supportive inter-
conditions (4). Given that the risk of devel- tary, we provide a public health perspec- actions between healthcare providers and
oping a chronic disease increases with tive on self-management support, iden- patients, proactive follow-up, and social
advancing age (5), the dramatic aging of tify examples of CDC investment in self- and physical environments that support
the U.S. population underscores the impor- management support activities, and dis- healthy behaviors such as having safe places
tance of chronic disease self-management cuss potential future directions. These to exercise, access to healthy foods, and
supports. Further, effective self manage- examples are provided to illustrate the social norms that combat stigma, promote
ment of chronic conditions is essential to breadth of CDC’s work in this area and are social participation, and support self-care
achieving a state of health, which is pro- not designed to serve as comprehensive list behaviors (9).
posed to reflect “the ability to adapt and to of CDC’s investment in self-management Self-management support interven-
self manage” (6). support. tions are provided in a variety of formats
An effective approach to improve pop- (e.g., one-to-one, small groups, telephone,
ulation health requires a strong focus on AN INTERNATIONAL FOUNDATION FOR online/mobile, self-study); and in a variety
self-management. CDC’s National Center UNDERSTANDING of settings (e.g., home, healthcare, work-
for Chronic Disease Prevention and Health Consistent with a public health perspec- site, community) (9, 12). Although the
Promotion includes among its four pri- tive, we advance an expanded definition of form and formats vary, the goal of self-
orities efforts to help ensure that “com- self-management support from the Inter- management support is consistent: to help
munities support and clinics refer patients national Framework for Chronic Condi- individuals and their personal support sys-
to programs that improve management of tion Self-Management Support. This defi- tem acquire and maintain the knowledge,
chronic conditions” (7). Self-management nition describes self-management support skills, and confidence to do what they need
(e.g., what individuals and families do on as a grouping of policies, programs, ser- to do to live as well as possible with their
a daily basis to feel better and pursue the vices, and structures that extend across chronic condition(s).
life they desire) (8) and self-management healthcare, social sectors, and communities
support (e.g., actions taken by others to to support and improve the way individuals ADVANCING THE STUDY AND
support individual self-management) (9) manage their chronic conditions (11). The APPLICATION OF SELF-MANAGEMENT
are critical strategies in meeting this pri- definition frames self-management sup- SUPPORT
ority objective. The U.S. Department of port within a social-ecological perspective The International Framework for Chronic
Health and Human Services recognized the underscoring individual, interpersonal, Condition Self-Management Support

www.frontiersin.org April 2015 | Volume 2 | Article 234 | 53

Brady et al. Self-management support: public health perspectives

Table 1 | Self-management support strategic directions: select CDC program examples.

Strategic direction Tactic CDC program examplesa

Involve consumers Community-based CDC’s prevention research centers (PRC) use community-based participatory research methods
participatory research as a foundation for their research (16). Evidence-based intervention programs such as Enhance
Fitness (17) to increase physical activity among older adults and PEARLs (18) to screen and
treat depression among older adults were developed at PRCs (http://www.cdc.gov/prc).

Audience research CDC collaborated with the Arthritis Foundation to support qualitative and quantitative market
research that provided insights into the types of services people with arthritis want to support
their self-management efforts (Listening to Consumers: What do People With Arthritis Want? A
Focus Group Report to: The Centers for Disease Control and Prevention (Arthritis Program) and
the Arthritis Foundation. Unpublished report by Fleishman Hillard, 2006; Receptivity to Existing
and Potential Programs and Services for People with Arthritis. A Report to the Arthritis
Foundation. Unpublished report by Fleishman Hillard, 2007).

CDC conducted audience research with people with various chronic conditions and determined
that non-disease specific, non-intervention specific messaging to increase the visibility of
self-management education would resonate with consumers and motivate them to seek more
information on specific interventions (Audience Research to Determine the Feasibility of
Developing a Marketing Campaign to Increase Visibility of Self-Management Education.
Unpublished report submitted to CDC by FHI 360, 2014). CDC is developing this type of broad
awareness campaign.

Expand reach and Tele-health Emory University developed Project UPLIFT, an effective tele-health intervention delivered by
range of services phone and Internet that helps adults with epilepsy and comorbid depression reduce their
depressive symptoms, and improve some well-being domains (19).

Self-study interventions The University of North Carolina and Stanford University developed and evaluated The Arthritis
Toolkit that provides the content of the small group-delivered Arthritis Self-Management
Program (ASMP) in a mail-delivered, self-study format (20). Currently, CDC is funding Stanford
University to develop a self-study version of the Chronic Disease Self-Management Program
(CDSMP).

Online interventions Stanford Universityb developed an online (virtual group) version of the ASMP (21), and national
disseminationa is being pilot-tested by the Arthritis Foundation.a (http://www.arthritistoday.org/
arthritis-self-management-program/).

The American Cancer Society pilot-tested a cancer-specific online version of the CDSMP, titled
Cancer: Thriving and Surviving, developed at Stanford University (https://cancer.selfmanage.org/
survivor/hl/hlMain).

Emory University PRC developed and tested, WebEase (Epilepsy Awareness Support and
Education), an online program that is available on the national Epilepsy Foundation web site
(22).

Advance evidence Meta-analysis CDC conducted meta-analyses of 24 ASMP and 23 CDSMP studies that documented robust
improvements in health outcomes and health behaviors across multiple studies (23, 24).

Intervention research PRCs, in collaboration with other universities, conducted effectiveness studies that
substantiated evidence-based community interventions such as the Arthritis Foundation
Exercise Program, Walk with Ease, Enhance Fitness, and First Steps to Active Health
(http://www.cdc.gov/arthritis/funded_science/completed/index.htm).

Improve Clinical-decision The University of Texas (Houston) PRC developed and tested MINDSET (Management
effectiveness and support tools Information & Decision Support Epilepsy Tool), a tablet-based tool for inputting data on (1)
appropriateness of seizures (e.g., history, management); (2) medicine (e.g., barriers, side effects), and (3) lifestyle
services (e.g., social support). The tool is designed to enhance patient-provider communication and
action planning to sustain or improve epilepsy self-management behaviors (25).

(Continued)

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 234 | 54
Brady et al. Self-management support: public health perspectives

Table 1 | Continued

Strategic direction Tactic CDC program examplesa

Comparative The University of North Carolina PRC conducted a study of CDSMP and ASMP among people
effectiveness studies with arthritis that documented the equivalence of these two interventions for people with
arthritis (26).

The University of Pittsburg PRC conducted a study of preventing falls among older adults that
compared 3 strategies: usual care, an education program, or an education-plus-exercise
program (http://www.caph.pitt.edu/wps/docs/falls/FP_AlbertfallsCDCpresentation9-25-11.pdf).

Strengthen Community-clinical CDC’s Heart Disease and Stroke Prevention Program is collaborating with the Vermont Blue
inter-sector linkages linkages Print for Health to explore the use of community-health workers as part of the primary care
team to help assess patients’ needs and coordinate community-based support services (27).

Linking public health The Healthy Aging Research Network, a thematic network of the PRC Program, developed and
and aging services implemented a national research and dissemination agenda related to the public health aspects
networks of healthy aging (28).

Linking mental health The Managing Epilepsy Well Network, a thematic network of the PRC Program, includes
and public health interdisciplinary teams of researchers who collaborate across mental health and epilepsy
sectors to develop and implement evidence-based self-management programs that target both
physical and mental health needs of people with epilepsy (29).

Linking multiple The Osteoarthritis Action Alliance, under the auspices of the Arthritis Foundation, provides a
stakeholders forum for multiple organizations to work collaboratively to advance the osteoarthritis public
health agenda including increasing physical activity and fostering self-management education
(http://www.oaaction.org).

Foster multi-sector Strategic frameworks The U.S. Department of Health and Human Services developed the document, The Multiple
commitment and Chronic Conditions: A Strategic Framework that outlines national strategies for improving
accountability health and quality of life for individuals with multiple chronic conditions (MCC), and cites the
use of proven self-care management and other services by individuals with MCC as one of its’
four strategic goals (10).

CDC staff participated in the development of the International Framework for Chronic Condition
Self-Management Support that highlights priority strategic directions to advance the research,
policy, and practice of self-management support (11).

National objectives Healthy People 2020 includes objectives to increase participation in self-management
education among select chronic disease populations including people with arthritis and diabetes
(http://www.healthypeople.gov/2020/topicsobjectives2020/objectiveslist.aspx?topicId=3;
http://www.healthypeople.gov/2020/topicsobjectives2020/objectiveslist.aspx?topicId=8).

Convening stakeholders CDC collaborated with the Arthritis Foundation to convene a broad stakeholder group that
developed Environmental and Policy Strategies to Increase Physical Activity among People with
Arthritis; this document recommends strategies for action in 6 sectors including community,
business, healthcare, transportation, parks and recreation, and mass media (http://www.
arthritis.org/files/documents/OA_Physical_Activity_Rpt_508_v1_TAG508.pdf).

Build infrastructure Professional training The Managing Epilepsy Well Network provided professional training that helped providers better
opportunities understand self-management strategies and how to implement at least three evidence-based
self-management programs (http://web1.sph.emory.edu/ManagingEpilepsyWell/index.php).

Capacity building All 50 states receive CDC funding to support the delivery of diabetes self-management
education through consolidated chronic disease funding (http://www.cdc.gov/chronicdisease/
about/statepubhealthactions-prevcd.htm).

Individual CDC programs focused on asthma, arthritis, diabetes, and heart disease have
supported at least 40 state health departments to disseminate CDSMP.

(Continued)

www.frontiersin.org April 2015 | Volume 2 | Article 234 | 55

Brady et al. Self-management support: public health perspectives

Table 1 | Continued

Strategic direction Tactic CDC program examplesa

The National Association of County and City Health Officials, the National Recreation and Parks
Association, and the Y-USA are testing their ability to serve as national delivery systems in
disseminating self-management support interventions. http://naccho.org/topics/HPDP/chronic
disease/cdsmp.cfm; http://www.nrpa.org/Grants-and-Partners/Recreation-and-Health/Arthritis-
Interventions; http://c.ymcdn.com/sites/www.chronicdisease.org/resource/resmgr/Arthritis_
Monthly_Reports/030513_ACHandoutDisseminatio.pdf).

Effective dissemination CDC collaborated with the National Association of Chronic Disease Directors to support a
strategies multi-site evaluation of state health department approaches to dissemination of
evidence-based interventions. This evaluation documented the effectiveness of working with
multi-site delivery systems, embedding interventions into routine operations, collaborating
with other chronic disease programs and prioritizing the expansion of reach (Strategic
approaches to expanding the reach of evidence-based interventions: results of a multi-state
evaluation) (Unpublished report submitted to the National Association of Chronic Disease
Directors by Westat, 2012.) (http://www.cdc.gov/arthritis/publications/reports.htm).

a
CDC funded these efforts unless otherwise noted.
b
CDC provided partial funding.

identifies seven key strategic directions designed to develop and test clinical- If self-management support interven-
to move self-management support for- decision support tools. In terms of efforts tions are to achieve their potential for
ward in research, policy, and practice at the to strengthen inter-sector linkages, CDC public health impact, they need to be inte-
local, regional, state, and national levels. supports community-clinical collabora- grated into comprehensive chronic dis-
These strategic directions are to involve tions and makes linkages across public ease management strategies at the national,
consumers, expand the reach and range of health sectors. CDC also helps to foster state, and local levels, and across sectors.
services, advance evidence, improve effec- multi-sector commitment and accountabil- Given the large and diverse population
tiveness and appropriateness of services, ity through the development of new frame- of people living with chronic conditions,
strengthen inter-sector linkages, foster works and guidelines. Finally, CDC invests engagement of multiple organizations
multi-sector commitment and account- in building infrastructure to deliver self- across various sectors is required to reach
ability, and build infrastructure. Using management support intervention pro- those in need. Ideally, self-management
this organizing structure (11), in Table 1, grams at the national, state, and local levels support will become an integral element
we highlight a few select but illustrative systems initiatives. of clinical care standards of care (30), part
examples of CDC’s contributions to the of the routine menu of services offered
Framework’s seven strategic directions. SUSTAINING SELF-MANAGEMENT by a variety of community agencies, and
Through funded research and other SUPPORT: GAPS AND OPPORTUNITIES an essential component of community
mechanisms, CDC and its partners have The need to advance efforts in self- chronic disease control efforts. Finally,
employed a variety of strategies to involve management support is well recognized sustaining self-management support will
consumers by using applied community- in the public health arena. However, chal- require the infrastructure as well as multi-
based participatory strategies in develop- lenges remain and several research ques- sectoral resources to reach people where
ing evidence-based programs and audi- tions are yet unanswered. Such questions they live, learn, work, and engage with their
ence research. To expand the range and include how to identify the essential ele- family and community. Creative financ-
reach of services, CDC supports the devel- ments of an intervention, how to best tar- ing mechanisms will need to be developed
opment, evaluation, and dissemination of get effective interventions to specific audi- or expanded to ensure wide availability of
a variety of small group, tele-health, self- ences, and how to determine the effect of evidence-based self-management support.
study, and online self-management sup- self-management support on critical public The Centers for Disease Control and
port tools. To advance evidence, CDC inves- health outcomes and biometric measures Prevention is supporting a wide variety of
tigators conduct systematic reviews of the such as hemoglobin A1c and blood pres- self-management support activities across
literature and CDC funds applied preven- sure. Additional comparative effectiveness multiple strategic directions. CDC sup-
tion research to establish or strengthen and cost effectiveness research studies ported activities exemplify a comprehen-
the evidence-base of programs and poli- of self-management support interventions sive view of self-management support that
cies. To improve the effectiveness and appro- are necessary. Importantly, selected papers encompasses both health-enhancing indi-
priateness of services, CDC supports com- in this special issue will help address these vidual behaviors and physical and social
parative effectiveness studies and research issues. environmental contexts that influence

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 234 | 56
Brady et al. Self-management support: public health perspectives

self-management behaviors. To advance health outcomes and quality of life. Public Health Centers for Disease Control and Prevention
self-management support, several impor- Rep (2011) 126(4):460–71. (US) (2011). Available from: http://www.cdc.gov/
11. Brady TJ, Mills S, Sargious P, Ziabakhsh S. An arthritis/docs/ASMP-executive-summary.pdf
tant areas of research need to be conducted,
international framework for chronic condition self 24. Brady TJ, Murphy LM, O’Colmain B, Beauchesne
broad-based organizational engagement management support: results from an interna- D, Daniels B, Greenberg M, et al. A meta-analysis
needs to occur, and delivery capacity infra- tional electronic consultation process [abstract]. of health status, health behaviors and health care
structure and financing mechanisms need Arthritis Rheum (2012) S10:S1011–2. doi:10.1002/ utilization outcomes of the chronic disease self-
to be established or expanded. Despite art.38216 management program. Prev Chronic Dis (2013)
12. Fisher EB, Brownson CA, O’Toole ML, Shetty G, 10:120112. doi:10.5888/pcd10.120112
these challenges, it remains essential to cre- Anwuri VV, Glasgow RE. Ecological approaches 25. Shegog R, Begley CE, Harding A, Dubinsky S,
ate self-management support services and to self-management: the case of diabetes. Am Goldmsith C, Omotola H, et al. Description and
environmental supports that allow people J Public Health (2005) 95:1523–35. doi:10.2105/ feasibility of MINDSET: a clinic decision aid for
to live well with their chronic condition. AJPH.2005.066084 epilepsy self-management. Epilepsy Behav (2013)
13. Elder GH, Johnson MK, Crosnoe R. The emer- 29:527–36. doi:10.1016/j.yebeh.2013.09.023
gence and development of life course theory. In: 26. Goeppinger J, Armstrong B, Schwartz T, Ensley D,
ACKNOWLEDGMENTS Mortimer JT, Shanahan MJ, editors. Handbook of Brady T. Self-management education for persons
The findings and conclusions in this com- the Life Course. New York, NY: Kluwer Acade- with arthritis: managing co-morbidities and elim-
mentary are those of the authors and do mic/Plenum (2003). p. 3–22. inating health disparities. Arthritis Rheum (2007)
not necessarily represent the official posi- 14. Lorig KR, Sobel DS, Stewart AL, Brown BW, Ban- 57(6):1081–8. doi:10.1002/art.22896
tion of the Centers for Disease Control and dura A, Ritter P, et al. Evidence suggesting that 27. Vermont Blueprint for Health: 2011 Annual
a chronic disease self-management program can Report [Internet]. Williston, VT: Department of
Prevention. improve health status while reducing hospitaliza- Vermont Health Access (2012). Available from:
tion: a randomized trial. Med Care (1999) 37:5–14. http://hcr.vermont.gov/sites/hcr/files/Blueprint%
REFERENCES doi:10.1097/00005650-199901000-00003 20Annual%20Report%20Final%2001%2026%
1. Briss PA, Zaza S, Pappaioanou M, Fielding J, 15. Stevens JA. A CDC Compendium of Effective 2012%20_Final_.pdf
Wright-De Agüero L, Truman BI, et al. Devel- Fall Interventions: What Works for Community- 28. Wilcox S, Altpeter M, Anderson LA, Belza B, Bryant
oping an evidence-based guide to community Dwelling Older Adults. 2nd ed. Atlanta, GA: Centers L, Jones DL, et al. The healthy aging research net-
preventive services-methods. Am J Prev Med for Disease Control and Prevention (US), National work: resources for building capacity for public
(2000) 18(1S):35–43. doi:10.1016/S0749-3797(99) Center for Injury Prevention and Control (2010). health and aging practice. Am J Health Promot
00119-1 16. Faridi Z, Grunbaum JA, Gray BS, Franks A, Simoes (2013) 28(1):2–6. doi:10.4278/ajhp.121116-CIT-
2. Kohatsu ND, Robinson JG, Torner JC. Evidence- E. Community-based participatory research: nec- 564
based public health: an evolving concept. Am J Prev essary next steps. Prev Chronic Dis (2007) 4: 29. CDC Prevention Research Centers. Managing
Med (2004) 27(5):417–21. doi:10.1016/j.amepre. A70. Available from: http://www.cdc.gov/pcd/ Epilepsy Well Network: Putting Collective Wisdom to
2004.07.019 issues/2007/jul/06_0182.htm Work for People with Epilepsy. Atlanta, GA: Centers
3. Anderson G. Chronic Care: Making the Case 17. Belza B, Snyder S, Thompson M, LoGerfo J. From for Disease Control and Prevention (US), National
for Ongoing Care. Princeton, NJ: Robert Wood research to practice: EnhanceFitness, an innovative Center for Chronic Disease Prevention and Health
Johnson Foundation (2010). Available from: community-based senior exercise program. Topics Promotion (2013). Available from: www.cdc.gov/
www.rwjf.org/content/dam/farm/reports/reports/ Geriat Rehab (2010) 26(4):299–309. epilepsy/pdfs/mew-booklet-tagged-508.pdf
2010/rwjf54583 18. Ciechanowski P, Wagner E, Schmaling K, Schwartz 30. National Committee for Quality Assurance. NCQA
4. Lochner KA, Cox CS. Prevalence of multiple S, Williams B, Diehr P, et al. Community- Patient-Centered Medical Home: A New Model
chronic conditions among Medicare beneficia- integrated home-based depression treatment in of Care Delivery, Patient-Centered Medical Homes
ries, United States, 2010. Prev Chronic Dis (2013) older adults: a randomized controlled trial. JAMA Enhance Primary Care Practices. Washington, DC:
10:120137. doi:10.5888/pcd10.120137 (2004) 291(13):1569–77. doi:10.1001/jama.291. National Committee for Quality Assurance (2014).
5. Ward BW, Schiller JS. Prevalence of multiple 13.1569 Available from: www.ncqa.org/Portals/0/PCMH%
chronic conditions among US adults: estimates 19. Thompson NJ, Walker ER, Obolensky N, Win- 20brochure-web.pdf
from the National Health Interview Survey, 2010. ning A, Barmon C, DiIorio C, et al. Distance deliv-
Prev Chronic Dis (2013) 10:120203. doi:10.5888/ ery of mindfulness-based cognitive therapy for
pcd10.120203 depression: project UPLIFT. Epilepsy Behav (2010) Conflict of Interest Statement: The authors declare
6. Huber M, Knottnerous JA, Green L, van der Horst 19(3):247–54. doi:10.1016/j.yebeh.2010.07.031 that the research was conducted in the absence of any
H, Leonard B, Lorig K, et al. How should we define 20. Goeppinger J, Lorig KR, Ritter PL, Mutatkar S, commercial or financial relationships that could be
health? BMJ (2011) 343:d4163. doi:10.1136/bmj. Villa F, Gizlice Z. Mail-delivered arthritis self- construed as a potential conflict of interest.
d4163 management tool kit: a randomized trial and
7. Centers for Disease Control and Prevention. 24/7: longitudinal follow-up. Arthritis Rheum (2009) This paper is included in the Research Topic, “Evidence-
Saving Lives. Protecting People.TM National Cen- 61(7):867–75. doi:10.1002/art.24587 Based Programming for Older Adults.” This Research
ter for Chronic Disease Prevention and Health 21. Lorig KR, Ritter PL, Laurent D, Plant K. The Topic received partial funding from multiple government
Promotion. Atlanta, GA: Centers for Disease internet-based arthritis self-management pro- and private organizations/agencies; however, the views,
Control and Prevention (US) (2013). Avail- gram: a one-year randomized trial for patients with findings, and conclusions in these articles are those of the
able from: www.cdc.gov/chronicdisease/about/ arthritis or fibromyalgia. Arthritis Rheum (2008) authors and do not necessarily represent the official posi-
public-health-approach.htm 59(7):1009–17. doi:10.1002/art.23817 tion of these organizations/agencies. All papers published
8. Brady TJ, Sniezek JE, Conn D. Enhancing patient 22. DiIorio C, Bamps Y, Escoffery C, Reisinger-Walker in the Research Topic received peer review from members
self-management in clinical practice. Bull Rheum E. Results of a randomized controlled trial: evaluat- of the Frontiers in Public Health (Public Health Edu-
Dis (2000) 49(9):1–4. ing WebEase, an online epilepsy self-management cation and Promotion section) panel of Review Editors.
9. Brady TJ. Strategies to support self-management program. Epilepsy Behav (2011) 22(3):469–74. doi: Because this Research Topic represents work closely asso-
in osteoarthritis. Am J Nurs (2012) 112:S54–60. 10.1016/j.yebeh.2011.07.030 ciated with a nationwide evidence-based movement in
doi:10.1097/01.NAJ.0000412653.56291.ab 23. Centers for Disease Control and Prevention. Sort- the US, many of the authors and/or Review Editors may
10. Parekh AK, Goodman RA, Gordon C, Koh HK, ing through the Evidence of the Arthritis Self- have worked together previously in some fashion. Review
The HHS Interagency Workgroup on Multiple Management Program and the Chronic Disease Editors were purposively selected based on their expertise
Chronic Conditions. Managing multiple chronic Self-Management Program: Executive Summary of with evaluation and/or evidence-based programming
conditions: a strategic framework for improving the ASMP/CDSMP Meta-Analyses. Atlanta, GA: for older adults. Review Editors were independent of

www.frontiersin.org April 2015 | Volume 2 | Article 234 | 57

Brady et al. Self-management support: public health perspectives

named authors on any given article published in this health perspectives. Front. Public Health 2:234. doi: Creative Commons Attribution License (CC BY). The
volume. 10.3389/fpubh.2014.00234 use, distribution or reproduction in other forums is per-
This article was submitted to Public Health Education mitted, provided the original author(s) or licensor are
Received: 16 June 2014; accepted: 28 October 2014; and Promotion, a section of the journal Frontiers in credited and that the original publication in this journal
published online: 27 April 2015. Public Health. is cited, in accordance with accepted academic practice.
Citation: Brady TJ, Anderson LA and Kobau R Copyright © 2015 Brady, Anderson and Kobau. This is No use, distribution or reproduction is permitted which
(2015) Chronic disease self-management support: public an open-access article distributed under the terms of the does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 234 | 58
 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00136

Evidence-based leadership council – a national

collaborative
Margaret Haynes 1 , Susan Hughes 2 , Kate Lorig 3 , June Simmons 4 , Susan J. Snyder 5 *, Lesley Steinman 6 ,
Nancy Wilson 7 , Roseanne DiStefano 8 , Jennifer Raymond 9 , Stephanie FallCreek 10 , Martha B. Pelaez 11 and
Don Smith 12
1
Elder Care Services, Partnership for Healthy Aging, MaineHealth, Portland, ME, USA
2
University of Illinois at Chicago School of Public Health, Chicago, IL, USA
3
Stanford University School of Medicine, Palo Alto, CA, USA
4
Partners in Care Foundation, San Fernando, CA, USA
5
Project Enhance, Senior Services, Seattle, WA, USA
6
University of Washington Health Promotion Research Center, Seattle, WA, USA
7
Baylor College of Medicine, Houston, TX, USA
8
Elder Services of the Merrimack Valley, Lawrence, MA, USA
9
Healthy Living Center of Excellence, Boston, MA, USA
10
Fairhill Partners, Cleveland, OH, USA
11
Healthy Aging Regional Collaborative, Health Foundation of South Florida, Miami, FL, USA
12
Area Agency on Aging of Tarrant County, Fort Worth, TX, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: evidence-based programs, health promotion, aging, self-management

Over many years, a number of aca- The EBLC is currently a group of for highest level of evidence (2). In addi-
demic/community partnerships have 11 individuals representing a total of tion to the ACL, the Centers for Disease
worked independently to develop, evaluate, 19 evidence-based programs (Chronic Control and Prevention (CDC) Arthri-
and bring to scale participant-centered, Disease Self-Management suite of Pro- tis Program (3), Substance Abuse and
evidence-based self-management, and grams, Matter of Balance, Enhance Fit- Mental Health Services Administration’s
health promotion programs offered in ness, Enhance Wellness, Healthy IDEAS, (SAMHSA) National Registry of Evidence-
community settings for older Americans. PEARLS, Fit & Strong!, HomeMeds, Based Programs (4), and the Agency for
Many of the programs developed by these Healthy MOVES) as well as four lead- Healthcare Research and Quality Innova-
partnerships have since become critical ers from organizations providing multiple tions Exchange (5) recommend these pro-
pieces of the infrastructure that sup- evidence-based programs (Health Foun- grams and find them to be the strongest of
ports older adults with chronic health dation of South Florida, Tarrant County evidence-based programs (6–14). The pro-
conditions. Indeed, community-based Area Agency on Aging, Elder Services grams represented by the EBLC are utilized
self-management support is an integral of the Merrimack Valley/Hebrew Senior by more than 1,700 agencies in the United
component of the Chronic Care Model Life, Fairhill Partners). EBLC members are States with nearly 400 agencies using more
(1) illustrated in Figure 1. This model employed by community-based organiza- than one program.
presents elements that can improve health tions, foundations, healthcare systems, uni- Together, the council represents more
outcomes for people with chronic condi- versities, and governmental entities and than 200 combined years of experience
tions, highlighting the need for connec- have been directly involved for many years in developing, evaluating, scaling, imple-
tions between healthcare and community in the development, evaluation, and scal- menting, and sustaining evidence-based
resources, integrating patient-centered, ing of their individual programs as well self-management programs. All of the pro-
evidence-based services that empower as implementation through community- grams have proven effectiveness in pub-
patients. And while these programs have based organizations. The individual pro- lished randomized controlled trial research
succeeded in finding their place in this gram developers met informally for several and all programs have been brought to
system working independently so far, years and in 2012 formed the EBLC. Over scale. The mission of the EBLC is to
the growth and maturation of the pro- the past year, community-based organi- increase delivery of evidence-based pro-
grams, combined with the changing zation leaders responsible for implement- grams that improve the health and well-
environment of healthcare, have prompted ing multiple evidence-based programs being of diverse populations. The EBLC is
new collaboration among the organiza- were asked to join and be part of the committed to the following values:
tions that manage and disseminate these council.
programs, specifically, the creation of All the programs represented by EBLC • Person Centeredness – individuals are
the Evidence-Based Leadership Council program developers meet the Administra- actively involved in programs and mak-
(EBLC). tion for Community Living’s (ACL) criteria ing a difference.

www.frontiersin.org April 2015 | Volume 2 | Article 136 | 59

Haynes et al. EBLC – a national collaborative

maintained data on organizations pro-

viding EBP trainings and workshops.
• Offer the potential for significant gains
in efficiency for program owners, who
can combine information about organi-
zations providing or interested in pro-
viding EBPs and can eliminate dupli-
cation of communications and com-
mon workflow processes (e.g., licensing,
training registration).
• Offer the potential for significant gains
in user-friendliness for organizations
providing or interested in providing
EBPs, by providing a single gateway
to the programs (including the com-
mon website) through which adoption
research, readiness assessment, licensing,
and training can be handled for one or
multiple programs at a time.

FIGURE 1 | The chronic care model. The EBLC’s vision for the future is an
ever increasing number of adults engaged
in evidence-based programs that inform,
• Effectiveness – evidence-based programs the new demands. Each program devel- activate, and empower them to improve
focus on outcomes/results. oper has experienced challenges to keep their health and maintain independence.
• Collaboration – multi-sector, multi- up with increasing expectations for plan- These programs will be embedded in a
organizational and interdisciplinary ning, training, and technical support, while permanent, sustainable infrastructure –a
(belief that health is achieved in the working within the confines of their parent national network supported by the EBLC’s
community, close to home and through organization and maintaining affordabil- technical assistance in implementation and
broad-based collaborations). ity for community-based organizations. dissemination, training, marketing, licens-
• Equity and access – social justice, respect Community-based organizations have had ing, and evaluation. Bringing years of
of diversity. their own set of challenges in sustaining experience and expertise in disseminating
• Sustainability. these programs. To bring the true promise participant-centered, evidence-based self-
of these programs to scale, there needs to management, and health promotion pro-
The EBLC has accomplished several be an integration of infrastructures, and grams in communities nationwide, the
important tasks, including: (1) performed a one-stop-shop to build and to assist EBLC is poised to help many more orga-
an initial mapping of all agencies (more implementing organizations. nizations with limited resources effectively
than 1,700) offering any of the 19 pro- The focus of the EBLC going for- address population health challenges.
grams as well as which programs are being ward will be to improve coordination
offered by each agency; (2) completed a ACKNOWLEDGMENTS
and efficiency around marketing, techni-
telephone survey of 15 of the agencies This work was supported in part by the
cal assistance (including readiness assess-
offering two or more programs to iden- Archstone Foundation, the National Coun-
ment, fidelity, implementation planning,
tify facilitators and barriers to implemen- cil on Aging, and the US Centers for Dis-
and evaluation), training, and licensing and
tation of multiple EBPs and approaches ease Control and Prevention. The findings
fee structures. An EBLC website is also
to support scaling up these programs; and conclusions in this article are those of
being developed to improve access to tools
(3) participated in federal meetings with the authors and do not necessarily repre-
and information in each of these areas. A
the ACL, the National Council on Aging sent the official position of the Centers for
shared data management platform is being
Self-Management Alliance and others; and Disease Control and Prevention.
expanded to include all programs in the
(4) held four in-person strategic planning EBLC. This platform will:
meetings as well as smaller subcommittee
REFERENCES
1. Wagner EH, Austin BT, Davis C, Hindmarsh M,
meetings and bi-weekly phone calls. • Facilitate the effort to identify a mini- Schaefer J, Bonomi A. Improving chronic illness
The EBLC believes that our commu- mal set of common data points for all care: translating evidence into action. Health Aff
nity care system has now reached a stage programs, which can be used to evaluate (Millwood) (2001) 20:64–78. doi:10.1377/hlthaff.
where the status quo is no longer accept- dissemination, reach and outcomes both 20.6.64
2. Disease Prevention and Health Promotion Ser-
able. As the demand for programs has within and across programs, vices OAA Title IIID. (2014). Available from:
increased, program infrastructures have, • Efficiently, feed a public-facing website http://www.aoa.gov/AoARoot/AoA_Programs/
for the most part, not grown to meet by providing unduplicated and jointly HPW/Title_IIID/index.aspx

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 136 | 60
Haynes et al. EBLC – a national collaborative

3. Intervention Programs. (2014). Available from: 11. Quijano LM, Stanley MA, Petersen NJ, Casado in the Research Topic received peer review from members
http://www.cdc.gov/arthritis/interventions.htm BL, Steinberg EH, Cully JA, et al. Healthy of the Frontiers in Public Health (Public Health Edu-
4. SAMHSA National Registry of Evidence-Based IDEAS: a depression intervention delivered by cation and Promotion section) panel of Review Editors.
Programs. (2014). Available from: http://www. community-based case managers serving older Because this Research Topic represents work closely asso-
nrepp.samhsa.gov/Index.aspx adults. J Appl Gerontol (2007) 26(2):139–56. doi: ciated with a nationwide evidence-based movement in
5. AHRQ Innovations Exchange. (2014). Available 10.1177/0733464807299354 the US, many of the authors and/or Review Editors may
from: http://innovations.ahrq.gov/ 12. Tennstedt S, Howland J, Lachman M, Peterson have worked together previously in some fashion. Review
6. Brown NJ, Griffin MR, Ray WA, Meredith S, Beers E, Kasten L, Jette A. A randomized, controlled Editors were purposively selected based on their expertise
MH, Marren J, et al. A model for improving med- trial of a group intervention to reduce fear of with evaluation and/or evidence-based programming
ication use in home health care patients. J Am falling and associated activity restriction in older for older adults. Review Editors were independent of
Pharm Assoc (1998) 38(6):696–702. adults. J Gerontol B Psychol Sci Soc Sci (1998) named authors on any given article published in this
7. Ciechanowski P, Wagner E, Schmaling K, Schwartz 53B(6):384–92. doi:10.1093/geronb/53B.6.P384 volume.
S, Williams B, Diehr P, et al. Community- 13. Wallace JI, Buchner DM, Grothaus L, Leveille S, Tyll
integrated home-based depression treatment in L, LaCroix AZ, et al. Implementation and effective- Received: 18 June 2014; accepted: 21 August 2014;
older adults: a randomized controlled trial. JAMA ness of a community-based health promotion pro- published online: 27 April 2015.
(2004) 291(13):1569–77. doi:10.1001/jama.291. gram for older adults. J Gerontol A Biol Sci Med Sci Citation: Haynes M, Hughes S, Lorig K, Sim-
13.1569 (1998) 53a(4):M301–6. mons J, Snyder SJ, Steinman L, Wilson N, DiSte-
8. Hughes SL, Seymour RB, Campbell R, Pollak N, 14. Yan T, Wilber KH, Wieckowski J, Simmons JW. fano R, Raymond J, FallCreek S, Pelaez MB and
Huber G, Sharma L. Impact of the fit and strong Results from the healthy moves for aging well Smith D (2015) Evidence-based leadership council – a
intervention on older adults with osteoarthritis. program: changes of the health outcomes. Home national collaborative. Front. Public Health 2:136. doi:
Gerontologist (2004) 44(2):217–28. doi:10.1093/ Health Care Serv Q (2009) 28(1–3):100–11. doi: 10.3389/fpubh.2014.00136
geront/44.2.217 10.1080/01621420903176136 This article was submitted to Public Health Education
9. Leveille SG, Wagner EH, Davis C, Grothaus L, Wal- and Promotion, a section of the journal Frontiers in
lace J, LoGerfo M, et al. Preventing disability and Conflict of Interest Statement: The authors declare Public Health.
managing chronic illness in frail older adults: a ran- that the research was conducted in the absence of any Copyright © 2015 Haynes, Hughes, Lorig , Simmons,
domized trial of a community-based partnership commercial or financial relationships that could be Snyder, Steinman, Wilson, DiStefano, Raymond, Fall-
with primary care. J Am Geriatr Soc (1998) 46: construed as a potential conflict of interest. Creek, Pelaez and Smith. This is an open-access article
1–9. distributed under the terms of the Creative Commons
10. Lorig K, Sobel DS, Stewart AL, Brown BW, This paper is included in the Research Topic, “Evidence- Attribution License (CC BY). The use, distribution or
Bandura A, Ritter P, et al. Evidence suggesting Based Programming for Older Adults.” This Research reproduction in other forums is permitted, provided the
that a chronic disease self-management program Topic received partial funding from multiple government original author(s) or licensor are credited and that the
can improve health status while reducing hospi- and private organizations/agencies; however, the views, original publication in this journal is cited, in accordance
talization: a randomized trial. Med Care (1999) findings, and conclusions in these articles are those of the with accepted academic practice. No use, distribution or
37(1):5–14. doi:10.1097/00005650-199901000- authors and do not necessarily represent the official posi- reproduction is permitted which does not comply with
00003 tion of these organizations/agencies. All papers published these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 136 | 61

 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00254

Working toward a multi-program strategy in fall prevention

(Bonita) Lynn Beattie*
National Council on Aging, Washington, DC, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: fall prevention, enlisting community programs that work, community health services, healthy aging, falls free initiative

The exceptional compilation of healthy potential depression or anxiety. Medica- how can we recruit growing numbers
aging articles contained within this tions to treat chronic diseases can also of senior participants, program leaders,
Research Topic are timely, and high- lead to an increased risk of falls through and mentors?
light many important ongoing health both the absolute number taken and the • How can we capture outcomes to pro-
care reform initiatives to improve the potential interactions (3–6). mote the reimbursement of programs
healthy behaviors of older adults and aging Research strongly suggests that peo- that can reduce health care costs and
boomers. The national discussion under- ple who exercise regularly live longer promote quality of life?
standably focuses on chronic conditions and healthier lives. Being physically active
including cost containment, improved and following an exercise program can MAKING A SUSTAINABLE DIFFERENCE
patient outcomes, and quality of life mea- reduce the risk of developing some dis- It is evident that adequately managing
sures. However, I would suggest that the eases and disabilities that often occur with expressions of chronic conditions and sup-
prevention of older adult falls and related age. Strength exercises build muscles and portive medication regimens can affect
injuries should be an integral part of the reduce the risk of osteoporosis. Flexibil- the risk of falls and fall-related injuries
discussion. By broadening the discussion ity or stretching exercises help keep the in older adults. I believe that there is
of effective management of chronic dis- body and joints flexible and often help to urgency to broaden the discussions on
eases and focusing on how to help inform, modulate pain (7). Not surprisingly, exer- chronic disease management and how to
educate, and support aging Americans, we cise – especially strength, balance, and flex- best apply disease management guidelines
could also reduce the growing number of ibility – is a key strategy in reducing the risk to fall prevention. Further, there is an
falls and falls-related injuries and deaths in of falls and serious injury. opportunity to capitalize on the invest-
this vulnerable population (1). Seminal research by Tinetti and col- ments of the U.S. Administration on Com-
While the evidence is strong that a small leagues noted that the cumulative number munity Living in the dissemination of
number of targeted prevention programs of falls risks (including but not limited sustainable, evidence-based health promo-
have significantly reduced falls in older to declining strength; balance/gait issues; tion, and chronic disease self-management
adults, few of these programs followed par- vision changes; postural blood pressure; programs.
ticipants for longer than 12 months (2). depression; arthritis; foot problems; mul- As the population of elderly grows to
However, in the absence of long-term out- tiple medications; and environmental haz- over 70 million by 2030 (9), there is
come data tracking the maintenance of ards) mattered (8). So, it seems that the value, even an urgency, to enlist commu-
behavior changes, it is difficult to evalu- questions worth exploring are: nity evidence-based programs and services
ate if we are promoting long-term healthy to offer older adults the opportunity to bet-
behaviors or just forestalling the onset of • Can we make a strong case for the ter manage their chronic disease, enhance
a fall. fall prevention contributions of com- their level of physical activity, and modify
The evidence is equally strong for link- munity programs effective in helping their risk of falls and injury.
ing the growing number of chronic con- older adults make behavior changes What is needed now is a more inclu-
ditions in older adults to falls. Chronic to enhance the management of their sive approach to the effective manage-
disease can significantly increase the risk of chronic conditions? ment of chronic disease and reduction
a variety of factors associated with those • Can we consider a multi-program, of fall risk; an approach that values and
diseases. This includes, but is not lim- longer-term community strategy that enfolds the broad spectrum of healthy
ited to functional limitations and disabil- helps to maintain behavior change, pro- aging program offerings. I believe that by
ities; chronic pain; sensory deprivations; motes physical activity, and helps to providing evidence-based prevention pro-
vision effects; and balance and gait dis- better manage medications and chronic grams to help older adults and their care-
turbances. Chronic disease manifestations conditions as a longer-term fall preven- givers make better choices, improve their
may also increase the risk of falls through tion strategy? health, and increase their quality of life
indirect effects such as reduced physical • How will seniors/caregivers view this will ultimately affect the rate of elderly
activity level, reduced social activities, and change in strategy? More importantly falls.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 254 | 62
Beattie Enfolding new fall prevention assets

REFERENCES falls risk. Rheumatologist (2009). Available at: in the Research Topic received peer review from members
1. Sibley KM, Voth J, Munce SE, Strauss SE, Jaglal http://www.cdc.gov/mmwr/preview/mmwrhtml/ of the Frontiers in Public Health (Public Health Edu-
SB. Chronic disease and falls in community- mm6317a3.htm?s_cid%3Dmm6317a3_x cation and Promotion section) panel of Review Editors.
dwelling Canadians over 65 years old: a 7. Amsterdam EA, Kappagoda CT. Benefits of exer- Because this Research Topic represents work closely asso-
population-based study exploring associations with cise in the elderly. In: Jugdutt BI, editor. Aging and ciated with a nationwide evidence-based movement in
number and pattern of chronic con- Heart Failure. New York: Springer Science+Business the US, many of the authors and/or Review Editors may
ditions. BMC Geriatr (2014) 14:22. Media (2014). p. 147–64. have worked together previously in some fashion. Review
doi:10.1186/1471-2318-14-22 8. Tinetti M, Williams TF, Mayewski R. Fall risk index Editors were purposively selected based on their expertise
2. Stevens JA, Sogolow ED. Preventing Falls – What for elderly patients based on number of chronic dis- with evaluation and/or evidence-based programming
Works: a CDC Compendium of Effective Community- abilities. Am J Med (1986) 80:429–34. doi:10.1016/ for older adults. Review Editors were independent of
Based Interventions From Around the World. 0002-9343(86)90717-5 named authors on any given article published in this
Atlanta: CDC, National Center for Injury Preven- 9. Vincent G, Velkoff V. The Next Four Decades, the volume.
tion (2008). Older Population in the United States: 2010 to 2050
3. Lawlor DA, Patel R, Ebrahim S. Association Population Estimates and Projections. (2010). Avail- Received: 16 June 2014; accepted: 09 November 2014;
between falls in elderly women and chronic able from: https://www.census.gov/prod/2010pubs/ published online: 27 April 2015.
disease and drug use: cross sectional study. p25-1138.pdf Citation: Beattie BL (2015) Working toward a multi-
BMJ (2003) 327(7412):712–7. doi:10.1136/bmj. program strategy in fall prevention. Front. Public Health
327.7426.1288-a Conflict of Interest Statement: The author 2:254. doi: 10.3389/fpubh.2014.00254
4. Sturnieks DL, Tiedemann A, Chapman K, Munro declares that the research was conducted in the This article was submitted to Public Health Education
B, Murray SM, Lord SR. Physiological risk factors absence of any commercial or financial relationships and Promotion, a section of the journal Frontiers in
for falls in older people with lower limb arthritis. that could be construed as a potential conflict of Public Health.
J Rheumatol (2004) 31(11):2272–9. doi:10.1046/j. interest. Copyright © 2015 Beattie. This is an open-access arti-
1532-5415.2001.49107 cle distributed under the terms of the Creative Commons
5. Barbour KE, Stevens J, Helmick C, Luo YH, Mur- This paper is included in the Research Topic, “Evidence- Attribution License (CC BY). The use, distribution or
phy LB, Hootman JM, et al. Falls and Fall Injuries Based Programming for Older Adults.” This Research reproduction in other forums is permitted, provided the
Among Adults with Arthritis – United States, 2012. Topic received partial funding from multiple government original author(s) or licensor are credited and that the
MMWR. Washington: U.S. Government Printing and private organizations/agencies; however, the views, original publication in this journal is cited, in accordance
Office (2014). findings, and conclusions in these articles are those of the with accepted academic practice. No use, distribution or
6. Davis GC. Reduce the danger of falls: com- authors and do not necessarily represent the official posi- reproduction is permitted which does not comply with
mon factors in arthritis patients increase tion of these organizations/agencies. All papers published these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 254 | 63

 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00194

Building the older adult fall prevention movement – steps

and lessons learned
Ellen C. Schneider 1 * and (Bonita) Lynn Beattie 2
1
Center for Health Promotion and Disease Prevention, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
2
Falls Free® Initiative, National Council on Aging, Washington, DC, USA

Edited by: Background: Falls are the leading cause of older adult injuries and injury-related deaths.
Matthew Lee Smith, The University of
Until 2004, the growing public health issue of older adult falls received little national atten-
Georgia, USA
tion. To elevate and focus on the issue, the National Council on Aging launched the Falls
Reviewed by:
Elisa Beth Hutson McNeill, Texas Free® Initiative, a group of national and state agencies working collaboratively to address
A&M University, USA older adult falls with evidence-based solutions. Since then, attention to older adult falls has
Miruna Petrescu-Prahova, University gained significant momentum.
of Washington, USA
Deborah Paone, Paone & Associates, Purpose: To describe the steps taken to create the momentum around fall prevention and
LLC, USA lessons learned that could be applied to supporting other older adult health-related issues.
*Correspondence:
Ellen C. Schneider , Center for Health Method/objectives: The Falls Free® Initiative took key steps to promote the older adult
Promotion and Disease Prevention, falls prevention movement, including initiating organized advocacy and supporting the
University of North Carolina at Chapel
development of state coalitions through increasing awareness of the issue, promot-
Hill, 1700 Martin Luther King
Boulevard, CB 7426, Chapel Hill, NC ing evidence-based programs, instituting evaluation, implementing systems change, and
27599-7426, USA providing tailored technical assistance.
e-mail: [email protected]
Results: Through the support of the Falls Free® Initiative and many partners, advocacy
efforts have increased federal funding for fall prevention, the majority of states have fall
prevention coalitions, and thousands of stakeholders are now engaged in fall prevention.
Select lessons learned include leveraging compelling data, choosing passionate leaders
for the movement, aligning the cause with partner missions, and being inclusive of all
stakeholders.
Conclusion: Although much progress has been made in the fall prevention movement, the
issue is growing along with the aging population. Efforts must continue to gain support
from all affected stakeholders to reduce older adult falls and fall-related injuries.
Keywords: fall prevention, fall prevention movement, Falls® Free Initiative, awareness, advocacy

In 2012, over 2.4 million older adults were treated in emergency Launched in 2005, the Falls Free® Initiative brought together
departments for falls; more than 722,000 or 30% of these patients national and state agencies to collaboratively address older adult
had to be hospitalized (1). Every 29 min, an older adult in the falls with evidence-based solutions; the authors of this article were
United States dies from fall-related injuries (2). Direct medical leaders in the effort. The Falls Free® Initiative has been particu-
costs for fall injuries total over $30 billion per year in the nation larly successful in advocacy at the national level and in supporting
and account for 6% of all medical expenditures for this age group the creation and development of state fall prevention coalitions
(3, 4). The risk of falling increases with age, and accelerates after age and local collaborative efforts across the country. The purpose of
85 years due to issues such as declining muscle strength, increased this paper is to describe the steps taken to create the momen-
frailty, poor eyesight, and limited movement (5). With an increased tum around fall prevention and lessons learned that could be
life expectancy among the growing baby boomer population, applied to supporting other older adult health-related issues. Steps
the problem of older adult falls has the potential to overwhelm include creating a national initiative; initiating advocacy efforts;
resources required to address the needs. and developing and supporting coalitions to increase awareness of
Until 2004, the issue of older adult falls received little national the issue, promote evidence-based programs, institute evaluation,
attention in part due to its complexity and lack of readily available and implement policy and systems change.
evidence-base interventions. As a growing public health issue, it
clearly needed a national effort to promote awareness and action. BACKGROUND OF THE FALLS FREE® INITIATIVE
Since then, attention to the issue of older adult falls has gained In the early 2000s, the Archstone Foundation, a private grant-
significant momentum through the work of many stakeholders making organization based in California, began funding local
around the country, primarily led through the National Council fall prevention initiatives in the state. The Archstone Founda-
on Aging’s (NCOA) Falls Free® Initiative (6). tion was pleased with the growth of statewide activities and the

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 194 | 64
Schneider and Beattie Building the fall prevention movement

subsequent development of a state fall prevention plan. The Foun- across the strategies presented in the National Action Plan, as well
dation approached the NCOA to design a similar initiative on a as emerging opportunities.
national level. NCOA, with funding from the Archstone Founda-
tion and the Home Safety Council, seated an advisory group of INITIATING ORGANIZED FALL PREVENTION ADVOCACY
leading fall prevention experts to begin planning and conducted In 2006, the Falls Free® Initiative recognized that the issue of
an environmental scan of organizations that were or should be older adult falls needed an active effort to advocate for appro-
working on fall prevention to identify key national stakeholders. priate national funding levels. Therefore, the National Falls Free®
Concurrently, leading researchers were commissioned to Advocacy Workgroup was formed and successfully advocated for
develop review papers based on the best available evidence on fall the passage of the Keeping Seniors Safe from Falls Act, signed into
prevention strategies targeted toward community-residing older law in April 2008 as PL 110-202 (8). The Act enfolded strategies
adults. The review papers focused on the topics of physical mobil- taken directly from the National Action Plan authorizing research,
ity, medications management, home safety, environmental safety demonstration programs, provider training, and public education
in the community, and additional cross-cutting areas for attention to prevent older adult falls. Although the Act passed, no funding
such as policy and advocacy. was appropriated with its enactment. The Workgroup continued
With continued financial support from the Archstone Foun- its advocacy efforts and successfully doubled fall prevention fund-
dation and Home Safety Council, NCOA convened the national ing for the Centers for Disease Control and Prevention’s (CDC)
Falls Free® Summit in Washington, DC, USA in December 2004. National Center for Injury Prevention and Control (NCIPC) from
Fifty-eight national organizations, professional associations, fed- $1 million in fiscal year 2008 to approximately $2 million in fiscal
eral agencies, and leading fall prevention experts were invited to year 2009 and subsequent years.
participate in this landmark summit to review the evidence and Centers for Disease Control and Prevention has used those
design a national blueprint for reducing falls among older adults. funds to translate and test evidence-based programs, conduct
As a result of the Summit, the National Action Plan (Plan avail- demonstration projects, and develop the STEADI (Stopping
able at http://www.ncoa.org/improve-health/center-for-healthy- Elderly Accidents, Deaths and Injuries) Tool Kit for health care
aging/content-library/FallsFree_NationalActionPlan_Final.pdf) providers (9). In 2014, continued advocacy efforts led to the allo-
was created with 9 goals and 36 evidence-based strategies key to cation of $5 million from the Affordable Act’s Prevention and
reducing older adult falls (7). Goals and strategies were offered for Public Health Fund for elder fall prevention to the Administra-
both providers and for older adults corresponding with the review tion for Community Living/Administration on Aging (AoA) (10).
papers on the evidence related to physical mobility, medications Funds are to increase the availability of and accessibility to effective
management, home and environmental safety, and cross-cutting programs and services in communities.
issues. The long range vision of the Plan was that older adults
would have fewer falls and fall-related injuries, maximizing their DEVELOPING STATE FALL PREVENTION COALITIONS
independence and quality of life (7). More than 8,000 print and CD In 2006, the Falls Free® Initiative accelerated with the addition
ROM copies of the Plan were distributed; the Plan was also posted of the State Coalitions on Fall Prevention Workgroup. At that
to the NCOA website where it has been downloaded over 125,000 time, only four states had fall prevention coalitions. These states
times (personal communication, Emily Dessem, National Council approached Falls Free® leadership and asked to join the effort.
on Aging, 2014 March 2). The purpose of the plan was to develop The State Coalitions on Fall Prevention Workgroup was formed
and enrich supplemental and complementary community-based and designed to facilitate collaboration between states working on
programs and services to provide a continuum of care aimed at similar issues. The State Coalition Workgroup members reported
reducing falls and fall-related injuries, not to undermine medical that developing a state or large regional coalition to address falls
interventions. and fall-related injuries offered a common forum for multidisci-
When the Plan was released, there was insufficient funding to plinary organizations to address falls, deter duplication of efforts,
mount a national campaign to promote action of its 36 strate- raise awareness, and facilitate necessary roles of resource coor-
gies. However, in response to the participants’ enthusiasm for dination, policy development, and systems change at the state
the Summit process and the Plan itself, and in an effort to pro- level (11).
mote the strategies, the Falls Free® Initiative was created (6). This To encourage other states to develop their own fall prevention
loose-knit collaborative of Summit attendees and their organi- coalitions, the Falls Free® Initiative developed a tool kit, Falls and
zations was charged with working toward the progress of one or Fall-Related Injuries Among Older Adults: A Practical Guide to State
more of the strategies that resonated with their organizational mis- Coalition Building to Address a Growing Public Health Issue (11).
sions. Since its inception, the Falls Free® Initiative has grown to Based on available evidence for coalition building, it was designed
over 70 national organizations, profession associations, and federal to enfold the strategies, experiences, and lessons learned of the
agencies (6). 10 fall prevention coalitions in existence by 2007 when the tool
The original 36 strategies remain relevant and evidence- kit was created. It includes three stages and nine recommended
supported. In 2008, the Falls Free® National Advisory Group con- steps, each with many subtasks, to initiate and build an effective
vened to review progress made over its 3-year history. Group mem- Fall Prevention Coalition. The tool kit still serves as the basis of
bers engaged in a rich, broad-based exchange of ideas; this deliber- NCOA technical assistance to states and local communities seeking
ation resulted in a number of recommendations and observations to build coalitions.

www.frontiersin.org April 2015 | Volume 2 | Article 194 | 65

Schneider and Beattie Building the fall prevention movement

In addition to the tool kit, Falls Free® Initiative leadership pro- SUPPORTING STATE FALL PREVENTION COALITIONS
vided individualized technical assistance to over 30 states that As the number of coalitions grew, the NCOA continually collab-
expressed interest in forming a fall prevention coalition. Tech- orated with them to provide targeted technical assistance and
nical assistance included structured calls or in-person meetings assist them to enhance their individual and collective impact.
with coalition leads to walk them through the steps in the Coali- The following discussion outlines the processes and tools NCOA’s
tion Building tool kit. Support was provided to answer questions Falls Free® Initiative leadership, with support and input from
about membership sectors, coalition structure, goals and objec- coalitions and partners, developed to support and sustain their
tives, funding, and evaluation. Formal quarterly calls were also efforts.
held with the full State Coalitions on Fall Prevention Workgroup
to problem solve and collaborate. INCREASING FALL PREVENTION AWARENESS
As new coalitions formed, they were added to the State Coali- In 2008, the 10 state-member Falls Free® State Coalitions on Fall
tions on Fall Prevention: Working Collaboratively to Make a Differ- Prevention Workgroup (15) requested assistance in declaring a
ence Compendium of Initiatives (12). This document was translated day of awareness; four states had already targeted the autum-
to an interactive map on the NCOA state fall prevention coalition nal equinox, which NCOA and the remaining states adopted. In
website (13). The website features background and contact infor- response, the National Advocacy Workgroup gained bipartisan
mation for each coalition so that potential partners can join or sponsorship of the first annual National Falls Prevention Aware-
have questions addressed about fall prevention activities in the ness Day (FPAD) resolution in the U.S. Senate and has obtained
state. Subsequently, NCOA worked with states to develop their bipartisan sponsorship every year since then. The number of states
own unique state profile of the impact of falls; the states included observing FPAD grew from 4 in 2007 to 11 in 2008, 22 in 2009, 36
demonstrate a powerful visual for advocacy purposes (14). in 2010, 43 in 2011, 46 in 2012, and 47 in 2013, plus the District of
In 2007, NCIPC and AoA entered into an interagency agree- Columbia (15).
ment to promote evidence-based fall prevention intervention According to a survey of state fall prevention coalition leads
including support of the Falls Free® Initiative. In the same year, conducted by the NCOA, an estimated 2,076,041 older adults were
NCIPC also named older adult falls as one of its top three pri- reached during FPAD activities in 2013, more than 511,000 partic-
ority areas. Making older adult falls a priority helped to engage ipated in evidenced-based fall prevention programs, over 17,000
the public health community and foster the development of fall were screened for falls risks, and more than 1.5 million older adults,
prevention coalitions in states with CDC Core Violence and Injury family caregivers, professionals, and policymakers were reached
Prevention Program grants; that funding could be used to support through advocacy events and education and awareness campaigns
fall prevention activities and Injury Community Planning Groups (16). Since 2011, NCOA has offered an annual webinar in advance
with falls as a priority. From 2006 to 2014, the number of active or of FPAD to generate creative partnerships and activities across the
developing state fall prevention coalitions grew from 4 to 43 (see country. Each year, webinar registration from across the country
Figure 1). has exceeded 1,000.

FIGURE 1 | States with active or developing fall prevention coalitions in 2014.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 194 | 66
Schneider and Beattie Building the fall prevention movement

States indicated that increasing awareness of fall prevention response to those requests, NCOA organized an Evaluation Com-
was an important goal of their coalitions (11), and several imple- mittee of the State Coalitions on Fall Prevention Workgroup
mented fall prevention awareness campaigns. To better understand to support state and local evaluation efforts. Members included
the fall prevention awareness campaigns that the states and their state coalition leads, researchers, advisors, and CDC staff. The
fall prevention coalitions implemented and the lessons learned Evaluation Committee developed guidelines to help state fall pre-
other states could apply to their awareness campaigns, Falls Free® vention coalitions evaluate the impact of their efforts, and foster
leadership interviewed 10 state agencies and 1 national organi- comparisons across states (20). The availability of a standard
zation between October 2008 and February 2009. As a result of evaluation process and strategies helps to develop state baseline
those interviews, Falls Prevention Awareness: Lessons Learned from measures, promote consistency in evaluation efforts across states,
State Coalitions on Fall Prevention (17) was developed to assist and provide data for advocacy or funding opportunities.
other states in developing fall prevention awareness campaigns. The Evaluation Guidelines contain two important products for
The document contains numerous lessons learned about target state coalitions, including the Falls Free® Logic Model and a stan-
audiences, messaging, media and methods, and recommendations dard set of survey questions (20). The Logic Model illustrates the
for FPAD activities. causal assumptions linking coalition activities to long-term, mea-
surable outcomes. State coalition leaders can choose to focus their
PROMOTING EVIDENCE-BASED PROGRAMS efforts using the logic model (adaptable to states’ specific needs)
The Falls Free® Initiative is housed within the Center for Healthy as a guide. The Logic Model articulates the relationship between
Aging (CHA); CHA’s mission is to promote evidence-based health the resources used to operate the coalition, the activities that the
promotion and disease prevention programs. The CHA has been coalition conducts, and the outcomes and impact that the coalition
working with a variety of state and local grantees since 2003 to will achieve. By demonstrating the progression, state coalitions
adopt and sustain programs making a difference in the health, can help stakeholders understand how their work leads to desired
independence, and quality of life of older adults. These programs outcomes.
provide measurable improvements in patient outcomes and build The standard set of survey questions measure progress of key
patient knowledge skills and confidence to manage problems. stakeholders including older adults, children of older adults, pri-
Since providers are increasingly being required by Centers for mary care providers, and state legislators. The question sets were
Medicare and Medicaid (CMS) to promote appropriate healthy selected from validated surveys and research activities, and states
behaviors, evidence-based programs can be a valuable asset for were asked to use the questions as designed since the standard set
provider referrals. In alignment with this requirement, a key strat- of questions must be the same to allow comparisons across states
egy of the Falls Free® Initiative is to increase access to quality and to demonstrate national impact. Three states (Kansas, New
programs and promote linkages and referrals from the health York, and New Hampshire) added a subset of questions from the
care community. A number of strategies within the State Policy standard set of survey questions to their state Behavioral Risk Fac-
Toolkit for Advancing Falls Prevention (discussed below) promote tor Surveillance System (BRFSS) survey. In 2012, 94 people from
this effort to affect fall prevention. across the country attended a webinar to learn how to use the
There are a variety of evidence-based programs recognized as Evaluation Guidelines (personal communication, Emily Dessem,
effective for fall prevention. A CDC Compendium of Effective Fall National Council on Aging, 2014 March 2).
Interventions: What Works for Community-Dwelling Older Adults,
2nd Edition represents a significant CDC investment in providing IMPLEMENTING FALL PREVENTION POLICY AND SYSTEMS CHANGE
access to programs that can work in a variety of community and Through these various efforts, significant progress was made in the
home settings and have been shown through randomized trials areas of increasing fall prevention awareness, creating multidisci-
to reduce falls (18). However, few of the 22 programs listed offer plinary networks, and identifying evaluation measures. However,
training, tools, and resources for successful implementation. Pro- to achieve systems change and long-term sustainability, state fall
grams ready to implement include Stepping On, Tai Chi: Moving prevention coalitions recognized that implementing policy change
for Better Balance and Otago. In addition, ACL/AoA recognizes was necessary, and subsequently requested assistance from Falls
A Matter of Balance and others available for funding under the Free® leadership in identifying, implementing, and advancing a
Older Americans Act, Title IIID (19). full array of fall prevention policies to pursue.
Ongoing Falls Free® collaborative activities promote shar- To assist the states in their policy development, Falls Free® lead-
ing among fall prevention coalitions of best practices, strategies, ership developed and released the State Policy Toolkit for Advancing
and funding opportunities for evidence-based programs. Active Falls Prevention, which incorporates the previously discussed Falls
partnership-building strategies are used to link coalitions with Free® Logic Model as a framework to advance policy change
national Falls Free® member organizations such as the American (21). The tool kit was disseminated to the State Workgroup on
Physical Therapy Association through their state chapters and local Fall Prevention Coalition members and on the NCOA website.
activities. A webinar was held in 2013 to train over 250 attendees on how
to use and implement policies in the policy tool kit. A survey of
INSTITUTING EVALUATION state fall prevention coalition leads was conducted prior to the
State fall prevention coalitions expressed the need to demonstrate webinar to ascertain policy implementation, and results indicate
the impact of their coalition work to a variety of stakeholders, that coalitions are actively pursuing evidence-based policies (see
coalition partners, funding organizations, and policymakers. In Figure 2).

www.frontiersin.org April 2015 | Volume 2 | Article 194 | 67

Schneider and Beattie Building the fall prevention movement

While there has been significant progress in raising awareness

of older adult falls, increasing the number of fall prevention coali-
tions, promoting evidence-based programs, instituting evaluation,
implementing fall prevention policy and systems change, and
enhancing federal funding for fall prevention efforts, much work
still must be done. Older adults themselves and their caregivers
need to take a more proactive role in fall prevention. Perhaps due
to the stigma of falling, lack of understanding that many falls can
be prevented, or limited knowledge on how to get involved, con-
sumers are not yet well engaged in offering new goals for advancing
this movement.
The escalating issue of older adult falls affects every state. As
the Falls Free® Initiative has demonstrated, one effective approach
is an inclusive, targeted coalition to bring partners together to
FIGURE 2 | Policy goals being “actively worked on” by state fall address the issue. A successful rallying activity is the annual
prevention coalitions are shown.
observance of FPAD. However, this one observance needs to be
leveraged into a more comprehensive approach to community fall
prevention.
DISCUSSION AND LESSONS LEARNED Despite the states’ request for coalition evaluation guidelines,
Although the Falls Free® Initiative played an important supportive
uptake has been limited. States may not have embraced the guide-
role in the fall prevention coalition movement, progress would not
lines due to lack of funding to conduct evaluations or lack of
have been possible without the work of thousands of people across
awareness that the guidelines are available.
the country, including the state fall prevention coalition leaders
Future research is needed to better understand how states and
and members, public health community, aging services network,
communities can implement policy and systems change to more
health care providers, researchers, and many, many others.
effectively implement falls prevention initiatives within and across
Creating a movement is not simple, but there are several lessons
sectors such as health care, aging services, and public health. Addi-
learned from the progress made thus far that could be applied to
tionally, the escalating issue of older adult falls is severely under-
other older adult health care issues at the state or local level (11):
funded, so partners and stakeholders must continue to advocate
for support.
• Identify and promote the issue. With its growing network of dedicated champions, the Falls
• Use available data to define the issue, convey its impact, and Free® Initiative will continue its collaborative efforts to address
design strategies. these areas of focus with the ultimate goal of reducing the num-
• Collect and share personalized stories of the impact of the ber of falls and fall-related injuries, increasing life expectancy, and
issues and how programs and services are making a difference improving quality of life among older adults in the United States.
in the lives of older adults.
• Advocate with legislators and decision-makers to promote REFERENCES
your issue; if your organization is not permitted to advocate, 1. Centers for Disease Control and Prevention. Injury Prevention & Control: Data
find partners who can. & Statistics (WISQARS). (2014). Available from: http://www.cdc.gov/injury/
• Engage partners and leaders. wisqars/
2. Centers for Disease Control and Prevention. Falls are a Major Threat for your
• Find partners whose missions align in some way with your
Patients. (2014). Available from: http://www.cdc.gov/HomeandRecreational
issue. Safety/pdf/steadi/falls_are_a_major_threat.pdf
• Choose engaged, passionate, determined leaders, and cham- 3. Guideline for the prevention of falls in older persons. American Geriatrics Soci-
pions to promote the issue. ety, British Geriatrics Society & American Academy of Orthopaedic Surgeons
• Clarify organizational relationships for lead roles and joint Panel on Falls Prevention. J Am Geriatr Soc (2001) 49:664–72. doi:10.1046/j.
1532-5415.2001.49115.x
planning activities. Develop partnerships between health
4. Centers for Disease Control and Prevention. Cost of Falls Among Older
care, aging, public health, and research networks. Adults. (2014). Available from: http://www.cdc.gov/homeandrecreationalsafety/
• Promote collaboration in the cross-agency planning, design, falls/fallcost.html
implementation, and evaluation of programs for older adults. 5. Centers for Disease Control and Prevention. Falls Among Older Adults: An
• Be inclusive; all stakeholders have a role to play. Overview. (2013). Available from: http://www.cdc.gov/homeandrecreational
safety/falls/adultfalls.html
• Identify solutions. 6. National Council on Aging. Falls Free® Initiative. (2014). Available from:
• Seek out and adapt measurable, feasible, evidence-based www.ncoa.org/FallsFreeInitiative
solutions. 7. National Council on Aging. Falls Free: Promoting a National Falls Prevention
• Increase the availability of tools, resources, and programs, Action Plan. (2005). Available from: http://www.ncoa.org/improve-health/
so affected individuals will have methods to reduce their center-for-healthy-aging/content-library/falls-free-promoting-a.html#sthash.
mNlUg5yD.dpuf
individual risks. 8. Government Printing Office. Public Law 110–202. (2008). Available from:
• Keep trying if you fail. It can take time for a movement to http://www.gpo.gov/fdsys/pkg/PLAW-110publ202/pdf/PLAW-110publ
take root, grow, and bloom. 202.pdf

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 194 | 68
Schneider and Beattie Building the fall prevention movement

9. Centers for Disease Control and Prevention. STEADI (Stopping Elderly Acci- 20. National Council on Aging. Evaluation Guidelines to Measure the Impact
dents, Deaths, & Injuries) Tool Kit for Health Care Providers. (2014). Available of State and Local Coalitions on Fall Prevention. (2012). Available
from: http://www.cdc.gov/homeandrecreationalsafety/Falls/steadi/index.html from: http://www.ncoa.org/improve-health/center-for-healthy-aging/content-
10. National Council on Aging. Aging Program Funding. (2014). Available at: library/evaluation-guidelines-1.html
http://www.ncoa.org/assets/files/pdf/public-policy–action/FY14-Request-v1. 21. Beattie BL, Schneider EC. State Policy Toolkit for Advancing Fall Prevention.
pdf?utm_source=FFEnews_1401&utm_medium=newsletter&utm_campaign= (2012). Available from: www.ncoa.org/fallspolicy
FFEnews)
11. National Council on Aging. Falls and Fall-Related Injuries Among Older Adults:
A Practical Guide to State Coalition Building to Address a Growing Public Health Conflict of Interest Statement: The authors declare that the research was conducted
Issue. (2014). Available from: www.ncoa.org/fallstoolkit in the absence of any commercial or financial relationships that could be construed
12. National Council on Aging. State Coalitions on Fall Prevention: Working as a potential conflict of interest.
Collaboratively to Make a Difference – Compendium of Initiatives. (2011). Avail-
able from: http://www.ncoa.org/improve-health/center-for-healthy-aging/ This paper is included in the Research Topic, “Evidence-Based Programming for Older
content-library/State-Coalitions-Compendium-2011.pdf Adults.” This Research Topic received partial funding from multiple government and
13. National Council on Aging. State Coalitions Map. (2014). Avail- private organizations/agencies; however, the views, findings, and conclusions in these
able from: http://www.ncoa.org/improve-health/center-for-healthy-aging/falls- articles are those of the authors and do not necessarily represent the official position
prevention/state-coalitions-map/ of these organizations/agencies. All papers published in the Research Topic received
14. National Council on Aging. Fall Prevention Profiles. (2014). Available peer review from members of the Frontiers in Public Health (Public Health Education
from: http://www.ncoa.org/improve-health/center-for-healthy-aging/content- and Promotion section) panel of Review Editors. Because this Research Topic repre-
library/falls-state-profiles/falls-prevention-state.html sents work closely associated with a nationwide evidence-based movement in the US,
15. National Council on Aging. NCOA Leadership Making a Difference: The Falls many of the authors and/or Review Editors may have worked together previously in
Free® Initiative. (2013). Available from: http://www.ncoa.org/assets/files/pdf/ some fashion. Review Editors were purposively selected based on their expertise with
NCOA-Falls-Free-Overview-6-2014.pdf evaluation and/or evidence-based programming for older adults. Review Editors were
16. National Council on Aging. Falls Prevention Awareness Day 2013: A independent of named authors on any given article published in this volume.
Compendium of State and National Activities. (2013). Available from:
http://www.ncoa.org/assets/files/pdf/center-for-healthy-aging/2013-FPAD-
Compendium-FINAL.pdf Received: 14 July 2014; accepted: 29 September 2014; published online: 27 April 2015.
17. Schneider EC, Beattie BL. Falls Prevention Awareness: Findings and Citation: Schneider EC and Beattie BL (2015) Building the older adult fall pre-
Lessons Learned from State Coalitions on Fall Prevention. (2009). Avail- vention movement – steps and lessons learned. Front. Public Health 2:194. doi:
able from: http://www.ncoa.org/improve-health/center-for-healthy-aging/ 10.3389/fpubh.2014.00194
content-library/Falls_Prevention_Lessons_Learned1-27-10.pdf This article was submitted to Public Health Education and Promotion, a section of the
18. Centers for Disease Control and Prevention. CDC Compendium of Effective Fall journal Frontiers in Public Health.
Prevention Interventions: What Works for Community-Dwelling Older Adults. 2nd Copyright © 2015 Schneider and Beattie. This is an open-access article distributed
ed. (2012). Available from: http://www.cdc.gov/HomeandRecreationalSafety/ under the terms of the Creative Commons Attribution License (CC BY). The use, dis-
Falls/compendium.html tribution or reproduction in other forums is permitted, provided the original author(s)
19. U.S. Administration on Aging. Disease Prevention and Health Promotion Ser- or licensor are credited and that the original publication in this journal is cited, in
vices (OAA Title IIID). (2013). Available from: http://www.aoa.acl.gov/AoA_ accordance with accepted academic practice. No use, distribution or reproduction is
Programs/HPW/Title_IIID/index.aspx permitted which does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 194 | 69

 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00191

Public health system perspective on implementation of

evidence-based fall prevention strategies for older adults
Sallie R. Thoreson 1 *, Lisa M. Shields 2 , David W. Dowler 2,3 and Michael J. Bauer 4
1
Colorado Department of Public Health and Environment, Denver, CO, USA
2
Oregon Health Authority, Portland, OR, USA
3
Multnomah County Health Department, Portland, OR, USA
4
New York State Department of Health, Albany, NY, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: state health departments, evidence-based strategy, older adults, fall prevention, health promotion

STATE HEALTH DEPARTMENTS’ directly providing programing by state- partners see and develop their roles in
MISSION AND STRATEGIES level agencies, they would partner with falls prevention, and providing the techni-
State health departments have tradition- local organizations to build infrastructure, cal expertise to share marketing strategies
ally worked in many areas of public health, change policies, and increase delivery and so partners could ensure their programs
including injury prevention (1). The public sustainability of the evidence-based pro- effectively reach the older adult audience.
health approach toward injury and disease grams. This commentary shares the expe- Lastly, each state health department applied
prevention directs programs to examine rience from our three states after 2.5 years evaluation techniques to provide feed-
surveillance data and then design, imple- of efforts to build clinical and community back to the partners on the positive out-
ment, and evaluate strategies to address prevention efforts to reduce falls in older comes of the programs, and to initiate
problems, such as falls among older adults adults. program changes when a strategy was not
(1, 2). The emphasis is to select evidence- working.
based strategies that have been success- SUCCESSES
fully tested in research settings and trans- The goal of the state health departments CHALLENGES MOVING FORWARD
lated into programs that are readily avail- was to go beyond “business as usual.” The All of the states developed program imple-
able for implementation. Fall prevention states worked to develop innovative part- mentation strategies to meet these chal-
among older adults has been acknowl- nerships to effectively reach target audi- lenges:
edged as a priority topic area, and one for ences. As illustrated with specific examples
which evidence-based strategies have been in Table 1, successful implementation of • It took substantial time and effort
identified (1, 3). the programs by each state health depart- to embed these programs into exist-
ment can be attributed to a number of ing infrastructure within the state
STATE ROLE IN FALL PREVENTION factors. First, each state ensured that inter- health departments and their partners.
In 2011, the Centers for Disease Con- nal support for the program was integrated The comprehensive integrated approach
trol and Prevention (CDC) funded state within the structure and function of the requiring simultaneous implementation
health departments in New York, Colorado, state health department. Second, the states of four programs was a definite chal-
and Oregon to implement evidence-based disseminated the programs through a vari- lenge.
older adult fall prevention programs over a ety of creative partnerships with health care • Recruiting and implementing STEADI
5-year period in several communities. The and community-based organizations not with health care providers was diffi-
grantees were tasked to bring or expand the traditionally involved with public health. cult. Health care entities were reluctant
evidence-based programs of Stepping On, Third, the states learned to understand and to partner with public health agencies
Tai Chi: Moving for Better Balance and the work with the needs of their partner orga- given the demands of the clinic prac-
Otago Exercise Program to community- nizations. An important lesson for work- tice and multiple initiatives already being
dwelling older adults (4). In addition, the ing with health care partnerships was to promoted. Part of the challenge was
grantees promoted the clinic-based stop- acknowledge their business goals and con- the need for rapid education of health
ping elderly accidents, deaths, and injuries sider initiatives meaningful to each organi- department staff in electronic medical
(STEADI) toolkit developed by the CDC zation. Next, the state health departments records and Medicare billing and cod-
to improve medical providers’ falls pre- made it a priority to assist the local partners ing. Additionally, identifying and moti-
vention assessment and treatment, empha- with embedding the evidence-based pro- vating champions within medical prac-
sizing referrals to the evidence-based pro- grams within their organizational struc- tices and physical therapy agencies to
grams in their communities (5). Each of ture. This entailed building a state infra- lead the process was problematic. In par-
the three states decided that instead of structure for instructor training, helping ticular, medical practices do not have

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 191 | 70
Thoreson et al. State view on falls prevention

Table 1 | Factors to successful implementation of a fall prevention program for older adults.

Examples of specific strategies employed by three state health departments (CO, OR, and NY) in implementing a fall prevention program for
older adults

Building public health State Health Department strategic plan includes a section recommending evidence-based programs for fall prevention
infrastructure for older adults
Injury Community Planning Group includes falls prevention as a priority topic
Public health toolkits for Accountable Care Organizations and Patient Centered Medical Homes include
recommendations for falls prevention interventions to meet quality standards and clinical incentive measures
Developing new Formed relationships with new partners, e.g., specific fitness centers, local parks and recreation departments,
partners community health workers, YMCAs, and home health agencies
Engaged with physician practice groups, professional associations, and health insurance companies to reach health
systems and individual physician practices
Worked with state-level professional organizations such as physical therapy association, primary care association,
pharmacy association, and state parks and recreation association to encourage their joint role in fall prevention
Developing capacity for For the Otago program, the University of North Carolina developed a web-based training for physical therapists and an
technical assistance on-line database to track Otago patients
Health department staff developed expertise on EHRs and the use of health care transformation initiatives to develop
system-wide improvements in health care
CDC developed a system to provide physicians with Maintenance of Certification and Continuing Medical Education
credits for participation in the STEADI program

Facilitating program Developed state-wide training systems to certify Stepping On and TCMBB instructors
uptake in organizations Many local parks and recreation departments added TCMBB to their regular class schedule
Developed physician and physical therapy champions who led their clinic teams in successfully implementing STEADI at
the practice level

Facilitating program Medicare-beneficiary fitness programs (Silver Sneakers and Silver and Fit) added TCMBB to their approved program list
uptake in systems in one state
Stepping On was adopted by hospital systems as a key injury prevention program for clinics and trauma centers
Stepping On was added as standard program by a Veterans Administration Medical Center

Reaching underserved Spanish-speaking health promoters and parish nurses were trained to deliver classes
populations Spanish language version of Stepping On is under development
Classes were offered at churches and senior residential housing complexes in addition to clinics and fitness centers
Small program subsidies were used to reach underserved seniors who are minorities, non-English speaking, or disabled

Evaluate programs for Data collection tools were developed to track programs
fidelity and success Clear and open communication with partners was established

a strong history of patient referral to and underserved elderly, such as minori- increase activities to penetrate the much
community-based programs, despite the ties, non-English speakers, and those larger state-wide older adult community
value of those programs being well docu- with disabilities. remains challenging. The key to success
mented in the medical and public health • Evidence-based programing and the will be to recognize fall prevention activ-
literature (6–8). need to maintain essential elements of ities as an essential service in patient care
• Implementing a clinical intervention adoption and fidelity were new con- and health promotion for older adults. The
such as Otago was challenging due to cepts to many community partners. This state health departments will continue to
Medicare billing requirements as well as is an area where state health depart- engage with community partners willing to
the lack of Otago experts in any of the ments provided technical assistance and make commitments to integrate fall pre-
funded state health departments. The direction (9). vention into their regular activities and
web-based Otago training was essential to identify sustainable sources of fund-
for training physical therapists, although ing and reimbursement to maintain these
it is still uncertain exactly how the CONCLUSION programs (8).
program is being implemented with The three states have demonstrated success
patients. in implementing evidence-based program- ACKNOWLEDGMENTS
• The states developed relationships with ing for fall prevention among older adults The state projects were funded by Grant
specific partners in order to ensure at the community level. Implementation of CE-11-1101 Core Violence and Injury
sustainable programing for vulnerable strategies to not only sustain but also to Prevention programs (Core VIPP) from

www.frontiersin.org April 2015 | Volume 2 | Article 191 | 71

Thoreson et al. State view on falls prevention

the Centers for Disease Control and 5. Centers for Disease Control and Prevention. and private organizations/agencies; however, the views,
Prevention. The contents of this arti- STEADI (Stopping Elderly Accidents, Deaths & findings, and conclusions in these articles are those of the
Injuries) Tool Kit for Health Care Providers. authors and do not necessarily represent the official posi-
cle are solely the responsibility of the
(2014). Available from: http://www.cdc.gov/ tion of these organizations/agencies. All papers published
authors and do not necessarily rep- homeandrecreationalsafety/Falls/steadi/index.html in the Research Topic received peer review from members
resent the official views of the Cen- 6. Moyer VA; U.S. Preventive Services Task Force. of the Frontiers in Public Health (Public Health Edu-
ters for Disease Control and Prevention. Prevention of falls in community-dwelling older cation and Promotion section) panel of Review Editors.
The authors would like to thank Lind- adults: U.S. Preventive Services Task Force Because this Research Topic represents work closely asso-
recommendation statement. Ann Intern Med ciated with a nationwide evidence-based movement in
sey Myers, Aerin LaCerte, and Barbara
(2012) 157:197–204. doi:10.7326/0003-4819-157- the US, many of the authors and/or Review Editors may
Gabella, CO, USA; Lisa Millet, Adrienne 9-201211060-00526 have worked together previously in some fashion. Review
Greene, and Joe Patton, OR, USA; Meaghan 7. Tinetti ME, Brach JS. Translating the fall preven- Editors were purposively selected based on their expertise
Tartaglia, Kara Burke, and Harrison Moss, tion recommendations into a covered service; can with evaluation and/or evidence-based programming
NY, USA for their contributions to the it be done, and who should do it? Ann Intern Med for older adults. Review Editors were independent of
(2012) 157(3):213–4. doi:10.7326/0003-4819-157- named authors on any given article published in this
project. 3-201208070-00014 volume.
8. Panel on Prevention of Falls in Older Persons, Amer-
REFERENCES ican Geriatrics Society and British Geriatrics Soci-
1. Bryan S, Williams AN, Porter JM, Patterson A. State ety. Summary of the updated American Geriatrics Received: 11 July 2014; accepted: 28 September 2014;
of the States: 2011 Report. Atlanta, GA: Safe States Society/British geriatrics society clinical practice published online: 27 April 2015.
Alliance (2013). guideline for prevention of falls in older persons. Citation: Thoreson SR, Shields LM, Dowler DW and
2. Centers for Disease Control and Preven- J Am Geriatr Soc (2011) 49:148–57. doi:10.1111/j. Bauer MJ (2015) Public health system perspective on
tion, National Center for Injury Prevention 1532-5415.2010.03234.x implementation of evidence-based fall prevention strate-
and Control. The Public Health Approach to 9. Wilson KM, Brady TJ, Lesesne C, on behalf of the gies for older adults. Front. Public Health 2:191. doi:
Violence Prevention (2014). Available from: NCCDPHP Work Group on Translation. An orga- 10.3389/fpubh.2014.00191
http://www.cdc.gov/violenceprevention/overview/ nizing framework for translation in public health: This article was submitted to Public Health Education
publichealthapproach.html the knowledge to action framework. Prev Chronic and Promotion, a section of the journal Frontiers in
3. Stevens JA. A CDC Compendium of Effective Fall Dis (2011) 8(2):A46. Public Health.
Interventions: What Works for Community-Dwelling Copyright © 2015 Thoreson, Shields, Dowler and Bauer.
Older Adults. 2nd ed. Atlanta, GA: Centers for Dis- Conflict of Interest Statement: The authors declare This is an open-access article distributed under the terms
ease Control and Prevention, National Center for that the research was conducted in the absence of any of the Creative Commons Attribution License (CC BY).
Injury Prevention and Control (2010). commercial or financial relationships that could be The use, distribution or reproduction in other forums is
4. Kaniewski M, Stevens JA, Parker EM, Lee construed as a potential conflict of interest. permitted, provided the original author(s) or licensor are
R. An introduction to the Centers for Dis- credited and that the original publication in this journal
ease Control and Prevention’s efforts to pre- This paper is included in the Research Topic, “Evidence- is cited, in accordance with accepted academic practice.
vent older adult falls. Front Public Health Based Programming for Older Adults.” This Research No use, distribution or reproduction is permitted which
(2015) 2:119. doi:10.3389/fpubh.2014.00119 Topic received partial funding from multiple government does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 191 | 72
 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00130

Falling for a balance partner

Sara B. May and Cherie A. Rosemond *
The University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: falling accident, fall prevention strategies, balance partner, older adults, opinion

The growing accumulation of knowledge apartment. Now prone to memory lapses, sunny, which suits me fine. We are working
about fall prevention strategies primarily Georgia had forgotten that she had diffi- with her physicians to find ways that Geor-
reflects a research perspective where inter- culty in completing the balance tests at the gia might be able to lessen the number of
vention data are rigorously collected and Building Better Balance Screening offered trips to the bathroom during the night,
analyzed. However, the voices of program in her apartment complex by the Land thereby lessening the chances of her taking
deliverers or participants who are part of of Sky Area Agency on Aging. Just a few another tumble.
these interventions are often aggregated months before Georgia went through her Although we are from diverse back-
and thus muted. With recognition of the balance screening, I had the opportunity grounds and cultures, we easily settled in
growing importance of patient-centered to attend Balance Partner training (on my to focus on our similarities, including hav-
care, we wanted to provide a personal birthday) sponsored by UNC’s Center for ing several chronic health conditions and a
reflection on the Balance Partner program, Health Promotion and Disease Prevention. desire to stay the healthiest we could for as
a CDC funded project to train peer lead- As it turned out, just when I was all set to long as possible. Georgia had given up tak-
ers in fall prevention. In the story below, find a fellow senior in need of volunteer ing baths, as she was unable to get up and
Sara was trained as a Balance Partner using assistance, Georgia was eager to reduce her out of the tub, even with grab bars on two
a curriculum developed at The Univer- risk of falling. We had a match! sides of it. I have had railings installed on
sity of North Carolina’s Center for Health Georgia and I started meeting in her short stairways on my porches and learned
Promotion and Disease Prevention. apartment once or twice a week. Since she to sit down when putting on my shoes to
The Balance Partner Program provides had moved into the apartment 3 years ear- avoid losing my balance while standing on
training and support to community vol- lier, three tumbles had undermined Geor- one foot. “I’m more aware of possible fall
unteers who are paired with a peer who gia’s confidence that she could avoid falling. causes now and am very careful using my
is at risk for falls. Together, Balance Part- She had walked with a cane for several walker,” Georgia says. She adds, “Sara has
ners plan strategies to decrease the like- years before needing the greater support of suggested I change to a closer pharmacy
lihood of a fall – strategies that could a walker. I, too, had three falls at home. so it’s easier to get my medications picked
include joining a balance exercise program, Thankfully, I was blessed with only bruises up. When Sara takes me shopping, she does
improving home safety, or getting a vision to show for the experiences. I was moti- the legwork for me, bringing me options of
check-up. By addressing social and emo- vated to help others stay upright after the items I’m looking for so I can save my
tional factors alongside knowledge about learning fall prevention techniques during energy for getting to and from her car.”
falls, the Balance Partner Program aims to physical therapy and implementing them Isolation is one of the most difficult
increase older adults’ overall adherence to around my home. I shared some of what I issues that Georgia has had to face since
fall prevention activities above the 50% rate learned with Georgia. she had to give up her driver’s license last
reported in literature (1). Sara, a 69-year- We outfitted Georgia with a night-light year. When I was asked to substitute as
old volunteer, was paired with Georgia, in her bathroom and two lightweight flash- a co-facilitator for a January session of a
who screened at high risk for falls dur- lights to help find her way around her Living Healthy with Chronic Conditions
ing a Building Better Balance Screening apartment when the electricity goes out. workshop at Georgia’s apartment complex,
in Asheville, NC, USA. As of this writing, She now keeps her apartment clutter-free, I persuaded her to come along and try it,
Sara and Georgia had worked together for avoiding throw rugs in which her walker suggesting that she sits close to the door
almost 6 months to implement strategies to might get tangled. I brought Georgia a since she was concerned that she might
reduce their falls risk. monthly wall calendar to help her keep need to leave in a hurry to go to the
track of upcoming doctor’s appointments bathroom. “I’m doing this for you,” she
A BLOSSOMING PARTNERSHIP AND and I drive her to them as often as my told me when I walked with her to the
FRIENDSHIP schedule and health permits. As a bonus, I first class, but attending the rest of them
“How did we get together?” the smiling get more exercise by helping Georgia in and has been for her. She loved the opportu-
81-year-old demure-looking lady asked me out of the car. She is only up for getting out nity to interact with other residents in a
during my Saturday visit in her Asheville when the weather is warm and the skies are small-group setting, especially since she is

www.frontiersin.org April 2015 | Volume 2 | Article 130 | 73

May and Rosemond Falling for a balance partner

hard of hearing. She has kept returning to community fall prevention interventions and private organizations/agencies; however, the views,
subsequent classes whether I am there co- suggests that social and emotional factors findings, and conclusions in these articles are those of the
authors and do not necessarily represent the official posi-
facilitating or not. In fact, that is where she such as isolation and boredom partially
tion of these organizations/agencies. All papers published
found her hall-walking buddies! explain low adherence rates (2). Making in the Research Topic received peer review from members
“I felt like it would help motivate me new friends, sharing talents, and having fun of the Frontiers in Public Health (Public Health Edu-
to have a Balance Partner,” Georgia said. are rarely reported in traditional research cation and Promotion section) panel of Review Editors.
“I want to stay out of a wheelchair and studies. This reflection reveals a personal Because this Research Topic represents work closely asso-
ciated with a nationwide evidence-based movement in
eventually go back to walking with only a account of why a falls prevention interven-
the US, many of the authors and/or Review Editors may
cane.” Little did we anticipate at our first tion effort based on supportive interactions have worked together previously in some fashion. Review
meeting that we would become fast friends. can be so powerful. Editors were purposively selected based on their expertise
Georgia advised me on a good hairstyle to with evaluation and/or evidence-based programming
make my hearing aid less visible to oth- for older adults. Review Editors were independent of
REFERENCES named authors on any given article published in this
ers, and I helped her pick out a new wig, 1. Nyman SR, Victor CR. Older people’s participa- volume.
which makes Georgia look picture-perfect. tion in and engagement with falls prevention inter-
Our outing to the wig shop even resulted ventions in community settings: an augment to
the cochrane systematic review. Age Ageing (2012) Received: 01 July 2014; accepted: 15 August 2014;
in her getting to relive old times with the 41(1):16–23. doi:10.1093/ageing/afr103 published online: 27 April 2015.
wig shop owner. When Georgia calls me 2. Dickinson A, Machen I, Horton K, Jain D, Mad- Citation: May SB and Rosemond CA (2015) Falling
to see how I am doing when she knows dex T, Cove J. Fall prevention in the community: for a balance partner. Front. Public Health 2:130. doi:
that I am sick and could use some cheering what older people say they need. Br J Community 10.3389/fpubh.2014.00130
up, it feels like the frosting on the cake of Nurs (2011) 16(4):174–80. doi:10.12968/bjcn.2011. This article was submitted to Public Health Education
16.4.174 and Promotion, a section of the journal Frontiers in
our relationship. Even though she prefers Public Health.
salty to sugary foods and I am the opposite, Copyright © 2015 May and Rosemond. This is an open-
we continue to enrich each other’s lives as Conflict of Interest Statement: The authors declare access article distributed under the terms of the Creative
we explore this “not for sissies” thing called that the research was conducted in the absence of any Commons Attribution License (CC BY). The use, dis-
commercial or financial relationships that could be tribution or reproduction in other forums is permitted,
aging in place.
construed as a potential conflict of interest. provided the original author(s) or licensor are credited
and that the original publication in this journal is cited,
CONCLUSION This paper is included in the Research Topic, “Evidence- in accordance with accepted academic practice. No use,
Helping others is often the pathway Based Programming for Older Adults.” This Research distribution or reproduction is permitted which does not
to helping oneself. Current research on Topic received partial funding from multiple government comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 130 | 74
 PERSPECTIVE ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00206

Setting the stage: measure selection, coordination, and

data collection for a national self-management initiative
Kristie P. Kulinski 1 *, Michele Boutaugh 2 , Matthew Lee Smith 3 , Marcia G. Ory 4 and Kate Lorig 5
1
National Council on Aging, Washington, DC, USA
2
Administration on Aging, Administration for Community Living, Atlanta, GA, USA
3
Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA, USA
4
Department of Health Promotion and Community Health Sciences, School of Public Health, Texas A&M Health Science Center, College Station, TX, USA
5
Stanford Patient Education Research Center, Department of Medicine, Stanford School of Medicine, Palo Alto, CA, USA

Edited by: This paper describes the history and rationale behind the development of a centralized
Will R. Ross, Washington University
data collection system for the national rollout of the Chronic Disease Self-Management
School of Medicine, USA
Program (CDSMP) through the American Recovery and Reinvestment Act of 2009 Com-
Reviewed by:
Lalit Raghunath Sankhe, Grant munities Putting Prevention to Work: CDSMP initiative. In addition to justifying the need
Government Medical College, India for solutions to the burgeoning burden of chronic disease in the United States, this paper
Lynda Anderson, Centers for Disease provides details about CDSMP and related self-management education programs, includ-
Control and Prevention, USA
ing their structure, facilitator training, and effectiveness. These topics set the stage for
*Correspondence:
the processes and procedures to create and manage the database for use at the national,
Kristie P. Kulinski , National Council on
Aging, 1901 L Street NW, 4th Floor, state, and local levels. Furthermore, this paper describes the processes related to selecting
Washington, DC 20036, USA variables, coordinating data collection, and utilizing data to inform research and policy.
e-mail: [email protected]
Keywords: chronic disease self-management, evidence-based program, data collection, intervention planning

RATIONALE the most common chronic conditions include high blood pres-
As more and more evidence-based interventions are being funded sure (58%), high cholesterol (45%), heart disease (31%), arthritis
as multi-site programs by both federal agencies and major foun- (29%), and diabetes (28%) (3).
dations, there is a growing need for uniform measures and pro- Older adults with chronic conditions face a number of barriers
tocols as well as centralized data collection systems. This paper in terms of coping with their illness and optimizing their health,
describes one such data system developed in response to the which include lack of social support, low skill levels for symptom
national rollout of the Chronic Disease Self-Management Pro- management, and low confidence in their abilities to manage their
gram (CDSMP) through the American Recovery and Reinvest- conditions (self-efficacy) (4). Increasingly, SM is being heralded as
ment Act of 2009 Communities Putting Prevention to Work: a key component in the improvement of health outcomes associ-
CDSMP initiative (ARRA CDSMP) (1). In addition to describ- ated with chronic disease. According to the Institute of Medicine,
ing the creation and management of the database used in this SM is defined as “the tasks that individuals must undertake to live
effort, we will discuss its uses at the national, state, and local levels well with one or more chronic conditions.” (5) Research demon-
as well as its utility for informing policy and research. Addition- strates the positive impact of SM programs on these tasks, which
ally, this paper will provide the necessary background for those include having the confidence to deal with the medical, role, and
wanting to understand the rationale behind this national initia- emotional management of their conditions (5, 6).
tive in terms of the burden of chronic conditions among older
Americans and self-management (SM) as a core requirement for THE CHRONIC DISEASE SELF-MANAGEMENT PROGRAM
dealing with such conditions. As a case example, the processes The CDSMP is perhaps the best known SM intervention (7). It was
related to selecting variables, developing a centralized data col- developed at Stanford University and is a peer-led, community-
lection system, training, and managing data will be described based intervention that helps individuals with chronic conditions
for this grand-scale translational rollout of an evidence-based learn skills and gain the confidence to manage and improve their
program. health (7). The program focuses on challenges that are common
to individuals living with any chronic condition, such as problem
BACKGROUND solving, decision making, symptom management, nutrition, exer-
BURDEN OF CHRONIC CONDITIONS cise, medication use, emotions, and communicating with health
Chronic conditions have become endemic in the United States, care professionals. In addition to the standard CDSMP, Stanford
with older adults bearing the greatest burden. Approximately 36% offers a comprehensive suite of chronic disease self-management
of adults age 18–34 have a chronic condition, compared to nearly education (CDSME) programs, with disease-specific variants for
92% in the population aged 65 and over (2). This same trend people living with diabetes, chronic pain, HIV/AIDS, cancer sur-
is observed with regard to multiple chronic conditions, with a vivors, and arthritis. Most of these also have culturally appropriate
range of 14% among the population aged 18–34 to nearly 77% Spanish versions. The programs are available in over 30 countries
in the older adult population (2). Among Medicare beneficiaries, and 25 languages.

www.frontiersin.org April 2015 | Volume 2 | Article 206 | 75

Kulinski et al. Setting the stage

Led by a pair of trained facilitators, many of whom also Intervention effectiveness

have chronic health conditions, these small, highly interactive The Chronic Disease Self-Management Program has been exten-
workshops meet once a week for six consecutive weeks. Dur- sively evaluated through randomized controlled trials (9, 10).
ing each 2.5-hour session, 10–15 participants focus on building Workshop participants experience significant improvements
the skills they need to manage their conditions. Fostering par- across several domains, including physical activity, symptom man-
ticipant self-efficacy is at the core of the intervention, achieved agement, communication with physicians, and general health.
through techniques such as skills mastery, peer modeling, rein- Additionally, the original research demonstrated that CDSMP par-
terpretation of physiological symptoms, and social persuasion. ticipants spend fewer days in the hospital, as well as a trend toward
Workshops are highly participative, with mutual success and sup- fewer outpatient visits and hospitalizations (10).
port building participants’ confidence in their ability to manage Further cementing the value of CDSMP, the program has been
their health and maintain active, fulfilling lives. Participants cre- successfully translated for implementation in a variety of commu-
ate a weekly action plan and try new behaviors such as exercise nity settings worldwide, with participants reporting results similar
monitoring. Each session includes an opportunity for feedback to the original research. A recent United States-based National
about progress and discussion of challenges. Table 1 provides an Study of CDSMP encompassed over 1000 participants drawn from
overview of the topics and activities covered during each workshop 145 workshops in 17 states (11). Sociodemographic, health status,
session. and behavioral data were collected at baseline, 6, and 12 months,
yielding a number of positive, significant improvements (12, 13).
Workshop facilitator training and infrastructure When aligned with the Institute for Healthcare Improvement’s
The program uses a train-the-trainer model consisting of Lay Triple Aim (13, 14), the following results are particularly note-
Leaders, Master Trainers, and T-Trainers (8). Lay Leaders can facil- worthy: better health – improvement in self-reported health, less
itate CDSMP workshops but cannot train others. They complete depression, and better quality of life; better care – improved com-
a structured training and must facilitate at least one workshop in munication with physicians, medication compliance, and health
the following year. Master Trainers can facilitate CDSMP work- literacy; and lower health cost – more than $360 per person net sav-
shops as well as train new Lay Leaders. As with CDSMP Lay ings after factoring in program costs (15). In addition to improving
Leaders, Master Trainers participate in a systematized training. participant health and decreasing health care costs, the outcomes
After training, they must facilitate at least two CDSMP work- of this national study reinforce that CDSMP has been effectively
shops within one year and conduct a Lay Leader training within translated from research to practice throughout the country.
18 months. Finally, T-Trainers are authorized to facilitate work-
shops, train new Lay Leaders, and train new Master Trainers. This NATIONAL INITIATIVES SUPPORTING CDSMP
role involves the completion of an apprenticeship with a Stan- IMPLEMENTATION
ford staff T-Trainer. Additionally, they must have facilitated at Over the past decade, community-based implementation of
least three Lay Leader trainings prior to their apprenticeship, co- CDSMP and its variants have received broad support through
lead a Master Trainer training within 12 months of completing funding from federal agencies [e.g., the Administration on Aging
the apprenticeship, and conduct a Master Trainer training every (AoA), a program division within the Administration for Commu-
two years. nity Living, and the Centers for Disease Control and Prevention

Table 1 | CDSMP workshop overview by session.

Week 1 Week 2 Week 3 Week 4 Week 5 Week 6

Overview of self-management and chronic health conditions X

Making an action plan X X X X X X
Relaxation/cognitive symptom management X X X X X
Feedback/problem solving X X X X X
Difficult emotions X X
Fitness/exercise X X
Better breathing X
Fatigue X
Eating well X
Advance directives X
Communication X
Medications X
Making treatment decisions X
Depression X
Informing the health care team X
Working with your health care professional X
Future plans X

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 206 | 76
Kulinski et al. Setting the stage

(CDC)], foundations (e.g., Atlantic Philanthropies, Archstone grant program, financed by the Prevention and Public Health
Foundation, Robert Wood Johnson Foundation, and Health Foun- Fund (PPHF), provides support to 22 states. Both the ARRA
dation of South Florida), and health care providers (e.g., Kaiser CDSMP and the PPHF grant programs have focused on not only
Permanente, Group Health Cooperative, and Dignity Health). chronic disease SM programs, including the generic CDSMP, but
Specific to the aging services network, AoA has supported states also programs developed for specific chronic conditions (arthri-
and community organizations in their efforts to develop infra- tis, diabetes, HIV/AIDs, and chronic pain), for Spanish-speaking
structure, workforce, and capacity to deliver CDSMP and other cultures, and in an online format. Table 2A highlights AoA
evidence-based programs. Since 2006, AoA has provided three funding history, although as previously noted a number of fed-
major competitive grant programs to states to support dissemina- eral and other sources of funding have also supported these
tion of evidence-based programs. The 2006–2012 Evidence-Based programs.
Disease and Disability Prevention Program (EBDDP) grants were
awarded to 24 states to support dissemination of CDSMP and DATA COLLECTION
evidence-based physical activity, fall prevention, nutrition, and SELECTING STANDARDIZED MEASURES
behavioral health programs. The national program infrastruc- The collection of standardized performance monitoring data has
ture was greatly expanded with the 2010–2013 American Recovery been a critical component of each of the aforementioned AoA ini-
and Reinvestment Act Communities Putting Prevention to Work: tiatives. While the specific measures collected have evolved over
CDSMP (ARRA CDSMP) grants awarded to 45 states, the Dis- time, the data collected by AoA grantees and their partners can
trict of Columbia, and Puerto Rico. The administration’s current be grouped within four categories: (1) workshop information;
2012–2015 Empowering Older Adults and Adults with Disabili- (2) participant information; (3) attendance; and (4) organization
ties through Chronic Disease Self-Management Education Programs data. The current standardized measures, which were approved

Table 2 | Support, data requirements, and rationale for AoA-CDSMP initiatives.

A: EVOLUTION OF AoA-SUPPORTED CDSMP INITIATIVES

Year Initiative Reach

2003 Evidence-Based Prevention Program for the Elderly Model Communities Project 14 Communities
2006 Evidence-Based Disease Prevention and Disability Program 16 States
2006–2007 Evidence-Based Disease Prevention and Disability Program (funding made available to 24 27 States
states by AoA, plus three states funded by Atlantic Philanthropies)
2010 American Recovery and Reinvestment Act Communities Putting Prevention to Work: 47 States/Territories
Chronic Disease Self-Management Program
2012 2012 Prevention and Public Health Funds: Empowering Older Adults and Adults with 22 States
Disabilities through Chronic Disease Self-Management Education Programs

B: AoA GRANTEE DATA COLLECTION REQUIREMENTS

Data type Elements collected

Workshop Information Host organization and implementation site name/location, workshop leader names, workshop start/end dates, use of
orientation session, workshop type, workshop language
Participant Information Date of birth, ZIP Code, sex, race, ethnicity, chronic conditions, caregiver status, disability status, number of people in
household, education level
Attendance Sessions attended by participant
Organization Data Organization type with regard to host organization and implementation site (list includes area agency on aging, county
health department, health care organization, faith-based organization, workplace, residential facility, and library)

C: RATIONALE FOR SELECTING DATA TYPES

Data type Rationale

Workshop Information Map delivery infrastructure, identify type of workshop offered, identify diversity of languages, monitor start/end dates
and number of workshop leaders as proxies for fidelity
Participant Information Accurately describe participant population, ensure adequate reach to target population, monitor demographic elements
that serve as proxies for health status and vulnerability (race/ethnicity, chronic conditions, caregiver status, disability
status, education level, etc.)
Attendance Track number of sessions attended by participant to determine completer status, identify organization and state
successes/challenges with participant retention
Organization Data Identify types of organizations involved in program delivery, monitor increase in delivery capacity and geographic reach

www.frontiersin.org April 2015 | Volume 2 | Article 206 | 77

Kulinski et al. Setting the stage

through the Office of Management and Budget (OMB) Paperwork receiving the forms is generally two weeks after the conclusion
Reduction Act, are listed in Table 2B. of a workshop. While the structure for data entry varies by
state, typically either (a) all data entry take place at the state
RATIONALE FOR UNIFORM DATA COLLECTION AND MONITORING level or (b) responsibility for data entry is divided regionally,
Data elements for the ARRA CDSMP initiative were carefully and with staff from selected organizations entering data on behalf
purposively chosen with the intent of balancing the critical need of their peers. Decisions as to which model to use are gen-
to monitor program operations and participant accrual with the erally based on adequacy of staffing for data entry/monitoring
desire to minimize data collection and reporting burden on pro- (e.g., is there sufficient staff time at a state level to devote to
gram deliverers and participants (see Table 2C). Considering the this task and keep up with demand, or does this task need
myriad studies reinforcing the effectiveness of the program in the to be parsed out regionally?). An additional consideration is
community when delivered with fidelity to the original model (6, the overall program management model, as some states cen-
12), emphasis was placed on collecting reach data versus addi- tralize program management at the state level, whereas others
tional outcome data. Moreover, grantees and their partners were take a decentralized approach with each region acting semi-
encouraged to invest their limited resources in program deliv- autonomously.
ery, infrastructure, and sustainability to ensure ongoing access to
CDSMP as opposed to engaging in costly outcomes measurement. DEVELOPING THE NATIONAL, ONLINE DATABASE
At the federal level, there is a strong emphasis on accountability Prior to the ARRA CDSMP initiative, data were collected via paper
and transparency to ensure that funds are being spent properly forms, which were mailed to a centralized location and entered into
and the desired reach and impact are being achieved. Therefore, an Excel spreadsheet. This spreadsheet was sent semi-annually by
the uniform collection of appropriate measures ensures that due each grantee to a central repository, where the data were cleaned,
diligence is performed in this regard. For example, an overar- analyzed, and shared back with AoA and their respective grantees.
ching goal of the ARRA CDSMP initiative was to reach 50,000 This system was rather burdensome with data transfers from mul-
program completers (i.e., those participants who attend four or tiple partners. There was also a considerable time lag that occurred
more of the six workshop sessions), with a particular emphasis on between the submission of semi-annual data and the receipt of the
engaging vulnerable and disadvantaged older adults (16). Because analyzed data.
detailed attendance information was collected on each participant, In 2010, with the advent of the ARRA CDSMP initiative and
it was easy to determine how many completers were reached. This the major expansion, a more efficient data system was needed for
attendance information was especially important because out- tracking the national rollout of CDSMP and assessing whether ini-
come measures were not collected in this initiative, thus workshop tiative goals were being met. Thus, the National Council on Aging
attendance served as a proxy variable indicating that participants (NCOA), the designated resource center for chronic disease SM
received an adequate intervention dose. Additionally, because out- education programs, developed an online national database. A cus-
comes were not directly measured, demographic variables such as tom application was developed by NCOA on the Salesforce.com
date of birth, living alone status, racial and ethnic status, educa- platform expressly for this purpose. Salesforce.com was selected
tion level, and number of chronic conditions served as proxies for reasons that include NCOA’s experience developing other
for health status and vulnerability. Collecting this participant data collection systems on the platform, cost-efficiency, web-based
and attendance information was deemed important for inform- access, and data security.
ing national and state leadership as to whether or not the target Presently, the database is available as a free resource for all
population was being reached/served by intervention workshops. states implementing these programs, regardless of funding source.
Upon request, users receive login information from NCOA, and
DATA COLLECTION TOOLS, COORDINATION, AND PROCESSES can then enter workshop and participant data from any computer
Timely and efficient collection and reporting of programmatic with Internet access. No software is required. All data are available
data are critical to ensure the success and value of the national data- in real-time, and data from any of the suite of in-person Stan-
base. The following OMB-approved data collection forms accom- ford University CDSME programs can be entered. The data are
pany the national database: (1) Workshop Information Cover stored securely and are de-identified at the participant level. The
Sheet; (2) Attendance Log; (3) Participant Information Survey; database does not contain participant names. Each participant is
and (4) Organization Data Form. assigned a random unique identifier and is linked to a workshop
At a local level, workshop leaders complete the Workshop though a separate unique identifier. Database users must view a
Information Cover Sheet. They are also responsible for using recorded training webinar prior to accessing the system. Technical
the Attendance Log to track participation at each session. Dur- support related to database utilization and data entry is available
ing the first session (or an orientation session, if applicable), via NCOA. Regular quality control activities, such as identifying
workshop leaders distribute a Participant Information Survey to erroneous duplicate workshops, are performed by NCOA and its
each participant. The completion of these brief, 10-item sur- database management partner, Senior Services. In addition, other
veys is optional and is not required for workshop participation. quality measures are built into the system, such as prompting users
Completed surveys are collected from the participants and sent to review workshop records with issues of concern such as work-
along with the Workshop Information Cover Sheet and Atten- shop start and end dates that are fewer than 6 weeks apart and
dance Log to a person responsible for entering the informa- participant ages that are younger than 18 years (the minimum
tion into the national database. Expected turnaround time for recommended participant age).

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 206 | 78
Kulinski et al. Setting the stage

In addition to the ability to easily enter workshop and par- CONCLUSION

ticipant data, users have access to a variety of standard reports Data collection for CDSMP and the suite of other Stanford Univer-
to inform program management and enhance quality assurance. sity CDSME programs has been essential to nationwide program
These reports can be filtered by elements such as date, county, and success and sustainability. With the inception of this database,
host organization and offer a comparison to state and national states and community-based organizations offering CDSME had
data. Beyond the standard reports that are available to all data- immediate, real-time access to their workshop and participant data
base users, NCOA staff can develop custom reports in response to for the first time. This proved to be an incredible value-add for the
information and data requests from AoA leadership and other key network. Not only does the data highlight program reach and
stakeholders. inform program planning, it is also critical in terms of attaining
additional resources to support implementation and infrastruc-
UTILITY OF COLLECTED DATA ture at national, state, and local levels. It is evident that the benefit
The NCOA and AoA, as well as other program funders and stake- gained from a national data collection system is certainly worth
holders, use the information from the various data collection tools the investment in development, training, and maintenance. Future
for numerous reasons, including to (1) comply with reporting grand-scale initiatives delivering evidence-based programs are
requirements mandated by the authorizing statutes; (2) collect encouraged to use this ARRA CDSMP experience when creating
data for performance measures used in the justification of the data collection and monitoring systems.
budget to Congress and by program, state, and national decision
makers; (3) effectively manage the program at the federal, state,
and local levels; (4) identify program implementation issues and REFERENCES
1. U.S. Department of Health and Human Services, Administration on Aging.
pinpoint areas for technical assistance activities; (5) identify best
ARRA – Communities Putting Prevention to Work: Chronic Disease Self-
practices in program implementation and building sustainable Management Program 2012. (2012). Available from: www.cfda.gov/index?s=
program delivery systems and to develop resources to enable cur- program&mode=form&tab=step1&id=5469a61f2c5f25cf3984fc3b94051b5f
rent and future program implementers to learn from and replicate 2. Machlin S, Cohen J, Beauregard K. Health Care Expenses for Adults with Chronic
these practices; and (6) provide information for reports to Con- Conditions, 2005. Statistical Brief #203. Rockville, MD: Agency for Healthcare
Research and Quality (2008).
gress, other governmental agencies, stakeholders, and to the public 3. Centers for Medicare and Medicaid Services. Chronic conditions among Medicare
about grantee progress. Beneficiaries. 2012 ed. Baltimore, MD: Chartbook (2012).
The uniform collection of these data elements using a coordi- 4. Bayliss EA, Ellis JL, Steiner JF. Barriers to self-management and quality of life
nated online system has great practicality and utility for reporting outcomes in seniors with multimorbidities. Ann Fam Med (2007) 5(5):395–402.
and providing real-time monitoring and feedback. Using these doi:10.1370/afm.722
5. National Research Council. 1st Annual Crossing the Quality Chasm Summit: A
data, AoA can perform state-based performance comparisons Focus on Communities. Washington, DC: The National Academies Press (2004).
related to delivery site engagement, participant reach, participant 6. Brady TJ, Murphy L, Beauchesne D, et al. Sorting through the Evidence for the
retention, and program embedment/sustainability. These data Arthritis Self-Management Program and the Chronic Disease Self-Management
can also enable program fidelity assessments to rapidly identify Program: Executive Summary of ASMP/CDSMP Meta-Analysis. Atlanta, GA:
Centers for Disease Control and Prevention (2011). Available from: http:
technical assistance needs and/or correct program drift.
//www.cdc.gov/arthritis/docs/asmp-executive-summary.pdf
Furthermore, these data can be used to develop webinars and 7. Chronic Disease Self-Management Program. (2014). Available from: http:
other resources for state grantees and their partners (for the //patienteducation.stanford.edu/programs/cdsmp.html
purposes of training, technical assistance, and/or strategic devel- 8. Stanford School of Medicine. Training Policies for Stanford Self-Management Pro-
opment) as well as generate standard and customized reports grams. (2014). Available from: http://patienteducation.stanford.edu/training/
trnpolicies.html
so grantees can identify local successes and opportunities for
9. Lorig KR, Sobel DS, Stewart AL, Brown BW Jr, Bandura A, Ritter P, et al. Evidence
improvement. More specifically, these reports have been (and con- suggesting that a chronic disease self-management program can improve health
tinue to be) used for quality control (e.g., identifying workshops status while reducing hospitalization: a randomized trial. Med Care (1999)
that offered a “session zero,” or introductory orientation session, to 37(1):5–14. doi:10.1097/00005650-199901000-00003
determine impact on participant retention), planning (e.g., iden- 10. Lorig KR, Ritter P, Stewart AL, Sobel DS, Brown BW Jr, Bandura A, et al. Chronic
disease self-management program: 2-year health status and health care uti-
tifying host organizations that are categorized as faith-based when lization outcomes. Med Care (2001) 39(11):1217–23. doi:10.1097/00005650-
looking to engage additional partners of this same type), and 200111000-00008
reach to a specific population (e.g., number of African American 11. National Study of the Chronic Disease Self-Management Program. A Brief
participants who indicate a diabetes diagnosis). Overview. Washington, DC: National Council on Aging (2013).
Beyond these data uses at a federal/national level, this uni- 12. Ory MG, Ahn S, Jiang L, Lorig K, Ritter P, Laurent DD, et al. National study of
the chronic disease self-management program: six month outcome findings. J
formly collected data also provide great benefit to state grantees Aging Health (2013) 25(7):1258–74. doi:10.1177/0898264313502531
and their partners. These data serve to inform key stakeholders 13. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, Whitelaw N, et al. Successes
about progress and challenges, guide quality control and assur- of a national study of the chronic disease self-management program: meet-
ance efforts and forward planning, and help justify the need ing the triple aim of health care reform. Med Care (2013) 51(11):992–8.
doi:10.1097/MLR.0b013e3182a95dd1
for, as well as attain, additional funding sources (through grant
14. Berwick DM, Nolan TW, Whittington J. The triple aim: care, health, and cost.
applications or other mechanisms). Furthermore, researchers have Health Aff (2008) 27:759–69. doi:10.1377/hlthaff.27.3.759
utilized these data to address a variety of topics, including program 15. Ahn S, Basu R, Smith ML, Jiang L, Lorig K, Whitelaw N, et al. The impact
participation of older adults with diabetes (17). of chronic disease self-management programs: healthcare savings through a

www.frontiersin.org April 2015 | Volume 2 | Article 206 | 79

Kulinski et al. Setting the stage

community-based intervention. BMC Public Health (2013) 13(1):1141. doi:10. and Promotion section) panel of Review Editors. Because this Research Topic repre-
1186/1471-2458-13-1141 sents work closely associated with a nationwide evidence-based movement in the US,
16. U.S. Administration on Aging. American Recovery and Reinvestment Act Com- many of the authors and/or Review Editors may have worked together previously in
munities Putting Prevention to Work Chronic Disease Self-Management Program some fashion. Review Editors were purposively selected based on their expertise with
Announcement. (2010). Available from: http://www.aoa.gov/AoARoot/Grants/ evaluation and/or evidence-based programming for older adults. Review Editors were
Funding/2010.aspx independent of named authors on any given article published in this volume.
17. Erdem E, Korda H. Self-management program participation by older adults
with diabetes. Fam Community Health (2014) 37(2):134–46. doi:10.1097/FCH. Received: 01 September 2014; accepted: 07 October 2014; published online: 27 April
0000000000000025 2015.
Citation: Kulinski KP, Boutaugh M, Smith ML, Ory MG and Lorig K (2015) Set-
Conflict of Interest Statement: The authors declare that the research was conducted ting the stage: measure selection, coordination, and data collection for a national
in the absence of any commercial or financial relationships that could be construed self-management initiative. Front. Public Health 2:206. doi: 10.3389/fpubh.2014.00206
as a potential conflict of interest. This article was submitted to Public Health Education and Promotion, a section of the
journal Frontiers in Public Health.
This paper is included in the Research Topic, “Evidence-Based Programming for Older Copyright © 2015 Kulinski, Boutaugh, Smith, Ory and Lorig . This is an open-access
Adults.” This Research Topic received partial funding from multiple government and article distributed under the terms of the Creative Commons Attribution License (CC
private organizations/agencies; however, the views, findings, and conclusions in these BY). The use, distribution or reproduction in other forums is permitted, provided the
articles are those of the authors and do not necessarily represent the official position original author(s) or licensor are credited and that the original publication in this
of these organizations/agencies. All papers published in the Research Topic received journal is cited, in accordance with accepted academic practice. No use, distribution or
peer review from members of the Frontiers in Public Health (Public Health Education reproduction is permitted which does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 206 | 80
 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00227

National dissemination of chronic disease

self-management education programs: an incremental
examination of delivery characteristics
Matthew Lee Smith 1 *, Marcia G. Ory 2 , SangNam Ahn 2,3 , Kristie P. Kulinski 4 , Luohua Jiang 5 , Scott Horel 2
and Kate Lorig 6
1
Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA, USA
2
Department of Health Promotion and Community Health Sciences, Texas A&M Health Science Center, School of Public Health, College Station, TX, USA
3
Division of Health Systems Management and Policy, School of Public Health, The University of Memphis, Memphis, TN, USA
4
National Council on Aging, Washington, DC, USA
5
Department of Epidemiology and Biostatistics, Texas A&M Health Science Center, School of Public Health, College Station, TX, USA
6
Department of Medicine, Stanford Patient Education Research Center, Stanford School of Medicine, Palo Alto, CA, USA

Edited by: With a near 20-year developmental history as an evidence-based program, the suite of
Sanjay P. Zodpey, Public Health
Chronic Disease Self-Management Education (CDSME) programs were selected in 2010
Foundation of India, India
for grand-scale dissemination in a federally supported initiative to improve the health of
Reviewed by:
Aida Mujkic, University of Zagreb, older Americans. The primary charge of this national effort was to establish a sustain-
Croatia able program delivery system for empowering American adults with one or more chronic
Ronit Elk, University of South conditions to better manage their health. The current study focused on a series of dis-
Carolina, USA
semination and implementation science research questions to: (1) examine the geographic
*Correspondence:
distribution of participation in this initiative across the Unites States; (2) describe workshop
Matthew Lee Smith, Department of
Health Promotion and Behavior, characteristics engaged to reach program participants in various settings; and (3) describe
College of Public Health, The personal characteristics of the first 100,000 participants. Each subsequent entering cohort
University of Georgia, 330 River was descriptively examined to indicate whether there was constancy or change in delivery
Road, 315 Ramsey Center, Athens,
sites and populations reached over time. Findings show a strengthening of the workshop
GA 30602, USA
e-mail: [email protected] delivery infrastructure in that it took 9.4 months to reach the first 25,000 participants in 853
counties compared to 5.4 months to reach the last 25,000 participants in 1,109 counties.
The workshop delivery characteristics and participant characteristics remained relatively
consistent across increments of 25,000 participants reached, although general trends were
observed for some variables. For example, after reaching the first 25,000 participants, sub-
sequent groups of 25,000 participants were reached more quickly. Additionally, workshops
were increasingly delivered in ZIP Codes with higher percentages of families residing below
the federal poverty line. As more participants were reached, more participants with chronic
conditions were enrolled. This national translational study illustrates the rapid expansion of
CDSME programs throughout the United States and capability to reach diverse populations
in a variety of settings.
Keywords: chronic disease self-management, evidence-based program, older adults, sustainability, program
implementation, program reach, evaluation

INTRODUCTION health status, modify their health behaviors, and reduce their
Seen as a critical part of primary care for the past 20 years (1, healthcare utilization and associated costs (6–9). The interac-
2), disease self-management programs have been associated with tive workshop sessions are designed to enhance three types of
a plethora of positive health outcomes among middle-aged and skills necessary for the everyday management of chronic condi-
older adults in the United States (3). While the healthcare system tions: medical management, emotional management, and social
is increasingly expected to provide chronic care (1), chronic dis- role management (6). While CDSMP remains the flagship pro-
ease self-management initiatives outside of the physician’s office gram, Stanford has translated it to be delivered online, in multi-
are now widely recognized as an effective complement to improve ple languages, and for specific diseases/conditions (e.g., diabetes,
health indicators and quality of life while reducing overall health- arthritis, chronic pain, HIV) (5). This collection of interven-
related complications and associated costs (4). One of the most tions (including CDSMP) comprises the suite of Chronic Disease
extensively tested programs, the Stanford Chronic Disease Self- Self-Management Education (CDSME) programs.
Management Program (CDSMP), is a 6-week program (5) that has Building on a nascent evidence-based prevention initiative sup-
strong evidence demonstrating its ability to improve participants’ ported by the U.S. Administration on Aging (AoA) beginning in

www.frontiersin.org April 2015 | Volume 2 | Article 227 | 81

Smith et al. CDSME program growth

2003 (10), funding was provided as part of the American Recovery Program initiative (12). Workshops were delivered in 45 states,
and Reinvestment Act of 2009 (ARRA) to disseminate CDSME Puerto Rico, and the District of Columbia (11). With support
programs in 45 states, Puerto Rico, and the District of Colum- from AoA, a centralized online data system was developed by
bia between 2010 and 2012 (11). Given the solid evidence base the National Council on Aging to collect data from participating
behind CDSMP, this jointly administered initiative of the AoA, the organizations (15). Each state identified several database users at
Centers for Disease Control and Prevention (CDC), and the Cen- the state- and/or regional-level who were responsible for entering
ters for Medicare and Medicaid Services (CMS) sought to bring workshop and participant data.
these evidence-based programs to scale for the important goal of
addressing the rapidly rising number of older adults struggling to MEASURES
manage their chronic conditions. The national goal of this ARRA- In recognition of the importance of minimizing assessment bur-
funded initiative was to reach at least 50,000 program completers den, the data collection effort was limited to a short informational
(i.e., attend four or more of the six workshops sessions). Each par- sheet about the delivery organization to be filled out by the deliv-
ticipating state and entity was assigned a target goal for program ery sites; a brief set of items describing participant characteristics
completers based on the size of its population of older Amer- such as age, sex, race/ethnicity, number and type of self-reported
icans. There was an expectation that certain delivery site types chronic conditions, living arrangements and ZIP Code (for par-
would be utilized (e.g., senior centers, healthcare organizations, ticipant residence and delivery site location); and attendance logs
residential facilities, educational institutions, faith-based organi- to document the specific sessions attended by each participant.
zations, and tribal centers), and special emphasis was placed on While the expectation was that each organization would collect all
recruiting and enrolling racial/minority and other underserved the data referenced above, due to privacy and other concerns at
populations. some locations, all of the data elements were not collected at all of
The goal of having over 50,000 adults complete CDSME pro- the sites (15). Further, because the completion of the participant
gram workshops was accomplished within the first 24 months of questionnaire was not a pre-requisite for attending the workshop,
this initiative across more than 1,000 United States counties (12). some delivery sites chose not to collect all data points and some
This accomplishment demonstrates the feasibility of a coordi- participants elected not to complete the questionnaire. However,
nated effort with the aging services network, the public health, to be counted as a“successful” completer (i.e., attending four of the
and healthcare sectors. This study examined participant accrual six offered workshop sessions), the workshop information sheet
of the first 100,000 participants enrolled in this national CDSME and attendance roster was required.
program roll out in four blocks (i.e., each representing 25,000
participants). Using this frame of progressing accrual blocks, the ANALYSES
purposes of this study were to: (1) examine the geographic distri- Statistical analyses for this study were performed using SPSS (ver-
bution of participation in this initiative across the Unites States; sion 21). Workshop and participant characteristics were compared
(2) describe workshop characteristics engaged to reach program between the first, second, third, and fourth group of 25,000 par-
participants in various settings; and (3) describe personal charac- ticipants reached. Additionally, maps were generated to illustrate
teristics of the first 100,000 participants. Each subsequent entering the cumulative geospatial distribution and accruement of CDSME
cohort was descriptively examined to indicate whether there was program participants and workshops for the first 25,000 partici-
constancy or change in delivery sites and populations reached pants, 50,000 participants, 75,000 participants, and all 100,000
over time. participants. Plots indicate workshop locations. Shading indicates
the number of participants reached in each state (i.e., darker shade
MATERIALS AND METHODS represents more participants reached). Hash markings represent
CHRONIC DISEASE SELF-MANAGEMENT EDUCATION (CDSME) the first year in which funding was received by state.
PROGRAMS
As described previously, CDSMP falls within a suite of CDSME RESULTS
programs that have been widely disseminated in the U.S. as a NATIONAL CDSME PROGRAM UPTAKE
method to empower patients with self-management skills to deal Figure 1 depicts the cumulative geospatial distribution of the first
with their chronic conditions (12, 13). Drawing upon Social 100,000 CDSME program enrollees by increments of 25,000 par-
Learning Theory (14), CDSMP is an evidence-based, peer-led ticipants. As can be seen, the first 25,000 participants were reached
intervention consisting of six highly participative classes held for by 2,226 workshops in 1,705 unique implementation sites over a
2.5 h each, once a week, for six consecutive weeks (13). Addi- 9.4-month period across 853 counties. At this stage in the inter-
tional details about the theory behind CDSME programs and their vention, only a few states had reached over 1,000 participants.
implementation can be found elsewhere (15). Comparatively, the last 25,000 participants were reached by 2,154
workshops in 1,769 unique implementation sites over a 5.4-month
DATA SOURCE AND STUDY POPULATION period across 1,109 counties. At this stage in the intervention, only
This study reports findings based on cross-sectional data collected a few states had not reached over 1,000 participants.
from the first 100,000 participants enrolled in the nationwide
delivery of CDSME programs as part of the American Recovery CDSME PROGRAM WORKSHOP CHARACTERISTICS
and Reinvestment Act of 2009 (i.e., Recovery Act) Communi- Table 1 presents workshop characteristics for the first 100,000
ties Putting Prevention to Work: Chronic Disease Self-Management CDSME program participants enrolled in the intervention. These

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 227 | 82
Smith et al. CDSME program growth

FIGURE 1 | Geospatial distribution of CDSMP workshops and participants by increments of 25,000 participants.

100,000 participants were reached by 8,702 workshops in 5,586 general trends were observed for some variables. For example,
unique implementation sites over a 25.9 month period across 1,786 after reaching the first 25,000 participants, it took shorter amounts
counties. The majority of participants enrolled in CDSMP work- of time to reach subsequent groups of 25,000 participants (i.e.,
shops (78.4%), followed by Diabetes Self-Management Program 9.37 months to reach the first 25,000 participants and 5.37 months
(DSMP) workshops (10.3%) and Tomando Control de su Salud to reach the last 25,000 participants). As more participants were
(Spanish CDSMP) workshops (8.9%). The largest proportion of reached, larger proportions participated in DSMP workshops (i.e.,
participants attended workshops at senior centers or area agencies 8.3% for the first 25,000 participants and 12.0% for the last 25,000
on aging (29.2%), followed by healthcare organizations (21.1%), participants) and fewer participated in Spanish-language work-
residential facilities (17.6%), community/multi-purpose facilities shops (i.e., 10.7% for the first 25,000 participants and 8.2% for the
(9.9%), faith-based organizations (8.4%), and other settings (e.g., last 25,000 participants). Additionally, workshops were increas-
correctional facilities malls, RV parks, fire departments, county ingly delivered in ZIP Codes with higher percentages of families
administration buildings, private residences, casinos, career cen- residing below the federal poverty line (i.e., an average of 10.76
ters). The majority of participants attended workshops delivered families below poverty for the first 25,000 participants and 11.46
in English (89.6%) and in metro settings (79.6%). On average, for the last 25,000 participants).
workshops included 12.69 (±4.18) participants, and participants
attended 4.38 (±1.72) sessions. The workshop completion rate CDSME PROGRAM PARTICIPANT CHARACTERISTICS
was 74.9%. Table 2 presents participant characteristics of the first 100,000
Workshop delivery characteristics remained relatively consis- CDSME program participants enrolled in the intervention. On
tent across increments of 25,000 participants reached, although average, the first 100,000 CDSME program participants were 67.09

www.frontiersin.org April 2015 | Volume 2 | Article 227 | 83

Smith et al. CDSME program growth

Table 1 | Workshop characteristics by reach increments of 25,000 participants.

Total Participant Participant Participant Participant

1–25,000 25,001–50,000 50,001–75,000 75,001–100,000

Number of unique counties served 1,786 853 1,048 988 1,109

Number of workshops delivered 8,702 2,226 2,138 2,184 2,154
Number of unique implementation sites 5,586 1,705 1,727 1,764 1,769
Time to enroll (in months) 25.91 9.37 6.00 5.17 5.37
Participants reached by CDSME workshop type
Arthritis self-management program (ASMP) 0.5% 0.8% 0.4% 0.1% 0.5%
Chronic disease self-management program (CDSMP) 78.4% 80.2% 79.5% 75.6% 78.3%
Chronic pain self-management program (CPSMP) 0.5% 0.0% 0.6% 0.4% 1.0%
Diabetes self-management program (DSMP) 10.3% 8.3% 9.4% 11.3% 12.0%
Spanish ASMP 0.1% 0.3% 0.0% 0.0% 0.0%
Tomando control de su dabetes (Spanish DSMP) 1.4% 0.8% 1.1% 2.5% 1.2%
Tomando control de su salud (Spanish CDSMP) 8.9% 9.5% 8.9% 10.1% 7.0%
Delivery site types
Senior center/AAA 29.2% 30.5% 27.4% 30.5% 28.3%
Healthcare organizations 21.1% 23.0% 20.5% 20.5% 20.5%
Residential facilities 17.6% 13.4% 18.5% 19.8% 18.8%
Community/multi-purpose facilities 9.9% 9.2% 10.2% 10.2% 10.0%
Faith-based organizations 8.4% 9.3% 8.9% 6.7% 8.7%
Educational institutions 2.3% 2.7% 2.1% 1.5% 2.8%
County health departments 1.3% 1.4% 1.3% 1.0% 1.4%
Tribal organizations 0.2% 0.2% 0.2% 0.1% 0.2%
Workplaces 0.5% 1.0% 0.6% 0.4% 0.2%
Other 9.5% 9.4% 10.2% 9.3% 9.1%

Workshop language
English 89.6% 89.3% 90.0% 87.4% 91.8%
Spanish 10.4% 10.7% 10.0% 12.6% 8.2%
Number of participants enrolled in workshop 12.69 (± 4.18) 12.61 (±4.22) 12.78 (±4.17) 12.66 (±4.18) 12.71 (±4.14)
Number of sessions attended 4.38 (± 1.72) 4.36 (± 1.75) 4.38 (± 1.71) 4.40 (± 1.68) 4.37 (± 1.72)

Successful completion (attend 4+ sessions)

No 25.1% 25.9% 25.3% 23.9% 25.4%
Yes 74.9% 74.1% 74.7% 76.1% 74.6%

Delivery site
Metro 79.6% 78.6% 77.6% 82.6% 79.6%
Non-Metro 20.4% 21.4% 22.4% 17.4% 20.4%
Percent of families below poverty 1128 (±5.39) 10.76 (±4.11) 11.41 (±5.49) 11.48 (±15.67) 11.46 (±16.05)

(±14.58) years of age; 12.0% were under age 50 years, 42.7% participants were reached by CDSME programs, more participants
were aged 65–79 years, and 19.9% were aged 80 years and older. with chronic conditions were enrolled, with the number of partic-
The majority of participants was female (77.9%), non-Hispanic ipants enrolling with three or more chronic conditions increasing
(82.6%), and white (66.0%). Approximately 22% of participants from 34.3% for the first 25,000 participants to 42.0% for the
were African American, 4.5% were Asian or Pacific Islander, 1.6% last 25,000 participants (i.e., participant 75,001–100,000). Addi-
American Indian or Native Alaskan, and 6.2% “other” or multiple tionally, as more participants were reached, the program enrolled
races. The majority of participants resided with other individuals smaller proportions of participants who lived alone (i.e., decreas-
(84.4%) and lived in metro areas (78.2%). On average, participants ing from 21.9% for the first 25,000 participants to 11.3% for the
self-reported 2.20 (±1.71) chronic conditions; 39.6% reported last 25,000 participants).
three or more co-morbidities.
Generally, participant characteristics remained consistent DISCUSSION
across increments of 25,000 participants reached; however, Self-management education has been recognized as a critical factor
trends were observed for some variables. For example, as more in empowering adults to improve their health and functioning (3).

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 227 | 84
Smith et al. CDSME program growth

Table 2 | Sample characteristics by reach increments of 25,000 participants.

Total Participant Participant Participant Participant

1–25,000 25,001–50,000 50,001–75,000 75,001–100,000

Age (average) 67.09 (±14.58) 67.37 (±14.31) 66.67 (±14.71) 67.28 (±14.62) 67.05 (±14.66)
Age group
Under 50 12.0% 11.0% 12.7% 12.2% 12.2%
50–64 25.3% 26.0% 26.3% 24.2% 24.6%
65–79 42.7% 42.7% 41.6% 43.2% 43.4%
80+ 19.9% 20.3% 19.5% 20.3% 19.7%
Sex
Male 22.1% 21.9% 23.2% 21.1% 22.1%
Female 77.9% 78.1% 76.8% 78.9% 77.9%
Hispanic ethnicity
No 82.6% 80.9% 83.3% 80.5% 85.6%
Yes 17.4% 19.1% 16.7% 19.5% 14.4%
Race
White 66.0% 69.2% 65.0% 63.9% 66.3%
African American 21.7% 19.3% 22.4% 22.7% 22.1%
Asian/Pacific Islander 4.5% 5.2% 4.8% 4.2% 4.1%
American Indian/Alaska native 1.6% 1.4% 1.6% 1.8% 1.5%
Other/multiple races 6.2% 5.0% 6.3% 7.4% 6.0%
Number of chronic conditions (average) 2.20 (±1.71) 1.96 (±1.63) 2.25 (±1.17) 2.28 (±1.74) 2.31 (±1.73)

Number of chronic conditions

0 Conditions 18.2% 22.5% 17.1% 17.0% 16.3%
1 Condition 20.9% 22.0% 21.0% 20.6% 19.8%
2 Condition 21.3% 21.2% 21.1% 20.9% 21.9%
3+ Condition 39.6% 34.3% 40.7% 41.5% 42.0%

Live alone
No 84.4% 78.1% 86.0% 85.0% 88.7%
Yes 15.6% 21.9% 14.0% 15.0% 11.3%

Participant residence
Metro 78.2% 77.8% 77.0% 80.9% 77.2%
Non-metro 21.8% 22.2% 23.0% 19.1% 22.8%

This study provides valuable dissemination and implementation them to think differently about program planning, participant
insights into the nature and progression of the largest ever national recruitment, and partnership development. Second, the stimulus
roll out of CDSME programs (i.e., highly effective evidence- money utilized in this initiative was essential for reaching this
based programs designed to help middle-aged and older adults recruitment goal, but it was also leveraged by funds from other
more effectively manage their chronic conditions). The aging ser- organizations with some non-traditional partners (e.g., health-
vices sector, in partnership with other healthcare, public health, care partners), which fostered growth by adopting and promoting
community, and faith-based organizations, proved to be a coor- CDSME programs as an integral care practice. Third, capitaliz-
dinated, efficient, and diverse delivery system capable of rapidly ing on the existing program delivery infrastructure established
reaching large numbers of older adults across the country. Exceed- by previous AoA initiatives, the broad network of delivery and
ing programmatic goals of having 50,000 participants complete funding partners has resulted in widespread financing of CDSME
CDSME program workshops (12), over 100,000 participants were programs by other government organizations. Fourth, workshops
reached more quickly than in previous efforts (16). Further, with were available in many local communities largely because of the
the exception of a predominant female participant population cooperation of the program developers to utilize and expand their
typically served with health promotion programs (16–18), par- training infrastructure (5). As seen in this initiative, the culmi-
ticipants were representative of the U.S. population and not just nation of leveraging opportunities led to the rapid dissemination
easy-to-reach subgroups. of CDSME programs by creating a highly collaborative commu-
The ability of this initiative to quickly reach 100,000 partici- nity structure that accelerated the speed of scalability across the
pants can be attributed to many factors. First, having each state set country to meet the needs of an increasingly diverse group of
ambitious yet feasible and attainable goals (19) can help stimulate participants.

www.frontiersin.org April 2015 | Volume 2 | Article 227 | 85

Smith et al. CDSME program growth

Past reports have shown CDSME programs have capacity to facilitates this type of grand-scale roll out, and what types of
serve large numbers of heterogeneous adults via a growing network participants are reached.
of delivery sites (8, 12, 20, 21). Success can be attributed, in part, Findings from this study capture the spread of CDSME pro-
to a community-driven delivery system that employed existing grams during a national, government-funded roll out and show the
networks for recruiting participants of varied ages, race/ethnicity; ability of this intervention to rapidly reach a diverse set of partici-
disease status; geographic region; and socio-economic status (22). pants using a well-coordinated delivery system. As of August 2014,
However, additional efforts are needed to help CDSME programs over 196,700 participants reached by CDSME programs through
gain major penetration among the over 35 million older Amer- 17,500 workshops in 1,200 counties across the United States. While
icans estimated to have at least one chronic condition (23). As this initiative capitalized and built upon previous efforts to create
such, this study suggests several actions that can help make the a delivery infrastructure for CDSME programs, this grand-scale
dissemination of CDSME programs part of routine care. dissemination has solidified the presence of CDSME programs
First, we must further examine and strengthen referral systems with great potential for long-term sustainability. While this ini-
to CDSME programs and the interconnectedness of the health- tiative has achieved impressive participant reach and completion,
care, public health, and aging services networks. Multi-institute it should be noted that many other organizations throughout the
funding initiatives that highly encourage/mandate multi-sectorial United States offered the intervention despite not receiving this
partnerships can set the stage for bridging such connections (11). ARRA funding. Because data from these organizations are not rep-
Second, we need to embrace the paradigm shift in provider-patient resented in the databased used in the current study, these findings
communications that emphasizes the value of “informed and are even more encouraging in that they underrepresent the actual
activated” patients working collaboratively with their prepared translation of CDSME programs nationwide. Continued efforts
practice team (1). This theme, initially articulated in Wagner’s are needed to track the progression and proliferation of this suite
chronic care model, is being revisited with the recent movement of programs to empower patients with self-management skills to
toward patient-centered care (24). Third, we need to be aware deal with their chronic conditions.
of the constraints facing today’s healthcare providers in terms of
shortened office visits and greater expectations for administrative ACKNOWLEDGMENTS
paperwork (25). Thus, we recommend easy-to-employ methods The American Recovery and Reinvestment Act of 2009 (i.e., Recov-
and mechanisms (e.g., automated systems) to help health care ery Act) Communities Putting Prevention to Work: Chronic Dis-
providers know where evidence-based programs like CDSMP are ease Self-Management Program initiative, led by the U.S. Admin-
offered. Also, guidelines are needed for identifying the types of istration on Aging in collaboration with the Centers for Disease
patients who are best suited for specific programs (e.g., informa- Control and Prevention and the Centers for Medicare and Med-
tion about the pros and cons of generic self-care programs versus icaid Services, allotted $32.5 million to support the translation of
disease-specific programs). While clinicians and other allied health the Stanford program in 45 States, Puerto Rico, and the District
providers should be trained about these guidelines and referral of Columbia. The National Council on Aging served as the Tech-
processes, it is also important that program participants report nical Assistance Resource Center for this initiative and collected
back to their healthcare providers about their experiences and de-identified data on program participation.
progress in such programs. Fourth, we must recognize that pro-
grammatic scalability needs to be paired with plans for achieving REFERENCES
sustainability over time. Thus, we recommend that national, state, 1. Wagner EH, Glasgow RE, Davis C, Bonomi AE, Provost L, McCulloch D, et al.
and local roll outs of evidence-based programs include sustain- Quality improvement in chronic illness care: a collaborative approach. Jt Comm
ability planning as a core element. Successful sustainability plans J Qual Improv (2001) 27(2):63–80.
2. Barr V, Robinson S, Marin-Link B, Underhill L, Dotts A, Ravensdale D, et al.
are those that build upon and leverage existing resources, often
The expanded Chronic Care Model: an integration of concepts and strategies
employing champions for recognizing and promoting new models from population health promotion and the Chronic Care Model. Hosp Q (2003)
of care (26). 7(1):73. doi:10.12927/hcq.2003.16763
There are study limitations that must be acknowledged. While 3. Brady TJ, Murphy L, O’Colmain BJ, Beauchesne D, Daniels B, Greenberg M,
this national effort afforded large numbers of participants, specific et al. A meta-analysis of health status, health behaviors, and healthcare utiliza-
tion outcomes of the chronic disease self-management program. Prev Chronic
data points were limited due to community concerns regarding
Dis (2013) 10:120112. doi:10.5888/pcd10.120112
burden. Additionally some data was missing due to local/regional 4. Brady T, Murphy L, Beauchesne D, Bhalakia A, Chervin D, Daniels B, et al. Sorting
constraints, and not necessarily individual refusal. Study data rep- Through the Evidence for the Arthritis Self-Management Program and the Chronic
resented a “snapshot” of an ongoing evolving evaluation process Disease Self-Management Program: Executive Summary of ASMP/CDSMP Meta-
at a particular time point. Underserved populations (e.g., African Analysis. Atlanta, GA: Centers for Disease Control and Prevention (CDC) (2011).
5. Stanford School of Medicine. Chronic Disease Self-Management Program (Bet-
Americans) were overrepresented in this study because of the focus ter Choices, Better Health® Workshop) [cited 2014 July 12]. Available from:
of the larger initiative to serve this subgroup of Americans. How- http://patienteducation.stanford.edu/programs/cdsmp.html
ever, males were underrepresented in this study, as they tradition- 6. Bodenheimer T, Lorig K, Holman H, Grumbach K. Patient self-management of
ally are in evidence-based programs delivered through the aging chronic disease in primary care. JAMA (2002) 288(19):2469–75. doi:10.1001/
services network (16–18). Despite these limitations inherent when jama.288.19.2469
7. Lorig KR, Sobel DS, Stewart AL, Brown BW Jr, Bandura A, Ritter P, et al. Evidence
using administrative records, we nevertheless believe this study suggesting that a chronic disease self-management program can improve health
represents a unique examination about how a national evidence- status while reducing hospitalization: a randomized trial. Med Care (1999)
based dissemination rolls out over time, what infrastructure 37(1):5–14. doi:10.1097/00005650-199901000-00003

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 227 | 86
Smith et al. CDSME program growth

8. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, Whitelaw N, et al. Successes from the communities putting prevention to work initiative. Ethn Dis (2012)
of a national study of the chronic disease self-management program: meet- 23(4):508–17.
ing the triple Aim of health care reform. Med Care (2013) 51(11):992–8. 22. Burdine JN, McLeroy K, Blakely C, Wendel ML, Felix MR. Community-based
doi:10.1097/MLR.0b013e3182a95dd1 participatory research and community health development. J Prim Prev (2010)
9. Ahn S, Basu R, Smith ML, Jiang L, Lorig K, Whitelaw N, et al. The impact 31(1):1–7. doi:10.1007/s10935-010-0205-9
of chronic disease self-management programs: healthcare savings through a 23. Anderson GF. Chronic Care: Making the Case for Ongoing Care. Princeton, NJ:
community-based intervention. BMC Public Health (2013) 13(1):1141. doi:10. Robert Wood Johnson Foundation (2010).
1186/1471-2458-13-1141 24. Bojadzievski T, Gabbay RA. Patient-centered medical home and diabetes. Dia-
10. Department of Health and Human Services. AoA Evidence-Based Prevention betes Care (2011) 34(4):1047–53. doi:10.2337/dc10-1671
Programs. Washington, DC: Administration on Aging. (2013) [cited 2014 July 25. Ahn S, Tai-Seale M, Huber C Jr, Smith ML, Ory MG. Psychotropic medication
8]. Available from: http://www.aoa.gov/AoA_Programs/HPW/Title_IIID/index. discussions in older adults’ primary care office visits: so much to do, so little time.
aspx Aging Ment Health (2011) 15(5):618–29. doi:10.1080/13607863.2010.548055
11. U.S. Department of Health and Human Services, Administration on 26. Scheirer MA. Is sustainability possible? A review and commentary on empirical
Aging. ARRA – Communities Putting Prevention to Work: Chronic Dis- studies of program sustainability. Am J Eval (2005) 26(3):320–47. doi:10.1177/
ease Self-Management Program. (2012) [cited 2014 June 14]. Avail- 1098214005278752
able from: www.cfda.gov/index?s=program&mode=form&tab=step1&id=
5469a61f2c5f25cf3984fc3b94051b5f
12. Ory MG, Smith ML, Kulinski KP, Lorig K, Zenker W, Whitelaw N. Self- Conflict of Interest Statement: The authors declare that the research was conducted
management at the tipping point: reaching 100,000 Americans with evidence- in the absence of any commercial or financial relationships that could be construed
based programs. J Am Geriatr Soc (2013) 61(5):821–3. doi:10.1111/jgs.12239 as a potential conflict of interest.
13. Lorig K. Living a Healthy Life with Chronic Conditions: Self-Management of Heart
Disease, Arthritis, Diabetes, Asthma, Bronchitis, Emphysema & Others. 3rd ed. This paper is included in the Research Topic, “Evidence-Based Programming for Older
Boulder, CO: Bull Pub. Company (2006). 382 p. Adults.” This Research Topic received partial funding from multiple government and
14. Bandura A. Social cognitive theory of self-regulation. Organ Behav Hum Decis private organizations/agencies; however, the views, findings, and conclusions in these
Process (1991) 50(2):248–87. doi:10.1016/0749-5978(91)90022-L articles are those of the authors and do not necessarily represent the official position
15. Kulinski KP, Boutaugh M, Smith ML, Ory MG, Lorig K. Setting the stage: mea- of these organizations/agencies. All papers published in the Research Topic received
sure seletion, coordination, and data collection for a national self-management peer review from members of the Frontiers in Public Health (Public Health Education
initiative. Front Public Health (2015) 2:206. doi:10.3389/fpubh.2014.00206 and Promotion section) panel of Review Editors. Because this Research Topic repre-
16. Smith M, Belza B, Altpeter M, Ahn S, Dickerson J, Ory M. Disseminating an sents work closely associated with a nationwide evidence-based movement in the US,
evidence-based disease self-management program for older Americans: impli- many of the authors and/or Review Editors may have worked together previously in
cations for diversifying participant reach through delivery site adoption. In: some fashion. Review Editors were purposively selected based on their expertise with
Maddock J, editor. Rijeka. Croatia: InTech (2012). p. 385–404. evaluation and/or evidence-based programming for older adults. Review Editors were
17. Smith ML, Ory MG, Belza B, Altpeter M. Personal and delivery site char- independent of named authors on any given article published in this volume.
acteristics associated with intervention dosage in an evidence-based fall risk
reduction program for older adults. Transl Behav Med (2012) 2(2):188–98. Received: 07 August 2014; accepted: 24 October 2014; published online: 27 April 2015.
doi:10.1007/s13142-012-0133-8 Citation: Smith ML, Ory MG, Ahn S, Kulinski KP, Jiang L, Horel S and Lorig K (2015)
18. Smith ML, Belza B, Braun KL, King S, Palmer RC, Sugihara NS, et al. National National dissemination of chronic disease self-management education programs: an
reach and dissemination of enhancefitness. Health Behav Policy Rev (2014) incremental examination of delivery characteristics. Front. Public Health 2:227. doi:
1(2):150–60. doi:10.14485/HBPR.1.2.7 10.3389/fpubh.2014.00227
19. The art of health promotion. Am J Health Promot (2004) 18(4): This article was submitted to Public Health Education and Promotion, a section of the
TAH–1–TAH–12. doi:10.4278/0890-1171-18.4.TAHP-1 journal Frontiers in Public Health.
20. Erdem E, Korda H. Self-management program participation by older adults Copyright © 2015 Smith, Ory, Ahn, Kulinski, Jiang , Horel and Lorig . This is an open-
with diabetes: Chronic Disease Self-Management Program and Diabetes Self- access article distributed under the terms of the Creative Commons Attribution License
Management Program. Fam Community Health (2014) 37(2):134–46. doi:10. (CC BY). The use, distribution or reproduction in other forums is permitted, provided
1097/FCH.0000000000000025 the original author(s) or licensor are credited and that the original publication in this
21. Korda H, Erdem E, Woodcock C, Kloc M, Pedersen S, Jenkins S. Racial and ethnic journal is cited, in accordance with accepted academic practice. No use, distribution or
minority participants in chronic disease self-management programs: findings reproduction is permitted which does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 227 | 87

 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2015.00019

Workshop characteristics related to Chronic Disease

Self-Management Education program attendance
Matthew Lee Smith 1 *, Marcia G. Ory 2 , Luohua Jiang 3 , Kate Lorig 4 , Kristie P. Kulinski 5 and SangNam Ahn 6,2
1
Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA, USA
2
Department of Health Promotion and Community Health Sciences, Texas A&M Health Science Center School of Public Health, College Station, TX, USA
3
Department of Epidemiology, School of Medicine, University of California Irvine, Irvine, CA, USA
4
Stanford Patient Education Research Center, Department of Medicine, Stanford University School of Medicine, Palo Alto, CA, USA
5
National Council on Aging, Washington, DC, USA
6
Division of Health Systems Management and Policy, School of Public Health, The University of Memphis, Memphis, TN, USA

Edited by: Using the national dissemination of Chronic Disease Self-Management Education (CDSME)
Sanjay P. Zodpey, Public Health
programs, the purposes of this study were to (1) document intervention attendance rates
Foundation of India, India
as related to the number of participants enrolled in the workshop and (2) compare the rela-
Reviewed by:
Shridhar Murlidharrao Kadam, Public tionship between workshop attendance and workshop size by delivery site rurality and type.
Health Foundation of India, India Data were analyzed from the first 100,000 middle-aged and older adults who participated
Michael Elliott, Saint Louis University in CDSME workshops spanning 45 states, Puerto Rico, and the District of Columbia as part
College for Public Health and Social
of the American Recovery and Reinvestment Act of 2009 Communities Putting Prevention
Justice, USA
to Work: Chronic Disease Self-Management Program initiative. Descriptive statistics are
*Correspondence:
Matthew Lee Smith, Department of reported for all participants, then separately by each delivery site type. Ratios between the
Health Promotion and Behavior, number of workshop participants and the number of workshop sessions attended were
College of Public Health, The calculated and graphed based on the rurality of delivery and separately for the leading five
University of Georgia, Health Science
delivery site types. Associations between the number of workshop participants and the
Campus, Wright Hall #345E, Athens,
GA 30602, USA number of sessions attended differed by delivery site rurality and type. Findings have impli-
e-mail: [email protected] cations for participant retention and workshop delivery costs, which can assist program
deliverers to strategically plan implementation efforts in their areas.
Keywords: chronic disease self-management, evidence-based program, older adults, intervention dose, evaluation

INTRODUCTION designated population of interest (14, 15). To counter recruitment

The recent movement toward evidence-based public health calls challenges often seen in research studies (16–18), there is now a
for a better understanding of the implementation and dissemina- growing literature on strategies to increase recruitment by facil-
tion of evidence-based programs (EBP) for older adults deliv- itating program adoption in a host of different delivery systems
ered in real world settings (1–4). EBP are interventions based reflecting where the population of interest resides and frequently
on research that were tested in clinical trials and translated into encounter in their daily lives.
community-based models, which receive the same intended health Despite the assumed importance of structured program fea-
benefits (5). EBP have common components, foremost of which tures such as class size or workshop delivery type, little is known
are essential intervention elements, materials, and procedures (6). about the programmatic impact of different delivery characteris-
More specifically, implementation features must be considered, tics on achieving recommended intervention doses. This is, in part,
which include having a well-defined program structure and time- because assumptions about ideal class size are often applied from
frame that enables the developers to track fidelity and others to prior literature without consideration of the specific intervention
uniformly deliver the program with replicable findings (7). Pro- focus, population, or setting. Delivery sites may be seen as imple-
gram developers often draw upon small group literature and adult mentation issues rather than researchable variables, an attitude
learning principles to define the ideal class size for intervention reinforced by the limited number of delivery site types included in
(8–10), which is often 12–16 participants (11). From our expe- most intervention studies.
rience working with program developers, determination of ideal The widespread availability of Chronic Disease Self-
class size is often more of an art than a science and is based on Management Education (CDSME) programs nationwide across
assumptions about ideal size to facilitate active group discussion. a multitude of settings has provided opportunity to examine the
Secondary concerns often revolve around cost implications of dif- programmatic impact of different delivery characteristics on par-
ferent class sizes in intervention studies because per-participant ticipants receiving the recommended intervention dose. Using
costs are influenced by the total number of participants enrolled the national dissemination of CDSME programs, the purposes
in workshops (12, 13). of this study were to (1) document intervention attendance rates
With a desire to take EBP to scale in order to make a public as related to the number of participants enrolled in the workshop
health impact, there is a need for widespread penetration in the and (2) compare the relationship between workshop attendance

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 19 | 88
Smith et al. CDSME program attendance

and workshop size by delivery site rurality (i.e., metro, non-metro) if they attended four or more of the six offered workshop sessions
and type (e.g., senior centers, healthcare organizations, residential (15, 21, 22).
facilities, faith-based organizations).
Class size
MATERIALS AND METHODS The number of participants enrolled in each CDSME workshop
PROGRAM DESCRIPTION was recorded (i.e., ranging from 1 to 20 participants). The max-
The Chronic Disease Self-Management Program (CDSMP) has imum number of participants allowed to be listed as enrollees in
been introduced and widely disseminated in the U.S. as a method any single workshop was 20. As a point of reference, the program
to empower patients with self-management skills to deal with their developers define the ideal class size as (i.e., between 10 and 15
chronic conditions (19). There is now a suite of CDSME programs participants) in the CDSMP fidelity manual (26).
licensed through the Stanford Patient Education Research Cen-
Delivery site types
ter, some of which are generic (e.g., CDSMP, Tomando Control
de su Salud) and others that are disease specific (e.g., diabetes, Data are presented for all 10 delivery site types (see Table 1), which
arthritis, chronic pain). Drawing upon Social Learning Theory were then assessed graphically based on the workshop rurality
(20), CDSME programs are evidence-based, peer-led interven- and independently for the leading five delivery site types based
tions consisting of six highly participative classes held for 2.5 h on participant enrollment (i.e., senior centers, healthcare organi-
each, once a week, for six consecutive weeks (19). CDSME pro- zations, residential facilities, community/multi-purpose facilities,
grams have resulted in improved health care and health (21, 22), and faith-based organizations). Data pertaining to CDSME pro-
while potentially saving healthcare costs (12). gram delivery site types were gathered administratively (24). Using
the ZIP code information provided by each delivery site, work-
DATA SOURCE AND STUDY POPULATION
shops were categorized as metro or non-metro based on the
Cross-sectional data for this study were obtained from a nation- rural–urban commuting area codes (RUCA) (27). The leading five
wide delivery of CDSME programs as part of the American CDSMP delivery site types included in analyses were senior centers
Recovery and Reinvestment Act of 2009 (i.e., ARRA) Communi- or area agencies on aging (AAAs), healthcare organizations, resi-
ties Putting Prevention to Work: Chronic Disease Self-Management dential facilities, community or multi-purpose centers (including
Program initiative (15). The US Administration on Aging led libraries), and faith-based organizations.
this initiative in collaboration with the Centers for Disease Con- Personal characteristics
trol and Prevention and the Centers for Medicare and Med- Personal characteristics of the participants included age, gender,
icaid Services to support the translation of CDSME programs race/ethnicity, and self-reported number of chronic conditions
in 45 states, Puerto Rico, and the District of Columbia (23). (i.e., arthritis, cancer, depression, diabetes, heart disease, hyper-
Federal funding for this initiative enabled participants to enroll tension, lung disease, stroke, osteoporosis, and other chronic
in CDSMP workshops free of charge. This initiative was orig- conditions).
inally designed to have 50,000 Americans complete at least
four out of six CDSME workshop sessions between 2010 and ANALYSES
2012 and to embed CDSME program delivery structures into Descriptive statistics were calculated for all participants, then sep-
statewide systems (15). For this study, data were analyzed from arately for each of the 10 delivery site types. Percentages are
the first 100,000 participants who attended CDSME program provided for categorical variables. Averages and standard devia-
workshops and responded to all relevant survey questions. Work- tions are provided for continuous and count variables. The average
shops included in study analyses began between January 2010 and number of workshop sessions attended by the size of the work-
February 2012. shop (i.e., the number of participants enrolled in each workshop)
As described elsewhere (24), all states receiving ARRA fund- was calculated and graphed based on the rurality of delivery and
ing for this initiative were assigned program completer target separately for the leading five delivery site types.
goals. It was expected that CDSME program workshops would
be delivered through certain site (e.g., senior centers, healthcare RESULTS
organizations, residential facilities, educational institutions, faith- CDSMP PARTICIPANT AND WORKSHOP CHARACTERISTICS BY
based organizations, and tribal centers). Each delivery site type DELIVERY SITE TYPE
recruited participants to enroll in workshops using their usual Of the first 100,000 participants reached in this initiative,
methods (e.g., flyers, emails, face-to-face). The majority of partic- 29.2% attended workshops at senior centers/AAAs, 21.1% at
ipants was introduced to the program during the first workshop healthcare organizations, 17.6% at residential facilities, 9.9%
session; however, some participants were introduced to the pro- at community/multi-purpose facilities (including libraries), and
gram during an optional pre-workshop session called a “session 8.4% at faith-based organizations. Smaller proportions of par-
zero” (25). ticipants attended workshops at educational institutions (2.3%),
county health departments (1.3%), workplaces (0.5%), and tribal
MEASURES centers (0.2%). Approximately 9.5% of participants attended
Workshop attendance CDSME workshops at delivery sites classified as “other” (e.g.,
Participants’ attendance was recorded to determine if the recom- correctional facilities malls, RV parks, fire departments, county
mended intervention dose was received. As defined by the program administration buildings, private residences, casinos, career
developers, a participant has“successfully”completed the program centers).

www.frontiersin.org April 2015 | Volume 3 | Article 19 | 89

 Smith et al.

Table 1 | Participant and workshop characteristics by delivery site type.

Total Senior Healthcare Residential Comm/ Faith-Based Educational Health Tribal Workplace Other
(n = 100,000) Center/AAA Organization Facility Multi-Purp/ Organization Institution Department Center (n = 541) (n = 9,516)
(n = 29,152) (n = 21,136) (n = 17,631) Library (n = 8406) (n = 2264) (n = 1274) (n = 189)
(n = 9,891)

PARTICIPANT CHARACTERISTICS
Age 67.1 (±14.58) 71.1 (±11.76) 61.9 (±14.49) 73.5 (±12.96) 65.7 (±14.01) 65.7 (±13.71) 61.8 (±15.85) 64.2 (±14.59) 64.6 (±13.46) 60.8 (±14.80) 59.0 (±17.21)
Female 77.9% 80.5% 73.3% 82.7% 78.2% 78.9% 82.8% 79.6% 72.0% 81.7% 68.8%
Race/Ethnicity
Latino/Hispanic 16.6% 12.4% 27.3% 10.4% 17.1% 13.0% 27.7% 7.7% 5.1% 5.4% 21.2%
Non-Hispanic White 54.4% 58.0% 51.7% 60.0% 49.9% 44.1% 58.6% 75.0% 24.3% 63.3% 48.6%

Frontiers in Public Health | Public Health Education and Promotion

African-American 21.8% 23.6% 14.6% 23.7% 20.8% 34.0% 10.6% 9.4% 7.3% 13.5% 22.0%

Asian/Pacific Islander 4.6% 3.6% 3.2% 3.6% 10.1% 7.1% 2.3% 4.8% 0.0% 14.4% 4.2%

American Indian/ 1.5% 1.2% 1.9% 1.0% 0.9% 0.8% 0.4% 1.5% 61.0% 1.7% 3.0%
Alaska native
Other race 1.2% 1.1% 1.3% 1.4% 1.1% 1.1% 0.6% 1.6% 2.3% 1.7% 1.0%

Average number of 2.2 (±1.71) 2.3 (±1.71) 2.1 (±1.68) 2.4 (±1.80) 2.1 (±1.70) 2.0 (±1.60) 2.0 (±1.66) 2.2 (±1.78) 2.7 (±1.80) 1.9 (±1.59) 2.0 (±1.63)
co-morbidities
WORKSHOP CHARACTERISTICS

Class size 12.7 (±4.18) 13.1 (±4.26) 12.1 (±4.13) 13.1 (±4.01) 12.5 (±4.18) 12.9 (±4.09) 12.9 (±4.10) 10.3 (±4.28) 9.6 (±2.74) 10.7 (±3.89) 12.8 (±4.12)
Number of sessions 4.4 (±1.72) 4.5 (±1.65) 4.2 (±1.78) 4.2 (±1.81) 4.4 (±1.70) 4.5 (±1.63) 4.5 (±1.70) 4.2 (±1.75) 4.2 (±1.69) 4.7 (±1.52) 4.7 (±1.66)
attended
Successful completion 74.9% 77.0% 72.1% 70.2% 74.7% 78.7% 77.3% 69.2% 69.3% 82.6% 79.7%

Proportion of 20.4% 22.4% 17.7% 16.4% 17.1% 22.2% 25.8% 36.7% 13.8% 31.8% 25.7%
workshops in
non-metro areas

April 2015 | Volume 3 | Article 19 | 90

CDSME program attendance
Smith et al. CDSME program attendance

On average, participants were age 67.1 (±14.6) years and higher average workshop attendance. In faith-based organizations,
had 2.2 (±1.7) self-reported chronic conditions. The majority however, there was no observed association between the number
of participants were female (77.9%) and non-Hispanic white of workshop participants and workshop attendance.
(54.4%). Almost 22% of participants were African-American
and 16.6% were Hispanic. Workshops at senior centers/AAAs DISCUSSION
and residential facilities enrolled participants with older than As demonstrated in a previous study, workshop size was associ-
average ages. Healthcare organizations, tribal centers, and sites ated with workshop attendance in national dissemination efforts
categorized as “other” enrolled larger proportions of male par- of EBP for older adults (13). As confirmed by other studies
ticipants. Healthcare organizations, community/multi-purpose (15, 22), most delivery sites reported workshop sizes in the ideal
facilities, educational institutions, and other sites enrolled larger range (i.e., between 10 and 15 participants) and large proportions
proportions of Hispanic participants. Senior centers/AAAs, res- of participants with successful completion rates. While general
idential facilities, and faith-based organizations enrolled larger findings in the current study indicate that workshops with fewer
proportions of African-American participants. Healthcare orga- participants had higher attendance rates, variability was noted
nizations, residential facilities, community/multi-purpose facili- by setting, especially among smaller workshops. Greater variabil-
ties, and tribal centers enrolled more participants from metro ity in smaller workshop as observed in workshops delivered in
areas. Workshops at senior centers/AAAs, residential facilities, non-metro areas, healthcare organizations, and faith-based orga-
and tribal centers enrolled participants with higher than average nizations. The strongest negative association was observed in
co-morbidities. residential facilities.
More than 20% of participants attended workshops delivered in Consistent with the RE-AIM planning and evaluation frame-
non-metro areas. The workshops in non-metro areas had smaller work (28, 29), wide-scale programmatic dissemination to diverse
average class size than those in metro areas, but no difference population subgroups often requires a multitude of community
in class attendance. On average, workshops had 12.7 (±4.2) par- partnerships representing various settings. It is not surprising that
ticipants, and participants attended an average of 4.4 (±1.7) of senior centers/AAAs and healthcare organizations serve as the
the six workshop sessions. The majority of participants success- predominant sites, given the sponsorship of this initiative by the
fully completed the workshop (74.9%), indicating they received Administration on Aging (23). However, it is more interesting to
the recommended intervention dose. Senior centers/AAAs, faith- consider the different delivery settings utilized in this initiative’s
based organizations, educational institutions, and delivery sites program implementation and dissemination activities (e.g., senior
categorized as “other” had higher than average workshop sizes and centers, healthcare organizations, residential facilities, faith-based
workshop attendance. Residential facilities also had higher than organizations). This study contributes to the emerging implemen-
average workshop sizes. Workplaces also had higher than average tation science literature by identifying and documenting the wide
workshop attendance. variability in workshop size and attendance based on different
setting types (30).
ASSOCIATIONS BETWEEN WORKSHOP SIZE AND ATTENDANCE As seen elsewhere (31, 32), certain delivery site types are known
As shown in Figure 1, associations between the number of work- to attract participants with certain characteristics, which make it
shop participants and the number of sessions attended differ by difficult to disentangle the impact of workshop size and atten-
workshop rurality. More specifically, in workshops in metro areas, dance from the types of participants who attend a particular
there was a negative correlation between participant number and delivery site type. Future research would benefit from qualita-
session attendance for smaller workshops (i.e., workshops ≤ 8 par- tive research to better understand what drives participants to
ticipants). The average number of sessions attended in non-metro one setting or another. For example, is participant attendance
organizations had higher variability, especially for smaller work- related to the supply of programing at different settings? Is it
shops. Associations between the number of workshop participants that participants identify with a particular organizational setting
and the number of sessions attended differed by delivery site type. and, therefore, attend workshops where they are most comfort-
More specifically, in senior centers/AAAs, there was a negative cor- able (33)? Or, is it simply a proximity issue in that participants
relation between participant number and session attendance for attend workshops that are closes to their home or work (34)?
smaller workshops (i.e., workshops ≤ 8 participants). Stated dif- Or, might it be a combination where participants are willing to
ferently, the fewer the participants enrolled in a workshop, the travel further distances to attend workshops delivered in a set-
higher the rate of session completion. For workshops with nine or ting of preference? These issues warrant further investigation at
more participants, the workshop size was not correlated with the the individual-level based on preference and the workshop-level
average number of sessions attended. The relationships between based on common characteristics associated with workshop size
participant number and session attendance in healthcare organi- and attendance.
zations and community/multi-purpose/libraries were similar to These study findings are illuminating in that they show the
that observed in senior centers/AAAs. However, the average num- interconnectedness of and interaction between workshop size,
ber of sessions attended in healthcare organizations had higher delivery site type, and intervention dosage. Findings indicate that
variability, especially for smaller workshops. there is no“one size fits all”rule of thumb regarding ideal workshop
For delivery sites located in residential facilities, the number size and that the recommended intervention dose can be obtained
of workshop participants was negatively associated with session at different delivery settings in workshop of differing sizes. Addi-
attendance. Generally, workshops with fewer participants had tional research is needed to determine the influence of workshop

www.frontiersin.org April 2015 | Volume 3 | Article 19 | 91

Smith et al. CDSME program attendance

FIGURE 1 | Average number of workshop sessions attended by class size by workshop rurality and leading five delivery site type.

size and attendance on known health-related benefits associated This has been seen in our calculations of the cost savings that can
with CDSME programs, controlling for workshop delivery site. be derived from CDSME programs based on variations in overall
Further, there is need for more sophisticated threshold analyses to per-participant costs, which is highly dependent upon class size
determine the critical class size for optimal attendance and how (12, 36). More specifically, based on a workshop cost of $3500
that may differ by delivery site type. USD, per-participant costs can range from about $219 USD for
Several study limitations can be noted. First, the cross-sectional larger workshops with 16 participants to $583 USD for smaller
nature of the study and lack of outcome data limited our ability to workshops with 6 participants (12). As such, because CDSME
determine causality and associate workshop size and attendance program workshops have a fair amount of fixed costs, regardless
with salient health outcomes. Second, there were a limited number of workshop size (e.g., associated with site coordination, partici-
of variables collected about the delivery sites and/or from partic- pant recruitment), hosting larger versus smaller workshops seems
ipants; thus, we were unable to investigate the greater context of to be more fiscally beneficial to organizations implementing these
factors related to delivery site type selection, reasons for atten- programs. These cost-related variations have implications for pro-
dance, or reasons why certain delivery sites held workshops of gram administrator and decision makers to finically plan future
certain sizes. However, the large number of workshops delivered dissemination efforts and identify necessary resources and part-
and participants enrolled in this national initiative provides an ini- ners to achieve participant recruitment goals. Further, because of
tial glimpse into study questions and suggests areas needing more its small group approach using the social learning theory (19, 20),
exploration. Third, this descriptive study was served as a prelim- workshop size should be considered to ensure the program oper-
inary examination of the relationship between workshop charac- ates as intended and participants receive anticipated intervention
teristics (delivery site type and number of participants) and class benefits.
attendance. Future studies with more sophisticated, inferential sta-
tistics that include more predictor variables are needed to compare CONCLUSION
these relationships by other factors (e.g., self-reported chronic The implementation processes in a national rollout of evidence-
condition types) and health-related improvements resulting from based CDSME programs are necessarily complex. Previous
intervention attendance. assumptions about the ideal class size need to be weighed in terms
This research has several practical implications. First, multi- of both programmatic and cost metrics, balancing the economies
pronged strategies are needed to improve participant retention of “going to scale” with the benefits of smaller class sizes in some
so participants can receive the recommended intervention dose, settings. Therefore, it is important to recognize how delivery
despite workshop enrollment size. These strategies should be tai- sites cater to different types of participants, which may in turn
lored approaches by delivery site types based on their specific influence program outcomes. Findings have implications for par-
participant characteristics and health-related status (35). Second, ticipant retention and workshop delivery costs, which can assist
while class size may not always be associated with intervention program deliverers to strategically plan implementation efforts in
dose, class size has implications for overall program costs (12, 36). their areas.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 19 | 92
Smith et al. CDSME program attendance

ACKNOWLEDGMENTS 17. Levkoff SE, Prohaska TR, Weitzman PF, Ory MG. Minority populations: Recruit-
The American Recovery and Reinvestment Act of 2009 (i.e., Recov- ment and retention in aging research. In: Maddox GL, editor. The Encyclopedia
of Aging. New York, NY: Springer Publishing Co (2000). p. 694–6.
ery Act) Communities Putting Prevention to Work: Chronic Dis-
18. Ory MG, Lipman PD, Karlen PL, Gerety MB, Stevens VJ, Singh MA, et al. Recruit-
ease Self-Management Program initiative, led by the US Admin- ment of older participants in frailty/injury prevention studies. Prev Sci (2002)
istration on Aging in collaboration with the Centers for Dis- 3(1):1–22. doi:10.1023/A:1014610325059
ease Control and Prevention and the Centers for Medicare and 19. Lorig K. Living a Healthy Life with Chronic Conditions: Self-Management of Heart
Medicaid Services, allotted $32.5 million to support the transla- Disease, Arthritis, Diabetes, Asthma, Bronchitis, Emphysema & Others. 3rd ed.
Boulder, CO: Bull Pub. Company (2006). 382 p.
tion of the Stanford program in 45 States, Puerto Rico, and the
20. Bandura A. Social cognitive theory of self-regulation. Organ Behav Hum Decis
District of Columbia. The National Council on Aging served as Process (1991) 50(2):248–87. doi:10.1016/0749-5978(91)90022-L
the Technical Assistance Resource Center for this initiative and 21. Ory MG, Ahn S, Jiang L, Lorig K, Ritter P, Laurent DD, et al. National study of
collected de-identified data on program participation. chronic disease self-management six-month outcome findings. J Aging Health
(2013) 25(7):1258–74. doi:10.1177/0898264313502531
22. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, Whitelaw N, et al. Successes
REFERENCES of a national study of the chronic disease self-management program: meet-
1. National Center for Chronic Disease Prevention and Health Promotion. Healthy ing the triple aim of health care reform. Med Care (2013) 51(11):992–8.
Aging Research Network: Putting Collective Wisdom to Work for Older Ameri- doi:10.1097/MLR.0b013e3182a95dd1
cans. Atlanta, GA: Centers for Disease Control and Prevention (CDC) (2011). 23. U.S. Department of Health and Human Services, Administration on Aging.
CS226787 p. ARRA – Communities Putting Prevention to Work: Chronic Disease Self-
2. National Council on Aging. Evidence-Based Health Promotion Programs for Older Management Program 2012. (2014). Available from: www.cfda.gov/index?s=
Adults: Key Factors and Strategies Contributing to Program Sustainability. Wash- program&mode=form&tab=step1&id=5469a61f2c5f25cf3984fc3b94051b5f
ington, DC: National Council on Aging (2012). 24. Kulinski KP, Boutaugh M, Smith ML, Ory MG, Lorig K. Setting the stage: mea-
3. Green LW, Ottoson JM, Garcia C, Hiatt RA. Diffusion theory and knowledge sure selection, coordination, and data collection for a national self-management
dissemination, utilization, and integration in public health. Annu Rev Public initiative. Front Public Health (2015) 2:206. doi:10.3389/fpubh.2014.00206
Health (2009) 30:151–74. doi:10.1146/annurev.publhealth.031308.100049 25. Jiang L, Smith ML, Chen S, Ahn S, Kulinski KP, Lorig K, et al. The role of Session
4. Brownson RC, Colditz GA, Proctor EK. Dissemination and Implementation Zero in successful completion of Chronic Disease Self-Management Program
Research in Health: Translating Science to Practice. New York, NY: Oxford Uni- workshops. Front Public Health (2015) 2:205. doi:10.3389/fpubh.2014.00205
versity Press (2012). 26. Stanford Patient Education Research Center. Program Fidelity Manual: Stan-
5. National Council on Aging Center for Healthy Aging. About Evidence-Based ford Self-Management Programs 2012 Update. Stanfrod, CA: Stanford School of
Programs. Washington, DC: (2014). Available from: http://www.ncoa.org/ Medicine (2012).
improve-health/center-for-healthy-aging/about-evidence-based-programs. 27. United States Department of Agriculture Economic Research Service.
html Rural-Urban Commuting Area Codes: Overview. (2014). Available from:
6. Schulz R, Czaja SJ, McKay JR, Ory MG, Belle SH. Intervention taxonomy http://www.ers.usda.gov/data-products/rural-urban-commuting-area-codes.
(ITAX): describing essential features of interventions. Am J Health Behav (2010) aspx#.U-ObPmPqlZU
34(6):811–21. doi:10.5993/AJHB.34.6.15 28. Glasgow RE, Klesges LM, Dzewaltowski DA, Bull SS, Estabrooks P. The future
7. National Council on Aging Center for Healthy Aging. Module 2: What is of health behavior change research: what is needed to improve translation of
Evidence-Based Health Promotion?: National Council on Aging. (2014). Available research into health promotion practice? Ann Behav Med (2004) 27(1):3–12.
from: http://www.ncoa.org/chamodules/series1/m2/module2.pdf doi:10.1207/s15324796abm2701_2
8. Northwest Center for Public Health Practice at the University of Washington 29. Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of health
School of Public Health. Effective Adult Learning: A Toolkit for Teaching Adults. promotion interventions: the RE-AIM framework. Am J Public Health (1999)
Seattle, WA: Robert Wood Johnson Foundation (2014). 89(9):1322–7. doi:10.2105/AJPH.89.9.1322
9. Burke SM, Carron AV, Patterson MM, Estabrooks PA, Hill JL, Loughead TM, 30. Israel BA, Schults AJ, Parker EA, Becker AB. BeckerReview of community-based
et al. Cohesion as shared beliefs in exercise classes. Small Group Res (2005) research: assessing partnership approaches to improve public health. Annu Rev
36(3):267–88. doi:10.1177/1046496404267942 Public Health (1998) 19:173–202. doi:10.1146/annurev.publhealth.19.1.173
10. Carron AV, Spink KS. The group size-cohesion relationship in minimal groups. 31. Smith M, Belza B, Altpeter M, Ahn S, Dickerson J, Ory M. Disseminating an
Small Group Res (1995) 26(1):86–105. doi:10.1177/1046496495261005 evidence-based disease self-management program for older americans: impli-
11. National Council on Aging. Center for Healthy Aging: Highlights. Wash- cations for diversifying participant reach through delivery site adoption. In:
ington, DC: (2014). Available from: http://www.ncoa.org/improve-health/ Maddock J, editor. Public Health – Social and Behavioral Health. Rijeka: InTech
center-for-healthy-aging/ (2012). p. 385–404.
12. Ahn S, Basu R, Smith ML, Jiang L, Lorig K, Whitelaw N, et al. The impact 32. Smith ML, Ory MG, Ahn S, Belza B, Mingo CA, Towne Jr. SD, et al. Reach-
of chronic disease self-management programs: healthcare savings through a ing diverse participants utilizing a diverse delivery infrastructure: a replication
community-based intervention. BMC Public Health (2013) 13(1):1141. doi:10. study. Front Public Health (2015) 3:77. doi:10.3389/fpubh.2015.00077
1186/1471-2458-13-1141 33. Cooney SM, Small SA, O’Connor C. Strategies for recruiting and retaining par-
13. Smith ML, Hochhalter AK, Cheng Y, Wang S, Ory MG. Programmatic influ- ticipants in prevention programs. What Works, Wisconsin Research to Practice
ences on outcomes of an evidence-based fall prevention program for older Series, 2. Madison, WI: University of Wisconsin–Madison/Extension (2007).
adults: a translational assessment. Transl Behav Med (2011) 1(3):384–93. Available from: http://whatworks.uwex.edu/attachment/whatworks_02.pdf
doi:10.1007/s13142-011-0058-7 34. Salazar CI, Smith ML, Perez AG, Ahn S, Ory MG. Geospatial characteristics of the
14. Ory MG, Altpeter M, Belza B, Helduser J, Zhang C, Smith ML. Perceptions chronic disease self-management program: reaching diverse ethnic populations
about the community application of RE-AIM in the promotion of evidence- in San Antonio, Texas. Tex Public Health J (2011) 63(1):16–20.
based programs for older adults. Eval Health Prof (2014). doi:10.1177/ 35. Smith ML, Wilson MG, DeJoy DM, Padilla H, Zuercher H, Corso P, et al.
0163278714542335 Chronic Disease Self-Management Program in the workplace: opportunities for
15. Ory MG, Smith ML, Kulinski KP, Lorig K, Zenker W, Whitelaw N. Self- health improvement. Front Public Health (2015) 2:179. doi:10.3389/fpubh.2014.
management at the tipping point: reaching 100,000 Americans with evidence- 00179
based programs. J Am Geriatr Soc (2013) 61(5):821–3. doi:10.1111/jgs.12239 36. Ahn S, Smith ML, Altpeter M, Post L, Ory MG. Healthcare cost savings esti-
16. Levkoff S, Sanchez H. Lessons learned about minority recruitment and retention mator tool for chronic disease self-management program: a new tool for pro-
from the centers on minority aging and health promotion. Gerontologist (2003) gram administrators and decision makers. Front Public Health (2015) 3:42.
43(1):18–26. doi:10.1093/geront/43.1.18 doi:10.3389/fpubh.2015.00042

www.frontiersin.org April 2015 | Volume 3 | Article 19 | 93

Smith et al. CDSME program attendance

Conflict of Interest Statement: The authors declare that the research was conducted evaluation and/or evidence-based programming for older adults. Review Editors were
in the absence of any commercial or financial relationships that could be construed independent of named authors on any given article published in this volume.
as a potential conflict of interest.
Received: 08 August 2014; accepted: 17 January 2015; published online: 27 April 2015.
Citation: Smith ML, Ory MG, Jiang L, Lorig K, Kulinski KP and Ahn S (2015) Work-
This paper is included in the Research Topic, “Evidence-Based Programming for Older shop characteristics related to Chronic Disease Self-Management Education program
Adults.” This Research Topic received partial funding from multiple government and attendance. Front. Public Health 3:19. doi: 10.3389/fpubh.2015.00019
private organizations/agencies; however, the views, findings, and conclusions in these This article was submitted to Public Health Education and Promotion, a section of the
articles are those of the authors and do not necessarily represent the official position journal Frontiers in Public Health.
of these organizations/agencies. All papers published in the Research Topic received Copyright © 2015 Smith, Ory, Jiang , Lorig , Kulinski and Ahn. This is an open-access
peer review from members of the Frontiers in Public Health (Public Health Education article distributed under the terms of the Creative Commons Attribution License (CC
and Promotion section) panel of Review Editors. Because this Research Topic repre- BY). The use, distribution or reproduction in other forums is permitted, provided the
sents work closely associated with a nationwide evidence-based movement in the US, original author(s) or licensor are credited and that the original publication in this
many of the authors and/or Review Editors may have worked together previously in journal is cited, in accordance with accepted academic practice. No use, distribution or
some fashion. Review Editors were purposively selected based on their expertise with reproduction is permitted which does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 19 | 94
 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00205

The role of Session Zero in successful completion of

Chronic Disease Self-Management Program workshops
Luohua Jiang 1 *, Matthew Lee Smith 2 , Shuai Chen 3 , SangNam Ahn 4,5 , Kristie P. Kulinski 6 , Kate Lorig 7 and
Marcia G. Ory 5
1
Department of Epidemiology and Biostatistics, School of Public Health, Texas A&M Health Science Center, College Station, TX, USA
2
Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA, USA
3
Department of Statistics, Texas A&M University, College Station, TX, USA
4
Division of Health Systems Management and Policy, School of Public Health, The University of Memphis, Memphis, TN, USA
5
Department of Health Promotion and Community Health Sciences, School of Public Health, Texas A&M Health Science Center, College Station, TX, USA
6
National Council on Aging, Washington, DC, USA
7
Stanford Patient Education Research Center, Department of Medicine, Stanford School of Medicine, Palo Alto, CA, USA

Edited by: Background: The Chronic Disease Self-Management Program (CDSMP) has been widely
Sue Ellen Levkoff, University of South
disseminated among various racial and ethnic populations. In addition to the six required
Carolina, USA
CDSMP workshop sessions, the delivery sites have the option to offer a Session Zero
Reviewed by:
Xiaoguang Ma, University of South (or zero class), an information session offered prior to Session One as a marketing tool.
Carolina, USA Despite assumptions that a zero class is helpful, little is known about the prevalence of
Nilesh Chandrakant Gawde, Tata these additional sessions or their impact on retaining participants in CDSMP workshops.
Institute of Social Sciences, India
This study aims to describe the proportion of CDSMP workshops that offered Session Zero
*Correspondence:
and examine the association between Session Zero and workshop completion rates.
Luohua Jiang, Department of
Epidemiology and Biostatistics, Texas Methods: Data were analyzed from 80,987 middle-aged and older adults collected during
A&M Health Science Center, 219
SRPH Administration Building, TAMU
a two-year national dissemination of CDSMP. Generalized estimating equation regression
1266, College Station, TX 77843-1266, analyses were conducted to assess the association between Session Zero and successful
USA workshop completion (attending four or more of the six workshop sessions).
e-mail: [email protected]
Results: On average, 21.04% of the participants attended workshops that offered Session
Zero, and 75.33% successfully completed the CDSMP workshop. The participants of the
workshops that offered Session Zero had significantly higher odds of completing CDSMP
workshops than those who were not offered Session Zero (OR = 1.099, P = <0.001)
after controlling for participants’ demographic characteristics, race, ethnicity, living status,
household income, number of chronic conditions, and workshop delivery type.
Conclusion: As one of the first studies reporting the importance of an orientation session
for participant retention in chronic disease management intervention projects, our find-
ings suggest offering an orientation session may increase participant retention in similar
translational efforts.
Keywords: attrition, retention, orientation session, evidence-based programs, chronic disease self-management
program

INTRODUCTION intervention benefits received by participants because of the lack

In recent years, an increasing number of interventions have been of adequate intervention dose (6, 7). Less rigorously controlled
deemed highly efficacious in the prevention and management than clinical trials, translational efforts pose special challenges for
of chronic diseases in randomized clinical trials (1). To dissem- participant engagement. For example, grand-scale translational
inate the findings of those clinical trials, the critical next step intervention efforts typically allocate fewer resources to intensively
is to examine whether the research-based studies can be trans- track and follow-up with participants over time, which may impact
lated into effective community-based programs that can recruit retention success.
and retain large numbers of participants with various chronic To address problems associated with participant attrition,
diseases. Program retention is often a challenge in controlled a wide range of studies have investigated factors related to
clinical studies (2), but it can be even more pronounced in large- retaining participants in clinical trials and observational stud-
scale implementation efforts of community-based interventions ies (8–15). A previous meta-analysis identified 12 basic themes
(3–5). Participant attrition not only threatens the internal valid- for successful retention in longitudinal studies, which include
ity and statistical power of a project, but also compromises the community involvement, contact and scheduling methods, and

www.frontiersin.org April 2015 | Volume 2 | Article 205 | 95

Jiang et al. Session Zero and participant retention

financial incentives (16). To date, however, the strategies for INDEPENDENT VARIABLES
successful retention in translational initiatives remain underex- Whether or not a workshop offered a Session Zero was recorded
plored (17, 18). administratively and included in the database along with work-
The Chronic Disease Self-Management Program (CDSMP) has shop attendance. Participants’ actual attendance of a Session Zero
been introduced and widely disseminated into US communities as was not recorded. If offered, the Session Zero was usually offered 1–
a method to empower patients to deal with their chronic condi- 4 weeks prior to the workshop and targeted those who had already
tions by enhancing their self-management skills (19). Drawing registered or who might have shown an interest in the workshop.
upon Social Learning Theory (20), CDSMP is an evidence-based, This orientation session was also used to recruit acquaintances
peer-led intervention consisting of six highly participative classes and/or family members of those who already registered for the
held for 2.5 h each, once a week, for six consecutive weeks (19). In workshop. The specific content of Session Zero varied by site; how-
addition to the six workshop sessions, some delivery sites are offer- ever, all of them should have provided an overview of the CDSMP
ing a Session Zero (or zero class), an information session offered workshop and its expectations for participation. Session Zero may
prior to Session One as a marketing tool (21). The primary pur- also be used to collect baseline data to reduce interference with
pose of Session Zero is to provide an overview of the workshop, Session One of the workshop.
explain expectations for workshop participation, and confirm Workshop delivery sites included area agencies on aging
commitment of those who are interested in or have already reg- (AAAs), healthcare organizations, residential facilities, community
istered for a workshop. This additional session also serves as an or multipurpose centers, faith-based organizations, educational
opportunity to collect baseline data from participants to alleviate institution, county health department, tribal center, workplace,
administrative burden on workshop instructors and ensure time and other (e.g., recreational center).
is not taken away from Session One of the workshop. Although Socio-demographic factors included age (in years), sex (male
designed as a recruitment tool, we believe that incorporating a vs. female), median household income (in $10,000 units), and
Session Zero to CDSMP workshops may boost participant reten- living arrangement (living with others vs. living alone). Partici-
tion rates because those who were not firmly committed to the pants’ health status was measured by their number of self-reported
workshop might decide to opt out of the program at this time. chronic conditions (i.e., arthritis, cancer, depression, diabetes,
The goals of the current study are to (1) describe the proportion heart disease, hypertension, lung disease, stroke, osteoporosis, and
of CDSMP workshops that offered Session Zero and (2) examine other chronic conditions).
the association between Session Zero and workshop completion
rates. STATISTICAL ANALYSIS
To compare the characteristics of the participants who attended
workshops with a Session Zero and those who attended work-
MATERIALS AND METHODS
shops without a Session Zero, we used χ2 tests for categorical
DATA SOURCE AND STUDY POPULATION
variables and two-sample t -tests for continuous variables. Because
Data for this study were obtained from a two-year nationwide
the participants were nested in workshops, generalized estimating
delivery of CDSMP as part of the American Recovery and Rein-
equation (GEE) regression models were employed to investigate
vestment Act of 2009 (i.e., Recovery Act) Communities Putting
the association between successful workshop completion and Ses-
Prevention to Work: Chronic Disease Self-Management Program
sion Zero attendance. Specifically, the dependent variable of these
Initiative (22). The U.S. Administration on Aging led this initia-
regression models was successful workshop completion, while the
tive in collaboration with the Centers for Disease Control and
independent variables were participant-level demographic and
Prevention (CDC) and the Centers for Medicare and Medicaid
health characteristics. Furthermore, delivery site type was also
Services (CMS) to support the translation of CDSMP in 45 states,
included as an independent variable in the second GEE regres-
Puerto Rico, and the District of Columbia (23). This initiative
sion model. All the models included an exchangeable working
was executed between 2010 and 2012 to embed CDSMP delivery
covariance to account for the intraclass correlation among partic-
structures into statewide systems (22). Within the first two years
ipants from the same workshop. Because the dependent variable is
of this initiative, more than 100,000 adults participated in 9305
a binary variable; GEE analyses were conducted using SAS GEN-
workshops in 1234 U.S. counties (22). For this study, administra-
MOD procedure with a logit link function (SAS 9.3, SAS Institute,
tive records were utilized to determine whether or not a Session
Inc., Cary, NC, USA).
Zero was held. Data were analyzed from 80,987 participants aged
50 years or older whose programmatic records contained data RESULTS
about Session Zero attendance. Table 1 shows the proportions of participants who attended Ses-
sion Zero and the workshop completion rates among the 10 types
MEASURES of delivery sites. Overall, 21.04% of the participants attended
DEPENDENT VARIABLE workshops with a Session Zero and 75.33% of participants suc-
CDSMP workshop attendance was the dependent variable for this cessfully completed the CDSMP workshop. Among the 10 differ-
study. As defined by the program developers and used in a vari- ent types of delivery sites, the largest proportion of participants
ety of studies (24, 25), successful completion was defined as when attending workshops with a Session Zero were at residential facili-
CDSMP participants attended four or more of the six workshop ties (26.27%), while the smallest proportion of participants attend-
sessions (22, 26), excluding Session Zero. ing workshops with a Session Zero were at tribal centers (9.76%).

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 205 | 96
Jiang et al. Session Zero and participant retention

Table 1 | Session Zero attendance and CDSMP workshop completion With respect to workshop completion rates, workplaces had the
rates by delivery site type. highest completion rate (82.44%) and tribal center had the lowest
completion rate (69.76%).
Workshop delivery site Total, N (%) Attended CDSMP As presented in Table 2, CDSMP participants who attended
workshops completion workshops with a Session Zero had significantly higher workshop
with a Session (%) completion rate than those who attended workshops without a
Zero (%) Session Zero (77.85% vs. 74.66%, P < 0.001). Participants who
Senior Center/AAA 24,653 (30.44) 25.81 77.27
attended workshops with a Session Zero were more likely to be
female, African American or other race group, Hispanic, and live
Health care organization 15,026 (18.55) 10.71 72.53
alone. In terms of chronic conditions, they were more likely to have
Residential facility 14,439 (17.83) 26.27 70.03 diabetes and hypertension, but less likely to have arthritis, cancer,
Community/multipurpose 8303 (10.25) 21.63 75.74 depression, and lung disease. The average numbers of chronic
Faith-based organization 7127 (8.80) 22.25 78.88 conditions were not significantly different based on Session Zero
Educational institution 1844 (2.28) 17.35 77.77 status. Finally, the participants who attended workshops with a
Session Zero were significantly older and had lower household
County health department 1013 (1.25) 19.64 69.89
incomes.
Tribal center 205 (0.25) 9.76 69.76 Table 3 illustrates the results of GEE regressions for workshop
Workplace 410 (0.51) 18.05 82.44 completion. As seen in Model 1, the participants of the workshops
Other 7967 (9.84) 16.08 80.48 that offered Session Zero had significantly higher odds of com-
Total 80,987 (100.00) 21.04 75.33 pleting CDSMP workshops than those who participated the work-
shops that did not offer a Session Zero (odds ratio [OR] = 1.087,
CDSMP, chronic disease self-management program; AAA, area agency on aging. P = 0.003). In addition, the likelihood of completing CDSMP

Table 2 | Baseline characteristics of CDSMP participants by Session Zero status.

Total (n = 80,987) Attended workshops Attended workshops P

without a Session with a Session
Zero (n = 63,946) Zero (n = 17,041)

Workshop completion 61,007 (75.33%) 47,740 (74.66%) 13,267 (77.85%) <0.001

Female 59,669 (78.17%) 46,571 (77.47%) 13,098 (80.77%) <0.001

Race <0.001
White 45,673 (65.10%) 37,430 (67.63%) 8243 (55.64%)
African American 15,929 (22.70%) 11,438 (20.67%) 4491 (30.32%)
Asian/Pacific Islander 3057 (4.36%) 2511 (4.54%) 546 (3.69%)
American Indian/Alaskan Native 1156 (1.65%) 969 (1.75%) 187 (1.26%)
Other 4344 (6.19%) 2997 (5.42%) 1347 (9.09%)

Hispanic 10,771 (15.78%) 7328 (13.49%) 3443 (24.74%) <0.001

Living alone 10,968 (13.55%) 7868 (12.32%) 3100 (18.19%) <0.001

Chronic conditions
Arthritis 34,769 (42.93%) 27,672 (43.27%) 7097 (41.65%) <0.001
Cancer 7585 (9.37%) 6135 (9.59%) 1450 (8.51%) <0.001
Depression 16,729 (20.66%) 13,819 (21.61%) 2910 (17.08%) <0.001
Diabetes 26,033 (32.14%) 19,453 (30.42%) 6580 (38.61%) <0.001
Heart disease 13,480 (16.64%) 10,630 (16.62%) 2850 (16.72%) 0.753
Hypertension 36,531 (45.11%) 28,647 (44.80%) 7884 (46.26%) <0.001
Lung disease 14,045 (17.34%) 11,231 (17.56%) 2814 (16.51%) 0.001
Stroke 4220 (5.21%) 3316 (5.19%) 904 (5.30%) 0.534

Mean (±SD) Mean (±SD) Mean (±SD)

Age 67.03 (±14.60) 66.58 (±14.79) 69.87 (±13.03) <0.001
Median income 5.07 (±1.30) 5.02 (±1.27) 4.87 (±1.40) <0.001
Number of chronic conditions 2.20 (±1.71) 2.29 (±1.71) 2.26 (±1.70) 0.060

CDSMP, Chronic Disease Self-Management Program.

www.frontiersin.org April 2015 | Volume 2 | Article 205 | 97

Jiang et al. Session Zero and participant retention

Table 3 | Generalized estimating equation regression models for successful workshop completion.

Model 1 Model 2

OR (95% CI) P OR (95% CI) P

Session Zero offered 1.087 (1.030, 1.147) 0.003 1.099 (1.041, 1.161) <0.001
Age 1.002 (1.001, 1.004) 0.003 1.004 (1.002, 1.005) <0.001
Female 1.089 (1.039, 1.141) <0.001 1.105 (1.054, 1.158) <0.001

Race <0.001a
White 0.989 (0.904, 1.082) 0.805 0.986 (0.900, 1.079) 0.751
African American 1.147 (0.904, 1.082) 0.007 1.133 (1.025, 1.251) 0.014
Asian/Pacific Islander 1.354 (1.179, 1.554) <0.001 1.293 (1.126, 1.485) < 0.001
American Indian/Alaska Native 0.916 (0.771, 1.088) 0.318 0.908 (0.760, 1.084) 0.284
Other Ref NA Ref NA
Hispanic 1.171 (1.087,1.263) <0.001 1.145 (1.063,1.236) <0.001
Living Alone 1.084 (0.977, 1.202) 0.127 0.930 (0.838, 1.031) 0.168
Median Income 0.979 (0.963, 0.994) 0.008 0.988 (0.972, 1.004) 0.128
Number of chronic conditions 1.005 (0.993, 1.017) 0.455 1.010 (0.998, 1.023) 0.093
Workshop delivery site <0.001a
Senior Center/AAA Ref NA
Health Care Organization 0.837 (0.785, 0.893) <0.001
Residential Facility 0.696 (0.655, 0.738) <0.001
Community/Multipurpose 0.901 (0.835, 0.971) 0.007
Faith-based organization 1.150 (1.057, 1.251) 0.001
Educational institution 1.003 (0.871, 1.155) 0.968
County health department 0.797 (0.666, 0.954) 0.013
Tribal center 0.815 (0.559, 1.187) 0.286
Workplace 1.627 (1.166, 2.271) 0.004
Other 1.260 (1.158, 1.370) <0.001

a
Overall Type 3 P value.
AAA, area agency on aging.

workshops was higher among older participants (OR = 1.002, served by each delivery site type (27, 28), as well as site staff
P = 0.003), females (OR = 1.089, P < 0.001), African Americans availability and facility constraints (e.g., space, time, competing
(OR = 1.147, P = 0.007),Asians and Pacific Islanders (OR = 1.354, commitments).
P < 0.001), and Hispanics (OR = 1.171, P < 0.001). Conversely, As suggested in a review of lessons learned from the National
the likelihood of completing the workshop was lower among those Institute of Aging’s Behavior Change Consortia (21), this study
with higher household incomes (OR = 0.979, P = 0.008). also demonstrates that participants who were offered orientation
After adding types of delivery site into the GEE regression sessoins were more likely to complete intervention protocols. This
model (Model 2), we found participants of the workshops that finding indicates offering a Session Zero may not only facilitate
offered Session Zero still had significantly higher odds of complet- participant recruitment, but also increase participant retention
ing CDSMP workshops (OR = 1.099, P < 0.001). Furthermore, in grand scale community-based program dissemination efforts.
the average workshop completion rates were significantly different Participants attending workshops with a Session Zero before the
among different delivery site types (P < 0.001), with residen- formal start of the workshop might have developed more sup-
tial facility had the lowest likelihood of completing the work- port for and positive views of the program because they were
shop (OR = 0.696, P < 0.001) while workplace had the highest given an opportunity to better understand the purpose, content,
likelihood (OR = 1.627, P = 0.004). and expectations of the workshop. Meanwhile, attending a Session
Zero may have given individuals who were not fully committed to
DISCUSSION the program a chance to re-evaluate their intention and opt out
The results of the current study show about one in five CDSMP of the program if they felt they were not completely ready for it
workshops in this national initiative offered a Session Zero. Among or thought it might not be beneficial for their preferences/needs.
the 10 delivery site types, senior centers/AAAs and residential facil- Therefore, the functions of a Session Zero with respect to reten-
ities had the highest rates of offering a Session Zero, while tribal tion might be twofold: (1) to strengthen the commitment of the
centers and healthcare organizations had the lowest rates. These participants by providing relevant information in advance and
differences might be related to variation in population subgroups (2) to serve as a screening tool to identify those who are truly

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 205 | 98
Jiang et al. Session Zero and participant retention

interested in the program and ready to participate. Future studies efficient two-prong approach to help with participant retention in
are warranted to study the details of these two potential functions future translational projects.
of Session Zero.
Our results regarding the relationships between participants’
ACKNOWLEDGMENTS
demographic characteristics and retention are consistent with the
This work was supported by the Administration on Aging
existing literature (11–13, 29–32). Specifically, we found that older
[90OP0001/03] and the National Institute of Child Health and
participants and females had higher workshop completion rates.
Human Development [R01HD047143].
The relationship between race/ethnicity and participant retention
in previous studies are mixed, although most report minority pop-
ulations were harder to retain (33), some have reported relatively REFERENCES
lower attrition rates among Hispanic participants (8, 10, 34). Here, 1. Brady TJ, Murphy L, O’Colmain BJ, Beauchesne D, Daniels B, Greenberg M,
et al. A meta-analysis of health status, health behaviors, and healthcare utiliza-
we found that African American, Asian Pacific Islanders, and His- tion outcomes of the chronic disease self-management program. Prev Chronic
panic participants had higher likelihood of completing the work- Dis (2013) 10:120112. doi:10.5888/pcd10.120112
shop successfully. Differences in CDSMP workshop attendance 2. Warner ET, Glasgow RE, Emmons KM, Bennett GG, Askew S, Rosner B, et al.
rates by participant demographics may have been reflective of the Recruitment and retention of participants in a pragmatic randomized interven-
type of delivery site at which workshops were attended. Other tion trial at three community health clinics: results and lessons learned. BMC
Public Health (2013) 13:192. doi:10.1186/1471-2458-13-192
studies have shown that certain delivery sites attract and serve 3. Escobar-Chaves SL, Tortolero SR, Masse LC, Watson KB, Fulton JE. Recruiting
different community subgroups (22, 27, 28). For example, better and retaining minority women: findings from the women on the move study.
workshop attendance among older participants may reflect that Ethn Dis (2002) 12(2):242–51.
larger proportions of older participants attend CDSMP workshops 4. Coday M, Boutin-Foster C, Goldman Sher T, Tennant J, Greaney ML, Saunders
SD, et al. Strategies for retaining study participants in behavioral intervention
at residential facilities (also with higher attendance rates).
trials: retention experiences of the NIH behavior change consortium. Ann Behav
The strengths of this study include a large sample size and Med (2005) 29(Suppl):55–65. doi:10.1207/s15324796abm2902s_9
diverse race and ethnicity representation included in the analy- 5. Glasgow RE, Nelson CC, Kearney KA, Reid R, Ritzwoller DP, Strecher VJ, et al.
sis. The large sample not only allowed us to have high power Reach, engagement, and retention in an Internet-based weight loss program in
to detect relatively small differences and associations, but also a multi-site randomized controlled trial. J Med Internet Res (2007) 9(2):e11.
doi:10.2196/jmir.9.2.e11
implies potential good generalizability of our findings. Further-
6. Mody L, Miller DK, McGloin JM, Freeman M, Marcantonio ER, Magaziner J,
more, the study sample included 16% Hispanics, 23% African et al. Recruitment and retention of older adults in aging research. J Am Geriatr
Americans, 4% Asian and Pacific Islanders, and 2% American Soc (2008) 56(12):2340–8. doi:10.1111/j.1532-5415.2008.02015.x
Indian and Alaskan Natives. The substantial diversity of the sample 7. King AC, Rejeski WJ, Buchner DM. Physical activity interventions targeting
composition further supports the generalizability of these study older adults. A critical review and recommendations. Am J Prev Med (1998)
15(4):316–33. doi:10.1016/S0749-3797(98)00085-3
results. 8. Hessol NA, Schneider M, Greenblatt RM, Bacon M, Barranday Y, Holman S,
Despite the study’s evident strengths, our findings need to be et al. Retention of women enrolled in a prospective study of human immun-
carefully interpreted in light of a few limitations. First, while data odeficiency virus infection: impact of race, unstable housing, and use of
were collected that indicated whether or not workshops offered human immunodeficiency virus therapy. Am J Epidemiol (2001) 154(6):563–73.
a Session Zero, participant attendance in these zero class sessions doi:10.1093/aje/154.6.563
9. Williams PL, Van Dyke R, Eagle M, Smith D, Vincent C, Ciupak G, et al.
was not recorded. This means that participant attendance in a Association of site-specific and participant-specific factors with retention of
Session Zero could not be directly linked to their workshop atten- children in a long-term pediatric HIV cohort study. Am J Epidemiol (2008)
dance data. Second, these results were based on observational 167(11):1375–86. doi:10.1093/aje/kwn072
data, which limit our ability to determine a causal relationship 10. McGuigan WM, Katzev AR, Pratt CC. Multi-level determinants of retention
in a home-visiting child abuse prevention program. Child Abuse Negl (2003)
between Session Zero attendance and CDSMP completion. Third,
27(4):363–80. doi:10.1016/S0145-2134(03)00024-3
the number of variables collected at baseline was relatively small; 11. O’Brien RA, Moritz P, Luckey DW, McClatchey MW, Ingoldsby EM, Olds DL.
therefore, although we were able to control for several important Mixed methods analysis of participant attrition in the nurse-family partnership.
participant characteristics in the regression analyses, the iden- Prev Sci (2012) 13(3):219–28. doi:10.1007/s11121-012-0287-0
tified associations in this study may be confounded by other 12. Anderson RT, Ory M, Cohen S, McBride JS. Issues of aging and adher-
ence to health interventions. Control Clin Trials (2000) 21(5 Suppl):171S–83S.
unmeasured variables. Last, we were not able to investigate the rela-
doi:10.1016/S0197-2456(00)00075-1
tionship between Session Zero attendance and changes in health- 13. Brown DR, Fouad MN, Basen-Engquist K, Tortolero-Luna G. Recruitment and
and healthcare-related outcomes among these CDSMP partici- retention of minority women in cancer screening, prevention, and treatment
pants because outcome measures were not available in this data trials. Ann Epidemiol (2000) 10(8 Suppl):S13–21. doi:10.1016/S1047-2797(00)
collection effort. 00197-6
14. Manson SM, Jiang L, Zhang L, Beals J, Acton KJ, Roubideaux Y. Special diabetes
In summary, including a Session Zero when delivering CDSMP program for Indians: retention in cardiovascular risk reduction. Gerontologist
workshops may be an important strategy for participant retention. (2011) 51(Suppl 1):S21–32. doi:10.1093/geront/gnq083
Our findings suggest hosting a Session Zero may have implications 15. Jiang L, Manson SM, Dill EJ, Beals J, Johnson A, Huang H, et al. Participant
for workshop attendance in similar translational efforts involv- and site characteristics related to participant retention in a diabetes prevention
ing evidence-based programs for older adults. Given potential translational project. Prev Sci (2014). doi:10.1007/s11121-013-0451-1
16. Robinson KA, Dennison CR, Wayman DM, Pronovost PJ, Needham DM. Sys-
challenges associated with retaining participants in grand scale tematic review identifies number of strategies important for retaining study
community-based interventions, offering a Session Zero before participants. J Clin Epidemiol (2007) 60(8):757–65. doi:10.1016/j.jclinepi.2006.
the formal start of the intervention might represent a feasible and 11.023

www.frontiersin.org April 2015 | Volume 2 | Article 205 | 99

Jiang et al. Session Zero and participant retention

17. Rosal MC, White MJ, Borg A, Scavron J, Candib L, Ockene I, et al. Trans- research. Gerontologist (2003) 43(Spec No 1):37–46. doi:10.1093/geront/43.
lational research at community health centers: challenges and successes in suppl_1.37
recruiting and retaining low-income Latino patients with type 2 diabetes into 31. Chang MW, Brown R, Nitzke S. Participant recruitment and retention in a pilot
a randomized clinical trial. Diabetes Educ (2010) 36(5):733–49. doi:10.1177/ program to prevent weight gain in low-income overweight and obese mothers.
0145721710380146 BMC Public Health (2009) 9:424. doi:10.1186/1471-2458-9-424
18. Davis RM, Hitch AD, Nichols M, Rizvi A, Salaam M, Mayer-Davis EJ. A collabora- 32. Gappoo S, Montgomery ET, Gerdts C, Naidoo S, Chidanyika A, Nkala B,
tive approach to the recruitment and retention of minority patients with diabetes et al. Novel strategies implemented to ensure high participant retention rates
in rural community health centers. Contemp Clin Trials (2009) 30(1):63–70. in a community based HIV prevention effectiveness trial in South Africa and
doi:10.1016/j.cct.2008.09.007 Zimbabwe. Contemp Clin Trials (2009) 30(5):411–8. doi:10.1016/j.cct.2009.05.
19. Lorig K, Holman H, Sobel D, Laurent D. Living a Healthy Life with Chronic Con- 002
ditions: Self Management of Heart Disease, Arthritis, Diabetes, Asthma, Bronchitis, 33. Probstfield JL, Frye RL. Strategies for recruitment and retention of participants
Emphysema and Others. 3rd ed. Boulder, CO: Bull Publishing Company (2006). in clinical trials. JAMA (2011) 306(16):1798–9. doi:10.1001/jama.2011.1544
20. Bandura A. Social cognitive theory of self-regulation. Organ Behav Hum Decis 34. Smith ML, Ahn S, Jiang L, Kulinski KP, Ory MG. Factors associated with
Process (1991) 50(2):248–87. doi:10.1016/0749-5978(91)90022-L Hispanic adults attending Spanish-language disease self-management pro-
21. Frank JC, Coviak CP, Healy TC, Belza B, Casado BL. Addressing fidelity in gram workshops and workshop completion. Front Public Health (2015) 2:155.
evidence-based health promotion programs for older adults. J Appl Geron (2008) doi:10.3389/fpubh.2014.00155
27(1):14–33. doi:10.1177/1090198114543007
22. Ory MG, Smith ML, Patton K, Lorig K, Zenker W, Whitelaw N. Self-management
at the tipping point: reaching 100,000 Americans with evidence-based programs. Conflict of Interest Statement: The authors declare that the research was conducted
J Am Geriatr Soc (2013) 61(5):821–3. doi:10.1111/jgs.12239 in the absence of any commercial or financial relationships that could be construed
23. U.S. Department of Health and Human Services Administration on Aging. as a potential conflict of interest.
ARRA – Communities Putting Prevention to Work: Chronic Disease Self-
Management Program 2012 April 15 (2014). Available from: www.cfda.gov/?s= This paper is included in the Research Topic, “Evidence-Based Programming for Older
program&mode=form&tab=step1&id=5469a61f2c5f25cf3984fc3b94051b5f Adults.” This Research Topic received partial funding from multiple government and
24. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, Whitelaw N, et al. Successes private organizations/agencies; however, the views, findings, and conclusions in these
of a national study of the chronic disease self-management program: meet- articles are those of the authors and do not necessarily represent the official position
ing the triple aim of health care reform. Med Care (2013) 51(11):992–8. of these organizations/agencies. All papers published in the Research Topic received
doi:10.1097/MLR.0b013e3182a95dd1 peer review from members of the Frontiers in Public Health (Public Health Education
25. Standford University, Patient Education in the Department of Medicine. Avail- and Promotion section) panel of Review Editors. Because this Research Topic repre-
able from: http://patienteducation.stanford.edu/programs/cdsmp.html sents work closely associated with a nationwide evidence-based movement in the US,
26. Ahn S, Basu R, Smith ML, Jiang L, Lorig K, Whitelaw N, et al. The impact many of the authors and/or Review Editors may have worked together previously in
of chronic disease self-management programs: healthcare savings through a some fashion. Review Editors were purposively selected based on their expertise with
community-based intervention. BMC Public Health (2013) 13(1):1141. doi:10. evaluation and/or evidence-based programming for older adults. Review Editors were
1186/1471-2458-13-1141 independent of named authors on any given article published in this volume.
27. Smith ML, Belza B, Altpeter M, Ahn S, Dickerson JB, Ory MG. Disseminat-
ing an evidence-based disease self-management program for older Americans: Received: 16 July 2014; accepted: 07 October 2014; published online: 27 April 2015.
implications for diversifying participant reach through delivery site adoption. In: Citation: Jiang L, Smith ML, Chen S, Ahn S, Kulinski KP, Lorig K and Ory MG (2015)
Maddock J, editor. Public Health: Social and Behavioral Health. Rijeka, Croatia: The role of Session Zero in successful completion of Chronic Disease Self-Management
InTech (2012). p. 385–404. Program workshops. Front. Public Health 2:205. doi: 10.3389/fpubh.2014.00205
28. Smith ML, Ory MG, Ahn S, Belza B, Mingo CA, Towne Jr. SD, et al. Reach- This article was submitted to Public Health Education and Promotion, a section of the
ing diverse participants utilizing a diverse delivery infrastructure: a replication journal Frontiers in Public Health.
study. Front Public Health (2015) 3:77. doi:10.3389/fpubh.2015.00077 Copyright © 2015 Jiang , Smith, Chen, Ahn, Kulinski, Lorig and Ory. This is an open-
29. Bailey JM, Bieniasz ME, Kmak D, Brenner DE, Ruffin MT. Recruitment and access article distributed under the terms of the Creative Commons Attribution License
retention of economically underserved women to a cervical cancer prevention (CC BY). The use, distribution or reproduction in other forums is permitted, provided
trial. Appl Nurs Res (2004) 17(1):55–60. doi:10.1016/j.apnr.2003.12.002 the original author(s) or licensor are credited and that the original publication in this
30. Warren-Findlow J, Prohaska TR, Freedman D. Challenges and opportunities in journal is cited, in accordance with accepted academic practice. No use, distribution or
recruiting and retaining underrepresented populations into health promotion reproduction is permitted which does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 205 | 100
 ORIGINAL RESEARCH
published: 27 April 2015
doi: 10.3389/fpubh.2015.00077

Reaching diverse participants

utilizing a diverse delivery
infrastructure: a replication study
Matthew Lee Smith 1* , Marcia G. Ory 2 , SangNam Ahn 2,3 , Basia Belza 4 , Chivon A. Mingo 5 ,
Samuel D. Towne Jr. 2 and Mary Altpeter 6
1
Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA,
USA, 2 Department of Health Promotion and Community Health Sciences, School of Public Health, Texas A&M Health
Science Center, College Station, TX, USA, 3 Division of Health Systems Management and Policy, School of Public Health, The
University of Memphis, Memphis, TN, USA, 4 Health Promotion Research Center, School of Nursing and School of Public
Health, University of Washington, Seattle, WA, USA, 5 Gerontology Institute, College of Arts and Sciences, Georgia State
University, Atlanta, GA, USA, 6 Center for Health Promotion and Disease Prevention, University of North Carolina at Chapel
Hill, Chapel Hill, NC, USA

This replication study examines participant recruitment and program adoption aspects
Edited by: of disease self-management programs by delivery site types. Data were analyzed
Sue Ellen Levkoff, from 58,526 adults collected during a national dissemination of the Stanford suite of
University of South Carolina, USA
chronic disease self-management education programs spanning 45 states, the District
Reviewed by:
Diane D. Stadler, of Columbia, and Puerto Rico. Participant data were analyzed using multinomial logistic
Oregon Health & Science University, regression to generate profiles by delivery site type. Profiles were created for the five
USA
Cathy H. Gong,
leading delivery site types, which included senior centers or area agencies on aging,
Australian National University, residential facilities, healthcare organizations, community or multi-purpose centers, and
Australia
faith-based organizations. Significant variation in neighborhood characteristics (e.g., rural-
*Correspondence:
ity, median household income, percent of the population age 65 years and older, percent
Matthew Lee Smith,
Department of Health Promotion and of the population i.e., non-Hispanic white) and participant characteristics (e.g., age,
Behavior, College of Public Health, sex, ethnicity, race, rurality) were observed by delivery site type. Study findings confirm
The University of Georgia, Health
Science Center, #345E Wright Hall,
that these evidence-based programs are capable of reaching large numbers of diverse
Athens, GA 30602, USA participants through the aging services network. Given the importance of participant
[email protected]
reach and program adoption to the success of translational research dissemination
initiatives, these findings can assist program deliverers to create strategic plans to engage
Specialty section:
This article was submitted to Public
community partners to diversify their participant base.
Health Education and Promotion,
Keywords: chronic disease self-management, evidence-based program, participant reach, program
a section of the journal Frontiers in
implementation
Public Health
Received: 10 March 2015
Accepted: 17 April 2015
Published: 27 April 2015
Introduction
Citation: The grand-scale dissemination of evidence-based programs in community settings is contingent
Smith ML, Ory MG, Ahn S, Belza B, upon the presence of a delivery infrastructure capable of serving a large and diverse set of partic-
Mingo CA, Towne SD Jr and
ipants. Developing and nurturing the delivery infrastructure is important to ensure a reliable and
Altpeter M (2015) Reaching diverse
participants utilizing a diverse delivery
sustainable community presence. Thus, in practice, community partners are encouraged to utilize
infrastructure: a replication study. many different delivery site types to ensure programs are available across geographic space. This
Front. Public Health 3:77. ensures a greater likelihood that programs are offered close to participants’ residences in familiar
doi: 10.3389/fpubh.2015.00077 settings that are easy to access (1, 2).

Frontiers in Public Health | www.frontiersin.org 101 April 2015 | Volume 3 | Article 77

Smith et al. Diverse delivery infrastructure

The types of agencies and organizations that adopt and deliver The aim of this replication study is to generate participant pro-
evidence-based programs can influence the types of participants files by delivery site types to assess common and unique recruit-
reached. As postulated by the RE-AIM Framework (3, 4), program ment characteristics using chronic disease self-management edu-
adoption and participant reach are closely related because a larger cation (CDSME) program data collected in 45 states, Puerto Rico,
number of participants can participate in a program if more and the District of Columbia from 2010 to 2012. To replicate
organizations adopt it and deliver workshops across a particular previous assessments with a more recent and expanded popula-
community or service region. Because certain organizations and tion frame (1), the purposes of this study were to: (1) describe
delivery sites typically serve constituents with varying charac- CDSMP delivery site types in terms of their workshop and
teristics (e.g., age, sex, race/ethnicity), diversifying the types of neighborhood-level characteristics; and (2) describe the personal
delivery sites offering workshops has potential to increase diver- and neighborhood-level characteristics of adults who enrolled in
sity among evidence-based program participants (1). Further, CDSME programs by delivery site type.
people are more likely to enroll in programs/services that are in
closer proximity to their residence. Thus, expanding the number
Materials and Methods
of engaged delivery sites spanning the geographic service region
may increase participant enrollment and program completion Program Description
(i.e., increase attendance to ensure adequate intervention dose is The CDSMP has been introduced and widely disseminated in the
received) (1). U.S. as a method to empower patients with self-management skills
This important issue was first examined in a study using data to deal with their chronic conditions (5). CDSMP is an evidence-
from the Administration on Aging (AoA)’s translation of the based, peer-led intervention consisting of six highly participative
Evidence-Based Disease and Disability Prevention (EBDDP) pro- classes held for 2.5 h each, once a week, for six consecutive weeks
gram collected through the aging services network in community- (5). CDSMP has resulted in improved healthcare and health (6, 7),
based settings (1). This federal funding for evidence-based while potentially saving healthcare costs (8). There is now a suite
programs facilitated the delivery of Chronic Disease Self- of CDSME programs licensed through the Stanford Patient Edu-
Management Program (CDSMP) in 27 States from 2006 to 2009, cation Research Center, some of which are generic (e.g., CDSMP,
which resulted in the development of a delivery infrastructure Tomando Control de su Salud) and others that are disease specific
for evidence-based programs to serve older adults in various (e.g., diabetes, arthritis, chronic pain). While the chronic condi-
community-based settings. The AoA led the EBDDP initiative in tion may vary, all of these programs are based upon the social
partnership with the Centers for Disease Control and Prevention learning theory (9), highly interactive, and apply the principles of
(CDC), Agency for Healthcare Research & Quality (AHRQ), Cen- goal setting, problem solving, and action planning.
ters for Medicare and Medicaid Services (CMS), Health Resources
& Services Administration (HRSA), Substance Abuse & Mental
Health Services Administration (SAMHSA), and over 30 private
Data Source and Study Population
foundations (1). Cross-sectional data for this study were obtained from a nation-
Findings from the initial study indicated that different delivery wide delivery of CDSME programs as part of the American Recov-
sites served areas with different neighborhood-level characteris- ery and Reinvestment Act of 2009 (i.e., Recovery Act) Communi-
tics and participants with different personal and neighborhood- ties Putting Prevention to Work: Chronic Disease Self-Management
level characteristics (1). More specifically, the initial study found Program initiative (10). The U.S. Administration on Aging led
that, relative to workshops delivered at senior centers/area agen- this initiative in collaboration with the CDC and CMS to support
cies on aging (AAA), the other delivery sites (i.e., residential the translation of CDSMP in 45 states, Puerto Rico, and the
facilities, healthcare organizations, community or multi-purpose District of Columbia (11). This initiative was originally designed
centers, and faith-based organizations) were less likely to be to have 50,000 Americans complete at least four out of six CDSMP
offered in rural areas. Workshops delivered at healthcare orga- sessions between 2010 and 2012 and to embed CDSMP delivery
nizations, community or multi-purpose centers, and faith-based structures into statewide systems (10). For this study, data were
organizations were more likely to be in more affluent areas. And, analyzed from the first 100,000 participants who attended CDSMP
workshops in residential facilities and faith-based organizations workshops and had complete data for study variables of interest.
were offered in areas with more non-Hispanic White residents As in the 2006–2009 initiative, systematic outcome data collection
compared to those offered in senior centers/AAA and community was not required because CDSMP is an evidence-based program.
or multi-purpose centers. In terms of participant characteristics, Thus, health-related outcomes are not reported in this study.
relative to workshops delivered at senior centers/AAA, healthcare Institutional Review Board approval for this study was obtained
organizations, community or multi-purpose centers, and faith- through Texas A&M University.
based organizations were more likely to reach younger partici-
pants. Healthcare organizations and community or multi-purpose Measures
centers were more likely to reach male participants. Community Delivery Site Types
or multi-purpose centers and faith-based organizations were more Data pertaining to CDSMP delivery site types were administra-
likely to reach African American participants, and healthcare tively collected (12). Only data from participants attending work-
facilities and faith-based organizations were more likely to reach shops in the five most prevalent delivery site types accounting for
participants residing in less affluent areas. approximately 85% of classes were compared in these analyses:

Frontiers in Public Health | www.frontiersin.org 102 April 2015 | Volume 3 | Article 77

Smith et al. Diverse delivery infrastructure

senior centers or AAA, residential facilities, healthcare organi- senior center or AAA sites served as the referent group). Odds
zations, community or multi-purpose centers, and faith-based ratios and 95% confidence intervals are reported.
organizations. These five delivery site types were consistent with
those included in the previous study (1). Results
Neighborhood Characteristics Neighborhood Characteristics by Delivery Site
Using participants’ residential ZIP Codes, geographic informa-
Type
tion system (GIS) software was used to generate neighborhood-
Neighborhood characteristics of the delivery site types are pre-
level variables for each participant. Neighborhood characteristics
sented in Table 1. Of the 58,526 participants included in this
included residential rurality (i.e., metro residence or non-metro
study, 36.5% attended workshops at senior centers or AAA,
residence based on the rural-urban commuting area codes
21.5% at residential facilities, 19.8% at healthcare organizations,
[RUCA]), median household income for residents residing in
12.1% at community or multi-purpose centers, and 10.2% at
the participants’ ZIP Code (i.e., interpreted in increments of
faith-based organizations. Seventy-nine percent of participants
$10,000), the percent of residents aged 65 years and older residing
attended workshops delivered in metro areas. On average, par-
in the participants’ ZIP Code, and the percent of non-Hispanic
ticipants attended workshops delivered in ZIP Codes where the
White residents residing in the participants’ ZIP Code (13). Using
median household income was $50,400 (±$13,070) and in areas
organizational ZIP Codes, GIS software was used to generate
where 13.9% (±5.6%) of the population was aged 65 years and
neighborhood-level variables for each delivery site (i.e., site rural-
older. On average, participants attended workshops delivered in
ity, median household income, percent of residents aged 65 years
ZIP Codes comprised of 69.0% (±25.4%) non-Hispanic White
and older, and percent of non-Hispanic White participants).
residents.
Personal Characteristics When comparing these neighborhood characteristics by deliv-
Personal characteristics of the participants included age, sex, race ery site type, a larger proportion of workshops in non-metro
(i.e., non-Hispanic White, African American, Asian or Pacific areas were delivered in healthcare organizations (23.6%), senior
Islander, American Indian or Alaska Native, Other/multiple centers/AAA (23.1%), and faith-based organizations (21.6%)
races), and ethnicity (i.e., Hispanic, non-Hispanic). Participants compared to community/multi-purpose centers (19.0%) and res-
also self-reported their living situation (i.e., lived alone, lived with idential facilities (15.7%). Little variation was observed based on
others). the average median household income of workshops by delivery
site types (i.e., range from $48,970 to $51,910). On average, work-
Analyses shops at faith-based organizations (65.8%) and community/multi-
All statistical analyses were performed using SPSS (version 22). Of purpose facilities (65.8% non-Hispanic White) were delivered in
the first 100,000 participants reached in this initiative, cases were more racially/ethnically diverse areas compared to workshops
immediately omitted for those who attended workshops hosted offered at healthcare organizations (73.0% non-Hispanic White).
at delivery sites other than the five most prevalent sites noted
above (n = 13,784). The following delivery site types were omitted Participant Characteristics by Delivery Site Type
from analyses: educational institutions (n = 2,264, 2.3%), county Personal characteristics of study participants are also presented
health departments (n = 1,274, 1.3%), workplaces (n = 541, 0.5%), in Table 1. Overall, the average age of participants was 68.6 years
and tribal organizations (n = 189; 0.2%). Further, delivery sites (±13.6). The majority of participants was female (79.4%), non-
categorized as “other” (n = 9,516, 9.5%) were omitted because of Hispanic (90.2%), non-Hispanic White (67.1%), and lived with
the potential difficulty to interpret findings associated with this others (96.3%). Approximately 78% of participants resided in
delivery site type. Of the remaining 86,216 cases, those with miss- metro areas. On average, participants resided in ZIP Codes where
ing data for age (n = 9,502), sex (n = 6,487), race (n = 14,278), eth- the median household income was $50,600 (±$13,170) and in
nicity (n = 18,154), living situation (n = 60), residential rurality areas where 14.1% (±5.7%) of the population was age 65 years
(n = 10,195), and delivery site rurality (n = 48) were subsequently and older. On average, CDSMP participants resided in ZIP Codes
omitted. Some participants had more than one of these exclu- comprised of 70.5% (±25.7%) non-Hispanic White residents.
sionary characteristics, thus the usable final sample was 58,526 When comparing participant characteristics by delivery site
middle-aged and older adults who attended CDSMP workshops type, residential facilities (73.4 years ± 12.8) and senior centers
at senior centers or AAA, residential facilities, healthcare orga- or AAA (71.0 years ± 11.79) recruited the oldest participants, on
nizations, community or multi-purpose centers, and faith-based average. The largest proportion of male participants was reached
organizations. in healthcare organizations (25.6%), whereas the smallest pro-
Frequencies were calculated for all major study variables, which portion was reached in residential facilities (17.1%). The great-
were examined in relationship to the program delivery site type. est ethnic participant diversity was observed among workshops
Differences for categorical variables were assessed using Pear- delivered at healthcare organizations (13.3%), community/multi-
son’s chi-squared tests. One-way analyses of variance (f statistics) purpose facilities (10.3%), and faith-based organizations (10.1%).
were used to identify mean differences for continuous variables. Relative to all other delivery site types, a substantially larger
Multinomial logistic regression was used to examine personal proportion of African American participants attended work-
characteristics and participants’ neighborhood-level characteris- shops at faith-based organizations (31.6%). Substantially, larger
tics associated with the type of delivery site they attended (i.e., proportions of Asian or Pacific Islander participants attended

Frontiers in Public Health | www.frontiersin.org 103 April 2015 | Volume 3 | Article 77

Smith et al. Diverse delivery infrastructure

TABLE 1 | Sample characteristics by delivery site type.

Total Senior Residential Healthcare Community or Faith-based χ 2 or P

(n = 58,526) center/AAA facility organization multi-purpose organization f
(n = 21,339) (n = 12,600) (n = 11,577) facility (n = 5,942)
(n = 7,068)

DELIVERY SITE CHARACTERISTICS

Metro (delivery site) 79.0% 76.9% 84.3% 76.4% 81.0% 78.4% 331.39 <0.001
Non-metro (delivery site) 21.0% 23.1% 15.7% 23.6% 19.0% 21.6%
Median household income 50.40 (±13.07) 48.97 (±13.26) 51.91 (±12.69) 50.82 (±12.51) 51.07 (±13.17) 50.75 (±13.64) 115.39 <0.001
for ZIP code (delivery site)
Percent of ZIP code 13.93 (±5.59) 13.97 (±5.39) 14.20 (±5.59) 13.51 (±4.80) 13.96 (±6.33) 13.97 (±6.63) 24.43 <0.001
population aged 65+
(delivery site)
Percent of ZIP code 68.95 (±25.35) 69.64 (±25.06) 67.33 (±26.67) 72.96 (±22.24) 65.83 (±26.53) 65.84 (±26.56) 139.84 <0.001
population non-Hispanic
White (delivery site)
PARTICIPANT CHARACTERISTICS
Age 68.58 (±13.56) 70.97 (±11.79) 73.44 (±12.81) 61.77 (±14.40) 66.03 (±13.76) 66.03 (±13.56) 1545.28 <0.001
Male 20.6% 19.1% 17.1% 25.6% 21.9% 21.6% 311.83 <0.001
Female 79.4% 80.9% 82.9% 74.4% 78.1% 78.4%
Non-Hispanic 90.2% 90.8% 92.7% 86.7% 89.7% 89.9% 256.58 <0.001
Hispanic 9.8% 9.2% 7.3% 13.3% 10.3% 10.1%
Non-Hispanic White 67.1% 68.4% 68.0% 74.1% 61.6% 54.1% 2029.69 <0.001
African American 21.5% 22.3% 23.2% 13.6% 20.9% 31.6%
Asian/Pacific islander 4.2% 3.0% 3.0% 2.5% 9.9% 7.6%
American Indian/Alaska 1.2% 1.1% 0.9% 2.0% 1.0% 0.9%
native
Other/multiple Races 5.9% 5.2% 5.1% 7.9% 6.6% 5.8%
Number of self-reported 2.13 (±0.98) 2.63 (±1.62) 2.77 (±1.69) 2.64 (±1.64) 2.46 (±1.62) 2.29 (±1.59) 101.77 <0.001
chronic conditions
Lives with others 96.3% 96.7% 96.5% 95.8% 95.3% 96.5% 38.24 <0.001
Lives alone 3.7% 3.3% 3.5% 4.2% 4.7% 3.5%
Metro (participant 77.9% 74.9% 83.9% 74.8% 81.1% 78.3% 478.07 <0.001
residence)
Non-metro (participant 22.1% 25.1% 16.1% 25.2% 18.9% 21.7%
residence)
Median household income 50.60 (±13.17) 49.11 (±13.30) 52.01 (±12.68) 51.14 (±12.78) 51.51 (±13.50) 50.80 (±13.57) 118.32 <0.001
for ZIP code (participant
residence)
Percent of ZIP code 14.13 (±5.72) 14.26 (±5.54) 14.14 (±5.70) 13.93 (±5.01) 14.10 (±6.14) 14.05 (±7.06) 6.81 <0.001
population aged 65+
(participant residence)
Percent of ZIP code 70.45 (±25.66) 71.12 (±25.29) 67.57 (±26.73) 74.88 (±23.26) 68.68 (±26.43) 67.57 (±26.86) 158.09 <0.001
population non-Hispanic
White (participant
residence)

Pearson’s chi-squared tests (χ 2 ) were used to identify significant distribution differences across delivery site types.
One-way analyses of variance (f statistics) were used to identify mean differences for continuous variables across delivery site types.

workshops at community/multi-purpose facilities (9.9%) and The average area-level (ZIP Code-level) median household
faith-based organizations (7.6%). The largest proportions of par- income varied by delivery site type. Participants who attended
ticipants living alone attended workshops at community/multi- workshops at senior centers or AAA ($49,110 ± $13,170) resided
purpose facilities (4.7%) and healthcare organizations (4.2%). in the least affluent areas, whereas those who attended work-
The largest proportions of participants residing in non-metro shops at residential facilities resided in the most affluent
areas were reached in healthcare organizations (25.2%), senior areas ($52,010 ± $12,680). The average area-level (ZIP Code-
centers/AAA (25.1%), and faith-based organizations (21.7%). level) race/ethnicity composition also varied. Participants who

Frontiers in Public Health | www.frontiersin.org 104 April 2015 | Volume 3 | Article 77

Smith et al. Diverse delivery infrastructure

attended workshops at faith-based organizations (67.6% non-

Upper

1.21

1.02

1.00

0.99
–
Hispanic White residents ± 26.9%) and residential facilities

95% CI
(67.6% non-Hispanic white residents ±26.7%) resided in the most

Faith-based organization
racially/ethnically diverse areas.

Lower

1.04

1.01

1.00

0.99
–
Delivery Site Type Profiles by Neighborhood
Characteristics

<0.001

<0.001

<0.001
0.004
P

–
Because senior center or AAA delivery site types were used as the
referent group for regression analyses, the descriptive profile for
this delivery site type by neighborhood characteristics is provided

1.000
1.121

1.014

1.000

0.993
OR
here (see Table 1). Approximately 77% of participants attended
workshops delivered in metro areas. On average, participants
attended workshops delivered in ZIP Codes where the median

Upper

1.02

1.02

1.00

1.00
Community or multi-purpose facility
household income was $48,970 (±$13,260) and in areas where

–
95% CI
14.0% (±5.4%) of the population was aged 65 years and older. On
average, participants attended workshops delivered in ZIP Codes

Lower
comprised of 69.6% (±25.1%) non-Hispanic White residents.

0.87

1.01

1.00

0.99
–
Utilizing multinomial logistic regression, profiles for resi-
dential facilities, healthcare organizations, community or multi-

<0.001

<0.001

<0.001
0.113
purpose centers, and faith-based organizations based on site

–
neighborhood characteristics are described below. The senior
center or AAA delivery site types were used as the referent group
(see Table 2).

1.000
0.941

1.016

1.000

0.993
OR
Residential Facilities

Upper
Compared to workshops delivered at senior centers or AAA, par-

1.23

0.97

1.00

1.01
–
ticipants were less likely to attend workshops delivered at residen-
95% CI

Odds ratios (OR) indicate the odds of a characteristic being associated with a delivery site type, relative to the referent group.
Healthcare organization

tial facilities in rural areas (OR = 0.718, P < 0.001). Participants Lower
who attended workshops at residential facilities did so in areas
1.08

0.96

1.00

1.01
–
that were more affluent (OR = 1.023, P < 0.001) and had smaller
proportions of the population who were non-Hispanic White
(OR = 0.996, P < 0.001).
<0.001

<0.001

<0.001

<0.001
TABLE 2 | Delivery site neighborhood characteristics associated with delivery site type.

Healthcare Organizations
Compared to workshops delivered at senior centers or AAA,
1.152
1.000
0.969

1.000

1.008
OR

participants were more likely to attend workshops delivered at

healthcare organizations in rural areas (OR = 1.152, P < 0.001).
Participants who attended workshops at healthcare organizations
Upper

0.77

1.03

1.00

1.00

did so in areas that were less affluent (OR = 0.969, P < 0.001) and
–
95% CI

had larger proportions of the population who were non-Hispanic

White (OR = 1.008, P < 0.001).
Residential facility

Lower

0.67

1.02

1.00

1.00
–

Community/Multi-Purpose Centers
Compared to workshops delivered at senior centers or AAA,
<0.001

<0.001

<0.001

<0.001

participants who attended workshops at community or multi-

P

purpose centers did so in areas that were more affluent

(OR = 1.016, P < 0.001) and had smaller proportions of the pop-
0.718
1.000
1.023

1.000

0.996

ulation who were non-Hispanic White (OR = 0.993, P < 0.001).

OR

Referent group: senior center/AAA.

Faith-Based Organizations
Median household income
for ZIP code (delivery site)

Compared to workshops delivered at senior centers or AAA,

population non-Hispanic
Non-metro (delivery site)

Significance: P < 0.001.

population aged 65+

participants were more likely to attend workshops delivered at

Percent of ZIP code

Percent of ZIP code

Metro (delivery site)

White (delivery site)

healthcare organizations in rural areas (OR = 1.121, P < 0.001).

Participants who attended workshops at faith-based organizations
(delivery site)

did so in areas that were more affluent (OR = 1.014, P < 0.001)
and had smaller proportions of the population who were non-
Hispanic White (OR = 0.993, P < 0.001).

Frontiers in Public Health | www.frontiersin.org 105 April 2015 | Volume 3 | Article 77

Smith et al. Diverse delivery infrastructure

Delivery Site Profiles by Personal Characteristics attended workshops delivered at community or multi-purpose
and Residential Characteristics centers were more likely to live alone (OR = 1.575, P < 0.001) and
Because senior center or AAA delivery site types were used as the less likely to reside in rural areas (OR = 0.787, P < 0.001). These
referent group for regression analyses, the descriptive profile for participants also resided in areas that had larger proportions of the
this site type by participant characteristics and their residential population who were age 65 and older (OR = 1.018, P < 0.001).
characteristics is provided here (see Table 1). The average age of
participants was 71.0 years (±11.79). The majority of participants Faith-Based Organizations
was female (80.9%), non-Hispanic (90.8%), non-Hispanic White Compared to workshops attended at senior centers or AAA,
(68.4%), and lived with others (96.7%). Approximately 75% of par- participants who attended workshops delivered at faith-based
ticipants resided in metro areas. On average, participants resided organizations were less likely to be older (OR = 0.971, P < 0.001).
in ZIP Codes where the median household income was $49,110 These individuals were more likely to be Hispanic (OR = 1.233,
(±$13,300) and in areas where 14.3% (±5.5%) of the population P < 0.001), African American (OR = 1.951, P < 0.001), or Asian
was age 65 years and older. On average, CDSMP participants or Pacific Islander (OR = 3.044, P < 0.001). Participants who
resided in ZIP Codes comprised of 71.1% (±25.3%) non-Hispanic attended workshops delivered at faith-based organizations resided
White residents. in areas that had larger proportions of the population who were
Utilizing multinomial logistic regression, profiles for resi- age 65 and older (OR = 1.015, P < 0.001).
dential facilities, healthcare organizations, community or multi-
purpose centers, and faith-based organizations based on site Discussion
neighborhood characteristics are described below. The senior Findings from this replication study support that CDSME pro-
center or AAA delivery site types were used as the referent group grams are capable of attracting and serving a large and diverse
(see Table 3). group of participants using coordinated delivery infrastruc-
ture through the aging services network (1). In particular, the
Residential Facilities evidence-based programs delivered in the nationwide delivery of
Compared to workshops attended at senior centers or AAA, par- CDSME programs as part of the American Recovery and Rein-
ticipants who attended workshops delivered at residential facilities vestment Act of 2009 initiative (12, 14, 15) reached many at-risk
were more likely to be older (OR = 1.017, P < 0.001) and female middle-aged and older adults in geographic regions of limited
(OR = 1.126, P < 0.001). These individuals were less likely to be affluence and those with larger minority populations. Results
Hispanic (OR = 0.771, P < 0.001) and less likely to be African indicate that certain delivery site types are more likely to serve
American (OR = 0.816, P < 0.001), or Asian or Pacific Islander geographic areas and participants with different characteristics,
(OR = 0.714, P < 0.001). Participants who attended workshops which highlights the importance of maintaining a diverse and
delivered at residential facilities were less likely to reside in rural dispersed collection of delivery sites in a given area/community to
areas (OR = 0.691, P < 0.001). These participants also resided in facilitate participants’ access to programs (16, 17). Other analyses
areas that had larger proportions of the population who were age of evidence-based programs for older adults reveal the mismatch
65 and older (OR = 1.009, P < 0.001) and smaller proportions between population needs and program availability (18). Thus,
of the population who were non-Hispanic white (OR = 0.993, continued efforts are needed to recruit new community partners
P < 0.001). to establish and grow the existing infrastructure while simulta-
neously nurturing and supporting the existing infrastructure to
Healthcare Organizations ensure a sustained community presence (19).
Compared to workshops attended at senior centers or AAA, par- This study is important in that it captures the continued growth
ticipants who attended workshops delivered at healthcare orga- and dispersion of CDSME programs from 2010 to 2012, the third
nizations were less likely to be older (OR = 0.951, P < 0.001) wave of evidence-based health promotion/disease prevention pro-
and female (OR = 0.734, P < 0.001). These individuals were more graming supported by the Administration for Community Living
likely to be American Indian or Alaska Native (OR = 1.423, (14, 15). The success of this intervention to reach over 100,000
P < 0.001), yet less likely to be African American (OR = 0.603, participants in such a short time period is largely attributed to the
P < 0.001) or Asian or Pacific Islander (OR = 0.675, P < 0.001). previous success of the ACL-supported evidence-based program
Participants who attended workshops delivered at healthcare initiatives, which builds upon the infrastructure that was estab-
organizations were more likely to live alone (OR = 1.270, lished from 2006 to 2009 (14). Continued monitoring of the reach
P < 0.001). These participants also resided in areas that had larger of CDSME programs enables the visualization of the evolution
proportions of the population who were non-Hispanic White of these programs as they are delivered throughout the United
(OR = 1.004, P < 0.001). States. From 2006 to 2009, approximately 29,000 participants were
reached across 27 states (17), while over 100,000 participants were
Community/Multi-Purpose Centers reached across 45 states and two territories from 2010 to 2012.
Compared to workshops attended at senior centers or AAA, The leading five delivery site types remained consistent across
participants who attended workshops delivered at community or these study periods, and senior centers and AAA consistently
multi-purpose centers were less likely to be older (OR = 0.970, served the largest proportion of participants. However, there were
P < 0.001). These individuals were more likely to be Asian some noteworthy changes in the areas served and the partici-
or Pacific Islander (OR = 3.040, P < 0.001). Participants who pants reached from 2006–2009 to 2010–2012. For example, in

Frontiers in Public Health | www.frontiersin.org 106 April 2015 | Volume 3 | Article 77

Frontiers in Public Health | www.frontiersin.org

Smith et al.
TABLE 3 | Personal characteristics associated with delivery site type.

Residential facility Healthcare organization Community or multi-purpose facility Faith-based organization

95% CI 95% CI 95% CI 95% CI

OR P Lower Upper OR P Lower Upper OR P Lower Upper OR P Lower Upper

Age 1.017 <0.001 1.02 1.02 0.951 <0.001 0.95 0.95 0.970 <0.001 0.97 0.97 0.971 <0.001 0.97 0.97
Female 1.126 <0.001 1.06 1.19 0.734 <0.001 0.69 0.78 0.921 0.016 0.86 0.99 0.896 0.003 0.83 0.96
Male 1.000 – – – 1.000 – – – 1.000 – – – 1.000 – – –
Hispanic 0.771 <0.001 0.70 0.85 1.157 0.001 1.06 1.26 1.006 0.915 0.91 1.12 1.233 <0.001 1.10 1.38
Non-Hispanic 1.000 – – – 1.000 – – – 1.000 – – – 1.000 – – –
Other/multiple races 1.046 0.436 0.93 1.17 0.981 0.726 0.88 1.09 1.119 0.084 0.99 1.27 1.118 0.125 0.97 1.29
American Indian/Alaska 0.854 0.184 0.68 1.08 1.423 <0.001 1.17 1.73 0.926 0.577 0.71 1.21 0.968 0.832 0.71 1.31
native
Asian/Pacific islander 0.714 <0.001 0.62 0.82 0.675 <0.001 0.58 0.78 3.040 <0.001 2.70 3.43 3.044 <0.001 2.66 3.48
African American 0.816 <0.001 0.76 0.88 0.603 <0.001 0.56 0.65 0.913 0.039 0.84 1.00 1.951 <0.001 1.79 2.13
Non-Hispanic White 1.000 – – – 1.000 – – – 1.000 – – – 1.000 – – –
107

Lives alone 1.109 0.099 0.98 1.25 1.270 <0.001 1.12 1.44 1.575 <0.001 1.38 1.80 1.115 0.181 0.95 1.31
Lives with others 1.000 – – – 1.000 – – – 1.000 – – – 1.000 – – –
Non-metro (participant 0.691 <0.001 0.65 0.74 1.047 0.154 0.98 1.11 0.787 <0.001 0.73 0.85 0.988 0.765 0.91 1.07
residence)
Metro (participant 1.000 – – – 1.000 – – – 1.000 – – – 1.000 – – –
residence)
Median household income 1.000 <0.001 1.00 1.00 1.000 <0.001 1.00 1.00 1.000 <0.001 1.00 1.00 1.000 <0.001 1.00 1.00
for ZIP code (participant
residence)
Percent of ZIP code 1.009 <0.001 1.01 1.01 0.999 0.629 0.99 1.00 1.018 <0.001 1.01 1.02 1.015 <0.001 1.01 1.02
population aged 65+
(participant residence)
Percent of ZIP code 0.993 <0.001 0.99 0.99 1.004 <0.001 1.00 1.01 0.998 0.002 1.00 1.00 1.002 0.002 1.00 1.00
population non-Hispanic
April 2015 | Volume 3 | Article 77

White (participant

Diverse delivery infrastructure

residence)

Referent group: senior center/AAA.

Significance: P < 0.001.
Odds ratios (OR) indicate the odds of a characteristic being associated with a delivery site type, relative to the referent group.
Smith et al. Diverse delivery infrastructure

the 2010–2012 initiative, residential facilities emerged as delivery study provides insight about the reach and adoption elements
site types more likely to enroll females and participants resid- of the RE-AIM Framework, additional investigations and data
ing in more affluent areas and areas with higher proportions of collection efforts are needed to understand the influence of this
people aged 65 years and older compared to senior centers or model on implementation, effectiveness, and maintenance in large
AAA. Healthcare organizations emerged as delivery sites more translational evidence-based program dissemination efforts.
likely to serve participants who reside alone and non-metro areas. It should be noted that this study was not an exact replication of
Faith-based organizations emerged as delivery sites more likely the earlier study (data from the years 2006 to 2009). The primary
to serve African Americans and Asian/Pacific Islanders. This reason for differences was that the variables collected from 2006 to
reveals a greater diversification of delivery sites, and resonates 2009 differed slightly from those collected from 2010 to 2012. For
with recommendations for capacity building and sustainability for example, data related to participants’ education were not collected
institutionalizing programmatic activities (20). in the 2010–2012 initiative. Therefore, this variable could not be
In terms of translating the success of these evidence-based pro- added to the analytic model. Among studies with older adults,
grams across multiple settings, the Consolidated Framework For education has been used as a proxy for social status because of
Implementation Research (CFIR) has been highlighted for poten- issues related to self-reported household income (either based on
tial use (20). This framework pulls from, among other things, missing data or that many older adults no longer work). Omit-
the idea that multiple theories can be combined to form a more ting education from the analyses reduced our ability to examine
comprehensive understanding of organizational characteristics participant-level social status data; however, neighborhood-level
associated with successful implementation of interventions (20). data were utilized. Another example was that the categories of
Future studies are encouraged to use this framework to identify race/ethnicity differed between the studies. The data collected
organizational features associated with successful adaption and in the 2010–2012 initiative asked participants to report ethnicity
implementation of community-based programs by different deliv- separately and included more race categories (consistent with
ery site types and programs. This would include the need to collect those in the U.S. Census) relative to the collapsed race/ethnicity
more comprehensive information about the delivery sites (e.g., item collected in the 2006–2009 initiative. While this change
culture, implementation climate). This information could also facilitated more nuanced analyses in the current study, it made
be used to develop targeted recommendations for organizations direct race/ethnicity comparisons between studies more difficult.
delivering these and other evidence-based programs (20). A last example of replication differences was that the 2006–2009
This study reinforces the value of using the RE-AIM Frame- initiative did not include participants’ self-reported chronic con-
work when planning, implementing, and evaluating grand-scale ditions. As such, the current study also omitted chronic conditions
translational initiatives to roll-out/disseminate evidence-based from the study analyses. However, because of the importance of
programs for older adults (21). More specifically, it supports participants’ chronic conditions for a disease self-management
the strong interdependence between program adoption and par- intervention, sensitivity analyses were performed that included
ticipant reach. However, this study did not examine the other self-reported chronic conditions in the participant-level multino-
important elements included in the RE-AIM Framework (e.g., mial regression model (tables not shown). On average, partici-
implementation, evaluation, and maintenance), all of which are pants reported 2.60 (±1.64) chronic condition diagnoses, with
of equal importance for the success of grand-scale program dis- 48.1% self-reporting three or more chronic conditions. All signif-
semination. icant relationships remained significant in these sensitivity anal-
There were limitations associated with this study. First, only yses. Relative to participants who attended workshops in senior
limited data were collected from participants and delivery sites, centers or AAA, those who attended workshops in residential
which limited our ability to fully assess the characteristics of par- facilities and healthcare organizations reported significantly more
ticipants and sites that participated in this initiative. For example, chronic conditions; whereas those who attended workshops in
outcomes were not collected for this grand-scale dissemination, community or multi-purpose centers or faith-based organizations
thus it is unknown whether certain delivery site types evoked reported significantly fewer chronic conditions.
better health benefits than others and/or for which participants
those benefits occurred. Second, there was considerable missing Conclusion
data for participant characteristics. This data collection issue was
also observed in the initial study (1), and is likely attributed to self- National replication studies are valuable for revealing the evolu-
report data collection occurring on-site and during workshop time tion of the infrastructure supporting evidence-based programs
at various locations across the country. Despite a coordinated data for older adults. Expanding upon current studies demonstrating
collection and reporting structure for this initiative, additional the potential of CDSMP to meet the Triple Aims of health care
efforts may be needed to increase data fidelity as well as reduce reform (7), this replication’s findings suggest fertile areas for
data collection burdens on program deliverers. Third, while there future study understanding about how delivery system character-
were many statistically significant relationships observed in this istics are related to programmatic processes and outcomes. Addi-
study, such significant relationships may be an artifact of the tional research is needed to identify the most effective strategies
large sample size and less about true differences across delivery for increasing organization-based recruitment efforts including
site types. However, in an effort to be conservative, it should be both personal incentives and policies providing sustained support
noted that only relationships meeting the P < 0.001 criteria were for CDSMPs for the increasingly diverse population of older
deemed statistically significant for this study. Fourth, while this Americans.

Frontiers in Public Health | www.frontiersin.org 108 April 2015 | Volume 3 | Article 77

Smith et al. Diverse delivery infrastructure

Acknowledgments served as the Technical Assistance Resource Center for this

initiative and collected de-identified data on program partici-
The American Recovery and Reinvestment Act of 2009 (i.e., pation. Because this is a replication study, we also recognize
Recovery Act) Communities Putting Prevention to Work: Chronic the original support from the Administration on Aging with
Disease Self-Management Program initiative, led by the U.S. assistance from the National Council on Aging for the eval-
Administration on Aging in collaboration with the Centers for uation of the EBDDP initiative under cooperative agreement
Disease Control and Prevention and the Centers for Medicare number 90OP0001/01. The findings and conclusions in this arti-
and Medicaid Services, allotted $32.5 million to support the cle are those of the author(s) and do not necessarily repre-
translation of the Stanford program in 45 States, Puerto Rico, sent the official position of AoA, ACL, NCOA, or any other
and the District of Columbia. The National Council on Aging agency.

References 15. Boutaugh ML, Lawrence LJ. Fostering healthy aging through evidence-based
prevention programs: perspectives from the administration for community
1. Smith ML, Belza B, Altpeter M, Ahn S, Dickerson JB, Ory MG. Disseminating living/administration on aging. Front Public Health (2015) 2:236. doi:10.3389/
an evidence-based disease self-management program for older Americans: fpubh.2014.00236
implications for diversifying participant reach through delivery site adoption. 16. Towne SD Jr, Smith ML, Ahn S, Ory MG. The reach of chronic disease self-
In: Maddock J, editor. Public Health: Social and Behavioral Health. Rijeka, management education programs to rural populations. Fron Public Health Educ
Croatia: InTech (2012) 385–404. Promot (0000) 2:172. doi:10.3389/fpubh.2014.00172
2. Smith ML, Bazzarre TL, Frisco J, Jackman BA, Cox NJ, Ory MG. Transfor- 17. Towne SD Jr., Smith ML, Ahn S, Altpeter M, Belza B, Kulinski KP, et al. National
mation of a rural community for active living. Fam Community Health (2011) dissemination of multiple evidence-based disease prevention programs: reach
34(2):163–72. doi:10.1097/FCH.0b013e31820e0d2e to vulnerable older adults. Front Public Health (2015) 2:156. doi:10.3389/fpubh.
3. Glasgow RE, McKay HG, Piette JD, Reynolds KD. The RE-AIM frame- 2014.00156
work for evaluating interventions: what can it tell us about approaches to 18. Towne SD, Smith ML, Yoshikawa A, Ory MG. Geospatial distribution of fall-
chronic illness management? Patient Educ Couns (2001) 44:119–27. doi:10. related hospitalization incidence in Texas. J Safety Res (2015) 53:11–16. doi:10.
1016/S0738-3991(00)00186-5 1016/j.jsr.2015.01.002
4. Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of health 19. Altpeter M, Schneider EC, Whitelaw N. Examining strategies to build and
promotion interventions: the RE-AIM framework. Am J Public Health (1999) sustain healthy aging programming collaboratives. Health Educ Behav (2014)
89:1322–7. doi:10.2105/AJPH.89.9.1322 41(1 Suppl):27S–33S. doi:10.1177/1090198114537065
5. Lorig K, Holman H, Sobel D, Laurent D. Living a Healthy Life with Chronic Con- 20. Damschroder LJ, Aron DC, Keith RE, Kirsh SR, Alexander JA, Lowery JC.
ditions: Self Management of Heart Disease, Arthritis, Diabetes, Asthma, Bron- Fostering implementation of health services research findings into practice: a
chitis, Emphysema and others. 3rd ed. Boulder, CO: Bull Publishing Company consolidated framework for advancing implementation science. Implement Sci
(2006). (2009) 4(1):50. doi:10.1186/1748-5908-4-50
6. Ory MG, Ahn S, Jiang L, Lorig K, Ritter P, Laurent DD, et al. National study of 21. Ory MG, Altpeter M, Belza B, Helduser J, Zhang C, Smith ML. Perceived
chronic disease self-management six-month outcome findings. J Aging Health utility of the RE-AIM framework for health promotion/disease prevention
(2013) 25(7):1258–74. doi:10.1177/0898264313502531 initiatives for older adults: a case study from the U.S. evidence-based disease
7. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, Whitelaw N, et al. Successes of prevention initiative. Front Public Health (2014) 2:143. doi:10.3389/fpubh.2014.
a national study of the chronic disease self-management program: meeting the 00143
triple aim of health care reform. Med Care (2013) 51(11):992–8. doi:10.1097/
MLR.0b013e3182a95dd1 Conflict of Interest Statement: The authors declare that the research was
8. Ahn S, Basu R, Smith ML, Jiang L, Lorig K, Whitelaw N, et al. The impact conducted in the absence of any commercial or financial relationships that could
of chronic disease self-management programs: healthcare savings through a be construed as a potential conflict of interest.
community-based intervention. BMC Public Health (2013) 13(1):1141. doi:10.
1186/1471-2458-13-1141 This paper is included in the Research Topic, “Evidence-Based Programming for Older
9. Bandura A. Social cognitive theory of self-regulation. Organ Behav Hum Decis Adults.” This Research Topic received partial funding from multiple government and
Process (1991) 50(2):248–87. doi:10.1016/0749-5978(91)90022-L private organizations/agencies; however, the views, findings, and conclusions in these
10. Ory MG, Smith ML, Patton K, Lorig K, Zenker W, Whitelaw N. Self- articles are those of the authors and do not necessarily represent the official position of
management at the tipping point: reaching 100,000 americans with evidence- these organizations/agencies. All papers published in the Research Topic received peer
based programs. J Am Geriatr Soc (2013) 61(5):821–3. doi:10.1111/jgs.12239 review from members of the Frontiers in Public Health (Public Health Education and
11. U.S. Department of Health and Human Services Administration on Promotion section) panel of Review Editors. Because this Research Topic represents
Aging. ARRA – Communities Putting Prevention to Work: Chronic Disease work closely associated with a nationwide evidence-based movement in the US,
Self-Management Program. (2012). Available at: www.cfda.gov/index?s= many of the authors and/or Review Editors may have worked together previously in
program&mode=form&tab=step1&id=5469a61f2c5f25cf3984fc3b94051b5f some fashion. Review Editors were purposively selected based on their expertise with
12. Kulinski KP, Boutaugh M, Smith ML, Ory MG, Lorig K. Setting the stage: mea- evaluation and/or evidence-based programming for older adults. Review Editors were
sure selection, coordination, and data collection for a national self-management independent of named authors on any given article published in this volume.
initiative. Front Public Health (2015) 2:206. doi:10.3389/fpubh.2014.00206
13. Area Health Resources Files (ARF). US Department of Health and Human Copyright © 2015 Smith, Ory, Ahn, Belza, Mingo, Towne and Altpeter. This is an
Services, Health Resources and Services Administration, Bureau of Health open-access article distributed under the terms of the Creative Commons Attribution
Professions, Rockville, MD. (2013). License (CC BY). The use, distribution or reproduction in other forums is permitted,
14. Boutaugh ML, Jenkins SM, Kulinski KP, Lorig KL, Ory MG, Smith ML. Closing provided the original author(s) or licensor are credited and that the original publica-
the disparity: the work of the administration on aging. Generations (2015) tion in this journal is cited, in accordance with accepted academic practice. No use,
38(4):107–18. distribution or reproduction is permitted which does not comply with these terms.

Frontiers in Public Health | www.frontiersin.org 109 April 2015 | Volume 3 | Article 77

 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00172

The reach of chronic-disease self-management education

programs to rural populations
Samuel D. Towne Jr.1 *, Matthew Lee Smith 2 , SangNam Ahn 1,3 and Marcia G. Ory 1
1
Department of Health Promotion and Community Health Sciences, Texas A&M Health Science Center, School of Public Health, College Station, TX, USA
2
Department of Health Promotion and Behavior, College of Public Health, University of Georgia, Athens, GA, USA
3
Division of Health Systems Management and Policy, School of Public Health, The University of Memphis, Memphis, TN, USA

Edited by: This study assessed the sociodemographic characteristics of rural residents who partici-
Renae L. Smith-Ray, University of
pated in chronic-disease self-management education (CDSME) program workshops and
Illinois at Chicago, USA
the extent to which CDSME programs were utilized by those with limited access to health
Reviewed by:
Jim P. Stimpson, University of care services. We analyzed data from the first 100,000 adults who attended CDSME
Nebraska Medical Center, USA program workshops during a national dissemination spanning 45 states, the District of
Pankaja Desai, University of Illinois at Columbia, and Puerto Rico. Approximately 24% of participants lived in rural areas. Overall,
Chicago, USA
42% of all participants were minorities; urban areas reached more minority participants
Jo Ann Shoup, Kaiser Permanente
Colorado, USA (48%) than rural areas (25%). The average age of participants was high in rural (age,
Renae L. Smith-Ray, University of µ = 66.1) and urban (age, µ = 67.3) areas. In addition, the average number of chronic con-
Illinois at Chicago, USA ditions was higher (p < 0.01) in rural (µ = 2.6 conditions) versus urban (µ = 2.4 conditions)
*Correspondence: areas. Successful completion of CDSME programs (i.e., attending four or more of the
Samuel D. Towne Jr., Department of
six workshop sessions) was higher (p < 0.01) in rural versus urban areas (78% versus
Health Promotion and Community
Health Sciences, Texas A&M Health 77%). Factors associated with higher likelihood of successful completion of CDSME pro-
Science Center, School of Public grams included being Black (OR = 1.25) versus White and living in rural (versus urban) areas
Health, 1266 TAMU, College Station, (OR = 1.09). Factors associated with lower likelihood of successful completion included
TX 77843/1266, USA
e-mail: [email protected]
being male (OR = 0.92) and residing in a primary care Health Professional Shortage Area
or HPSA (versus a non-HPSA) (OR = 0.93). Findings highlight the capability of CDSME pro-
grams to reach rural residents, yet dissemination efforts can be further enhanced to ensure
minorities and individuals in a HPSA utilize this program. Tailored strategies are needed to
increase participant recruitment and retention in rural areas to overcome traditional barriers
to health service access.
Keywords: chronic-disease self-management, evidence-based program, rural, intervention dose, older adults

INTRODUCTION exemplified by areas designed as rural highly overlapping with

While it is known that individuals with chronic diseases are more health professional shortage area (HPSA) and medically under-
likely to utilize health care services (1–3), we are still learning served area designations (11, 12). Using geographic information
about their use of health promotion resources available in com- systems (GIS), researchers have identified geospatial barriers hin-
munity settings. Further, less is known about the unique commu- dering rural area residents, especially minority older adults, from
nity characteristics and infrastructures that influence the delivery accessing resources (e.g., longer distances, lower availability of
and adoption of evidence-based chronic-disease self-management health care providers) (13).
education (CDSME) programs in traditionally underserved areas Prior research has documented the benefits of delivering
and populations. evidence-based programs (EBP) in rural communities [e.g.,
Compared to metropolitan or urban areas, there is limited improving health-related outcomes (14), falls efficacy (15)].
research about aging in rural areas. And, studies about rural pop- However, the extent to which CDSME programs are deliv-
ulations are primarily demographic or epidemiological in focus. ered in rural areas remains unknown. Because of the known
Disproportionately, more older adults live in rural areas (15% in effectiveness of CDSME programs (e.g., improved health out-
rural, 12% in urban) (4), and rural areas have less health care comes, lower hospitalization, better chronic-disease manage-
service availability and fewer health care providers compared to ment) (16–19), it is important to identify whether residents
urban areas (5–7). Relative to those living in urban areas, rural of rural areas have access to these EBP, especially in vulnera-
area residents are disproportionately affected by poor health out- ble rural areas with fewer health-related resources and services.
comes and health care access barriers, which contributes to them Additionally, even when EBP are available in rural areas, it is
having higher disease rates, disability rates, and risk factors for important to assess whether or not participants in these areas
poor health outcomes (8–10). attend enough sessions to receive adequate intervention dose.
Studies have shown that rural areas traditionally encounter This is especially important considering individuals in rural
geographic barriers limiting access to health care resources, as areas may have greater distances to resources (e.g., health care

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 172 | 110
Towne et al. Rural CDSME programs

resources), which may act as a barrier to program participa- City/Town, Small Rural Town, and Isolated Small Rural Town. These
tion (8, 13). multiple rurality measurements allowed us to identify differences
As such, the objectives of this study were to: (1) assess the within rural areas with a greater degree of specificity in selected
extent to which CDSME programs were utilized by rural residents analysis.
and identify characteristics of these rural residents as compared Primary Care HPSA are defined based on geographic area, pop-
to their urban counterparts; (2) investigate the geographic distri- ulation groups, and facilities, with more detailed definitions avail-
bution of CDSME program participation based on the rurality of able from the Health Resources Services Administration (HRSA)
participants’ residence; and (3) examine factors associated with (http://www.hrsa.gov/shortage/) (29). Primary Care HPSAs were
successful workshop completion. defined as either full, partial, or non-HPSA at the county level.
A full-HPSA is defined as an entire county designated as a HPSA
MATERIALS AND METHODS versus partial-HPSA. A non-HPSA is a county not designated as
PROGRAM DESCRIPTION a HPSA.
With the goal of improving self-management skills among adults Areas served by CDSME were defined as unique ZCTA/ZIP
with chronic conditions, CDSME programs have been widely codes where at least one participant was located. These were spread
delivered across the US (20). The CDSME program suite of nationwide throughout 9,599 unique ZCTA/ZIP Codes.
evidence-based self-management programs, developed at Stan-
ford University Patient Education Research Center, uses the Social Dependent variable
Learning Theory (21) to deliver these peer-led interventions (i.e., Our primary dependent variable was successful workshop com-
six sessions, once a week at 2.5 h each for six consecutive weeks) pletion. Participant’s attendance was recorded to determine if
(20). The results of participation in this program include improved adequate intervention dose was received. As defined by the pro-
health, health care utilization (e.g., lower rate of hospitalizations) gram developers, a participant has “successfully” completed the
(19, 22), and health care cost savings (23). program if they attended four or more of the six offered workshop
sessions (19, 22, 25, 30).
DATA SOURCE AND STUDY POPULATION
We conducted a cross-sectional analysis using data collected via Sociodemographics
the national delivery (45 states, Puerto Rico, and the District of Personal characteristics of the participants included age, sex, and
Columbia) (24) of the CDSME programs. As part of the Amer- race/ethnicity. We used one variable for race and ethnicity with
ican Recovery and Reinvestment Act of 2009, CDSME programs categories of non-Hispanic White, non-Hispanic Black or African
were delivered via the Communities Putting Prevention to Work: American, non-Hispanic Native American or Alaskan Native,
Chronic-Disease Self-Management Program initiative led by the US non-Hispanic Asian American, and “other” race/ethnicity cate-
Administration on Aging in partnership with the Centers for Dis- gory (including non-Hispanic Native Hawaiian or other Pacific
ease Control and Prevention (CDC) and the Centers for Medicare Islander, those identified as “other,” and those identified as belong-
and Medicaid Services (CMS) (25). Analyses were conducted using ing to multiple race/ethnic groups), and Hispanic. We also
data on the first 100,000 participants targeted in this initiative (25). included living arrangement to specify whether participants lived
Institutional Review Board approval for this study was given by alone or lived with others.
Texas A&M University. The number of chronic conditions among participants was
identified as having any one or more of the following chronic dis-
MEASURES eases: diabetes, heart disease, hypertension, lung disease, arthritis,
Geospatial variables cancer, or“other”(another chronic disease). We summed the num-
Geospatial analyses were those examining differences across rural- ber of chronic diseases into one variable and grouped it into the
ity. We were interested in characterizing participants and delivery following categories: one condition, two conditions, three condi-
sites by rural and urban categories. To accomplish this, the 2013 tions, four conditions, and five or more conditions (due to small
Area Health Resource File (AHRF) was used to identify geographic sample sizes with six chronic conditions).
characteristics (i.e., rural residency, health professional resources)
(26). We defined rurality based on county and separately ZCTA Statistical analyses
(ZIP Code Tabulation Areas)/ZIP Codes. For counties, urban We conducted analyses on the first 100,000 participants reached
influence codes (UIC) were merged with data from the National in this initiative who had observations with complete data on all
Council on Aging (NCOA) using Federal Information Processing variables of interest. Those with missing data for age (n = 12,447),
Standards (FIPS) Codes. We compare results using both county sex (n = 8,826), race/ethnicity (n = 12,124), living arrangement
and ZIP Code levels of rurality. We used county-level rurality (n = 1,605), number of chronic conditions (n = 1,539), and geo-
in fully adjusted analyses. We dichotomized UIC into Metro- graphic identifiers (n = 12,314) were omitted. Some participants
politan (UIC = 1–2) and Non-Metropolitan (UIC = 3–12) (27). had more than one of these exclusionary characteristics. There-
For ZCTA/ZIP Codes, we merged Rural-Urban Commuting Area fore our final sample size was 82,044. Analyses on observations
Codes (RUCA) into urban and non-urban (large rural cities, small with missing information (e.g., missing rurality) were not con-
rural towns, isolated small rural towns) areas (28). We also coded ducted because our primary goal was to measure outcomes across
rurality into more than a two-way split (i.e., rural and urban). study characteristics (e.g., rurality). We did not attempt to measure
We coded rurality into a 4-way split including Urban, Large Rural program success independent of study characteristics.

www.frontiersin.org April 2015 | Volume 2 | Article 172 | 111

Towne et al. Rural CDSME programs

We used independent sample t -tests and Chi Square for bivari- The bulk of participants were female (approximately 78.0%).
ate comparisons. Logistic regression analyses were used to investi- Approximately 48.9% of participants lived alone. In general, par-
gate factors associated with successful workshop completion. We ticipants had at least two chronic conditions, where the average
used logistic regression to predict the dichotomous outcome of number of chronic conditions was 2.5.
successful completion (versus not attending at least four of the six When compared by the geography of residence, participants
workshop sessions). Fully adjusted analyses (logistic regression) residing in rural areas were younger (p < 0.01) on average com-
includes participant race/ethnicity, rurality (county-level), HPSA pared to those in urban areas (approximately 66.1 years versus
status, participant sex, living arrangement (living alone or not), 67.3 years). The percent of individuals living alone was higher
participant age, and the number of chronic conditions. SAS ver- (p < 0.01) in rural areas (i.e., ranging from 52.1 to 52.6% in rural
sion 9.4 was used for all analysis (31). ArcGIS version 10.2 was areas versus 47.9 to 47.8% in urban areas by ZCTA/ZIP Code and
used for mapping (32). county, respectively). Participants residing in rural areas had more
(p < 0.01) chronic conditions on average compared to those in
RESULTS urban areas (approximately 2.6 conditions versus 2.4 conditions).
Overall, 1,721 counties throughout the US had a CDSME pro- Table 2 presents the successful completion rates by rurality. Suc-
gram available to residents, while 1,421 counties did not offer a cessful completion of the CDSME program was uniformly high
CDSME workshop. There were 922 rural counties and 799 urban at 77.3% overall; however, it was slightly higher in rural areas
counties offering CDSME workshops. Rural counties without a (77.9%) than in urban areas (77.1%). When we specified a 4-level
CDSME workshop totaled 1,130 versus 291 for urban counties. categorization for rurality, we found participants residing in large
Here, 74.3% of areas lacking a CDSME workshop were rural. rural towns (78.4%) and isolated small rural towns (78.3%) had
Some states had more workshop clustering, and others had wider higher successful completion rates than those participants residing
coverage throughout the states (e.g., South Carolina and North in small rural towns (76.6%).
Carolina). The distribution of rural CDSME program participants Table 3 presents the distribution of areas with a CDSME
varied across the US (see Figure 1). Analysis across rurality indi- program presence (i.e., having one or more CDSME workshops
cated that approximately 22.1% (using county-level rurality) to available in the county) by rurality. The majority of areas with
24.4% (using ZCTA/ZIP Code-level rurality) of CDSME program CDSME workshops were urban (70.0%). Approximately 9.3% of
participants resided in rural areas. all CDSMP workshops were located in isolated small rural towns,
Characteristics of participants across rurality are provided in and approximately 8.2% were located in small rural towns. The
Table 1. Age ranged from 18 to over a 100 across all observations. average number of participants in areas with a CDSME workshop

FIGURE 1 | Distribution of the chronic-disease self-management program by ZIP Code/ZCTA and rurality.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 172 | 112
Towne et al. Rural CDSME programs

Table 1 | Distribution of key characteristics across rurality.

Rural Urban Total

ZIP Code/ZCTA County ZIP Code/ZCTA County

Sample size (n = 19,982) 24.38%* (n = 18,111) 22.09%* (n = 61,991) 75.62%* (n = 63,862) 77.91%* (n = 81,973) 100%
Number of chronic conditions 2.59* 2.59* 2.42 2.43 2.46
Sex (% Female) 77.70% 78.00% 78.06% 77.96% 77.97%
Age 66.06* 66.13* 67.32 67.27 67.01
Living alone 52.12%* 52.75%* 47.88% 47.83% 48.92%
Race/ethnicity
White 74.52% 75.48% 52.19% 52.57% 57.63%
Black 13.20% 13.21% 22.33% 22.06% 20.11%
AIAN 2.56% 2.48% 0.92% 0.99% 1.32%
Asian 1.07% 1.17% 3.99% 3.88% 3.28%
Other 6.77% 6.36% 13.11% 13.04% 11.56%
Hispanic 1.87% 1.29% 7.46% 7.46% 6.10%
HPSA
Full HPSA 8.48% 7.59% 34.12% 35.01% 42.60%
Partial HPSA 11.36% 10.26% 33.58% 34.68% 44.94%
Non-HPSA 4.52% 4.23% 7.94% 8.23% 12.46%

*Indicates significantly different (p < 0.01) from urban areas using independent group t-test for continuous variables (number of chronic conditions, percent female,
age, and percent living alone). The overall sample size is different (p < 0.01) by rurality (Chi Square).

Table 2 | Successful completion rates by rurality. Table 3 | Distribution of CDSMP sites (unique ZCTA/ZIP codes with a
participant) by rurality.
Successful Standard Total
completion deviation (n = 82,044) Average Standard Range Total
number of deviation (n = 9,599)
Rurality participants
Urban 77.1% 0.42 62,051
Large rural city/town 78.4% 0.41 10,054 Rurality
Small rural town 76.6% 0.42 5,900 Urban 9.23 13.67 1 208 6,725 (70.01%)
Isolated small rural town 78.3% 0.41 4,039 Large rural 8.46 28.11 1 884 1,192 (12.42%)
city/town
Operational definition of rurality (4-way) includes Urban: RUCA 1.0, 1.1, 2.0, 2.1, Small rural town 7.40 9.37 1 62 791 (8.24%)
4.1, 5.1, 7.1, 8.1, and 10.1; Large Rural City/Town: 3.0, 4.0, 4.2, 5.0, 5.2, 6.0, 6.1, Isolated small 4.54 5.50 1 45 891 (9.28%)
7.2, 8.2, and 10.2; Small Rural Town: 7.0, 7.3, 7.4, 8.0, 8.3, 8.4, 9.0, 9.1, 9.2, and
rural town
10.3; Isolated Small Rural Town: 10.0, 10.4, 10.5, and 10.6.
Operational definition of rurality (4-way) includes Urban: RUCA 1.0, 1.1, 2.0, 2.1,
4.1, 5.1, 7.1, 8.1, and 10.1; large rural city/town: 3.0, 4.0, 4.2, 5.0, 5.2, 6.0, 6.1,
by rurality (calculated at the ZCTA/ZIP Code) was 9.2 participants 7.2, 8.2, and 10.2; Small Rural Town: 7.0, 7.3, 7.4, 8.0, 8.3, 8.4, 9.0, 9.1, 9.2, and
in urban areas, which was almost twice the amount of participants 10.3; isolated small rural town: 10.0, 10.4, 10.5, and 10.6.
in isolated small rural towns. Among areas with a CDSME work-
shop present, the range of the average number of participants in included being male (OR = 0.92) and residing in a full-HPSA
urban areas was much higher than small rural towns or isolated (OR = 0.93) versus a non-HPSA.
small rural towns (1–208 participants versus 1–62 participants
and 1–45 participants, respectively). However, the highest range DISCUSSION
in the number of participants in a ZCTA/ZIP Code was measured Our findings support earlier work about rural–urban differences
in areas identified as a large rural city/town (1–884 participants). in access to health-related resources (33). As expected, CDSME
Table 4 presents factors associated with successful completion programs were less prevalent in rural versus urban areas. How-
of the CDSME program. A greater likelihood of successful com- ever, this study highlights that CDSME workshops are reaching
pletion was associated with being Black (OR = 1.25), or another rural areas in the US, although this reach is less than 25% of
race/ethnicity (OR = 1.32) versus being non-Hispanic White. A all rural areas. This is critical because CDSME programs have
greater likelihood of successful completion was also associated been shown to facilitate improvements in health status and other
with living in a rural county (OR = 1.10). Factors associated with health-related outcomes among adults. CDSME programs assist
lower likelihood of successful completion of the CDSME program participants to set goals, problem solve and do action planning

www.frontiersin.org April 2015 | Volume 2 | Article 172 | 113

Towne et al. Rural CDSME programs

Table 4 | Likelihood of successful completion of the CDSMP. make CDSME programs available rural residents’ homes include
embedding programs into existing local community infrastruc-
Odds p-Value Confidence tures such as health care clinics or agricultural extension health
ratio intervals (95%) services. Engaging multiple delivery sites in these communities
(through the aging services network and public health system) is
Race
encouraged. For example, offering programs in faith-based orga-
White (referent)
nizations have been shown to improve participant reach (34).
Black 1.249* <0.0001 1.194 1.305
Embedding these programs in as regular offerings in organiza-
AIAN 0.923 0.0023 0.804 1.060
tions where rural residents frequently attend may increase their
Asian 1.209** 0.0342 1.098 1.331
participation and foster long-term program sustainability.
Other 1.318* <0.0001 1.246 1.395
Another strategy to better serve rural communities with
Hispanic 0.994 0.0008 0.927 1.067
CDSME programs includes cross-training workshop facilitators
Rurality
to be certified to an array of EBP (e.g., disease self-management,
Rural county 1.095* <0.0001 1.051 1.140
fall prevention). Cross-training these facilitators can increase the
HPSA
capacity of rural areas to deliver a collection of diverse EBP, each
Non-HPSA (referent)
of which differ in purpose to meet the needs of rural residents and
Partial HPSA 0.988 0.1588 0.936 1.042
their caregivers. While increasing the availability of EBP in rural
Full HPSA 0.926* 0.0002 0.877 0.977
communities is essential, increasing access (and repeated access)
Sex
to workshops is of equal importance. Once recruited into the pro-
Female (referent)
gram, additional efforts are needed to ensure participants remain
Male 0.924* <0.0001 0.888 0.961
in the program long enough to receive sufficient intervention
Household status
dose for desired effects. Possible strategies to improve participants’
Lives with others (referent)
Lives alone 1.017 0.3376 0.983 1.052
access to and retention within workshops may include the cre-
ation of participant “buddy systems,” exploring options for free
*Indicates significant differences (p < 0.01) using logistic regression. or low-cost transportation services (e.g., shared rides or volunteer
**Indicates significant differences (p < 0.05) using logistic regression. drivers), including technologically driven approaches, or holding
meetings in community settings where older adults are already
congregating.
that can help in medical, emotional, and social role management More research should be conducted to identify differences in
of chronic conditions (16–18). how programs are delivered in rural versus urban areas (e.g.,
Rural residents face several issues related to health care and dis- strategies for recruitment and retention of different community
ease prevention program access (5–7). Identifying efficient ways partners; targeting different delivery settings; and determining
to bridge access issues in rural areas is critically important for ideal but feasible class size). Further investigation is also needed
those who are older and have one or more chronic conditions. to assess the health-related impact of programs in rural versus
Improving the rural reach of EBP is one example of bridging this urban areas, with special attention to cost-benefit issues. Future
gap and linking rural residents to appropriate health care ser- efforts should also examine whether differences by region or US
vices intended to improve-health outcomes (34). Thus, examining territory exist (e.g., comparisons between continental US and
strategies that bolster participation rates in rural and urban areas is Hawaii/Puerto Rico).
warranted. More research is needed to identify why rural residents
had somewhat higher completions rates when compared to urban LIMITATIONS
residents. Overall, rural adults may be harder to reach and have The measure of rurality used in health services research is an
other barriers related to social support, as exemplified by rural par- important consideration in studies about rurality because the des-
ticipants reporting higher rates of living alone (35, 36). In addition, ignated selection has potential to change areas of comparison and
the somewhat higher rates in the number of chronic conditions influence study findings (42). Our definition of rurality varied
among rural residents may make this population potentially more across the level of analysis. We used both a county-level mea-
vulnerable to self-care issues and in need of CDSME programs. sure (UIC) and the ZCTA/ZIP Code-level measure (RUCA), which
In the current study, the smaller number of participants in rural assessed rurality in both larger areas (i.e., counties) and on a more
versus urban ZCTA/ZIP Codes may be related to the smaller num- micro-level (i.e., ZCTA/ZIP Codes). Thus, our use of different lev-
ber of eligible participants in these areas (i.e., population density els of rurality in this study provides a more complete picture of
and geographic isolation) (37) and the difficulty of some potential geospatial differences. While CDSME workshops were delivered in
participants getting to centralized locations (e.g., longer distance, Puerto Rico, the measure of rurality used (i.e., 2006 RUCA Codes)
limited transportation) (38–41). To adequately serve rural popu- was not available for Puerto Rico (43). As such, we were unable to
lations, efforts are needed to ensure these programs are delivered provide accurate estimates of delivery by rurality for this area in
in areas closer to potential/existing participants’ homes. Offering the current study.
these programs in closer proximity to rural participants’ residences Data presented in the current study is based on the level of rural
has potential to increase attendance rates because it can reduce the residents reached by the CDSME programs only among those who
time and distance traveled to get to workshop sessions. Strategies participated in this initiative. We do however, provide the rural

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 172 | 114
Towne et al. Rural CDSME programs

reach by geographic distribution (i.e., reach within areas). Fur- 7. MacDowell M, Glasser M, Fitts M, Nielsen K, Hunsaker M. A national view of
ther, distances traveled by participants to attend workshops were rural health workforce issues in the USA. Rural Remote Health (2010) 10(3):1531.
8. Gamm L, Hutchison L, Dabney BJ, et al. Rural Healthy People 2010: A companion
not measured, thus we could not determine if time or distance
document to Healthy People 2010. College Station, TX: The Texas A&M Univer-
traveled influenced workshop attendance. Additionally, the level sity System Health Science Center, School of Rural Public Health, Southwest
of missing data is not uncommon to community-based interven- Rural Health Research Center (2003).
tions (44–46). While there was substantial missing data, this may 9. Norton CH, McManus MA. Background tables on demographic character-
have been attributed more to the sites’ administrative ability to col- istics, health status, and health services utilization. Health Serv Res (1989)
23(6):725–56.
lect field data than from individual data refusal (47). Because the
10. Jones CA, Parker TS, Ahearn M, Mishra AK, Variyam JN. Health Status and
analyses performed in this study were not longitudinal, we could Health Care Access of Farm and Rural Populations., EIB-57. Washington, DC:
not measure changes in the rural reach of the CDSME programs U.S. Department of Agriculture, Economic Research Service (2009). p. 1–64.
over time. Designing such longitudinal analyses is highly recom- 11. Bennett KJ, Olatosi B, Probst JC. Health Disparities: A Rural – Urban Chartbook.
mended as a next step in identifying whether progress is being South Carolina: Rural Health Research Center (2008).
12. Lengerich EJ, Wyatt SW, Rubio A, Beaulieu JE, Coyne CA, Fleisher L, et al. The
made in reaching rural residents. We acknowledge that because Appalachia cancer network: cancer control research among a rural, medically
of our large sample size seemingly small comparative differences underserved population. J Rural Health (2004) 20(2):181–7. doi:10.1111/j.1748-
were statistically significant. To be more conservative and protect 0361.2004.tb00026.x
against Type I error, we used a p-value of 0.01 in all study analyses. 13. Towne SD, Smith ML, Ory MG. Geographic variations in access and utilization
of cancer screening services: examining disparities among American Indian and
Alaska Native Elders. Int J Health Geogr (2014) 13:18. doi:10.1186/1476-072X-
CONCLUSION 13-18
The current study helps lessen the gap in what is known about 14. Smith ML, Quinn C, Gipson R, Wilson AD, Ory MO. Serving rural communities
the rural reach of CDSME programs and factors affecting suc- for falls prevention: the dissemination of a matter of balance in the Brazos Valley
cessful completion. Findings highlight the capability of CDSME Region of Texas. Tex Pub Health J (2011) 63(1):54–8.
15. Smith ML, Ahn SN, Sharkey JR, Horel S, Mier N, Ory MG. Successful falls pre-
programs to reach rural residents, yet dissemination efforts can
vention programming for older adults in Texas Rural–Urban Variations. J Appl
be enhanced to ensure minorities and individuals in HPSAs utilize Gerontol (2012) 31(1):3–27. doi:10.1177/0733464810378407
this program. Tailored strategies are needed to increase participant 16. Lorig KR, Sobel DS, Stewart AL, Brown BW, Bandura A, Ritter P, et al. Evidence
recruitment and retention in rural areas to overcome traditional suggesting that a chronic disease self-management program can improve health
barriers to health service access. Assessing the infrastructure in status while reducing hospitalization: a randomized trial. Med Care (1999)
37(1):5–14. doi:10.1097/00005650-199901000-00003
rural areas may be helpful for identifying viable partners for those 17. Lorig KR, Sobel DS, Ritter PL, Laurent D, Hobbs M. Effect of a self-management
seeking to deliver EBP to residents of rural areas, creating greater program on patients with chronic disease. Eff Clin Practi (2001) 4(6):256–62.
uptake, reach, and sustainability. 18. Lorig KR, Ritter P, Stewart AL, Sobel DS, Brown BW Jr, Bandura A, et al. Chronic
disease self-management program: 2-year health status and health care uti-
lization outcomes. Med Care (2001) 39(11):1217–23. doi:10.1097/00005650-
ACKNOWLEDGMENTS
200111000-00008
The American Recovery and Reinvestment Act of 2009 (i.e., Recov- 19. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, Whitelaw N, et al. Successes
ery Act) Communities Putting Prevention to Work: Chronic- of a national study of the chronic disease self-management program: meet-
Disease Self-Management Program initiative, led by the U.S. ing the triple aim of health care reform. Med Care (2013) 51(11):992–8.
Administration on Aging in collaboration with the Centers for doi:10.1097/MLR.0b013e3182a95dd1
20. Lorig K, Holman H, Sobel D, Laurent D. Living a Healthy Life with Chronic Con-
Disease Control and Prevention and the Centers for Medicare and
ditions: Self Management of Heart Disease, Arthritis, Diabetes, Asthma, Bronchitis,
Medicaid Services, allotted $32.5 million to support the transla- Emphysema and others. 3rd ed. Boulder, CO: Bull Publishing Company (2006).
tion of the Stanford program in 45 States, Puerto Rico, and the 21. Bandura A. Social cognitive theory of self-regulation. Organ Behav Hum Decis
District of Columbia. The National Council on Aging served as Process (1991) 50(2):248–87. doi:10.1016/0749-5978(91)90022-L
the Technical Assistance Resource Center for this initiative and 22. Ory MG, Ahn S, Jiang L, Lorig K, Ritter P, Laurent DD, et al. National study of
chronic disease self-management six-month outcome findings. J Aging Health
collected de-identified data on program participation. (2013) 25(7):1258–74. doi:10.1177/0898264313502531
23. Ahn S, Basu R, Smith ML, Jiang L, Lorig K, Whitelaw N, et al. The impact
REFERENCES of chronic disease self-management programs: healthcare savings through a
1. Anderson G, Horvath J. The growing burden of chronic disease in America. community-based intervention. BMC Public Health (2013) 13(1):1141. doi:10.
Public Health Rep (2004) 119(3):263–70. doi:10.1016/j.phr.2004.04.005 1186/1471-2458-13-1141
2. Broemeling AM, Watson DE, Prebtani F. Population patterns of chronic health 24. U.S. Department of Health and Human Services Administration on Aging.
conditions, co-morbidity and healthcare use in Canada: implications for policy ARRA – Communities Putting Prevention to Work: Chronic Disease Self-
and practice. Healthc Q (2007) 11(3):70–6. doi:10.12927/hcq.2008.19859 Management Program. Washington, DC. (2012).
3. Nolte E, McKee M. Caring For People with Chronic Conditions: A Health System 25. Ory MG, Smith ML, Patton K, Lorig K, Zenker W, Whitelaw N. Self-management
Perspective. Berkshire: McGraw-Hill International (2008). at the tipping point: reaching 100,000 Americans with evidence-based programs.
4. Glasgow N, Berry EH. Rural Aging in 21st Century America. Heidelberg: Springer J Am Geriatr Soc (2013) 61(5):821–3. doi:10.1111/jgs.12239
(2013). doi:10.1007/978-94-007-5567-3 26. Area Health Resources Files (ARF). US Department of Health and Human Ser-
5. Council on Graduate Medical Education. Physician Distribution and Health Care vices, Health Resources and Services Administration. Rockville, MD: Bureau of
Challenges in Rural and Inner City Areas: Tenth Report to Congress and the Depart- Health Professions (2013).
ment of Health and Human Services Secretary. Rockville, MD: Health Resources 27. United States Department of Agriculture Economic Research Service. Available
and Services Administration, US Dept of Health and Human Services (1998). from: http://www.ers.usda.gov/data-products/urban-influence-codes.aspx#
6. Knapp KK, Paavola FG, Maine LL, Sorofman B, Politzer RM. Availability of pri- .UwahVfldV8E
mary care providers and pharmacists in the United States. J Am Pharm Assoc 28. WWAMI Rural Health Research Center. RUCA Data: Using RUCA Data. Avail-
(1999) 39(2):127–35. able from: http://depts.washington.edu/uwruca/ruca-uses.php

www.frontiersin.org April 2015 | Volume 2 | Article 172 | 115

Towne et al. Rural CDSME programs

29. Health Resources and services Administration (HRSA). Primary Medical Care 43. WWAMI Rural Health Research Center. Available from: http://depts.
HPSA Designation Overview. Available from: http://bhpr.hrsa.gov/shortage/ washington.edu/uwruca/ruca-projects.php (2014).
hpsas/designationcriteria/primarycarehpsaoverview.html 44. Edwards KE, Gibson N, Martin J, Mitchell S, Andersson N. Impact of
30. Stanford School of Medicine. Chronic Disease Self-Management Program (Better community-based interventions on condom use in the Tłicho region of
Choices, Better Health®Workshop) (2014). Available at: http://patienteducation. Northwest Territories Canada. BMC Health Serv Res (2011) 11(Suppl 2):S9.
stanford.edu/programs/cdsmp.html doi:10.1186/1472-6963-11-S2-S9
31. SAS Institute Inc. Sas 9.4 by SAS Institute Inc. Cary, NC: SAS Institute Inc. 45. Mayworm AM, Sharkey JD. Gender-specific mental health outcomes of a
Copyright © 2002–2013. community-based delinquency intervention. J Juv Just (2013) 3(1).
32. ESRI Inc. ArcGIS: Release 10.2 by ESRI Inc. Redlands, CA: Environmental Sys- 46. Temel S, van Voorst SF, Jack BW, Denktas S, Steegers EA. Evidence-based
tems Research Institute. Copyright ©1999–2013. preconceptional lifestyle interventions. Epidemiol Rev (2014) 36(1):19–30.
33. Towne SD, Lee C, Smith ML, Pulczinski J, Swierc SM, Coughlin R, et al. Health- doi:10.1093/epirev/mxt003
related quality of life & well-being. In: Bolin J, editor. Rural Health People 2020: 47. Kulinski K, Boughtaugh M, Smith ML, Ory MG, Lorig K. Setting the stage: mea-
A Companion Document to Healthy People 2020, Vol. 2, Chapter 14. Southwest sure selection, coordination, and data collection for a national self-management
Rural Health Research Center, School of Public Health, Texas A&M University initiative. Front Public Health (2015) 2:206. doi:10.3389/fpubh.2014.00206
System Health Science Center (2014).
34. Smith ML, Belza B, Altpeter M, Ahn S, Dickerson JB, Ory MG. Disseminat-
ing an evidence-based disease self-management program for older Americans: Conflict of Interest Statement: The authors declare that the research was conducted
implications for diversifying participant reach through delivery site adoption. in the absence of any commercial or financial relationships that could be construed
In: Maddock J, editor. Public Health: Social and Behavioral Health. Rijeka: InTech as a potential conflict of interest.
(2012). p. 385–404.
35. Eshbaugh EM. Perceptions of living alone among older adult women. J Commun This paper is included in the Research Topic, “Evidence-Based Programming for Older
Health Nurs (2008) 25(3):125–37. doi:10.1080/07370010802221685 Adults.” This Research Topic received partial funding from multiple government and
36. Culo S. Risk assessment and intervention for vulnerable older adults. B C Med J private organizations/agencies; however, the views, findings, and conclusions in these
(2011) 53(8):421–5. articles are those of the authors and do not necessarily represent the official position
37. United States Department of Agriculture. Economic Research Service. What is of these organizations/agencies. All papers published in the Research Topic received
Rural (2014). Available from: http://www.ers.usda.gov/topics/rural-economy- peer review from members of the Frontiers in Public Health (Public Health Education
population/rural-classifications/what-is-rural.aspx#.UyCRHPldV8E and Promotion section) panel of Review Editors. Because this Research Topic repre-
38. Freeman VA, Patterson D, Slifkin RT. Issues in Staffing Emergency Medical Ser- sents work closely associated with a nationwide evidence-based movement in the US,
vices: Results from a National Survey of Local Rural and Urban EMS Directors. many of the authors and/or Review Editors may have worked together previously in
Report no. 93. Chapel Hill: University of North Carolina at Chapel Hill, US some fashion. Review Editors were purposively selected based on their expertise with
Department of Health and Human Services (2008). evaluation and/or evidence-based programming for older adults. Review Editors were
39. Genovesi AL, Hastings B, Edgerton EA, Olson LM. Pediatric emergency care independent of named authors on any given article published in this volume.
capabilities of Indian Health Service emergency medical service agencies serv-
ing American Indians/Alaska Natives in rural and frontier areas. Rural Remote Received: 21 July 2014; accepted: 17 September 2014; published online: 27 April 2015.
Health (2014) 14(2):2688. Citation: Towne SD Jr, Smith ML, Ahn S and Ory MG (2015) The reach of chronic-
40. Peck J, Alexander K. Maternal, Infant, and Child Health in Rural Areas: A Lit- disease self-management education programs to rural populations. Front. Public Health
erature Review. Rural Healthy People 2010: A Companion Document to Healthy 2:172. doi: 10.3389/fpubh.2014.00172
People 2010 (Vol. 2) (2010). This article was submitted to Public Health Education and Promotion, a section of the
41. Velaga NR, Beecroft M, Nelson JD, Corsar D, Edwards P. Transport poverty meets journal Frontiers in Public Health.
the digital divide: accessibility and connectivity in rural communities. J Transp Copyright © 2015 Towne, Smith, Ahn and Ory. This is an open-access article dis-
Geography (2012) 21:102–12. doi:10.1016/j.jtrangeo.2011.12.005 tributed under the terms of the Creative Commons Attribution License (CC BY). The
42. Smith ML, Dickerson JB, Wendel ML, Ahn S, Pulczinski JC, Drake KN, et al. The use, distribution or reproduction in other forums is permitted, provided the original
utility of rural and underserved designations in geospatial assessments of dis- author(s) or licensor are credited and that the original publication in this journal is cited,
tance traveled to healthcare services: implications for public health research and in accordance with accepted academic practice. No use, distribution or reproduction is
practice. J Environ Public Health (2013) 2013:960157. doi:10.1155/2013/960157 permitted which does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 172 | 116
 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00257

Factors associated with successful completion of the

Chronic Disease Self-Management Program among
middle-aged and older Asian-American participants:
a national study
SangNam Ahn 1,2 *, Matthew Lee Smith 3 , Jinmyoung Cho 2,4 , Luohua Jiang 5 , Lindsey Post 1 and
Marcia G. Ory 2
1
Division of Health Systems Management and Policy, School of Public Health, The University of Memphis, Memphis, TN, USA
2
Department of Health Promotion and Community Health Sciences, School of Public Health, Texas A&M Health Science Center, College Station, TX, USA
3
Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA, USA
4
Center for Applied Health Research, Baylor Scott & White Healthcare, Temple, TX, USA
5
Department of Epidemiology, School of Medicine, University of California Irvine, Irvine, CA, USA

Edited by: Asian-Americans are a small but fast-growing population in the United States who are
Shane Andrew Thomas, Monash
increasingly experiencing multiple chronic diseases. While the evidence-based Chronic
University, Australia
Disease Self-Management Program (CDSMP) has been disseminated among various racial
Reviewed by:
Laura Rudkin, University of Texas and ethnic populations, few studies specifically investigate participants with an Asian back-
Medical Branch, USA ground. The study aims to identify characteristics of middle-aged and older Asian-American
Geraldine Sanchez Aglipay, University CDSMP participants (older than 50 years) and investigate factors related to successful
of Illinois at Chicago, USA
workshop completion (i.e., attending 4+ of the 6 sessions) among this population. Data
*Correspondence:
were analyzed from 2,716 middle-aged and older Asian-Americans collected during a 2-year
SangNam Ahn, Division of Health
Systems Management and Policy, national dissemination of CDSMP. Multilevel logistic regression analyses were conducted to
School of Public Health, The identify individual- and workshop-level covariates related to successful workshop comple-
University of Memphis, 133 Robison tion. The majority of participants were female, living with others, and living in metro areas.
Hall, Memphis, TN 38152-3530, USA
The average age was 71.3 years old (±9.2), and the average number of chronic conditions
e-mail: [email protected]
was 2.0 (±1.5). Successful completion of CDSMP workshops among participants was asso-
ciated with their number of chronic conditions (OR = 1.10, P = 0.011), living in non-metro
areas (OR = 1.77, P = 0.009), attending workshops from area agencies on aging (OR = 1.56,
P = 0.018), and attending a workshop with higher completion rates (OR = 1.03, P < 0.001).
This study is the first large-scale examination of Asian-American participants enrolled in
CDSMP and highlights characteristics related to intervention attendance among this under-
studied minority population. Knowing such characteristics is important for serving the
growing number of Asian-Americans with chronic conditions.
Keywords: Asian-Americans, chronic disease management, Chronic Disease Self-Management Program,
evidence-based programs

INTRODUCTION and White (64%) counterparts, the burden of chronic conditions

Almost three-quarters of Asian-American adults are foreign-born, among Asian-Americans should be carefully scrutinized based on
representing many countries of origin, including China, India, and population projections (2–4). As the total population of Asian-
the Philippines (1). Collectively, these subgroups constitute the Americans increases, it is expected that a greater number of
fastest-growing ethnic group in the country, representing almost Asian-Americans will suffer from chronic conditions.
6% of the U.S. population (1). The Asian-American population The Chronic Disease Self-Management Program (CDSMP)
in the U.S., which was estimated at 18.9 million in 2012 (2), has been introduced and widely disseminated in the U.S. as a
grew by 46% from 2000 to 2010, and is outpacing the growth method to empower patients with self-management skills to deal
of other racial/ethnic groups. Between 2011 and 2012, the rate of with their chronic conditions (5). Drawing upon social learning
population increase was 2.9% for Asian-Americans, 2.2% for His- theory (6), CDSMP is an evidence-based, peer-led intervention
panics; 2.2% for Native Hawaiians and Other Pacific Islanders; consisting of six highly participative classes held for 2.5 h each,
1.5% for American-Indians and Alaska Native; and 1.3% for once a week, for six consecutive weeks (5). CDSMP has resulted
African-Americans (2). This population growth warrants further in improved healthcare and health (7, 8), while potentially sav-
study of health conditions among Asian-Americans. Although the ing healthcare costs (9). While CDSMP has been successfully
prevalence rate of chronic conditions among Asian-Americans disseminated among diverse populations, there are few studies
(42%) is lower than their African-American (77%), Latino (68%), focusing specifically on the characteristics of middle-aged or older

www.frontiersin.org April 2015 | Volume 2 | Article 257 | 117

Ahn et al. CDSMP among Asian-American Participants

Asian-Americans enrolled in CDSMP or examining the factors total and were coded as “other” for purpose of the study and their
associated with completing CDSMP in this population. Previous low distribution (<13%). Workshop composition varied in the
studies have shown that Asian-Americans complete CDSMP at a proportion of Asian-Americans participating. We hypothesized
somewhat higher rate than the general participant population and that workshops with more racial/ethnic homogeneity might have
at about the same rate as White participants (10). Thus, the objec- higher completion rates due to shared culture and language (14,
tive of the study was to analyze the dataset more closely to (1) 15). Workshops with larger proportions of participants success-
identify characteristics of Asian-American CDSMP participants fully completing the intervention might also signify greater social
in the 2010–2012 national dissemination of CDSMP in the U.S.; cohesion and support (i.e., higher completion workshop). We also
and (2) identify the factors associated with CDSMP completion hypothesized that Asian-American participants in workshops with
among middle-aged and older Asian-American participants. higher overall completion rates would have higher completion
rates themselves (16). As such, we computed the percentages of
METHODS Asian-Americans and successful completers in each workshop. To
DATA SOURCE AND STUDY POPULATION avoid endogeneity issue, we excluded the current participant from
Cross-sectional data for this study were retrospectively obtained their workshop when calculating the workshop completion rate.
from a nationwide delivery of CDSMP as part of the American In other words, the resulting workshop completion rate repre-
Recovery and Reinvestment Act of 2009 (i.e., Recovery Act) Com- sents the average completion rate among the classmates of each
munities Putting Prevention to Work: CDSMP initiative (11). The participant. The proportions of Asian-American participants and
U.S. Administration on Aging led this initiative in collaboration average workshop completion rates were included in analyses as
with the Centers for Disease Control and Prevention and the Cen- workshop-level covariates.
ters for Medicare and Medicaid Services to support the translation
of CDSMP in 45 states, Puerto Rico, and the District of Columbia STATISTICAL ANALYSIS
(12). This initiative was conducted between 2010 and 2012 with the To compare the characteristics of the participants who com-
goal of reaching the diverse population of the Americans embed- pleted the CDSMP workshop to those who did not, we used
ding the delivery structures into statewide systems (11). Within the χ2 -tests for categorical independent variables and two-sample
first 2 years of this initiative, there were more than 100,000 adults t -test for continuous independent variables. Multilevel logistic
participating in 9,305 workshops in 1,234 U.S. counties (11). For regression models were used to investigate the association between
this study, data were analyzed from 2,716 Asian-American partic- successful workshop completion and individual-level, as well as
ipants (i.e., aggregate Asian ethnic groups) who aged 50 years or workshop-level covariates. First, all individual-level independent
older and responded to all relevant survey questions. variables were introduced into a multilevel logistic regression
model (Model 1). Then, we generated another multilevel logis-
MEASURES tic regression model after including workshop-level variables
DEPENDENT VARIABLE (Model 2). The proportion of variance explained (PVE) at the
Chronic Disease Self-Management Program workshop attendance workshop level by different levels of variables was calculated as:
was the dependent variable for this study. Successful completion PVE = (V 0 − V 1 )/V 0 × 100, where V 0 is the second-level vari-
was defined as attendance at four or more of the six workshop ance of the Null Model, and V 1 is the second-level variance of the
sessions, which is consistent with definitions used by the program adjusted model (17). Multilevel analyses were conducted using the
developers and in a variety of other studies (7–9, 11). “gllamm” command in Stata 11 (18).

INDIVIDUAL- OR NEIGHBORHOOD-LEVEL COVARIATES RESULTS

As individual-level covariates, socio-demographic factors included Table 1 shows frequency distributions of independent variables
age (in years), sex (male vs. female), and living arrangement among the total population, and then divided by workshop
(living with others vs. living alone). Health status was mea- completion status; 79.5% of participants successfully completed
sured by the number of self-reported chronic conditions (i.e., CDSMP workshops (n = 2,159) and 20.5% did not (n = 557). As
arthritis, cancer, depression, diabetes, heart disease, hypertension, a whole, Asian-American participants were predominantly female
lung disease, stroke, osteoporosis, and other chronic conditions). (73.1%), living with others (97.5%), and living in metro areas
As neighborhood-level covariates, median household income (in (91.3%). The average age was 71.3 years old (±9.2), and the average
$10,000 units) was included based on the participants’ ZIP Code. number of chronic conditions was 2.0 (±1.5).
Rural–Urban Commuting Area Codes based on participants’ ZIP Chronic Disease Self-Management Program completion was
Code information were used to categorize participants’ residence not significantly different in terms of age, sex, neighborhoods
(metro vs. non-metro) (13). median income, or living arrangement. Considering individual-
level variables, participants living in non-metro areas had sig-
WORKSHOP-LEVEL COVARIATES nificantly higher completion rates than their urban counterparts
Workshop delivery sites included area agencies on aging (P = 0.009). The number of chronic conditions was higher among
(AAA)/senior centers, healthcare organizations, residential facil- those who completed CDSMP workshops relative to those who
ities, community or multi-purpose centers, faith-based organiza- did not (P = 0.009). All workshop-level variables significantly dif-
tions, educational institutions, recreational centers, tribal centers, ferentiated between those who completed CDSMP workshops
and workplaces. The last four delivery sites made up <13% of the and who did not. Those who attended AAA delivery sites were

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 257 | 118
Ahn et al. CDSMP among Asian-American Participants

Table 1 | Characteristics of middle-aged and older Asian-American Chronic Disease Self-Management Program participants by CDSMP
completion (N = 2,716).

Total (N = 2,716) CDSMP non-completion CDSMP completion

(N = 557) (N = 2,159)

N (%) N (%) N (%) P-value*

Female 1, 985 (73.1) 397 (71.3) 1, 588 (73.6) 0.280

Living alone 68 (2.5) 19 (3.4) 49 (2.3) 0.124
Rural–urban commuting area codes 0.009
Metro 2, 479 (91.3) 524 (94.1) 1, 955 (90.6)
Non-metro 237 (8.7) 33 (5.9) 204 (9.5)
Workshop delivery site 0.010
Other 341 (12.6) 77 (13.8) 264 (12.2)
Area agency on aging/senior centers 741 (27.3) 123 (22.1) 618 (28.6)
Health care organization 379 (14.0) 77 (13.8) 302 (14.0)
Residential facilities 431 (15.9) 111 (19.9) 320 (14.8)
Community/multi-purpose 720 (26.5) 148 (26.6) 572 (26.5)
Faith-based organization 104 (3.8) 21 (3.8) 83 (3.8)

Mean (±SD) Mean (±SD) Mean (±SD) P -valuea

Age in years (from 50 to 101) 71.3 (±9.2) 71.2 (±9.5) 71.4 (±9.2) 0.670
Neighborhood median income in $10,000 unit (from 2.2 to 11.6) 6.2 (±1.4) 6.2 (±1.5) 6.2 (±1.3) 0.466
Number of chronic conditions (from 0 to 10) 2.0 (±1.5) 1.9 (±1.4) 2.1 (±1.5) 0.009
% Of Asian-Americans in the workshop 57.5 (±36.8) 53.2 (±36.7) 58.6 (±36.7) 0.002
Workshop completion rate (%) 74.9 (±18.7) 66.4 (±18.7) 77.1 (±18.0) <0.001

*P-value for χ2 -test comparing the participants who completed the CDSMP workshop and who did not. Other workshop delivery sites include educational institutions,
recreational centers, tribal centers, and workplaces.
a
P-value for two-sample t-test comparing the participants who completed the CDSMP workshop and who did not.

more likely to complete CDSMP workshops, whereas those who Meanwhile, the second-level variance explained by the workshop-
attended residential facilities or other sites had lower completion level variables (in Model 2) was (1.41–0.16)/1.41 × 100 = 88.7%,
rates (P = 0.010). In addition, CDSMP workshop completion was indicating that the workshop-level variables explained 88.7%
positively associated with the percentage of Asian-American par- of the variability found at the second-level compared with the
ticipants (P = 0.002) and the workshop completion rate in each Model 1.
workshop (P < 0.001).
Table 2 shows results of the multilevel logistic regressions DISCUSSION
including the Null Model (i.e., intercept-only model), Model 1 This retrospective analysis is unique in that it identifies corre-
(only including individual-level variables), and Model 2 (includ- lates significantly related to CDSMP completion at individual- and
ing both individual-level and workshop-level variables). In Model workshop-levels among Asian-American participants. The study
1, a higher number of chronic conditions and living in non- findings are especially relevant given that Asian-Americans are the
metro areas increased the odds of completing CDSMP work- fastest-growing population in the U.S. The current study reveals
shops (OR = 1.09, P = 0.030; OR = 1.84, P = 0.025, respectively). that Asian-American participants are similar to other CDSMP
In Model 2, the odds of completing CDSMP workshops increased participants in this national dissemination in terms of being pre-
among participants with a higher number of chronic conditions dominantly female, living with others, residing in metro areas, and
(OR = 1.10, P = 0.011), living in non-metro areas (OR = 1.77, having multiple chronic conditions (7, 11). Workshop comple-
P = 0.009), and those who attended a workshop with a higher tion was positively associated with number of chronic conditions,
completion rate (OR = 1.03, P < 0.001). Those who attended attending workshops from AAA, attending higher completion
workshops from AAA (OR = 1.56, P = 0.018) were more likely to workshops, and rural residence. These factors independently or
complete CDSMP workshops compared to those who attended in combination contributed to the 80% CDSMP completion rate
CDSMP workshops from other delivery sites (i.e., educational among Asian-American participants. Recent studies utilizing the
institution, recreational center, tribal center, and workplace). The same dataset reported that the average CDSMP completion rate
second-level variance explained by the individual-level variables was 75% (n = 89,861) (10, 11), which is slightly lower than that
(in Model 1) was (1.45–1.41)/1.45 × 100 = 2.8%, indicating that (i.e., 80%) of the Asian-American participants shown in the cur-
the individual-level variables explained 2.8% of the variabil- rent study. They also found no significant difference in comple-
ity found at the second-level compared with the Null Model. tion rates between Asian-American and White participants (10).

www.frontiersin.org April 2015 | Volume 2 | Article 257 | 119

Ahn et al. CDSMP among Asian-American Participants

Table 2 | Individual and workshop characteristics associated with successful completion of Chronic Disease Self-Management Program
(CDSMP) among middle-aged and older Asian-American participants (N = 2,716).

Empty model Model 1 Model 2

Adjusted OR (95% CI) Adjusted OR (95% CI) Adjusted OR (95% CI)

Individual characteristics
Age – 1.00 (0.99–1.01) 1.00 (0.99–1.02)
Sex
Male – Ref. Ref.
Female – 1.12 (0.88–1.43) 1.08 (0.86–1.35)
Number of chronic conditions – 1.09 (1.01–1.18) 1.10 (1.02–1.18)
Neighborhood median income – 1.06 (0.96–1.18) 1.04 (0.96–1.12)
Living arrangement
Living with others – Ref. Ref.
Living alone – 0.69 (0.36–1.31) 0.71 (0.40–1.27)
Rural–urban commuting area codes
Metro – Ref. Ref.
Non-metro – 1.84 (1.08–3.15) 1.77 (1.15–2.71)

Workshop characteristics
Workshop delivery site
Other – – Ref.
Area agency on aging/senior centers – – 1.56 (1.08–2.25)
Health care organization – – 1.42 (0.94–2.15)
Residential facilities – – 1.12 (0.76–1.67)
Community/multi-purpose – – 1.35 (0.91–2.00)
Faith-based organization – – 1.21 (0.64–2.27)
% Of Asian-Americans in the workshop – – 1.00 (0.99–1.01)
Workshop completion rate (%) – – 1.03 (1.02–1.03)
Between-area variation (SE) 1.45 (0.30) 1.41 (0.30) 0.16 (0.14)

Other workshop delivery sites include educational institutions, recreational centers, tribal centers, and workplaces.
OR are printed in bold if P < 0.05.
OR, odds ratio; 95% CI, 95% confidence interval; Ref., referent.

However, there has been no further analysis of the factors that Rural residence seems to be an important factor related to
might influence completion rates among Asian-Americans. CDSMP completion among Asian-Americans. The current study
The positive association between the participants’ health status reveals 8.7% of participants reported living in non-metro areas,
(i.e., represented by the number of chronic conditions) and their which is two times greater than the 2010 Census estimates of
likelihood of completing the program is consistent with recent rural-residing Asian-Americans (4.3%) (22). In the current study
CDSMP evaluations (8). This result is relatively encouraging given including both individual- and workshop-level variables, Asian-
that an earlier study found those with chronic conditions had American participants who lived in non-metro areas were 77%
lower attendance rates of behavioral interventions in the com- more likely to complete CDSMP compared to those who lived
munity (19). Mobility issues related to chronic conditions have in metro areas. There are multiple potential explanations for this
been cited as probable barriers to intervention adherence (20, 21). finding. First, there could be more social integration in rural com-
There is little evidence to help us disentangle the positive associa- munities so that participants could be more likely to know leaders,
tion between severe health conditions and workshop completion. organizers, and participants, leading to a natural support system
However, when study participants engage in highly participative for encouraging attendance. Similarly, it could be easier to “get
workshops focusing on goal setting and problem solving skills the word out” to participants in rural communities once rural
needed by those coping with multiple chronic conditions, their informants were reached. Alternatively, if fewer participants are
motivations to attend these workshops may outweigh other fac- enrolled in rural workshops, it might be easier for leaders to pro-
tors, which would limit attendance, such as pain, depression, or vide reminders about workshop sessions and communicate with
mobility issues. As such, this research highlights the importance of participants outside of workshop time. In our additional analysis,
recognizing the value of prevention across the disease continuum average number of participants enrolled in rural workshops (i.e.,
and especially targeting those with multiple chronic conditions for 11.7) was significantly lower than that of urban workshops (i.e.,
CDSMP. 14.2) (P < 0.001). Second, the general lack of access to healthcare

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 257 | 120
Ahn et al. CDSMP among Asian-American Participants

in rural areas (23, 24) might encourage participants in rural current study also found that utilizing the multilevel analysis (i.e.,
areas to take advantage of available behavioral interventions like individual- and workshop-levels) is highly recommended since
CDSMP. More in-depth analysis is needed to reveal which aspects including workshop-level variables explained more than 88% of
of the non-metro CDSMP facilitated more successful completion. the second-level variance.
Along this line, future studies should further investigate the extent Despite the study’s unique contribution to the literature, some
to which rural residence affects completion of evidence-based limitations should be considered. First, these results were based on
programs among Asian and other smaller racial/ethnic groups. cross-sectional data, which limit our ability to determine a causal
When workshop-level variables were included in the analy- relationship between any of the variables and CDSMP completion.
sis, one of the interesting results was the significant association Second, the study participants are not nationally representative,
between program completion and specific delivery site. Asian- which limits the generalizability of these study findings. Third, our
American CDSMP participants were more likely to complete data relied on program participants’ self-reported measurements
CDSMP when they attended workshops from AAA. In a sense, that may generate recall bias or social desirability bias. However,
AAAs or senior centers are especially advantaged in delivering we were not able to find problematic patterns related to this con-
evidence-based programs to minority groups because AAAs have cern. Third, and most importantly, it is over-simplistic to lump all
had a longer experience of providing educational services or con- Asian-Americans together in one category. Because of the multi-
gregate meals that have attracted diverse populations (25). This plicity of nationalities and unique cultures in Asia, it is difficult
early advantage is strengthened by new mandates for AAAs to to make generalizations or draw conclusions about such a broadly
deliver evidence-based programs for diverse groups of seniors defined, diverse population. Additionally, we did not include mea-
(25). Nevertheless, the association between delivery sites and sures, which would clarify participants’ level of acculturation that
CDSMP completion requires another look while considering par- likely influence the variables’ effects on program participation or
ticipants’ residence (metro vs. non-metro). In an additional analy- workshop completion. Unfortunately, the available data did not
sis, we found that AAA (28.2%) and community/multi-purpose allow distinctions among different Asian-Americans; however, we
centers (28.8%) were the two most common delivery sites in metro recommend that sub-classifications of Asian-American partici-
areas, whereas other (i.e., educational institution, recreational cen- pants be collected and analyzed in future studies. Nevertheless,
ter, tribal center, and workplace) (35.2%) and residential facilities the primary purpose of this study was to explore the character-
(19.0%) were two most common delivery sites in non-metro areas. istics of Asian-American CDSMP participants and contributing
This may indicate that adding the workshop characteristics in factors related to program completion, and as such offer initial
Model 2 does not seem to confound the relationship between insights that can be explored further.
rural residence and workshop completion anymore when look-
ing at the relatively small changes of odds ratios of rural residence CONCLUSION
covariate (from 1.84 in Model 1 to 1.77 in Model 2). Nevertheless, The underlying value of our study is the potential to improve the
these results may require additional study to address unanswered implementation and dissemination of successful evidence-based
questions: which delivery sites can reach out to diverse popula- programs among Asian-Americans. A major study conclusion is
tions? What factors related to delivery sites might be associated that completion rates among Asian-American CDSMP partici-
with more successful completion of CDSMP (e.g., rural residence pants were high (approximately 80% of participants), but they
or specific delivery sites or both)? could be improved with careful targeting of these populations
As part of the workshop-level variables, Asian-American par- based on health status, participant’s residence, and workshop
ticipants who attended workshops with higher completion rates delivery sites. In this way, our findings can inform policy mak-
were more likely to complete the program. There are very few ers, program coordinators, and workers in the field who want to
studies explaining variation of completing interventions in terms expand CDSMP utilization among Asian-Americans. The cru-
of percentage of intervention completers among specific racial and cial next step will be focusing on improving implementation
ethnic groups. One of the plausible explanations would be related and dissemination of CDSMP among diverse segments of Asian-
to efficient or engaging CDSMP leaders. These leaders may instruct Americans. Such actions can help the growing population of
workshops in a way that provides participants with more enjoy- Asian-Americans achieve improvements in health and health care
ment or educational benefit from each session, thereby increasing outcomes.
participants’ motivation to complete the program. Alternatively, ACKNOWLEDGMENTS
when a large portion of participants in a workshop were dili- This work was supported by the Administration on Aging
gently engaging in each session, other“less-interested”participants (90OP0001/03); and the National Institute of Child Health and
may have been effectively encouraged to complete the program Human Development (R01HD047143). The authors also thank
through a form a positive peer pressure. These factors indepen- Ms. Patti Smith, who helped to review this manuscript.
dently or together, coupled with an assumption that Asian people
feel more comfortable in a group (i.e., collectivism rather than REFERENCES
individualism) and tend to follow the group (26) might boost 1. Pew Research Center. Pew Research Social & Demographic Trends: The Rise of
attendance and increase complete rates of CDSMP. While it was Asian Americans (2013). Available from: http://www.pewsocialtrends.org/2012/
06/19/the-rise-of-asian-americans/
assumed that Asian-American participants in classes with higher 2. Census Bureau US. Asians Fastest-Growing Race or Ethnic Group in 2012, Cen-
proportions of Asian-Americans would be more likely to complete sus Bureau Reports (2013). Available from: http://www.census.gov/newsroom/
CDSMP, this a priori hypothesis was not supported. Moreover, the releases/archives/population/cb13-112.html

www.frontiersin.org April 2015 | Volume 2 | Article 257 | 121

Ahn et al. CDSMP among Asian-American Participants

3. Collins KS, Fund C. Diverse Communities, Common Concerns: Assessing Health 20. Jerant AF, Friederichs-Fitzwater MM, Moore M. Patients’ perceived barriers
Care Quality for Minority Americans. New York, NY: Commonwealth Fund to active self-management of chronic conditions. Patient Educ Couns (2005)
(2002). 57(3):300–7. doi:10.1016/j.pec.2004.08.004
4. National Center for Health Statistics. Older Americans 2012: Key Indicators of 21. Jones F. Strategies to enhance chronic disease self-management: how can we
Well-being (2012). Available from: http://www.agingstats.gov/Main_Site/Data/ apply this to stroke? Disabil Rehabil (2006) 28(13–14):841–7. doi:10.1080/
2012_Documents/Docs/EntireChartbook.pdf 09638280500534952
5. Lorig K, Holman H, Sobel D, Laurent D. Living a Healthy Life with Chronic Con- 22. Housing Assistance Council. Rural Research Brief: Race & Ethnicity in Rural
ditions: Self Management of Heart Disease, Arthritis, Diabetes, Asthma, Bronchitis, America (2012). Available from: http://www.ruralhome.org/storage/research_
Emphysema and Others. 3rd ed. Boulder, CO: Bull Publishing Company (2006). notes/rrn-race-and-ethnicity-web.pdf
6. Bandura A. Social cognitive theory of self-regulation. Organ Behav Hum Decis 23. Rosenthal TC, Fox C. Access to health care for the rural elderly. JAMA (2000)
Process (1991) 50(2):248–87. doi:10.1016/0749-5978(91)90022-L 284(16):2034–6. doi:10.1001/jama.284.16.2034
7. Ory MG, Ahn S, Jiang L, Lorig K, Ritter P, Laurent DD, et al. National study of 24. Chan L, Hart LG, Goodman DC. Geographic access to health care for rural
chronic disease self-management six-month outcome findings. J Aging Health Medicare beneficiaries. J Rural Health (2006) 22(2):140–6. doi:10.1111/j.1748-
(2013) 25(7):1258–74. doi:10.1177/0898264313502531 0361.2006.00022.x
8. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, Whitelaw N, et al. Successes 25. Smith LA, McCaslin R. Enhancing baccalaureate student learning in research
of a national study of the chronic disease self-management program: meet- and gerontology through partnership with an area agency on aging. J Teach Soc
ing the triple aim of health care reform. Med Care (2013) 51(11):992–8. Work (2011) 31(1):79–88. doi:10.1080/08841233.2010.539149
doi:10.1097/MLR.0b013e3182a95dd1 26. Power D, Schoenherr T, Samson D. The cultural characteristic of individual-
9. Ahn S, Basu R, Smith ML, Jiang L, Lorig K, Whitelaw N, et al. The impact ism/collectivism: a comparative study of implications for investment in oper-
of chronic disease self-management programs: healthcare savings through a ations between emerging Asian and industrialized Western countries. J Operat
community-based intervention. BMC Public Health (2013) 13(1):1141. doi:10. Manage (2010) 28(3):206–22. doi:10.1016/j.jom.2009.11.002
1186/1471-2458-13-1141
10. Erdem E, Korda H. Self-management program participation by older adults
with diabetes: chronic disease self-management program and diabetes self- Conflict of Interest Statement: The authors declare that the research was conducted
management program. Fam Community Health (2014) 37(2):134–46. doi:10. in the absence of any commercial or financial relationships that could be construed
1097/FCH.0000000000000025 as a potential conflict of interest.
11. Ory MG, Smith ML, Patton K, Lorig K, Zenker W, Whitelaw N. Self-management
at the tipping point: reaching 100,000 Americans with evidence-based programs. This paper is included in the Research Topic, “Evidence-Based Programming for Older
J Am Geriatr Soc (2013) 61(5):821–3. doi:10.1111/jgs.12239 Adults.” This Research Topic received partial funding from multiple government and
12. U.S. Department of Health and Human Services Administration on Aging. private organizations/agencies; however, the views, findings, and conclusions in these
ARRA – Communities Putting Prevention to Work: Chronic Disease Self- articles are those of the authors and do not necessarily represent the official position
Management Program (2012). Available from: http://www.cfda.gov/?s=program of these organizations/agencies. All papers published in the Research Topic received
&mode=form&tab=step1&id=5469a61f2c5f25cf3984fc3b94051b5f peer review from members of the Frontiers in Public Health (Public Health Education
13. U.S. Department of Agriculture Economic Research Service. ERS/USDA Briefing and Promotion section) panel of Review Editors. Because this Research Topic repre-
Room: Measuring Rurality: Rural-Urban Commuting Area Codes (2014). Avail- sents work closely associated with a nationwide evidence-based movement in the US,
able from: http://www.ers.usda.gov/data-products/rural-urban-commuting- many of the authors and/or Review Editors may have worked together previously in
area-codes.aspx some fashion. Review Editors were purposively selected based on their expertise with
14. Barrera M Jr, Castro FG, Strycker LA, Toobert DJ. Cultural adaptations of evaluation and/or evidence-based programming for older adults. Review Editors were
behavioral health interventions: a progress report. J Consult Clin Psychol (2013) independent of named authors on any given article published in this volume.
81(2):196. doi:10.1037/a0027085
15. Cardemil EV. Cultural adaptations to empirically supported treatments: a Received: 16 June 2014; paper pending published: 21 September 2014; accepted: 10
research agenda. Sci Rev Mental Health Pract (2010) 7(2):8–21. November 2014; published online: 27 April 2015.
16. Burke BL, Vassilev G, Kantchelov A, Zweben A. Motivational interviewing with Citation: Ahn S, Smith ML, Cho J, Jiang L, Post L and Ory MG (2015) Factors asso-
couples. In: Miller WR, Rollnick S, editors. Motivational Interviewing. New York, ciated with successful completion of the Chronic Disease Self-Management Program
NY: The Guilford Press (2002). p. 347–61. among middle-aged and older Asian-American participants: a national study. Front.
17. Sanz-Barbero B, García LO, Hernández TB. The effect of distance on the use Public Health 2:257. doi: 10.3389/fpubh.2014.00257
of emergency hospital services in a Spanish region with high population dis- This article was submitted to Public Health Education and Promotion, a section of the
persion: a multilevel analysis. Med Care (2012) 50(1):27–34. doi:10.1097/MLR. journal Frontiers in Public Health.
0b013e31822d5e03 Copyright © 2015 Ahn, Smith, Cho, Jiang , Post and Ory. This is an open-access
18. StataCorp. Stata Statistical Software: Release 11. College Station, TX: StataCorp article distributed under the terms of the Creative Commons Attribution License (CC
LP (2009). BY). The use, distribution or reproduction in other forums is permitted, provided the
19. Virgil KM. Community-Based Exercise Program Attendance and Exercise Self- original author(s) or licensor are credited and that the original publication in this
Efficacy in African American Women. Indianapolis: Indiana University – Purdue journal is cited, in accordance with accepted academic practice. No use, distribution or
University Indianapolis (2013). reproduction is permitted which does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 257 | 122
 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00174

Chronic Disease Self-Management Education (CDSME)

program delivery and attendance among urban-dwelling
African Americans
Chivon A. Mingo 1 *, Matthew Lee Smith 2 , SangNam Ahn 3,4 , Luohua Jiang 5 , Jinmyoung Cho 4,6 ,
Samuel D. Towne Jr.4 and Marcia G. Ory 4
1
Gerontology Institute, College of Arts and Sciences, Georgia State University, Atlanta, GA, USA
2
Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA, USA
3
Division of Health Systems Management and Policy, School of Public Health, The University of Memphis, Memphis, TN, USA
4
Department of Health Promotion and Community Health Sciences, School of Public Health, Texas A&M Health Science Center, College Station, TX, USA
5
Department of Epidemiology and Biostatistics, School of Public Health, Texas A&M Health Science Center, College Station, TX, USA
6
Baylor Scott & White Health, Center for Applied Health Research, Temple, TX, USA

Edited by: Background: Older African Americans carry a disproportionate share of chronic diseases.
Renae L. Smith-Ray, University of
The purpose of this study was to identify the characteristics of urban-dwelling African
Illinois at Chicago, USA
Americans with chronic disease participating in Chronic Disease Self-Management Educa-
Reviewed by:
Pankaja Desai, University of Illinois at tion (CDSME) programs and to examine factors related to successful program completion
Chicago, USA (i.e., attending at least four of the six sessions).
Debra Evelyn Krotish, University of
South Carolina School of Medicine, Methods: Data were analyzed from 11,895 African Americans who attended a CDSME pro-
USA gram at one of the five leading delivery sites (i.e., senior center, health care organization,
Renae L. Smith-Ray, University of residential facility, community location, faith-based organization). Logistic regression analy-
Illinois at Chicago, USA
ses were used to assess the associations of demographic, delivery site, and neighborhood
*Correspondence:
Chivon A. Mingo, Gerontology
characteristics with CDSME program successful completion.
Institute, Georgia State University,
Results: Approximately, half of the African American participants were aged 65–79 years,
P.O. Box 3984, Atlanta, GA
30302-3984, USA 83% were female, and 92% lived alone. Approximately, 44% of participants had three
e-mail: [email protected] or more chronic conditions and 35% resided in an impoverished area (i.e., 200% below
federal poverty level). Successful completion of the CDSME program was associated with
being between the ages of 50–64 and 65–79 years, being female, living alone, living in
an impoverished community, and attending a CDSME program at a residential facility or
community center.
Conclusion: Findings highlight the unique patterns of attendance and delivery within the
context of self-management interventions among this unique and traditionally underserved
target population. Understanding such patterns can inform policy and practice efforts to
engage more organizations in urban areas to increase CDSME program adoption. Partic-
ularly, employing strategies to implement CDSME programs across all delivery site types
may increase reach to African American participants.
Keywords: African American, urban, chronic disease self-management, delivery site, evidence-based program

INTRODUCTION The drastic increase in the number of older adults, particularly

Currently, over 43 million US residents are 65+ years of age with those from diverse racial/ethnic groups, is associated with bur-
that number projected to increase to approximately 80 million geoning rates of chronic disease. Approximately, 50% of aging
by the year 2040 (1). Along with the overall growth of the aging adults report two or more chronic conditions, and data show
population, a significant increase in the racial/ethnic diversity of that the prevalence of having two or more comorbidities is
this population is also occurring. Racial/ethnic minority popula- higher among aging African Americans than other racial/ethnic
tions are projected to increase to 20.2 million making up 28% of groups (i.e., Whites, Hispanics) (2). Moreover, African Americans
the aging population in 2030 (1). Specifically, between the years are more likely to be diagnosed with a chronic condition at a
of 2012 and 2030, it is projected that the non-Hispanic African younger age and be more physically disabled than Whites (3).
American population aged 65 years and above will increase 104% Although African Americans are at a greater risk for a chronic
in comparison to the 54% projected increase for non-Hispanic disease diagnosis and negative health outcomes associated with
Whites in the same age group (1). chronic conditions in comparison to Whites, African Americans

www.frontiersin.org April 2015 | Volume 2 | Article 174 | 123

Mingo et al. Chronic Disease Self-Management Education program

are underrepresented in access to chronic disease self-management significant proportion of African Americans reside in urban areas
education programs as a result of a variety of biopsychosocial and overwhelmed with violence, dilapidated housing, limited options
sociocultural factors (e.g., racism, low socioeconomic status, dis- for fresh fruits and vegetables, and in close proximity to toxic
crimination, unequal access to goods and services, lack of trust waste sites (15–17). Further, research has shown that specifically
in the health care system) (4–6). It is plausible that the context urban-dwelling aging African Americans have greater levels of dis-
in which one experiences negative health outcomes may impact ability as a result of chronic conditions in comparison to not only
their perceptions about health, health behaviors, and disease man- Whites but also aging African Americans living outside of urban
agement (7–9). However, very little attention has been given to areas (18, 19).
cultural milieu and unique experiences for African Americans that The aforementioned findings suggest the importance of
may influence utilization, participation, and even completion of extending CDSME research in an effort to understand the pat-
evidenced-based health prevention and management programs terns of chronic disease management among urban-dwelling aging
[e.g., Chronic Disease Self-Management Program (CDSMP)]. African Americans. Studying African Americans specifically in this
Considerable efforts have been made nationally to address context is a unique contribution to the literature in that African
issues surrounding the management of chronic illnesses among Americans embody cultural similarities but also intragroup dif-
underserved and disadvantaged individuals (10–13). In fact, pub- ferences. It is important for researchers, health care providers, and
lic health efforts have focused on reducing the burden of chronic policy makers to recognize the vast array of characteristics that
disease for all and minimizing health disparities in chronic disease may be represented in aging African Americans in order to pro-
among racial/ethnic minorities by redirecting society’s attention vide services that are appealing, feasible, and that lead to successful
to the benefits of evidence-based health prevention and manage- completion of effective behavioral health programs.
ment programs. Notably, there has been a concerted effort by Therefore, the objective of this study is to (1) determine the
the US Administration on Aging, Centers for Disease Control characteristics of urban-dwelling African Americans with at least
and Prevention (CDC), and Centers for Medicare and Medicaid one chronic condition who participated in a CDSME program,
Services (CMS) to disseminate and implement Chronic Disease and (2) determine factors associated with successfully completing
Self-Management Education (CDSME) programs nationally (14). the program.
As a result, a national study of the CDSMP was conducted
resulting in greater racial/ethnic diversity among participants than MATERIALS AND METHODS
had been seen in previous research focused on self-management DATA SOURCE AND STUDY POPULATION
(10, 11, 13). Specifically, the national study included approxi- Cross-sectional data for this study was obtained from a nationwide
mately 45% of participants who self-identified as African Amer- delivery of CDSME programs as part of the American Recovery
ican, Latino, or other minority groups (11). Although substan- and Reinvestment Act of 2009 (i.e., Recovery Act) Communities
tial efforts to widely disseminate and implement the program Putting Prevention to Work: Chronic Disease Self-Management Pro-
have been made, there is still a great deal of information to be gram initiative (20). The US Administration on Aging led this
learned about how individuals living with multiple chronic condi- initiative in collaboration with CDC and CMS to support the
tions (MCC) successfully manage their conditions by participating translation of CDSME programs in 45 states, Puerto Rico, and the
in and completing evidenced-based self-management programs District of Columbia (14). This initiative was originally designed
(e.g., CDSMP) and how this may vary by race/ethnicity. to have 50,000 Americans complete CDSME program sessions
While much has been written about national dissemination between 2010 and 2012 and to embed CDSME program delivery
and implementation of the CDSME programs (10–12) as well as structures into statewide systems (20). Based on the aforemen-
the benefits of completing a CDSME program (13), there is lit- tioned unique challenges facing the target population, the focus of
tle empirical data that examine the effectiveness of the program this study is on the 11,895 urban-dwelling African Americans who
or the unique predictors of participation and successful comple- attended the program at one of the five leading delivery sites (i.e.,
tion among African Americans with chronic conditions. Moreover, senior center, health care organization, residential facility, com-
few studies have reported on completers (i.e., individuals who munity location, faith-based organization) and reported having at
complete four out of six sessions) and non-completers (i.e., indi- least one chronic condition. Specifically, 2,170 unique workshops
viduals who stop attending the program prior to completing four were delivered to African American participants in this study. Of
sessions) within this context. Thus, the fact that African Ameri- these workshops, 754 were delivered in senior centers/AAA, 518
cans are already disproportionately affected by chronic conditions in healthcare organizations, 479 in residential facilities, 242 in
is made worse by the relative gap in what is known about the community/multi-purpose facilities, and 177 in faith-based orga-
management of chronic diseases among this target population. nizations. Participants attending other sites were small in sample
In addition to African Americans facing unique challenges size and omitted from the analyses in this study. Institutional
(e.g., racism, discrimination, cultural mistrust) that may influence Review Board approval for this study was given by Texas A&M
health behaviors, health status, and utilization of health care (4–6), University.
African Americans living in urban areas may add another layer of
complexity that is not well addressed or understood in the chronic PROGRAM DESCRIPTION
disease self-management literature. African Americans who reside The CDSMP, a program included within a larger suite of Stanford’s
in urban areas often experience unique health-related vulnerabil- CDSME programs, has been introduced and widely disseminated
ities in comparison to Whites and other minority racial/groups. A in the US as a method to empower patients with self-management

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 174 | 124
Mingo et al. Chronic Disease Self-Management Education program

skills to deal with their chronic conditions (21). Drawing upon Dependent variable
Social Learning Theory (22), CDSME programs are evidence- Successful completion. Completion of the CDSME program was
based, peer-led interventions consisting of six highly participative the dependent variable of interest in this study. Attending at least
classes held for 2.5 hours each, once a week, for six consecutive four of the six classes was considered successful completion of
weeks (21). During the tenure of these programs, participants the program. This criterion is consistent with criterion used in
receive a copy of the book, Living a Healthy Life with Chronic previous work focused on the CDSMP (11, 12, 20).
Conditions, Fourth Edition (23) as well as an audio relaxation
CD titled Relaxation for Mind and Body (24). In addition, con-
STATISTICAL ANALYSIS
tent for the workshops focus on teaching individuals ways to
First, frequencies were examined to assess demographics, health
deal with frustration, isolation, pain, and fatigue. Furthermore,
status, and utilization by delivery site type among the total sample
participants are taught how to develop an action plan to meet
of urban-dwelling African Americans with chronic disease. Next,
intended goals; how to develop an individualized exercise pro-
independent samples t -test, one-way ANOVA, and chi-square
gram; how to appropriately use medications; how to solve chronic
analyses were conducted to examine differences in sample charac-
disease-related problems; and how to communicate with family,
teristics by number of chronic conditions (i.e., one, two, or three or
friends, and health care providers. In particular, the CDSMP has
more chronic conditions) and by CDSME programs completion
resulted in improved health care (e.g., exercise, communication
status (i.e., non-successful completion, successful completion).
with physician) and health (e.g., pain, self-reported health, fatigue,
Specifically, independent samples t -test analyses and one-way
disability, depression) (11–13), while potentially saving health care
ANOVA were used for continuous variables and chi-square analy-
costs (25).
ses were used for categorical variables. Subsequently, a multiple
logistic regression model was employed to determine the asso-
MEASURES ciation between the independent variables (i.e., age, sex, living
Independent variables situation, number of chronic conditions, delivery site, poverty
In an effort to identify characteristics associated with participat- level) and successful completion of CDSME programs.
ing and successfully completing the CDSME program that are
specific to urban-dwelling African Americans with chronic condi-
tions, various measures (i.e., demographics, health status, delivery RESULTS
site types) were included. Descriptive statistics for the sample of urban-dwelling African
Americans (N = 11,895) are displayed in Table 1. Descriptive
statistics are presented for the total sample and stratified by the
Demographics. Participants were asked to self-report age (i.e.,
number of self-reported chronic conditions and CDSME program
in years), sex (i.e., male or female), and living arrangement (i.e.,
completion status. On average, participants were 68 (±12.06)
living alone, living with others). In addition, residential ZIP Codes
years of age. The majority of the participants were female (83%)
provided by participants were used to determine percentage of res-
and lived with someone (92%). On average, participants reported
idents within the ZIP Code that fell below a 200% federal poverty
having 2.5 (±1.41) chronic conditions with approximately 44%
level, and the percent of African American residents residing
reporting 3 or more. Arthritis (50%), diabetes (48%), and hyper-
within the participants’ ZIP Code.
tension (64%) were the three most commonly reported chronic
conditions among the total sample. The largest percentage of par-
Health status. Health status was determined using the partic- ticipants attended the CDSME program at a senior center (37%)
ipants’ self-reported chronic conditions. Participants were pre- or residential facility (24%). Thirty-five percent of our sample
sented with a list of chronic conditions (i.e., arthritis, cancer, resided in an impoverished area (i.e., 200% below federal poverty
depression, diabetes, heart disease, hypertension, lung disease, level).
stroke, osteoporosis, other) and asked to indicate whether or not When examining differences in characteristics across number
they had been diagnosed with each condition by a health care of chronic conditions (i.e., one chronic condition, two chronic
provider. The number of self-reported chronic conditions were conditions, three or more chronic conditions), a statistically sig-
summed to create one count variable, and the prevalence of each nificant difference was found among the three groups on age, sex,
individual disease was calculated based on participants’self-report. living situation, disease prevalence, poverty level, and delivery site
type. Notably, larger proportions of participants reporting three
Delivery site type. Delivery site information was collected or more chronic conditions were ages 65–79 years (53%), female
through administrative procedures. For the purposes of this study, (86%), and living with someone (98%). As expected, larger pro-
analyses focused on the five leading delivery sites (i.e., senior center, portions of participants who reported having any of the chronic
health care organization, residential facility, community location, conditions ultimately had more comorbidities. A larger propor-
faith-based organization). Participants attending CDSME pro- tion of participants who attended workshops at senior centers
grams at any other delivery site types (e.g., workplaces, educational and faith-based organizations (40%, 16%; respectively) reported
institutions, tribal centers) were omitted because of inadequate only one chronic condition. Conversely, a larger proportion of
case sizes. Reports of participants attending the CDSME program participants who attended a workshop at a health care organi-
at any delivery sites or delivery sites labeled as “other” were omitted zation or community center (17%, 13%; respectively) reported
from analyses due to the complexities of interpretation. two chronic conditions. Lastly, a larger proportion of participants

www.frontiersin.org April 2015 | Volume 2 | Article 174 | 125

Frontiers in Public Health | Public Health Education and Promotion

Mingo et al.
Table 1 | Sample characteristics by number of chronic conditions and CDSME program completion.

Total 1 Chronic 2 Chronic 3+ Chronic X 2 or f p Non-successful Successful X 2 or t p

(n = 11,895) condition conditions conditions completion completion
(n = 3,331) (n = 3,294) (n = 5,270) (n = 2,445) (n = 9,450)

Age (years) 68.46 (±12.06) 66.95 (±13.70) 68.46 (±12.21) 69.41 (±10.69) 42.57 <0.001 67.21 (±12.69) 68.78 (11.87) −5.53 <0.001
Under 50 6.6% 10.8% 6.9% 3.7% 180.53 <0.001 8.4% 6.1% 35.89 <0.001
50–64 26.5% 25.5% 27.2% 26.8% 29.4% 25.8%
65–79 49.7% 46.6% 47.8% 53.0% 46.0% 50.7%
80+ 17.1% 17.2% 18.1% 16.5% 16.2% 17.4%
Sex
Male 16.8% 22.1% 16.0% 13.9% 100.52 <0.001 19.6% 16.1% 16.69 <0.001
Female 83.2% 77.9% 84.0% 86.1% 80.4% 83.9%
Lives alone
No 92.3% 79.3% 97.3% 97.5% 1115.18 <0.001 96.4% 91.3% 70.28 <0.001
Yes 7.7% 20.7% 2.7% 2.5% 3.6% 8.7%

Number of chronic conditions 2.53 (±1.41) – – – – – 2.64 (±1.47) 2.51 (±1.40) 3.93 <0.001
1 28.0% – – – – – 25.8% 28.6% 8.42 0.015
2 27.7% – – – 27.8% 27.7%
3+ 44.3% – – – 46.4% 43.8%

Disease prevalence
Arthritis 50.3% 14.9% 44.7% 76.3% 3139.78 <0.001 52.1% 49.9% 3.98 0.046
Cancer 7.4% 2.0% 4.9% 12.4% 369.93 <0.001 7.9% 7.3% 1.21 0.271
Depression 14.4% 3.1% 8.2% 25.3% 960.76 <0.001 18.1% 13.4% 35.38 <0.001
Diabetes 47.9% 34.8% 38.4% 62.1% 777.04 <0.001 44.2% 48.9% 16.87 <0.001
Heart disease 17.5% 2.3% 9.5% 32.2% 1465.08 <0.001 18.5% 17.3% 2.05 0.152
Hypertension 64.1% 26.3% 66.2% 86.6% 3234.57 <0.001 65.1% 63.8% 1.37 0.242
Lung disease 18.9% 3.9% 8.1% 35.2% 1651.38 <0.001 21.7% 18.2% 15.30 <0.001

Chronic Disease Self-Management Education program

Stroke 7.7% 1.3% 4.3% 13.9% 530.15 <0.001 8.6% 7.5% 3.67 0.055
Osteoporosis 10.0% 1.6% 5.3% 18.1% 729.44 <0.001 10.4% 9.8% 0.78 0.378
Other 15.0% 9.9% 10.4% 21.1% 276.60 <0.001 16.8% 14.6% 7.27 0.007

Percent of ZIP Code under 200% poverty level 34.59 (±8.22) 35.14 (±8.02) 34.33 (±8.21) 34.40 (±8.34) 10.73 <0.001 34.72 (±8.15) 34.55 (±8.24) 0.89 0.375
April 2015 | Volume 2 | Article 174 | 126

Percent of African American residents in ZIP Code 47.48 (±29.20) 47.80 (±29.49) 47.81 (±29.22) 47.06 (±29.01) 0.96 0.381 48.67 (±27.70) 47.17 (±29.07) 2.24 0.025

Delivery site type

Senior Center/AAA 36.5% 40.0% 34.5% 35.6% 178.72 <0.001 31.2% 37.9% 82.70% <0.001
Healthcare organization 15.5% 13.7% 16.6% 15.8% 18.0% 14.8%
Residential facility 24.1% 18.6% 22.7% 28.4% 29.4% 22.7%
Community/multipurpose-facility/library 11.2% 11.6% 12.5% 10.0% 10.6% 11.3%
Faith-based organization 12.8% 16.0% 13.7% 10.2% 10.8% 13.3%
Mingo et al. Chronic Disease Self-Management Education program

who attended the CDSME program at a residential facility (28%) (OR = 1.24, p < 0.001) and those that live alone (OR = 2.38,
reported three or more chronic conditions. p < 0.001). Number of chronic conditions was not significantly
Table 1 also reports differences in completion status across all associated with successful completion. Compared to senior cen-
independent variables. Significant differences in non-successful ters, attending the CDSME program at a residential facility or a
completion/successful completion were found across age, sex, community center decreased the odds of successfully complet-
living situation, number of chronic conditions, prevalence of ing the program (OR = 0.70, p < 0.001; OR = 0.61, p < 0.001). In
disease by disease type, and type of delivery site. A larger pro- addition, living in an impoverished neighborhood reduced the
portion of participants who successfully completed the CDSME odds of successfully completing the CDSME program (OR = 0.99,
program were older (p < 0.001), female (p < 0.001), report liv- p < 0.024).
ing alone (p < 0.001), and report fewer chronic conditions
(p < 0.001). A significantly smaller proportion of successful com- DISCUSSION
pleters reported arthritis (p = 0.05), depression (p < 0.001), lung IDENTIFICATION OF SIMILARITIES AND DIFFERENCES
disease (p < 0.001), whereas a larger proportion of successful An initial research objective was to identify the characteristics of
completers reported having diabetes (p < 0.001). There were no African American participants and the program delivery infra-
significant differences found in cancer, heart disease, hyperten- structure that served them. In many regards, the urban-dwelling
sion, stroke, and osteoporosis disease prevalence between suc- African American participants with chronic disease had profiles
cessful and non-successful completers. Lastly, significant dif- similar to participants seen in earlier CDSMP studies (11, 20, 26,
ferences were found across delivery sites between those who 27). For example, participants were older, predominantly female,
successfully completed CDSME programs and those who did reported their most prevalent chronic conditions as hyperten-
not. A larger proportion of participants who successfully com- sion, arthritis, and diabetes, and were more likely to partici-
pleted CDSME programs attended senior centers, whereas a pate in the program at senior centers, residential facilities and
smaller proportion of successful completers attended CDSME health care organizations. Notably, in our sample participating
programs at other delivery sites (i.e., health care organization, in CDSME programs at a residential facility was more preva-
residential facility, community center, faith-based organization) lent than participating at a health care organization; the order
(p < 0.001). of prevalence was reverse in the overall national level sample
Table 2 provides results of the logistic regression analyses. Com- (20, 27).
pared to participants below the age of 50 years, being between Although the participants in our study were demographically
the ages of 50–64 and 65–79 years decreased the odds of success- similar in many ways to those in previous research describing
fully completing the CDSME program [odds ratio (OR) = 0.71, populations and delivery characteristics, this study clearly demon-
p = 0.001 and OR = 0.85, p = 0.025; respectively). However, the strates unique characteristics that are worthy of highlighting, and
odds of successfully completing the program increased for females important to consider in policy, practice, and future research.

Table 2 | Factors associated with successful completion.

95% CI

OR P Lower Upper

Age: under 50 years 1.00 – – –

Age: 50–64 years 0.71 0.001 0.578 0.863
Age: 65–79 years 0.85 0.025 0.738 0.980
Age 80+ years 1.02 0.763 0.897 1.160
Male 1.00 – – –
Female 1.24 <0.001 1.106 1.397
Live alone: no 1.00 – – –
Live alone yes 2.38 <0.001 1.892 3.007
Number of chronic conditions: 1 1.00 – – –
Number of chronic conditions: 2 1.04 0.454 0.932 1.172
Number of chronic conditions: 3+ 1.05 0.365 0.944 1.171
Senior Center/AAA 1.00 – – –
Healthcare organization 0.92 0.280 0.785 1.073
Residential facility 0.70 <0.001 0.590 0.833
Community/multipurpose facility/library 0.61 <0.001 0.523 0.720
Faith-based organization 0.87 0.155 0.719 1.054
Percent of ZIP Code under 200% poverty level 0.99 0.024 0.988 0.999

OR, odds ratio; 95% CI, 95% confidence interval; attending fewer than four workshop sessions, referent.

www.frontiersin.org April 2015 | Volume 2 | Article 174 | 127

Mingo et al. Chronic Disease Self-Management Education program

Particularly, the burden of chronic disease among urban-dwelling status, it is important to note the implications for the health
African Americans cannot be ignored. While the ranking of of African American males. Unfortunately, the health of African
chronic condition prevalence rates for specific disease type was American males has been likened to that of individuals living in
similar to CDSME program participants in other studies (20), the developing countries (33). African American males fare worse than
actual prevalence rates for the individual chronic conditions were other segments of the population (e.g., African American females,
substantially higher. National prevalence rates indicate that irre- Whites, and other racial/ethnic minority groups) on almost every
spective of racial/ethnic group 21% of US community-dwelling chronic condition (33). While there is a dismal focus on African
adults report having three chronic conditions and approximately American men in chronic disease self-management research (13),
5% report four or more (28). Overall, 44% of our sample reported our findings suggest that it is paramount that emphasis is put on
having three or more chronic conditions. Moreover, our research empowering urban-dwelling African American men to engage in
findings indicated that participants with different comorbidity health promotion that would ultimately lead to positive health
levels (i.e., one, two, three or more) participated in the CDSME outcomes and providing other health-related benefits (e.g., symp-
program at different delivery site types. Specifically, larger propor- tom management, reduction of health care cost, reduction in
tions of participants who attended the CDSME at senior centers emergency room visits, reduction in work disability) at both the
and faith-based organizations reported only having one chronic individual and societal level.
condition, whereas a larger proportion of those participating at
a residential facility reported three or more chronic conditions. ASSOCIATIONS WITH CHRONIC CONDITIONS
This is not overwhelmingly surprising in that residential facili- When examining health status in this study, individuals who com-
ties may be servicing individuals living within their community. pleted the CDSME program had fewer total number of chronic
Oftentimes, decisions to move into residential facilities particu- conditions and reported lower prevalence rates of arthritis, depres-
larly for aging individuals is predicated on disabilities and health sion, and lung disease. Interestingly, successful completers in com-
challenges associated with common chronic conditions (29). parison to non-completers reported higher rates of diabetes and no
significant differences for the other conditions. Moreover, the total
ASSOCIATIONS WITH PROGRAM COMPLETENESS number of chronic conditions was not associated with successful
In addition to examining differences across chronic conditions, completion. Erdem and Korda (26) reported similar findings in a
our study also closely examined differences in characteristics for study that examined characteristics of participants with diabetes
successful completers and non-completers of the program as well who completed both the CDSMP and Diabetes Self-Management
as identified factors that were associated with successful comple- Program (DSMP). Future research is warranted to further examine
tion. Results indicated that successful completers of the program the impact of MCC on participation and completion of behavioral
were older than those who did not complete the program. How- health interventions.
ever, the regression analyses showed that compared to participants
below the age of 50 years, being between the ages of 50–64 and 65– ASSOCIATIONS WITH ENVIRONMENTAL CHARACTERISTICS
69 years decreased the odds of successfully completing CDSME While there were no significant differences in successful com-
programs. pleters and non-completers based on neighborhood level poverty,
Considering that African Americans are diagnosed with chronic living in an impoverished community was associated with a
conditions much earlier in life in comparison to other groups (6), lower likelihood of successfully completing the program. Research
it may be that those between the age of 50–64 and 65–79 years have has consistently documented the association between socioeco-
been dealing with the condition(s) for a longer period of time and nomic factors and health, particularly highlighting the vulner-
either feel that they have learned how to successfully manage their abilities of living in poverty (34). It is plausible that living in
health or have a perception that nothing can really be done for areas of concentrated poverty introduces additional barriers that
their condition. For example, in a study examining perceptions may interfere with not only participating but completing health
of arthritis, the authors found that individuals were less likely to programs that are beneficial. As previously stated, a significant
believe a person diagnosed with arthritis could improve their con- proportion of urban-dwelling African Americans live in impover-
dition with better health care (30). Perceptions of this type would ished areas that are associated with violence, poor housing, and
make one more vulnerable to not completing the program. Studies limited access and availability to options that promote health
have shown that the perception of chronic conditions and symp- (15–17).
toms associated with chronic conditions influences health care To our knowledge, no research has closely examined the asso-
decisions among African Americans (30, 31). In addition to age, a ciation of living in an impoverished community with completing
larger proportion of females and those living alone were successful the CDSME program among urban-dwelling African Americans.
completers. Moreover, being female and living alone increased the Findings not only suggest the need for future research in this area
odds of successfully completing the CDSME program. Our find- but also provide preliminary results that should be considered
ings on gender differences are consistent with findings in previous when implementing and disseminating the CDSME program in
research (26, 32). certain communities. Providing additional support that would
foster increased completion rates of the CDSME program to res-
IMPLICATIONS FOR REACHING MALES idents of this community could lead to positive health outcomes
Due to the limited number of African American male participants for a population that is at risk for continued health and health care
and the findings highlighting gender differences in completion disparities.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 174 | 128
Mingo et al. Chronic Disease Self-Management Education program

ASSOCIATIONS WITH DELIVERY SITE how participants were recruited to participate in the program
Finally, our analyses indicated that delivery site type is an impor- is unavailable. It may not just be the differences in the types
tant factor for urban-dwelling African Americans when consider- of delivery sites that result in significant differences in success-
ing program completion. Senior centers had the largest proportion ful completion of the CDSME program, but also the methods
of completers whereas faith-based organizations and community that delivery sites use to initially recruit participants in the pro-
centers had the smallest proportion of completers. However, it gram. While lay leaders are trained to offer the program in a
is important to note among African Americans only that faith- standardized manner, another limitation to consider would be
based organizations and community centers had the smallest unaccounted variance in program implementation that may or
proportion of program participants. When examining the asso- may not impact participants to remain in the program. Lastly, lim-
ciation between delivery sites and successful completion, findings ited information is available about the lay leaders who served as
indicated that the odds of successfully completing the program instructors for the workshops. Therefore, an additional limitation
were significantly decreased if one attended the CDSME pro- of the study is the inability to examine the impact of instructor-
gram at a residential facility or community center compared to level factors on completion rates (e.g., race, gender, age, health
attending the program at a senior center. Interestingly, there was status).
no significant relationship between successful completion and
faith-based organizations or health care organizations. This is CONCLUSION
in contrast to another study including participants from mul- The American Recovery and Reinvestment Act Communities Putting
tiple racial/ethnic groups that found that completion rates for Prevention to Work: Chronic Disease Self-Management Program
the CDSME program are lowest at residential facilities but high- (14) has made a large public impact on the ongoing national
est at faith-based organizations (26). Other studies focused on dissemination and implementation of CDSME programs. The
delivery preferences of a self-management program among aging present study yields findings that can contribute to the ongo-
individuals who lived in an urban area have found that, in com- ing research, practice, and policy efforts associated with this
parison to Whites, African Americans were more likely to prefer initiative. Particularly, our findings indicate that for vulnera-
a self-management program that would be delivered at a local ble populations such as urban-dwelling African Americans, the
church or a health care organization (35). Notably, implement- influence of the individual, social, and environmental context in
ing programs within faith-based organizations have shown to which one may experience CDSME programs must be consid-
be effective in increasing utilization of health promotion pro- ered. Strategies to encourage employment of CDSME programs
grams for African Americans (36–41). Therefore, the finding across all available delivery site types can foster participation
that faith-based and health care organizations have no associa- and completion. Working closely with health care providers and
tion with successful completion may be more about reach and community gatekeepers to inform individuals about the avail-
less about preference for participating or completing at that ability and benefits of completing CDSME programs is one
site type. method for moving forward. In addition, considerations for pro-
Therefore, one promising strategy for increasing the reach of gram modifications that would still yield similar outcomes, but
CDSME programs may be to work more closely with faith-based foster greater levels of completion should be discussed. Also,
organizations and health care organizations in urban African it is likely that putting policies in place that allow for allo-
American communities. Embedding the program in existing infra- cation of resources that would provide support to individu-
structures (e.g., church health ministry, senior ministry) may yield als in impoverished communities may also yield positive out-
greater participation and completion of the program. Findings comes. In summary, our study indicates that in order to increase
indicating that the odds of successfully completing the program are reach and positively impact, a diverse population, practice, pol-
lowered for those participating at a residential facility or at a com- icy, and research strategies must consider the cultural milieu
munity center may be a result of a number of barriers. For example, for African Americans that ultimately influence chronic disease
residential facility participants may have a number of health com- self-management.
plications resulting in the inability to successfully complete the
program. Individuals participating in CDSME programs at a local ACKNOWLEDGMENTS
community center may have other family responsibilities or trans- Funding: The American Recovery and Reinvestment Act of 2009
portation issues that could serve as a barrier to participating in a (i.e., Recovery Act) Communities Putting Prevention to Work:
program for six consecutive weeks. Chronic Disease Self-Management Program initiative, led by the
U.S. Administration on Aging in collaboration with the Centers for
STUDY LIMITATIONS Disease Control and Prevention and the Centers for Medicare and
Given that many of the measures were collected using self-reports, Medicaid Services, allotted $32.5 million to support the translation
it is possible that participants may over- or underreport health sta- of the Stanford program in 45 States, Puerto Rico, and the District
tus variables. However, self-report of chronic conditions is used of Columbia. The National Council on Aging served as the Tech-
as a valid measure in studies examining health in aging indi- nical Assistance Resource Center for this initiative and collected
viduals (42). In addition, the cross-sectional nature of the data de-identified data on program participation. This study was sup-
does not allow for us to determine a causal relationship between ported by a grant from the National Institutes of Health, 5P30
the independent variables and outcome variables. Although the AG015281, and the Michigan Center for Urban African American
delivery site type is available, lack of information concerning Aging Research.

www.frontiersin.org April 2015 | Volume 2 | Article 174 | 129

Mingo et al. Chronic Disease Self-Management Education program

REFERENCES 22. Bandura A. Social cognitive theory of self-regulation. Organ Behav Hum Decis
1. Administration on Aging. A Profile of Older Americans. (2013). p. 1–15. Available Process (1991) 50(2):248–87. doi:10.1016/0749-5978(91)90022-L
from: http://www.aoa.gov/Aging_Statistics/Profile/2013/docs/2013_Profile.pdf 23. Lorig K, Holman H, Sobel D, Laurent D, Gonzalez V, Minor M. Living a
2. Fried VM, Bernstein AB, Bush MA. Multiple Chronic Conditions among Adults Healthy Life with Chronic Conditions: Self-Management of Heart Disease,
Aged 45 and over: Trends over the Past 10 Years. (2013). Available from: Arthritis, Diabetes, Depression, Asthma, Bronchitis, Emphysema and Other Phys-
http://www.cdc.gov/nchs/data/databriefs/db100.htm ical and Mental Health Conditions. Boulder, CO: Bull Publishing Company
3. Centers for Disease Control and Prevention. Minority Health: Black or (2012).
African American Populations. (2013). Available from: http://www.cdc.gov/ 24. Regan C, Seidel R. Relaxation for Mind and Body: Pathways to Healing [CD].
minorityhealth/populations/REMP/black.html Boulder, CO: Bull Publishing Company (2012).
4. Becker G, Gates RJ, Newsom E. Self-care among chronically ill African Ameri- 25. Ahn S, Basu R, Smith ML, Jiang L, Lorig K, Whitelaw N, et al. The impact
cans: culture, health disparities, and health insurance status. Am J Public Health. of chronic disease self-management programs: healthcare savings through a
(2004) 94(12):2066–73. community-based intervention. BMC Public Health (2013) 13(1):1141. doi:10.
5. Dell JL, Whitman S. A history of the movement to address health disparities. 1186/1471-2458-13-1141
In: Whitman S, Shah AM, Benjamins MR, editors. Urban Health: Combating 26. Erdem E, Korda H. Self-management program participation by older adults
Disparities with Local Data. New York: Oxford University Press (2011). p. 8–30. with diabetes: chronic disease self-management program and diabetes self-
6. Rooks RN, Whitfield KE. Health disparities among older African Americans: management program. Fam Community Health (2014) 37(2):134–46. doi:10.
past, present, and future perspectives. In: Whitfield KE, editor. Closing the Gap: 1097/FCH.0000000000000025
Improving the Health of Minority Elders in the New Millennium. Washington, 27. Korda H, Erdem E, Woodcock C, Kloc M, Pedersen S, Jenkins S. Racial and ethnic
DC: Gerontological Society of America (2004). p. 45–54. minority participants in chronic disease self-management programs: findings
7. McLeroy KR, Bibeau D, Steckler A, Glanz K. An ecological perspective on from the communities putting prevention to work initiative. Ethn Dis (2013)
health promotion programs. Health Educ Q (1988) 15(4):351–77. doi:10.1177/ 23(4):508–17.
109019818801500401 28. Ward BW, Schiller JS. Prevalence of multiple chronic conditions among US
8. Solar O, Irwin A. A conceptual framework for action on the social adults: estimates from the national health interview survey, 2010. Prev Chronic
determinants of health. Social Determinants of Health Discussion. Paper Dis (2013) 10:1–15. doi:10.5888/pcd10.120203
2. (2010). Available from: http://www.who.int/sdhconference/resources/ 29. Ball MM, Perkins MM, Hollingsworth C, Whittington FJ, King SV. Pathways
ConceptualframeworkforactiononSDH_eng.pdf to assisted living: the influence of race and class. J Appl Gerontol (2009)
9. Probst JC, Moore CG, Glover SH, Samuels ME. Person and place: the com- 28(1):81–108. doi:10.1177/0733464808323451
pounding effects of race/ethnicity and rurality on health. Am J Public Health. 30. Mingo CA, McIlvane JM, Haley WE. Impact of the diagnostic label of
(2004) 94(10):1695–703. osteoarthritis on perceptions of disability and willingness to help a par-
10. Smith ML, Belza B, Altpeter M, Ahn S, Dickerson JB, Ory M. Disseminating an ent in African Americans and whites. Arthritis Care Res (2006) 55(6):913–9.
evidence-based disease self-management program for older Americans: impli- doi:10.1002/art.22352
cations for diversifying participant reach through delivery site adoption. In: 31. Mingo CA, McIlvane JM, Haley WE, Luong M-LN. Impact of race and diag-
Maddock J, editor. Public Health: Social and behavioral health. Rijeka: InTech nostic label on older adults’ emotions, illness beliefs, and willingness to help
(2012). p. 385–404. a family member with osteoarthritis. J Appl Gerontol (2013). doi:10.1177/
11. Ory MG, Ahn S, Jiang L, Lorig K, Ritter P, Laurent DD, et al. National study of 0733464812471893
chronic disease self-management six-month outcome findings. J Aging Health 32. Hawkins J, Kieffer EC, Sinco B, Spencer M, Anderson M, Rosland A-M.
(2013) 25(7):1258–74. doi:10.1177/0898264313502531 Does gender influence participation? Predictors of participation in a com-
12. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, Whitelaw N, et al. Successes munity health worker diabetes management intervention with African Amer-
of a national study of the chronic disease self-management program: meet- ican and Latino adults. Diabetes Educ (2013) 39(5):647–54. doi:10.1177/
ing the triple aim of health care reform. Med Care (2013) 51(11):992–8. 0145721713492569
doi:10.1097/MLR.0b013e3182a95dd1 33. Gadson SL. The third world health status of black American males. J Natl Med
13. Brady TJ, Murphy L, O’Colmain BJ, Beauchesne D, Daniels B, Greenberg M. Assoc (2006) 98(4):488–91.
A meta-analysis of health status, health behaviors, and health care utilization 34. World Health Organization. Social Determinants of Health: The Solid Facts.
outcomes of the chronic disease self-management program. Prev Chronic Dis (2003). Available from: http://www.euro.who.int/__data/assets/pdf_file/0005/
(2013) 10:E07. doi:10.5888/pcd10.120112 98438/e81384.pdf
14. US Department of Health and Human Services. American Recovery and 35. Mingo CA, McIlvane JM, Jefferson M, Edwards LJ, Haley WE. Preferences
Reinvestment Act Communities Putting Prevention to Work: Chronic Disease for arthritis interventions: identifying similarities and differences among
Self-Management Program. (2012). Available from: http://www.aoa.gov/AoA_ African Americans and whites with osteoarthritis. Arthritis Care Res (2013)
programs/HPW/ARRA/index.aspx 65(2):203–11.
15. Cooper RS. Health and the social status of blacks in the United States. Ann 36. Campbell MK, Demark-Wahnefried W, Symons M, Kalsbeek WD, Dodds J,
Epidemiol (1993) 3(2):137–44. doi:10.1016/1047-2797(93)90126-O Cowan A, et al. Fruit and vegetable consumption and prevention of cancer:
16. Johnson BL, Coulberson SL. Environmental epidemiologic issues and minority the black churches united for better health project. Am J Public Health (1999)
health. Ann Epidemiol (1993) 3(2):175–80. doi:10.1016/1047-2797(93)90133-O 89(9):1390–6. doi:10.2105/AJPH.89.9.1390
17. Zenk SN, Schulz AJ, Hollis-Neely T, Campbell RT, Holmes N, Watkins G, et al. 37. Campbell MK, Hudson MA, Resnicow K, Blakeney N, Paxton A, Baskin M.
Fruit and vegetable intake in African Americans income and store characteris- Church-based health promotion interventions: evidence and lessons learned.
tics. Am J Prev Med (2005) 29(1):1–9. doi:10.1016/j.amepre.2005.03.002 Annu Rev Public Health (2007) 28:213–34. doi:10.1146/annurev.publhealth.28.
18. Miller DK, Carter ME, Miller JP, Fornoff JE, Bentley JA, Boyd SD, et al. Inner-city 021406.144016
older blacks have high levels of functional disability. J Am Geriatr Soc (1996) 38. Irvin TM, Sutherland MS, Harris GJ. A faith-based arthritis self-help program
44(10):1166–73. for rural African Americans. Am J Health Stud (2006) 21(2):107–14.
19. Miller DK, Wolinsky FD, Malmstrom TK, Andresen EM, Miller JP. Inner 39. Darnell JS, Chang C-H, Calhoun EA. Knowledge about breast cancer and partici-
city, middle-aged African Americans have excess frank and subclinical pation in a faith-based breast cancer program and other predictors of mammog-
disability. J Gerontol A Biol Sci Med Sci (2005) 60(2):207–12. doi:10.1093/gerona/ raphy screening among African American women and Latinas. Health Promot
60.2.207 Pract (2006) 7(3 Suppl):201S–12S. doi:10.1177/1524839906288693
20. Ory MG, Smith ML, Patton K, Lorig K, Zenker W, Whitelaw N. Self-management 40. Kerner JF, Dusenbury L, Mandelblatt JS. Poverty and cultural diversity: chal-
at the tipping point: reaching 100,000 Americans with evidence-based programs. lenges for health promotion among the medically underserved. Annu Rev Public
J Am Geriatr Soc (2013) 61(5):821–3. doi:10.1111/jgs.12239 Health (1993) 14:355–77. doi:10.1146/annurev.pu.14.050193.002035
21. Lorig K, Holman H, Sobel D, Laurent D. Living a Healthy Life with Chronic Con- 41. Kotecki CN. Developing a health promotion program for faith-based commu-
ditions: Self Management of Heart Disease, Arthritis, Diabetes, Asthma, Bronchitis, nities. Holist Nurs Pract (2002) 16(3):61–9. doi:10.1097/00004650-200204000-
Emphysema, and Others. 3rd ed. Boulder, CO: Bull Publishing Company (2006). 00011

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 174 | 130
Mingo et al. Chronic Disease Self-Management Education program

42. Simpson CF, Boyd CM, Carlson MC, Griswold ME, Guralnik JM, Fried LP. many of the authors and/or Review Editors may have worked together previously in
Agreement between self-report of disease diagnoses and medical record valida- some fashion. Review Editors were purposively selected based on their expertise with
tion in disabled older women: factors that modify agreement. J Am Geriatr Soc evaluation and/or evidence-based programming for older adults. Review Editors were
(2004) 52(1):123–7. doi:10.1111/j.1532-5415.2004.52021.x independent of named authors on any given article published in this volume.

Received: 17 July 2014; accepted: 18 September 2014; published online: 27 April 2015.
Conflict of Interest Statement: The authors declare that the research was conducted Citation: Mingo CA, Smith ML, Ahn S, Jiang L, Cho J, Towne SD Jr. and Ory MG
in the absence of any commercial or financial relationships that could be construed (2015) Chronic Disease Self-Management Education (CDSME) program delivery and
as a potential conflict of interest. attendance among urban-dwelling African Americans. Front. Public Health 2:174. doi:
10.3389/fpubh.2014.00174
This paper is included in the Research Topic, “Evidence-Based Programming for Older This article was submitted to Public Health Education and Promotion, a section of the
Adults.” This Research Topic received partial funding from multiple government and journal Frontiers in Public Health.
private organizations/agencies; however, the views, findings, and conclusions in these Copyright © 2015 Mingo, Smith, Ahn, Jiang , Cho, Towne and Ory. This is an open-
articles are those of the authors and do not necessarily represent the official position access article distributed under the terms of the Creative Commons Attribution License
of these organizations/agencies. All papers published in the Research Topic received (CC BY). The use, distribution or reproduction in other forums is permitted, provided
peer review from members of the Frontiers in Public Health (Public Health Education the original author(s) or licensor are credited and that the original publication in this
and Promotion section) panel of Review Editors. Because this Research Topic repre- journal is cited, in accordance with accepted academic practice. No use, distribution or
sents work closely associated with a nationwide evidence-based movement in the US, reproduction is permitted which does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 174 | 131

 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00155

Factors associated with Hispanic adults attending

Spanish-language disease self-management program
workshops and workshop completion
Matthew Lee Smith 1 *, SangNam Ahn 2,3 , Luohua Jiang 4 , Kristie P. Kulinski 5 and Marcia G. Ory 3
1
Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA, USA
2
Division of Health Systems Management and Policy, School of Public Health, The University of Memphis, Memphis, TN, USA
3
Department of Health Promotion and Community Health Sciences, Texas A&M Health Science Center, School of Public Health, College Station, TX, USA
4
Department of Epidemiology and Biostatistics, Texas A&M Health Science Center, School of Public Health, College Station, TX, USA
5
National Council on Aging, Washington, DC, USA

Edited by: Many factors influence ways in which middle-aged and older Hispanic adults prefer to
Sue Ellen Levkoff, University of South
receive health-related information. While Spanish-language disease management pro-
Carolina, USA
grams are increasingly offered in community and healthcare settings, less is known about
Reviewed by:
Marta Enid Pagan-Ortiz, Environment their utilization among the Hispanic population. This study aimed to identify participant and
and Health Group, USA workshop factors associated with middle-aged and older Hispanic adults attending Spanish-
Dharma E. Cortes, Northeastern language disease self-management program workshops and receiving the recommended
University, USA
intervention dose (i.e., successful workshop completion is defined as attending four or
*Correspondence:
more of the six workshop sessions). Data were analyzed from 12,208 Hispanic adults
Matthew Lee Smith, Department of
Health Promotion and Behavior, collected during a national dissemination of the Stanford suite of Chronic Disease Self-
College of Public Health, The Management Education (CDSME) programs spanning 45 states, the District of Columbia,
University of Georgia, 330 River and Puerto Rico. Two logistic regression analyses were performed. Over 65% of par-
Road, 315 Ramsey Center, Athens,
ticipants attended Spanish-language workshops, and 78.3% of participants successfully
GA 30602, USA
e-mail: [email protected] completed workshops. Relative to participants in English-language workshops, participants
who attended Spanish-language CDSME workshops were more likely to successfully com-
plete workshops, as were those aged 80 years and older, females, and those who lived
alone. Participants who were aged 50–79 years and female were significantly more likely
to attend Spanish-language workshops than their counterparts under age 50. Conversely,
those with more chronic conditions were less likely to attend Spanish-language workshops.
Those who attended workshops with more participants and where the Hispanic popula-
tion was less affluent were more likely to attend Spanish-language workshops. This study
provides insight into Spanish-language CDSME program recruitment and utilization with
implications for program adoption in underserved Hispanic community settings.
Keywords: chronic disease self-management, evidence-based program, Hispanic adults, intervention dose, Spanish
language

INTRODUCTION (12). Despite the growing availability of evidence-based disease

The United States is becoming increasingly more racially and eth- prevention programs for seniors (13, 14), language and/or cul-
nically diverse (1). The Hispanic population is the largest and tural barriers may prevent Hispanic individuals from accessing
fastest growing minority group in the United States (2). This pop- these services (10).
ulation is expected to represent nearly one-third of the American English or Spanish-language preferences for receiving health
population and one-fifth of the older adult population by 2050 information and materials among Hispanic individuals vary by a
(3). Growth rates are anticipated to be even higher in some parts multitude of factors (15–17), but less is known about language-
of America such as the Texas–Mexico border (4). based preferences for evidence-based programs among this pop-
The pattern of chronic disease differs among minority groups, ulation. As such, this study draws from national data to examine
and Hispanic individuals often acquire chronic conditions at participant and workshop characteristics associated with Hispanic
younger ages than their non-Hispanic white counterparts (5). individuals’ attending Spanish-language disease self-management
Additionally, as a group, Hispanic individuals are disproportion- program workshops. Further, this study examines if this partici-
ately burdened by chronic conditions including obesity, diabetes, pant subgroup received the recommended intervention dose (i.e.,
and heart disease (6–9). They are also less likely to have access to successfully completed the workshop by attending four or more of
health care (10, 11) or evidence-based health promotion programs the six workshop sessions).

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 155 | 132
Smith et al. Hispanic adults and Spanish-led workshops

MATERIALS AND METHODS Diabetes (i.e., Spanish version of the diabetes self-management
PROGRAM DESCRIPTION program), and Curso de Manejo Personal de la Artritis (i.e., Spanish
The Chronic Disease Self-Management Program (CDSMP) is version of the arthritis self-management program).
one intervention in a suite of Chronic Disease Self-Management
Education (CDSME) programs licensed through the Stanford Personal characteristics
Patient Education Research Center. CDSMP has been introduced Personal characteristics of the participants included age group (i.e.,
and widely disseminated in the U.S. as a method to empower under 50 years, 50–64 years, 65–79 years, 80+ years), sex (i.e., male,
patients with self-management skills to deal with their chronic female), living situation (i.e., lives alone, lives with others), and
conditions (18). CDSMP is an evidence-based, peer-led interven- self-reported number of chronic conditions (i.e., ranging from 0
tion consisting of six highly participative classes held for 2.5 h each, to 10). Chronic condition types included arthritis, cancer, depres-
once a week, for consecutive 6 weeks. (18) CDSMP has resulted in sion, diabetes, heart disease, hypertension, lung disease, stroke,
improved health care and health (19, 20), while potentially saving osteoporosis, and other chronic conditions.
healthcare costs (21). While some of the CDSME programs are
general (e.g., CDSMP), others are disease specific (e.g., diabetes, Delivery site types
arthritis, chronic pain). While the chronic condition may vary, Data pertaining to CDSME program delivery site types were gath-
all CDSME programs are based upon social learning theory (22), ered administratively, as described previously. Delivery site types
highly interactive, and apply the principles of goal setting, problem included healthcare organizations, senior centers or area agen-
solving, and action planning (22). cies on aging (AAAs), residential facilities, community or multi-
purpose centers (including libraries), faith-based organizations,
DATA SOURCE AND STUDY POPULATION educational institutions, and site types classified as “other” (e.g.,
Cross-sectional data for this study were obtained from a nation- correctional facilities malls, RV parks, fire departments, county
wide delivery of CDSME programs as part of the American administration buildings, private residences, casinos, career
Recovery and Reinvestment Act of 2009 (i.e., ARRA) Communi- centers).
ties Putting Prevention to Work: Chronic Disease Self-Management
Program initiative (13, 23). The U.S. Administration on Aging led Neighborhood characteristics
this initiative in collaboration with the Centers for Disease Con- Using participants’ residential ZIP Codes, geographic informa-
trol and Prevention and the Centers for Medicare and Medicaid tion system (GIS) software was used to generate neighborhood-
Services to support the translation of CDSME programs in 45 level variables for each participant. Neighborhood characteris-
states, Puerto Rico, and the District of Columbia (24). For this tics included residential rurality (i.e., metro residence or non-
study, cases were only drawn from Hispanic participants within metro residence based on the rural–urban commuting area
the first 100,000 participants enrolled in CDSME program work- codes (RUCA) (26)) and the percent of Hispanic families below
shops and who had complete data on variables of interest. Based the federal poverty line residing in the participants’ ZIP Code
on these inclusion criteria, the final analytic sample was 12,208 (27). Using organizational ZIP Codes, GIS software was used
middle-aged and older Hispanic adults who attended a CDSMP to generate neighborhood-level variables for each delivery site
workshop. (i.e., site rurality, percent of Hispanic families below the federal
poverty line).
MEASURES
Dependent variables ANALYSES
Two dependent variables were used for this study. Participants’ All statistical analyses were performed using SPSS (version 21).
attendance was recorded to determine if the recommended inter- Of the first 100,000 participants reached in this initiative, all
vention dose was received. As defined by the program developers, non-Hispanic cases (n = 86,191) were immediately omitted from
a participant has “successfully” completed the program if they analyses based on specified study aims, which left 13,809 Hispanic
attended four or more of the six offered workshop sessions (13, 19, participants. Of these Hispanic participants, those with missing
20, 25). Therefore, successful program completion was used as the data for age (n = 661), sex (n = 291), living situation (n = 8), resi-
first dependent variable in this study (i.e., non-successful comple- dential rurality (n = 973), delivery site rurality (n = 11), and class
tion served as the referent group). The second dependent variable size (n = 85) were omitted. Some participants had more than
was the workshop language in which attended. Workshops are one of these exclusionary characteristics, thus the final sample
offered in approximately 20 languages worldwide (25). Although was 12,208 middle-aged and older Hispanic adults who attended
CDSME program workshops are available in a variety of languages a CDSMP workshop. When comparing characteristics between
other than English (e.g., Mandarin Chinese, Korean, Farsi, Taga- Hispanic participants in the analytic sample with Hispanic partic-
log), the most predominant non-English workshop language is ipants omitted from analyses, participants in the analytic sample
Spanish. Therefore, participants’ enrollment in Spanish-language were significantly younger, lived with others, and had more chronic
workshops was used as the second dependent variable in this study conditions. No significant differences were observed based on
(i.e., enrollment in English-language workshops served as the ref- participants’ sex or the rurality of their residence.
erent group). Spanish-language CDSME programs offered in this For participants meeting study inclusion criteria, frequencies
nationwide rollout included Tomando Control de su Salud (i.e., were calculated for all major study variables, which were initially
Spanish version of CDSMP), Programa de Manejo Personal de la examined in relationship to participants’ successful workshop

www.frontiersin.org April 2015 | Volume 2 | Article 155 | 133

Smith et al. Hispanic adults and Spanish-led workshops

completion and the workshop language in which participants served as the referent group). Compared to participants under
attended. Pearson’s chi-square tests were performed to assess dif- age 50 years, those who were aged 50–64 years (OR = 1.68,
ferences between categorical independent variables. Independent P < 0.001) and 65–79 years (OR = 1.29, P = 0.002) were signifi-
sample t -tests were used to examine mean differences for contin- cantly more likely to attend Spanish-language workshops. Female
uous variables. Two logistic regression analyses were performed to participants were also more likely to attend Spanish-language
identify factors associated with attending Spanish-language work- workshops (OR = 1.26, P < 0.001), whereas, those with fewer
shops (i.e., attending English-language workshops served as the chronic conditions (OR = 0.85, P < 0.001) and who resided in
referent group) and successful workshop attendance (i.e., non- non-metro areas (OR = 0.28, P < 0.001) were less likely to attend
successful attendance served as the referent group). Odds ratios Spanish-language workshops. Relative to those who attended
and 95% confidence intervals are reported. workshops in healthcare organizations, participants who attended
workshops at all other delivery site types, except residential facil-
RESULTS ities, were significantly more likely to attend Spanish-language
SAMPLE CHARACTERISTICS workshops (P < 0.001). Participants in workshops with more par-
Sample characteristics of study participants are presented in ticipants (OR = 1.03, P < 0.001) and those attending workshops
Table 1. Of the 12,208 study participants, 65.1% attended Spanish- at delivery sites in areas with higher percentages of Hispanic fam-
language workshops and 78.3% successfully completed the pro- ilies below the federal poverty line (OR = 1.15, P < 0.001) were
gram (i.e., attended four or more of the six offered workshop significantly more likely to attend Spanish-language workshops.
sessions). Over 55% of participants were aged 64 years or younger
and 78.4% was female. On average, participants self-reported 1.96 SUCCESSFUL WORKSHOP COMPLETION
(±1.54) chronic conditions. The majority of participants lived Significant differences were observed when comparing sample
with others (92.2%) and resided in metro areas (93.2%). The characteristics by workshop completion in bivariate analyses
largest proportion of these Hispanic participants attended work- (see Table 1). A significantly larger proportion of participants
shops at healthcare organizations (32.1%), followed by senior aged 65–79 years (χ2 = 26.16, P < 0.001) and female participants
centers or AAAs (22.3%), residential facilities (11.1%), and com- (χ2 = 22.42, P < 0.001) successfully completed workshops. Sig-
munity or multi-purpose centers (10.1%). On average, workshops nificantly larger proportions of participants who lived alone
had 13.14 (±4.08) participants, and participants attended 4.49 (χ2 = 20.41, P < 0.001) and lived in metro areas (χ2 = 4.72,
(±1.64) of the six workshop sessions. P = 0.030) successfully completed workshops. On average, those
who successfully completed workshops resided in (t = −5.10,
ATTENDING SPANISH-LANGUAGE WORKSHOPS P < 0.001) and attended delivery sites in (t = −4.34, P < 0.001)
Significant differences were observed when comparing sample areas with higher percentages of Hispanic families below the
characteristics by workshop language in bivariate analyses (see federal poverty line. Larger proportions of participants who suc-
Table 1). A significantly larger proportion of participants who cessfully completed workshops did so at senior centers or AAAs
enrolled in Spanish-language workshops also received the recom- and other delivery sites, whereas smaller proportions of partici-
mended intervention dose (i.e., attended four or more of the six pants who successfully completed workshops did so at healthcare
workshop sessions) (χ2 = 58.52, P < 0.001). Significantly larger facilities and residential facilities (χ2 = 86.17, P < 0.001). On aver-
proportions of younger participants (χ2 = 149.44, P < 0.001) and age, participants who successfully completed workshops were in
female participants (χ2 = 59.47, P < 0.001) attended Spanish- workshops with fewer participants (t = 3.09, P = 0.002).
language workshops. On average, participants who attended Table 3 presents the logistic regression modeling factors asso-
Spanish-language workshops had fewer chronic conditions ciated with successful workshop completion (i.e., attending fewer
(t = 14.36, P < 0.001). Significantly larger proportions of par- than four workshops served as the referent group). Compared
ticipants who lived alone (χ2 = 14.09, P < 0.001) and lived in to participants under age 50 years, those who were aged 85 years
metro areas (χ2 = 374.95, P < 0.001) attended Spanish-language and older were significantly more likely to be successful com-
workshops. On average, those attending Spanish-language work- pleters (OR = 1.36, P < 0.001). Female participants (OR = 1.22,
shops resided (t = −40.79, P < 0.001) and attended delivery P < 0.001), those who lived alone (OR = 1.34, P = 0.002), and
sites (t = −41.30, P < 0.001) in areas with higher percentages those who resided in areas with higher percentages of Hispanic
of Hispanic families below the federal poverty line. Larger pro- families below the federal poverty line (OR = 1.05, P = 0.002)
portions of participants who attended Spanish-language work- were more likely to successfully complete workshops. Rela-
shops did so at healthcare organizations and educational institu- tive to those who attended workshops in healthcare organiza-
tions, whereas smaller proportions of participants who attended tions, participants who attended workshops at all other delivery
Spanish-language workshops did so at senior centers or AAAs and site types, except educational institutions, were significantly less
residential facilities (χ2 = 464.88, P < 0.001). On average, par- likely to successfully complete workshops (P < 0.05). Participants
ticipants who attended Spanish-language workshops had larger enrolled in Spanish-language workshops were significantly more
class sizes (t = −10.61, P < 0.001) and attended more workshop likely to successfully complete workshops (OR = 1.50, P < 0.001),
sessions (t = −7.56, P < 0.001). whereas, those in workshops with larger class sizes (OR = 0.98,
Table 2 presents the logistic regression modeling factors P < 0.001) were significantly less likely to successfully complete
associated with Hispanic participants’ enrollment in Spanish- workshops. Those attending workshops at delivery sites in areas
language workshops (i.e., attending English-language workshops with higher percentages of Hispanic families below the federal

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 155 | 134
www.frontiersin.org

Smith et al.
Table 1 | Sample characteristics by workshop language and completion.

Total Workshop language Workshop completion

(n = 12,208)

English Spanish χ2 or t P Not successful Successful χ2 or t P

(n = 4,262) (n = 7,946) (n = 2,652) (n = 9,556)

Work shop completion 58.52 <0.001

Not successful 21.7% 25.6% 19.6% – – – –
Successful 78.3% 74.4% 80.4% – – – –
Work shop language 58.52 <0.001
English 34.9% – – – – 41.2% 33.2%
Spanish 65.1% – – – – 58.8% 66.8%
Age 149.44 <0.001 26.16 <0.001
Under 50 25.6% 19.5% 28.9% 26.9% 25.2%
50–64 29.8% 30.0% 29.7% 31.3% 29.4%
65–79 34.8% 39.6% 32.2% 30.8% 35.9%
80+ 9.8% 11.0% 9.1% 11.0% 9.4%
Sex 59.47 <0.001 22.42 <0.001
Male 21.6% 25.6% 19.5% 25.0% 20.7%
Female 78.4% 74.4% 80.5% 75.0% 79.3%
Living situation 14.09 <0.001 20.41 <0.001
Lives with others 92.2% 93.4% 91.5% 94.3% 91.6%
Lives alone 7.8% 6.6% 8.5% 5.7% 8.4%
Rurality (participant residence) 374.95 <0.001 4.72 0.030
Metro 93.2% 87.2% 96.4% 92.3% 93.5%
Non-metro 6.8% 12.8% 3.6% 7.7% 6.5%
Number of chronic conditions 1.96 (±1.54) 2.24 (±1.62) 1.81 (±1.47) 14.36 <0.001 1.96 (±1.57) 1.96 (±1.53) −0.22 0.825
Percent of Hispanics below poverty (participant residence) 10.41 (±10.26) 6.30 (±5.66) 12.61 (±11.43) −40.79 <0.001 9.59 (±8.96) 10.63 (±10.58) −5.10 <0.001
Delivery site type 464.88 <0.001 86.17 <0.001
Healthcare Organization 32.1% 24.7% 36.0% 38.2% 30.4%
Senior center/AAA 22.3% 29.5% 18.4% 20.1% 22.9%

Hispanic adults and Spanish-led workshops

Residential facility 11.1% 13.2% 10.0% 12.1% 10.9%
Community/Multi-Purpose Center 10.1% 10.1% 10.2% 10.2% 10.1%
April 2015 | Volume 2 | Article 155 | 135

Faith-based Organization 6.6% 5.1% 7.5% 5.6% 6.9%

Educational Institution 4.1% 1.3% 5.6% 3.5% 4.2%
Other 13.6% 16.0% 12.3% 10.3% 14.5%
Rurality (delivery site location) 390.99 <0.001 0.38 0.537
Metro 93.7% 87.7% 96.9% 93.4% 93.7%
Non-metro 6.3% 12.3% 3.1% 6.6% 6.3%
Class size 13.14 (±4.08) 12.60 (±4.16) 13.43 (±4.00) −10.61 <0.001 13.35 (±4.06) 13.08 (±4.08) 3.09 0.002
Number of sessions attended 4.49 (±1.64) 4.33 (±1.70) 4.57 (±1.61) −7.56 <0.001 1.76 (±0.82) 5.25 (±0.79) –195.38 <0.001
Percent of Hispanics below poverty (delivery site location) 10.51 (±10.25) 6.35 (±5.64) 12.73 (±11.42) −41.30 <0.001 9.80 (±9.20) 10.70 (±10.52) −4.34 <0.001
Smith et al. Hispanic adults and Spanish-led workshops

Table 2 | Factors associated with enrollment in Spanish-language Table 3 | Factors associated with successful workshop
workshops. completion.

OR P 95% CI OR P 95% CI

Lower Upper Lower Upper

Age: under 50 1.00 – – – Age: under 50 1.00 – – –

Age: 50–64 1.68 <0.001 1.41 1.99 Age: 50–64 1.09 0.320 0.92 1.30
Age: 65–79 1.29 0.002 1.10 1.51 Age: 65–79 1.14 0.103 0.97 1.34
Age: 80+ 0.97 0.681 0.83 1.13 Age: 80+ 1.36 <0.001 1.17 1.59
Male 1.00 – – – Male 1.00 – – –
Female 1.26 <0.001 1.15 1.40 Female 1.22 <0.001 1.10 1.35
Lives with others 1.00 – – – Lives with others 1.00 – – –
lives alone 1.12 0.179 0.95 1.33 Lives alone 1.34 0.002 1.11 1.61
Metro (participant-level) 1.00 – – – Metro (participant-level) 1.00 – – –
Non-metro (participant-level) 0.28 <0.001 0.21 0.39 Non-metro (participant-level) 0.81 0.210 0.57 1.13
Number of chronic conditions 0.85 <0.001 0.82 0.87 Number of chronic conditions 1.01 0.530 0.98 1.04
Percent of Hispanics below 0.98 0.287 0.95 1.02 Percent of Hispanics below 1.05 0.002 1.02 1.08
poverty (participant-level) poverty (participant-level)
Delivery site: Healthcare 1.00 – – – Workshop: English 1.00 – – –
Organization Workshop: Spanish 1.50 <0.001 1.36 1.66
Delivery site: senior 2.27 <0.001 1.98 2.60 Delivery site: Healthcare 1.00 – – –
center/AAA Organization
Delivery site: residential facility 1.07 0.400 0.92 1.24 Delivery site: senior center/AAA 0.52 <0.001 0.45 0.61
Delivery site: 1.60 <0.001 1.35 1.90 Delivery site: residential facility 0.74 0.001 0.63 0.88
Community/Multi-Purpose Delivery site: 0.60 <0.001 0.50 0.73
Center Community/Multi-Purpose Center
Delivery site: Faith-Based 1.96 <0.001 1.64 2.34 Delivery site: Faith-based 0.64 <0.001 0.53 0.78
Organization Organization
Delivery site: Educational 3.46 <0.001 2.82 4.24 Delivery site: Educational 0.81 0.060 0.64 1.01
Institution Institution
Delivery site: other 5.14 <0.001 3.72 7.11 Delivery site: other 0.71 0.013 0.55 0.93
Metro (delivery site-level) 1.00 – – – Metro (delivery site-level) 1.00 – – –
Non-metro (delivery site-level) 0.74 0.072 0.53 1.03 Non-metro (delivery site-level) 1.19 0.341 0.83 1.69
Class size 1.03 <0.001 1.02 1.04 Class size 0.98 <0.001 0.97 0.99
Percent of Hispanics below 1.15 <0.001 1.11 1.19 Percent of Hispanics below 0.96 0.004 0.93 0.99
poverty (delivery site-level) poverty (delivery site-level)

poverty line (OR = 0.96, P = 0.004) were also significantly less health-related information delivered in Spanish (15–17). While
likely to successfully complete workshops. the proportion of Hispanic participants electing to attend Spanish-
language workshops is substantial, further inspection of the larger
DISCUSSION initiative (28) reveals that the majority of workshops delivered
Hispanic participants represented 17.4% of the first 100,000 par- to the first 100,000 participants were English-language CDSMP
ticipants reached through this ARRA implementation effort (28), (78.4%) and the Diabetes Self-Management Program (10.3%),
a percentage that is representative of the overall Hispanic pop- whereas only about 10% were specialized Spanish versions of
ulation in the United States (29). However, relative to the larger the CDSME (i.e., Tomando Control de su Salud, Programa de
population reached in this initiative, participants in our sample Manejo Personal de la Diabetes, and Curso de Manejo Personal de
are younger (25.6% under age 50 compared to 12.0% in the la Artritis). Therefore, it remains to be determined if the number
larger group) (28). This finding is important because it rein- of Hispanic participants would have been larger if more Spanish-
forces that Hispanics in the United States are acquiring chronic language workshops were available across the country, or if older
conditions at younger ages and living with those conditions for Hispanics are becoming increasingly assimilated and comfortable
longer periods of time (5), thus highlighting the necessity for self- with English for health-related information.
management programs. With approximately two-thirds of sam- The overall completion rates among Hispanic participants were
ple participants (n = 7,946) attending Spanish-language CDSME higher than for the total population of CDSME program partici-
program workshops, this study supports previous studies’ assump- pants (28) (i.e., 74.9% completion among all participants, 78.3%
tions about preferences among Hispanic individuals for receiving completion among Hispanic participants, and 80.4% completion

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 155 | 136
Smith et al. Hispanic adults and Spanish-led workshops

among Spanish-language workshop participants), and can be should also identify the participant and delivery characteristics
attributed, in part, to the availability of Spanish-language work- associated with greater health benefits received. Some evidence
shops. This study identified participant and workshop charac- suggests that Hispanic participants in evidence-based programs
teristics associated with attendance at Spanish speaking versus have greater benefits than White non-Hispanic participants (12,
English-language workshops. Relative to English-language work- 39), but it is not clear whether these advantages are due to baseline
shops, Spanish-language workshops attracted a different popu- disadvantages of Hispanic participants, or the way the classes are
lation base (e.g., younger, female, fewer chronic conditions) and structured or made available in community settings.
were held in different settings (e.g., more urban, less affluent set-
tings). These differences may be attributed to other characteristics CONCLUSION
associated with successful Spanish-language workshop comple- The American Recovery and Reinvestment Act of 2009 (i.e.,ARRA)
tion such as attending more workshops in healthcare facilities and Communities Putting Prevention to Work: Chronic Disease Self-
senior centers/AAAs or attending workshops with larger class sizes. Management Program initiative shows the potential for reaching
Several research and practice implications emerge from this Hispanic participants in a variety of delivery sites. This study
study. First, in future research, it will be important to stratify provides insight into Spanish-language CDSME program recruit-
Hispanic participants by ethnic origin to identify characteris- ment and utilization with implications for program adoption in
tics contributing to their program enrollment, attendance, and underserved Hispanic community settings. To grow the numbers
benefits. As indicated in new census designations (30), there is of Hispanic participants reached, it may be important to increase
growing awareness of the importance of differentiating among the capacity of communities and organizations to deliver Spanish-
different Hispanic populations (e.g., Mexican, Mexican American, language programs and utilize culturally tailored and appropriate
Chicano, Puerto Rican, or Cuban), as well as the degree of accul- recruitment materials and channels.
turation (e.g., native or recent immigrant) and available social
support (31). Including these types of measures is also impor-
ACKNOWLEDGMENTS
tant to address inherent biases associated with the current study,
The American Recovery and Reinvestment Act of 2009 Communi-
in that they are likely attributed to Spanish-language workshop
ties Putting Prevention to Work: Chronic Disease Self-Management
preferences.
Program initiative, led by the U.S. Administration on Aging in
Second, additional efforts are needed to understand differ-
collaboration with the Centers for Disease Control and Preven-
ences between surface and deeper intervention approaches, which
tion and the Centers for Medicare and Medicaid Services, allotted
identify language as a defining characteristic as opposed to other
$32.5 million to support the translation of the Stanford program in
intervention strategies that resonate with cultural preferences (32,
45 States, Puerto Rico, and the District of Columbia. The National
33). These elements are especially important for tailoring partici-
Council on Aging served as the Technical Assistance Resource Cen-
pant recruitment and delivery efforts, which may be more or less
ter for this initiative and collected de-identified data on program
feasible based on the delivery site type and socio-economics of the
participation.
residents and service area. Further investigations are warranted
to better understand program preferences among this popula-
tion within the context of existing delivery systems and referral REFERENCES
1. U.S. Census Bureau. The Next Four Decades: The Older Population in the United
patterns. Healthcare settings are already primed to reach diverse States: 2010 to 2050. Washington, DC: U.S. Census Bureau (2010). 138 p.
populations because of their capacity to use bilingual patient 2. Humes K, Jones N, Ramirez R. Overview of Race and Hispanic Origin: 2010.
navigators and community health workers (34). This study’s find- Washington, DC: U.S. Department of Commerce Economics and Statistics
ings suggest that healthcare systems had more capacity to deliver Administration (2011).
3. Passel JS, Cohn DV. US Population Projections: 2005-2050. Washington, DC: Pew
Spanish-language workshops, in contrast to faith-based organi-
Research Center (2008).
zations. Given the traditional importance of religion and church 4. United States-México Border Health Commission. Health Disparities and the
involvement within the Hispanic population (35–38), it may be US-México Border: Challenges and Opportunities. El Paso, TX: United States-
necessary to increase delivery capacity at faith-based organiza- México Border Health Commission (2010).
tions to reach and enroll more Spanish-speaking CDSME program 5. Angel JL, Whitfield KE. The Health of Aging Hispanics: The Mexican-Origin Pop-
ulation. New York, NY: Springer (2007). 294 p.
participants.
6. Agency for Healthcare Research and Quality. 2013 National Healthcare Dispar-
Third, it will be important to understand how the geography ities Report. Rockville, MD: U.S. Department of Health and Human Services
of service delivery affects program utilization. Prior research has (2014).
revealed a service gap in predominantly Hispanic residential areas, 7. Federal Interagency Forum on Aging-Related Statistics. Population: Indicator
which was combined with a tendency for Hispanic participants 1. Number of Older Americans (2012) [cited 2014 July 8]. Available from:
http://www.agingstats.gov/Main_Site/Data/2012_Documents/Population.aspx
to travel further distances to attend CDSMP classes (9). Strate- 8. Administration on Aging. A Statistical Profile of Hispanic Older Americans Aged
gies and partnerships may be needed in certain areas and settings 65+ (2014) [updated 2010; cited 2014 July 9]. Available from: http://www.aoa.
to coordinate transportation for participants without means of gov/Aging_Statistics/minority_aging/Facts-on-Hispanic-Elderly.aspx
travel, which can increase participant retention rates. 9. Salazar C, Smith ML, Perez A, Ahn S, Ory MG. Geospatial characteristics of the
Fourth, in future research and practice efforts, it will be impor- chronic disease self-management program: reaching diverse ethnic population
in San Antonio, Texas. Tex Public Health Assoc J (2011) 63:16–20.
tant to examine whether Spanish- versus English-language work- 10. Livingston G, Minushkin S, Cohn DV, Center PH. Hispanics and Health Care
shops are more effective in terms of achieving positive health and in the United States: Access, Information and Knowledge. Washington, DC: Pew
quality of life outcomes among participants. Further investigations Hispanic Center (2008).

www.frontiersin.org April 2015 | Volume 2 | Article 155 | 137

Smith et al. Hispanic adults and Spanish-led workshops

11. Escarce J, Kapur K. Access to and quality of health care. In: Tienda M, Mitchell 30. U.S. Census Bureau. Hispanic Origin (2012) [cited 2014 August 5]. Available
F, editors. Hispanics and the Future of America. (Chap. 10), Washington, DC: from: http://www.census.gov/population/hispanic/
National Academies Press (2006). p. 410–46. 31. Gonzalez-Barrera A, Lopez MH. A Demographic Portrait of Mexican-
12. Smith ML, Ahn S, Mier N, Jiang L, Ory MG. An evidence-based program to Origin Hispanics in the United States. Pew Research Hispanic Trends Project
reduce fall-related risk among older adults: a comparison of program efficacy [Internet] (2013). Available from: http://www.pewhispanic.org/2013/05/01/
by ethnicity. Calif J Health Promot (2012) 10(1):28–44. a-demographic-portrait-of-mexican-origin-hispanics-in-the-united-states/
13. Ory MG, Smith ML, Kulinski KP, Lorig K, Zenker W, Whitelaw N. Self- 32. Mier N, Ory MG, Medina A. Anatomy of culturally sensitive interventions
management at the tipping point: reaching 100,000 Americans with evidence- promoting nutrition and exercise in Hispanics: a critical examination of
based programs. J Am Geriatr Soc (2013) 61(5):821–3. doi:10.1111/jgs.12239 existing literature. Health Promot Pract (2010) 11(4):541–54. doi:10.1177/
14. National Council on Aging. Evidence-Based Health Promotion Programs for Older 1524839908328991
Adults: Key Factors and Strategies Contributing to Program Sustainability. Wash- 33. Mier N, Ory MG, Toobert DJ, Smith ML, Osuna D, McKay JR, et al. A qualitative
ington, DC: National Council on Aging (2012). case study examining intervention tailoring for minorities. Am J Health Behav
15. DuBard CA, Gizlice Z. Language spoken and differences in health status, access (2010) 34(6):822–32.
to care, and receipt of preventive services among US Hispanics. Am J Public 34. Riddick S. Improving access for limited English-speaking consumers: a review
Health (2008) 98(11):2021–8. doi:10.2105/AJPH.2007.119008 of strategies in health care settings. J Health Care Poor Underserved (1998)
16. Pearson WS, Ahluwalia IB, Ford ES, Mokdad AH. Language preference as a pre- 9(5):S40–61. doi:10.1353/hpu.2010.0672
dictor of access to and use of healthcare services among Hispanics in the United 35. Koenig H, King D, Carson VB. Handbook of Religion and Health. 2nd ed. New
States. Ethn Dis (2008) 18(1):93–7. York, NY: Oxford University Press (2012). 1192 p.
17. Cheng EM, Chen A, Cunningham W. Primary language and receipt of recom- 36. Koenig HG, Pargament KI, Nielsen J. Religious coping and health status in
mended health care among Hispanics in the United States. J Gen Intern Med medically ill hospitalized older adults. J Nerv Ment Dis (1998) 186(9):513–21.
(2007) 22(Suppl 2):283–8. doi:10.1007/s11606-007-0346-6 doi:10.1097/00005053-199809000-00001
18. Lorig K. Living a Healthy Life with Chronic Conditions: Self-Management of Heart 37. Musgrave CF, Allen CE, Allen GJ. Spirituality and health for women of color.
Disease, Arthritis, Diabetes, Asthma, Bronchitis, Emphysema & Others. 3rd ed. Am J Public Health (2002) 92(4):557–60. doi:10.2105/AJPH.92.4.557
Boulder, CO: Bull Publishing Company (2006). 8, 382 p. 38. Ahn S, Hochhalter AK, Moudouni DK, Smith ML, Ory MG. Self-reported phys-
19. Ory MG, Ahn S, Jiang L, Lorig K, Ritter P, Laurent DD, et al. National study of ical and mental health of older adults: the roles of caregiving and resources.
chronic disease self-management six-month outcome findings. J Aging Health Maturitas (2012) 71(1):62–9. doi:10.1016/j.maturitas.2011.10.011
(2013) 25(7):1258–74. doi:10.1177/0898264313502531 39. Smith ML, Cho J, Salazar CI, Ory MG. Changes in quality of life indicators
20. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, Whitelaw N, et al. Successes among chronic disease self-management program participants: an examination
of a national study of the chronic disease self-management program: meet- by race and ethnicity. Ethn Dis (2013) 23(2):182–8.
ing the triple aim of health care reform. Med Care (2013) 51(11):992–8.
doi:10.1097/MLR.0b013e3182a95dd1
21. Ahn S, Basu R, Smith ML, Jiang L, Lorig K, Whitelaw N, et al. The impact Conflict of Interest Statement: The authors declare that the research was conducted
of chronic disease self-management programs: healthcare savings through a in the absence of any commercial or financial relationships that could be construed
community-based intervention. BMC Public Health (2013) 13(1):1141. doi:10. as a potential conflict of interest.
1186/1471-2458-13-1141
22. Bandura A. Social cognitive theory of self-regulation. Organ Behav Hum Decis This paper is included in the Research Topic, “Evidence-Based Programming for Older
Process (1991) 50(2):248–87. doi:10.1016/0749-5978(91)90022-L Adults.” This Research Topic received partial funding from multiple government and
23. Kulinski KP, Boughtaugh M, Smith ML, Ory MG, Lorig K. Setting the stage: mea- private organizations/agencies; however, the views, findings, and conclusions in these
sure seletion, coordination, and data collection for a national self-management articles are those of the authors and do not necessarily represent the official position
initiative. Front Public Health (2015) 2:206. doi:10.3389/fpubh.2014.00206 of these organizations/agencies. All papers published in the Research Topic received
24. U.S. Department of Health and Human Services, Administration on peer review from members of the Frontiers in Public Health (Public Health Education
Aging. ARRA – Communities Putting Prevention to Work: Chronic Dis- and Promotion section) panel of Review Editors. Because this Research Topic repre-
ease Self-Management Program (2012) [cited 2014 June 14]. Avail- sents work closely associated with a nationwide evidence-based movement in the US,
able from: www.cfda.gov/index?s=program&mode=form&tab=step1&id= many of the authors and/or Review Editors may have worked together previously in
5469a61f2c5f25cf3984fc3b94051b5f some fashion. Review Editors were purposively selected based on their expertise with
25. Stanford School of Medicine. Chronic Disease Self-Management Program (Better evaluation and/or evidence-based programming for older adults. Review Editors were
Choices, Better Health® Workshop) (2014) [updated 2014; cited 2014 July 12]. independent of named authors on any given article published in this volume.
Available from: http://patienteducation.stanford.edu/programs/cdsmp.html
26. United States Department of Agriculture Economic Research Service. Rural- Received: 11 August 2014; paper pending published: 01 September 2014; accepted: 08
Urban Commuting Area Codes: Overview (2014) [updated 2014; cited September 2014; published online: 27 April 2015.
2014 August 7]. Available from: http://www.ers.usda.gov/data-products/ Citation: Smith ML, Ahn S, Jiang L, Kulinski KP and Ory MG (2015) Factors
rural-urban-commuting-area-codes.aspx#.U-ObPmPqlZU associated with Hispanic adults attending Spanish-language disease self-management
27. U.S. Census Bureau. Census 2000, Summary File 1, Table S1702; gener- program workshops and workshop completion. Front. Public Health 2:155. doi:
ated by Scott Horel; using American FactFinder (2012). Available from: http: 10.3389/fpubh.2014.00155
//factfinder2.census.gov This article was submitted to Public Health Education and Promotion, a section of the
28. Smith ML, Ory MG, Ahn S, Kulinski KP, Jiang L, Horel S, et al. National dissem- journal Frontiers in Public Health.
ination of chronic disease self-management education programs: an incremen- Copyright © 2015 Smith, Ahn, Jiang , Kulinski and Ory. This is an open-access article
tal examination of delivery characteristics. Front Public Health (2015) 2:227. distributed under the terms of the Creative Commons Attribution License (CC BY).
doi:10.3389/fpubh.2014.00227 The use, distribution or reproduction in other forums is permitted, provided the original
29. U.S. Census Bureau. State & County QuickFacts: USA (2014) [updated 2014; author(s) or licensor are credited and that the original publication in this journal is cited,
cited 2014 August 3]. Available from: http://quickfacts.census.gov/qfd/states/ in accordance with accepted academic practice. No use, distribution or reproduction is
00000.html permitted which does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 155 | 138
 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00294

Methods for streamlining intervention fidelity checklists:

an example from the Chronic Disease Self-Management
Program
SangNam Ahn 1,2 *, Matthew Lee Smith 3 , Mary Altpeter 4 , Basia Belza 5 , Lindsey Post 1 and Marcia G. Ory 2
1
Division of Health Systems Management and Policy, School of Public Health, The University of Memphis, Memphis, TN, USA
2
Department of Health Promotion and Community Health Sciences, Texas A&M Health Science Center, School of Public Health, College Station, TX, USA
3
Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA, USA
4
Center for Health Promotion and Disease Prevention, The University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
5
School of Nursing, Biobehavioral Nursing and Health Systems, The University of Washington, Seattle, WA, USA

Edited by: Maintaining intervention fidelity should be part of any programmatic quality assurance (QA)
Katherine Henrietta Leith, University
plan and is often a licensure requirement. However, fidelity checklists designed by original
of South Carolina, USA
program developers are often lengthy, which makes compliance difficult once programs
Reviewed by:
Jay E. Maddock, University of Hawaii become widely disseminated in the field. As a case example, we used Stanford’s origi-
at Manoa, USA nal Chronic Disease Self-Management Program (CDSMP) fidelity checklist of 157 items to
Katherine Henrietta Leith, University demonstrate heuristic procedures for generating shorter fidelity checklists. Using an expert
of South Carolina, USA
consensus approach, we sought feedback from active master trainers registered with
*Correspondence:
the Stanford University Patient Education Research Center about which items were most
SangNam Ahn, Division of Health
Systems Management and Policy, essential to, and also feasible for, assessing fidelity. We conducted three sequential surveys
School of Public Health, The and one expert group-teleconference call. Three versions of the fidelity checklist were cre-
University of Memphis, 133 Robison ated using different statistical and methodological criteria. In a final group-teleconference
Hall, Memphis, TN 38152-3530, USA
call with seven national experts, there was unanimous agreement that all three final ver-
e-mail: [email protected]
sions (e.g., a 34-item version, a 20-item version, and a 12-item version) should be made
available because the purpose and resources for administering a checklist might vary from
one setting to another. This study highlights the methodology used to generate shorter
versions of a fidelity checklist, which has potential to inform future QA efforts for this and
other evidence-based programs (EBP) for older adults delivered in community settings.
With CDSMP and other EBP, it is important to differentiate between program fidelity as
mandated by program developers for licensure, and intervention fidelity tools for providing
an “at-a-glance” snapshot of the level of compliance to selected program indicators.
Keywords: intervention fidelity, quality assurance, Chronic Disease Self-Management Program, aging, evidence-
based programs, expert consensus

INTRODUCTION the actual implementation processes at the state or national level

Chronic conditions have received nationwide attention because other than gross indicators of program completion or adherence
of their adverse impact on individuals’ daily functioning, social to recommendations regarding class size (9, 10). Thus, the pri-
interaction, and self-reported quality of life (1) as well as their mary purpose of this project was to address intervention fidelity
association with rising healthcare costs (2). Self-management has and describe a methodological approach to streamline a fidelity
been viewed as a key factor enabling patients to deal with the checklist. As a heuristic example, we used the fidelity checklist
everyday challenges of chronic conditions through medical, emo- contained within the CDSMP Fidelity Toolkit (11). A secondary
tional, and/or role management (3, 4). Despite national calls for purpose of this project was to use this methodological process
more attention to public health strategies that empower Ameri- to evoke expert opinions about how leaders in the aging services
cans to be more involved in their own health (5), many Americans field view the fidelity and quality assurance (QA) processes. As a
do not inherently possess the skills for actively engaging in self- note, we will consistently use “intervention” fidelity throughout
management behaviors that can help ameliorate the effects of the current study because we focus on a fidelity checklist assess-
living with chronic diseases. ing processes beyond the program itself (e.g., training, before the
Evidence-based programs such as the Chronic Disease Self- program, after the program, and evaluation).
Management Program (CDSMP) have gained national and inter-
national recognition for helping people with chronic conditions PUBLIC HEALTH INITIATIVES AND EVALUATIONS
learn self-management skills (6, 7). While positive outcomes of The U.S. Administration on Aging (AoA) Evidence-Based Disease
CDSMP have been well documented (8), less is known about Prevention Grant Programs, initiated in 2003, have stimulated the

www.frontiersin.org April 2015 | Volume 2 | Article 294 | 139

Ahn et al. Streamlining intervention fidelity checklists for practice

development and implementation of evidence-based programs evidence-based health promotion programs such as CDSMP (29)
(EBP) for seniors, which dispel earlier myths that health pro- or EnhanceFitness (30).
motion efforts were futile in older populations (12). With this
greater national appreciation for the potential of evidence-based STRATEGIES FOR ENSURING CDSMP PROGRAM FIDELITY
health promotion programing for improving health and function- Programmatic adherence to implementation aspects of CDSMP is
ing among older adults, the research questions have shifted from supported by a centralized training and certification system that
“do we know what works?” to “can we do what is known to work?” provides for scripted small-group and participatory workshops
This change in focus is now seen with service providers having (2.5 h a week for 6 consecutive weeks) focused on self-management
ready access to a growing list of EBP, which have been widely tested strategies that provide medical, emotional, and role management
in community and clinical settings and within the aging services skills (4). In regard to the training and certification system, it is
network (13, 14). noted that there are three hierarchical levels of trainers (31). First,
Yet, the translation of scientifically tested research findings to a person can be a certified (lay) leader when she or he completes
community-based programs is often slow, fragmented, and sub- 4-day Leader Training and facilitates one 6-week workshop within
ject to speculation by the practitioner community (15). As more 12 months from the training date. Second, master trainer certifi-
EBP are offered by diverse host agencies in more diverse commu- cation can be obtained when a person completes 4.5-day Master
nities, evidence is mounting that their successful dissemination Training and facilitates two 6-week workshops within 12 months
occurs sporadically (15–18). Translational research is coming of of completion of training. Third, a person can be a T-Trainer when
age, and models such as RE-AIM are being formulated to serve she or he completes 4.5-day apprenticeship under supervision of
as guiding frameworks for planning implementation efforts and a Stanford approved certifying T-Trainer and conducts at least
evaluating the public health impacts of EBP (19–21). More specif- one Master Training within 12 months completion. In addition,
ically, the RE-AIM framework seeks to identify and overcome the the program coordinator is another important workforce mem-
challenges facing program planners and practitioners when mov- ber who plays a key role in implementing CDSMP. The program
ing an EBP from the research setting in which it was developed coordinator, who may be a master trainer or lay leader, typi-
to the less-than-perfect, resource-limited, and real-world practice cally engages in a variety of tasks such as: identifying community
environment (19–22). partners, recruiting and supervising workshop leaders and par-
The RE-AIM framework contains the following five key ele- ticipants, arranging for workshop sites, monitoring intervention
ments: reach, effectiveness, adoption, implementation, and main- fidelity, and evaluating program processes and outcomes (32).
tenance (21, 23). Some studies examine all five elements, while A standardized resource material (e.g., the program guide “Liv-
others examine outcomes using one or two key elements (16). ing a Healthy Life with Chronic Conditions” revised in 2012)
Our current study focuses on the “I” in RE-AIM, program imple- (4) helps provide general guidance behind the theory and activ-
mentation processes, specifically fidelity monitoring, which can ities. An implementation manual provides more detailed guid-
be neglected because of funding and logistic issues in large-scale ance to trainers (33), and a fidelity manual outlines mandatory
community-based disease prevention efforts for older adults. program requirements. The 2010 CDSMP Fidelity Toolkit (11)
contains a fidelity checklist with key aspects listed in the follow-
INTERVENTION FIDELITY ing link: http://patienteducation.stanford.edu/licensing/Fidelity_
As EBP become widely disseminated, there has been growing atten- ToolKit2010.pdf . This fidelity checklist as part of the Toolkit is
tion to program fidelity in implementation science (16). In terms called “Must Do’s Fidelity Checklist” and provides guidance for
of translational research, there has been a strong programmatic personnel with regard to the implementation of CDSMP (e.g.,
emphasis on fidelity, which can be defined as the adherence of program coordinators, leaders, master trainers, or T-trainers). Per-
actual treatment delivery to the protocol originally developed (24). sonnel are advised to go through the list and check “Yes” for all the
A breach in intervention fidelity, defined as the adherent and com- items they are currently doing, and are encouraged to incorporate
petent delivery of an intervention by the interventionists (e.g., these items with their fidelity plan for the future if they are not able
trainers, course leaders, and program coordinators) as set forth to implement the entire fidelity task right away. These 157 items
in the intervention manual (25), threatens licensure and makes it are chronologically categorized under 7 headings (16 subheadings
difficult to interpret study results. For example, if the program is or blocks): (1) personnel; (2) delivery before training; (3) fidelity
not delivered as intended, it is difficult to know if the resulting during training; (4) fidelity after training; (5) fidelity during work-
health outcomes can be attributed to receipt of the intervention shops; (6) fidelity for leaders and master trainer retention; and (7)
or to some other variation in the intervention’s delivery (26). fidelity after workshops. Each heading was further divided into a
While maintaining fidelity during program implementation is couple of subheadings.
essential, ensuring the feasibility of monitoring fidelity is also Implementing EBP can require detailed monitoring and track-
important, especially for organizations with relatively limited ing information, placing substantial administrative burdens on
capacity to administer the intervention (27). Fidelity to treat- program coordinators (15). As such, shorter fidelity tools were
ment or intervention delivery is one subset of overall treatment developed by some states implementing CDSMP including Mis-
fidelity (28) and has often been monitored through observation, souri1 and New Jersey2 . However, none of these fidelity tools
interviews, self-assessed fidelity checklists, and pairing of trained
facilitators (26). Recent articles have highlighted the importance 1 http://www.ncoa.org/chamodules/documents/MOCDSMPQITool.pdf

of having high resource commitment to better monitor fidelity in 2 http://www.ncoa.org/chamodules/documents/NJ_PeerLeaderChecklist.doc

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 294 | 140
Ahn et al. Streamlining intervention fidelity checklists for practice

Table 1 | Four rounds to streamline CDSMP fidelity checklists (original 157 items).

Round Methods Reduction criteria Number of participants Number of items left

One Survey Items related to master trainer training (T-training)/language 114 master trainers 148
literacy
Items that did not meet either statistical or practical 114 master trainers 116
approaches
Two Survey Items not ranked “very high” in “feasibility” step and not 47 master trainers 34/20/12
selected as part of a predetermined number of items in
“endorsement” step
Three Survey Items not related to their perceptions about the most critical 7 experts 34
items for assessing program fidelity
Four Conference call Version selection based on organizational resources 7 experts 34a /20/12

a
Most preferred.

was systematically tested. In our role as technical advisors to the a telephone conference to obtain feedback, advice, and concerns
AoA Evidence-Based Disease Prevention Grant Programs, we were from their experiences delivering and overseeing programs. In this
asked to explore methods for streamlining a fidelity checklist (e.g., last round, seven of out of eight experts were able to join the group-
CDSMP) and use these methodological processes to seek expert teleconference call with the study team. Each of the four rounds to
opinions about how leaders in the aging services field view the streamline the CDSMP fidelity checklist contained multiple steps
fidelity and QA processes. as shown below.

MATERIALS AND METHODS ROUND ONE: STREAMLINING THE ORIGINAL CSDMP BY “ENSURING”
The expert consensus method refers to a multi-phase approach THE OVERALL FIDELITY OF CDSMP
for statistically analyzing pooled opinions that minimizes biases Survey process
inherent in other systems of summarizing expert viewpoints (34, The survey in Round One was conducted in November and
35). We used this method to gather and analyze expert opinion December 2010 with master trainers (n = 114) who identified
on streamlining the 157-item CDSMP fidelity checklist in 2010– through the Stanford CDSMP master trainer listserv. These experts
2011. We employed three rounds of data collection using Qualtrics helped us identify the items in the original 157-item CDSMP
software (36) to streamline the original 157-item CDSMP fidelity fidelity checklist, which they believed to be most important to
checklist without losing essential fidelity items but improving fea- ensuring the overall fidelity of the program. Participants were
sibility of administration. As a final effort, we held a telephone asked to rank statements on a scale of one (least important) to
conference to solicit expert opinion for making final recommen- five (most important). During this initial step, we eliminated 8
dations regarding the use of fidelity checklists. Table 1 displays items related to master trainer training and another item referring
these four rounds of checklist streamlining and the number of to language literacy, which resulted in an initial portfolio of 148
items remaining after each round. Human Subjects approval was unique items to consider. To obtain the most relevant information
obtained from the Texas A&M University Institutional Review from participants, the survey included skip patterns that presented
Board. participants with a list of items most appropriate to them based on
their roles (i.e., master trainers or master trainers/program coor-
PARTICIPANTS dinators). Because of their universal relevance, some items were
As the credentialing unit, the Stanford University Patient Educa- presented to every group.
tion Research Center compiles a list of all CDSMP master trainers
and manages a listserv for exchange of information. Using this Reduction criteria
distribution list, we invited all active master trainers in 2010 to To streamline the CDSMP fidelity checklist, we used both statisti-
participate in our study. In Round One, we collected 119 responses cal and practical approaches based upon participants’ ranking of
(114 master trainers, 5 others), where 5 responses were elimi- each checklist item. First, the statistical approach involved elimi-
nated from the final analysis since they were not master trainers. nating items based on their distance from the mean score. Because
Twenty-six out of 114 were both master trainers and program of testing the statistical significance of multiple comparisons, we
coordinators. Out of these 114 master trainers, 47 agreed to par- used the Bonferroni technique by adjusting the significance level
ticipate in the second survey. In Round Two, 24 out of 47 master (0.05) to avoid the risk of Type I error (37). Second, the practi-
trainers responded to the survey (51% response rate). From this cal version involved selecting only those responses that were rated
group, nine master trainers were willing to respond to the third sur- as four or five in terms of importance. We eliminated any items
vey. In Round Three, seven out of nine master trainers responded that “failed” to meet criteria for either the statistical or practical
to the third survey (78% response rate). In Round Four, eight out cut-offs. For instance, there were 26 items in a question block (or
of nine master trainers from Round Three agreed to participate in subheading) asking the importance of fidelity before lay leader

www.frontiersin.org April 2015 | Volume 2 | Article 294 | 141

Ahn et al. Streamlining intervention fidelity checklists for practice

training. Relying on the statistical approach, the mean of these ROUND THREE: FINALIZING THE SHORTENED CDSMP FIDELITY
items was 4.15, and the adjusted significance level for this question CHECKLISTS
block was 0.0019 (i.e., 0.05÷26). This process helped us identify We surveyed nine experts to review the three shortened versions of
six items that showed significantly lower importance ratings rela- the fidelity checklist and to help us address some remaining ques-
tive to the mean after adjusting multiple comparisons. In the same tions related to their perceptions about the most critical items for
question block, the practical approach helped us identify another assessing intervention fidelity using Fidelity-34 as the referent. We
six items that were not rated as four or five. Of the 12 items iden- had seven experts respond to the short survey, and the response
tified for elimination, 2 items were mutually exclusive and 5 items rate was 78%. A report of findings was prepared indicating that
were common to both the statistical and practical approaches. Fidelity-34 was the one that experts felt best balanced the inclusion
Thus, we were able to eliminate seven items that failed to meet the of key fidelity items with the feasibility of administration.
statistical or practical cutoff. In other question blocks (e.g., fidelity
during workshops), we continued to use both approaches to elim- ROUND FOUR: CONFIRMING THE NECESSITY OF A SHORTER CDSMP
inate items that failed to meet the statistical criteria or the practical FIDELITY CHECKLIST
criteria and we were able to delete 32 (22%) out of 148 checklist We held a telephone conference with seven experts identified in
items. At the conclusion of Round One, we asked for volunteers to Round Three to report on our findings and solicit expert opin-
complete a second survey used to continue this expert consensus ion regarding the use of fidelity checklists. This conference call
process. was critically important for obtaining feedback and advice from
those with experience in delivering and overseeing programs. The
ROUND TWO: STREAMLINING THE ORIGINAL CDSMP BY experts confirmed a preference for Fidelity-34, but felt that the
ASCERTAINING “FEASIBILITY” OF ADMINISTRATION AND other two shorter versions should be available since the purpose
“ENDORSEMENT” OF PREDETERMINED NUMBER OF ITEMS and resources for administering the fidelity checklist might vary
Survey process from one setting to another.
The second survey was conducted in June 2011. In this round,
we solicited the opinions of the 47 master trainers who agreed to EXPERT CONSENSUS RESULTS: SUMMARY RECOMMENDATIONS
participate after Round One. Round Two involved asking partici- Table 2 displays the three versions of the CDSMP fidelity checklist
pants to provide feedback via a two-step process that assessed both (i.e., Fideity-12, Fidelity-20, and Fidelity-34). To sum up these
“feasibility” and “endorsement.” The first step asked master train- streamlining processes, we included 14 question blocks out of
ers to rank how feasible they think it is to monitor each fidelity the original 16 blocks (subheadings) that related to personnel to
checklist item. Feasibility response categories included “not at all,” administer and the chronological tasks of CDSMP. In total, 36
“somewhat,” and “very high.” Based on the items identified as hav- items were included in three versions of fidelity checklist. Among
ing “very high” feasibility in the first step, the master trainers were these 36 items, 2 items were only included in the Fidelity-20 (“Have
then asked to select a predetermined number of items (total 34) all Leaders facilitate at least once a year,” and “Have trainees partic-
from each question block that they would endorse for inclusion ipate in two practice teaching activities during training”). Just like
in the final, shorter fidelity checklist versions. We had 24 master Fidelity-34, Fidelity-20 included at least one item of each question
trainers respond to our second survey, resulting in a 51% response block. However, Fidelity-12, as the shortest version of CDSMP
rate. fidelity checklist, did not include any items from four question
blocks: “Fidelity after lay leader training;” “Fidelity before master
Reduction criteria training;” “Fidelity when counseling out leaders/master trainers
Employing the practical methodology based on the feasibility and during training;” and “Fidelity during workshops: Physical envi-
endorsement responses, we were able to generate a shorter Fidelity- ronment and material resources.” Although Fidelity-34 was highly
12 checklist (12 items), a medium-length Fidelity-20 checklist (20 recommended in Rounds Three and Four, CDSMP experts from
items), and a longer version Fidelity-34 checklist (34 items). First, the telephone conference also recommended providing organiza-
we calculated the percent of participants who endorsed each item tional partners with all three versions and allowing them to select
relative to the total number of endorsements received for each the best checklist based on their resources. For instance, organiza-
question block (using a scale from 0 to 100%). We also calculated tions with time restraints and limited staffing may prefer shorter
the percent of participants who endorsed each item received rela- checklist versions to the longer version, whereas organizations with
tive to the total number of feasibility ratings of “very high” (using better staffing and time resources may want to utilize the more
a scale from 0 to 100%). The aggregate of this process was assessed thorough version of the checklist.
on a combined scale of 0–200%. Fidelity-34 included 34 items that
received the highest aggregate numbers of each question block (1, DISCUSSION
2, or 3 items based on number of items in each question block). Using the original CDSMP fidelity checklist as a case example,
Due to our desire to streamline the checklist further, we selected this research provides a methodology for streamlining fidelity
only items with scores of 90 and 100 as cut-points on the combined checklists that have many unique items, making field implementa-
0–200 scale. Fidelity-20 (90 as a cut-off) included 20 items while tion resource-intensive and challenging. We see the methodology
Fidelity-12 (100 as a cut-off) included 12 items. At the conclusion described in this paper as our key contribution, which can be
of Round Two, 9 out of 24 master trainers agreed to participate in applied to different EBPs. It should be noted that these shortened
the survey in Round Three. checklists are for research only and require further field-testing

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 294 | 142
Ahn et al. Streamlining intervention fidelity checklists for practice

Table 2 | Streamlined fidelity checklists using expert consensus method: a CDSMP case studya .

Checklist version

Fidelity-12 Fidelity-20 Fidelity-34

Question block #1: program coordinator qualifications

Ql They are very familiar with both the program fidelity and program implementation manuals X X X
Q2 They have observed a Leader or Master Training X X X

Question block #2: lay leader qualifications

Q3 They are not afraid to speak in front of groups – – X
Q4 They read, write, and speak the language of the workshop participants X X X
Q5 They are able to attend all 4 days of training and complete two practice teachings during training as X X X
prospective leaders

Question block #3: fidelity before lay leader training

Q6 Apply for, renew, or confirm receipt of organization’s program licenseb – – X
Q7 Adhere to recommended schedule for leader trainings (total of 4 days: recommended 2 days per week for – X X
2 weeks)
Q8 Have two certified master trainers who are committed to conduct entire training sessions X X X
Q9 Inform participants that their full attendance and participation is required on all training – – X
Q10 Prepare a complete leader’s manual for each participant – – X

Question block #4: fidelity after lay leader training

Q11 Have all leaders facilitate at least once a year – X –
Q12 Not let those leaders with whom there are concerns facilitate workshopsb – X X

Question block #5: master trainer qualifications

Q13 They are willing and available to attend a 4.5-day-master training X X X
Q14 They either have led two workshops as a leader either before coming to master training or are willing – – X
and available to lead two workshops within 1 year after master training

Question block #6: fidelity before master training

Q15 Prepare master trainer manuals and leader manuals for each participant – – X
Q16 Determine the most recent training materials are being used for training (most current version are 3rd – – X
edition, living a healthier life with chronic conditions book and CDSMP manual (2006)
Q17 Follow the Stanford Patient Education Research Center’s checklist for master trainings (obtained upon – – X
confirmation of training request)
Q18 Inform participants their full attendance and participation is required on all training days – X X

Question block #7: fidelity during master training

Q19 Have trainees participate in two practice teaching activities during training – X –
Q20 Have trainees complete the second practice teaching session and demonstrate a minimum set of core X X X
competency as observed by a master trainer or T-trainerb
Q21 Make sure that training must be at least 27 h, usually over 4.5 days – – X
Q22 Have training offered by two certified T-trainers – – X
Q23 Understand and agree with the importance of program fidelity – – X

Question block #8: fidelity after master training

Q24 Conduct one leader training a yearb X X X

Question block #9: fidelity in judging trainee competence during training

Q25 Adheres to the curriculum (also includes appropriate presentation of charts) X X X
Q26 Facilitates group contributions particularly in the following types of activities: brainstorming, action – X X
planning, action plan feedback, and problem solving
Q27 Models activities appropriately – – X

Question block #10: fidelity when counseling out leaders/master trainers during training
Q28 Observe and document problem behaviors – – X
Q29 Give the trainee specific reasons and examples of why they are concerned – – X

(Continued)

www.frontiersin.org April 2015 | Volume 2 | Article 294 | 143

Ahn et al. Streamlining intervention fidelity checklists for practice

Table 2 | Continued

Checklist version

Fidelity-12 Fidelity-20 Fidelity-34

Q30 Tell the trainee what she/he did well, but also tell her/him clearly how they are expected to improve – – X

Question block #11: fidelity for lay leader and master trainer retention
Q31 Have defined protocols for resolution of potential personality conflicts, communication problems, X X X
improper behavior with participants and co-leaders or co-trainers is in place

Question block #12: fidelity during workshops: physical environment and material resources
Q32 Have the necessary number and quality of educational materials and supplies – – X
Q33 Offer the workshop 2.5 h a week over 6 weeks – X X

Question block #13: fidelity during workshops: lay leader performance

Q34 Have two leaders teach the workshops X X X
Q35 Ensure that leaders use facilitation techniques appropriately and effectively – X X

Question block #14: fidelity after workshops

Q36 Track leader activity (i.e., programs they teach, retention rates)b X X X

X, include; –, exclude.
a
CDSMP sites must comply with licensure and fidelity requirements as defined by Stanford University Patient Education Research Center. The purpose of
this research was to demonstrate the use of expert consensus method to streamline intervention fidelity monitoring checklists and to improve monitoring
of evidence-based program fidelity. For the most current CDSMP licensure information, please visit Stanford University Patient Education Research Center
(http://patienteducation.stanford.edu/licensing/).
b
Some of these items represent a slight modification from the Stanford Self-Management Fidelity ToolKit (2010) to fit the current question format.

before specific endorsements can be made. The CDSMP fidelity In addition to generating three abbreviated versions of CDSMP
checklist has been updated since the time of this study, and a fidelity checklist, this study also demonstrated an effective use of
new fidelity manual was developed that stresses the importance of expert consensus method for generating consensus from a broad
setting intervention fidelity within an overall fidelity plan. These field of CDSMP experts with varying experience and perspec-
updated materials also distinguish “must do” fidelity strategies tives. When dealing with a diversity of opinions, problems may
from those that are“nice to do”to strengthen program fidelity (33). arise due to conflicting viewpoints, self-censorship due to lack of
For the most current requirements, please refer to the Stanford anonymity, incomplete feedback loops or poor communication,
University Patient Education Center website3 . or lack of defined statistical methods for attributing quantita-
Applying the expert consensus technique, we consolidated the tive values to subjective factors. Many of these potential pitfalls
157-item CDSMP fidelity checklist into three abbreviated versions were avoided by using three iterative rounds of online surveys.
without sacrificing fidelity items deemed essential by master train- Because respondents did not communicate directly, they were free
ers. Due to its overall length, the original Stanford checklist was to express their opinions. Statistical and practical methods were
often used more as a self-assessment reminder rather than an used to give each respondent’s perspective equal weight in reach-
actual fidelity checklist. Given the importance of program imple- ing consensus. In the Round Four telephone conference, expert
mentation as a core component of program evaluation (21, 23), participants did directly communicate, but because much of the
we believe shorter fidelity checklists will prove beneficial to cur- work related to consolidating the list was already concluded, the
rent and future program leaders and coordinators who are trying participants were able to reach consensus easily.
to implement EBP with limited financial or time resources. At the end of the process, we were able to identify the most
Though the abbreviated checklists are likely less time- relevant and applicable items and garner experts’ endorsement of
consuming than the original version, further improvements may the abbreviated checklists as useful. The methods employed in this
be needed. First, any checklist should be seen as just one element study could be used as a model for administrators of other EBP
in an overall QA plan (38). Additionally, one might explore dif- aiming to reduce the length of a fidelity checklist for program
ferent delivery modalities. For example, enabling users to access monitoring. However, it is important to note that CDSMP sites
fidelity checklists online could further enhance their usefulness. must comply with licensure requirements and be familiar with the
An online monitoring system would allow for real-time review official CDSMP Implementation and Fidelity Manuals, which will
and feedback so that program coordinators and trainers can eval- need to be updated on a regular basis, if programmatic or licensure
uate their progress and fidelity as implementation is occurring, requirements change.
making changes when necessary to adhere to program guidelines. Several limitations should be considered. First, our relying
on small-group processes should be understood for its lack of
true representation of the general CDSMP master trainer popu-
3 http://patienteducation.stanford.edu/licensing/ lation. Nevertheless, we believe the current study contributes to

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 294 | 144
Ahn et al. Streamlining intervention fidelity checklists for practice

the literature related to streamlining a fidelity checklist because ACKNOWLEDGMENTS

of the relative large sample size (n = 114) in Round One survey, This work was supported by the Administration on Aging
and multiple rounds of survey, and a final telephone conference [90OP0001/03]; and the National Institute of Child Health and
with content experts. Second, our study is potentially limited by Human Development [R01HD047143]. The opinions expressed
focusing on one major stakeholder group, i.e., master trainers. The are those of the authors and do not reflect those of the funding
inclusion of various opinions from CDSMP completers, lay lead- agency. The authors acknowledge the university members of the
ers, T-trainers, program developers, and researchers in academia CDC’s Prevention Research Centers Healthy Aging Research Net-
in the Round Three survey could have ensured better represen- work (PRC-HAN). The authors also thank Ms. Patti Smith, who
tation by various CDSMP implementers and stakeholders. Third, helped to review this manuscript.
the CDSMP program was revamped after the initiation of this
study. While some aspects of the program have been improved REFERENCES
upon, we believe the basic elements have remained consistent 1. Rothrock NE, Hays RD, Spritzer K, Yount SE, Riley W, Cella D. Relative to
to ensuring fundamental intervention fidelity based on what is the general US population, chronic diseases are associated with poorer health-
needed before, during, and after workshop delivery. Hence, the related quality of life as measured by the patient-reported outcomes measure-
need for shortened fidelity checklists is still relevant. Fourth, the ment information system (PROMIS). J Clin Epidemiol (2010) 63:1195–204.
doi:10.1016/j.jclinepi.2010.04.012
study was conducted within a specific time period when federal 2. Kaiser Family Foundation. Health Care Costs: A Primer, Key Information on
funding was available for CDSMP implementation. Such changes Health Care Costs and Their Impact. Menlo Park, CA: The Henry J. Kaiser Family
in the availability of resources and program demands may influ- Foundation (2012).
ence, which fidelity checklist items are most feasible to monitor. 3. Bodenheimer T, Lorig K, Holman H, Grumbach K. Patient self-management of
chronic disease in primary care. JAMA (2002) 288:2469. doi:10.1001/jama.288.
However, the ever-changing context and limited funding resources
19.2469
might make three short versions of the CDSMP fidelity check- 4. Lorig K, Holman H, Sobel D, Laurent D. Living a Healthy Life with Chronic Con-
list helpful and valuable. Fifth, there were noticeable dropouts ditions: Self Management of Heart Disease, Arthritis, Diabetes, Asthma, Bronchitis,
between streamlining rounds. Since data were not collected from Emphysema and Others. Boulder, CO: Bull Publishing Company (2006).
the purposive list of participants, it was not possible to make real 5. Kulinski K, Smith ML, Ory MG. The Chronic Disease Self-Management Program:
Process, Organization, and Data Collection. Washington, DC: National Council
comparisons between responders and non-responders. Sixth, we
on Aging (2014).
note that generating one intervention fidelity tool may not be the 6. Lorig K. Chronic Disease Self-Management Program: Insights from the Eye of the
most effective or efficient way to capture all aspects of fidelity, Storm. Stanford, CA: Patient Education Research Center, School of Medicine,
which could be implementation fidelity for leaders or program Stanford University (2014).
fidelity for program coordinators and agency administrators. Last, 7. Ory MG, Smith ML, Patton K, Lorig K, Zenker W, Whitelaw N. Self-management
at the tipping point: reaching 100,000 Americans with evidence-based programs.
we do not know the level of expertise and knowledge among par- J Am Geriatr Soc (2013) 61:821–3. doi:10.1111/jgs.12239
ticipated master trainers since we did not collect this information. 8. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, Whitelaw N, et al. Suc-
Nevertheless, we strongly believe master trainers are appropri- cesses of a national study of the Chronic Disease Self-Management Program:
ate experts given the required level of training and experience meeting the triple aim of health care reform. Med Care (2013) 51:992–8.
with delivering CDSMP. To retain their certificates, master trainers doi:10.1097/MLR.0b013e3182a95dd1
9. Smith ML, Ory MG, Belza B, Altpeter M. Personal and delivery site char-
are required to conduct 4-day Leader training within 18 months acteristics associated with intervention dosage in an evidence-based fall risk
of original training and conduct either a 4-day Leader Train- reduction program for older adults. Transl Behav Med (2012) 2:188–98.
ing, a 6-week series of community workshops, or a Leader cross doi:10.1007/s13142-012-0133-8
training (31). 10. Smith ML, Hochhalter AK, Cheng Y, Wang S, Ory MG. Programmatic influences
on outcomes of an evidence-based fall prevention program for older adults:
a translational assessment. Transl Behav Med (2011) 1:384–93. doi:10.1007/
CONCLUSION s13142-011-0058-7
Fidelity is critical to the successful dissemination of EBP. The chal- 11. Stanford University Patient Education Research Center. Stanford Self-
lenge for the field is balancing resources for program delivery with Management Fidelity Tool Kit. San Jose, CA: Stanford University Patient Educa-
those for assuring intervention fidelity. In this study we explore tion Research Center (2010).
12. Ory MG, Hoffman M, Hawkins M, Sanner B, Mockenhaupt R. Challenging aging
one avenue for reducing the administrative resources needed for
stereotypes: designing and evaluating physical activity programs. Am J Prev Med
maintaining fidelity. The study demonstrates the importance of (2003) 25:164–71. doi:10.1016/S0749-3797(03)00181-8
finding ways to streamline intervention fidelity checklists for EBP, 13. National Council on Aging. Evidence-Based Health Promotion Programs for Older
suggesting several key points. First, an expert consensus method Adults. Washington, DC: NCOA (2012).
is a viable approach to assessing the usability of fidelity checklists. 14. National Council on Aging. Chronic Disease Self-Management Program Point of
Contact Map. Arlington, VA: NCOA (2014).
Second, online software (e.g., Qualtrics) can be used for efficient 15. Glasgow RE, Emmons KM. How can we increase translation of research into
data collection, analyses, and tracking of participant response with practice? Types of evidence needed. Annu Rev Public Health (2007) 28:413–33.
built-in reminder systems. Third, it is important to have input doi:10.1146/annurev.publhealth.28.021406.144145
from stakeholders with various roles to have a comprehensive pic- 16. Gaglio B, Shoup JA, Glasgow RE. The RE-AIM framework: a systematic review
ture of intervention fidelity. We encourage researchers to apply of use over time. Am J Public Health (2013) 103:e38–46. doi:10.2105/AJPH.2013.
301299
this expert consensus model to other EBP, and to conduct fur- 17. Jernigan VBB. Community-based participatory research with Native American
ther study of the reliability, validity, and practicability of fidelity communities: the Chronic Disease Self-Management Program. Health Promot
checklists. Pract (2009) 11(6):888–99. doi:10.1177/1524839909333374

www.frontiersin.org April 2015 | Volume 2 | Article 294 | 145

Ahn et al. Streamlining intervention fidelity checklists for practice

18. Green LW, Glasgow RE. Evaluating the relevance, generalization, and applicabil- 34. Karceski S, Morrell M, Carpenter D. The expert consensus guideline series: treat-
ity of research: issues in external validation and translation methodology. Eval ment of epilepsy. Epilepsy Behav (2001) 2:A1–50. doi:10.1016/j.yebeh.2010.07.
Health Prof (2006) 29:126. doi:10.1177/0163278705284445 016
19. Belza B, Toobert DJ, Glasgow RE. RE-AIM for Program Planning: Overview and 35. Frances A, Kahn DA, Carpenter D, Ross R, Docherty JP. The expert consensus
Applications. Washington, DC: National Council on Aging (2007). practice guideline project: a new method of establishing best practice. J Psychiatr
20. Ory GM. Evidence-Based Health Planning 101. Arlington, VA: National Council Pract (1996) 2:295–306.
on Aging (2011). 36. Snow J, Mann M. Qualtrics Survey Software: Handbook for Research Professionals.
21. Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of health Provo, UT: Qualtrics Labs, Inc (2013).
promotion interventions: the RE-AIM framework. Am J Public Health (1999) 37. Vogt WP. Dictionary of Statistics & Methodology: A Nontechnical Guide for the
89:1322–7. doi:10.2105/AJPH.89.9.1322 Social Sciences. Thousand Oak, CA: Sage Publications (2005).
22. Ory MG, Altpeter M, Belza B, Helduser J, Zhang C, Smith ML. Perceived utility 38. National Council on Aging. Checklist: Recommended Components for an “Ideal”
of the RE-AIM framework for health promotion/disease prevention initiatives Quality Assurance (QA) Plan. Arlington, VA: NCOA (2014).
for older adults: a case study from the U.S. evidence-based disease prevention
initiative. Front Public Health (2015) 2:143. doi:10.3389/fpubh.2014.00143
23. Glasgow RE, Klesges LM, Dzewaltowski DA, Estabrooks PA, Vogt TM. Evaluat- Conflict of Interest Statement: The authors declare that the research was conducted
ing the impact of health promotion programs: using the RE-AIM framework to in the absence of any commercial or financial relationships that could be construed
form summary measures for decision making involving complex issues. Health as a potential conflict of interest.
Educ Res (2006) 21:688–94. doi:10.1093/her/cyl081
24. Borrelli B. The assessment, monitoring, and enhancement of treatment fidelity This paper is included in the Research Topic, “Evidence-Based Programming for Older
in public health clinical trials. J Public Health Dent (2011) 71:S52–63. doi:10. Adults.” This Research Topic received partial funding from multiple government and
1111/j.1752-7325.2011.00233.x private organizations/agencies; however, the views, findings, and conclusions in these
25. Santacroce SJ, Maccarelli LM, Grey M. Intervention fidelity. Nurs Res (2004) articles are those of the authors and do not necessarily represent the official position
53:63–6. doi:10.1097/00006199-200401000-00010 of these organizations/agencies. All papers published in the Research Topic received
26. Frank JC, Coviak CP, Healy TC, Belza B, Casado BL. Addressing fidelity in peer review from members of the Frontiers in Public Health (Public Health Educa-
evidence-based health promotion programs for older adults. J Appl Gerontol tion and Promotion section) panel of Review Editors. Because this Research Topic
(2008) 27:4–33. doi:10.1177/1090198114543007 represents work closely associated with a nationwide evidence-based movement in
27. Mowbray CT, Holter MC, Teague GB, Bybee D. Fidelity criteria: develop- the US, many of the authors and/or Review Editors may have worked together pre-
ment, measurement, and validation. Am J Eval (2003) 24:315–40. doi:10.1177/ viously in some fashion. Review Editors were purposively selected based on their
109821400302400303 expertise with evaluation and/or evidence-based programming for older adults. Review
28. Bellg AJ, Borrelli B, Resnick B, Hecht J, Minicucci DS, Ory M, et al. Enhancing Editors were independent of named authors on any given article published in this
treatment fidelity in health behavior change studies: best practices and recom- volume.
mendations from the NIH behavior change consortium. Health Psychol (2004)
23:443–51. doi:10.1037/0278-6133.23.5.443
29. Tomioka M, Braun KL, Compton M, Tanoue L. Adapting Stanford’s Chronic Dis- Received: 19 September 2014; accepted: 29 December 2014; published online: 27 April
ease Self-Management Program to Hawaii’s multicultural population. Gerontol- 2015.
ogist (2012) 52:121–32. doi:10.1093/geront/gnr054 Citation: Ahn S, Smith ML, Altpeter M, Belza B, Post L and Ory MG (2015)
30. Tomioka M, Braun KL. Implementing evidence-based programs a four-step pro- Methods for streamlining intervention fidelity checklists: an example from the
tocol for assuring replication with fidelity. Health Promot Pract (2012) 14:850–8. Chronic Disease Self-Management Program. Front. Public Health 2:294. doi:
doi:10.1177/1524839912469205 10.3389/fpubh.2014.00294
31. Stanford University Patient Education Research Center. Leader & Trainer Certifi- This article was submitted to Public Health Education and Promotion, a section of the
cation Guidelines. San Jose, CA: Stanford University Patient Education Research journal Frontiers in Public Health.
Center (2013). Copyright © 2015 Ahn, Smith, Altpeter, Belza, Post and Ory. This is an open-access
32. Stanford University Patient Education Research Center. Implementation Manual. article distributed under the terms of the Creative Commons Attribution License (CC
San Jose, CA: Stanford University Patient Education Research Center (2008). BY). The use, distribution or reproduction in other forums is permitted, provided the
33. Stanford University Patient Education Research Center. Program Fidelity Man- original author(s) or licensor are credited and that the original publication in this
ual: Stanford Self-Management Programs 2012 Update. San Jose, CA: Stanford journal is cited, in accordance with accepted academic practice. No use, distribution or
University Patient Education Research Center (2012). reproduction is permitted which does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 294 | 146
 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00237

Factors supporting implementation among CDSMP

organizations
Deborah Paone*
Paone & Associates, LLC, Minneapolis, MN, USA

Edited by: Reaching individuals who can benefit from evidence-based health promotion and disability
Matthew Lee Smith, The University of
prevention programs is a goal of federal, state, and local agencies as well as researchers,
Georgia, USA
providers, community agencies, and other stakeholders. Implementation effectiveness at
Reviewed by:
Samuel D. Towne, The University of the organizational level must be achieved in order to reach these individuals and sustain
South Carolina, USA the program. This mixed methods study examined eight organizations within two states
Danice Brown Greer, The University that successfully implemented the Chronic Disease Self-Management Program (CDSMP)
of Texas at Tyler, USA
and sustained it from 4 to 10 years. There were two types of organizations: aging ser-
*Correspondence:
vices and health care. Internal and external implementation factors and influences were
Deborah Paone, Paone & Associates,
LLC, 10413 Rhode Island Circle, explored. Additional examination of state activities (as a key external agent supporting
Minneapolis, MN 55438, USA CDSMP implementation) was conducted. The examination found agreement among the
e-mail: deborahpaone@ eight organizations regarding why they had adopted the CDSMP – citing the alignment
paoneandassociates.com
between the program and their organizations’ mission and purpose to improve health sta-
tus and promote better self-care, and the demonstrated value (benefits) of the program.
Organizations were also alike in that they described the importance of an internal cham-
pion and supportive senior leader. Organizations differed in how they experienced and
valued peer support and collaborative networks. Organizations also differed in how they
filled their CDSMP workshops. Internal drivers and capability were more often discussed
as facilitating successful implementation than external factors. However, state activities
and external support enabled successful adoption – particularly funding and training. The
primary challenges identified by this set of organizations included difficulty in recruiting par-
ticipants (filling workshops) and irregular or insufficient funding sources. These challenges
were identified as significant and represented barriers to sustaining the program.
Keywords: implementation, evidence-based health promotion, organizational capacity, implementation factors,
sustainability, chronic disease self-management program

INTRODUCTION sustained implementation will require ongoing training, coaching,

Reaching individuals who can benefit from evidence-based health feedback, data, and other systems working in tandem to regularly
promotion (EBHP) and disability prevention programs is an maintain the desired behavior [(1), p. 4]. Durlak and DuPre focus
important goal for public health. Stakeholders for successful on environment/context and implementation structure and fac-
EBHP program dissemination and implementation include: the tors that influence the implementation process including: commu-
individual/consumer, program manager or champion within an nity participation/collaboration, provider characteristics, innova-
organization, the organizational executive, a purveyor or exter- tion characteristics, organizational capacity, and technical assis-
nal agent such as the state department of health or of aging, tance/training (2). Greenhalgh and colleagues describe a good
funding organizations, national program centers maintaining “innovation to system fit” as a key factor where the existing values,
fidelity-monitoring, federal agencies, and policymakers. Deci- norms, strategies, goals, skill mix, supporting technologies, and
sions made at each level can change the landscape for effective ways of working are aligned (3).
implementation. Other internal factors influencing implementation success at
Evidenced-based programs can be viewed as complex innova- the organizational level include: organizational leadership, orga-
tions – those requiring multiple inputs within an organization nizational climate, staff capability, staff buy-in, and acceptability
or system. The path from adoption to sustainability of evidence- to the consumer, patient, or client (4–8). External factors found
based programs is often characterized by a series of fits and starts, to be important to include: technical assistance and availability of
with internal and external forces affecting progress. For example, adequate resources (9, 10). Community-based organizations, in
Fixsen, Blasé, and colleagues place special emphasis on human particular, may have additional challenges or constraints requiring
capability and systems [emphasis added] that support the prac- adaptation to the type or level of technical assistance, or to the pro-
titioner/worker implementing the program (1). Because imple- tocol itself (10, 11). For example, one study of community-based
mentation is so dependent on human behavior, successful and organizations found that barriers to EBHP program adoption

www.frontiersin.org April 2015 | Volume 2 | Article 237 | 147

Paone Factors supporting implementation among CDSMP organizations

included: resource constraints, program adaptation challenges, Given the multiple program components and requirements for
and conflicts with organizational culture (12). both instructors and organizational sponsors of CDSMP, the need
for specific marketing or referral methods to attract participants
CHRONIC DISEASE SELF-MANAGEMENT PROGRAM into the program, and the specific funding needed to sustain it, this
The Chronic Disease Self-Management Program (CDSMP) is EBHP program can be considered a complex innovation. Embed-
an evidence-based program for adults with chronic disease to ding the program and sustaining it requires ongoing commitment
encourage these individuals to better manage and maintain their by the organization to continue to invest in the training, materi-
health status. The development of the CDSMP evolved from als, and outreach to keep the workshops filled and facilitated by
knowledge and practice experience gained from the Arthritis Self- instructors who meet the protocol requirements.
Management Program [(13), p. 680]. The CDSMP is designed to
build on the strengths and capability of individuals – including STUDY PURPOSE
belief in their own abilities, knowledge of what to do regarding The ARRA-funded national dissemination and implementation
their condition, and behavior skills to address situations that arise effort for CDSMP provides an opportunity to study the experi-
(13). The program was tested in a randomized controlled trial of ences of organizations and a dataset, which can be mined. This
952 subjects receiving the CDSMP from community-based pro- study uses that dataset as a starting point to identify a set of orga-
gram sites in the 1990s followed by another study of 831 subjects nizations that effectively implemented and sustained the CDSMP.
followed over several years through a longitudinal trial. In both To be considered successful, the organizations had to have offered
trials, CDSMP proved to have significant positive effects on par- at least four workshops in the 2-year timeframe, with a completion
ticipants’ self-efficacy, levels of exercise, self-reported health, and rate of 65% or higher. All organizations had to continue to offer
other health status measures (14). The participant group also had the program at the time of the interviews (2013).
fewer hospital days (14, 15). Key informant interviews with organizational managers
One estimate puts a dollar value of potential medical care cost responsible for the program provided qualitative data. Using a
savings at over $4.2 billion – savings that could be realized from set of internal and external factors that previously have been iden-
better health management if just 10% of persons with chronic tified in the literature as important facilitators, this study exam-
conditions participated in the CDSMP (16, 17). In addition to the ined commonalities and differences among two different types of
medical cost savings, there are quality of life benefits for individu- organizations implementing the CDSMP. It focused on common
als who are more actively engaged in their health management. A internal facilitators and also explored the type and perceived value
study of the health-related outcomes of a sample of 687 CDSMP of external support provided by a key state agent – the department
participants found that significant improvements were observed charged with dissemination of the program.
for health outcomes such as depression, self-assessed health, and The research question was:
unhealthy physical days (18).
The CDSMP follows a 6-week, 2.5 h/week group format and is • “What affected implementation success of the evidence-based
guided by a tightly scripted protocol that is delivered by certified CDSMP among eight organizations located in two states – exam-
instructors. Each week, the workshop focuses on a specific self-care ining a defined set of implementation factors (internal and
management and educational topic. Instructors (two instructors external)?”
are required for every workshop) follow guidelines and partic-
ipants set a goal each week to pursue. Participants report on The purpose of this article is to offer insight on implementation
progress they have made, week by week, to the other participants of CDSMP from the organizational perspective. Understanding
in the group. The format includes facilitated interaction and group more about what factors or influences positively support organi-
sharing. Participants often encourage each other and offer insights zations on the CDSMP implementation “journey” from adoption
into the way they have managed their own conditions. to sustainability can help identify what needs to be enhanced,
Dissemination of the CDSMP was fostered through a collabo- what barriers exist, how some organizations have overcome these
rative initiative (called “Communities Putting Prevention to Work: barriers, and what lessons they have learned. Such insight can
CDSMP”) funded under the American Reinvestment and Recov- help enhance external supports, such as policy, technical assis-
ery Act (ARRA). Two-year grants (2010–2012) totaling $27 million tance, public health marketing, or fidelity-monitoring as well as
were awarded to 45 states. A program evaluation of this national clarify internal organizational elements that were important. This
dissemination initiative for CDSMP was conducted in 2012–2013 knowledge may help increase the likelihood that organizations will
(19). This process evaluation focused primarily on the states’ (pur- effectively implement and sustain the program.
veyors’) activities. The state units on aging and the state or local
public health departments most often performed this role. These MATERIALS AND METHODS
external agents provided technical assistance, training, fidelity- This mixed methods study examined implementation of CDSMP
monitoring, marketing, and other support for a defined period to by eight organizations located in two states (identified as State #1
implementing organizations (19).1 Such external support is one and State #2) and the support offered by their state agency to facil-
factor that was examined in the study described in this article. itate dissemination and implementation. The two states remain
unnamed to protect the identity of the respondents. There were
1 It should be noted that organizations do not have to be a part of a federal initiative two types of organizations included in this sample: aging services
to implement CDSMP. organizations (ASOs), including three area agencies on aging and

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 237 | 148
Paone Factors supporting implementation among CDSMP organizations

one other aging services provider, and health care organizations to conduct. Interviews were recorded and transcribed. The study
(HCOs), including three hospital/clinic systems and one health was submitted to the University of North Carolina Institutional
care center. Review Board and determined to be exempt.

DATA SOURCES AND SAMPLE SELECTION Factors probed in the state representative interviews were:
The ARRA dataset (secondary data) and results from an electronic
survey (primary data) were used to conduct several iterations of • Drivers of CDSMP dissemination and implementation in the
review in order to select the study sample of organizations. The state.
ARRA dataset provided information, by state, on the number and • Type and level of assistance provided to implementing organi-
type of organizations that participated in CDSMP implementation zations in the state.
from 2010 to 2012 through the ARRA network. States were selected • Peer support and communication – whether and how this was
that had participated in a previous national EBHP initiative (from fostered within the state.
2006 to 2007). This was done in order to maximize the likelihood
that the state had invested time and resources to create structures Factors probed in the organizational interviews included:
or processes that fostered dissemination and implementation of
• Drivers for adoption within organization – the “will” to do this.
these EBHP programs.
• Program fit.
There were 24 states that participated in both the prior EBHP
• Ease of use of the protocol.
initiative and the ARRA grant. The pool was further narrowed
• Value.
to seven states that had at least six ASOs and HCOs. The ARRA
• External support (particularly state agency support and peer
dataset was then used to find organizations that met a set of criteria
networking).
indicating implementation effectiveness for CDSMP. These crite-
ria were: the organization offered at least four workshops within State agency representatives were interviewed first (April/May
the 2 years of the ARRA initiative, had at least a 65% completion 2013). This allowed the investigator to ask questions about the
rate, and continued to offer the workshop in 2013. In addition, the external context in which the implementing organizations had
CDSMP program manager within the organization had to have been operating. It provided a picture of the state’s activities in
institutional memory of the implementation process. Institutional fostering CDSMP from the state’s perspective, prior to hearing
memory is defined as knowledge of the organization’s motivation, from the organizations. The aging services division was the entity
climate, and/or steps to beginning the program by virtue of being responsible for CDSMP dissemination within State #1. Two rep-
employed by the organization during the timeframe when this resentatives from this agency participated in the key informant
occurred. interview, including the program coordinator who had been in
that role for 3 years and the director of the division. One represen-
KEY INFORMANTS AND INTERVIEW PROTOCOLS tative from State #2 participated in the key informant interview.
The qualitative data source was comprised of 10 semi-structured This individual had served as the program coordinator since 2007
interviews conducted by telephone by the investigator. The two and was from the department of public health.
types of key informants were: (1) state representatives who were Organizational representatives were interviewed second. There
the responsible managers for the CDSMP in their state, and were eight organizational key informants (one per organization)
(2) organizational representatives who were the managers of the who participated in interviews between June and August, 2013. All
CDSMP within their implementing organizations. had been involved in the implementation of CDSMP for their
Two semi-structured interview protocols were designed with organizations for at least 3 years and all were CDSMP Master
questions probing a set of pre-defined implementation factors, Trainers.
drawing from the work of Durlak and DuPre, Fixsen, Green-
halgh, and Damschroder (1–3, 20). The investigator pilot-tested RESULTS
the instrument with a CDSMP manager who did not participate The results from the state key informant interviews provide con-
in the study. The investigator also had the instrument reviewed text and background to the organizational data and therefore are
by a national program manager providing technical support to offered first.
CDSMP implementing organizations.
Many of the items on the interview protocol had adjectival STATES’ PERSPECTIVES – ADOPTION AND EARLY EFFORTS
responses scaled (best to worst) with a corresponding weight from The technical assistance and dissemination support to implement-
+2 to −2. This five-point scale is consistent with the scale used by ing organizations from State #1 focused on building the capacity
Damschroder (20). and infrastructure for CDSMP. Drivers for the state included
Training in interview techniques was not required as the inves- interest in helping elders to stay active and healthy-support for
tigator was a seasoned interviewer, having conducted more than CDSMP and other EBHP programs were included in the State
100 interviews over 20 years of experience in health services evalu- Plan. The state began offering “mini-grants” to aging services
ation – of both health services professionals as well as laypersons. providers interested in CDSMP through a competitive applica-
Key informants provided verbal and written consent. The ques- tion process. The state tapped into the existing network of Area
tions were provided to each key informant at least 1 week prior Agencies on Aging. The primary support provided to implement-
to the scheduled interview. Each interview took about an hour ing organizations were start-up grants (to cover workshop direct

www.frontiersin.org April 2015 | Volume 2 | Article 237 | 149

Paone Factors supporting implementation among CDSMP organizations

costs), Master Trainer and Peer Leader training sessions, and the STATES’ PERSPECTIVES – IMPLEMENTATION ISSUES AND
CDSMP workbooks, which were to be given or loaned to partic- SUSTAINABILITY
ipants. This state focused on having a corps of Master Trainers Since the ARRA funding ended, one state has focused on support-
trained by Stanford University. The Master Trainers would then ing the Master Trainers and ensuring fidelity-monitoring. This
train workshop leaders. state also includes a calendar of workshop offerings on their gov-
Drivers for CDSMP came from the public health depart- ernment website as well as all the forms that CDSMP providers
ment in State #2. The attraction was the evidence base and the need. With the ARRA funding ended, the other state does not pay
defined purpose/focus for this program, which emphasized per- license costs for CDSMP providers, nor does it compensate organi-
sonal engagement in one’s own health. This state contracted with zations for training costs. The state representative explained that
external agencies to provide technical assistance and support to it has a philosophy of local authority and control and also that
implementing CDSMP organizations. Through these agencies, the program sustainability requires embedding at the organizational
state provided training, workshop materials, marketing support, level. Each organization is expected to create its own business plan
and fidelity-monitoring. The state also required implementing to address CDSMP (as well as to support other health promotion,
organizations to participate in peer collaboration and informa- disability prevention programing). The state still provides some
tion sharing. Initially, this state paid for the license cost of each funding support to the external agency providing technical assis-
funded organization under ARRA. With state support the number tance and peer collaboration facilitation and marketing assistance.
of Master Trainers grew substantially. This state also maintains policy support for CDSMP (e.g., it is in
Comments by these state representatives about early dissemi- the State Plan).
nation efforts included: The state representatives offered the following insights about
organizational implementation of CDSMP and sustainability:
We had programs that encouraged health and wellness of
seniors – but before 2006 people were not aware of CDSMP. I think that an across-the-board issue in implementation is
We did not have this evidence-based program. There was staff turnover and agency redirection as a result . . . Any time
only 1 Master Trainer in the whole state. We needed to build [senior] management changes there is a question – will they
capacity and infrastructure. We started with natural partners see the value?
who had an interest. There is a very high investment upfront to become a CDSMP
We included this kind of focus in our State Plan. There were provider organization – heavy staff or volunteer training and
major goals around empowering older people to stay active certification, etc. That is also an ongoing issue – keeping the
and healthy. volunteers certified and active. They have many reasons why
they might drop out including their own health issues.
We had several organizations that were committed to
evidence-based programing and knew about CDSMP. In fact One thing I’ve seen is if the organization doesn’t truly have
our first training session was led by one of them. We created the buy-in of the higher level administration, it will struggle
partnerships with these organizations and partnered with when the funding ends. Grants are good for start-up, but a
them very closely. sustainability plan is needed.
We found interest among organizations that had already had
ORGANIZATIONAL PERSPECTIVES
a successful track record of offering the program. It then
The results from the electronic survey showed that most of the
grew very organically from one organization to another. As
eight organizations (75%) had begun offering CDSMP between
it [funding] was made available [for implementing organiza-
2006 and 2009 (one began before 2006 and one started in 2010). All
tions] we worked with organizations all over the state – rural,
of the organizations had offered at least four CDSMP workshops
metro, etc.
in the 2-year time period. The range in number of workshops
One of these states contracted separately with a consultant offered spanned from a low of 5 to a high of 21 in this time
agency to identify data elements for tracking and monitoring period. All eight organizations had an overall completion rate2
the program. Each funded CDSMP provider organization was of 65% or higher, with a range from 66 to 78%. In addition, all
required to submit data to the state office on these elements. In of the organizations reported that they followed the program with
2010, this state adopted a name for CDSMP to be used consistently fidelity. Thus the electronic survey confirmed these eight organi-
statewide – this name was branded. In that same year, the state pur- zations met the criteria for inclusion – they represented a group
chased a multi-organizational license for CDSMP for their state. of successful implementers with extensive experience.
This meant that many organizations that with their own single-
organization licenses through Stanford switched to operate under Adoption and fit
the state’s license. The examination found agreement among the eight organiza-
When asked about how information sharing was fostered tions on why the organization had adopted the CDSMP – citing
among CDSMP implementing organizations, these state repre- alignment between the program and the organization’s purpose
sentatives described their role as conveners and facilitators – pro- related to improving health and promoting better self-care. Many
viding forums for these organizations to gather and communi-
cate. This included regional meetings, newsletters, and electronic 2 Completion was defined as the participant completing 4 of the 6 sessions of
list-serves. CDSMP.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 237 | 150
Paone Factors supporting implementation among CDSMP organizations

organizations had begun implementing the program prior to the of participants). Despite this complexity, the program was often
ARRA grant funding in 2010 – thus the grant facilitated the work described by informants as relatively easy to adopt. This was true
that this set of organizations had already begun (it was not the even though many of these organizations had not had any prior
reason that they adopted the program). Informants frequently experience implementing an EBHP program.
discussed the organizational leadership and internal champion for All informants mentioned that the heavily scripted workshop
the program. This strong champion for the program (sometimes sessions, well-developed content of CDSMP, and required train-
it was the respondent) was instrumental in getting their organiza- ing assisted in implementation. This ease of use was noted by both
tion to adopt and implement CDSMP. For example, this comment experienced informants (e.g., community health educators who
was offered: said they had used evidence-based protocols extensively) and by
informants who said they had never used evidence-based proto-
I was the champion for the program and then I influenced
cols. Figure 2 provides a side-by-side comparison across the eight
others. I think the evidence-based nature and availability of
organizations on “ease of use” of the CDSMP protocol.
the training was what attracted me to CDSMP.
Even though we had never done an evidence-based class, I
From the beginning and continuing through the time of the would say the protocol was very easy to follow. The guidelines
interviews – organizational respondents said that CDSMP was were very clear.
seen as a good fit. Even so, CDSMP was often discussed as being
I would say the protocol was very easy because of the part-
somewhat unusual compared to the organization’s other services.
nership we had. When we first started we were under another
The structured protocol of CDSMP was what set this program
license-holder’s license . . . they provided us with technical
apart from the organizations’ usual health education and wellness
support and trainings, and the manuals. That made it easy.
services.
This was somewhat easy in that it was scripted and heavily
This program is a great fit – it fits extremely well. However directed.
I would also say that (especially in the beginning) – in some
ways it was new. It was outside the norm (the group work- Value
shop with a structured protocol) of what we typically did, Comments about value focused on the participants and benefits
how we typically provided education. they received from the program. Respondents talked about see-
The program fits well with the organization. This is because ing participants make progress on their personal health goals and
the core concept of CDSMP is one of promoting the maintain a commitment to a healthier lifestyle. Respondents also
individual’s self-management. discussed program value in terms of alignment with the future
Really this is a perfect fit with our organization. All of our vol- direction for the organization – many mentioned health care
unteers are 55 and older and this program is designed around reform and the growing awareness of the need to achieve bet-
the idea of peer leaders. That fit perfectly. ter population health management, prevent disability or decline
(Figure 3). Comments are offered below:
Figure 1 provides a side-by-side comparison of the respon-
dents’ ratings on “Fit” of CDSMP with their organizations. This program is of extremely high value. It has proven results.
Implementation of CDSMP requires a number of components It also attracts volunteers . . .
(e.g., organizational licensure, instructor training and use of pre- As we go down the health care reform path, I think this kind
scribed guidelines, session scripts and materials, and recruitment of program will be even more valued.

2 2

1.5 1.5

1 1

0.5 0.5

0
0
Org: A B C D E F G H
2=Fits extremely well
2=Very easy
1=Fits well 1=Somewhat easy
0=Neutral 0=Neutral
-1=Not a great fit -1=Somewhat difficult
-2=Does not fit at all -2=Very difficult

FIGURE 1 | Organizational respondents’ rating of CDSMP program FIGURE 2 | Organizational respondents’ rating of “ease of use” of
“Fit” with their organization. CDSMP protocol during their implementation experience.

www.frontiersin.org April 2015 | Volume 2 | Article 237 | 151

Paone Factors supporting implementation among CDSMP organizations

2 2

1.5 1.5

1
1

0.5
0.5

0
0 A B C D E F2 G H
2= Very helpful
2=Extremely high value
1=high value,
1= Somewhat helpful
0=Some value 0 = Neutral
-1=Lile value -1 = Limited Helpfulness
-2=No value at all -2 = Not helpful/harmful

FIGURE 3 | Organizational respondents’ rating of “value of CDSMP” to FIGURE 4 | Organizational respondents’ rating of “level of external
their organizations. support and helpfulness” during their implementation experience.

External support across the state. We are doing fidelity monitoring the same
One objective of this research was to assess the importance of way across the state and have set up a method to do that, as a
external support to the organizations implementing CDSMP. The peer group. We communicate regularly.
support from the state agency disseminating CDSMP was par- A graphic depiction of the ratings from all eight organiza-
ticularly probed, as the state department responsible for CDSMP tions on the level and helpfulness of external support (a combined
dissemination was considered a key purveyor of the program. score from −2 to +2, corresponding to the adjectival responses)
Most of the informants indicated that their state agency had is shown in Figure 4.
assisted with: funding, marketing, training, fidelity-monitoring,
supplies, and peer networking. However, the way that these infor- CHALLENGES
mants valued this support differed from“very helpful”to“neutral.” Organizational key informants were asked about their imple-
One reason given for the lukewarm rating given by a respondent mentation challenges. There was substantial consistency among
was that the support from the state had diminished over time. this set of eight respondents about the challenges they faced in
In addition to the support provided by the state, organiza- implementing CDSMP and in sustaining the program.
tions named other sources of external support including: Stanford
University (served as a source of information, provided sup- Recruitment/lack of demand
plies/materials, and guided fidelity-monitoring), local organiza- Recruiting participants and filling workshops was the number one
tions such as libraries, senior housing facilities, senior centers and challenge described by five of the eight organizational informants.
hospitals (helped with logistics, provided space, and helped with Seven of the eight organizational informants said that it took “very
marketing), and the county health department (helped with peer significant effort” (−2) or “significant effort” (−1) to fill the work-
leader training). shops. Stanford fidelity guidelines recommend that class size be
Respondents differed in how they experienced or perceived the from 8 to 16 participants to optimize peer support and problem-
level of peer support/collaboration their organizations received. solving. Six of the eight organizational representatives said that
Several remarked they did not receive much of this type of external they have had to cancel a class at some point in time due to insuffi-
support. A few said that there was extensive support and collabo- cient registration. These findings are consistent with other studies
ration with similar organizations. This may indicate differences in (19, 21).
the type or level of support offered – or it may be a function of the The lack of demand for CDSMP was seen as being an effect
individual’s or organization’s commitment to and efforts around of at least two things. First, very few individuals with chronic
engaging in peer networks and collaborative activities. Comments disease self-identify as needing the program – that is the indi-
included: vidual hearing or reading about CDSMP does not interpret the
program as being relevant for them. Marketing to consumers
The peer collaboration is not growing. It was initially high,
directly was challenging. This group of organizational managers
but as the program grew, it became minimal.
instead often sought out other collaborative agencies, such as
We have had modest peer collaboration. seniors center managers, senior housing facility managers, case
There is extensive peer collaboration . . . We meet monthly managers, or health coaches to describe and promote the pro-
via conference calls and share information and strategies . . . gram as well as encourage participation among their clientele.
We’ve worked at making this CDSMP operate consistently Second, there was a lack of awareness on the part of physicians

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 237 | 152
Paone Factors supporting implementation among CDSMP organizations

and other clinical providers about CDSMP and its’ benefit. The was covering the costs entirely out of their core operational
organizational informants said that they rarely had direct refer- budget – which also relies heavily on grants.
rals from physicians to the CDSMP workshops (except where the Three of the HCOs talked about the lack of external fund-
program was a referral option within the health system’s medical ing and hospital budget issues. These organizations discussed the
information system). shorter-term focus of their organizations where community health
Informants described extensive efforts to market the program education is not seen as core – more of a community benefit.
and educate adults about the benefits of the program. The four Therefore CDSMP and other health promotion, disease preven-
aging service organizations more often described their “sales” and tion programs are vulnerable to budget cuts. Despite this, one
“outreach” efforts – going to senior centers, retirement housing HCO respondent saw the potential for CDSMP. She said it was
facilities, doctor’s offices and putting up fliers, including informa- becoming more relevant for where the health care system is going
tion in newsletters, and networking with local social services agen- in terms of accountability for population health. Despite their lack
cies. The four HCOs more often described their internal health of current external funding, there was some optimism around the
system connections as sources of referrals, including physicians growing awareness and support for this type of program among
working in the health system and health coaches. the health care organizational respondents.
Getting the workshops/classes filled is difficult – getting the Without additional funding we can’t do this on an ongoing
number of participants we need to hold the class. They need basis . . . I only expect to do 1 or 2 [workshops] this year. This
it, but they don’t understand that – it has to be sold. is down from the 18 workshops we did in 2011 and 2012.
It takes significant effort – marketing and recruitment to fill Hospitals are under a lot of budget restrictions. They are less
the workshops. able to provide this kind of community benefit now. We’ve
We have a system of referrals within the clinic. If a provider had some reductions in staff in community health education.
wants to refer he or she can click on the classes we are offering I don’t think this is self-sustaining – not so far.
through our electronic system – then we get the referral and I think there is some potential demand for this – under ACOs
follow-up. there is a commitment to population health and every mem-
ber within their population. There is a basic level of service
A side-by-side comparison across the eight organizations of
to be provided. CDSMP could be part of that.
ratings on the “Demand/Recruitment” factor is shown in Figure 5.
ORGANIZATIONAL ADVICE
Funding
Organizational informants reflected on lessons learned. They
Many of the key informants also discussed challenges with fund-
used various strategies to address implementation challenges or
ing the CDSMP. The ASOs and HCOs differed in where they
enhance their programs.
obtained funds to support the program. However, they were alike
Strategies and advice included:
in commenting that funds received did not cover full costs.
Three of the aging services providers had partial funding of
• Have a strong program champion internally.
their CDSMP through Older American Act Title III-D funds.
• Build and maintain support at all levels internally especially
Donations and small fees as well as supplemental state grant funds
senior administration and managers or clinical professionals
were other sources of revenue to cover costs. One organization
who can serve as referral sources internally.
• Pursue a variety of ways to extend reach and improve visibility
2 of CDSMP in order to build external referral sources and tap
1.5 into collaborative resources (e.g., volunteers, building space for
1 workshop locations, etc.).
0.5 • Recruit, train, and retain strong workshop leaders (staff or
0 volunteers).
-0.5
• Conduct ongoing marketing and outreach to make target pop-
ulation groups (potential participants) aware of the program.
-1
• Measure results. Present a “return on investment” or value
-1.5
proposition to key stakeholders.
-2

DISCUSSION
2=No trouble (to fill workshop)/lie effort A study of eight organizations, purposively selected because of
1-Some effort
0=Neutral
long-term successful experience with CDSMP, revealed that inter-
-1=Somewhat difficult/significant effort nal drivers and capability were more often discussed as facilitating
-2=Extensive difficulty/very significant effort
successful implementation than external factors.
FIGURE 5 | Organizational respondents’ ratings on “demand/
FACILITATING FACTORS
recruitment factor” – pertaining to the level of effort to fill CDSMP
classes during their implementation of this program. Common facilitating factors for adoption and successful imple-
mentation of the CDSMP included:

www.frontiersin.org April 2015 | Volume 2 | Article 237 | 153

Paone Factors supporting implementation among CDSMP organizations

• Program-to-organization fit. External support

• Organizational leadership. External support also facilitated implementation among these
• Training and well-developed materials. eight organizations. Key external supports described as “very help-
ful” by these eight organizations were: (1) funding, (2) training
These factors are consistent with other studies examining pro- and workshop materials, and (3) fidelity-monitoring. The value of
gram implementation and sustainability. For example, a review of peer support and collaboration varied among this set of respon-
19 empirical studies of health-related programs found the follow- dents – for some organizations this type of support had been
ing factors to be important to the organizations (the study focused (and continued to be) very important. Other organizations had
on organizations continuing the program at least 2 years follow- not participated and/or did not rely on peer support very much.
ing the ending of external funding): program champion, program These findings are consistent with other research identifying key
fit with organizational mission, perceived value/benefits to clients, supports for implementation, including the perceived benefits of
and support of stakeholders (22). using the EB program and collaborative technical assistance or
program supports that are matched or tailored to the organization
Program-to-organization fit (12). This external support may be particularly important to the
These eight organizational informants discussed alignment organization in the adoption and early implementation phases.
between the focus and purpose of CDSMP and the overall purpose
or mission of their organizations as a facilitating factor in adoption BARRIERS AND THREATS TO SUSTAINABILITY
and implementation effectiveness. Others studying implementa- Common barriers and threats to sustainability included difficulties
tion success have discussed the importance of fit between the in recruiting participants, and lack of funding for the program –
innovation (program) and the organization – particularly the fit including lack of participant health insurance coverage for this
with the purpose or values of the organization (3, 4). This may type of EBHP program. It is likely that these challenges are linked.
be an important baseline criterion for an organization to consider
when considering an EBHP program to adopt.
Lack of demand
Organizational leadership Recruiting participants to the workshops was a key challenge
The managers responsible for CDSMP discussed both their among the organizations in this study. Seven of the eight orga-
own leadership as internal champions of the program, and nizations said that getting participants into the workshops was the
the leadership from their supervisors, department directors, or number one challenge. Organizations said that there is low aware-
senior executives – who demonstrated their support for CDSMP ness of the program among both the lay public and physicians – a
adoption and the implementation process. These administrators key referral source.
remained committed to offering the program even with limited The need for better marketing and distribution systems for
funding. public health programs has been identified elsewhere. In a study
Supportive leadership has been identified elsewhere as an orga- of 32 community-based prevention programs only modest pene-
nizational characteristic linked to successful implementation (23). tration occurred in the marketplace, which limited impact. The
The importance of champions and organizational leadership has researchers called for more effective approaches that “employ
been found to be a facilitator to health promotion practices being a reinforcing combination of both high-risk (targeting) and
adopted, implemented, and maintained. For example, in a study of population-wide strategies” [(24), p. 571]. Others working in
five Canadian provincial efforts to adopt a chronic disease preven- public health have pointed out the stark contrast between the
tion initiative, the research team found that there was “remarkable sophistication of marketing and distribution systems for products
consistency in the top factors identified as facilitators and barri- and services in the business sector and the “unassigned, underem-
ers to health promotion capacity building” [(7), p. 470]. Internal phasized, and underfunded” dissemination strategies in the public
organizational factors were most frequently mentioned as facilitat- health sector [(25), p. 215].
ing implementation (more than external factors). Organizational
respondents particularly noted the importance of having skilled, Lack of funding/insurance
committed staff and supportive senior leadership (7). This may Since many people with chronic conditions have Medicare as
be another baseline criterion for organizations when considering their primary insurance, the fact that CDSMP is not covered by
EBHP program adoption. Medicare is an impediment.3 Medicare beneficiaries (and physi-
cians) may believe that if a service or program is not covered by
Training and materials Medicare, then that service has not been shown to have enough
Among these organizations, the CDSMP protocol and training benefit to the patient/consumer to warrant coverage. This has
materials were described as well-developed, easy-to-use, and excel- been shown to be true in other studies where the lack of insur-
lent guides. These materials and training sessions worked well for ance coverage contributed to underuse of proven services, such
both staff members and volunteers. Researchers of implementa-
tion effectiveness have discussed the importance of having quality 3 The Medicare program does provide coverage for patient education and rehabilita-
tools and training (e.g., manuals, guides, worksheets, education, tion, most commonly for specific defined time periods and usually related to a new
skills development, etc.) to support organizational performance diagnosis, illness, surgical procedure, or injury, or an exacerbation of an existing
and implementation effectiveness (23). condition/issue.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 237 | 154
Paone Factors supporting implementation among CDSMP organizations

as secondary prevention programs in cardiac rehabilitation (26). findings would add strength to this examination. This limita-
Other researchers studying implementation and sustainability tion was addressed into some degree by: (1) conversations with
have noted that fiscal support is a critical external factor in some or experts in the field who are familiar with CDSMP and its imple-
all stages of adoption, implementation, and sustainability (6, 27). mentation history and challenges, (2) careful crafting of the key
While the focus or mission of health care and ASOs may informant interview instruments, (3) feedback from the national
be to assist individuals to improve or maintain health, they are dataset program manager familiar with organizations implement-
reimbursed largely for addressing problems (after-the-fact), not ing this program and with state agents, and (4) review of published
preventing them. HCOs’ reimbursement comes primarily from studies about CDSMP, particularly recent program evaluations
illness care/treatment not prevention (28). For social service orga- to identify important external factors. Other limitations include
nizations such as area agencies on aging that provide direct services the small sample size of organizations and the exclusive focus
to elders, provision of services is primarily based on an older indi- on “successful” implementing organizations. This research would
vidual requesting help for an existing problem or need (relying on be strengthened by examining additional organizations of many
grant and OAA funding and skewed to those financially vulnera- types, located within other states, and having variable success in
ble). Thus, the CDSMP runs into the same challenge that many implementation – using the same interview protocol including the
public health interventions face: lack of funding for prevention. scaled response options.
It is the author’s opinion, that without a regular source of
funding or payment for service, the “value” of CDSMP has to be CONCLUSION
demonstrated one person and one provider at a time. The lack of Reflecting on the lessons learned from these eight successful
payment for EBHP programs such as CDSMP may be interpreted CDSMP organizations, recommendations related to enhancing
by the lay public or by physicians as a signal that the program does internal and external supports are offered.
not provide sufficient value in terms of health status improve- Supportive elements for adoption and early implementation
ment or effect. Given these forces at work, participation remains efforts that drove these organizations were very consistent, espe-
low and each referral/registration to a CDSMP workshop is hard- cially organizational leadership and the perceived value and fit of
won. Organizations expend extra effort to get the program costs CDSMP with the mission and purpose of the organizations. Advice
covered for those who do elect to participate. Without demand, by this set of successful organizations included clarifying the ben-
there is little pressure to pay for these programs. Thus, the cycle efit of the program using both participant and organizational
perpetuates. metrics. Thus recommendations to enhance internal capability
This issue goes beyond what the single manager within an to support effective implementation include:
implementing organization can address alone or even what a pro-
• Identify an internal program champion who has the ability
gram coordinator at the state level can solve. It calls for a systems
to help drive adoption and ensure senior level buy-in and
approach – where the stakeholders are aware of common overall
commitment to the program;
objectives, their roles and boundaries in producing results, and the
• Make the case that the program is a good fit with the organiza-
accountability of component parts to one another (29–31).
tion’s mission and purpose;
It is also important that policy and technical assistance is
• Clearly identify the value of the program in terms that make
informed by and supportive of practice in the field. The Centers for
sense to key stakeholders – e.g., to the participant, organiza-
Medicaid and Medicare Services (CMS) has commissioned stud-
tion, funders, and policymakers. Measure and report this value
ies of EBHP programs that include CDSMP, but notes challenges,
consistently and repeatedly to enhance demand and solidify the
particularly how to directly fund community-based wellness and
foundation of support – which will help raise awareness of the
prevention programs for this and other programs (32). CMS calls
value and benefits of CDSMP in the local area and should help
for more research to: “develop a sustainable framework for sup-
in referrals to the program.
porting a health ecosystem of community-based providers, while
not exposing the Medicare program to undue risk” [(32), p. 72]. External supports were also clearly important for adoption and
Meanwhile the infrastructure to support CDSMP may be early implementation efforts among these organizations. Training
eroding. The infrastructure investment under ARRA facilitated and fidelity-monitoring were especially noted, as was funding to
regional and local training, grew state, and organizational expertise get the program up and running. Organizations noted that they
on how to run these programs, fostered fidelity-monitoring peer could have also used help building awareness of the program.
collaboration and shared learning, produced Master Trainers and Therefore, recommendations for program sustainability in terms
peer workshop leaders in every state, and engaged implementing of external supports include:
organizations to commit to and market the program. As evidenced
by the response from these eight organizations in just two states, • Enhance supportive policy at the federal and state level for
these external supporting and infrastructure components seem to CDSMP and programs like it that focus on improving preven-
be shrinking. tion and self-management behaviors of individuals with chronic
conditions and engaging individuals in their own care through
LIMITATIONS fostering organizational readiness;
The primary limitation of this study is due to there being a • Maintain support for training and fidelity-monitoring as a
single investigator. A second researcher to review and confirm funded external support that appears to be key to both imple-
the categorization and coding of comments and to interpret the mentation and sustainability;

www.frontiersin.org April 2015 | Volume 2 | Article 237 | 155

Paone Factors supporting implementation among CDSMP organizations

• Improve visibility/public awareness of the program through Network, Louis De La Parte Florida Mental Health Institute, University of Florida
national campaigns or other methods to lessen the burden on (2005).
2. Durlak JA, DuPre EP. Implementation matters: a review of research on the influ-
each organization to make this program known;
ence of implementation on program outcomes and the factors affecting imple-
• Accelerate efforts to make benefit changes to the Medicare pro- mentation. Am J Community Psychol (2008) 41(3):327–50. doi:10.1007/s10464-
gram to include CDSMP as a defined benefit for persons with 008-9165-0
chronic conditions to reduce the barriers around funding the 3. Greenhalgh T, Robert G, Macfarlane F, Bate P, Kyriakidou O. Diffusion
program and to embed it in the fabric of the organizations that of innovations in service organizations: systematic review and recommen-
dations. Milbank Q (2004) 82(4):581–629. doi:10.1111/j.0887-378X.2004.
have done the hard work of adopting and implementing it.
00325.x
4. Klein KJ, Sorra JS. The challenge of innovation implementation. Acad Manage
Reaching individuals who can benefit from EBHP and disability Rev (1996) 21(4):1055–80. doi:10.2307/259164
prevention programs is an important public health goal. As health 5. Simpson DD. A conceptual framework for transferring research to
care and social support “systems” within the U.S. move haltingly practice. J Subst Abuse Treat (2002) 22(4):171–82. doi:10.1016/S0740-5472(02)
00231-3
forward toward more accountability for producing outcomes in
6. Simpson DD. A framework for implementing sustainable oral health promotion
health status among defined population (patient/client) groups, interventions. J Public Health Dent (2011) 71(s1):S84–94. doi:10.1111/j.1752-
programs like CDSMP will become more relevant as a strategy 7325.2011.00234.x
for population health management. At that time, perhaps the pro- 7. Robinson K, Driedger M, Elliott S, Eyles J. Understanding facilitators of and
gram will be seen as a fundamental service of these organizations, barriers to health promotion practice. Health Promot Pract (2006) 7:467–576.
doi:10.1177/1524839905278955
with funding allocated through internal budgeting processes.
8. Aarons G, Horowitz J, Dlugoz L, Ehrhart M. The role of organizational processes
Until that time, organizations willing to adopt such programs in dissemination and implementation research. In: Brownson R, Colditz G, Proc-
must be supported effectively. Key external supports, such as train- tor E, editors. Dissemination and Implementation Research in Health: Translating
ing, materials, and funding provide the bedrock for dissemination Science to Practice. (Vol. Chapter 7), New York, NY: Oxford University Press
and implementation [(11), p. 46]. National marketing campaigns (2012). p. 128–53.
9. Mitchell R, Florin P, Stevenson J. Supporting community-based prevention and
or other external marketing supports are clearly needed. These health promotion initiatives: developing effective technical assistance systems.
findings are consistent with a more extensive review of activi- Health Educ Behav (2002) 29:620–39. doi:10.1177/109019802237029
ties at the state (intermediary agent) level on CDSMP diffusion 10. Wilcox S, Dowda M, Leviton L, Bartlett-Prescott J, Bazzare T, Campbell-Voytal
and dissemination. In a final evaluation report about CDSMP to K, et al. Active for life: final results from the translation of two physical activ-
the Administration on Aging (focusing on states’ activities), the ity programs. Am J Prev Med (2008) 35(4):340–51. doi:10.1016/j.amepre.2008.
07.001
authors also recommended a “centralized or coordinated process 11. Whitelaw N. System change and organizational capacity for evidence-based
for recruitment, intake, referral, and registration/enrollment” practices: lessons from the field. Generations (2010) 34(1):43–50.
[(19), p. 97]. Both studies call for a coordinated, systems approach. 12. Ramanadhan S, Crisostomo J, Alexander-Molloy J, Gandelman E, Grullon M,
Without this greater effort, individuals with chronic conditions, Lora V, et al. Perceptions of evidence-based programs among community-based
organizations tackling health disparities: a qualitative study. Health Educ Res
medical providers, and potential referral organizations within a
(2011) 27(4):717–28. doi:10.1093/her/cyr088
given region are likely to remain unaware of the program and its 13. Lorig K. Chronic disease self-management a model for tertiary prevention. Am
value. This is a tremendous missed opportunity for public health Behav Sci (1996) 39(6):676–83. doi:10.1177/0002764296039006005
and disability prevention. 14. Lorig KR, Sobel DS, Stewart AL, Brown BW Jr, Bandura A, Ritter P, et al. Evidence
Chronic disease represents one of the top public health issues suggesting that a chronic disease self-management program can improve health
status while reducing hospitalization: a randomized trial. Med Care (1999)
domestically and globally. Stronger public policy to ensure there
37(1):5. doi:10.1097/00005650-199901000-00003
is an infrastructure to support EBHP programs that have demon- 15. Lorig KR, Ritter P, Stewart AL, Sobel DS, Brown BW Jr, Bandura A, et al. Chronic
strated effectiveness with chronic disease populations should be a disease self-management program: 2-year health status and health care uti-
public health priority. lization outcomes. Med Care (2001) 39(11):1217–23. doi:10.1097/00005650-
200111000-00008
ACKNOWLEDGMENTS 16. Ory M, Ahn S, Jiang L, Ritter PL, Whitelaw N, Lorig K. Successes of a national
study of the chronic disease self-management program: meeting the triple
The author would like to thank the state and organizational repre- aim of health care reform. Med Care (2013) 51(11):992–8. doi:10.1097/MLR.
sentatives who shared their insights and implementation experi- 0b013e3182a95dd1
ences and Kristie Kulinski from the National Council on Aging for 17. Ahn S, Basu R, Smith M, Jiang L, Lorig K, Whitelaw N, et al. The impact
her data support. The author thanks Nancy Whitelaw, Ph.D., and of chronic disease self-management programs: Healthcare savings through a
community-based intervention. BMC Public Health (2013) 13:1141. doi:10.
Mary Altpeter, Ph.D., for their review and suggestions on this man-
1186/1471-2458-13-1141
uscript and for serving on the doctoral dissertation committee, 18. Ory M, Ahn S, Jiang L, Whitelaw N, Lorig K, Matthew LS. National study of
which guided the development and execution of this study. Finally, chronic disease self-management program (CDSMP): one year changes in health
thanks to faculty members from the Health Policy and Manage- outcomes for older adults. Presentation Session #289097 at 141st APHA Annual
ment Department of the Gillings School of Public Health from Meeting and Exposition. Boston, MA (2013).
19. Woodstock C, Korda H, Erdem E, Pedersen S, Kloc M, Tollefson E. Chronic
the University of North Carolina-Chapel Hill for their guidance
Disease Self-Management Program Process Evaluation. Washington, DC: Admin-
throughout the doctoral program. istration for Community Living (2013).
20. Damschroder LJ, Aron DC, Keith RE, Kirsh SR, Alexander JA, Lowery JC. Foster-
REFERENCES ing implementation of health services research findings into practice: a consol-
1. Fixsen D, Naoom S, Blase K, Friedman R, Wallace F. Implementation Research: idated framework for advancing implementation science. Implement Sci (2009)
A Synthesis of the Literature. Tampa, FL: The National Implementation Research 4(1):50. doi:10.1186/1748-5908-4-50

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 237 | 156
Paone Factors supporting implementation among CDSMP organizations

21. Frank J, Lau C. Empowering Older People to Take More Control of Their Health 32. Centers for Medicare & Medicaid. Report to Congress: The Centers for
Through Evidence-Based Prevention Programs: A Capping Report. Washington, Medicare & Medicaid Services’ Evaluation of Community-Based Wellness and
DC: National Council on Aging (2013). Prevention Programs under Section 4202(b) of the Affordable Care Act. (2013).
22. Scheirer MA. Is sustainability possible? A review and commentary on empirical Available from: http://innovation.cms.gov/Files/reports/CommunityWellness
studies of program sustainability. Am J Eval (2005) 26(3):320–47. doi:10.1177/ RTC.pdf
1098214005278752
23. Wandersman A, Chien V, Katz J. Toward an evidence-based system for innova-
tion support for implementing innovations with quality: tools, training, tech- Conflict of Interest Statement: The author declares that the research was conducted
nical assistance, and quality assurance/quality improvement. Am J Community in the absence of any commercial or financial relationships that could be construed
Psychol (2012) 50:445–59. doi:10.1007/s10464-012-9509-7 as a potential conflict of interest.
24. Merzel C, D’Afflitti J. Reconsidering community-based health promotion:
promise, performance, and potential. Am J Public Health (2003) 93(4):557–74. This paper is included in the Research Topic, “Evidence-Based Programming for Older
doi:10.2105/AJPH.93.4.557 Adults.” This Research Topic received partial funding from multiple government and
25. Krueuter M, Casey C, Bernhardt J. Enhancing dissemination through marketing private organizations/agencies; however, the views, findings, and conclusions in these
and distribution systems: a vision for public health. In: Brownson R, Colditz G, articles are those of the authors and do not necessarily represent the official position
Proctor E, editors. Dissemination and Implementation Research in Health: Trans- of these organizations/agencies. All papers published in the Research Topic received
lating Science to Practice. (Vol. Chapter 11), New York, NY: Oxford University peer review from members of the Frontiers in Public Health (Public Health Educa-
Press (2012). p. 213–22. tion and Promotion section) panel of Review Editors. Because this Research Topic
26. Balady GJ, Ades PA, Bittner VA, Franklin BA, Gordon NF, Thomas RJ, et al. represents work closely associated with a nationwide evidence-based movement in
Referral, enrollment, and delivery of cardiac rehabilitation/secondary preven- the US, many of the authors and/or Review Editors may have worked together pre-
tion programs at clinical centers and beyond: a presidential advisory from the viously in some fashion. Review Editors were purposively selected based on their
American heart association. Circulation (2011) 124:2951–60. doi:10.1161/CIR. expertise with evaluation and/or evidence-based programming for older adults. Review
0b013e31823b21e2 Editors were independent of named authors on any given article published in this
27. Aarons G, Hurlburt M, Horwitz S. Advancing a conceptual model of evidence- volume.
based practice implementation in public service sectors. Adm Policy Ment Health
(2011) 38:4–23. doi:10.1007/s10488-010-0327-7 Received: 15 July 2014; accepted: 28 October 2014; published online: 27 April 2015.
28. McGinnis JM, Williams-Russo P, Knickman J. The case for more active pol- Citation: Paone D (2015) Factors supporting implementation among CDSMP
icy attention to health promotion. Health Aff (2002) 21(2):78–93. doi:10.1377/ organizations. Front. Public Health 2:237. doi: 10.3389/fpubh.2014.00237
hlthaff.21.2.78 This article was submitted to Public Health Education and Promotion, a section of the
29. Churchman CW. The Systems Approach. New York, NY: Dell Publishing journal Frontiers in Public Health.
Co (1968). Copyright © 2015 Paone. This is an open-access article distributed under the terms of
30. Deming WE. The New Economics for Industry, Government, Education. Cam- the Creative Commons Attribution License (CC BY). The use, distribution or repro-
bridge, MA: MIT Center for Advanced Engineering Study (1993). duction in other forums is permitted, provided the original author(s) or licensor are
31. Ulrich W. Some difficulties of ecological thinking, considered from a sys- credited and that the original publication in this journal is cited, in accordance with
tems perspective: a plea for critical holism. Syst Pract (1993) 6:583–611. accepted academic practice. No use, distribution or reproduction is permitted which
doi:10.1007/BF01059480 does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 237 | 157

 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00140

Examining sustainability factors for organizations that

adopted Stanford’s Chronic Disease Self-Management
Program
Michiyo Tomioka* and Kathryn L. Braun
Office of Public Health Studies, University of Hawai‘i at Mānoa, Honolulu, HI, USA

Edited by: In 2006, funds were received to replicate Stanford’s Chronic Disease Self-Management
Matthew Lee Smith, The University of
Program (CDSMP) among eldercare providers in Honolulu. This case study, conducted
Georgia, USA
1 year after the close of the initial 3-year replication grant, explored factors for sustaining
Reviewed by:
Angie Hochhalter, Scott & White the delivery of CDSMP, with an aim to create guidelines for cultivating sustainability. Face-
Healthcare, USA to-face semi-structured interviews were conducted with one representative from each of
Lynda Anderson, Centers for Disease eight eldercare agencies, with the representative specified by the agency. Representatives
Control and Prevention, USA
discussed the presence and strength (low, medium, or high) of sustainability factors, includ-
*Correspondence:
ing readiness, champions, technical assistance, perceived fit of CDSMP with their agency,
Michiyo Tomioka, Office of Public
Health Studies, University of Hawai‘i CDSMP modifiability, perceived benefits of CDSMP, and other. Only three of the eight
at Mānoa, 1960 East-West Road, agencies (38%) were still offering CDSMP by the end of 2010. Agencies who sustained
Biomed D-209, Honolulu, HI 96822, CDSMP rated higher on all sustainability factors compared to those that did not sustain
USA
the program. Additional factors identified by representatives as important were funding
e-mail: [email protected]
and ongoing access to pools of elders from which to recruit program participants. When
replicating evidence-based programs, sustainability factors must be consciously nurtured.
For example, readiness must be cultivated, multiple champions must be developed, agen-
cies must be helped to modify the program to best fit their clientele, evaluation findings
demonstrating program benefit should be shared, and linkages to funding may be needed.
Keywords: chronic disease, health promotion, evidence-based, minority groups, sustainability

INTRODUCTION Effectiveness, Adoption, Implementation, and Maintenance) help

Demand for preventing, delaying the onset, and managing chronic guide replication processes and evaluation (19).
diseases has escalated. Attention is being given to expanding At the same time, researchers have summarized and advanced
replication of evidence-based health promotion programs, those definitions of program sustainability, identified factors associ-
proven to work, to address chronic disease (1). Several federal ated with sustainability, and developed conceptual frameworks
agencies recommend that service providers adopt evidence-based to understand program sustainability (15, 17, 20–22). According
health promotion programs rather than “reinvent the wheel” to literature review of sustainability research by Wiltsey-Stirman
in efforts to help older adults maintain health and indepen- et al. (18), one of the most cited definitions of sustainability
dent living for as long as possible (2, 3). Yet, studies on how evolved from work of Scheirer (15) and Shediac-Rizhallah and
organizations learn about, adapt, and sustain such programs are Bone (17) and is:“the integration of the new program into ongoing
limited (4, 5). organizational systems.”
In replicating an evidence-based program, organizations need Scheirer’s framework for program sustainability posits four
adequate knowledge and skills in adapting the program to fit local phases of program adoption: (1) initiation; (2) implementation;
circumstances while maintaining fidelity and evaluating the pro- (3) level of use (full or partial); and (4) sustainability (sustained,
gram to assure that it achieves the outcomes promised in the discontinued, or replaced) (15). Based on her literature review,
original research (6–8). Much of the extant literature outlines the agencies that sustain new programs likely agree that the program
many challenges of translating scientific knowledge to commu- can be modified to fit their organization, see the program as fit-
nity practice (9–13). These include: (1) resistance to new practice ting their organization’s mission, perceive program benefits, have
modalities; (2) lack of organizational buy-in; (3) lack of specific champions for the program, and have access to technical assistance
goals and standards in translating the evidence; and (4) rigidity while adopting the program.
of evidence-based practice that cannot be molded to meet specific The purpose of this study was to describe and determine the
needs of the applied setting or target population (4, 6, 14). important factors that supported or hindered sustainability of the
These adaptation barriers also influence the long-term contin- Stanford’s Chronic Disease Self-Management Program (CDSMP)
uation of evidence-based programs, and more attention is being among eldercare service providers in Hawai‘i. CDSMP was devel-
focused on ways to assure widespread availability of evidence- oped to empower people with various chronic diseases to take
based programs (15–18). Useful models such as RE-AIM (Reach, control of their health (23). Participants attend six 2.5-h sessions

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 140 | 158
Tomioka and Braun Examining sustainability factors for organizations

(one per week). Facilitators share knowledge and use motivational Honolulu County Area Agency on Aging than the other four (E, F,
interviewing techniques to engage participants, who make weekly G, and H), because they had been funded by the Honolulu County
action plans to help themselves take small steps toward chang- Area Agency on Aging for other programs. Five providers (A, B, C,
ing a behavior of their choice. Numerous studies of CDSMP have E, and G) were involved with HHAP during the statewide planning
shown that people who participate in this program feel better, of CDSMP adoption, whereas three providers (D, F, and H) joined
have better control over the symptoms of their chronic diseases, HHAP when CDSMP training was held. Each provider chose the
and are better able to talk to their physicians (23, 24). Although employee to be interviewed in this study.
the original test of CDSMP was conducted with Caucasians (25),
it has been successfully adapted to fit Hawai‘i’s multicultural MEASURES
communities (26). The investigators developed an interview guide that asked about
The implementation of CDSMP in Hawai‘i was supported by the five sustainability factors identified by Scheirer (15): (1) cham-
Hawai‘i Healthy Aging Partnership (HHAP), formed in 2003 to pions; (2) technical assistance; (3) perceived fit of the program; (4)
increase access to health promotion programs among Hawai‘i program modifiability; and (5) perceived benefits of the program.
older adults with chronic conditions. HHAP members include A sixth factor, readiness to replicate, was added, as it was consid-
professionals from government offices for aging and public health, ered important to HHAP partners. After discussing each factor,
elder care agencies, and the university. The process of CDSMP participants were asked to rate its importance to sustainability as
adaptation began in August 2006 when HHAP was awarded a 3- low, medium, or high. Finally, they are asked to identify other
year grant from the Administration on Aging (AoA). HHAP mem- sustainability factors (Table 1).
bers developed CDSMP implementation and evaluation plans, Readiness refers to an individual’s and agency’s sense of pre-
assessed readiness to implement CDSMP, and coordinated train- paredness to replicate the program. A champion is an agency
ing for CDSMP leaders. Implementation in two of Hawai‘i’s four employee who plays a key role in adapting, delivering, and/or sus-
counties began in July 2007. It was expanded statewide in 2008 taining CDSMP in the agency. Technical assistance refers to help
with funds from the National Council on Aging (NCOA). The sus- employees could access when they had questions or encountered
tainability phase began in 2009, when the original implementation problems in CDSMP implementation and sustainment. Perceived
funding ended. The purpose of this study, guided by Scheirer’s sus- fit of the program implies a similarity between the new interven-
tainability framework, was to better understand the process and tion and the parent organization’s mission and culture. Program
factors associated with sustainability of CDSMP in Hawai‘i. modifiability refers to the level of satisfaction that the agency has
with the modifications that it can make to the evidence-based pro-
MATERIALS AND METHODS gram (e.g., to better fit its clientele and agency structure) without
DESIGN, SETTINGS, PARTICIPANTS jeopardizing the behavior-changing components of the program.
Although CDSMP was implemented statewide, the findings Perceived benefits of the program include feelings of staff and
reported here were gathered as a part of the Honolulu case study, clients (which may or may not be based on evaluation data) that
which was conducted to understand the process of organizational the program is making a positive impact.
change for CDSMP adoption. In the Honolulu case study, data These factors were identified by Scheirer (15), with the excep-
from state and county government, community organizations, tion of readiness (item 1). Readiness as a sustainability factor was
and older adults were examined to investigate “how” and “why” identified through our review of the sustainability literature and
organizations in Honolulu adapted, implemented, and sustained discussion with funders, who considered organizational readiness
CDSMP successfully or unsuccessfully (27). The Honolulu case an important first step in adoption of CDSMP.
study was approved by University of Hawai‘i Institutional Review Although providers continue to offer CDSMP, data for this
Board. This paper reports on a portion of data collected, specifi- study of sustainability factors were collected in late 2010. At
cally from the eight service providers in Honolulu that started to that time, the first author (MT) conducted face-to-face, semi-
replicate CDSMP in 2007. structured interviews with the eight community provider repre-
For the Honolulu case study, we identified three phases of the sentatives in Honolulu. The interview questions were sent to the
CDSMP adoption path. The first 6 months of 2007 was considered representatives ahead of the interview to help them prepare. All
as the Initiation Phase, when HHAP began planning replication eight representatives provided a written consent. Interviews were
and training agency personnel in CDSMP delivery. The Deliv- held at the representatives’ offices and took about an hour. Six
ery Phase ran from June 2007 to June 2009, during which time individuals allowed their interviews to be audio taped and, for
staff members from multiple agencies were trained and then deliv- two, hand notes were made. All interviews were transcribed into
ered CDSMP to older adults and participated in ongoing fidelity text files.
monitoring and evaluation. The Sustainability Phase began in July
2009, when original funding ended. This paper reports on the ANALYSIS
Sustainability Phase of new-program adoption. Ratings of low, medium, or high were noted for each respon-
The eight providers included multi-purpose social service orga- dent for each sustainability factor. The discussion of each factor
nizations (designated as A and D), community health centers and the discussion of other possible factors were read indepen-
(designated as B, F, G, and H), a community college (designated dently by two researchers. For the most part, discussion of a priori
as E), and a community meals program (designated as C). Four sustainability factors served to give examples of, expand on, and
providers (A, B, C, and D) had a closer relationship with the contextualize each factor, which was useful in understanding its

www.frontiersin.org April 2015 | Volume 2 | Article 140 | 159

Tomioka and Braun Examining sustainability factors for organizations

Table 1 | Summary of sustainability factors assessed.

Sustainability factors Sample questions

1. Readiness Describe your “readiness” to replicate CDSMP. For example, how adequate was training in the program, data
collection, and program monitoring forms? How prepared was your agency?

2. Championsa Describe your experience with program champions for CDSMP? Who and how many people from your organization
were helping with CDSMP, and in what ways? What did these champions do? Comment on their effectiveness.

3. Technical assistancea How does your organization have access to technical assistance to sustain the program? Comment on the availability
and usefulness of technical assistance as you replicated CDSMP.

4. Program-organization fita How does CDSMP match your organization’s culture or mission? Comment on the level of “fit” between CDSMP and
your agency.

5. Program modifiabilitya Describe your ability to change or modify CDSMP that fit your clients and your agency. Describe your experience
making program modifications while trying to maintain fidelity to the original CDSMP design.

6. Perceived program benefitsa How did organizational leaders and worker feel CDSMP impacted your clients? How do you think CDSMP benefited
the people you served? In what ways has your involvement in CDSMP benefited clients, staff, and your organization?

7. Other (open-ended) How do you think CDSMP will be sustained by your agency? What are the major factors that contributed to long-term
sustainability?

a
Identified by Scheirer (15).

ranking (low, medium, or high). Analysis of responses to the (63%) who scored their organization as “high” had noted in their
open-ended question about other possible factors required the two discussion of readiness that their staff had spent time learning
researchers to discern themes in the data and then code responses about the concept of evidence-based programing prior to pro-
into themes. These were discussed in a meeting, and differences gram adoption, felt well trained in CDSMP and data collection,
were resolved by re-reading the interview transcript together and were motivated to pilot CDSMP in their community, had identified
having further discussion until consensus was reached. There were potential CDSMP participants, had established policies and pro-
no major disagreements during the analysis process. cedures related to CDSMP, and had purchased a CDSMP license.
The three providers who scored medium or low remembered some
RESULTS uncertainty within their organization and perhaps some miscom-
The progress of the eight providers who volunteered to repli- munication with HHAP as to the costs associated with CDSMP
cate CDSMP in Honolulu is shown in Figure 1. One provider licensing, the coordination of CDSMP workshops, and the need
(H) dropped out during the Initial Phase (the first 6 months) to participate in fidelity monitoring and evaluation.
because of the organization decided that it could not dedicate
staff time to deliver CDSMP. Thus, only seven of the eight orga- CHAMPIONS
nizations entered the 2-year Delivery Phase. Two providers (E and Respondents agreed that having champions was very important to
F) dropped out before the end of that phase. For Provider E, two sustainability, and the organizational representatives that reported
staff members completed CDSMP training and offered it twice high champion effectiveness were most likely to be from organiza-
in the community, but the organization felt that it was too time tions that sustained CDSMP. The transcript-analysis process, how-
consuming to recruit and track clients and chose to discontinue. ever, distinguished three types of champions, including program
Provider F was not able to fully implement the program with its champions, participant champions, and supervisor champions.
fidelity monitoring and evaluation requirements. Of the five enter- Respondents defined a program champion as someone who
ing the Sustainment Phase in mid-2009, only three providers (A, had been trained to lead CDSMP, had a passion for it, was very
B, D) were still sustaining CDSMP in late 2010. Provider G had committed, was able to promote it, and had the drive to expand
replaced CDSMP with another program, and Provider C discon- it. Although all organizations had staff trained in CDSMP, not all
tinued the program. Provider F decided to reengage with CDSMP could identify a program champion at the time of the interview,
at this time. while some agencies reported as many as three program champi-
During the analysis process, it became clear that organizations ons. The three sustained programs reported having more than one
that sustained CDSMP had more supports throughout the process program champion at their organization.
than the organizations that were unable to sustain CDSMP. Table 2 Interview findings suggested that the most successful program
shows a summary of rating results from the interview. champions had relatively flexible schedules, which allowed them
to offer CDSMP during or outside of work hours, and their job
READINESS descriptions included CDSMP. They had strong skills in team-
Thinking back to the Delivery Phase, providers described and rated work and took time to educate other branches of the organization
their level of readiness to replicate CDSMP. The five providers in CDSMP to increase organizational buy-in. They advocated for

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 140 | 160
Tomioka and Braun Examining sustainability factors for organizations

FIGURE 1 | Providers progress over the 4 years.

other staff in the agency to attend CDSMP leader training. They their story about CDSMP. By talking about the benefits they
networked with CDSMP leaders at other agencies, which helped received, they helped to recruit other participants for the program.
them find a substitute or second workshop leader when needed. Many providers felt that the word-of-mouth strategy was the most
They also were seen as role models by program participants and effective approach to attract new participants.
provided support to CDSMP leaders at other organizations. As one A supervisor champion was described as a manager who
provider noted: “Both champions have been enthusiastic and . . . try supported the delivery of CDSMP within the organization and
to promote, try to get more funding, try to get more people, are supported the program champions in leading CDSMP sessions.
involved . . . and they wish they were able to do more.” Supervisor champions always sent an agency representative to the
The two providers who reported no program champion might statewide HHAP meetings and saw CDSMP as a valuable service to
have had a program champion initially, but this person changed offer agency clients. Interview analysis suggested that all sustained
jobs or became too busy to lead and advocate for CDSMP. One agencies had a supervisor champions. Respondents for the other
provider said: “We don’t have a champion because we had to take four agencies felt that their supervisors were not strong champi-
care of other things. . .Champion requires that one person constantly ons. Two of them said that their supervisors oversaw a variety of
pushes CDSMP.” Sustained organizations reported that they would projects and remained relatively uninvolved with CDSMP repli-
be able to continue to sustain the program as long as they had lead- cation efforts. The other two providers reported that they did
ers and trainers on board. To facilitate this, HHAP continues to not get any support from their supervisors because their supervi-
provide CDSMP leader trainings and encourage organizations to sors had unrealistic expectations of CDSMP, for example, that it
continuously send new staff or volunteers to training so that they would take less time than it did, would be more modifiable that
could keep enough leaders within their organizations. it was, or would generate revenue. The discrepancy between the
A participant champion was described as someone who had expectations of these supervisors and those of the staff trained
graduated the CDSMP (attending four or more sessions out of in CDSMP discouraged staff from championing the program in
six), realized benefits from the program, and was willing to share their agencies.

www.frontiersin.org April 2015 | Volume 2 | Article 140 | 161

Tomioka and Braun Examining sustainability factors for organizations

assistance
PROGRAM-ORGANIZATION FIT
Technical

Five providers (63%) reported that CDSMP fit well with their

M
M
H
H
H
H
H
H
organizational mission and goals. They valued the program’s
concept of empowerment and its goal to help clients improve
health and maintain independence through self-management.
Perceived
benefits

They also felt that their participants would appreciate the pro-

M
M
–
H
H
H
H
H
gram’s motivational interviewing approach.
Two providers rated program-organizational fit as medium.
They felt that CDSMP fit their general mission – to improve
modifiability

the well-being of older adults – but not the type of service they
Program

provided, in this case meal delivery and college courses.

M
M
H
H
H
H
H
H

PROGRAM MODIFIABILITY
Five out of eight providers (63%) reported that the modifications
organization fit

of CDSMP by HHAP (with permission of Stanford) to fit Hawai‘i’s

multicultural population helped them attract enrollees. This was
Program-

H
H
H
H
L
L
L
L

done by including local examples and expressions of local culture

in the curriculum (26). One provider gave this example: “. . . our
participants are not fluent English speakers . . . so it takes double the
time to explain things . . .. We serve local food that they like or fit
Participants

with their culture, they feel happy even though they work hard dur-
Yes
Yes
Yes

No
No
No
No

ing the session. We also offer certificate of completion with leis.a little
champ

recognition for their hard work and they were very happy about
it.” Providers who reported high program modifiability also spent
time to develop local marketing tools that included pictures of
Supervisor

local older adults and symbols that resonated with Hawai‘i’s cul-
support

M
H
H
H
H
L
L

tures. The three providers who rated medium or low in program

Champions

modifiability noted disappointment that the program materials

were not available in the languages of their target groups (e.g., the
Effectiveness

various Pacific Islander languages, like Samoan, Chuukese, and

Marshallese) and that the program required two leaders to deliver
M
H
H
H
H
L

L
L

each workshop. They also reported that the structured outline and
scripted format of the workshop were too foreign for their clients.

PERCEIVED BENEFITS
No. of prog

One agency was not able to rate this factor because it dropped out
champ

2
3
2
1
0
1
0
0

prior to implementing CDSMP. Of the seven remaining providers,

five rated CDSMP benefit to clients as “high” and perceived
CDSMP was a good investment. They reported seeing improve-
Readiness

ments in the health of their clients and related success stories of

participants who had lost weight, were exercising more, and/or
M
H
H
H
H

L
L
L

were keeping better track of their health. One provider noted: “I

see them really change and keep hearing their stories . . . We had a
Table 2 | Summary of interview rating results.

quite a few that really struggled with making action plans. Boy, when
Org

D
G

H
B

C
A

E
F

they get it, they get it and they got so excited, you know, the first time
they come back, they were so proud that they got something accom-
H, high; M, medium; L, low; ?, unknown.
Dropped out in implementation phase

plished. I just think . . . you know, to me, it’s really had a big impact
on people’s lives.” They also saw benefits for their staff, many of
whom had incorporated CDSMP tools (e.g., problem solving and
Dropped out in initial phase

developing action plan) into their daily activities and had gained
confidence in public speaking. One provider used some of the
CDSMP tools for staff training. They also mentioned the benefit
of receiving evaluation results from HHAP specific to their agency
Discontinued

to share with supervisors and funders. Hearing good stories from

Reengaged
Sustained
Outcome

Replaced

their participants and seeing the positive evaluation data further

boosted their confidence in replicating CDSMP. The two providers
who were rating this factor as “medium” noted that some of the

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 140 | 162
Tomioka and Braun Examining sustainability factors for organizations

agency’s staff and clients were unable to grasp or apply CSDMP DISCUSSION
self-management strategies. Evidence-based health promotion programs are developed in
research settings, and replicating them in real-world settings can
TECHNICAL ASSISTANCE be challenging (4, 5). This study examined factors related to sus-
Six of the eight providers (63%) reported that the technical assis- tainability of CDSMP by Honolulu County providers a year after
tance that they received from HHAP was very useful. Techni- initial funding ended. Depending on presence and strength of
cal assistance was provided through HHAP’s monthly meetings, these factors, providers varied in their sustainment of CDSMP as
from the evaluation team, and from Stanford. At monthly meet- shown in Figure 1.
ings, providers were able to exchange ideas on how to leverage As Scheirer (15) proposed, new-program sustainability can
resources, to recruit and retain participants and leaders, and be enhanced by having many champions (and several types of
to carry-out the evaluation protocol. Meeting attendance also champions), ensuring program fit with the organizational mis-
helped providers to develop strong skills in working with other sion, allowing some modifications to the program so it can better
providers. These relationships were useful when they needed to fit clientele, seeing benefits of the program, and having access to
find a substitute leader for a CDSMP session or borrow books technical assistance. In addition to these five factors, this study
and other program materials for a workshop. A provider said: identified three more factors that appear to contribute to sustain-
“We are fortunate to have partners for CDSMP. I can call [Agency ability of CDSMP – readiness, access to additional funding, and
name], if I have questions. They also helped us and clarify things access to potential participants.
for us. One time, we did not have CDSMP books, and I called Readiness can be cultivated by providing training about
them . . . It was very helpful. “Some providers reported that the evidence-based programing, fidelity monitoring, and program
Stanford website and email listserv were useful and helped moti- evaluation, along with training in the intervention to be adopted
vate them to continue offering CDSMP. The providers who (13). For non-profit organizations, external funding to support
rated technical assistance as medium indicated that they wanted added programs is essential, especially in light of cutback to
more support to recruit participants and more clarification of social services and the piloting of the “reimbursement-for-service”
program requirements. model by many eldercare service providers. Although continued
funding was not cited explicitly by the majority of studies of
OTHER SUSTAINABILITY FACTORS program sustainability reviewed by Scheirer (15), she noted that
Interviewees were asked to identify other factors related to sus- many of these programs had in fact found alternative funding to
tainability. Two were identified – having access to potential par- maintain new programs.
ticipants (four providers) and having access to additional funding Exhausting potential clients can occur in agencies that work
(seven providers). within small geographic areas, have a fixed number of clients
Because most of the providers were offering CDSMP to their that they are allowed to serve, and/or have low client turnover
existing clientele, they initially did not encounter problems finding (28). This is especially true in Hawai‘i, where providers receiving
participants. However, at some point they had provided CDSMP funds through the AAAs are contracted to serve a specified num-
to all willing clientele. For the most part, agency clients were willing ber elders, often in a defined target area. With low turnover in
to participate in the program, and attendance was high (26), but clientele, all willing participants can participate in CDSMP over
“once we went through all of the participants, then we did not have the course of several years. Also, some service providers in Hawai‘i
any more new participants . . . we do not have a large turnover in our serve elders who speak languages for which CDSMP is not avail-
clients, so we could not expand our numbers.” Provider B noted that able. It may be that CDSMP is more sustainable in Hawai‘i’s health
their organization was established to serve a specific community. maintenance organizations that serve thousands of clients, as a
After all willing clients completed CDSMP, their organization had portion of their clients would likely be diagnosed with chronic
to consider the advantages and disadvantages to enrolling people disease each year. Meanwhile, HHAP members have expressed a
from other communities. Provider A felt that the CDSMP was desire to learn about and replicate other evidence-based programs
important enough to continue because it fit so well with their pro- that could benefit their clientele. Already, a number of providers
gram. This provider’s solution was to conduct CDSMP workshop in the state are replicating EnhanceFitness with good success (29).
with a combination of participants who have done it before and This study explored CDSMP sustainability among eight elder-
any new participants that they could find. care providers in one of Hawai‘i’s four counties, and only one
Additional funding was also a critical factor to sustain CDSMP. representative from each organization was interviewed. Although
One provider stated “in the long-term, always funding is needed, organizations selected to be interviewed the individual most
things cost money and staff time, plus money for licensing [and] closely involved in CDSMP adaptation, the interviewee may not
when the new books come out . . . all of those things cost something.” represent the whole organization.
Although most of the providers were able to access additional Also, because the interview asked about sustainability after
funding through the HHAP’s awarded grants, sustained organiza- initial funding ended, the results may have been compromised
tions’interviewees also wrote proposals to other funders to support by inability to recall events, especially for those organizations
CDSMP. They also used cost-saving strategies, such as holding that discontinued CDSMP, and by social desirability bias. Future
workshops at no-cost sites and creating a library of workshop examination of new-program sustainability would benefit from
materials that could be loaned to (rather than purchased for or prospective study and inclusion of multiple representatives of an
by) participants. organization.

www.frontiersin.org April 2015 | Volume 2 | Article 140 | 163

Tomioka and Braun Examining sustainability factors for organizations

Despite the limitations, this study was able to confirm the 14. Grol R. Successes and failures in the implementation of evidence-based
importance of the sustainability factors proposed by Scheirer (15), guidelines for clinical practice. Med Care (2001) 39(8):II.46–54. doi:10.1097/
00005650-200108002-00003
and added three more, which may be specific to the Hawai‘i
15. Scheirer MA. Is sustainability possible?: a review and commentary on empirical
context of CDSMP replication. The clear message from this study studies of program sustainability. Am J Eval (2005) 26(3):320–47. doi:10.1177/
is that planning for sustainability should start before replicat- 1098214005278752
ing evidence-based programs. It requires tremendous effort to 16. Scheirer MA, Dearing JW. An agenda for research on the sustainability of pub-
translate evidence-based programs, to build provider capacity, to lic health programs. Am J Public Health (2011) 101(11):2059–67. doi:10.2105/
AJPH.2011.300193
implement a new program (or new practice) in real-world set-
17. Shediac-Rizkallah MC, Bone LR. Planning for the sustainability of community-
ting, and to sustain it. These findings can help guide healthcare based health programs: conceptual frameworks and future directions for
workers and organizations to plan and sustain the adoption of research, practice and policy. Health Educ Res (1998) 13(1):87–108. doi:10.1093/
evidence-based programs. her/13.1.87
18. Wiltsey-Stirman SW, Kimberly J, Cook N, Calloway A, Castro F, Charns M.
The sustainability of new programs and innovations: a review of the empiri-
ACKNOWLEDGMENTS
cal literature and recommendations for future research. Implement Sci (2012)
This study was based on the doctoral thesis of the first author. We 7(17):1–19. doi:10.1186/1748-5908-7-17
acknowledge the Hawai‘i Healthy Aging Partnership, a coalition 19. Ory MG, Evashwick CJ, Glasgow RB, Sharkey JR. Pushing the Boundaries of
of the Hawai‘i Executive Office on Aging, the Area Agencies on Evidence-Based Research: Enhancing the Application and Sustainability of Health
Aging, the Hawai‘i State Department of Health, the University of Promotion Programs in Diverse Populations. In Behavioral change: An Evidence-
Based Handbook For Social and Public Health. Edinburgh: Elsevier (2005). p.
Hawai‘i, and service providers dedicated to expand health promo- 267–93.
tion options for older adults in Hawai‘i. Funding for the Hawai‘i 20. Akerlund KM. Prevention program sustainability: the state’s perspective. J Com-
Health Aging Partnership has been received from U.S. Adminis- munity Psychol (2000) 28(3):353–62. doi:10.1002/(SICI)1520-6629(200005)28:
tration on Aging, National Council on Aging, State of Hawai‘i, 3<353:AID-JCOP9>3.0.CO;2-6
County of Kaua’i, and County of Maui. 21. Johnson K, Hays C, Center H, Daley C. Building capacity and sustainable preven-
tion innovations: a sustainability planning model. Eval Program Plann (2004)
27(2):135–49. doi:10.1016/j.evalprogplan.2004.01.002
REFERENCES 22. Mancini JA, Marek LI. Sustaining community-based programs for families: con-
1. Altpeter M, Bryant L, Schneider E, Whitelaw N. Evidence-based health practice: ceptualization and measurement. Fam Relat (2004) 53(4):339–47. doi:10.1111/
knowing and using what works for older adults. Home Health Care Serv Q (2006) j.0197-6664.2004.00040.x
25(1–2):1–11. doi:10.1300/J027v25n01_01 23. Stanford Patient Education Research Center. Selected Publications [Internet].
2. Brownson RC, Baker EA, Leet TL, Gillespie KN, True WR. Evidence-Based Public Palo Alto, CA: School of Medicine Patient Education Research Center (2013).
Health. 2nd ed. Oxford, UK: Oxford University Press (2011). 293 p. Available from: http://patienteducation.stanford.edu/bibliog.html
3. U.S. Administration on Aging. Facts AoA Evidence-Based Prevention Program. 24. Ory MG, Smith ML, Kulinski KP, Lorig K, Zenker W, Whitelaw N. Self-
Washington, DC: U.S. Department of Health and Human Services, Adminis- management at the tipping point: reaching 100,000 Americans with evidence-
tration on Aging (2010). Available from: http://www.aoa.gov/AoARoot/Press_ based programs. J Am Geriatr Soc (2013) 61(5):821–3. doi:10.1111/jgs.12239
Room/Products_Materials/pdf/fs_EvidenceBased.pdf 25. Lorig KR, Sobel DS, Stewart AL, Brown BW Jr, Bandura A, Ritter P, et al. Evidence
4. Glasgow RE, Lichtenstein E, Marcus AC. Why don’t we see more translation of suggesting that a chronic disease self-management program can improve health
health promotion research to practice? Rethinking the efficacy-to-effectiveness status while reducing hospitalization: a randomized trial. Med Care (1999)
transition. Am J Public Health (2003) 93(8):1261–7. doi:10.2105/AJPH.93.8. 37(1):5–14. doi:10.1097/00005650-199901000-00003
1261 26. Tomioka M, Braun KL, Compton M, Tanoue L. Adapting Stanford’s chronic dis-
5. Green LW, Glasgow RE. Evaluating the relevance, generalization, and applicabil- ease self-management program to Hawaii’s multicultural populations. Geron-
ity of research: issues in external validation and translation methodology. Eval tologist (2012) 52(1):121–32. doi:10.1093/geront/gnr054.U.S
Health Prof (2006) 29(1):126–53. doi:10.1177/0163278705284445 27. Tomioka M. Organizational Change in the Aging Network to Whom it May Con-
6. Alkema GE, Frey D. Implications of translating research into practice: a medica- cern: Promote Evidence-Based Programs [Dissertation]. Honolulu (HI): Univer-
tion management intervention. Home Health Care Serv Q (2006) 25(1–2):33–54. sity of Hawaii (2014). 186 p.
doi:10.1300/J027v25n01_03 28. Freedman AM, Kuester SA, Jernigan J. Evaluating public health resources:
7. Glasgow RE, Klesges LM, Dzewaltowski DA, Bull SS, Estabrooks P. The future what happens when funding disappears? Prev Chronic Dis (2013) 10:130130.
of health behavior change research: what is needed to improve translation of doi:10.5888/pcd10.130130
research into health promotion practice? Ann Behav Med (2004) 27(1):3–12. 29. Tomioka M, Sugihara N, Braun KL. Replicating the enhance fitness physical
doi:10.1207/s15324796abm2701_2 activity program in Hawai‘i’s multicultural population: 2007-2010. Prev Chronic
8. Dusenbury L, Brannigan R, Falco M, Hansen W. A review of research on fidelity Dis (2012) 9:110155. doi:10.5888/pcd9.110155
of implementation: implications for drug abuse prevention in school settings.
Health Educ Res (2003) 18(2):237–56. doi:10.1093/her/18.2.237 Conflict of Interest Statement: The authors declare that the research was conducted
9. Doig GS, Simpson F. Efficient literature searching: a core skill for the prac- in the absence of any commercial or financial relationships that could be construed
tice of evidence-based medicine. Intensive Care Med (2003) 29(12):2119–27. as a potential conflict of interest.
doi:10.1007/s00134-003-1942-5
10. Dunn AL, Blair SN. Translating evidenced-based physical activity interven- This paper is included in the Research Topic, “Evidence-Based Programming for Older
tions into practice: the 2010 challenge. Am J Prev Med (2002) 22(4):8–9. Adults.” This Research Topic received partial funding from multiple government and
doi:10.1016/S0749-3797(02)00432-4 private organizations/agencies; however, the views, findings, and conclusions in these
11. Elliott DS, Mihalic S. Issues in disseminating and replicating effective preven- articles are those of the authors and do not necessarily represent the official position
tion programs. Prev Sci (2004) 5(1):47–53. doi:10.1023/B:PREV.0000013981. of these organizations/agencies. All papers published in the Research Topic received
28071.52 peer review from members of the Frontiers in Public Health (Public Health Education
12. Goodman RM. Community-based participatory research: questions and chal- and Promotion section) panel of Review Editors. Because this Research Topic repre-
lenges to an essential approach. J Public Health Manag Pract (2001) 7(5): sents work closely associated with a nationwide evidence-based movement in the US,
v–vi. many of the authors and/or Review Editors may have worked together previously in
13. Simpson DD, Flynn PM. Moving innovations into treatment: a stage-based some fashion. Review Editors were purposively selected based on their expertise with
approach to program change. J Subst Abuse Treat (2007) 33(2):111–20. doi:10. evaluation and/or evidence-based programming for older adults. Review Editors were
1016/j.jsat.2006.12.023 independent of named authors on any given article published in this volume.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 140 | 164
Tomioka and Braun Examining sustainability factors for organizations

Received: 15 June 2014; accepted: 28 August 2014; published online: 27 April 2015. Copyright © 2015 Tomioka and Braun. This is an open-access article distributed under
Citation: Tomioka M and Braun KL (2015) Examining sustainability factors for orga- the terms of the Creative Commons Attribution License (CC BY). The use, distribution
nizations that adopted Stanford’s Chronic Disease Self-Management Program. Front. or reproduction in other forums is permitted, provided the original author(s) or licensor
Public Health 2:140. doi: 10.3389/fpubh.2014.00140 are credited and that the original publication in this journal is cited, in accordance with
This article was submitted to Public Health Education and Promotion, a section of the accepted academic practice. No use, distribution or reproduction is permitted which
journal Frontiers in Public Health. does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 140 | 165

 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00179

Chronic Disease Self-Management Program in the

workplace: opportunities for health improvement
Matthew Lee Smith 1 *, Mark G. Wilson 1 , David M. DeJoy 1 , Heather Padilla 1 , Heather Zuercher 1 ,
Phaedra Corso 2 , Robert Vandenberg 3 , Kate Lorig 4 and Marcia G. Ory 5
1
Workplace Health Group, Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA, USA
2
Department of Health Policy and Management, College of Public Health, The University of Georgia, Athens, GA, USA
3
Department of Management, Terry College of Business, The University of Georgia, Athens, GA, USA
4
Stanford Patient Education Research Center, Department of Medicine, Stanford School of Medicine, Palo Alto, CA, USA
5
Department of Health Promotion and Community Health Sciences, Texas A&M Health Science Center School of Public Health, College Station, TX, USA

Edited by: Disease management is becoming increasingly important in workplace health promotion
Harshad Thakur, Tata Institute of
given the aging workforce, rising chronic disease prevalence, and needs to maintain a pro-
Social Sciences, India
ductive and competitive American workforce. Despite the widespread availability of the
Reviewed by:
Xiaoyun Liu, Peking University, China Chronic Disease Self-Management Program (CDSMP), and its known health-related bene-
Nilesh Chandrakant Gawde, Tata fits, program adoption remains low in workplace settings.The primary purpose of this study
Institute of Social Sciences, India is to compare personal and delivery characteristics of adults who attended CDSMP in the
*Correspondence: workplace relative to other settings (e.g., senior centers, healthcare organizations, residen-
Matthew Lee Smith, Department of
tial facilities). This study also contrasts characteristics of CDSMP workplace participants to
Health Promotion and Behavior,
College of Public Health, The those of the greater United States workforce and provides recommendations for translat-
University of Georgia, 330 River ing CDSMP for use in workplace settings. Data were analyzed from 25,664 adults collected
Road, 315 Ramsey Center, Athens, during a national dissemination of CDSMP. Only states and territories that conducted work-
GA 30602, USA
shops in workplace settings were included in analyses (n = 13 states and Puerto Rico).
e-mail: [email protected]
Chi-squared tests and t -tests were used to compare CDSMP participant characteristics by
delivery site type. CDSMP workplace participant characteristics were then compared to
reports from the United States Bureau of Labor Statistics. Of the 25,664 CDSMP partici-
pants in this study, 1.7% (n = 435) participated in workshops hosted in worksite settings.
Compared to CDSMP participants in non-workplace settings, workplace setting partici-
pants were significantly younger and had fewer chronic conditions. Differences were also
observed based on chronic disease types. On average, CDSMP workshops in workplace
settings had smaller class sizes and workplace setting participants attended more work-
shop sessions. CDSMP participants in workplace settings were substantially older and a
larger proportion were female than the general United States workforce. Findings indicate
opportunities to translate CDSMP for use in the workplace to reach new target audiences.
Keywords: chronic disease self-management, evidence-based program, workplace wellness, evaluation, transla-
tional research

INTRODUCTION 77% of older adults currently have two or more comorbidities

Chronic diseases are multi-dimensional and affect all aspects of (5). Recognizing the growing prevalence of obesity and other car-
people’s lives, especially work (1, 2). People with chronic diseases diovascular risk factors (e.g., hypertension, hypercholesterolemia,
are constantly required to make decisions that affect their health, diabetes) and related chronic conditions among working-aged
which have ramifications for work performance and employabil- Americans (6), interventions are needed to arm middle-aged and
ity. It has been reported that, depending on the chronic condition older employees with skills and strategies to manage their diseases
involved, between 22 and 49% of employees experience difficul- and associated symptoms.
ties meeting physical work demands, while between 27 and 58% Disease management is increasingly recognized as an important
have problems meeting psychosocial work requirements (3). These component of workplace health promotion given our aging work-
problems can lead to job loss or premature departure from the force, the prevalence of chronic conditions, and the importance of
workforce. maintaining a productive and competitive American workforce
To complicate matters, the American workforce is aging. The (7–11). Currently, most workplace-based disease management
Bureau of Labor Statistics (BLS) predicts that between 2006 and programs are offered by health insurance providers and operate
2016, the number of workers aged 65–74 years will increase by largely independent of other on-site health promotion activities
83%, and those aged 75 and older will increase by 84% (4). Many (2). Mounting evidence supports the effectiveness and growing
of these older workers will have one or more chronic diseases; importance of disease management programing in workplace

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 179 | 166
Smith et al. CDSMP in the workplace

settings (12–15). Unfortunately, workplace-delivered disease man- DATA SOURCE AND STUDY POPULATION
agement activities may have limited reach because they are expen- Cross-sectional data for this study were obtained from a nation-
sive and require medically trained providers/facilitators (2). These wide delivery of CDSMP as part of the American Recovery and
activities may also have narrow scopes (i.e., focus on one specific Reinvestment Act of 2009 (i.e., Recovery Act) Communities Putting
disease or condition). Prevention to Work: Chronic Disease Self-Management Program ini-
Stanford’s Chronic Disease Self-Management Program tiative (16). The United States Administration on Aging led this
(CDSMP) is among the most widely disseminated and researched initiative in collaboration with the Centers for Disease Control
evidence-based programs (16, 17) and is extremely effective in and Prevention (CDC) and the Centers for Medicare and Medic-
helping individuals better manage their chronic disease and related aid Services to support the delivery of CDSMP in 45 states, Puerto
complications (18–20). Developed based on over 20 years of Rico, and the District of Columbia (17). This initiative was origi-
research at Stanford University, CDSMP is currently offered in nally designed to have 50,000 Americans complete at least 4 out of
over 30 countries and a variety of languages1 . Traditionally deliv- 6 CDSMP sessions between 2010 and 2012 and to embed CDSMP
ered through the aging services network, this robust program has delivery structures into statewide systems (16).
been delivered in a wide variety of community settings (e.g., senior For this study, data were analyzed from 25,664 participants
centers, healthcare organizations, residential facilities, faith-based who attended CDSMP workshops in 13 states and 1 territory
organizations, and tribal centers). CDSMP has the advantages of that delivered the program in workplace settings (i.e., to reduce
being inexpensive and easily disseminated. It is not disease or con- threats for systematic bias associated with state-specific delivery
dition specific and can be delivered effectively by lay leaders with infrastructures or preferences) and had no missing data for vari-
minimal training using a train-the-trainer model. ables of interest. We also utilized 2012 BLS data from the United
To date, CDSMP has not been widely implemented by work- States Department of Labor to compare CDSMP participant
place settings or incorporated into workplace health promotion characteristics to those of the larger American workforce2 .
programing (21). In part, low-implementation rates in workplace
settings may be a consequence of CDSMP being primarily deliv- MEASURES
ered through the aging services network, which predominately tar- The primary variable of interest in this study was whether or
gets older adults, many of whom may no longer in the workforce. not CDSMP participants attended program workshops in work-
Additionally, the standard CDSMP structure and format (i.e., 2.5 h place settings. Data from states that did not deliver one or more
sessions, once a week for six consecutive weeks) may not appear CDSMP workshops were omitted from study analyses. Among
amenable to widespread implementation in work organizations. included states, workshop delivery site type was dichotomized into
For these reasons, it is important to investigate characteristics of worksite settings and non-worksite settings. Non-worksite settings
CDSMP uptake in workplace settings and explore opportunities included senior centers, area agencies on aging (AAA), healthcare
for reaching and better serving the American workforce. organizations, residential facilities, community or multipurpose
Using data from the first 100,000 participants collected during centers, faith-based organizations, educational institutions, and
a 2-year national dissemination of CDSMP, the primary purpose tribal centers. Other workshop-level variables of interest included
of this study was to compare personal and delivery characteristics the number of participants enrolled in the workshop (i.e., contin-
of adults who attended CDSMP in the workplace relative to other uous number ranging from 1 to 20 individuals) and the number
settings (e.g., senior centers, healthcare organizations, residential of workshop sessions attended (i.e., “successful completion” is
facilities). To contextualize CDSMP implementation in workplace defined as attending 4 or more of the 6 possible sessions) (20).
settings, this study also contrasts characteristics of CDSMP work- Participant characteristics of interest in this study included age
place participants relative to those of the greater United States (i.e., measured continuously in years as well as categories con-
workforce. Building upon these findings, we highlight potential sistent with those reported by the United States Department of
opportunities for translating CDSMP for use in workplace settings Labor), sex, ethnicity, and race. Rural–urban commuting area
to overcome traditional barriers, reach new customer markets, codes based on participants’ ZIP were used to categorize partic-
and improve work performance indicators while maintaining the ipants’ residence (metro vs. non-metro). The number and type
program’s well-documented effectiveness. of self-reported chronic conditions was also recorded (i.e., arthri-
tis, cancer, depression, diabetes, heart disease, hypertension, lung
MATERIALS AND METHODS disease, stroke, osteoporosis, and other chronic conditions).
PROGRAM DESCRIPTION
The CDSMP has been introduced and disseminated in the United
ANALYSES
States as a method to empower patients with self-management
To compare the characteristics of the participants who attended
skills to deal with their chronic conditions (22). Drawing upon
CDSMP workshops in workplace settings and those who partic-
social learning theory (23), CDSMP is an evidence-based, peer-led
ipated in other settings, we used chi-square tests for categorical
intervention consisting of six highly participative classes held for
variables and independent-sample t -tests for continuous vari-
2.5 h each, once a week, for six consecutive weeks (22). CDSMP
ables. Only data for the following states and territories were
has resulted in improved healthcare and health (18, 20), while
included in analyses: Alabama, Arizona, Florida, Hawaii, Maine,
potentially saving healthcare costs (19).

1 http://patienteducation.stanford.edu/programs/cdsmp.html 2 http://www.bls.gov/data/#employment

www.frontiersin.org April 2015 | Volume 2 | Article 179 | 167

Smith et al. CDSMP in the workplace

Maryland, Massachusetts, Missouri, New Jersey, North Carolina, average 6.7 years younger); however, they were substantially older
Oregon, Puerto Rico, Virginia, and Wisconsin. Statistical analyses than the American workforce (i.e., 5.2% of the 2012 workforce
for this descriptive study were performed using SPSS (version 21). aged 65 and older compared to 51.4% of CDSMP workplace
setting participants). Further, when compared to the American
RESULTS workforce, males and Hispanics were underrepresented in CDSMP
CDSMP PARTICIPATION IN WORKPLACE SETTINGS workplace settings. This finding highlights potential opportuni-
Of the 25,664 participants who attended CDSMP workshops in ties to expand program reach to new target audiences. When
these 13 states and 1 territory, 435 (1.70%) did so at a workplace CDSMP was delivered in workplace settings, a significantly larger
setting. As seen in Table 1, the average age of workplace setting par- proportion of participants successfully completed the workshop
ticipants was 61.12 (±14.69) years. The majority of workplace set- relative to those in non-workplace settings (i.e., 85.1% compared
ting participants was under age 65 years (58.6%), female (80.9%), to 78.1%). Clearly there is potential to expand CDSMP reach and
non-Hispanic (96.1%), and white (61.6%). Over 63% reported adoption among the American workforce.
living in metro areas. On average, workplace setting participants
self-reported having 2.29 (±1.50) chronic conditions, with 41.4% OPPORTUNITIES TO TRANSLATE CDSMP FOR USE IN THE WORKPLACE
reporting 3 or more coexisting conditions. The most frequently Agencies such as the CDC are promoting coordinated approaches
reported chronic conditions were hypertension (43.9%), arthri- to workplace health that encompass interventions to address the
tis (40.0%), diabetes (28.0%), and depression (22.5%). Almost multi-factorial influences of health risk and employee wellness
35% of participants reported some other chronic condition. On (25). Aligned with CDC’s goal of increasing the number of science-
average, CDSMP workshop held in workplace settings had 10.82 based initiatives in worksites (26), implementing and evaluating
(±3.89) participants. On average, these participants attended 4.84 CDSMP in workplace settings is a viable strategy to improve
(±1.46) of the 6 workshop sessions, with 85.1% successfully employee health using a proven evidence-based intervention. Even
completing the workshop. though there is considerable need in worksite health promotion
Compared to CDSMP participants in non-workplace settings, for efficacious disease management programs, CDSMP has not
workplace setting participants were significantly younger and had been tested in a format conducive for broad-based worksite dis-
fewer chronic conditions. A significantly smaller proportion of semination. If CDSMP were appropriately tailored to the needs
workplace setting participants had arthritis, hypertension, stroke, of middle-aged and older workers and delivered through work-
whereas a significantly larger proportion of these participants had place settings, this translated version would have potential to
other chronic condition types. A significantly larger proportion reduce healthcare utilization and boost work productivity and
of participants in workplace settings were non-Hispanic and non- retention. This combination of benefits coupled with relatively
white, although a significantly larger proportion of non-workplace low-delivery costs and scalability should be attractive to almost
setting participants were African American. On average, CDSMP all employers and employer groups (i.e., leverage for making “a
workshops in workplace settings had significantly fewer partici- business case” to adopt CDSMP). Further, this model has poten-
pants, and participants in workplace settings attended significantly tial to be extremely cost-effective and yield substantial returns on
more workshop sessions. investment.
Compared to 2012 estimates from the United States Depart-
ment of Labor, a larger proportion of CDSMP participants in UNIQUE NATURE OF WORKSITES
workplace settings were over age 50 years, female, non-Hispanic, Although there is considerable potential for offering CDSMP in
and non-white. workplaces, in order to maximize program effectiveness, it needs
to be translated to accommodate the unique nature of worksite
DISCUSSION settings. Generally speaking, the typical worker is paid a certain
Findings from this descriptive study indicate that CDSMP adop- amount of money, to work a defined period of time, to accomplish
tion is low in the workplace, with merely 1.7% of participants in specific tasks or outcomes that will benefit the organization’s goals
this sample attending workshops in workplace settings. Over 66% and enable them to support themselves and their families. Orga-
of workplace setting participants had two or more chronic condi- nizations, in turn, are focused on maximizing the outcomes and
tions, which indicates the need for a widely available, high-quality minimizing the costs to achieve those outcomes, most of which
disease self-management intervention. While significantly smaller are driven by the people, environment, and materials needed to
proportions of workplace setting participants had arthritis, hyper- produce the product or outcome. So, both the worker and orga-
tension, and stroke relative to non-workplace setting participants nization have a strong economic incentive, time constraints, and
(conditions more prevalent in the older adult population), it is interrelated goals and/or outcomes that are restricted by the envi-
interesting that rates of self-reported chronic conditions among ronment in which they operate. These factors vary from organi-
workplace setting participants were generally comparable to rates zation to organization and job to job. As a result, any intervention
among non-workplace setting participants. This aligns with pre- implemented in worksites must be tailored to these unique charac-
vious reports indicating the American workforce is developing teristics. For CDSMP to be effective in worksite settings, it must be
chronic conditions and accruing more comorbidities during their cost-effective, not too disruptive of work schedules, and achieve
extended time on the job before their delayed retirement (24). varying work-related outcomes (both individual and organiza-
Compared to non-workplace setting participants in this study, tional). And, most importantly, it must do so within the constraints
workplace setting participants were significantly younger (i.e., on of the workplace environment.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 179 | 168
Smith et al. CDSMP in the workplace

Table 1 | National sample characteristics by CDSMP delivery site type.

U.S. Workforce Total CDSMP Worksite setting Non-worksite χ2 or t P

Statistics (2012)a participants (n = 435) setting (n = 25,229)
(n = 25,664)

Age – 67.70 (±14.35) 61.12 (±14.69) 67.82 (±14.31) 9.67 <0.001

Under 50 66.8% 11.2% 20.0% 11.0% 108.82 <0.001
50–64 28.0% 23.9% 38.6% 23.7%
6574 4.2% 30.2% 20.0% 30.4%
75+ 1.0% 34.7% 21.4% 35.0%
Sex
Male 53.0% 21.2% 19.1% 21.2% 1.14 0.286
Female 47.0% 78.8% 80.9% 78.8%

Hispanic ethnicity
No – 89.0% 96.1% 88.9% 22.70 <0.001
Yes 15.4% 11.0% 3.9% 11.1%
Race
White 80.5% 66.7% 61.6% 66.8% 85.66 <0.001
African American 11.2% 20.9% 14.0% 21.0%
Asian/Pacific Islander 5.5% 4.7% 12.9% 4.5%
American Indian/Alaska naive – 1.0% 2.1% 0.9%
Other multiple races – 6.8% 9.4% 6.7%
Number of chronic conditions _ 2.50 (±1.65) 2.29 (±1.50) 2.50 (±1.65) 2.90 0.004
0 – 9.2% 9.2% 9.2% 3.94 0.268
1 – 21.4% 24.6% 21.3%
2 – 23.9% 24.8% 23.9%
3+ – 45.5% 41.4% 45.6%

Participant residence
Metro – 78.5% 63.4% 78.7% 59.01 <0.001
Non-metro – 21.5% 36.6% 21.3%
Number of participant enrolled in workshop – 12.59 (±4.03) 10.82 (±3.89) 12.62 (±4.03) 9.23 <0.001

Number of sessions attended – 4.53 (±1.63) 4.84 (±1.46) 4.53 (±1.63) −4.48 <0.001
Successful completion: no – 21.8% 14.9% 21.8% 12.16 <0.001
Successful completion: yes – 78.2% 85.1% 78.1%

Disease prevalence
Arthritis – 47.1% 40.0% 47.2% 8.99 0.003
Cancer – 10.9% 12.0% 10.9% 0.49 0.486
Depression – 23.1% 22.5% 23.1% 0.09 0.770
Diabetes – 32.1% 28.0% 32.1% 3.29 0.070
Heart disease – 18.6% 16.6% 18.6% 1.18 0.277
Hypertension – 48.7% 43.9% 48.8% 4.12 0.042
Lung disease – 19.1% 16.1% 19.1% 2.51 0.113
Stroke – 5.4% 3.2% 5.4% 4.12 0.042
Osteoporosis – 14.6% 12.2% 14.7% 2.10 0.148
Other – 30.3% 34.7% 30.2% 4.07 0.044

a
Unadjusted estimates of employed persons (U.S. Bureau of Labor Statistics; http://www.bls.gov/data/#employment).
–, Workforce data not available for comparison purposes.

CDSMP TRANSLATION modifying the session length (and thereby increasing the number
As with any translation, it is imperative to maintain the program’s of sessions) and incorporating worksite-specific strategies. This
integrity, which assures that the original intervention effects will translation should also include efforts to complement existing
be achieved. This could be accomplished by keeping the content, workshop content to include topics, skills, and examples more
program duration, and number of contact hours constant, but relevant to working-aged individuals. It would afford researchers

www.frontiersin.org April 2015 | Volume 2 | Article 179 | 169

Smith et al. CDSMP in the workplace

and evaluators an opportunity to introduce and assess new out- Disease Self-Management Program initiative, led by the United
comes measures related to work performance and productivity. States Administration on Aging in collaboration with the Centers
Such modifications may overcome barriers to workplace adop- for Disease Control and Prevention and the Centers for Medicare
tion as well as foster more universal cross-industry appeal in and Medicaid Services, allotted $32.5 million to support the trans-
small rural jobsites and Fortune 500 companies alike. In the event lation of the Stanford program in 45 States, Puerto Rico, and the
that a workplace-based CDSMP were created, the new interven- District of Columbia. The National Council on Aging served as
tion would need to be standardized (as with any evidence-based the Technical Assistance Resource Center for this initiative and
program), with careful attention given to implementation man- collected de-identified data on program participation.
uals, fidelity standards, leader training (i.e., new and/or “bridge” Research reported in this publication was supported by the
trainings for T-trainers, master trainers, lay leaders), workshop National Heart, Lung, And Blood Institute of the National Insti-
materials, and evaluation tools and protocol. tutes of Health under Award Number R01HL122330. The content
is solely the responsibility of the authors and does not neces-
Translation benefits sarily represent the official views of the National Institutes of
A key advantage of offering CDSMP to working adults would be Health.
that the program could reach younger individuals and those who
are earlier in the time course of their chronic conditions, thereby
reducing the likelihood of costly and debilitating complications. A REFERENCES
1. Detaille SI, van der Gulden JWJ, Engels JA, Heerkens YF, van Dijk FJ. Using inter-
workplace-based CDSMP has the potential to reach large numbers
vention mapping (IM) to develop a self-management programme for employees
of working adults, which is not occurring through current work- with a chronic disease in the Netherlands. BMC Public Health (2010) 10:353.
place disease management delivery models or traditional CDSMP doi:10.1186/1471-2458-10-353
delivery channels. 2. Richling DE. Disease management for employed populations. 2nd ed. In: Pronk
Although delivered in workplace settings, a workplace-based NP, editor. ACSM’s Worksite Health Handbook: A Guide to Building Healthy and
Productive Companies. Champaign, IL: Human Kinetics (2009). p. 286–95.
CDSMP would benefit from ongoing collaboration with the aging 3. Lerner D, Allaire S, Reisine S. Work disability resulting from chronic health
services network and local community-based organizations to conditions. J Occup Environ Med (2005) 47(3):253–64. doi:10.1097/01.jom.
ensure long-term program sustainability. Further, a workplace- 0000150206.04540.e7
based CDSMP would readily complement other existing on-site 4. Bureau of Labor Statistics. Older Workers [Internet]. (2008). Available from:
health promotion initiatives that target healthy lifestyle behaviors http://www.bls.gov/spotlight/2008/older_workers/
5. Agency for Health Research and Quality. Medical Expenditure Panel Survey
and healthy decision-making. Household Component: 2010 Full Year Consolidated Data File. Rockville, MD
(2010). Available from: http://meps.ahrq.gov/data_stats/download_data_files_
LIMITATIONS detail.jsp?cboPufNumber=HC-138
As with any study, there were limitations that should be addressed. 6. Bhattacharya J, Choudhry K, Lakdawalla D. Chronic disease and severe disability
Because of the grand scale nature of this national initiative, only among working-age populations. Med Care (2008) 46(1):92–100. doi:10.1097/
self-report sociodemographics and administrative records were MLR.0b013e3181484335
7. Goetzel RZ, Pei X, Tabrizi MJ, Henke RM, Kowlessar N, Nelson CF, et al.
collected. No outcome data were obtained. Of the 45 states and 2 Ten modifiable health risk factors are linked to more than one-fifth of
territories involved in this study, only 13 states and Puerto Rico employer-employee health care spending. Health Aff (2012) 31(11):2474–84.
offered workshops in workplace settings, thus, only data from these doi:10.1377/hlthaff.2011.0819
areas were included in study analyses to reduce possible bias when 8. Special Committee on Health, Productivity, and Disability Management.
making comparisons. However, because 2012 BLS data contained Healthy workforce/healthy economy: The role of health, productivity, and dis-
ability management in addressing the nation’s health care crisis. J Occup Environ
data from all states and territories, comparisons with CDSMP par- Med (2009) 51(1):114–9. doi:10.1097/JOM.0b013e318195dad2
ticipant data were less than ideal. Further, BLS data only contained 9. Collins JJ, Baase CM, Sharda CE, Ozminkowski RJ, Nicholson S, Billotti GM,
a few variables to which CDSMP participant data could be com- et al. The assessment of chronic health conditions on work performance,
pared (e.g., no information about rurality or disease diagnoses). absence, and total economic impact for employers. J Occup Environ Med (2005)
47(6):547–57. doi:10.1097/01.jom.0000166864.58664.29
This limited these authors’ ability to fully realize the aims of this
10. Hoffman C, Rice D, Sung H-Y. Persons with chronic conditions. Their preva-
study. lence and costs. J Am Med Assoc (1996) 276(18):1473–9. doi:10.1001/jama.276.
18.1473
CONCLUSION 11. Hwang W, Weller W, Ireys H, Anderson G. Out-of-pocket medical spending for
This study provides a unique glimpse into the under-explored care of chronic conditions. Health Aff (2001) 20(6):267–78. doi:10.1377/hlthaff.
realm of CDSMP delivered in workplace settings. Findings sug- 20.6.267
12. Nyman JA, Abraham JM, Jeffery MM, Barleen NA. The effectiveness of a health
gest considerable opportunities for translating CDSMP for use
promotion program after 3 years: evidence from the University of Minnesota.
in workplace settings to overcome traditional barriers, reach new Med Care (2012) 50(9):772–8. doi:10.1097/MLR.0b013e31825a8b1f
target audiences, and improve work performance indicators while 13. Caloyeras JP, Liu H, Exum E, Broderick M, Mattke S. Managing manifest dis-
maintaining the program’s effectiveness. While the recommenda- eases, but not health risks, saved PepsiCo money over seven years. Health Aff
tions put forth in this paper are those of the authors, additional (2014) 33(1):124–31. doi:10.1377/hlthaff.2013.0625
14. Varekamp I, Verbeek JH, van Dijk FJH. How can we help employees with chronic
workplace-based CDSMP translation efforts are inevitable.
diseases to stay at work? A review of interventions aimed at job retention and
based on an empowerment perspective. Int Arch Occup Environ Health (2006)
ACKNOWLEDGMENTS 80(2):87–97. doi:10.1007/s00420-006-0112-9
The American Recovery and Reinvestment Act of 2009 (i.e., 15. Pelletier K. A review and analysis of the clinical and cost-effectiveness studies
Recovery Act) Communities Putting Prevention to Work: Chronic of comprehensive health promotion and disease management programs at the

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 179 | 170
Smith et al. CDSMP in the workplace

worksite: update VI 2000-2004. J Occup Environ Med (2005) 47(10):1051–8. 26. Centers for Disease Control and Prevention [Internet]. About Work@Health
doi:10.1097/01.jom.0000174303.85442.bf (2014). Available from: http://www.cdc.gov/workathealth/about.html
16. Ory MG, Smith ML, Patton K, Lorig K, Zenker W, Whitelaw N. Self-management
at the tipping point: Reaching 100,000 Americans with evidence-based pro- Conflict of Interest Statement: The authors declare that the research was conducted
grams. J Am Geriatr Soc (2013) 61(5):821–3. doi:10.1111/jgs.12239 in the absence of any commercial or financial relationships that could be construed
17. ARRA. Communities Putting Prevention to Work: Chronic Disease Self- as a potential conflict of interest.
Management Program [Internet]. US Department of Health and Human Ser-
vices, Administration on Aging (2012). Available from: https://www.cfda.gov/ This paper is included in the Research Topic, “Evidence-Based Programming for Older
index?s=program&mode=form&tab=core&id=04e3dca0a3218583cc472c28f Adults.” This Research Topic received partial funding from multiple government and
5d26644 private organizations/agencies; however, the views, findings, and conclusions in these
18. Ory M, Ahn S, Jiang L, Lorig K, Ritter P, Laurent D, et al. National study articles are those of the authors and do not necessarily represent the official position
of chronic disease self-management. J Aging Health (2013) 25(7):1258–74. of these organizations/agencies. All papers published in the Research Topic received
doi:10.1177/0898264313502531 peer review from members of the Frontiers in Public Health (Public Health Education
19. Ahn S, Basu R, Smith ML, Jiang L, Lorig K, Whitelaw N, et al. The impact and Promotion section) panel of Review Editors. Because this Research Topic repre-
of chronic disease self-management programs: healthcare savings through a sents work closely associated with a nationwide evidence-based movement in the US,
community-based intervention. BMC Public Health (2013) 13:1141. doi:10. many of the authors and/or Review Editors may have worked together previously in
1186/1471-2458-13-1141 some fashion. Review Editors were purposively selected based on their expertise with
20. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, Whitelaw N, et al. Successes evaluation and/or evidence-based programming for older adults. Review Editors were
of a national study of the chronic disease self-management program: meet- independent of named authors on any given article published in this volume.
ing the triple aim of health care reform. Med Care (2013) 51(11):992–8.
doi:10.1097/MLR.0b013e3182a95dd1 Received: 16 July 2014; accepted: 21 September 2014; published online: 27 April 2015.
21. Smith ML, Ory MG, Ahn S, Kulinski KP, Jiang L, Horel S, et al. National dissem- Citation: Smith ML, Wilson MG, DeJoy DM, Padilla H, Zuercher H, Corso P, Van-
ination of chronic disease self-management education programs: an incremen- denberg R, Lorig K and Ory MG (2015) Chronic Disease Self-Management Program
tal examination of delivery characteristics. Front Public Health (2015) 2:227. in the workplace: opportunities for health improvement. Front. Public Health 2:179.
doi:10.3389/fpubh.2014.00227 doi: 10.3389/fpubh.2014.00179
22. Lorig K, Holman H, Sobel D, Laurent D. Living a Healthy Life with Chronic This article was submitted to Public Health Education and Promotion, a section of the
Conditions. 3rd ed. Boulder, CO: Bull Publishing Company (2006). journal Frontiers in Public Health.
23. Bandura A. Social cognitive theory of self-regulation. Organ Behav Hum Decis Copyright © 2015 Smith, Wilson, DeJoy, Padilla, Zuercher, Corso, Vandenberg , Lorig
Process (1991) 50(2):248–87. doi:10.1016/0749-5978(91)90022-L and Ory. This is an open-access article distributed under the terms of the Creative
24. Sliverstein M. Meeting the challenges of an aging workforce. Am J Ind Med Commons Attribution License (CC BY). The use, distribution or reproduction in other
(2008) 51:269–80. doi:10.1002/ajim.20569 forums is permitted, provided the original author(s) or licensor are credited and that
25. Centers for Disease Control and Prevention [Internet]. Workplace Health Model the original publication in this journal is cited, in accordance with accepted academic
(2013). Available from: http://www.cdc.gov/workplacehealthpromotion/model/ practice. No use, distribution or reproduction is permitted which does not comply with
index.html these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 179 | 171

 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00214

Meeting the challenge of cancer survivorship in public

health: results from the evaluation of the chronic disease
self-management program for cancer survivors
Betsy C. Risendal 1 *, Andrea Dwyer 1 , Richard W. Seidel 2 , Kate Lorig 3 , Letoynia Coombs 4 and Marcia G. Ory 5
1
Community and Behavioral Health, Colorado School of Public Health, Aurora, CO, USA
2
Carilion Clinic, Virginia Tech Carilion School of Medicine, Roanoke, VA, USA
3
School of Medicine, Stanford University, Stanford, CA, USA
4
Colorado Health Outcomes, University of Colorado, Aurora, CO, USA
5
Health Promotion and Community Health Sciences, Texas A&M Health Science Center School of Public Health, College Station, TX, USA

Edited by: Introduction: Self-management has been identified as an important opportunity to improve
Matthew Lee Smith, The University of
health outcomes among cancer survivors. However, few evidence-based interventions are
Georgia, USA
available to meet this need.
Reviewed by:
Diane Elliot, Oregon Health and Methods: The effectiveness of an adapted version of the Chronic Disease Self-
Science University, USA
Miruna Petrescu-Prahova, University
Management Program for cancer survivors called Cancer Thriving and Surviving was
of Washington, USA evaluated in a randomized trial. Outcomes were assessed at baseline and 6-months
*Correspondence: post program via written survey among 244 participants in Colorado. Repeated measures
Betsy C. Risendal , Community and analysis was used to analyze pre/post program change.
Behavioral Health, Colorado School of
Public Health, 13001 E. 17th Place, Results: Statistically significant improvement was observed among those in the inter-
MS F538, Aurora, CO 80045, USA vention in the following outcomes: Provider communication (+16.7% change); depres-
e-mail: [email protected]
sion (−19.1%); energy (+13.8%); sleep (−24.9%) and stress-related problems (−19.2%);
change over time was also observed in the controls for energy, sleep, and stress-related
outcomes though to a lesser degree. Effect sizes of the difference in change over time
observed indicate a net beneficial effect for provider communication (0.23); and decreases
in depression (−0.18); pain (−0.19); problems related to stress (−0.17); and sleep (−0.20).
Conclusion: Study data suggest that the self-management support from adaptation of
the CDSMP can reach and appeal to cancer survivors, improves common concerns in this
population, and can fill an important gap in meeting the ongoing need for management of
post-diagnosis issues in this growing segment of the U.S. population.
Keywords: cancer survivorship, self-management support, patient education, community-based research,
effectiveness trial

INTRODUCTION also concluded that the care of cancer survivors is fragmented

The estimated lifetime risk of developing cancer is 45% among and poorly coordinated and that self-management support can
men and 38% among women, with an expected total of 1.6 mil- help promote the delivery of quality care and improved health
lion new cancer cases in 2012 (1). More individuals are living outcomes in this population. Further, cancer survivors die from
longer due to improvements in early detection and treatment, and non-cancer causes at a rate higher than the general population (3,
therefore, the number of cancer survivors in the U.S. has dramat- 4) likely due to the long-term side effects of cancer and its treat-
ically increased. Current estimates suggest that there are over 13 ment and risk factors common to both cancer and non-cancer
million survivors alive today in the U.S., with an estimated 18 mil- causes of death.
lion at the end of the decade; an estimated 65% of all survivors The Chronic Care Model [CCM; (5)] is a rigorously evaluated
live 5 years or more. This dramatic increase in the survivor popu- and widely adopted approach to care management for chronic
lation has consequences for both the health of survivors and the conditions and features self-management support as one of the
healthcare system. For example, many survivors experience late key components for assuring quality healthcare. Self-management
and long-term effects from cancer and its treatment. Pain, fatigue, is defined as comprehensive engagement of the patient in prob-
depression, impaired physical function, and fear of recurrence are lem solving, decision making, and daily health-related behaviors
among the most common consequences of cancer as described in in partnership with their healthcare provider and community (6).
the landmark report by the Institute of Medicine, “From Cancer A 2007 review by Nolte et al. (7) found many benefits from self-
Patient to Cancer Survivor: Lost in Transition” (2). This Report management programs are also relevant to survivorship such as:

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 214 | 172
Risendal et al. Self-management program for cancer survivors

improved knowledge, acquisition of skills, symptom management, contacted by COAW personnel (located in Denver, also within
and ability to self-monitor health and healthcare needs. the Front Range) to assess interest and eligibility for participation
The Chronic Disease Self-Management Program (CDSMP) in the program. Participants were allocated to the intervention
is one of the few evidence-based interventions available across or control groups for the analytic evaluation. Inclusion in the
a variety of health-related conditions for comprehensive self- program required participants to be over the age of 21 years and
management support (8). While there are now specialized ver- diagnosed with cancer that required radiation, surgical, or adju-
sions for some chronic conditions such as diabetes, chronic pain, vant chemotherapy treatment, but not to be in active treatment at
HIV/AIDS, and arthritis1 , an adapted version for cancer survivors the time of enrollment. Persons currently receiving maintenance
has only recently been developed for use and testing in the U.S. The therapies for cancer delivered after completion of primary treat-
purpose of the current paper is to describe the findings from the 6- ment (such as anti-hormonal treatments) were eligible. Support
month outcome evaluation of cancer thriving and surviving (CTS) persons/caregivers of the above were also allowed to attend. All
among over 200 cancer survivors in the post-treatment phase who persons had to speak and read/write in English, and also agree
participated in a randomized trial in Colorado between 2011 and to attend in-person classes and arrange transportation to attend
2013. In this paper, we report the effectiveness of the evidence- classes. Persons in end-of-life care or currently undergoing active
based CDSMP translated to cancer survivors by comparing the treatment for cancer were excluded, as were individuals over the
magnitude of the effect observed in the intervention vs. control age of 79 years. Approval to conduct the research was obtained
group over time. by the Colorado Multiple Institutional Review Board; participants
provided signed informed consent. No incentives were offered to
MATERIALS AND METHODS potential participants.
INTERVENTION
Developed by researchers from the Stanford Patient Education INTERVENTION DELIVERY
Research Center at Stanford University, the model for the CDSMP Twenty-seven workshops were delivered in Colorado between
program entails a series of six weekly small-group sessions led by August 2011 and January 2013. Each workshop consisted of six
trained facilitators. The model is based on social cognitive theory 2.5 h sessions led by two facilitators as described above. Facilitators
(9) to focus on building skills, sharing experiences, and support were periodically observed by Master Trainers and provided
among the participants to maximize engagement. Sessions follow written feedback to monitor fidelity and quality assurance.
a standardized curriculum detailed in a program manual to pro-
mote fidelity to the following program elements: brainstorming, DATA COLLECTION
action plan formulation, action plan feedback, problem solving, Written self-administered surveys were collected from participants
and decision making (10). at baseline and for final follow-up measure (6 months after pro-
In brief, adaptations to the CDSMP for cancer survivors were gram completion). Instruments were from the Stanford CDSMP
guided by the research of Foster et al. (11) and the subsequent Evaluation2 and have been widely used in many health and aging
conceptual model (12) to include restoration of self-confidence, studies (13, 14) and are viewed as pragmatic measures (15).
adjustment to changed self, and confidence to self-manage cancer- Participants were asked at baseline to self-report demographic
related problems. The resultant CTS curriculum was initially characteristics (age, gender, marital status, race/ethnicity) as well
developed by Macmillan Cancer support in the U.K and subse- as cancer-related history (caregiver, time since diagnosis, type of
quently modified by the Stanford Patient Education Center to cancer, co-morbid conditions).
incorporate language more common to the U.S.
Researchers at the Colorado School of Public Health (CSPH) STUDY DESIGN
partnered with the Consortium for Older Adult Wellness (COAW) This study deployed a randomized controlled trial design, where
to deliver the program. COAW is a community-based agency with participants were randomized in a 2:1 ratio following consent to
state-wide license to deliver the evidence-based CDSMP. Individ- the intervention vs. control group. Since the purpose of this study
uals who were already trained and licensed to provide the CDSMP was to evaluate effectiveness rather than efficacy, we intention-
workshops and who were also cancer survivors completed a 2- ally sought to maximize the number of participants receiving the
day cross-training program led by the Stanford Patient Education intervention so that we could gain more experience to inform
Center to ensure fidelity to the model. implementation and also to improve the generalizability of the
results by broadening the characteristics and delivery while still
RECRUITMENT utilizing a valid comparison group. Participants were random-
Cancer survivors throughout the “Front Range” of Colorado, ized to group assignment using a random number generator by
where roughly two-thirds of state’s population resides, were the research coordinator who was separate from the intervention
approached in a variety of outreach methods including: inter- delivery. Caregivers/support persons were randomized as a pair
actions with cancer center staff and brochures left at medical with their survivor so they could attend sessions together, and
offices, mailed to homes using mailing lists from local cancer sur- therefore, not counted toward the 2:1 ratio. Persons who consented
vivor programs, distributed at cancer survivor local events, and and were randomly assigned to the control group were offered to
media. Potential respondents identified from these routes were attend the CTS workshops after the final evaluation assessment was

1 http://patienteducation.stanford.edu/programs/cdsmp.html 2 http://patienteducation.stanford.edu/research

www.frontiersin.org April 2015 | Volume 2 | Article 214 | 173

Risendal et al. Self-management program for cancer survivors

collected at 6-months following consent; thus their data served as • Health status: we asked respondents to rate their health on a
the control for those randomized to the intervention group. This scale of excellent (1) to poor (5). A low value on this scale
design was chosen to facilitate retention of controls over the 6- indicates better health; a decrease or negative change from the
month time period between consent and assessment of the main base-line period to the final period for this variable.
outcome measures at 6 months post-program (in order to mirror • Health symptomatology:
the CDSMP evaluation plan), and to gain more experience with o Energy: we asked patients five questions about their level of
intervention delivery. energy: (1) Do you feel worn out?, (2) Did you have a lot
of energy?, (3) Did you feel tired?, (4) Do you have enough
OUTCOMES energy to do the things you wanted to do?, and (5) Did you
Our hypothesis was that the intervention would produce improve- feel full of pep?. Responses to these items range from none of
ment in outcomes directly related to health beliefs and behav- the time (0) to all of the time (5). If the respondent replied
iors related to physical activity (days active/minutes active), self- to at least three of these five items, the scale was calculated as
efficacy, and communication with providers. Secondary outcomes the mean of the non-missing items with the two negatively
of interest included self-reported health and symptoms (health worded items (1 and 3) reversed coded. A high score on this
status, depression, energy, pain, sleep, and stress). scale represents more energy. An increase or positive change
The following describes measures employed in this study: for scale is desirable.
o Pain, stress, and sleep problems: these three visual scales
• Days active, minutes active: days active, minutes active: respon-
ranged from no problem (0) to very big problem (10). A high
dents were asked how many days in the past week they were
score on these scales represents more problems. A decrease
physically active or exercising for at least 30 min and how many
or negative change on this scale was desired.
total minutes in the past week they were physically active or
• Depression: participants completed the eight-item Personal
exercising, including brisk walking, running, dancing, bicy-
Health Questionnaire Depression Scale (16). The items’
cling, water exercise, etc., that may cause faster breathing or
responses ranged from Not at all (0) to Nearly Everyday (3).
heartbeat, or feeling warmer. For the current analyses, we are
Sum scores ranged from 0 to 24. Higher scores indicate more
using continuous count data for number of minutes exercised
severe depression. A decrease or negative change in these items
and number of days exercised. Respondents were asked how
was desired.
many days in past week they were physically active or exer-
cising for at least 30 min and how many total minutes in the STATISTICAL ANALYSIS
past week they were physically active or exercising (includ- Pearson Chi Square and Fisher exact tests were used in Table 1 to
ing brisk walking, running, dancing, bicycling, water exercise, compare the demographic characteristics of the intervention and
etc.) that may cause faster breathing or heartbeat, or feeling control groups.
warmer. For the current analyses, we are using continuous count In order to determine if the outcome variables showed the
data for number of minutes exercised and number of days change in the desired direction over time, we used repeated mea-
exercised. sures analysis with an unstructured variance covariance matrix.
• Participant care seeking behaviors (communication with physi- This method models the correlations between repeated observa-
cians): communication with a physician was measured using a tions from the same individual. It also utilizes available data for all
three-item scale, which asked participants if they did the follow- participants, regardless of final measure completion allowing for a
ing things when visiting a physician: prepare a list of questions, form of intent to treat analysis, which reduces potential drop-out
ask questions about things they want to know or do not under- bias. A case is only excluded if they did not answer a sufficient
stand, and discuss personal problems. Scores for these items number of items on both the pre- and the post-test. If they have
ranged from never (0) to always (5). If respondents answered enough data for either time period, they were included in the sam-
at least two of these items, the scale was calculated as the mean ple. Data from participants were only excluded if they did not
of the non-missing items. Higher scores represent better com- supply an adequate number of responses required for each instru-
munication with a physician. An increase or positive change is ment; if they had enough responses for either time point the data
desirable. were included in the analysis. Each of the 10 outcomes described
• Self-efficacy: this was measured using a six-item scale, which above served as dependent variables in models with no intercepts
asked participants how confident they were keeping fatigue, and a time period (baseline, final) by group assignment interaction
physical discomfort, pain, emotional distress, and other symp- as the independent variable. Parameters resulting from this model
toms and health problems caused by cancer diagnosis and include an estimated mean for each group at each time period
treatment from interfering with the things they want to do; they (17). Contrasts were estimated to determine change from baseline
were also asked about their confidence doing different tasks and to final and differences between groups.
activities needed to manage their cancer diagnosis and treat- We conducted additional analyses to determine if the effect of
ment to reduce their need to see a doctor. Responses to these the intervention was moderated by age. These models were similar
items ranged from Not at all confident to (1) to Totally confi- to the models described above except a three-way interaction of
dent (10). If respondents answered at least four of these items, age group (<65, 65+), treatment group (intervention vs. control),
the scale was calculated as the mean of the non-missing items. and time period since diagnosis replaced the two-way interaction.
Higher scores represented greater confidence. An increase or No interaction effect was observed, so the original analyses are
positive change was desired. presented.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 214 | 174
Risendal et al. Self-management program for cancer survivors

Table 1 | Characteristics at baseline among study participants*, by

treatment group (n, %).

Characteristic Intervention Control p-Value Characteristic Intervention Control p-Value

(n = 169*) (n = 89*) (n = 169*) (n = 89*)

Age (years) Lung 9 (5.3) 1 (1.1) 0.17

<50 33 (19.5) 19 (21.4) 0.93 Leukemia 6 (3.6) 4 (4.5) 0.74
50–64 81 (47.9) 41 (46.1) Melanoma 4 (2.4) 1 (1.1) 0.66
65+ 55 (32.5) 29 (32.6) ***Other 43 (25.4) 6 (6.7) 0.0003
Sex
Male 38 (22.5) 9 (10.1) 0.01 *Includes persons diagnosed with cancer (excludes participating caregivers).

Female 131 (77.5) 80 (89.9) **Anyone who did check black or white including those who checked Asian,
Native American, or other.
Marital status
***Including cancer of the cervix, bladder, bone, brain, esophagus, kidney, liver,
Married/partner 106 (62.7) 47 (52.8) 0.19
pancreas, thyroid, or other.
Single 62 (36.7) 42 (47.2)
1 (0.6)
Hispanic ethnicity 13 (7.7) 6 (6.7) 0.78 Effect sizes were calculated using Cohen’s d (18), which is
Race defined as the difference between two means divided by the pooled
White 145 (85.8) 74 (83.2) 0.57 SD of the groups. Analysis was conducted using the Mixed Pro-
Black 14 (8.3) 5 (5.6) 0.43 cedure of SAS 9.4. Unlike statistical significance, effect size is not
Other** 11 (6.5) 11 (12.4) 0.11 dependent on sample size for interpretation. Effect size is a quanti-
Insurance tative measure of the relative strength of the intervention whereby
Medicaid 10 (5.9) 3 (3.4) 0.55 a larger absolute effect size value always indicates a stronger effect.
Medicare 83 (49.1) 31 (34.8) 0.03
HMO (Kaiser) 28 (16.6) 8 (9.0) 0.09 RESULTS
Private 63 (37.3) 34 (38.2) 0.88 RECRUITMENT AND RANDOMIZATION
VA/Other 4 (2.4) 2 (2.3) 1 The activities described above resulted in 493 referrals (see Con-
None 6 (3.6) 3 (3.4) 1 sort Diagram, Figure 1). Since this was an effectiveness study, the
Employment eligibility criteria were quite broad and only 12 of these individ-
Working 47 (27.8) 38 (42.7) 0.04 uals were ineligible (reasons included: did not receive treatment
Not working 49 (29.0) 15 (16.9) for cancer, over age 79, and still in treatment). A total of 158
Retired 49 (29.0) 25 (28.1) subjects ultimately did not enroll as follows: did not show for
Other 11 (6.5) 2 (2.3) first session to sign consent form (n = 37); not interested after
Missing 13 (7.7) 9 (10.1) learning more (n = 31); unreachable/voicemail left (n = 38); bad
Self-rated health timing/inconvenient time/location (n = 34); in cancer treatment
Excellent 13 (7.7) 8 (9.0) 0.76 because cancer returned following initial outreach (n = 18). This
Very good 52 (30.8) 29 (32.6) resulted in a total of 323 eligible subjects enrolled, including 267
Good 77 (45.6) 35 (39.3) persons diagnosed with cancer and 56 caregivers/supporters. Ran-
Fair 25 (14.8) 15 (16.9) domization resulted in 169 survivors (and 29 of their caregivers)
Poor 2 (1.2) 1 (1.1) assigned to the intervention and 89 survivors (and 15 of their care-
Missing 0 (0.0) 1 (1.1) givers) assigned to the control. Only the survivors (not caregivers)
Years since diagnosis 18 (10.7) in each group were utilized for the comparisons described in this
<1 73 (43.2) 7 (7.9) 0.87 paper (see below).
1–3 44 (26.0) 41 (46.1)
4–9 29 (17.2) 24 (27.0) PARTICIPATION/COMPLETION
10+ 5 (3.0) 13 (14.6) The average number of participants in each workshop was
Missing 4 (4.5) 8.2 ± 2.7; half the sessions had 7–9 participants (48.2%) with
Cancer type the remainder of workshops approximately evenly split between
Breast 66 (39.1) 66 (74.2) <0.0001 5and 6 (29.6%) or 10 or greater participants (22.2%). The major-
Lymph./Hodgkins 27 (16.0) 4 (4.5) 0.01 ity of participants (84%) completed four or more of the six
Prostate 12 (7.1) 4 (4.5) 0.41 sessions in each workshop (data not shown). Of the 169 per-
Colorectal 11 (6.5) 2 (2.3) 0.23 sons diagnosed with cancer who were assigned to the inter-
Endometrial/uterine 7 (4.1) 1 (1.1) 0.27 vention, 117 completed the final program measure (69.2%). A
Ovary 9 (5.3) 0 (0.0) 0.03 similarly high percentage of persons diagnosed with cancer and
Multiple myeloma 6 (3.6) 2 (2.3) 0.72 assigned to the control group (n = 89) completed the final measure
(n = 72; 81.0%). Baseline characteristics of completers and non-
(Continued) completers were compared (data not shown) with only one factor

www.frontiersin.org April 2015 | Volume 2 | Article 214 | 175

Risendal et al. Self-management program for cancer survivors

(gender) significantly different between groups (more women who did not perform the initial data entry and demonstrated a
in the intervention group); differences by intervention vs. con- detected error rate of <1%. A total of 448 surveys (258 base-
trol group in regard to completion were not observed. Base- line, 190 post-program) were collected. The number of surveys
line characteristics by group assignment are shown in Table 1; with complete data for specific outcomes ranged from 427 to
the groups were quite similar in accordance with major char- 448, indicating that missing data within available surveys was
acteristics with the exception of gender and cancer type. Dif- minimal.
ferences in insurance type and employment status were also
present. OUTCOME MEASURES
Baseline, final values measured at 6-month post-program, and
MISSING DATA/DATA ENTRY ERRORS change (%) values over time observed among participants in
Five percent of surveys entered were checked at random for data the intervention as compared to the control group are shown in
entry errors; these checks were performed by study personnel Table 2. Statistically significant change over time among partici-
pants in the intervention was observed in the following outcomes:
provider communication, depression, energy, sleep, and stress.
Program Referrals/Contacts
493 Where change was observed in the controls, they were smaller
among most outcomes.
|
Ineligible
12 EFFECT SIZES
Eligible
481
Effect sizes calculated by Cohen’s d are shown in Table 3 for
Decided not to join the intervention, control, and the difference in degree of change
158
| between the two groups. A beneficial effect was observed over time
Consented among participants in the intervention for many outcomes, con-
323
sistent with the results in Table 2. For example, medium effect sizes
Randomized to Intervention Randomized to Lagged Control (0.5–0.75) were shown for provider communication, depression,
Survivors Caregivers Survivors Caregivers
173 36 94 20
energy, sleep, and stress. In contrast, small effect sizes (0.16–
| | 0.35) or no effects were observed in the control for these same
Baseline Baseline
outcomes.
Survivors Caregivers Survivors Caregivers When the effect size for the difference in change in the inter-
169* (97.7%) 29 (80.5%) 89* (94.7%) 15 (75.0%)
| | vention group relative to the change in the control was evaluated,
a small effect was observed in regard to provider communica-
Final (6 month measure) Final (6 month measure
Survivors Caregivers Survivors Caregivers
tion (0.23), sleep (−0.20), and very small effect for stress-related
117* (69.2%) 20 (69.0%) 72* (81.0%) 7 (46.7%) problems (−0.17).
* Subjects included in repeated measures outcome analysis
In addition to the results shown in Tables 2 and 3, we also
examined the role of interaction between age, time since diagno-
FIGURE 1 | Study participation. sis, and outcomes of interest and did not detect possible effect
modification (data not shown).

Table 2 | Baseline, final, change, and % change between 6-month outcome measures, by intervention vs. control.

Intervention (n = 169) Lagged control (n = 89)

6-month outcome Baseline value Final value Change Change (%) Base-line value Final value Change Change (%)
mean (SE) mean (SE) mean (SE) mean (SE) mean (SE) mean (SE)

Days active 3.0 (0.2) 3.0 (0.2) 0.1 (0.2) 1.7 2.9 (0.2) 2.8 (0.3) −0.1 (0.3) −4.3
Minutes active 143.4 (12.3) 192 (27.3) 48.7 (28.2) 34.0 124.4 (17.0) 155.1 (35.0) 30.6 (36.5) 24.6
Self-efficacy 70.2 (1.7) 72.4 (2.0) 2.3 (1.9) 3.2 73.6 (2.3) 77.7 (2.6) 4.0 (2.5) 5.5
Provider communication 3.1 (0.1) 3.7 (0.1) 0.5** (0.1) 16.7 3.5 (0.1) 3.8 (0.1) 0.3 (0.1)* 7.4
Depression 8.5 (0.4) 6.9 (0.5) 1.62** (0.5) −19.1 7.8 (0.6) 7.2 (0.6) −0.7 (0.6) −8.5
Health status 2.7 (0.1) 2.7 (0.1) 0.0 (0.1) 0.4 2.7 (0.1) 2.6 (0.1) −0.1 (0.1) −2.2
Energy 2.2 (0.1) 2.5 (0.1) 0.3** (0.1) 13.8 2.1 (0.1) 2.4 (0.1) 0.3 (0.1)* 12.9
Pain 3.2 (0.2) 3.0 (0.2) −0.3 (0.2) −8.0 3.5 (0.3) 3.7 (0.3) 0.2 (0.3) 5.9
Sleep problems 5.3 (0.2) 4.0 (0.3) −1.3** (0.2) −24.9 5.6 (0.3) 4.9 (0.3) −0.7 (0.3)* −12.7
Stress problems 5.1 (0.2) 4.1 (0.2) −1.0** (0.22) −19.1 5.4 (0.3) 4.8 (0.3) −0.6 (0.3)* −10.3

*Statistically significant change between baseline and final measures; p<0.05.

**Statistically significant change between baseline and final measures; p < 0.001; repeated measures analysis.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 214 | 176
Risendal et al. Self-management program for cancer survivors

Table 3 | Effect sizes (Cohen’s d ) in 6-month outcomes, by intervention 0.11), suggesting a net small/medium benefit since the trend in the
vs. control and change between groups. controls was to worsen over time. Although sample size does not
directly impact the calculation of effect size, the finding of effect
Outcome Effect size in Effect size Effect size size differences is considered meaningful when observed in larger,
intervention observed of difference in well-designed studies such as reported here.
group in control change between The heterogeneous nature of our study population in this
(n = 169) group intervention intentionally pragmatic design allows us to estimate the benefits
(n = 89) and control of the program in the real world by examining effect sizes
(i.e., effectiveness). However, it also dampens the ability to detect
Days active 0.03 −0.05 0.06
statistically significant differences because by design, it does not
Minutes active *0.25 0.12 0.06
use carefully constructed homogeneous study populations to min-
Self-efficacy 0.17 *0.23 0.08
imize variation as in efficacy trials. The ability to demonstrate
Health status 0.02 −0.08 0.08
statistical significance in an effectiveness evaluation is strongly
Provider communication **0.75 *0.29 *0.23
influenced by the number of persons in relevant subgroups where
Depression **−0.50 *−0.16 *−0.18
the intervention may be more or less efficacious; however, these
Energy **0.51 *0.35 0.03
subgroups are not necessarily known to the researchers or able to
Pain −0.17 0.11 *−0.19
be detected in the real-world setting of the evaluation. The fact that
Sleep problems **−0.72 *−0.31 *−0.20
we that we did not see statistically significant difference between
Stress problems **0.63 *−0.28 *−0.17
the change over time in the intervention vs. control is likely a
*Borderline/small effect (0.2). consequence of the heterogeneous nature of our real-world study
**Medium effect (0.5). population; but the difference in effect sizes represents the impact
of the intervention by measuring the magnitude of this difference
DISCUSSION observed over time in the two groups and highlights the external
These outcome analyses of the adapted version of the Stanford validity of our findings.
CDSMP for cancer survivors indicate demonstrable beneficial Other studies of the CDSMP have similarly evaluated effect
effects in many outcomes (Tables 2 and 3); further, no out- size to evaluate the impact of the program. A 2008 Cochrane
comes worsened following participation in the intervention when Collaboration review (11) of self-management education inter-
we evaluated the group-level comparisons. The outcomes that ventions demonstrated effect sizes observed in multiple reports
improved with the CTS program (e.g., provider communica- of other populations similar to or smaller than those observed in
tion, depression, sleep problems, and stress-related problems; see the current study. For example, of the 17 randomized trials of lay-
Table 3) are particularly salient to the challenges faced by cancer led self-management programs in this review demonstrated effect
survivors. For example, fragmented and poorly coordinated sys- sizes for pain of 0.10 (current study 0.11) and depression of 0.16
tems of care make provider communication an important skill for (current study 0.18).
the cancer survivor. Sleep, depression, and pain are commonly Although we did not observe an effect with self-efficacy as
reported symptoms as described in the previously cited IOM observed in other trials (ranging from 0.30 to 0.40), we did observe
report, and it is notable that a non-medical, relatively inexpensive an improvement in provider communication (0.23). Cancer sur-
and brief educational intervention delivered in the community set- vivors in the post-treatment period neither have the frequency nor
ting had positive impact on these common yet potentially serious regularity of health system interaction as with other chronic con-
issues. ditions such as asthma and diabetes and therefore may not have
Improvement over time between the baseline and 6-month had ample opportunity to use their self-management skills, which
measure was also observed in three domains among the controls, could be the cause of the neutral scores on this domain. However,
although generally to a lesser degree than those in the intervention the improvement observed in provider communication is a related
(see Table 2). For example, we observed a 19.1% mean difference and similarly important skill for this population. Provider com-
depression scores measured by the PHQ-8 over time in the inter- munication is a necessary component of the Chronic Care Model,
vention group in contrast to only 8.5% decline in the controls over which promotes collaboration between patients and providers in
time. Further, the difference over time was statistically significant partnership to achieve improved outcomes (5). This is especially
in the intervention but not in the control group. This difference important in survivors who may experience both late and long-
in the magnitude of effect over time observed in the intervention term side effects from treatment that can change over time, and
vs. control groups is illustrated by comparing effect size. Effect may require ongoing vigilance and care.
sizes take into account the size of measurement error in the data Another observation from our study could be explored in future
but do not rely on sample size or statistical significance for their research is our observation of improvement over time in the
interpretation; therefore, they are meaningful when evaluating the control group, which although was to a lesser degree, was sta-
relative impact of an intervention. In the case of depression, for tistically significant in three constructs (energy, sleep, and stress).
example, the effect size in the intervention group over time was a Other researchers have suggested that positive adjustment or post-
medium/large effect (−0.50) vs. a small effect (−0.16) in controls traumatic growth over time following a stressful event such as
(Table 3). Similarly, striking differences in effect size among the cancer can occur (19–21). Thus, one possible explanation for
intervention vs. control groups were observed for pain (−0.17 vs. this finding is that survivors have accepted a “new normal” and

www.frontiersin.org April 2015 | Volume 2 | Article 214 | 177

Risendal et al. Self-management program for cancer survivors

therefore the increase over a time is a reflection of this perception. from diagnoses through to treatment and survivorship, lack
Additionally, there may be some endogenous aspects to a survivor’s of an evidence base to guide follow-up surveillance and deci-
improvement that could be capitalized upon in future iterations sion making, complex late and long-term side effects requir-
of the program. ing detailed patient history and records, and limited oncol-
Limitations of this study are that we may not have quantita- ogy system capacity. While the delivery of self-management
tively measured all the outcomes of relevance in this population. programs to date has been driven by innovations in primary
For example, we did not directly measure social support or the care, a recent emerging trend in support of needed system
unique benefits among caregivers such as family communication. and policy change for cancer survivors is the establishment of
Additionally, while the majority of respondents completed the final patient-centered medical homes (PCMH) in the oncology set-
measure, we were unable to measure final outcomes in all respon- ting (24). Future research is needed to support policy change
dents. However, we utilized repeated measures analysis to utilize to ensure that patients receive self-management support that
data from all respondents regardless of completion to minimize is tailored to their cancer needs across oncology and a vari-
this potential source of bias. Additionally, we chose to include sur- ety of other settings, driven by patient needs and preferences.
vivors in the post-treatment stage only to support the unmet need Additional research is also needed to understand which out-
for transition support. Although it is reasonable to expect that comes are most relevant in this population toward demonstrat-
similar benefits would be observed in survivors at other points ing cost-effectiveness that can inform needed system and policy
in the continuum, additional evaluations with survivors at other change.
time points should be conducted. Decreased emergency room visits and hospitalizations are of
A recent review of 16 self-management programs that have been relevance to cost in other chronic illness populations (14), but
utilized with a variety of cancer survivor populations promotes outcomes such as overuse of care/screening may be even more
the use of the Chronic Care Model and particularly support for important for cancer survivors. Patterns of care outcomes are dif-
self-management in addressing needs across the continuum from ficult to track in with multiple payor systems, but policy changes
diagnosis to survivorship (22). Aspects of self-management high- to support the collection and analysis where possible in Medic-
lighted in this review as beneficial for survivors are also highly vis- aid/Medicare or other single-payor systems should be pursued
ible “active ingredients” in the CTS program and include: goal set- to further evaluate outcomes from self-management support for
ting, realistic action plans, partnering with providers, and identify- cancer survivors. The CTS has enormous potential to be widely
ing aspects of health and healthcare that patients can self-manage disseminated by tapping into existing channels in the community
with confidence. Although the attention to self-management inter- and among providers that have already been established with the
ventions in this population is increasing, this review concludes CDSMP Program; however, the successful implementation of self-
that there is an urgent need for the translation of these inter- management interventions such as the CTS is reliant upon buy-in
ventions into practice, particularly in the post-treatment period. by oncology providers, survivors, and the healthcare system to
The authors suggest that interventions at this point in time can recognize benefits such as those observed in the current report. As
be especially helpful in easing transition to less regular contact evidence continues to mount on the effectiveness of the CDSMP
with oncologists and dealing with the psychosocial and functional in other chronic disease populations (25, 26), and models of sur-
challenges into survivorship. vivorship care continue to develop, policy and system support for
Contemporary views of effectiveness have evolved to suggest self-management as a vital and viable component in successful
that it is influenced not only by efficacy, but reach of the program transition to survivorship is needed.
as well as implementation with fidelity (23). Our enrollment of
over 300 cancer survivors and caregivers to this effectiveness study ACKNOWLEDGMENTS
and the diversity of the study population according to cancer type, This publication was supported by the Centers for Disease Con-
time since diagnosis, age, and other characteristics as shown in trol and Prevention (CDC) and the National Cancer Institute
Table 1 suggests that this program can reach and appeal to the gen- through the Cancer Prevention and Control Research Network, a
eral cancer population. By partnering with a community agency network within the CDC’s Prevention Research Centers Program
with state-wide reach for delivery of the original CDSMP Program (University of Colorado, U48DP001938; Texas A&M University,
with certified facilitators and extensive experience in delivering U48DP001924). The contents of this article are solely the respon-
the program, we were further able to deliver the new adapted ver- sibility of the authors and do not necessarily represent the official
sion in keeping with the principles of original program method. views of the CDC or NCI. Additionally, we thank Stanford Patient
When taken in sum, these outcome and implementation data Education Research Center for program development and tech-
demonstrate that survivors who participate in the CTS program nical assistance and the Consortium for Older Adult Wellness in
experience a small but measurable net gain over time in impor- Colorado for program dissemination.
tant survivorship domains in comparison to those who receive
no intervention, and that the program can fill an important gap REFERENCES
in meeting the ongoing need for management of post-diagnosis 1. American Cancer Society. Cancer Facts and Figures 2012. Atlanta: ACS (2012).
issues in this growing segment of the U.S. population. 2. Hewitt M, Greenfield S, Stovall E. From Cancer Patient to Cancer Survivor: Lost
in Transition. The National Academies Press (2005).
The implementation of self-management support is partic- 3. Stull VB, Snyder DC, Demark-Wahnefried W. Lifestyle interventions in can-
ularly challenging in the cancer environment for a number of cer survivors: designing programs that meet the needs of this vulnerable and
reasons including the use of multiple specialty care providers growing population. J Nutr (2007) 137(1 Suppl):243S–8S.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 214 | 178
Risendal et al. Self-management program for cancer survivors

4. Patnaik JL, Byers T, Diguiseppi C, Denberg TD, Dabelea D. The 21. Bellizzi KM, Smith AW, Reeve BB, Alfano CM, Bernstein L, Meeske K, et al.
influence of comorbidities on overall survival among older women diagnosed Posttraumatic growth and health-related quality of life in a racially diverse
with breast cancer. J Natl Cancer Inst (2011) 103(14):1101–11. doi:10.1093/jnci/ cohort of breast cancer survivors. J Health Psychol (2010) 15(4):615–26.
djr188 doi:10.1177/1359105309356364
5. Wager EH. Chronic disease management: what will it take to improve car for 22. McCorkle R, Ercolano E, Lazenby M, Schulman-Green D, Schilling LS, Lorig K,
chronic illness? Eff Clin Pract (1998) 1(1):2–4. et al. Self-management: enabling and empowering patients living with cancer as
6. Lorig KR, Holman H. Self-management education: history, definition, out- a chronic illness. CA Cancer J Clin (2011) 61(1):50–82. doi:10.3322/caac.20093
comes, and mechanisms. Ann Behav Med (2003) 26(1):1–7. doi:10.1023/A: 23. Glasgow RE, Lictenstein E, Marcus AC. Re-thinking the efficacy to effective-
1021718304633 ness transition. Am J Public Health (2003) 93(8):1261–7. doi:10.2105/AJPH.93.
7. Nolte S, Elsworth GR, Sinclair AJ, Osborne RJ. The extent and breadth of ben- 8.1261
efits from participating in chronic disease self-management course: a national 24. Sprandio J. Oncology patient-centered medical home. J Oncol Pract (2012)
patient-reported outcomes survey. Patient Educ Couns (2007) 65(3):351–60. 8(3S):47s–9s. doi:10.1200/JOP.2012.000590
doi:10.1016/j.pec.2006.08.016 25. Ory MG, Ahn S, Jiang L, Lorig K, Ritter P, Laurent DD, et al. National Study of
8. Brady TJ, Murphy L, O’Colmain BJ, Beauchesne D, Daniels B, Greenberg M, Chronic Disease Self-Management: six-month outcome findings. J Aging Health
et al. A meta-analysis of health status, health behaviors, and health care utiliza- (2013) 25(7):1258–74. doi:10.1177/0898264313502531
tion outcomes of the chronic disease self-management program. Prev Chronic 26. Ory MG, Smith ML, Patton K, Lorig K, Zenker W, Whitelaw N. Self-management
Dis (2013) 10:1201–12. doi:10.5888/pcd10.120112 at the tipping point: reaching 100,000 Americans with evidence-based programs.
9. Bandura A. Self-Efficacy: The Exercise of Control. New York: W.J. Freeman (1997). J Am Geriatr Soc (2013) 61(5):821–3. doi:10.1111/jgs.12239
10. Risendal B, Dwyer A, Seidel R, Lorig K, Katzenmeyer C, Coombs L, et al. Adap-
tation of the chronic disease self-management program for cancer survivors:
feasibility, acceptability, and lessons for implementation. J Cancer Educ (2014) Conflict of Interest Statement: Dr. Lorig receives royalties from Bull Publishing
29:762–71. doi:10.1007/s13187-014-0652-8 Company (Boulder, CO, USA) for the book that is included in the CDSMP delivery.
11. Foster C, Wright D, Hill J, Hopkinson J, Roffe L. Psychosocial implications of The research was conducted in the absence of any commercial or financial rela-
living 5 years or more following a cancer diagnosis: a systematic review of the tionships that could be construed as a potential conflict of interest among other
research evidence. Eur J Cancer (2009) 18(3):223–47. doi:10.1111/j.1365-2354. authors.
2008.01001.x
12. Foster C, Fenlon D. Recovery and self-management following a cancer diagnosis. This paper is included in the Research Topic, “Evidence-Based Programming for Older
Br J Cancer (2011) 105:S21–8. doi:10.1038/bjc.2011.419 Adults.” This Research Topic received partial funding from multiple government and
13. Lorig K, Stewart A, Ritter P, González V, Laurent D, Lynch J. Outcome Measures private organizations/agencies; however, the views, findings, and conclusions in these
for Health Education and other Health Care Interventions. Thousand Oaks CA: articles are those of the authors and do not necessarily represent the official position
Sage Publications (2012). of these organizations/agencies. All papers published in the Research Topic received
14. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, Whitelaw N, et al. Successes peer review from members of the Frontiers in Public Health (Public Health Education
of a national study of the chronic disease self-management program: meet- and Promotion section) panel of Review Editors. Because this Research Topic repre-
ing the triple aim of health care reform. Med Care (2013) 51(11):992–8. sents work closely associated with a nationwide evidence-based movement in the US,
doi:10.1097/MLR.0b013e3182a95dd1 many of the authors and/or Review Editors may have worked together previously in
15. Kulinski KP, Boutaugh M, Smith ML, Ory MG, Lorig K. Setting the stage: mea- some fashion. Review Editors were purposively selected based on their expertise with
sure selection, coordination, and data collection for a national self-management evaluation and/or evidence-based programming for older adults. Review Editors were
initiative. Front Public Health (2015) 2:206. doi:10.3389/fpubh.2014.00206 independent of named authors on any given article published in this volume.
16. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression
severity measure. J Gen Intern Med (2001) 16:606–13. doi:10.1046/j.1525-1497.
2001.016009606.x Received: 18 June 2014; accepted: 14 October 2014; published online: 27 April 2015.
17. Fairclough D. Design and Analysis of Quality of Life Studies in Clinical Trials. 2nd Citation: Risendal BC, Dwyer A, Seidel RW, Lorig K, Coombs L and Ory MG (2015)
ed. London, UK: Chapman and Hall, CRC Press (2012). Meeting the challenge of cancer survivorship in public health: results from the evalua-
18. Rosnow RL, Rosenthal R. Computing contrasts, effect sizes, and counternulls tion of the chronic disease self-management program for cancer survivors. Front. Public
on other people’s published data: general procedures for research consumers. Health 2:214. doi: 10.3389/fpubh.2014.00214
Pyschol Methods (1996) 1:331–40. doi:10.1037/1082-989X.1.4.331 This article was submitted to Public Health Education and Promotion, a section of the
19. Cordova MJ, Andrykowski MA. Responses to cancer diagnosis and treatment: journal Frontiers in Public Health.
posttraumatic stress and posttraumatic growth. Semin Clin Neuropsychiatry Copyright © 2015 Risendal, Dwyer, Seidel, Lorig , Coombs and Ory. This is an open-
(2003) 8(4):286–96. access article distributed under the terms of the Creative Commons Attribution License
20. Barskova T, Oesterreich R. Post-traumatic growth in people living with a (CC BY). The use, distribution or reproduction in other forums is permitted, provided
serious medical condition and its relations to physical and mental health: the original author(s) or licensor are credited and that the original publication in this
a systematic review. Disabil Rehabil (2009) 31(21):1709–33. doi:10.1080/ journal is cited, in accordance with accepted academic practice. No use, distribution or
09638280902738441 reproduction is permitted which does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 214 | 179

 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00181

Implementing a chronic disease self-management

program into China: the Happy Life Club™
Colette Joy Browning 1,2 *, Hui Yang 2 , Tuohong Zhang 3 , Anna Chapman 2 , Shuo Liu 4 , Joanne Enticott 2 and
Shane Andrew Thomas 2,5
1
Royal District Nursing Service Institute, St Kilda, VIC, Australia
2
School of Primary Health Care, Monash University School of Primary Health Care, Monash University, Melbourne, VIC, Australia
3
Research Center for Ageing and Health Services, Department of Global Health, Peking University, Beijing, China
4
Department of Social Medicine and Management, School of Public Health, Peking University, Beijing, China
5
The University of Adelaide, Adelaide, SA, Australia

Edited by: China is experiencing population aging, increased prevalence of chronic diseases, and
Matthew Lee Smith, The University of
reductions in the frequency of healthy lifestyle behaviors. In response to these signifi-
Georgia, USA
cant transitions, China is implementing major reforms in health care services with a focus
Reviewed by:
Martin Dempster, Queen’s University on strengthening primary health care. In this paper, we describe a 12-month diabetes
Belfast, UK management program, the Happy Life Club™ (HLC™), implemented in a primary health
Emily Joy Nicklett, University of care setting in Beijing, that uses doctor and nurse health coaches trained in behavior
Michigan, USA
change techniques and motivational interviewing (MI). This paper reports the results of
*Correspondence:
this pilot study and discusses issues involved in the implementation of Chronic Diseases
Colette Joy Browning, School of
Primary Health Care, Building 1 270 Self-Management Programs in China. The intervention group showed improvements in
Ferntree Gully Road, Notting Hill, VIC HbA1c levels at 6 months and both the control and intervention groups showed reductions
3168, Australia in waist circumference over time. Systolic blood pressure improved over time in the inter-
e-mail: colette.browning@
vention group. The intervention group showed improvement in quality of life across the
monash.edu
intervention period and both groups showed decreases in psychological distress across
the intervention. Doctor visits increased between baseline and 6 months, but there was
no change in doctor visits between 6 and 12 months for both groups. The effects were
modest, and further investigations are required to evaluate the long-term impact of health
coach approaches in China.
Keywords: chronic disease self-management, motivational interviewing, diabetes, older people, China

INTRODUCTION that in 2010, the prevalence of DM was 11.6%, representing 113.9

China is now following the trajectory of many Western coun- million adults. The prevalence of DM increases with age in Chi-
tries in terms of population aging, increased prevalence of chronic nese adults: in those aged 70 years and over it is estimated that
diseases and reductions in the population frequency of healthy 21.8% of women and 22% of men have diabetes (9). The Chinese
lifestyle behaviors. For example, over the period 1998–2008, the government has recognized the need for new approaches to the
incidence of diabetes mellitus (DM) tripled in China and, from management of DM including self-care education and the incor-
1991 to 2006, physical activity levels decreased by 32% (1, 2). In poration of healthy lifestyle interventions into routine care (10,
response to these significant transitions, China is implementing 11). However, such approaches need a trained workforce of health
major reforms in health care services for its 1.4 billion citizens. professionals who understand and embrace patient-centered care
The primary health care reforms, first announced in 2009, aim and who possess the requisite skills in behavior change and coun-
to deliver basic chronic disease care through community health seling principles and practice. Such an approach, involving the
services with referral of complex cases to the tertiary hospital sys- training of doctors and nurses in patient-centered care and behav-
tem (3, 4). Chronic illness management approaches in China are ior change techniques, has been piloted and developed into a
neither typically patient centered nor do they include a central diabetes self-management program in Beijing, China: the Happy
role for the patient in the self-management of their condition (5). Life Club™ (HLC™). This program is based on a similar program
Furthermore, patients are often dissatisfied with the medical ser- developed in Australia (12).
vices they receive while doctors focus on providing medications The HLC™ program involves nurses and doctors in primary
to manage chronic diseases rather than the facilitation of behavior care settings, trained in motivational interviewing (MI) and
change to moderate or control these conditions (6, 7). behavior change techniques, delivering face-to-face and telephone
Our work in China has focused on designing diabetes manage- coaching to patients with T2DM. MI is a way of communicating
ment programs that can be delivered effectively and efficiently in with patients that is collaborative in style and focused on how
primary health care settings. Over the last decade the number of patients talk about change (13). In order to facilitate change, it
cases of DM in China has increased to the extent that China now is assumed that the patient needs to elicit their own ideas about
has the highest number of DM cases globally. Xu et al. (8) estimated change as they will then be more likely to act. MI is founded upon

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 181 | 180
Browning et al. The Happy Life Club

an attitude of acceptance and compassion. It aims to strengthen system (17). The training program consists of a self-learning pack-
the patient’s motivation to achieve a goal by resolving ambivalence. age and health coach skills workshops. The self-learning package
MI is linked to the stages of change approach whereby the coach included key concepts in patient-centered care, health psychology
assists the patient to work through stages of change, from no inten- and behavior change approaches, the epidemiology of diabetes,
tion to change to a commitment to change and action (14). The and the role of MI in behavior change. The self-learning package
MI approach has been shown to be effective in achieving glycemic was followed by a 2-day intensive MI workshop. This workshop
control in adults with T2DM but to date the approach has not been covered the concepts and spirit of MI including: promoting a
tested widely in the Chinese context (15). Therefore, the purpose patient-centered approach and a collaborative coach style that
of this paper is to report findings from the pilot HLC™ study and focused on the stage of behavior change of targeted lifestyle behav-
discuss issues involved in implementing a CDSMP developed for iors relevant to chronic disease management; eliciting patient’s
Western primary care in a Chinese setting. intrinsic motivation to change; promoting client choice; building
self-efficacy; and resolving patient ambivalence. The workshop
MATERIALS AND METHODS included the application and practise of MI core skills: the use of
SAMPLE AND PROGRAM SETTING open-ended questions; affirmation; reflection, and summarizing
The data reported in this article are based on a 12-month pilot across the behavior change process. During the implementation
study of n = 100 patients of age 55 years and over with Type of the HLC™, refresher workshops were conducted and, 1 month
2 Diabetes Mellitus (T2DM). The pilot study was conducted in after the initial training, the coaches participated in a further
Fangzhuang. The Fangzhuang community is located in the south half-day advanced training workshop.
of Beijing and has a resident population of 110,000; 21.4% of
which are aged 60 years and above. The community has an estab-
MEASURES AND STATISTICAL PROCEDURES
lished community health service system that includes a large
Community Health Center (CHC) or community hospital which, Clinical, self-reported health, and well-being measures, and health
administrated by the local government, functions as the main pri- service use were collected at baseline, 6 and 12 months. Clinical
mary health care provider. The CHC includes five community measures included HbA1c, blood pressure, waist circumference,
health stations (CHSs), which aim to serve the health needs of and BMI. Quality of life was measured using the WHOQOL-BREF
the local communities. Participants were approached consecu- (18). Psychological well-being was measured using the Kessler Psy-
tively as they attended their usual diabetes appointment and asked chological Distress Scale (K10) (19). Participants were asked how
if they wished to participate in the study. Recruitment contin- often they had visited the doctor in the last 6 months. Differences
ued until 100 patients had agreed to participate. The patients between the control and intervention groups at baseline, 6, and
were randomly allocated to the intervention group or the con- 12 months were assessed using repeated measures ANOVA with
trol group (see below). Health professionals in the CHC and the the control and intervention groups as between-subjects factors.
CHSs are government employed doctors and nurses. The pilot Effect sizes were calculated.
study was subsequently expanded into a pragmatic cluster ran-
domized controlled trial (16). The study was approved by the RESULTS
Monash University Human Research Ethics Committee. Table 1 shows the baseline characteristics of the participants
(n = 100). At the 6-month follow-up, n = 5 participants were lost
THE INTERVENTION to follow-up: one participant died and four participants moved
The HLC program uses trained health coaches. In the pilot study, house and could not be contacted. There were no differences
the control group received usual care provided by a family physi- between the groups in terms of key demographic variables at base-
cian where patients are typically referred to diabetes specialists line except that the control group participants were more highly
and/or Traditional Chinese Medicine (TCM) practitioners. The educated. Sixty-seven percent of the total sample was women.
intervention group received telephone and face-to-face coaching
in addition to usual care. The key components of the interven-
tion were patient-centered care and the use of MI (13) to help Table 1 | Baseline characteristics of participants.
effect change in diet, physical activity, and general chronic dis-
ease self-management behaviors. In the first 3 months, participants Baseline characteristics Control Intervention Total
received two face-to-face and two telephone coaching sessions per
Participants, n 50 50 100
month after which, as the participants gained confidence in self-
Age in years, mean ± SD 63.3 ± 7.8 65.8 ± 7.5 64.2 ± 7.7
management, the frequency diminished. Overall, the intervention
Female 33 34 67
group received a maximum of 19 telephone coaching and 18 face-
Married (including de facto) 47 44 91
to-face coaching sessions. The intervention ran for 12 months.
Retired 44 46 90
COACH TRAINING Education
The health coaches (experienced doctors and nurses) received a Primary or less 7 13 20
certified training program. Doctors and nurses were chosen to Secondary/high school 25 34 59
deliver the intervention as they are by far the main providers Tertiary/technical 18 3 21
of health care in China. Other health professionals such as dia- Duration of diabetes in 8.2 ± 6.1 9.0 ± 6.3 8.6 ± 6.2
betes educators are virtually non-existent in the Chinese health years, mean ± SD

www.frontiersin.org April 2015 | Volume 2 | Article 181 | 181

Browning et al. The Happy Life Club

Table 2 | Mean scores and SD for clinical, self-reported health, and well-being measures and doctor visits at baseline, 6 and 12 months for the
control and intervention groups (n = 100).

Measure Baseline (N = 100) 6 months (N = 95) 12 months (N = 95) Effect Size Effect Size
from group from time
(partial eta (partial eta
squared) squared)

Control Intervention Control Intervention Control Intervention

HbA1c 7.00 ± 0.81 7.16 ± 1.11 6.96 ± 0.92 6.88 ± 1.10 7.16 ± 1.16 6.88 ± 0.88 0.036 (small 0.029 (small
effect) effect)

Systolic blood 128.2 ± 12.7 132.0 ± 15.4 128.6 ± 15.0 129.2 ± 12.1 129.1 ± 12.6 125.0 ± 12.4 0.054 (small 0.026 (small
pressure (mmHg) effect) effect)

Diastolic blood 76.5 ± 6.9 77.6 ± 8.1 75.9 ± 6.7 76.6 ± 6.9 76.5 ± 7.8 76.2 ± 7.4 0.008 0.007
pressure (mmHg)

Waist circumference 91.82 ± 7.10 90.56 ± 9.19 89.86 ± 7.25 89.02 ± 8.89 89.13 ± 7.14 88.59 ± 9.26 0.004 0.088 (small-
(cm) medium effect)

BMI (kg/m2) 25.37 ± 2.64 25.60 ± 3.40 25.47 ± 2.35 25.63 ± 3.27 25.44 ± 2.39 25.60 ± 3.41 0.001 0.001

WHOQOL-BREF 3.40 ± 0.85 3.34 ± 0.67 3.40 ± 0.74 3.44 ± 0.77 3.36 ± 0.49 3.83 ± 0.81 0.047 (small 0.044 (small
effect) effect)

K10 19.0 ± 6.4 17.3 ± 6.3 16.9 ± 6.7 15.1 ± 6.2 15.9 ± 5.0 13.8 ± 3.7 0.001 0.159 (medium
effect)

Number of community 5.38 ± 3.33 5.17 ± 2.49 5.70 ± 4.21 6.88 ± 4.95 5.68 ± 3.87 6.98 ± 4.97 0.016 0.032 (small
doctor visits effect)

Partial eta squared values: small effects indicated by 0.02, medium effects by 0.13, and large by 0.26.

Table 2 shows a comparison between the control and interven- groups. The effect size for changes over time was medium. In terms
tion groups at baseline, 6, and 12 months on key clinical and health of visits to the community doctor in the last 6 months, both groups
measures. showed a significant increase in doctor visits between baseline and
There was a significant interaction effect between HbA1c 6 months (F = 4.844, p = 0.030), but no change between 6 and
and group over the period baseline to 6 months (F = 7.098, 12 months. The effect size for changes over time was small.
p = 0.009). The intervention group showed significant improve-
ment in HbA1c levels between baseline and 6 months. How- DISCUSSION
ever, the effect size was small. Neither group showed changes The pilot study demonstrated that a CDSMP using Western con-
in HbA1c over the period 6–12 months. There was a signifi- cepts of behavior change and MI has an effect on the management
cant interaction effect between systolic blood pressure and group T2DM particularly in terms of the key physiological parameters of
(F = 5.194, p = 0.006), indicating that the intervention group sig- HbA1c levels and systolic blood pressure. However, the effect sizes
nificantly improved over time compared to the control group. were small. By 6 months, the intervention group had achieved
Again, the effect size was small. Diastolic blood pressure and BMI the goal of an HbA1c <7% and this may have contributed to
did not change across the intervention period for either group; no further significant reductions in HbA1c at 12 months. The
however, waist circumference decreased for both groups over intervention group also showed improvements in quality of life
time (F = 8.591, p < 0.001). There was no significant difference across the intervention and both groups showed reductions in
between the groups in terms of decrease in waist circumference. psychological distress.
The effect size was small to medium. Both groups showed improvements in some of the clinical and
There was a significant interaction effect between quality of life health indicators. This may be due to participation effects, with no
and group (F = 4.612, p = 0.011). The intervention group showed differential effect due to the intervention. In China, people with
improvement in quality of life across the intervention. The effect T2DM do not regularly monitor their condition, including HbA1c
size was small. The control group showed no significant change in levels, due to cost. The control group received feedback about their
quality of life across the intervention period. Both groups showed HbA1c levels and this may have motivated them to implement
a decrease in scores on the K10 between baseline and 6 months self-management approaches (Hawthorne effect). The study was
(F = 11.306, p < 0.001) and between 6 and 12 months (F = 4.577, conducted in a residential area where there was the potential for
p = 0.035), but there were no significant differences between the contamination between the groups. The participants lived in the

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 181 | 182
Browning et al. The Happy Life Club

same building or residential area and potentially had close inter- modest effects found in this pilot study, we were pleasantly sur-
action when shopping or participating in community activities. prised by how well our CDSMP was adopted by both the coaches
Participants in the intervention group may have discussed their and the patients. We could not put it better than one of our
coaching with other residents. While the coaches were asked not coaches:
to use MI with their other patients, it is difficult to control this.
MI is a powerful tool. Since using it, both me and the patient
Coaches may have used the techniques with patients outside the
have opened our mouths, and have more conversations. . . We
intervention group.
work together and start talking about what small steps to take,
While one of the aims of a CDSMP is to reduce hospital-based
and what is the easiest way to go. . . Patients then have their
specialists’ visits in order to reduce health care costs, in our pilot
ideas of targets and a plan. I encourage them constantly. . . It
community, doctor (general practitioner) visits increased. We con-
does not work immediately, but several months later, I find
cluded that this increase was largely due to an improvement in
they really get success. . . if one can maintain a new behav-
doctor–patient relationships. In China, dissatisfaction with the
ior for 3–6 months, the behavior seems to be a stable life
services provided by hospital-based doctors is very high (20). The
habit.
HLC pilot may have increased the participant’s confidence in gain-
ing a higher quality of care from their community doctors thus The patients showed similar positive views about their experiences
increasing primary health service use. of the intervention:
There are few trials of behavioral and psychological approaches
Before this project, I was quite negative about my disease and
to the management of T2DM in China. One recent but small 12-
for everything I just relied on my doctors. But now, I can
week intervention (n = 40) that used cognitive behavior therapy
manage the disease by myself. If my blood sugar level is high
(CBT), found that the CBT group showed reductions in fast-
I will try to find the reason by myself first . . . because I am
ing glucose, HbA1c, and depression compared to the usual care
the person who knows me better. I do not feel the disease is a
group (21). A systematic review of lifestyle interventions aimed
huge burden to me anymore and that is really good. (Female,
at preventing T2DM in developing countries (22) identified only
61 years old, duration of T2DM 8 years)
one Chinese study (23). There is a pressing need to rigorously
evaluate the different behavioral and psychological approaches Our qualitative results support the view that patients appear to
to the management of T2DM in China, particularly when inter- have benefited from the approach in terms of changing health
ventions that have only been proven effective in Western settings behaviors and gaining confidence in managing their T2DM.
are used. Patients in the intervention group were also able to reach the
We have attempted to address the issues raised by this pilot HbA1c goal of 7% and improve systolic blood pressure. How-
study in the full pragmatic cluster randomized controlled trial ever, we need stronger evidence to conclude that our approach
(11) where there is more geographical separation between the will lead to long-term changes in T2DM management.
groups. The full trial is also sufficiently powered to detect dif-
ferences between the groups. The pilot study was conducted in a REFERENCES
relatively high SES urban area in Beijing. Its applicability in rural 1. Yang G, Kong L, Zhao W, Wan X, Zhai Y, Chen LC, et al. Emergence of chronic
and low SES areas is unknown and requires further testing. In the non-communicable diseases in China. Lancet (2008) 372(9650):1697–705.
doi:10.1016/S0140-6736(08)61366-5
full trial, which includes sampling across SES groups, the influence 2. Ng SW, Norton EC, Popkin BM. Why have physical activity levels declined
of SES will be examined. among Chinese adults? Findings from the 1991-2006 China health and nutri-
CDSM approaches require a health workforce that is highly tion surveys. Soc Sci Med (2009) 68(7):1305–14. doi:10.1016/j.socscimed.2009.
trained in communication skills, patient-centered approaches, and 01.035
3. Editorial. China must not neglect reform of primary care. Lancet (2005)
behavior change frameworks and skills. Traditionally, there has
365(9468):1362. doi:10.1016/S0140-6736(05)66346-5
been little focus on these skills in health care practitioner train- 4. Liu Q, Wang B, Kong Y, Cheng KK. China’s primary health-care reform. Lancet
ing in China (24). Behavior change and psychological approaches (2011) 377:2064–6. doi:10.1016/S0140-6736(11)60167-0
to the prevention and management of chronic illnesses are lim- 5. Yang H, Thomas SA, Browning CJ. Chronic disease management in commu-
ited in China as there are only 2.4 professionals with psychology nity health. In: Li ZX, Zhang Y, Yang H, editors. Community Health Services
Management. Beijing: People’s Military Medicine Publisher (2010). p. 65–97.
training per 1 million of the population (25) compared to 3500
6. Bhattacharyya O, Dulu Y,Wong S. Evolution of primary care in China 1997-2009.
nurses and physicians per 1 million of the population (17). Con- Health Policy (2011) 100(2011):174–80. doi:10.1016/j.healthpol.2010.11.005
sequently, psychological and behavioral approaches to health and 7. Browne D. The long march to primary health care in China: from collectivism
illness have not been widely endorsed either by medical practition- to market economics. Public Health (2001) 115:2–3. doi:10.1038/sj.ph.1900742
ers, policy makers, or the general population. We were therefore 8. Xu Y, Wang L, Bi Y, Li M, Wang T, Wang L, et al. Prevalence and control of
diabetes in Chinese adults. JAMA (2013) 310(9):948–59. doi:10.1001/jama.2013.
interested in whether a Western model of CDSM would suc- 168118
cessfully translate into a Chinese setting. It would seem that 9. Yang W, Lu J, Weng J. Prevalence of diabetes among men and women in China.
health literacy concerning chronic illnesses in China is improv- N Engl J Med (2014) 362:1090–101. doi:10.1056/NEJMoa0908292
ing, especially among women and those with higher SES (“a 10. International Diabetes Federation. New Diabetes Figures in China: IDF
long illness turns a patient into a doctor”) (26). In addition, Press Statement. IDF (2010). Available from: http://www.idf.org/press-releases/
idf-press-statement-china-study
the holistic mind–body approach embodied in TCM, where it 11. Pan XR, Yang WY, Li GW, Liu J. Prevalence of diabetes and its risk factors in
is assumed that health is governed by emotion and thoughts, China, 1994, national diabetes prevention and control cooperative group. Dia-
is actually consistent with CDSM approaches (27). Despite the betes Care (1997) 20(11):1664–9. doi:10.2337/diacare.20.11.1664

www.frontiersin.org April 2015 | Volume 2 | Article 181 | 183

Browning et al. The Happy Life Club

12. Kelly J, Menzies D, Taylor S. The good life club: methodology and study 25. Han B, Kan Z. Psychology in China. Psychologist (2007) 20:734–6. Avail-
design – a discussion. Aust J Prim Health (2003) 9(2&3):186–91. doi:10.1071/ able from: http://www.thepsychologist.org.uk/archive/archive_home.cfm/
PY03045 volumeID_20-editionID_154-ArticleID_1280-getfile_getPDF/thepsychologist/
13. Miller WR, Rollnick S. Motivational interviewing: Helping people change. New 1207chin.pdf
York: Guilford Press (2013). 26. Song Y, Ma W, Yi X, Wang S, Sun X, Tian J, et al. Chronic diseases knowledge and
14. Prochaska JO, Velicer WF. The transtheoretical model of health behavior change. related factors among the elderly in Jinan, China. PLoS One (2013) 8(6):e68599.
Am J Health Promot (1997) 12(1):38–48. doi:10.4278/0890-1171-12.1.38 doi:10.1371/journal.pone.0068599
15. Ismail K, Winkley K, Rabe-Hesketh S. Systematic review and meta- 27. Tan C, Chen W, Wu Y, Chen S. Chinese medicine for mental disorder
analysis of randomised controlled trials of psychological interventions to and its applications in psychosomatic diseases. Altern Ther Health Med
improve glycaemic control in patients with type 2 diabetes. Lancet (2004) (2013) 19(1):59–69. Available from: http://todayspractitioner.com/wp-
363(9421):1589–97. doi:10.1016/S0140-6736(04)16202-8 content/uploads/2013/11/Chinese-Medicine-for-Mental-Disorder-and-its-
16. Browning CJ, Chapman A, Cowlishaw S, Li Z, Thomas SA, Yang H, et al. The Applications-in-Psychosomatic-Diseases-191_ATM_p59_69tan.pdf
happy life club™ study protocol: a cluster randomised controlled trial of a
type 2 diabetes health coach intervention. BMC Public Health (2011) 11:90. Conflict of Interest Statement: The authors declare that the research was conducted
doi:10.1186/1471-2458-11-90 in the absence of any commercial or financial relationships that could be construed
17. Hou J, Michaud C, Li Z, Dong Z, Sun B, Zhang J, et al. Transformation of the edu- as a potential conflict of interest.
cation of health professionals in China: progress and challenges. Lancet (2014)
384(9945):819–27. doi:10.1016/S0140-6736(14)61307-6 This paper is included in the Research Topic, “Evidence-Based Programming for Older
18. World Health Organization. WHO Quality of Life-BREF (WHOQOL-BREF). Adults.” This Research Topic received partial funding from multiple government and
(1991). Available from: http://www.who.int/substance_abuse/research_tools/ private organizations/agencies; however, the views, findings, and conclusions in these
whoqolbref/en/ articles are those of the authors and do not necessarily represent the official position
19. Kessler RC, Anderws G, Colpe LJ, Hiripi E, Mroczek DK, Normand SLT, et al. of these organizations/agencies. All papers published in the Research Topic received
Short screening scales to monitor population prevalence and trends in non- peer review from members of the Frontiers in Public Health (Public Health Education
specific psychological distress. Psychol Med (2002) 32:959–76. doi:10.1017/ and Promotion section) panel of Review Editors. Because this Research Topic repre-
S0033291702006074 sents work closely associated with a nationwide evidence-based movement in the US,
20. Thomas SA, Yang H, Browning C, Zhang T, Xiaofei L, Ren Y, et al. Quality of many of the authors and/or Review Editors may have worked together previously in
care in different community health facilities in China: from the patient’s point some fashion. Review Editors were purposively selected based on their expertise with
of view. J Chin Gen Pract (2007) 10(27):1760–4. doi:10.3969/j.issn.1007-9572. evaluation and/or evidence-based programming for older adults. Review Editors were
2007.21.002 independent of named authors on any given article published in this volume.
21. Sun H, Zhang Q. Effects of psychological interventions on depression and glu-
cose metabolism of low-income people with type 2 diabetes. Chin Clin Pract Received: 16 July 2014; accepted: 22 September 2014; published online: 27 April 2015.
Med (2009) 3(11):96–7. doi:10.3760/cma.j.issn1673-8799.2009.11.62 Citation: Browning CJ, Yang H, Zhang T, Chapman A, Liu S, Enticott J and Thomas
22. Rawal LB, Tapp RJ, Williams ED, Chan C, Yasin S, Oldenburg B. Prevention SA (2015) Implementing a chronic disease self-management program into China: the
of type 2 diabetes and its complications in developing countries: a review. Int Happy Life Club™. Front. Public Health 2:181. doi: 10.3389/fpubh.2014.00181
J Behav Med (2012) 19:121–33. doi:10.1007/s12529-011-9162-9 This article was submitted to Public Health Education and Promotion, a section of the
23. Li G, Zhang P, Wang J, Gregg EW, Yang W, Gong Q, et al. The long-term journal Frontiers in Public Health.
effect of lifestyle interventions to prevent diabetes in the China Da Qing Copyright © 2015 Browning , Yang , Zhang , Chapman, Liu, Enticott and Thomas.
diabetes prevention study: a 20-year follow-up study. Lancet (2008) 371:1783–9. This is an open-access article distributed under the terms of the Creative Commons
doi:10.1016/S0140-6736(08)60766-7 Attribution License (CC BY). The use, distribution or reproduction in other forums is
24. Qian M, Juhua M. Problems and solutions to the teaching of a doctor-patient permitted, provided the original author(s) or licensor are credited and that the original
communication course. Res Med Educ (2012) 9(8):1047–9. doi:10.3760/cma.j. publication in this journal is cited, in accordance with accepted academic practice. No
issn.2095-1475.2010.08.012 use, distribution or reproduction is permitted which does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 181 | 184
 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00162

Implementing chronic disease self-management

approaches in Australia and the United Kingdom
Colette Joy Browning 1,2 * and Shane Andrew Thomas 2,3
1
Royal District Nursing Service Institute, St Kilda, VIC, Australia
2
School of Primary Health Care, Monash University School of Primary Health Care, Monash University, Notting Hill, VIC, Australia
3
The University of Adelaide, Adelaide, SA, Australia
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: chronic illness prevalence, self-management policy and practice, training and workforce needs, Australia, United Kingdom

Most nations have responded to the cur- in a shared care model. Other primary their development and progression. This
rent and projected burden of chronic ill- health practitioners, including psycholo- is reflected in the Australian Institute for
nesses by promoting patient centered care gists and allied health practitioners, have Health and Welfare’s 2012 report on Risk
and self-management approaches (1). In access to Medicare funding for patients fol- Factors contributing to Chronic Disease
the current paper, we focus on Australia lowing GP referral and shared case man- (7). It asserts:
and the UK where chronic illness has a agement. For Australian patients with a
The development of chronic diseases
major impact upon the burden of disease chronic illness, the GP may devise, in con-
is strongly associated with the behav-
on individuals, society and its institutions, sultation with the patient, a Chronic Dis-
ioural risk factors of smoking, physi-
and the use of health services1 . Thus of ease Management Plan3 and/or a Team
cal inactivity, poor diet and the harm-
necessity, Australian and UK health pol- Care Arrangements Plan4 . The Plans iden-
ful use of alcohol. These behaviours
icy, funding, and service delivery have a tify the patient’s health care needs, spec-
can contribute to the development
strong focus upon chronic disease and its ify the services to be provided by the GP
of biomedical risk factors, such as
treatment and prevention. It is notewor- and other health professionals, and out-
high blood pressure, obesity and high
thy that both the UK and Australia fund line the actions that the patient needs
cholesterol. (p. 9)
individual health care costs through uni- to take. Detailed Health Assessments are
versal insurance paid from general taxation also funded utilizing Health Assessment A very useful aspect of this approach is
rather than via a user pays model and this Proformas. that it not only focuses on the epidemiol-
strongly and positively impacts on service Australia’s universal insurance access to ogy of chronic illnesses but it also focuses
access and equity. Further, both countries’ support the diagnosis and management of on the epidemiology of the underlying
policies on chronic disease management chronic illness is a stand out feature of its risk and protective factors that directly
have been influenced by Wagner’s Chronic health system (5). While chronic illnesses influence the development and progres-
Care Model (3) and Lorig’s chronic disease have major well-being, social, and finan- sion of the illnesses. This is a useful and
self-management program (4). cial effects, in Australia, the costs to the appropriate focus that has been reflected
individual of health care are minimized in the activities of many government-
AUSTRALIA compared to other countries, although it funded bodies such as VicHealth5 since
The Australian health system relies heav- seems that the new national government the mid 1980s. VicHealth is a state agency
ily on its primary health care system and may attempt to reduce costs to govern- focused on health promotion. While it
the Medicare2 universal insurance cover- ment by increasing the contributions of was initially funded by tobacco taxes and
age scheme to deliver and fund health individuals to their health care costs (6). focused on smoking cessation, VicHealth
care services to its citizens and permanent A further standout feature of the Aus- has expanded into much broader programs
residents. Primary health care physicians tralian approach to chronic illness has of health promotion and prevention of
(General Practitioners: GPs) provide the been the recognition that the key to long- chronic disease. The other Australian states
bulk of medical services and primary health term population control of chronic illness have also established similar bodies focused
care in Australia. Access to specialist med- is best obtained through modification of on chronic disease reduction (e.g., Health-
ical care is obtained by referral from GPs risk and protective factors underpinning way in Western Australia, and programs

1 For an overview of specific country approaches to chronic disease management the reader is referred to the European Observatory on Health Systems and Policies paper

on managing chronic illness in eight countries including European countries, Australia and the United Kingdom (2).
2 http://www.humanservices.gov.au/customer/dhs/medicare
3 http://www.humanservices.gov.au/customer/services/medicare/chronic-disease-management-plan
4 http://www.health.gov.au/internet/main/publishing.nsf/content/mbsprimarycare-chronicdiseasemanagement
5 http://www.vichealth.vic.gov.au

www.frontiersin.org April 2015 | Volume 2 | Article 162 | 185

Browning and Thomas Disease self-management in Australia and the UK

such as OPAL – Obesity Prevention and government’s success will often the integration of health and social care
Lifestyle in South Australia). Most states depend on their ability to implement and was based on extensive community
run significant programs in smoking cessa- effective behaviour change interven- consultations. Patients with high, medium,
tion, obesity reduction, sexual health, and tions whilst, at the same time, avoiding and low risk of poor outcomes were
lifestyle modification. significant harmful side effects. linked to three different tiers of inter-
At the national level, the Sharing Health vention, respectively: case management,
Care Initiative (SHCI) (8) 2002–2004 pro- In Australia, while there is a strong disease management, and self-care man-
vided a focus upon alternative approaches psychologist workforce, the training in agement (17). Case management is the
to CDSM for the purposes of formulating behavior change principles and concepts responsibility of “Community Matrons”
national policy. The $36.2 million initia- amongst other clinicians, especially med- (nurses) who are responsible for health and
tive tested a range of chronic disease self- ical practitioners, is patchy. The Happy Life social care, and support self-management
management models that could be suit- Club™(12) clinical research program and to reduce hospital admissions.
able for the Australian health care sys- its Australian predecessor the Good Life
Despite early optimism that the EPP
tem and incorporation in the subsequent Club (13) have demonstrated how train-
would improve health and reduce health
2005 National Chronic Disease Strategy ing doctors and nurses can deliver robust
care costs, some critics have questioned its
(NCDS) (9). The Initiative was a model for improvements in chronic illness such as
value. Griffiths and colleagues (18) noted
evidence-based policy and it led to a high diabetes. Thus, we consider that training
that while evaluations of the UK EPP found
degree of agreement about the appropri- of the wider clinical workforce in behavior
improvements in patient self-efficacy, self
ate frameworks and policies for CDSM in change principles and practice is a priority
rated health, and the use of health ser-
Australia. for the effective prevention and manage-
ment of chronic illness in Australia. vices remained unchanged. They also noted
However, while the basic policy set- that UK CDSMP led by health profession-
tings and approaches concerning chronic als have shown stronger effects in people
illness have been agreed for some time UNITED KINGDOM
with specific chronic conditions such as
in Australia, the structure of the bod- Primary health care is fundamental to the
heart disease and diabetes. Greenhalgh (19)
ies coordinating these efforts is again in delivery of health services in the UK. The
argued that the EPP in the UK has failed
flux. Recently, the Australian Government National Health Service funds primary and
to take account of the impact of economic
announced that it was going to discontinue specialist care and patients register with a
conditions, social support, health literacy,
the national Australian National Preven- practice of their choice (14). As in Australia,
and cultural norms in CDSM. She pro-
tive Health Agency (ANPHA)6 and relocate British GPs play a “gatekeeping” triage role
posed a social ecology approach whereby
its functions within the Commonwealth through a referral system to specialists.
the responsibility for the prevention and
Department of Health and to terminate the Health services are essentially free except
management of chronic illness rests with
National Partnership Agreement on Pre- for medications and dental and optometry
individuals, health professionals, and the
ventive Health (10). The National Part- care. This contrasts with Australia where
some patient co-payments for GP services wider society and recognizes the social
nership Agreement on Preventive Health
are now common. determinants of health. In a recent review
announced by the Council of Australian
The modern UK health policy approach of EPP, Vadie (20) noted that the EPP has
Government on 29 November 2008 was
to chronic illness management occurred at not fostered alliances between patients and
also terminated on 30 June 2014.
a similar time to Australia’s response. In health professionals and generally there has
Notwithstanding the strong and long-
1999, in recognition of the growing preva- been a lack of engagement with the pro-
standing emphasis upon behavior risk and
lence of chronic illness and the complex- grams by clinicians. Further, the program
protective factors, Australia has a short-
ity of patient needs, the UK government has failed to reach those who are most
age of practitioners trained in the use of
proposed more involvement by patients disadvantaged (20).
behavior change techniques to promote the
prevention and effective management of in decision making about their care (15). Despite these criticisms, self-
chronic illness. While there is wide recog- An outcome of this approach was the management approaches are strongly
nition of the benefits of behavior change Expert Patient Program (EPP), which com- endorsed within the UK health care system
approaches, it falls short of the whole of menced in 2002, and was based largely and CDSMP have evolved and incorpo-
government approach taken by, for exam- on Lorig’s generic lay led CDSMP (4). rated new models in response to early
ple, the UK House of Lords Science and With the growing recognition of the need criticisms. Currently, a number of not-
Technology Select Committee review (11) for integrated care for people with mul- for-profit agencies are engaged with the
of behavior change approaches. The Com- tiple chronic conditions, UK health pol- NHS in delivering innovative CDSMP pro-
mittee noted that: icy later explicitly incorporated Wagner’s grams. For example, Self-management UK
Chronic Care Model and the Kaiser risk (21) is a key provider of self-management
The aim of much government pyramid model into its chronic disease programs in the UK. It also provides a con-
policy is to bring about changes management approach (16, 17). The NHS sultancy service for NHS clinicians to help
in people’s behaviour and so a Health and Social Care model focused on them design and implement programs that

6 www.anpha.gov.au/

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 162 | 186
Browning and Thomas Disease self-management in Australia and the UK

are locally responsive. Their program Self both countries have shown support for co-payments-a-triple-fail-for-the-commission-
Management for Life attempts to address CDSMP and programs have evolved over of-audit-26195
7. Australian Institute of Health and Welfare. Risk
early criticisms of the EPP by encouraging the last 15 years to respond to gaps in
Factors Contributing to Chronic Disease. Cat No.
better communication between the patient delivery; however, the current Australian PHE 157. Canberra: AIHW (2012).
and the health care team. Similarly, the Government seems somewhat less com- 8. Australian Government Department of Health.
UK Health Foundation has developed mitted to preventive approaches than its National Evaluation of the Sharing Health
Co-creating Health that aims to imbed predecessor as evidenced by the down- Care Initiative Final Technical Report. Canberra:
Australian Government Department of Health
self-management in mainstream health grading of the ANPHA (see text foot- (2005).
services (22). It incorporates Wagner’s note 6) and its greater reliance upon co- 9. National Health Priority Action Council. National
Chronic Care Model, self-management payment patient funding initiatives. A key Chronic Disease Strategy. Canberra: Australian
support, and collaboration between the issue for the delivery of CDSMP is the Government Department of Health and Ageing
(2006).
service providers and patients in planning quality of clinician skills and training. In
10. Council of Australian Governments. National
and delivery. The model trains patients in the modern crowded curriculum, many Partnership Agreement on Preventive Health.
self-management, trains clinicians in self- medical and health care undergraduate (2008). Available from: http://www.federalfinan-
management support skills, and addresses degrees pay scant attention to effective cialrelations.gov.au/content/npa/health_preven-
system level processes to support effective patient–clinician communication, behav- tive/national_partnership.pdf
11. House of Lords. 2nd Report of Session 2010–12.
and efficient chronic disease management. ior change skills, patient centered care, and Behaviour Change Report, Science and Technology
An innovative key feature of the model social determinants of health despite the Select Committee. London: The Stationery Office
is co-production where both the patient recognition of their importance in patient Limited (2011).
and the clinician training are delivered by care (23). A recent review of behavior 12. Browning C, Thomas S. Six-month outcome data
for the Good Life Club project: an outcomes study
a clinician and a layperson living with a change counseling curricula for medical
of diabetes self-management. Aust J Prim Health
chronic condition. students found that the majority of stud- (2003) 9(3):192–8. doi:10.1071/PY03046
The UK experience recognizes the ies reported only eight or less curricu- 13. Browning C, Chapman A, Cowlishaw S, Li X,
importance of changing practice among lum hours devoted to these fundamen- Thomas S, Yang H, et al. The Happy Life Club™
clinicians as well as changing patient tal skills (24). In order to embed CDSM study protocol: a cluster randomised controlled
trial of a type 2 diabetes health coach interven-
behaviors. The evaluation of the Co- approaches in our health systems, it is nec- tion. BMC Public Health (2011) 11:90. doi:10.1186/
creating Health program identified four essary to create a workforce that under- 1471-2458-11-90
ways to promote clinician practice change stands the importance of these approaches 14. Roland M, Guthrie B, Thomé D. Primary med-
to support the sustainability of self- in delivering quality health outcomes and ical care in the United Kingdom. J Am Board Fam
Med (2012) 25:S6–11. doi:10.3122/jabfm.2012.02.
management approaches and embed these who will champion genuine partnership
110200
approaches within the health care services approaches to the management of chronic 15. Department of Health. Saving Lives, Our Healthier
(22). Targeting whole teams, utilizing influ- illness. Nation. London: Stationery Office (1999).
ential clinicians, providing support to clin- 16. Department of Health. Supporting People with
icians after the initial training program, REFERENCES Long-Term Conditions. An NHS and Social Care
1. Newman S, Steed L, Mulligan K. Self-manage- Model to Support Local Innovation and Integration.
and incorporating self-management skills ment interventions for chronic illness. Lancet London: The Stationery Office (2005).
training in health care education were (2004) 364:1523–37. doi:10.1016/S0140-6736(04) 17. Rosen R, Asaria P, Dixon A. Improving Chronic Dis-
key recommendations. The emphasis on 17277-2 ease Management: An Anglo-American exchange.
early skills training and ongoing profes- 2. Nolte E, Knai C, McKee M. Managing Chronic London: King’s Fund (2007).
Conditions: Experience in 8 Countries. Observa- 18. Griffiths C, Foster G, Ramsay J, Eldridge S,
sional development in the area of self-
tory Study Series No. 15 European Observatory on Taylor S. How effective are expert patient (lay
management support and behavior change Health Systems and Policies. WHO: Copenhagen led) education programmes for chronic dis-
principles is a strong feature of the current (2008). ease? BMJ (2007) 334:1254–1256. doi:10.1136/
UK approach. 3. Wagner E, Austin B, Von Korff M. Organizing care bmj.39227.698785.47
for patients with chronic illness. Milbank Q (1996) 19. Greenhalgh T. Chronic illness: beyond the expert
CONCLUSION 74:511–44. doi:10.2307/3350391 patient. BMJ (2009) 338:629–31. doi:10.1136/
4. Lorig KR, Sobel DS, Stewart AL, Brown BW bmj.b49
Both Australia and the UK face simi-
Jr, Bandura A, Ritter P, et al. Evidence suggest- 20. Vadiee M. The UK “Expert Patient Program” and
lar challenges in terms of the increas- ing that a chronic disease self-management pro- self-care in chronic disease management: an analy-
ing prevalence of chronic conditions and gram can improve health status while reduc- sis. Eur Geriatr Med (2012) 3:3201–5. doi:10.1016/
patients with complex health and social ing hospitalization. A randomized trial. Med j.eurger.2012.02.003
care needs. Both countries have adopted Care (1999) 37:5–14. doi:10.1097/00005650- 21. Self management UK. (2014). Available from: http:
199901000-00003 //selfmanagementuk.org/services/programmes
models of chronic disease management 5. Thomson S, Osborn R, Squires D, Jun M. Inter- 22. Newbronner L, Chamberlain R, Borthwick R,
that have their origins in the US. However national Profiles of Health Care Systems. The Baxter M, Sanderson D. Sustaining and Spread-
Australia and the UK fund their health sys- Commonwealth Fund (2013). Available from: ing Self-Management Support: Lessons from Co-
tems largely through general taxation and http://www.commonwealthfund.org/~/media/ Creating Health Phase 2. London: Health Founda-
therefore are in a better position than most files/publications/fund-report/2013/nov/1717_ tion (2013).
thomson_intl_profiles_hlt_care_sys_2013_v2.pdf 23. Institute of Medicine. Improving Medical Educa-
nations to provide accessible and equitable 6. Duckett S. GP Co-Payments: A Triple Fail for the tion: Enhancing the Behavioral and Social Science
health care for people living with the bur- Commission of Audit. The Conversation. (2014). Content of Medical School Curricula. Washington,
den of chronic illnesses. Governments in Available from: http://theconversation.com/gp- DC: National Academies Press (2004).

www.frontiersin.org April 2015 | Volume 2 | Article 162 | 187

Browning and Thomas Disease self-management in Australia and the UK

24. Hauer K, Carney P, Chang A, Satterfield J. Behavior position of these organizations/agencies. All papers pub- Citation: Browning CJ and Thomas SA (2015) Imple-
change counseling curricula for medical trainees: lished in the Research Topic received peer review from menting chronic disease self-management approaches in
a systematic review. Acad Med (2012) 87:956–68. members of the Frontiers in Public Health (Public Health Australia and the United Kingdom. Front. Public Health
doi:10.1097/ACM.0b013e31825837be Education and Promotion section) panel of Review Edi- 2:162. doi: 10.3389/fpubh.2014.00162
tors. Because this Research Topic represents work closely This article was submitted to Public Health Education
Conflict of Interest Statement: The authors declare associated with a nationwide evidence-based movement and Promotion, a section of the journal Frontiers in
that the research was conducted in the absence of any in the US, many of the authors and/or Review Editors Public Health.
commercial or financial relationships that could be may have worked together previously in some fash- Copyright © 2015 Browning and Thomas. This is an
construed as a potential conflict of interest. ion. Review Editors were purposively selected based on open-access article distributed under the terms of the
their expertise with evaluation and/or evidence-based Creative Commons Attribution License (CC BY). The
This paper is included in the Research Topic, “Evidence- programming for older adults. Review Editors were inde- use, distribution or reproduction in other forums is per-
Based Programming for Older Adults.” This Research pendent of named authors on any given article published mitted, provided the original author(s) or licensor are
Topic received partial funding from multiple government in this volume. credited and that the original publication in this journal
and private organizations/agencies; however, the views, is cited, in accordance with accepted academic practice.
findings, and conclusions in these articles are those of Received: 23 June 2014; accepted: 11 September 2014; No use, distribution or reproduction is permitted which
the authors and do not necessarily represent the official published online: 27 April 2015. does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 162 | 188
 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2015.00027

Cost-effectiveness of the chronic disease self-management

program: implications for community-based organizations
Rashmita Basu 1 *, Marcia G. Ory 2 , Samuel D. Towne Jr.2 , Matthew Lee Smith 3 , Angela K. Hochhalter 1 and
SangNam Ahn 4
1
Baylor Scott & White Health, Temple, TX, USA
2
Department of Health Promotion and Community Health Sciences, School of Public Health, Texas A&M Health Science Center, College Station, TX, USA
3
Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA, USA
4
Division of Health Systems Management and Policy, School of Public Health, The University of Memphis, Memphis, TN, USA

Edited by: Chronic conditions are the leading cause of growing healthcare spending, disability, and
Sue Ellen Levkoff, University of South
death in the U.S. In the wake of national health reform, policy makers and healthcare
Carolina, USA
professionals are becoming increasingly concerned in containing healthcare costs while
Reviewed by:
Sharyl Kidd Kinney, University of improving quality of patient care. A basic policy question is whether the Chronic Disease
Oklahoma College of Public Health, Self-Management Program (CDSMP), a widely distributed evidenced-based self-managed
USA program, can be cost-effective in managing chronic conditions while improving quality of
Sue Ellen Levkoff, University of South
life. Utilizing data from the National Study of CDSMP, the primary objective of the current
Carolina, USA
study is to estimate cost-effectiveness of the CDSMP program among individuals with at
*Correspondence:
Rashmita Basu, Baylor Scott & White least one chronic condition. The second objective is to determine how cost-effectiveness
Health, 2401 South 31st Street, ratios vary by depression status. EuroQol-5D (EQ-5D) was used to measure health-related
Temple, TX 76508, USA quality of life (HRQOL) of CDSMP participants, which was then converted to quality-
e-mail: [email protected]
adjusted life years (QALYs) for cost-effectiveness analysis. Participants who completed the
CDSMP program experienced higher EQ-5D scores from baseline to 12-month follow-up
(increased from 0.736 to 0.755; p < 0.001). The incremental cost-effectiveness ratio (ICER)
ranges from $83,285 to $31,285 per QALYs, which can be comparable to the common
benchmark of $50,000/QALYs. ICER by baseline depression status indicates that it will
cost more per QALYs gained for those diagnosed with depression based on their Patient
Health Questionnaire-8 score. However, cautions should be taken while considering this
point estimate too literally because the average cost for CDSMP participants was a rough
estimate and based on several simplifying assumptions. Identifying cost-effective strate-
gies that can lower the burden of chronic disease among community-dwelling adults is
critical for decision makers in allocating limited resources. Policy makers and community
organizations can use this information to guide funding decisions and delivery of CDSMP
programs for individuals with multiple chronic health conditions.
Keywords: chronic disease self-management, cost-effectiveness analysis, health-related quality of life, older adults,
EQ-5D and quality-adjusted life years

INTRODUCTION population surveillance of health-related quality of life (HRQOL)

With the rapid aging of the baby boomer cohort, it is estimated that (8). HRQOL is a multi-dimensional measure, which is defined as
one in five Americans will be 65 years or older by 2030 (1). Simul- “perceived physical and mental health over time” (9). It can be
taneously, the existence of multiple chronic conditions among considered as a part of a person’s overall quality of life that is
Americans 65 years or older is becoming increasingly prevalent, determined by his or her health status. Because HRQOL addresses
with 60–75% of older adults having at least two chronic conditions physical and mental health of a large number of individuals, it
(2), many of which are preventable (1). Moreover, the number can offer current health data that public health agencies need
of Americans with chronic conditions is projected to increase by to assess population health. In light of the growing prevalence
37% by the year 2030 (3). More than 75% of total healthcare costs of chronic illness, healthcare burdens, and concerns for pro-
are attributable to the treatment of chronic illnesses (3). Further- moting population health, service providers and policy makers
more, chronic conditions among older adults are associated with are pursuing cost-effective ways to design self-management pro-
lower quality of life and increased limitations in activities of daily grams that can improve the health and well-being of the popu-
living (4–7). lation (10, 11). As healthcare costs continue to rise for treating
With a mission to promote health and the quality of life in chronic diseases, identifying ways to manage the progression of
Americans, the U.S. Department of Health and Human Services’ multiple chronic conditions among older adults is critical and
Center for Disease Control and Prevention (CDC) has supported time-sensitive (12, 13).

www.frontiersin.org April 2015 | Volume 3 | Article 27 | 189

Basu et al. Cost-effectiveness of CDSMP

To promote health and the quality of life of community- healthy behavior such as physical exercise for maintaining and
dwelling older adults, federal, state, and local stakeholders are improving strength, flexibility, and endurance; and (c) appropri-
implementing evidence-based initiatives to engage individuals in ate use of medications, effective communication with healthcare
managing chronic health conditions while improving health out- professionals (24, 28). For the purpose of this study, participants
comes (14, 15). One such approach is the implementation of with complete information on indicators of HRQOL at baseline,
self-management programs that improve health and quality of 6-month, and 12-month follow-up were included.
life while simultaneously reducing costly healthcare utilization
(16–20). These self-management programs have the potential to Study sample
embrace the triple aim goals of healthcare (better care, improved As a part of translating this intervention, CDSMP included
patient care experience, and lower cost of care) that will enhance 1,170 community-dwelling individuals at baseline, 6 months, and
population health (21). 12 months across the nation. A total of 825 (71%) participants
The Chronic Disease Self-Management Program (CDSMP), completed 12-month follow-up assessment including HRQOL
one of the well-studied evidence-based programs, improves health measures and approximately 77% (n = 903) participants com-
status and chronic illness symptoms while showing promise for pleted 6-month follow-up (29). While attrition was minimal for
lowering healthcare spending through the reduction in hospital- a community-based translational research study, HRQOL infor-
ization (22, 23). Although evidence suggests that CDSMP can mation at the 6-month follow-up data was missing for 77 partici-
improve health outcomes among patients with chronic diseases pants (N = 748 contributed to the final analyses). Few differences
(24–26), little is known about the cost-effectiveness of improv- were observed based on data attrition. Participants who com-
ing HRQOL among CDSMP participants. Moreover, program pleted follow-up assessments at 6-month and 12-month tended
effectiveness may vary when chronic diseases are accompanied to be older, and completers of the 6-month assessment were
by depression because individuals with depression are less likely more likely to be non-Hispanic White (15). Institutional Review
to complete the self-management education programs than those Board approval for the National Study was obtained at Stanford
without depression (27). Thus, the current study has two goals: (1) University and Texas A&M University.
to perform an economic evaluation of the CDSMP by utilizing a
cost-effectiveness analysis of HRQOL among CDSMP participants MEASURES
from baseline to 6-month and 12-month follow-up; and (2) to Health-related quality of life measures
examine how the intervention effectiveness varies for participants In the current study, we focus on the healthy-days measure of
with or without depression at baseline. HRQOL because it captures the key concepts of population health
and well-being. This construct is aligned with one of CDSMP’s
MATERIALS AND METHODS main objectives of empowering program participants to bet-
NATIONAL STUDY OF CDSMP AS STUDY BASIS ter manage their chronic conditions and experience a higher
The current study utilized a change in HRQOL measures at three quality of life. Healthy-days measures are important compo-
time points (baseline, at 6-month, and at 12-month) to examine nents that assess HRQOL. The HRQOL includes a set of four
the cost-effectiveness of the intervention among middle-aged and questions (8):
older adults enrolled in the National Study of the Chronic Disease
Self-Management Program (CDSMP). Data were analyzed from 1. Would you say that in general your health is; Excellent, Very
workshops delivered nationwide by 22 licensed sites in 17 states good, Good, Fair, or Poor?
across the nation from August 2010 to April 2011. CDSMP work- 2. Now thinking about your physical health, which includes physi-
shops were supported by various federal, state and local sources, cal illness and injury, for how many days during the past 30 days
healthcare organizations, and community agencies. The eligibil- was your physical health not good?
ity criteria and recruitment, intervention delivery, and referral 3. Now thinking about your mental health, which includes stress,
activities are described elsewhere (22). Sites already licensed to depression, and problems with emotions, for how many days
deliver CDSMP were selected and then agreed to participate in the during the past 30 days was your mental health not good?
National Study, delivering the manualized workshops following 4. During the past 30 days, for about how many days did poor
standardized intervention protocols and submitting data for study physical or mental health keep you from doing your usual
purposes. Data were collected in person before the start of the activities, such as self-care, work, or recreation?
intervention (baseline) and at 6 and 12 months post-intervention
by mail/phone. Investigators had no role in leader training, work- In the current study, we utilize an “unhealthy days” summary
shop recruitment, or program implementation. Each CDSMP measure, which is based on the second and third questions, esti-
delivery site recruited people for workshops in their usual fash- mates the overall number of days when physical and mental health
ion, which included referrals from organizations serving older was not good. We then calculate the number of days estimated to
adults (e.g., senior centers, healthcare facilities, and social ser- be healthy, which is the complement to unhealthy days measure
vice organizations as well as self-referrals from other recruitment (total number of “healthy days” limits to maximum of 30 days
activities including flyers, brochures, and health fairs). The inter- as this is the maximum possible value that this measure could
vention was designed to focus on content areas including (a) possibly take). These items have been extensively used for eval-
techniques to manage typical responses to chronic health prob- uating program objectives in other studies (8, 30–33) and the
lems such as frustration, fatigue, pain, and isolation: (b) improving validity of these measures has been confirmed in population based

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 27 | 190
Basu et al. Cost-effectiveness of CDSMP

samples. Participants responded to each item with the number of directly in the cost-effectiveness analyses (41). Since the CDSMP
days ranging from 0 to 30. survey did not include preference-based measure of EQ-5D, the
non-preference-based scores of “healthy days” measure was con-
Cost measures verted to preference-based EuroQol 5D (EQ-5D) utilizing the
The average cost per CDSMP participant varies by the number method proposed by Jia and Lubetkin (42). The EQ-5D is an
of enrolled participants per workshop with the estimated per- internationally developed preference-based (29, 43) method that
participant cost ranging from $219 to $583 (23). In the National provides a measure of utility scores to calculate QALYs which
Study of CDSMP, 145 workshops had an average size of 12.7 is used in cost-effectiveness analyses (44). EQ-5D estimates are
(±4.18) participants, with the majority of workshops (66.2%) obtained from healthy days by matching the cumulative distribu-
having between 8 and 16 participants. A detailed description of tions of the two HRQOL measures and EQ-5D from Behavioral
the cost measures reported by CDSMP delivery sites appears else- Risk Factor Surveillance System and Medical Expenditure Panel
where (23). Based on extant literature confirmed by experts in Survey datasets (42). For example, we obtained EQ-5D utility
the delivery field, we estimated program costs at $350 per par- score corresponding to number of “healthy days” measures in our
ticipant, assuming an average of 10 participants in each CDSMP sample. Detail description of the estimation method including
workshop (23). These program costs typically include licensure underlying assumptions can be found in Jia and Lubetkin (42).
costs, trained peer personnel, materials, and any space rental Utility values range from 1 (best possible health state) through 0
costs (34). Because the CDSMP was a community-based pro- (death) (44). We then used EQ-5D scores to calculate QALY for
gram and goal that this type of self-management program is to the calculation of cost-effectiveness ratios.
provide evidence regarding resources needed to deliver within Estimation of participant specific QALYs was based on parti-
the community, cost data are collected at the aggregate level. In tioning the study period into the number of follow-up assessments
the case of CDSMP, per-participant costs were aggregated at the and weighting each time interval by the individual’s utility score
workshop level. Individual-level cost data are less valuable for during that period of time (45). It is assumed that changes in util-
the effective implementation of this type of community-based ity values are linear over time, which is the most commonly used
program. method in cost-effectiveness analysis. Individual-level QALYs are
then estimated by applying the area-under the curve approach.
Other participant-level measures Details of this method can be found elsewhere (42). The general
Participants’ demographic characteristics measured at baseline expression for calculating QALYs using individual data that are
included age, gender, race/ethnicity, and number of chronic con- fully observed (i.e., no censoring) can be written as follows (45):
ditions. Measures and a sample questionnaire can be found in
Xn  (Q1 + Q t+1 ) (Tt+1 + Tt ) 
English and Spanish (35). Depressive symptoms were measured QALY = × (1)
using the Patient Health Questionnaire (PHQ) (36). Self-rated t =0 2 T
items (9 DSM-IV criteria) scored from 0 (not at all) to 3 (nearly
everyday) were added to determine overall PHQ score of study where, n is the number of utility measurements over the study
participants at baseline. A score greater than or equal to 10 was period (i.e., 1 year), Q t is the individual utility score (i.e., EQ-5D
considered clinically depressed because this cut-off point of 10 score) obtained in the t th measurement, T is the total duration of
has a sensitivity and specificity of 88% in detecting a diagnosis study period expressed in terms of total number of time units in a
of major depression in primary care patients (36). The reliability year (e.g., months), T t is the time period in which the t th measure-
of the PHQ-9 is high, with a Cronbach’s α of 0.89 and test–retest ment takes place (expressed as number of time units in a year). In
reliability of 0.84 (37). We used an eight-item version of the PHQ our case, n = 2 (i.e., first interval from baseline to 6-month and
(38), which excludes the item that asks patients if they have been second interval from 6-month to 12-month), T = 12 (i.e., number
bothered by “thoughts that you would be better off dead or of of months in a year), and three time points as T 0 = 0, T 1 = 6, and
hurting yourself in some way.” Scores for the eight-item version of T 2 = 12. For example, in our current study, QALYs are obtained
the PHQ range from 0 to 24, and are highly correlated with scores by substituting mean EQ-5D scores and controlling for baseline
on the nine-item version (r = 0.997) (38). utility (45):

Incremental QALY =
ANALYSIS
HRQOL, EQ-5D, and QALYs
Xn  (0.743 − 0.736) 6 (0.743 − 0.736)
× +
The CDC-derived measure of HRQOL is one of the most com- t =0 2 12 2
monly used outcome measures for evaluating burden of disease (0.755 − 0.743)

6 6
in public health research. However, a single measurement such as × + × = 0.007 (2)
12 2 12
quality-adjusted life year (QALY) is considered as a more useful
measure for cost-effectiveness analyses (38–40). This is because Cost-effectiveness analysis
QALY uses preference-based measures of HRQOL, which uses The strong need to control healthcare costs for the treatment of
summary scores (i.e., utility values) to represent population prefer- chronic diseases led us to search for interventions that produce
ences for different health states. Because the number of unhealthy greatest value, based on comparative economic evaluation (46).
days are not preference-based measures of HRQOL (as asked in the Cost-effectiveness analysis (CEA) is a type of economic evalua-
CDSMP survey), the CDC “healthy days” measures cannot be used tion method that can be utilized to assess whether money is well

www.frontiersin.org April 2015 | Volume 3 | Article 27 | 191

Basu et al. Cost-effectiveness of CDSMP

spent in a particular health promotion program (47). Funding Table 2 represents summary statistics for healthy days and cor-
agencies may continue to support programs on the basis of this responding EQ-5D measures at baseline, 6-month, and 12-month
information or may find additional interventions that can produce during the study. Both healthy days (17.9–19.2) and correspond-
the best outcomes with available resources. The most widely used ing EQ-5D scores (0.743–0.755) were significantly improved from
method for CEA is the incremental cost-effectiveness ratio (ICER), baseline to 12-month) period (with a p-value <0.001); however,
which compares differences in cost to differences in effectiveness no significant improvement was observed for these measures from
between two competing interventions and therefore relevant for baseline to 6-month.
policy making decisions. In the absence of a control group, we are Table 3 presents the similar statistics by baseline depression
comparing health gains compared to no intervention (i.e., “doing status. Changes in mean healthy days and EQ-5D scores were
nothing”). Therefore, in this case, ICER was calculated compared examined by utilizing paired t -test by baseline depression sta-
to baseline and measures the effectiveness of CDSMP in improving tus. On average, participants with depression at baseline reported
QALYs compared to “doing nothing.” lower number of healthy days and their corresponding EQ-5D
An important first step of calculating ICER is to quantify its scores were also lower than participants who had no depression
average cost of a program in order to relate the cost to spe- at baseline. However, both groups (depression versus no depres-
cific measures of the program (48). Considering zero cost for sion at baseline) showed significant improvement in healthy days
“no intervention,” the numerator of ICER is the average cost and and EQ-5D scores from baseline to the12-month period. These
denominator includes the mean effectiveness of the program (48, results indicate that CDSMP improves population health sta-
49). In our case, the numerator is the mean program cost spent tus among individuals with multiple chronic conditions through
per CDSMP participant and denominator is QALYs estimates. The better disease self-management strategies.
QALYs is particularly useful in quantifying program effectiveness Table 4 shows the cost-effectiveness results for the CDSMP
and is the most commonly used measure of treatment effectiveness intervention. Incremental cost-effectiveness ratios (ICER) are cal-
in CEA literature (50, 51). The ICER for each outcome measure culated for the overall group as well as by baseline depression
was calculated by dividing per person CDSMP workshop costs by status. These ratios explain how much each additional QALYs
the QALYs. Therefore ICER can be indicated as: gained with CDSMP will cost. Overall, ICER ranges from $83,285
to $31,285 per QALYs gained for participants in the CDSMP
Average cost spent per CDSMP participant −$0 program with the median of $50,000/QALYs. ICER by baseline
ICER= (3)
QALYs gained adjusted for baseline utility score depression status indicates that it will cost more per QALYs gained
for those diagnosed with depression based on PHQ-8 score.
RESULTS
Table 1 describes participants’ characteristics at baseline. In total, DISCUSSION
1,170 participants completed the baseline assessment. On aver- Prior evidence suggests that CDSMP can significantly improve
age, participants were 65 years old, nearly 83% were female, and health outcomes for individuals with a variety of chronic con-
had an average of 13 years of education. Ethno-racial composition ditions (21, 24). However, economic efficacy of the CDSMP
included 55% non-Hispanic white, 16% African American, 22% on HRQOL is not well known. The current study developed
Hispanic, and 6.5% others. About 79% reported two or more con- a preference-based EQ-5D measure of HRQOL from healthy
ditions and 79.1% of participants attended four or more workshop days to quantify the cost-effectiveness of a CDSMP program for
sessions. improving QALYS gained for individuals with multiple chronic
conditions. Although there is no universally acceptable threshold
value for cost-effectiveness ratio (52), costs range from $50,000 to
Table 1 | Sample characteristics at baseline (N = 1170). $75000 per QALYs gained have been considered as an acceptable
value for resources expended (48).
Variables % Mean (SD)
Health-related quality of life is recognized as an important
Age (in years) 65.4 (14.3) measure in public health as well as clinical research because it
Female 82.7 includes a population-based approach that addresses physical and
Race/ethnicity – mental health of a large number of individuals over time. More-
Non-Hispanic White 55.2 over, converting non-preference-based measures of HRQOL to
African American 16.0 a preference-based measure provides a way to compare the effi-
Hispanic 22.3 cacy of CDSMP to other evidenced-based disease management
Other 6.5 programs in the literature. As shown in the current study, the eco-
Workshop completion rate 79.1 nomic value of CDSMP, as measured in dollars per QALYs gained,
Education (1–23) 12.9 (3.8) may have far reaching effects when magnified across the U.S. Thus,
At least two chronic conditions 79.0 finding ways to improve the reach of the CDSMP among espe-
PHQ-8 depression (0–24) 6.6 (5.5) cially vulnerable individuals (e.g., rural, minority, low income) is
Healthy days (0–30) 17.9 (11.5) a critical path of research for future studies.
EQ-5D (0–1)a 0.736 (0.156) Policy makers are interested in finding ways to improve
the health of individuals with multiple chronic conditions as
a
In our sample, EQ-5D value ranges from 0.411 to 0.995. a significant share of healthcare dollars are attributed to the

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 27 | 192
Basu et al. Cost-effectiveness of CDSMP

Table 2 | Changes in mean (SD) of healthy days and EQ-5D scores among CDSMP participants during the study period.

HRQOL measures Baseline 6 months 12 months p-value for the change

Baseline and 6 months Baseline and 12 months

Healthy days (0–30) 17.9 (11.5) 18.5 (11.4) 19.2 (11.1.) 0.25 <0.001
EQ-5D (0–1) 0.736 (0.156) 0.743 (0.156) 0.755 (0.152) 0.32 <0.001

Table 3 | Changes in mean (SD) of healthy days and EQ-5D scores by depression status at baseline.

HRQOL Depression at baseline (PHQ-8 ≥10) Difference in scores: Difference in scores:

measures baseline and 6-months baseline and 12-months

Yes No Yes No Yes No

Baseline 6-months 12-months Baseline 6-months 12-months

Healthy days 16.4 (11.5) 17.3 (11.4) 17.9 (11.2) 24.1 (8.8) 24.9 (8.8) 27.2 (6.9) 0.33 0.52 0.01 0.001
EQ-5D 0.721 (0.15) 0.728 (0.15) 0.737 (0.15) 0.825 (0.11) 0.830 (0.13) 0.861 (0.10) 0.29 0.96 0.007 0.005

treatment of chronic medical conditions. Deploying resources there is evidence that people do not always value the use of
with the goal of population-based health management will facil- leisure time and it is also hard to estimate the value given the
itate efficient allocation of resources in such a way that will availability of survey instruments. The theoretical notion is that
lower overall healthcare cost, and improve quality of care expe- high motivation and retirement lower the time cost of partici-
rience (50). The CDSMP provides a mechanism to deliver cost- pating in this type of health promotion program. Literature also
effective evidence-based strategies to those who may benefit suggests excluding time cost of participants in physical activity
most (e.g., being older and having co-morbid conditions). Poor interventions (51).
quality of life and other mental health concerns has broader Another limitation is that the calculation of QALYs was not
effects than immediate impacts on individuals with chronic adjusted for possible confounding factors which could potentially
conditions. influence cots and outcome measures (45). Although the use of
multiple regression method would control for this imbalance, the
LIMITATIONS lack of a control group of CDSMP intervention makes this method
This study builds upon an existing translational National Study of infeasible in the current study. There are many more unmeasured
CDSMP, which was not originally designed as an economic cost- benefits of the CDSMP that are not captured in the outcomes
effectiveness study. Hence, several variables typically included in presented in the current study. For example, participants typically
economic analyses were not present (e.g., a comparison group report many positive aspects of their participation including new
or individual cost measures). While there was some participant social interactions. As such, the complete value of this evidence-
attrition over time, the impacts appear minimal in terms of the based program is not fully measured in the current analysis and
diversity of participants in the study. may be targeted for future study.
As an accommodation to available data, our study has assumed Using the generally accepted cost-effectiveness ratio of
a standard per-participant costs now cited in the CDSMP liter- $50,000/QALYs, results of this study indicate that CDSMP is
ature (23). Although ICER values seem very attractive, cautions potentially cost-effective for individuals with multiple chronic
should be used to interpret too literally because these values conditions. Utilizing the most widely used generic measure of
can change substantially depending on changes in cost estimates HRQOL from a population-based health days measure, the cur-
assumption and point estimates. We acknowledge that this is a rent study quantifies cost-effectiveness of CDSMP. However, due
rough estimate that excludes a full consideration of all poten- to the fact that CERs evaluate how a program’s costs compare to
tial costs. One consideration is whether to include the oppor- its outcomes, judgments about whether the outcomes achieved are
tunity cost of participating in the CDSMP program. Here, the worth than the cost are subjective and dependent on several factors
opportunity cost would be the value of participation time in (e.g., current needs and resources).
the workshop and which could be calculated based on wage for- Nevertheless, we feel that this study makes a major contribution
gone or the value of leisure time forgone. Since the majority of as one of the first studies to quantify the benefits of CDSMP in
CDSMP participants are older adults, we can make the assump- terms of a preference-based quality of life measure and examine
tion that there is limited (if any) opportunity cost involved in the impacts for those with co-morbid depressive symptomatol-
terms of forgone wages as they are likely to be out of labor force. ogy. It provides a foundation for future cost-effectiveness studies
So, the value of leisure time could be the only way to calculate of self-management programs for adults with multiple chronic
the opportunity cost of participating in the program. However, conditions.

www.frontiersin.org April 2015 | Volume 3 | Article 27 | 193

Basu et al. Cost-effectiveness of CDSMP

ACKNOWLEDGMENTS

ICER = average cost/QALYs

The American Recovery and Reinvestment Act of 2009 (i.e., Recov-

Min

$8423
ery Act) Communities Putting Prevention to Work: Chronic Dis-
ease Self-Management Program initiative, led by the U.S. Admin-
istration on Aging in collaboration with the Centers for Disease
Median

$13461
Control and Prevention and the Centers for Medicare and Med-
Not depressed at baseline

icaid Services, allotted $32.5 million to support the translation of

QALYs = 0.026b

$22423
the Stanford program in 45 States, Puerto Rico, and the District of
Max

Columbia. This research was supported by a grant to the National

Council on Aging, with subcontracts to Texas A&M University and
Stanford University.
$219
Min
ICER by depression status based on PHQ-8 score

Average cost-ranges

REFERENCES
1. Centers for Disease Control and Prevention. The State of Aging and Health in
Median

America 2013. Atlanta (GA): CDC. (2013). Available from: http://www.cdc.gov/

$350

aging/pdf/state-aging-health-in-america-2013.pdf
2. Ralph NL, Mielenz TJ, Parton H, Flatley AM, Thorpe LE. Multiple chronic con-
ditions and limitations in activities of daily living in a community-based sam-
$583
Max

ple of older adults in New York City, 2009. Prev Chronic Dis (2013) 10:E199.
doi:10.5888/pcd10.130159
3. Anderson GF. Chronic Care: Making the Case for Ongoing Care. Princeton, NJ:
$36500
ICER = average cost/QALYs

Robert Wood Johnson Foundation (2010). Available from: http://www.rwjf.org/

Min

content/dam/farm/reports/reports/2010/rwjf54583
4. Barile JP, Thompson WW, Zack MM, Krahn GL, Horner-Johnson W, Bowen
SE. Multiple chronic medical conditions and health-related quality of life in
Median

older adults, 2004–2006. Prev Chronic Dis (2013) 10:E162. doi:10.5888/pcd10.

$58333

120282
5. Dominick KL, Ahern FM, Gold CH, Heller DA. Relationship of health-related
Depressed at baseline

quality of life to health care utilization and mortality in older adults. Aging Clin
QALYs = 0.006a

Exp Res (2002) 14:499–508. doi:10.1007/BF03327351

$97166
Max

6. Unutzer J, Donald LP, Diehr P, Simon G, Grembowski SD, Katon W.

Quality adjusted life years in older adults with depressive symptoms and
chronic medical disorders. Int Psychogeriatr (2000) 12:15–33. doi:10.1017/
S1041610200006177
$219
Min
Average cost-ranges

7. Fortin M, Lapointe L, Hudon C, Vanasse A, Ntetu AL, Maltais D. Multimor-

bidity and quality of life in primary care: a systematic review. Health Qual Life
= 0.006.
= 0.026.

Outcomes (2004) 2:51. doi:10.1186/1477-7525-2-51

Median

8. Moriaty DG, Zack MM, Kobau R. The centers for disease control and pre-
$350

vention’s healthy days measures – population tracking of perceived physi-




cal and mental health over time. Health Qual Life Outcomes (2003) 1:37.
12
6

12
6
×

doi:10.1186/1477-7525-1-37
×
$583
Max

(0.737−0.728)

(0.861−0.830)

9. Centers for Disease Control and Prevention. Measuring Healthy Days: Popula-
tion Assessment of Health-Related Quality of Life [Internet]. Atlanta (GA): CDC
2

(2000). Available from: http://www.cdc.gov/hrqol/pdfs/mhd.pdf

$31,285

+
ICER = average cost/QALYs

10. Harris JR, Wallace RB. The Institute of Medicine’s new report on living well with
+
Min

12
6

12
6

chronic illness. Prev Chronic Dis (2012) 9:E148. doi:10.5888/pcd9.120126

×
×

11. Parekh AK, Barton MB. The challenge of multiple comorbidity for the US health
(0.728−0.721)

(0.861−0.825)

care system. JAMA (2010) 303:1303–4. doi:10.1001/jama.2010.381

Table 4 | ICER for QALYs for CDSMP participants.

2
$50,000
Median

12. Parekh AK, Goodman RA, Gordon C, Koh HK. Managing multiple chronic
conditions: a strategic framework for improving health outcomes and qual-
+
+

ity of life. Public Health Rep (2011) 126(4):460–71. Available from: http:
12
6

12
6
×

//www.publichealthreports.org/cphce/126_4_460.pdf
×
QALYs = 0.007

 (0.728−0.721)
$83,285

 (0.861−0.825)

13. Patient Protection and Affordable Care Act of 2010. Pub L. 111-148. 124 Stat.
Max
Overall

119. (2010). Available from: http://www.gpo.gov/fdsys/pkg/PLAW-111publ148/

2

pdf/PLAW-111publ148.pdf
14. Barlow J, Turner A, Edwards R, Gilchrist M. A randomised controlled trial of
Incremental QALYdepression=yes =
Incremental QALYdepression=no =

lay-led self-management for people with multiple sclerosis. Patient Educ Couns
$219
Min
Average cost-ranges

(2009) 77:81–9. doi:10.1016/j.pec.2009.02.009

15. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, Whitelaw N, et al. Suc-
cesses of a national study of the chronic disease self-management program:
Median

meeting the triple aim of health care reform. Med Care (2013) 51:992–8.
$350

doi:10.1097/MLR.0b013e3182a95dd1
16. Brady TJ, Murphy L, O’Colmain BJ, Beauchesne D, Daniels B, Greenberg M,
et al. A meta-analysis of health status, health behaviors, and health care utiliza-
Max

$583

tion outcomes of the chronic disease self-management program. Prev Chronic

Dis (2013) 10:120112. doi:10.5888/pcd10.120112
b
a

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 27 | 194
Basu et al. Cost-effectiveness of CDSMP

17. Bodenheimer T, Lorig K, Holman H, Grumbach K. Patient self-management of 38. Kroenke K, Strine TW, Spitzer RL, Williams JB, Berry JT, Mokdad AH. The
chronic disease in primary care. JAMA (2002) 288:2469–75. doi:10.1001/jama. PHQ-8 as a measure of current depression in the general population. J Affect
288.19.2469 Disord (2009) 114:163–73. doi:10.1016/j.jad.2008.06.026
18. Reeves D, Kennedy A, Fullwood C, Bower P, Gardner C, Gately C, et al. Predicting 39. The Secretary’s Advisory Committee on National Health Promotion and Dis-
who will benefit from an expert patients programme self-management course. ease Prevention Objectives for 2020. Phase I Report: Recommendations for the
Br J Gen Pract (2008) 58:198–203. doi:10.3399/bjgp08X277320 Framework and Format of Healthy People 2020 [Internet]. (2008). Available from:
19. Labrecque M, Rabhi K, Laurin C, Favreau H, Moullec G, Lavoie K, et al. http://www.healthypeople.gov/sites/default/files/PhaseI_0.pdf
Can a self-management education program for patients with chronic obstruc- 40. Metzler M, Kanarek N, Highsmith K, Straw R, Bialek R, Stanleyl J, et al. Com-
tive pulmonary disease improve quality of life. Can Respir J (2011) 18: munity health status indicators project: the development of a national approach
e77–81. to community health. Prev Chronic Dis (2008) 5:A94.
20. Ahmed S, Bartlett SJ, Ernst P, Paré P, Kanter GM, Perreault R, et al. Effect of a 41. Jia H, Lubetkin EI. Trend in quality-adjusted life-years lost contributed by smok-
web-based chronic disease management system on asthma control and health- ing and obesity. Am J Prev Med (2010) 38:138–44. doi:10.1016/j.amepre.2009.
related quality of life: study protocol for a randomized controlled trial. Trials 09.043
(2011) 12:260. doi:10.1186/1745-6215-12-260 42. Jia H, Lubetkin EI. Estimating EuroQol EQ-5D scores from population healthy
21. Berwick DM, Nolan TW, Whittington W. The triple aim: care, health and cost. days data. Med Decis Making (2008) 28:491–9. doi:10.1177/0272989X07312708
Health Aff (Millwood) (2008) 27(3):759–69. doi:10.1377/hlthaff.27.3.759 43. Kind P, Dolan P, Gudex C, Williams A. Variations in population health sta-
22. Lorig KR, Sobel DS, Stewart AL, Brown BW Jr, Bandura A, Ritter P, et al. Evidence tus: results from a United Kingdom national questionnaire survey. BMJ (1998)
Suggesting that a chronic disease self-management program can improve health 316:736–41. doi:10.1136/bmj.316.7133.736
status while reducing hospitalization: a randomized trial. Med Care (1999) 44. Goldsmith KA, Dyer MT, Buxton MJ, Sharples LD. Mapping of the EQ-
37(1):5–14. doi:10.1097/00005650-199901000-00003 5D index from clinical outcome measures and demographic variables in
23. Ahn S, Basu R, Smith ML, Jiang L, Lorig K, Whitelaw N, et al. The impact patients with coronary heart disease. Health Qual Life Outcomes (2010) 8:54.
of chronic disease self-management programs: healthcare savings through a doi:10.1186/1477-7525-8-54
community-based intervention. BMC Public Health (2013) 13:1141. doi:10. 45. Manaca A, Hawkings N, Sculpher MJ. Estimating mean QALYs in trial-based
1186/1471-2458-13-1141 cost-effectiveness analysis: the importance of controlling for baseline utility.
24. Ory MG, Ahn S, Jiang L, Lorig K, Ritter P, Larent DD, et al. National study of Health Econ (2005) 14:487–96. doi:10.1002/hec.944
chronic disease self-management six-month outcome findings. J Aging Health 46. Cellini SR, Kee JE. Cost-effectiveness and cost-benefit analysis. In: Wholey JS,
(2013) 25:1258–74. doi:10.1177/0898264313502531 Hatry HP, Newcomer KE, editors. Handbook of Practical Program Evaluation.
25. Lorig K, Holman H, Sobel D, Laurent DD. Living a Healthy Life with Chronic San Francisco, CA: John Wiley & Sons Inc (2010). 493 p.
Conditions: Self Management of Heart Disease, Arthritis, Diabetes, Asthma, Bron- 47. Haddix A, Teutsch SM, Corso PS. Prevention Effectiveness: A Guide to Deci-
chitis, Emphysema and Others. 3rd ed. Boulder, CO: Bull Publishing Company sion Analysis and Economic Evaluation. 2nd ed. Oxford: Oxford University Press
(2006). (2003). 264 p.
26. Lorig K, Ritter PL, Pifer C, Werner P. Effectiveness of the chronic disease self- 48. Ory MG, Smith ML, Kulinski KP, Lorig K, Zenker W, Whitelaw N. Self-
management program for persons with a serious mental illness: a translation management at the tipping point: reaching 100,000 Americans with evidence-
study. Community Ment Health J (2013) 50:96–103. doi:10.1007/s10597-013- based programs. J Am Geriatr Soc (2013) 61:821–3. doi:10.1111/jgs.12239
9615-5 49. Drummond MF, Stoddart GL, Torrance GW. Methods for the Economic Evalua-
27. Erdem E, Korda H. Self-management program participation by older adults tion of Healthcare Programmes. Oxford: Oxford University Press (1987).
with diabetes: chronic disease self-management program and diabetes self- 50. Brownson CA, Hoerger TH, Fisher EB, Kilpatrick KE. Cost-effectiveness of dia-
management program. Fam Community Health (2014) 37:134–46. doi:10.1097/ betes self-management programs in community primary care settings. Diabetes
FCH.0000000000000025 Educ (2009) 35:761–9. doi:10.1177/0145721709340931
28. Smith ML, Ory MG, Larsen R. Older women in a statewide evidence-based 51. Woolf SH. A closer look at the economic argument for disease prevention. JAMA
falls prevention program: who enrolls and what benefits are obtained? Womens (2009) 301:536–8. doi:10.1001/jama.2009.51
Health Issues (2010) 20:427–34. doi:10.1016/j.whi.2010.07.003 52. Boheler C, Milton KE, Bull FC, Fox-Rushby JA. The cost of changing
29. Rabin R, de Charro F. EQ-5D: a measure of health status from the EuroQol physical activity behavior: evidence from a “physical activity pathway” in a pri-
Group. Ann Med (2001) 33:337–43. doi:10.3109/07853890109002087 mary care setting. BMC Public Health (2011) 11:370. doi:10.1186/1471-2458-
30. Smith ML, Ory MG, Ahn S, Bazzarre TL, Resnick B. Older adults’ participation 11-370
in a community-based falls prevention exercise program: relationships between
the EASY tool, program attendance, and health outcomes. Gerontologist (2011) Conflict of Interest Statement: Rashmita Basu has no conflicts of interest to dis-
51:809–21. doi:10.1093/geront/gnr084 close. Dr. Basu has no financial relationships with entities that could be perceived
31. Smith ML, Ahn S, Sharkey JR, Horel S, Mier N, Ory MG. Successful falls pre- to influence, or that give the appearance of potentially influencing, what she wrote
vention programming for older adults in Texas: rural-urban variations. J Appl in the submitted work. Dr. Basu has no patents or copyrights to declare (whether
Gerontol (2012) 31:3–27. doi:10.1177/0733464810378407 pending, issued, licensed, and/or receiving royalties) relevant to the work. Dr. Basu
32. Smith ML, Ahn S, Mier N, Jiang L, Ory MG. An evidence-based program to has no other relationships or activities that readers could perceive to have influ-
reduce fall-related risk among older adults: a comparison of program efficacy enced, or that give the appearance of potentially influencing, what she wrote in the
by ethnicity. Calif J Health Promot (2012) 10:28–43. submitted work. Marcia G. Ory has no conflicts of interest to disclose. Texas A&M
33. NCOA. About Evidence-Based Programs. Available from: http://www.ncoa.org/ University received a grant (PI: Dr. Ory) to evaluate the National Study of CDSMP.
improve-health/center-for-healthy-aging/about-evidence-based-programs. Dr. Ory has no financial relationships with entities that could be perceived to influ-
html ence, or that give the appearance of potentially influencing, what she wrote in the
34. Page TF, Palmer RC. Cost analysis of chronic disease self-management pro- submitted work. Dr. Ory has no patents or copyrights to declare (whether pending,
grams being delivered in South Florida. Health Educ J (2014) 73:228–36. issued, licensed, and/or receiving royalties) relevant to the work. Dr. Ory has no
doi:10.1177/0017896912471047 other relationships or activities that readers could perceive to have influenced, or
35. Stanford School of Medicine. Training Policies for Stanford Self-Management that give the appearance of potentially influencing, what she wrote in the submitted
Programs. Palo Alto, CA: Patient Education Research Center (2014). Available work. Samuel D. Towne Jr. has no conflicts of interest to disclose. Neither myself
from: http://patienteducation.stanford.edu/training/trnpolicies.html nor this institution at any time received payment or services from a third party for
36. Kroenke K, Spitzer RL. The PHQ-9; a new depression diagnostic and severity any aspect of the submitted work. Dr. Towne has no financial relationships with
measure. Psychiatr Ann (2002) 32(9):509–15. entities that could be perceived to influence, or that give the appearance of poten-
37. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of brief depression tially influencing, what he wrote in the submitted work. Dr. Towne has no patents or
severity measure. J Gen Intern Med (2001) 16:606–13. doi:10.1046/j.1525-1497. copyrights to declare (whether pending, issued, licensed, and/or receiving royalties)
2001.016009606.x relevant to the work. Dr. Towne has no other relationships or activities that readers

www.frontiersin.org April 2015 | Volume 3 | Article 27 | 195

Basu et al. Cost-effectiveness of CDSMP

could perceive to have influenced, or that give the appearance of potentially influ- in the US, many of the authors and/or Review Editors may have worked together
encing, what he wrote in the submitted work. Matthew Lee Smith has no conflicts previously in some fashion. Review Editors were purposively selected based on their
of interest to disclose. Neither myself nor this institution at any time received pay- expertise with evaluation and/or evidence-based programming for older adults. Review
ment or services from a third party for any aspect of the submitted work. Dr. Smith Editors were independent of named authors on any given article published in this
has no financial relationships with entities that could be perceived to influence, or volume.
that give the appearance of potentially influencing, what he wrote in the submitted
work. Dr. Smith has no patents or copyrights to declare (whether pending, issued, Received: 14 October 2014; accepted: 02 February 2015; published online: 27 April
licensed, and/or receiving royalties) relevant to the work. Dr. Smith has no other 2015.
relationships or activities that readers could perceive to have influenced, or that give Citation: Basu R, Ory MG, Towne SD Jr., Smith ML, Hochhalter AK and
the appearance of potentially influencing, what he wrote in the submitted work. Ahn S (2015) Cost-effectiveness of the chronic disease self-management program:
SangNam Ahn has no conflicts of interest to disclose. implications for community-based organizations. Front. Public Health 3:27. doi:
10.3389/fpubh.2015.00027
This paper is included in the Research Topic, “Evidence-Based Programming for This article was submitted to Public Health Education and Promotion, a section of the
Older Adults.” This Research Topic received partial funding from multiple govern- journal Frontiers in Public Health.
ment and private organizations/agencies; however, the views, findings, and conclusions Copyright © 2015 Basu, Ory, Towne, Smith, Hochhalter and Ahn. This is an open-
in these articles are those of the authors and do not necessarily represent the official access article distributed under the terms of the Creative Commons Attribution License
position of these organizations/agencies. All papers published in the Research Topic (CC BY). The use, distribution or reproduction in other forums is permitted, provided
received peer review from members of the Frontiers in Public Health (Public Health the original author(s) or licensor are credited and that the original publication in this
Education and Promotion section) panel of Review Editors. Because this Research journal is cited, in accordance with accepted academic practice. No use, distribution or
Topic represents work closely associated with a nationwide evidence-based movement reproduction is permitted which does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 27 | 196
 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2015.00042

Healthcare cost savings estimator tool for chronic disease

self-management program: a new tool for program
administrators and decision makers
SangNam Ahn 1,2 *, Matthew Lee Smith 3 , Mary Altpeter 4 , Lindsey Post 1 and Marcia G. Ory 2
1
Division of Health Systems Management and Policy, School of Public Health, The University of Memphis, Memphis, TN, USA
2
Department of Health Promotion and Community Health Sciences, Texas A&M University Health Science Center, School of Public Health, College Station, TX, USA
3
Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA, USA
4
Center for Health Promotion and Disease Prevention, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA

Edited by: Chronic disease self-management education (CDSME) programs have been delivered to
Sanjay P. Zodpey, Public Health
more than 100,000 older Americans with chronic conditions. As one of the Stanford
Foundation of India, India
suite of evidence-based CDSME programs, the chronic disease self-management program
Reviewed by:
Cheryll Diann Lesneski, UNC-Chapel (CDSMP) has been disseminated in diverse populations and settings. The objective of this
Hill, USA paper is to introduce a practical, universally applicable tool to assist program administrators
Erica T. Sosa, University of Texas at and decision makers plan implementation efforts and make the case for continued program
San Antonio, USA
delivery. This tool was developed utilizing data from a recent National Study of CDSMP to
*Correspondence:
estimate national savings associated with program participation. Potential annual health-
SangNam Ahn, Division of Health
Systems Management and Policy, care savings per CDSMP participant were calculated based on averted emergency room
School of Public Health, The visits and hospitalizations. While national data can be utilized to estimate cost savings, the
University of Memphis, 133 Robison tool has built-in features allowing users to tailor calculations based on their site-specific
Hall, Memphis, TN 38152-3530, USA
data. Building upon the National Study of CDSMP’s documented potential savings of $3.3
e-mail: [email protected]
billion in healthcare costs by reaching 5% of adults with one or more chronic conditions,
two heuristic case examples were also explored based on different population projections.
The case examples show how a small county and large metropolitan city were not only
able to estimate healthcare savings ($38,803 for the small county; $732,290 for the large
metropolitan city) for their existing participant populations but also to project significant
healthcare savings if they plan to reach higher proportions of middle-aged and older adults.
Having a tool to demonstrate the monetary value of CDSMP can contribute to the ongoing
dissemination and sustainability of such community-based interventions. Next steps will be
creating a user-friendly, internet-based version of Healthcare Cost Savings Estimator Tool:
CDSMP, followed by broadening the tool to consider cost savings for other evidence-based
programs.
Keywords: chronic disease, chronic disease self-management program, healthcare cost, healthcare cost savings
estimator tool, return on investment

BACKGROUND AND RATIONALE their health behaviors, enhance their health outcomes, and reduce
Adults with chronic conditions are the primary users of health- healthcare utilization (5, 6). Topics covered in CDSMP include
care in US and account for two-thirds of total healthcare spending coping skills and symptom control (7). Coping strategies such
(1). Healthcare costs in US, as measured by the percentage of as action planning and feedback, behavior modeling, problem-
gross domestic product (GDP), essentially doubled in <30 years solving techniques, and decision-making are applicable to all
from 9.2% in 1980 to 17.6% in 2009 (2). People with three or chronic conditions. CDSMP participants are also taught how to
more chronic conditions have 14.6 times more hospital stays control their symptoms through relaxation techniques, healthy
than patients with no chronic conditions, and patients with co- eating, sleep and fatigue monitoring, medication management,
morbidities spend 25 times more nights in the hospital than adults exercise, and improved communication with providers. Led by
with no chronic conditions (3). Coupled with our rapidly aging two peer facilitators, CDSMP is a highly interactive program that
society, this trend, if not curtailed, will lead to one of every three engages participants for six weekly sessions for two and a half
dollars spent in America paying for healthcare by 2040, with at hours per session (8). Each CDSMP delivery site recruited people
least 65% of that spending going toward patients with multiple for workshops in their usual manner including self-referrals from
chronic conditions (4). flyers, brochures, and health fairs as well as referrals from orga-
The chronic disease self-management program (CDSMP) has nizations serving older adults (e.g., senior centers, social service
been introduced to help patients with chronic conditions improve organizations) (8). Additional information regarding sampling,

www.frontiersin.org April 2015 | Volume 3 | Article 42 | 197

Ahn et al. Healthcare cost savings estimator tool

recruitment, training, and fidelity assessment can be found in Recovery and Reinvestment Act of 2009 (i.e., ARRA) Communities
previous work done by the authors (8). Putting Prevention to Work: Chronic Disease Self-Management Pro-
Since its introduction, CDSMP has been made available in all gram initiative (17). Data from 1,170 CDSMP participants were
US states (9) and 25 countries (10–12). More recent studies doc- used to estimate health cost savings associated with the program
umented CDSMP participants’ improvements in the Triple Aim (15). Of the 1,170 participants at baseline, approximately 77%
components of healthcare reform (i.e., better healthcare, better (n = 903) and 71% (n = 825) completed the 6- and 12-month
health outcome, better value) (8). CDSMP can improve health- assessments, respectively (8). On each assessment, participants
care delivery and patient experience through increased patient– were asked to self-report any ER visits and hospitalizations in the
physician communication, better education about medication previous 6 months. These items were included to identify changes
utilization, and patient empowerment, and self-efficacy (8, 13). in participants’ healthcare utilization at three time points. The
There is also a rapidly growing body of evidence demonstrat- health benefits and financial effects of this National Study have
ing substantially better health outcomes for CDSMP participants, been documented in previous studies (8, 15).
which include improved self-reported general health, fewer social Based on data from the National Study of CDSMP, a six-step
activity limitations, more physical activity, and decreases in depres- process was developed for assessing potential cost savings (15).
sion, fatigue, and pain (8, 13, 14). These benefits have been demon- An excel-based tool was constructed that used National Study
strated among participants with a variety of chronic conditions, data to summarize potential national savings as a default; how-
across the full spectrum of socioeconomic status, and in multiple ever, users are able to override the default by inputting their own
types of delivery settings (14). To better value, healthcare-related numbers to estimate the savings accrued by offering CDSMP in
cost savings are achieved as healthcare utilization decreases, evi- their service area. More details about the data required of users
denced by reduced hospitalizations, emergency room (ER) visits, for tailored estimates are provided below. This excel-based tool
and lengths of hospital stays (14, 15). A recent study estimated is publicly available at: http://cdsmp-cost-tool.herokuapp.com/
annual net cost savings of $364 per CDSMP participant, which static/files/CDSMP_Cost_Estimator.xls.
would amount to a national savings of $3.3 billion assuming 5%
of adults with one or more chronic conditions participated in the HEALTHCARE COST SAVINGS ESTIMATOR TOOL
intervention (15). GENERIC SIX-STEP HEALTHCARE COST SAVINGS ESTIMATOR TOOL
While the cost savings associated with CDSMP delivery and MODELED FROM THE NATIONAL STUDY OF CDSMP
participation can be calculated, there is no universal tool or stan- The following six-step method was developed as a practical way for
dardized method for easily estimating program cost savings among identifying program costs and potential cost savings for evidence-
CDSMP participants. Such a tool would be of great benefit to based programs, utilizing the National Study of CDSMP as the case
program administrators responsible for allocating resources for example (15). These data are the basis of the tool’s creation.
evidence-based programs. More specifically, a tool estimating cost
savings including training, personnel, and material costs (16) Step 1: examine the pattern of ER visits and hospitalizations
could help program deliverers estimate the average per participant among CDSMP participants (n = 1,170) in the first and second
program costs. Even more, if there were a tool that guided users 6-month periods.
step-by-step through the process and allowed them to tailor esti- Table 1 shows changes in ER visits from baseline (18%) to
mates by “filling in the blanks” based on their specifications and 6 months (13%) to 12 months (13%). ER visits between base-
available data, program administrators could more confidently line to 6 months (5%) and baseline to 12 months (5%) were
demonstrate the effectiveness of CDSMP at containing costs in significantly reduced (15). Table 1 also shows changes in hospi-
their communities and service areas. Additionally, the tool could talizations from baseline (14%) to 6 months (11%) to 12 months
help program administrators be strategic when selecting partic- (14%). Hospitalizations between baseline and 6 months (3%)
ipant recruitment goals and/or targeting particular participant were significantly reduced (15).
groups (e.g., based on their healthcare utilization patterns or geo- Step 2: Identify age-adjusted mean costs for ER visits and hospi-
graphic location), identify returns on investment, justify funding talizations from 2010 Medical Expenditure Panel Survey (MEPS).
requests, and prepare for program scalability and sustainability Table 1 shows mean costs for ER visits and hospitalizations
within their organization and/or community. To support these from 2010 MEPS. The MEPS data were selected for this study
strategic planning efforts, this paper: (1) describes the develop- because this is the most complete source of data related to the cost
ment of the Healthcare Cost Savings Estimator Tool: CDSMP (i.e., and use of healthcare in US at the time of this study (18). First, we
tool); (2) illustrates how the tool can be tailored by users and intro- identified the age distribution in the National Study of CDSMP:
duces two heuristic case examples to show how context impacts 10% were 18-44 years of age, 31% were 45-64 years of age, and
potential cost savings; and (3) describes the recommended uses of 59% were 65 years of age or older. Then we identified mean costs
the tool and potential challenges to be considered. of ER visits by the aforementioned age categories from the 2010
MEPS dataset and found $1,513 as the age-adjusted cost of ER
METHODS: HEALTHCARE COST SAVINGS ESTIMATOR TOOL visits. The age-adjusted value was used to calculate a more accu-
ORIGINS AND CREATION rate cost of ER visits based on the age distribution and mean costs
Data from the National Study of CDSMP, conducted from 2010 to of ER visits of each age category. Thus, total cost savings associ-
2012 among 22 licensed sites within 17 states, were used to create ated with ER visits per person at two time periods amounted
the tool. The National Study of CDSMP was part of the American to $151.31 [first 6 months (5% reduction × $1,513) + second

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 42 | 198
Ahn et al. Healthcare cost savings estimator tool

Table 1 | Healthcare cost savings estimator tool: the national study of CDSMP.

CDSMP health cost savings estimatora

National study case example

N % Change in %

1. EXAMINE THE PATTERN OF HEALTH CARE UTILIZATION WITHIN YOUR POPULATION

Emergency room (ER) visits

Include number of participants at baseline 1170
Include number of participants reported visiting ER at baseline 211 18%
Include number of participants at 6 months 903
Include number of participants reported visiting ER at the first 6 months 118 13% 5%
Include number of participants at 12 months 825
Include number of participants reported visiting ER at the second 6 months 108 13% 5%

Hospitalizations
Include number of participants reporting hospitalization at baseline 164 14%
Include number of participants reporting hospitalization at the first 6 months 100 11% 3%
Include number of participants reporting hospitalization at the second 6 months 116 14% 0%

2. IDENTIFY MEAN COSTS FOR HEALTH CARE UTILIZATION FROM 2010 MEDICAL EXPENDITURE PANEL SURVEY (MEPS)

Age distribution
Include % for those 18–44 years of age 10%
Include % for those 45–64 years of age 31%
Include % for those 65+ years of age 59%

ER visits
Mean costs of ER visits for those 18–44 years of age $ 1,465.00
Mean costs of ER visits for those 45–64 years of age $ 1,738.00
Mean costs of ER visits for those 65+ years of age $ 1,403.00
Age-adjusted cost of ER visits $ 1,513.05
Cost savings associated with ER visits per person at the first 6 months $ 75.65
Cost savings associated with ER visits per person at the second 6 months $ 75.65
Total cost savings associated with ER visits per person at two time periods $ 151.31

Hospitalizations
Mean costs of hospitalizations for those 18–44 years of age $ 11,501.00
Mean costs of hospitalizations for those 45–64 years of age $ 21,462.00
Mean costs of hospitalizations for those 65+ years of age $ 18,554.00
Age-adjusted cost of hospitalizations $ 18,750.18
Cost savings associated with hospitalizations per person at the first 6 months $ 562.51
Cost savings associated with hospitalizations per person at the second 6 months $–
Total cost savings associated with hospitalizations per person at two time periods $ 562.51

3. ESTIMATE COSTS SAVED FROM REDUCED UTILIZATION FOR THE PERIOD OF TIME YOU ARE INTERESTED IN EXAMINING

Based on national information, potential annual health care savings per CDSMP participant from $ 713.81
averting ER visits ($ 151.31) and hospitalizations ($ 562.51) can be estimated

4. ESTIMATE AVERAGE ANNUAL PROGRAM DELIVERY COSTS

Estimated program delivery costs per person in the National CDSMP study $ 350.00

5. DEDUCT ANNUAL PROGRAM COSTS FROM ESTIMATED HEALTH CARE UTILIZATION SAVINGS

Based on national information and using average CDSMP costs per participant ($ 350.00), net $ 363.81
cost savings related to ER visits and hospitalizations per CDSMP participant can be estimated

(Continued)

www.frontiersin.org April 2015 | Volume 3 | Article 42 | 199

Ahn et al. Healthcare cost savings estimator tool

Table 1 | Continued

N % Change in %

6. EXTRAPOLATE TO NATIONAL SAVINGS USING CENSUS DATA COMBINED WITH MEPS DATA
Number of American adults from census data by age 234,564,071 100%
18–44 112,806,642 48%
45–64 81,489,445 35%
65+ 40,267,984 17%
Estimated % of American adults having at least 1 chronic condition from MEPS data by age 77%
18–44 71%
45–64 84%
65+ 94%
Number of American adults aged 18 and older having at least one chronic condition 180,614,335
Cost savings if you could reach ALL American adults age 18+ having at least one chronic $ 65,709,373,342.03
condition
Include % of this population you want to reach 5%
Based on per participant program annual net savings ($ 363.81) for the population you want to $ 3,285,468,667.10
reach (5%), national health care savings can be estimated

a
Be aware of potential limitations when presenting your data.
The cost estimates presented must be treated as general estimates, as they are not based on precise cost expenditures. Yet, we feel they are robust for purposes
of providing ballpark health care utilization costs, program delivery expenses, and estimated net savings to support the widespread dissemination and sustainability
of evidence-based chronic disease self-management programs.

6 months (5% reduction × $1,513)]. Using the same MEPS extrapolated from per participant annual net savings to national
dataset, we identified $18,750 as an age-adjusted cost of hos- savings using Census and MEPS data. We first identified the age
pitalizations, and total cost savings associated with hospital- distribution of American adults from 2010 Census data: 18–44
izations per person at two time periods amounted to $562.50 (112.8 million, 48%), 45–64 (81.5 million, 35%), and 65+ (40.3
[first 6 months (3% reduction × $18,750) + second 6 months million, 17%). From the 2010 MEPS data, we tallied percentages of
(0% × $18,750)]. American adults having at least 1 chronic condition: 18–44 (71%),
Step 3: estimate costs saved from reduced ER visits and hospital- 45–64 (84%), and 65+ (94%). Thus, the age-adjusted number
izations for two 6-month periods of CDSMP. of American adults aged 18 and older having at least 1 chronic
Table 1 shows that $714 was the potential annual healthcare condition was 180.6 million [i.e., (112.8 million × 71%) + (81.5
savings per CDSMP participant from averting ER visits ($151) million × 84%) + (40.3 million × 94%)]. Finally, $3.3 billion in
and hospitalizations ($563). healthcare costs may be saved by averting ER visits and hos-
Step 4: Estimate average annual CDSMP costs. pitalizations if the CDSMP could reach 5% of this population
Table 1 shows that we suggest $350 as the estimated program (180.6 million × 5% × $364). It is also important to note that the
delivery cost per person in the National Study of CDSMP based national extrapolation in Step 6 can be replaced by local projec-
on best estimates from experts and field reports (15). It should tions based on participant reach and age distributions of those
be noted that estimates were based on the cost of $3,500 per projected participants.
CDSMP workshop, thus the cost per participant ranged from
$219 for workshops of 16 participants, $350 for workshops of 10 TAILORING THE HEALTHCARE COST SAVINGS ESTIMATOR
participants, and $583 for workshops of 6 participants. TOOL
Step 5: Deduct the average annual CDSMP costs (#4) from the The information requested of users wanting to tailor their region-
estimated cost savings due to reduced ER visits and hospitaliza- specific estimates is described in Table 2. Users are asked to provide
tions (#3). data points including the number of CDSMP participants, ER vis-
The potential annual net healthcare cost savings of $364 per its, and hospitalizations at baseline, 6- and 12-month; participant
participant was found by deducting the annual per participant age distribution at baseline; and estimated program delivery cost
program costs ($350) from the estimated annual per participant per participant. This information will be used to estimate net cost
healthcare savings ($714) (Table 1). savings in Step 5 based on their current data sources. In Step 6, this
Step 6: Extrapolate to national savings using Census data among tool can be further tailored by estimating new net cost savings and
American adults (with a population size of 234.5 million age projecting total healthcare net savings in the next 12 months based
18 years and above) having at least 1 chronic condition combined on the expected number of participants (e.g., 200) to be enrolled
with MEPS data. and their anticipated age distribution (e.g., increasing reach of
middle-aged participants by 10%).
Table 1 shows the amount of money that might be saved by The data points described above are derived from various data
implementing the program nationally. To calculate this figure, we sources including the user’s assessment data (i.e., collected from

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 42 | 200
Ahn et al. Healthcare cost savings estimator tool

Table 2 | Data points, data sources, formats, and recommendations needed for users to tailor cost estimates.

Data point Data source Format in tool Recommendation for measurement

Number of CDSMP participants User’s assessment Open-ended Collected at baseline and 6 months
data (and 12 months, if possible)
Number of ER visits User’s assessment Open-ended Collected at baseline and 6 months
data (and 12 months, if possible)
Number of hospitalizations User’s assessment Open-ended Collected at baseline and 6 months
data (and 12 months, if possible)
Participant age distribution User’s assessment Open-ended Categorized as 18–44, 45–64, and
data 65 years and older
Estimated program delivery cost (per participant) User’s administrative Drop-down Choices of $219, $250, $292, $350,
data menu $438, $583, or open-ended (override)
Number of eligible individuals (aged 18+ years with 1+ chronic User’s projection Open-ended Open-ended
conditions) to be served by CDSMP in next 12 months about reach

participants using questionnaires at baseline and follow-up time among participants (n = 125). Ms. Jones also wants to know how
points); user’s administrative data (i.e., gathered from delivery sites much healthcare costs could be saved if she knows the expected
and administrative records about workshop characteristics); and number of participants (n = 200) to be enrolled in CDSMP in
the user’s projections about per participant cost to deliver the pro- next 12 months and the age distribution of these participants. The
gram and reach (i.e., the projected number of participants (as well six-step process taken by Ms. Jones utilizing the excel-based tool
as the new age distribution of participants) the user anticipates is described below. Ms. Jones entered relevant numbers marked
enrolling in the forthcoming 12-month period). in diagonal stripe based on her data and projections for her target
The tool comes complete with a set of step-by-step instructions region/service area (Table 3). She consulted her records and the
about data to be entered for tailored estimates. Drop-down menus recollection of her colleagues and partners to gather data including
are provided to ensure default values (e.g., those calculated from number of CDSMP participants (at baseline, 6 and 12 months), ER
the National Study of CDSMP described above) exist from which visits, hospitalizations, baseline age distribution, estimated pro-
to calculate estimates; however, users can override the drop-down gram delivery cost, and the expected number of participants (and
menu options by entering their own responses from their data. The their anticipated age distribution) to be enrolled in CDSMP in
more user data that is entered, the more tailored the cost savings next 12 months.
estimates will be. It should be noted that cost estimates generated
with this Tool are only estimates and the Tool does not calculate
actual cost savings. It is also noted that there should be at least 100 Step 1: Examine the pattern of ER visits and hospitalizations
participants to make the estimates stable. among CDSMP participants (n = 125) in the first and second
Two heuristic case examples (i.e., a small county and a metro- 6-month periods
politan city) are described below to show how users can utilize the Table 3 shows changes in ER visits from baseline (16%) to
Tool with their own data to create tailored cost savings estimates for 6 months (11%) to 12 months (11%). ER visits were reduced
their existing and future CDSMP participant populations. These between baseline to 6 months (5%) and baseline to 12 months
examples reinforce how the context and methods of CDSMP deliv- (5%). Table 3 also shows changes in hospitalizations from base-
ery impact potential cost savings. They also demonstrate the value line (12%) to 6 months (10%) to 12 months (11%). Hospitaliza-
of the Tool for demonstrating potential savings when the age dis- tions were also reduced between baseline and 6 months (2%) and
tribution of the projected participant population is adjusted to baseline to 12 months (1%).
target older age groups. One case example concerns the Depart- Step 2: Identify mean costs for ER visits and hospitalizations from
ment of Public Health located in a small county while the other 2010 medical expenditure panel survey (MEPS)
case example pertains to an academic institution located in an Table 3 shows mean costs for ER visits and hospitalizations
urban area. from 2010 MEPS by taking into account the age distribution
of the small county as noted above. Then Ms. Jones identified
CASE EXAMPLE #1 (DEPARTMENT OF PUBLIC HEALTH IN A SMALL mean costs of ER visits by the aforementioned age categories
COUNTY) from the 2010 MEPS dataset and found $1,514.77 as the age-
Ms. Jones is the director of the Department of Public Health in adjusted cost of ER visits. Thus, total cost savings associated with
a small county with a population size of 7,774, in which 56% ER visits per person through the two time periods amounted
of adults are 18–44 years of age, 23% are 45–64 years of age, and to $151.48 [first 6 months (5% reduction × $1,514.77) + second
21% are 65 years of age or older according to the Census. She 6 months (5% reduction × $1,514.77)]. Using the same MEPS
wants to know how much her CDSMP program might be reduc- dataset, she identified $15,273.16 as an age-adjusted cost of hospi-
ing healthcare costs by averting ER visits and hospitalizations talizations, and total cost savings associated with hospitalizations

www.frontiersin.org April 2015 | Volume 3 | Article 42 | 201

Ahn et al. Healthcare cost savings estimator tool

Table 3 | Healthcare cost savings estimator tool: a small county.

Healthcare cost savings estimator tool: CDSMPa

National study case example Your local example: if you have data, please
enter relevant numbers in cells marked in
diagonal stripe for your population. This
spreadsheet will make automatic
calculations for you.

N % Change N % Change
in % in %

1. EXAMINE THE PATTERN OF HEALTH CARE UTILIZATION WITHIN YOUR POPULATION

Emergency room (ER) visits

Include number of participants at baseline 1170 125
Include number of participants reported visiting ER at baseline 211 18% 20 16%
Include number of participants at 6 months 903 115
Include number of participants reported visiting ER at the first 118 13% 5% 13 11% 5%
6 months
Include number of participants at 12 months 825 105
Include number of participants reported visiting ER at the second 108 13% 5% 12 11% 5%
6 months

Hospitalizations
Include number of participants reporting hospitalization at baseline 164 14% 15 12%
Include number of participants reporting hospitalization at the first 100 11% 3% 11 10% 2%
6 months
Include number of participants reporting hospitalization at the 116 14% 0% 12 11% 1%
second 6 months

2. IDENTIFY MEAN COSTS FOR HEALTH CARE UTILIZATION FROM 2010 MEDICAL EXPENDITURE PANEL SURVEY (MEPS)

Age distribution Indicate the age distribution for your population

Include % for those 18–44 years of age 10% Indicate % for those 18–44 56%
Include % for those 45–64 years of age 31% Indicate % for those 45–64 23%
Include % for those 65+ years of age 59% Indicate % for those 65+ 21%

ER visits
Mean costs of ER visits for those 18–44 years of age $ 1,465.00 $ 1,465.00
Mean costs of ER visits for those 45–64 years of age $ 1,738.00 $ 1,738.00
Mean costs of ER visits for those 65+ years of age $ 1,403.00 $ 1,403.00
Age-adjusted cost of ER visits $ 1,513.05 $ 1,514.77
Cost savings associated with ER visits per person at the first $ 75.65 $ 75.74
6 months
Cost savings associated with ER visits per person at the second $ 75.65 $ 75.74
6 months
Total cost savings associated with ER visits per person at two time $ 151.31 $ 151.48
periods

Hospitalizations
Mean costs of hospitalizations for those 18–44 years of age $ 11,501.00 $ 11,501.00
Mean costs of hospitalizations for those 45–64 years of age $ 21,462.00 $ 21,462.00
Mean costs of hospitalizations for those 65+ years of age $ 18,554.00 $ 18,554.00
Age-adjusted cost of hospitalizations $ 18,750.18 $ 15,273.16
Cost savings associated with hospitalizations per person at the $ 562.51 $ 305.46
first 6 months

(Continued)

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 42 | 202
Ahn et al. Healthcare cost savings estimator tool

Table 3 | Continued

N % Change N % Change
in % in %

Cost savings associated with hospitalizations per $– $ 152.73

person at the second 6 months
Total cost savings associated with hospitalizations per $ 562.51 $ 458.19
person at two time periods
3. ESTIMATE COSTS SAVED FROM REDUCED UTILIZATION FOR THE PERIOD OF TIME YOU ARE INTERESTED IN EXAMINING
Based on national information, potential annual health $ 713.81 Potential annual health $ 609.67
care savings per CDSMP participant from averting ER care savings
visits ($ 151.31) and hospitalizations ($ 562.51) can be ($ 151.48 + $ 458.19)
estimated

4. ESTIMATE AVERAGE ANNUAL PROGRAM DELIVERY COSTS

Estimated program delivery costs per person in the $ 350.00 Select your closest $ 438.00
National CDSMP study program cost per person
from the drop-down menu

5. DEDUCT ANNUAL PROGRAM COSTS FROM ESTIMATED HEALTH CARE UTILIZATION SAVINGS

Based on national information and using average $ 363.81 Net cost savings $ 171.67
CDSMP costs per participant ($ 350.00), net cost ($ 609.67 − $ 438.00)
savings related to ER visits and hospitalizations per
CDSMP participant can be estimated

6. EXTRAPOLATE TO NATIONAL SAVINGS USING CENSUS DATA COMBINED 6. CALCULATE YOUR SAVINGS BASED
WITH MEPS DATA ON POPULATION TO REACH AND NEW
AGE DISTRIBUTION

Number of American adults from Census data by age 234,564,071 100% Number of potential 200 100%
participants reflecting their
age distribution
18-44 112,806,642 48% 18–44 102 51%

45-64 81,489,445 35% 45–64 66 33%

65+ 40,267,984 17% 65+ 32 16%

Estimated % of American adults having at least 1 77% Net Cost Savings based on 194.02
chronic condition from MEPS data by age population to reach and
new age distribution
18-44 71%
45-64 84%

65+ 94%

Number of American adults aged 18 and older having at 180,614,335

least 1 chronic condition

Cost savings if you could reach ALL American adults age $ 65,709,373,342.03
18+ having at least 1 chronic condition
Include % of this population you want to reach 5%

Based on per participant program annual net savings ($ $ 3,285,468,667.10 Your healthcare net cost $38,803.06
363.81) for the population you want to reach (5%), savings by averting ER
national health care savings can be estimated visits and hospitalizations
attributed to CDSMP

a
Be aware of potential limitations when presenting your data.
The cost estimates presented must be treated as general estimates, as they are not based on precise cost expenditures. Yet, we feel they are robust for purposes
of providing ballpark health care utilization costs, program delivery expenses, and estimated net savings to support the widespread dissemination and sustainability
of evidence-based chronic disease self-management programs.

www.frontiersin.org April 2015 | Volume 3 | Article 42 | 203

Ahn et al. Healthcare cost savings estimator tool

per person through two time periods amounted to $458.19 participants (with their anticipated new age distribution) to be
[first 6 months (2% reduction × $15,273.16) + second 6 months enrolled in CDSMP in next 12 months.
(1% × $15,273.16)].
Step 3: Estimate costs saved from reduced ER visits and hospital- Step 1: Examine the pattern of ER visits and hospitalizations
izations for two 6-month periods of CDSMP among CDSMP participants (n = 500) in the first and second
Table 3 shows that $609.67 was the potential annual healthcare 6-month periods
savings per CDSMP participant by averting ER visits ($151.48) Table 4 shows changes in ER visits from baseline (15%) to
and hospitalizations ($458.19). 6 months (12%) to 12 months (11%). ER visits were reduced
Step 4: Estimate average annual CDSMP costs between baseline to 6 months (3%) and baseline to 12 months
Table 3 shows that Ms. Jones suggests $438 as the estimated (4%). Table 4 also shows changes in hospitalizations from base-
program delivery costs per person based on the average number line (16%) to 6 months (11%) to 12 months (15%). Hospital-
of participants in each workshop and the organizational capacity izations were reduced between baseline and 6 months (5%) and
of providing CDSMP. baseline to 12 months (1%).
Step 5: Deduct the average annual CDSMP costs (#4) from the Step 2: Identify mean costs for ER visits and hospitalizations from
estimated cost savings from reduced ER visits and hospitaliza- 2010 medical expenditure panel survey (MEPS)
tions (#3) Table 4 shows mean costs for ER visits and hospitaliza-
The potential annual net healthcare cost savings of $171.67 per tions from 2010 MEPS accounting for the age distribution
participant was found by deducting the annual per participant of the large city as noted above. Then Mr. Smith identified
program cost ($438) from the estimated annual per participant mean costs of ER visits by the aforementioned age categories
healthcare savings ($609.67) (Table 3). from the 2010 MEPS dataset and found $1,501.24 as the age-
Step 6: Project healthcare cost savings based on the expected adjusted cost of ER visits. Thus, total cost savings associated
number of participants to be enrolled in next 12 months and with ER visits per person at two time periods amounted to
their anticipated age distribution. $105.09 [first 6 months (3% reduction × $1,501.24) + second
6 months (4% reduction × $1,501.24)]. Using the same MEPS
After acknowledging $171.67 as per person net cost savings dataset, he identified $13,941.12 as an age-adjusted cost of hos-
among CDSMP participants in Step 5, Ms. Jones wants to project pitalizations, and total cost savings associated with hospital-
healthcare cost savings when reaching 200 people in the next izations per person at two time periods amounted to $836.47
12 months (Table 3). She also decides to recruit 10% more middle- [first 6 months (5% reduction × $13,941.12) + second 6 months
aged adults (i.e., from 23 to 33%) after realizing that the costs of (1% × $13,941.12)].
ER visits and hospitalization for this population is more expensive Step 3: Estimate costs saved from reduced ER visits and hospital-
than their younger or older counterparts (also shown in Table 3). izations for two 6-month periods of CDSMP
As a result, when reaching more middle-aged population [i.e., Table 4 shows that $941.55 was the potential annual health-
33% compared to younger (51%) and older adults (16%)], Ms. care savings per CDSMP participants from averting ER visits
Jones estimates $194.02 as the new net cost savings, and concludes ($105.09) and hospitalizations ($836.47).
CDSMP could potentially help save $38,804 (i.e., 200 × $194.02). Step 4: Estimate average annual CDSMP costs
This equates to approximately $4,000 more healthcare cost savings Table 4 shows that Mr. Smith suggests $250 as the estimated
than the age distribution of her existing CDSMP participant pool. program delivery cost per person based on the average number
of participants in each workshop and the organizational capacity
CASE EXAMPLE #2 (ACADEMIC INSTITUTION IN AN URBAN AREA) of providing CDSMP.
Mr. Smith is a director of the Healthy Aging Network at an acad- Step 5: Deduct the average annual CDSMP costs (#4) from the
emic institution providing CDSMP in a metropolitan city with estimated cost savings from reduced ER visits and hospitaliza-
a population size of 940,764, in which 72% of adults are 18– tions (#3)
44 years of age, 16% are 45–64 years of age, and 12% are 65 years The potential annual net healthcare cost savings of $691.55 per
of age or older according to the Census. He wants to know how participant was found by deducting the annual per participant
much his CDSMP program could potentially reduce healthcare program costs ($250) from the estimated annual per participant
costs by averting ER visits and hospitalizations among participants healthcare savings ($941.55) (Table 4).
(n = 500). Mr. Smith also wants to project healthcare costs saved Step 6: Project healthcare cost savings based on the expected
if he knows the expected number of participants to be enrolled in number of participants to be enrolled in next 12 months and
CDSMP in next 12 months (n = 1,000) and their anticipated age their anticipated age distribution.
distribution. The six-step process taken by Mr. Smith utilizing the
Excel-based Tool is described below. Mr. Smith entered relevant Mr. Smith estimates $691.55 as per person net cost savings
numbers marked in diagonal stripe based on his data and projec- among CDSMP participants in Step 5. Now, Mr. Smith wants to
tions for his target region/service area (Table 4). He consulted his project healthcare cost savings when reaching 1,000 people in the
records and the recollection of his colleagues and partners to gather next 12 months (Table 4). He also decides to recruit 10% more
data including number of CDSMP participants (at baseline, 6 and middle-aged adults (i.e., from 16% to 26%) after realizing that the
12 months), ER visits, hospitalizations, baseline age distribution, costs of ER visits and hospitalization for this population is more
estimated program delivery cost, and the expected number of expensive than their younger or older counterparts (also shown

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 42 | 204
Ahn et al. Healthcare cost savings estimator tool

Table 4 | Healthcare cost savings estimating tool: a metropolitan city.

Healthcare cost savings estimator tool: CDSMPa

National Study Case Example Your Local Example: If you have data, please enter
relevant numbers in cells marked in diagonal
stripe for your population. This spreadsheet will
make automatic calculations for you

N % Change N % Change
in % in %

1. EXAMINE THE PATTERN OF HEALTH CARE UTILIZATION WITHIN YOUR POPULATION

Emergency room (ER) visits
Include number of participants at baseline 1170 500
Include number of participants reported visiting ER at 211 18% 75 15%
baseline
Include number of participants at 6 months 903 450
Include number of participants reported visiting ER at the 118 13% 5% 55 12% 3%
first 6 months
Include number of participants at 12 months 825 400
Include number of participants reported visiting ER at the 108 13% 5% 44 11% 4%
second 6 months

Hospitalizations
Include number of participants reporting hospitalization 164 14% 80 16%
at baseline
Include number of participants reporting hospitalization 100 11% 3% 50 11% 5%
at the first 6 months
Include number of participants reporting hospitalization 116 14% 0% 60 15% 1%
at the second 6 months
2. IDENTIFY MEAN COSTS FOR HEALTH CARE UTILIZATION FROM 2010 MEDICAL EXPENDITURE PANEL SURVEY (MEPS)
Age distribution Indicate the age distribution for your population
Include % for those 18-44 years of age 10% Indicate % for those 18-44 72%
Include % for those 45-64 years of age 31% Indicate % for those 45-64 16%
Include % for those 65+ years of age 59% Indicate % for those 65+ 12%
ER Visits
Mean costs of ER visits for those 18-44 years of age $ 1,465.00 $ 1,465.00
Mean costs of ER visits for those 45-64 years of age $ 1,738.00 $ 1,738.00
Mean costs of ER visits for those 65+ years of age $ 1,403.00 $ 1,403.00
Age-adjusted cost of ER visits $ 1,513.05 $ 1,501.24
Cost savings associated with ER visits per person at the $ 75.65 $ 45.04
first 6 months
Cost savings associated with ER visits per person at the $ 75.65 $ 60.05
second 6 months
Total cost savings associated with ER visits per person at $ 151.31 $ 105.09
two time periods

Hospitalizations
Mean costs of hospitalizations for those 18-44 years of $ 11,501.00 $ 11,501.00
age
Mean costs of hospitalizations for those 45-64 years of $ 21,462.00 $ 21,462.00
age
Mean costs of hospitalizations for those 65+ years of age $ 18,554.00 $ 18,554.00
Age-adjusted cost of hospitalizations $ 18,750.18 $ 13,941.12
Cost savings associated with hospitalizations per person $ 562.51 $ 697.06
at the first 6 months

(Continued)

www.frontiersin.org April 2015 | Volume 3 | Article 42 | 205

Ahn et al. Healthcare cost savings estimator tool

Table 4 | Continued

N % Change N % Change
in % in %

Cost savings associated with hospitalizations per $– $ 139.41

person at the 2nd 6 months
Total cost savings associated with hospitalizations $ 562.51 $ 836.47
per person at two time periods
3. ESTIMATE COSTS SAVED FROM REDUCED UTILIZATION FOR THE PERIOD OF TIME YOU ARE INTERESTED IN EXAMINING
Based on national information, potential annual $ 713.81 Potential annual health $ 941.55
health care savings per CDSMP participant from care savings
averting ER visits ($ 151.31) and hospitalizations ($ 105.09 + $ 836.47)
($ 562.51) can be estimated

4. ESTIMATE AVERAGE ANNUAL PROGRAM DELIVERY COSTS

Estimated program delivery costs per person in the $ 350.00 Select your closest $ 250.00
National CDSMP study program cost per person
from the drop-down menu

5. DEDUCT ANNUAL PROGRAM COSTS FROM ESTIMATED HEALTH CARE UTILIZATION SAVINGS

Based on national information and using average $ 363.81 Net cost savings $ 691.55
CDSMP costs per participant ($ 350.00), net cost ($ 941.55 − $ 250.00)
savings related to ER visits and hospitalizations per
CDSMP participant can be estimated

6. EXTRAPOLATE TO NATIONAL SAVINGS USING 6. CALCULATE YOUR SAVINGS BASED ON

CENSUS DATA COMBINED WITH MEPS DATA POPULATION TO REACH AND NEW AGE
DISTRIBUTIOIN
Number of American adults from Census data by 234,564,071 100% Number of potential 1,000 100%
age participants reflecting their
age distribution
18-44 112,806,642 48% 18-44 670 67%

45-64 81,489,445 35% 45-64 260 26%

65+ 40,267,984 17% 65+ 70 7%

Estimated % of American adults having at least 1 77% Net Cost Savings based on 732.29
chronic condition from MEPS data by age population to reach and
new age distribution
18-44 71%
45-64 84%

65+ 94%
Number of American adults aged 18 and older 180,614,335
having at least 1 chronic condition
Cost savings if you could reach ALL American $ 65,709,373,342.03
adults age 18+ having at least 1 chronic condition
Include % of this population you want to reach 5%
Based on per participant program annual net $ 3,285,468,667.10 Your healthcare net cost $732,289.00
savings ($ 363.81) for the population you want to savings by averting ER
reach (5%), national health care savings can be visits and hospitalizations
estimated attributed to CDSMP

a
Be aware of potential limitations when presenting your data.
The cost estimates presented must be treated as general estimates, as they are not based on precise cost expenditures. Yet, we feel they are robust for purposes
of providing ballpark health care utilization costs, program delivery expenses, and estimated net savings to support the widespread dissemination and sustainability
of evidence-based chronic disease self-management programs.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 42 | 206
Ahn et al. Healthcare cost savings estimator tool

at Table 4). As a result, when reaching more middle-aged pop- reporting features and introduce it via channels such as a national
ulation [i.e., 26% compared to younger (67%) and older adults webinar, relevant health and aging services organizational web-
(7%)], Mr. Smith estimates $732.29 as the new net cost savings, sites, and social media. As a follow-up, we envision an online
and concludes CDSMP could potentially help save $732,289 (i.e., tutorial will be created to help different key stakeholders under-
1,000 × $732.29). This equates to approximately $40,000 more stand how to use (and/or collect) local or state data to estimate
healthcare cost savings than the age distribution of his existing the amount of healthcare costs saved by CDSMP for specific pop-
CDSMP participant pool. ulations of interest. Ideally, early users will provide feedback and
suggestions to help us improve the Tool and maximize its utility.
DISCUSSION Moreover, it would be also beneficial if we could extrapolate the
When users input the appropriate values for their situation, they methodology of this Tool to create new tools to estimate healthcare
will be able to use this Tool to customize estimated cost savings cost savings associated with specific chronic conditions or partic-
related to reduced healthcare utilization for participants antici- ipants of other evidence-based programs (e.g., enhance fitness, a
pated to enroll in the CDSMP within the next 12 months. More- matter of balance).
over, they can then forecast net healthcare savings of an expanded
recruitment or delivery effort (i.e., reaching even more partici- TOOL LIMITATIONS AND POTENTIAL CHALLENGES
pants and middle-aged or older participants in next 12 months). The data and methods used to develop this Tool have limitations
As illustrated in the heuristic case examples of this manuscript, the that should be acknowledged. First, ER visits and hospitalizations
Tool can be used to estimate cost savings for CDSMP programs in were self-reported healthcare measures that could be biased. How-
different size communities. For a program planner or coordinator ever, self-reported data can be fairly accurate for these utilization
to customize the Tool, they will be required to supply setting- measures, as evidenced in a national study examining the con-
specific data that can be obtained from various sources. This may cordance between self-reported and Medicare administrative data
involve review of their past records as well as consultations with for those with Medicare claims data (Jiang et al., under review).
community and academic partners to ensure accurate informa- In this prior study, we identified moderate [for ER visits; kappa
tion and projections are entered into the Tool. The availability of statistics (0.45–0.61)] to substantial [for hospitalization; kappa
CDSMP baseline and follow-up data (including ER and hospi- (0.69–0.79)] concordance among 119 CDSMP participants (19).
talization utilization) and documented per participant program Next, all estimates applied in the Tool have been based on the sta-
costs are important for tailoring cost savings estimates. Similarly, tic 2010–2012 National Study of CDSMP, 2010 MEPS, and 2010
additional tailoring is possible with access to data about the age Census datasets. The primary reason for this is that the national
distribution of the target community and the mean costs for ER initiative to implement CDSMP started in 2010. Therefore, we syn-
and hospitalizations, which may be available at a county-level. chronously utilized the 2010 MEPS and Census dataset. To keep
Capacity to conduct such tailoring may suggest a need for tech- estimates current, inflation estimators will need to be built into
nical assistance to guide program coordinators and planners to future iterations of the Tool. We reiterate our caution that cus-
resources about how and where to locate information document- tomization of healthcare expenditures should only be attempted
ing healthcare utilizations of ER visits and hospitalizations or other if there are sufficient numbers of participants with linked health-
necessary information for customization (e.g., age-adjusted mean care utilization data (we would recommend a lower threshold of
costs of ER visits and hospitalizations among adults with at least at least 100 participants). The Excel-based Tool is also limited in
one chronic condition, per person CDSMP program cost). that it generates the number of participants to reach, which is not
Healthcare savings data should be extremely useful for pro- directly linked to the target population in a given community or
gram administrators and key decision makers. Concrete estimates region. Stated another way, the Tool does not currently calculate
of achieved savings can bolster the impact of self-reported data on the proportion or percentage of the population to be reached in the
program successes. The savings estimates can also assist program community based on the projected number of participants iden-
administrators and decision makers in developing a strong busi- tified by the user. This may be a needed feature for public health
ness case to obtain funding for CDSMP and recruiting partners policy makers whose “unit of analysis” could be a proportion of
or sponsors from other organizations who can also benefit from population to reach rather than a specific number of populations.
reduced healthcare spending and over-utilization. The Tool will An updated version of Excel-based Tool or Web-based Tool will
allow program coordinators to set performance goals and monitor reflect this feature. Additionally, the current Tool is not yet tested
progress in relation to the efficiency required to achieve the desired among users in broader fields, though plans for testing it are being
return on investment. Finally, we anticipate that users will share currently being developed.
their results internally to their organization, externally to the com-
munity, and across geographic regions to raise public awareness CONCLUSION
about the value of CDSMP. Given findings from previous studies, CDSMP could save a signifi-
cant amount of healthcare costs by averting ER visits and hospital-
FUTURE DIRECTIONS izations, if even only a small portion of the population was reached
We believe developing an accessible and user-friendly web-based (15). These results are quite encouraging in that they demonstrate
version of this Tool will be important for attracting a national cadre a positive return on investment for CDSMP nationally. The cre-
of potential users to estimate healthcare savings. When translating ation of this Tool contributes to the field by introducing a user-
this Tool to a web-based interface, we plan to offer a variety of friendly resource to help program administrators and decision

www.frontiersin.org April 2015 | Volume 3 | Article 42 | 207

Ahn et al. Healthcare cost savings estimator tool

makers more easily estimate healthcare savings among their 13. Chan WLS, Hui E, Chan C, Cheung D, Wong S, Wong R, et al. Evaluation
existing and planned CDSMP implementation efforts. of chronic disease self-management programme (CDSMP) for older adults in
Hong Kong. J Nutr Health Aging (2011) 15:209–14. doi:10.1007/s12603-010-
0257-9
ACKNOWLEDGMENTS 14. Gordon C, Galloway T. Review of Findings on Chronic Disease Self-Management
This work was supported by the Administration on Aging Program (CDSMP) Outcomes: Physical, Emotional & Health-Related Quality of
(90OP0001/03); and the National Institute of Child Health and Life, Healthcare Utilization and Costs. Washington, DC: Centers for Disease Con-
Human Development (R01HD047143). We thank members of trol and Prevention; National Council on Aging (2008).
15. Ahn S, Basu R, Smith ML, Jiang L, Lorig K, Whitelaw N, et al. The impact
the CDC-funded Prevention Research Center’s Healthy Aging of chronic disease self-management programs: healthcare savings through a
Research Network for their insights into evidence-based pro- community-based intervention. BMC Public Health (2013) 13:1141. doi:10.
graming and feedback on the development and application of 1186/1471-2458-13-1141
this community cost calculator. The authors also thank Ms. Car- 16. Lorig K. Chronic Disease Self-Management Program: Insights from the Eye of the
olyn Josephine Riordan (“Carrie Jo”), who helped to review this Storm. Stanford, CA: Patient Education Research Center, School of Medicine,
Stanford University (2014).
manuscript. 17. U.S. Department of Health and Human Services Administration on Aging.
ARRA – Communities Putting Prevention to Work: Chronic Disease Self-
REFERENCES Management Program. Washington, DC: DHHS Administration for Community
1. Tinetti ME, Fried TR, Boyd CM. Designing health care for the most common Living (2012).
chronic condition – multimorbidity. JAMA (2012) 307:2493–4. doi:10.1001/ 18. Agency for Health Research and Quality (AHRQ). Medical Expenditure Panel
jama.2012.5265 Survey: Survey Background. Washington, DC (2013).
2. Martin A, Lassman D, Whittle L, Catlin A. Recession contributes to slowest 19. Ory MG, Radcliff TA, Jiang L, Lorden A, Smith ML, Horel S, et al. Pilot Evalua-
annual rate of increase in health spending in five decades. Health Aff (2011) tion of the Chronic Disease Self-Management Program. Washington, DC: National
30:11–22. doi:10.1377/hlthaff.2010.1032 Council on Aging (2013).
3. Schneider KM, O’Donnell BE, Dean D. Prevalence of multiple chronic con-
ditions in the United States’ Medicare population. Health Qual Life Outcomes Conflict of Interest Statement: The authors declare that the research was conducted
(2009) 7:82. doi:10.1186/1477-7525-7-82 in the absence of any commercial or financial relationships that could be construed
4. Emanuel EJ. Where are the health care cost savings? JAMA (2012) 307:39–40. as a potential conflict of interest.
doi:10.1001/jama.2011.1927
5. Lorig K, Sobel DS, Stewart AL, Brown BW Jr, Bandura A, Ritter P, et al. Evidence This paper is included in the Research Topic, “Evidence-Based Programming for Older
suggesting that a chronic disease self-management program can improve health Adults.” This Research Topic received partial funding from multiple government and
status while reducing hospitalization: a randomized trial. Med Care (1999) private organizations/agencies; however, the views, findings, and conclusions in these
37:5–14. doi:10.1097/00005650-199901000-00003 articles are those of the authors and do not necessarily represent the official position
6. Brady TJ, Murphy L, Beauchesne D, Bhalakia A, Chervin D, Daniels B, et al. of these organizations/agencies. All papers published in the Research Topic received
Sorting through the evidence for the arthritis self-management program peer review from members of the Frontiers in Public Health (Public Health Education
and the chronic disease self-management program. Executive Summary of and Promotion section) panel of Review Editors. Because this Research Topic repre-
ASMP/CDSMP Meta-Analysis. Atlanta, GA: Centers for Disease Control and sents work closely associated with a nationwide evidence-based movement in the US,
Prevention (2011). 24 p. many of the authors and/or Review Editors may have worked together previously in
7. Lorig K, Holman H, Sobel D, Laurent D. Living a Healthy Life with Chronic Con- some fashion. Review Editors were purposively selected based on their expertise with
ditions: Self Management of Heart Disease, Arthritis, Diabetes, Asthma, Bronchitis, evaluation and/or evidence-based programming for older adults. Review Editors were
Emphysema and Others. Boulder, CO: Bull Publishing Company (2006). independent of named authors on any given article published in this volume.
8. Ory MG, Ahn S, Jiang L, Smith ML, Ritter P, Whitelaw N, et al. Successes of a
national study of the chronic disease self-management program: meeting the Received: 26 August 2014; accepted: 17 February 2015; published online: 27 April 2015.
triple aim of health care reform. Med Care (2013) 51:992–8. doi:10.1097/MLR. Citation: Ahn S, Smith ML, Altpeter M, Post L and Ory MG (2015) Healthcare
0b013e3182a95dd1 cost savings estimator tool for chronic disease self-management program: a new tool
9. Boutaugh M. National Dissemination of Evidence-Based Programs. Washington, for program administrators and decision makers. Front. Public Health 3:42. doi:
DC: National Council on Aging (2012). 10.3389/fpubh.2015.00042
10. Nielsen AJ, Jensen CB. Travelling frictions. Sci Technol (2013) 26:61–79. This article was submitted to Public Health Education and Promotion, a section of the
11. Fu D, Fu H, McGowan P, Shen YE, Zhu L, Yang H, et al. Implementation journal Frontiers in Public Health.
and quantitative evaluation of chronic disease self-management programme in Copyright © 2015 Ahn, Smith, Altpeter, Post and Ory. This is an open-access article
Shanghai, China: randomized controlled trial. Bull World Health Organ (2003) distributed under the terms of the Creative Commons Attribution License (CC BY).
81:174–82. The use, distribution or reproduction in other forums is permitted, provided the original
12. Wilson PM. The UK expert patients program: lessons learned and implications author(s) or licensor are credited and that the original publication in this journal is cited,
for cancer survivors’ self-care support programs. J Cancer Surviv (2008) 2:45–52. in accordance with accepted academic practice. No use, distribution or reproduction is
doi:10.1007/s11764-007-0040-z permitted which does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 42 | 208
 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00176

Linking evidence-based program participant data with

Medicare data: the consenting process and correlates of
retrospective participant consents
Philip Lloyd Ritter 1 *, Marcia G. Ory 2 , Matthew Lee Smith 3 , Luohua Jiang 4 , Audrey Alonis 1 ,
Diana D. Laurent 1 and Kate Lorig 1
1
Stanford Patient Education Research Center, Stanford School of Medicine, Stanford, CA, USA
2
School of Public Health, Texas A&M University, College Station, TX, USA
3
College of Public Health, The University of Georgia, Athens, GA, USA
4
Department of Epidemiology & Biostatistics, School of Public Health, Texas A&M Health Science Center, College Station, TX, USA

Edited by: As part of a nation-wide study of the Chronic Disease Self-Management Program (National
Roger A. Harrison, University of
Study ), older participants were asked to consent to have their Medicare data matched
Manchester, UK
with study data. This provided an opportunity to examine the consenting process and
Reviewed by:
Shu-Chun Chuang, National Health compare consenters, refusers, and non-responders. We compared the three groups on
Research Institutes, Taiwan a large number of variables. These included demographic, National Study participation,
Daniel Francis Fahey, California State health indicator, health behavior, and health-care utilization variables. We assessed differ-
University San Bernardino, USA
ences in 6-month change scores for time-varying variables. We also examined whether
*Correspondence:
asking participants to consent prior to the final questionnaire impacted completion of that
Philip Lloyd Ritter , 1000 Welch Road,
Suite 204, Palo Alto, CA 94304, USA questionnaire. Of 616 possible participants, 42% consented, 44% refused, and 14% failed
e-mail: [email protected] to respond. Differences by ethnicity were found, with Hispanics more likely to consent.
There was a consistent tendency for those who participated most in the National Study
to consent. With the exception of number of chronic diseases, there was no evidence of
health indicators or health behaviors being associated with consenting. Participants with
more physician visits and more nights in the hospital were also more likely to consent.
Those asked to consent before the 12-month follow-up questionnaire were less likely to
complete that questionnaire than those who were asked after. Fewer than half consented
to link to their Medicare data. The greater willingness to consent by those who participated
most suggests that willingness to consent may be part of program engagement. Consen-
ters had more diseases, more MD visits, and more nights in the hospital, suggesting that
greater contact with the medical system may be associated with willingness to consent.
This indicates that examinations of Medicare data based only on those willing to consent
could introduce bias. Asking for consent appears to reduce participation in the larger study.
Keywords: chronic disease self-management, patient education, Medicare, consenting, cost analysis

INTRODUCTION outcomes have been published elsewhere (3, 4, 5, Ory et al. in

Most of what we know about the effectiveness of evidence-based prepartion).
chronic disease self-management programs (CDSMP) comes from After completion of the initial intervention and during the col-
self-reports of health and health-care outcomes experienced by lection of follow-up questionnaires, the Centers for Medicare &
participants (1). In 2010, the National Council on Aging (NCOA), Medicaid Services (CMS) contracted with NCOA for a pilot study
the Stanford Patient Education Research Center and Texas A&M to examine the feasibility of matching National Study partici-
Health Science Center’s Program on Healthy Aging initiated a pants with their CMS data. Linking with administrative claims
major longitudinal nation-wide U.S. study of participants in a data would provide an alternative and potentially more precise
dissemination of the Stanford CDSMP (2). The primary pur- method for examination of health-care utilization and associated
poses of the project were to inform NCOA’s technical assistance costs savings attributed to program participation.
work and assessing the impact of the program when offered Because consent to match study data with CMS data was not
in a variety of “real world” settings across the nation. Base- obtained at the beginning of the National Study, all potential
line enrollment of study participants began in August 2010 subjects had to provide supplemental consent for the specific
and ended in April 2011, with subsequent collection of 6- and purposes of having their CDSMP data linked to CMS Medicare
12-month follow-up survey data. That study is known as the Administrative Data. A subset of National Study participants who
U.S. National Study of the CDSMP (referred to as the National were at least 65.5 years of age at the beginning of the National
Study), and details of the intervention and the self-reported Study were invited to enroll in the CMS study. This paper reports

www.frontiersin.org April 2015 | Volume 2 | Article 176 | 209

Ritter et al. Linking participants with Medicare data

about this consenting process and how consenters differed from: DATA ANALYSES
(a) those who actively declined to participate (refusers); and (b) Primary analyses compare those who consented to participate in
those who did not respond (non-responders). Institutional Review the CMS study with those who were eligible to participate (were
Board (IRB) approval was obtained at Stanford University and enrolled or likely to be enrolled in Medicare) but did not con-
Texas A&M University for the initial National Study and for the sent. The latter group consisted of two subsets, those who actively
subsequent consenting study reported in this paper. declined to participant and those who did not respond to con-
sent requests. Consequently, two additional sets of comparisons
MATERIALS AND METHODS were conducted comparing: (1) those who consented to those
CONSENTING PROTOCOL who actively declined; and (2) those who either consented and/or
There were several steps to the consenting process. We started declined (responded) to those who did not respond to the mailings
by mailing consent requests to 188 participants who had recently and phone calls. Given the study emphasis on who would actively
completed their 6-month questionnaires and had been 65.5 or consent to have their data linked, only those able to give consents
older at the beginning of the National Study. The process would (e.g., living participants) were included in these analyses.
continue as other participants completed or would have completed Comparisons between groups of individuals (consenters,
6- or 12-month follow-up questionnaires. These first mailings refusers, and non-responders) were made using demographic,
occurred in August and September 2011. Potential participants CDSMP workshop participation, health indicator, health behavior,
were asked to provide the last four digits of their social security and health-care utilization variables (described below). Differ-
number (SSN) and to consent to allow their study identifying ences between groups were tested using independent sample t -
information to be used to obtain Medicare claims data. There was test for continuous variables, chi-squares for categorical variables,
an initial assumption that having a partial SSN would acceler- and non-parametric (Wilcoxon) tests for low frequency medical
ate the matching process. After 3 weeks and several follow-up or utilization variables.
attempted follow-up contacts by telephone, only 23% of the initial The consent forms were first mailed to all potential partici-
188 potential participants had consented. Feedback from partici- pants after they had the opportunity to complete 6-month or,
pants revealed some concerns about providing SSNs. We therefore in the case of the earliest National Study recruitment cohorts,
suspended the consenting process and modified the protocol for 12-month follow-up questionnaires. Thus, we were also able to
those who had not yet responded and for subsequent mailings. In examine whether 6-month changes were related to whether par-
the revised protocol, we asked participants for permission to match ticipants consented, refused, or did not respond. We compared
their study identifying data with their Medicare data using name, mean changes on two health indicators, three health behaviors,
gender, address, and date of birth – four identifiers that we hypoth- and three health-care utilization measures.
esized would yield fairly accurate matches with CMS records.
Detailed information on processes for linking various administra- MEASURES
tive data sets can be found elsewhere (6, 7). Given the low initial Demographic variables included age, gender, number of years of
response rate and stated concerns among older adults about reveal- education, and ethnic identification (African-American, Hispanic,
ing such highly identifiable information, requests for any part of or non-Hispanic white). CDSMP program participation was mea-
participants’ SSNs were dropped. The following six-step protocol sured in a number of ways. Both the mean number of workshop
was followed for the remainder of the study. sessions attended (out of a possible six) and completion of the pro-
gram (defined as having attended at least four of the six sessions)
Step 1: Each potential participant received a short hand- were tabulated. Assuming that those who had previously con-
addressed note explaining the CMS study and telling them that in sented to be in a sub-study might differ from those who had not, we
a few days they would receive a gray envelope containing the study calculated the percentage of National Study participants who were
details, consent forms, and a small gift. The gray envelope was also participating in a sub-study for people with Type 2 diabetes
used so that the mailing could not be confused with the National and had agreed to furnish blood samples for testing hemoglobin
Study questionnaires, which were sent out in white envelopes. A1c levels (8). Finally, we tabulated the proportions of participants
Step 2: Three days later consent forms were mailed along with a who completed 6- and 12-month follow-up questionnaires as part
gift of four “forever” stamps. of the larger National Study.
Step 3: Five to 10 days later at least two calls were made. Messages Three health indicators were measured. These consisted of the
were left on the second calls if participants were not yet reached. mean number of comorbid conditions reported, PHQ-8 depres-
Step 4: Two weeks later those who had not responded received a sion, and self-reported general health. The PHQ-8 consists of eight
post card reminder. items, which are summed resulting in a range of 0–24 (9). The self-
Step 5: One month after the first mailing, those who had not reported general health measure consists of a single-item ranging
responded received a second consent-form mailing. from one (excellent) to five (poor) and was originally used in the
Step 6: Approximately 6 weeks after the initial mailing, phone National Health Interview Survey (10). For each of the three mea-
calls were made to participants. At least three attempts were sures, a higher score is less desirable (more conditions, greater
made to reach each participant. Consents could be obtained on depressive symptoms, and worse overall health).
the phone if study participants allowed the research assistant to The three health behaviors were whether exercised during
read the entire five-page consent statement prior to accepting via the past week, communication with physician and medication
verbal consent. adherence. The exercise measure was a single-item that asked if

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 176 | 210
Ritter et al. Linking participants with Medicare data

the participant had participated in physical activity or exercise

within the last week. Communication with physician scale is a 3- 676 Eligible for Medicare based on age
item, 6-point scale and was developed to evaluate the CDSMP and at baseline
related programs and has been described by Lorig and colleagues
(11). Medication adherence was the sum of four questions regard-
ing medication use (12). A higher score indicates less medication 639 Mailed consent forms 30 Previously withdrew or
adherence. were lost from study
We also examined three measures of health-care utilization: 6 Known to be deceased
1 not sent
physician visits, emergency department visits, and nights of hos-
pitalization in the previous 6-months. These self-report measures
have been found to be relatively unbiased when compared to health
616 Eligible (had Medicare or 21 Had no Medicare
provider records in an earlier study (13).
did not reply)
We calculated completion rates of 12-month follow-up ques- 2 Addi onal deceased
tionnaires for those who were asked to consent before 12-month
follow-up and those with consent forms sent after the 12-month
follow-up period. This was to help ascertain if the consent process
might have affected participation in follow-up within the larger 529 Responded to mailing or 87 Did not respond
study. follow-up phone calls (14% of total eligible)
(86% of total eligible)
RESULTS
PARTICIPANTS
269 Declined to par cipate
At the time CMS consent requests and forms were mailed, there (44% of total eligible)
were a total of 639 National Study participants who were the appro- 46% by mailing back form
priate age to have Medicare (65.5 or older). These people were 54% by follow-up phone calls
mailed CMS consent requests between August and December 2011
(Figure 1). Of the 639, 21 subsequently indicated that they were
not participating in Medicare for a variety of reasons but mainly
because they were still employed and/or had other medical insur- 260 Consented to par cipate 356 Did not consent
(42% of total eligible) (58% total of eligible)
ance, including veterans’ benefit. This left 618 participants with
65% by mailing back form
Medicare. An additional two had died before receiving the mail- 35% by follow-up phone calls
ing, as had 6 participants who were known to have died before
the mailing. The eight deceased individuals (six who were never
FIGURE 1 | Status of CDSMP participants invited to participate in
sent consent forms and two who were) are not included in these Medicare cost study.
analyses. Thus, there were 616 participants (618 minus the 2 who
were discovered to have been deceased) who could have actively
consented to participate. Of these, 260 consented, 169 by mail, and potential CMS study participants. The last three columns present
91 by phone. Two-hundred sixty-nine actively declined, while 87 the P-values for three sets of comparisons. The first column com-
did not respond. In summary, of the 616 eligible participants, 42% pares those who consented with all those who did not consent
consented to participate, 44% declined to participate, and 14% did (both refusers and non-responders). The second column shows
not respond. results from the comparison of those who consented with those
NON-RESPONDERS who actively declined to consent. The last column examines the
Of the 87 who did not respond, 12 indicated they did not want to comparison of those who responded with those who did not
be called or hung up and were put on a “do-not-call” list. There respond.
were 20 participants who were contacted and indicated they would Among the demographic variables, there was little differ-
return the forms but never did. These included seven who men- ence in age, education, or gender. The non-response group had
tioned being ill and seven who indicated they were very busy, higher proportions of African-Americans and Hispanics. In addi-
including two with deaths in the family. Two thought they had tion, among those who responded, the consenters had greater
returned the forms, four said they had not received the forms and proportions of African-Americans and Hispanics than did the
four requested we call back but were not reached. The remaining 45 refusers, although the differences were only marginally signifi-
were never contacted (failed to respond to mail or phone messages, cant for African-Americans (p = 0.057). The proportion Hispanic
had no or full answering machines, or had no or disconnected was significantly higher for consenters when compared to both
phone numbers). refusers and to all others. Described in another way (not shown
in the table), Hispanics were more likely to consent than non-
DIFFERENCES BETWEEN CONSENTERS, REFUSERS, AND Hispanics (53 versus 40%, p = 0.026). African-Americans were
NON-RESPONDERS AT BASELINE more likely to not respond than non-African-Americans (22 ver-
Table 1 shows the mean of continuous measures or the per- sus 13%, p = 0.027). Non-Hispanic whites had the lowest level of
centage of categories at baseline for each of the three groups of non-response (11 versus 20% for others, p = 0.002).

www.frontiersin.org April 2015 | Volume 2 | Article 176 | 211

Ritter et al. Linking participants with Medicare data

Table 1 | Baseline participant and workshop characteristics among CDSMP national study participants by consented, refused, or failed to
respond to invitation to participate in Medicare cost study.

Baseline characteristic Consented Refused No response P -value (A) P -value (B) P -value (C)
(N = 260) (N = 269) (N = 87) consent versus consent versus response versus
no consent decline no response

DEMOGRAPHIC CHARACTERISTICS
Mean age 73.7 (5.07) 74.2 (5.08) 73.7 (5.17) 0.419 0.315 0.707
Mean years of education 13.1 (3.97) 13.2 (3.23) 12.6 (4.4) 0.997 0.623 0.261
Percent male 16.5% 16.7% 16.1% 0.991 0.953 0.900
Percent African-American 15.0% 12.6% 23.0% 0.388 0.432 0.057
Percent Hispanic 18.9% 9.29% 21.8% 0.031 0.002 0.098
Percent non-Hispanic White 61.5% 68.3% 47.6% 0.660 0.104 0.002

WORKSHOP PARTICIPATION
Mean number of sessions attended (0–6) 4.92 (1.61) 4.23 (1.96) 4.13 (1.89) <0.001 <0.001 0.037
Completed program (4 +) 85.5% 72.9% 69.0% <0.001 <0.001 0.033
Participated in HbA1c study 12.7% 9.29% 5.75% 0.084 0.211 0.137
Completed 6-month questionnaire 95.0% 90.0% 48.3% <0.001 0.029 <0.001
Completed 12-month questionnaire 92.3% 80.3% 35.6% <0.001 <0.001 <0.001

HEALTH INDICATORS
Number of Chronic diseases 3.03 (1.55) 2.78 (1.45) 2.44 (1.18) 0.005 0.052 0.001
PHQ depression 5.35 (4.56) 4.79 (4.43) 6.06 (5.10) 0.513 0.155 0.061
General health 3.07 (0.882) 2.99 (0.916) 3.14 (0.904) 0.516 0.282 0.302

HEALTH BEHAVIORS
% Exercised (past week) 76.5% 76.1% 71.3% 0.809 0.908 0.313
Communication with MD 2.66 (1.33) 2.81 (1.32) 2.57 (1.46) 0.421 0.202 0.276
Medication adherence 0.808 (1.05) 0.732 (0.971) 0.779 (1.04) 0.440 0.392 0.935
HEALTH-CARE UTILIZATION
# of physician visits 3.80 (3.54) 3.29 (3.20) 2.52 (3.07) 0.012 0.081 0.009
# ED visits 0.142 (0.411) 0.205 (0.610) 0.118 (0.359) 0.658 0.486 0.526
# of hospital nights 0.946 (4.08) 0.300 (1.02) 0.977 (4.90) 0.045 0.063 0.468

For means, standard deviations are given in parentheses. Percentages are the percent within each of the three categories (consenters, refusers, and non-responders)
that belong to the variable (e.g., 16.5% of consenters were male compared to 16.7% of refusers).
P-values are from chi-square test for categorical variables and from independent sample t-tests for continuous variables, except number of ED visits and number of
hospital nights, which are from Wilcoxon rank sum tests. P(A) compares those who consented (N = 260) with all who did not consent (N = 356). P(B) compares those
who consented (N = 260) with those who actively declined to consent (N = 269). P(C) compares those who responded (N = 529) with those who did not respond
(N = 87).
P-values less than 0.05 are shown in italics.

There were a number of significant differences in workshop non-consenters and to non-responders. Consenters also had
participation indicators. Consenters attended more sessions, were a higher number of hospital nights than those who did
more likely to have completed the program and more likely to not consent.
return 6- and 12-month questionnaires. Those who had already
consented to participate in the diabetes A1c study were also more SIX-MONTH CHANGES IN HEALTH INDICATORS, BEHAVIORS, AND
highly represented among consenters than non-consenters, but the UTILIZATION
difference was not statistically significant (p = 0.084). No significant differences were found in 6-month changes in
The mean number of comorbid chronic conditions was greater the two health indicators (depression and self-reported overall
among those who consented and lower among those who did not health) and three health behaviors (exercise, communication with
respond. The other two health indicators (depression and self- physician, and medication adherence) among the three groups
reported overall health) did not differ significantly among the (Table 2). Among health-care utilization measures, those who
three groups. Similarly, there were no statistical differences among consented had a 6-month increase in emergency department visits
baseline health behaviors. compared to those who refused to consent or did not consent over-
There were two significant differences in baseline self-reported all. Although not significant, non-responders had greater reduc-
health-care utilization. Consenters had a higher mean num- tions in hospitalizations than did consenters, while those who
ber of physician visits in the last 6 months compared to all actively refused slightly increased their nights of hospitalization. As

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 176 | 212
Ritter et al. Linking participants with Medicare data

Table 2 | Six-month changes, among CDSMP national study participants by consented, refused, or failed to respond to invitation to participate
in Medicare cost study.

Baseline measure Consented Refused No response P -value (A) P -value (B) P -value (C)
(N = 246) (N = 241) (N = 42) consent versus consent versus response versus
no consent decline no response

HEALTH INDICATORS
PHQ depression −0.614 (3.85) −0.575 (4.05) −0.610 (4.28) 0.923 0.914 0.981
General health −0.069 (0.721) −0.575 (4.05) −0.095 (0.932) 0.805 0.721 0.799

HEALTH BEHAVIORS
% Exercised (past week) 0.094 (0.465) 0.075 (0.450) 0.122 (0.557) 0.762 0.644 0.620
Communication with MD 0.122 (1.14) 0.201 (1.15) −0.283 (0.986) 0.922 0.466 0.023
Medication adherence −0.036 (1.01) −0.074 (1.04) 0.214 (1.18) 0.958 0.682 0.107
HEALTH-CARE UTILIZATION
No. of physician visits 0.150 (3.55) 0.148 (3.82) 0.366 (3.00) 0.925 0.995 0.713
No. ED visits 0.029 (0.602) −0.113 (0.196) 0.0 (0.392) 0.046 0.032 0.754
No. of hospital nights −0.154 (5.77) 0.188 (2.38) −0.50 (2.39) 0.835 0.696 0.518

Standard deviations are given in parentheses. P-values are from independent sample t-tests, except number of ED visits and number of hospital nights, which are
from Wilcoxon rank sum tests. P (A) compares those who consented (N = 246) with all who did not consent (N = 281). P (B) compares those who consented (N = 246)
with those who actively declined to consent (N = 241). P (C) compares those who responded (N = 487) with those who did not respond (N = 40).
P-values less than 0.05 are shown in italics.

noted above, non-response for consents was associated with lower data and guiding future initiatives desiring to link data sources
return of 6-month questionnaires – only 48% of non-responders post hoc.
had completed 6-month questionnaires compared to 92% of The most notable differences between consenters and non-
responders (p < 0.001). Thus, the reduction in hospitalizations consenters were among the workshop participation variables.
among the non-responders may reflect a biased subset of all Consenters (versus non-consenters) and responders (versus
non-responders. non-responders) attended more sessions and were more likely to
complete the program and both 6- and 12-month follow-up ques-
DIFFERENCES BETWEEN TIMING OF THE CONSENT REQUESTS tionnaires. This is not unexpected and suggests that those more
There were 356 participants who were asked to consent after engaged with the program or with their health-care are more likely
completing the 6-month study period but before being asked to to be willing to share their Medicare information.
complete 12-month questionnaires. These consisted of all those There were little differences in demographic conditions
who had entered the National Study in 2011. There were 251 between the three groups, with the exception of ethnicity. Hispanic
participants who were asked to consent after completing the 12- and African-Americans were less likely to respond. In contrast,
month follow-up period (those who entered the study during among those who responded, members of these two minority
2010). Of those who were asked to consent after 12 months in the groups were more likely to consent. While non-Hispanic white
study, 84% had completed 12-month questionnaires. In contrast, participants were more likely to respond, they were also more
only 76% of those who were asked to consent before 12 months likely to decline to consent.
eventually completed a 12-month questionnaire (p = 0.020 from Although few statistically significant differences in health indi-
chi-square). cators and health behaviors were found between consenters and
There were no statistically significant associations between the non-consenters, participants who consented reported more ill-
proportions of participants who consented and when participants nesses or more contact with the medical system. The consenters
were asked. Among those who responded, the proportion who had higher mean number of self-reported conditions and physi-
consented was 49.5% for those asked before 12 months and 48.6% cian visits at baseline as well as less decreases in ED visits at
for those who had completed the 12-month follow-up period (not 6 months than those who did not consent.
shown in tables). While non-responders had a mean of 0.5 days decrease in hospi-
tal nights at 6-months, over 50% of the 6-month data was missing
DISCUSSION for that group. Thus, in our case, any attempt to estimate possible
RESULTS changes in medical expenditures for non-respondents would be
These data present a unique opportunity to examine factors asso- subject to bias resulting from the high attrition rate. The likelihood
ciated with older adults’ willingness to consent to have their that consenters were both more engaged with their health-care self-
programmatic data linked to administrative claims data. This management and were likely to have greater numbers of chronic
information is important for identifying potential systematic conditions would introduce further bias into studies of Medicare
biases in assessing programmatic impacts using administrative utilization.

www.frontiersin.org April 2015 | Volume 2 | Article 176 | 213

Ritter et al. Linking participants with Medicare data

IMPLICATIONS: THE CONSENTING PROCESS 6% of the 87 non-respondents refused further contact (5 partici-
There is increased concern about the third arm of the Triple Aims pants who were put on the do-not-call list) and subsequently did
for Health-Care, e.g., wanting to document that effective interven- not complete 12-month questionnaires after having completed
tions can be provided for better value (14) and lower costs. Thus, 6-month questionnaires. Thus, we are aware of at least 17 spe-
cost effectiveness of interventions is becoming more important. cific participants in the ongoing study who were likely lost to
To determine costs and cost effectiveness, at least for older adults, follow-up as a result of being contacted with a request to con-
examining Medicare claims data is treated as a “gold standard.” sent. Of those who were asked to consent after completion of
To gain such access, participants must usually sign an informed the 12-month follow-up period, 84% had completed 12-month
consent. Little is known about the population that consents to questionnaires. In contrast, only 76% of those who were asked to
examination of their claims data as opposed to those who decline. consent before 12 months eventually completed a 12-month ques-
This study opens a window into these differences. tionnaire. This suggests that as many as 28 out of 356 participants
The best variable for matching data is SSN or at least the last four did not complete 12-month questionnaires and likely would have
digits of this number. In our study, only 23% of the initial potential if they had not been asked to consent. It appears clear that the
participant population was willing to disclose this number within consenting process contributed to attrition in the larger study. For
3 weeks involving multiple contacts. Even after exhaustive follow- the National Study, where consenting at recruitment was no longer
up involving as many as eight attempted contacts, only 42% of an option, there likely would have had less effect on participation
the population was willing to consent to having any data used for in follow-up questionnaires if we had delayed the consent process
matching to Medicare data, while 44% actively refused consent. for all participants until after all follow-up was completed.
Of equal importance, we found several significant differences
between those who consented and those who did not. Of particular LIMITATIONS
interest are both the baseline differences and 6-month differences The study to match Medicare data with National Study data
in changes in self-reported health-care utilization. If these differ- was conceived and initiated after the National Study was well
ences are mirrored in Medicare claims data, it brings to question underway. Thus, we lost the opportunity to learn if consent rates
conclusions regarding the cost effectiveness of these evidence- might have been different had participants been asked to consent
based interventions. We must acknowledge that such conclusions at the time of enrollment in the larger intervention and study.
represent only those who consent and that consenters may repre- There were little differences in rates of consenting between those
sent less than half the population. Furthermore, this population who were asked 6 months after entering the study versus those
differs in several ways from those who do not consent. who were asked to consent after the 12-month follow-up period,
This study highlights limitations in using Medicare or claims but it is possible consent rates would have been higher at baseline.
data as the sole standard for assessing cost outcomes, if consent However, based on the greater attrition rate among those who were
is required. Unfortunately, in a free society without a nationalized asked to consent before the final follow-up questionnaire, it is likely
health service database, it is almost impossible to secure unbiased that asking for consent to match to Medicare data earlier might
estimates of costs. It is beyond the scope of this paper to discuss have reduced participation rates during the initial enrollment in
the problems with self-report, billing, or insurance payments. All the overall study.
have well-known problems. We would suggest that the solution to This study was limited to the consenting process and comparing
this conundrum is to use two or more methods of estimating costs consenters versus others. As noted in the Section “Materials and
and triangulating outcomes. Methods,” we do not address the actual matching of participant
There are at least two other disadvantages for a retrospective data with CMS Medicare data for those who consented to allow
consenting process for seniors enrolled in evidence-based pro- such matching. The matching process is described elsewhere (6).
grams. First, the personnel costs must be considered. In the case of Nor do we attempt to offer solutions for several issues raised. The
this study, it took one-and-one-half full-time positions over more findings suggest the need for future research on the problem of
than 3 months to attempt to consent just over 600 people. increased attrition among those asked to allow matching, and on
The second disadvantage is the potential of people opting not the problem of differences between consenters and non-consenters
to participate in studies, programs, or treatment when consent to resulting in bias.
examine claims data is required. In the fall of 2013, the Agency for We present a large number of comparisons in Table 1. Because
Community Living (ACL) began asking participants in evidence- of the exploratory nature of this study, we have not attempted to
based community programs funded by the agency to voluntarily adjust for multiple comparisons. Thus caution should be exercised
consent to having their ZIP Codes and birth dates matched with in drawing conclusions from any single statistically significant
Medicare data. While it is not known if people did not attend pro- result. Of more importance are the patterns in the results, specif-
grams because of this request, ACL did receive many complaints ically the tendency of consenters to be more involved with the
from sites and the consenting process was dropped when CMS medical system and to be more involved or engaged in the inter-
decided the data would not be needed. vention and larger National Study. Further study of the ethnic
In the study presented here, we estimate that at least 7% of those differences in consenting would be highly desirable.
who had completed 6-month questionnaires and were contacted
by phone with a request to consent, both refused to consent and CONCLUSION
asked to be dropped from the original National Study before com- Fewer than half the eligible participants consented to link their
pleting 12-month questionnaires (12 participants). In addition, name, gender, age, and ZIP Code to Medicare data. Those who

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 176 | 214
Ritter et al. Linking participants with Medicare data

consented were significantly different in several ways from those 8. Lorig K, Ritter PL, Ory MG, Whitelaw N. Effectiveness of the chronic disease
who chose not to consent or who did not respond. In particular, self-management program for with type 2 diabetes: a translation study. Diabetes
Educ (2013) 39(5):655–63. doi:10.1177/0145721713492567
consenters may have had more contact with the medical system
9. Kroenke K, Strine TW, Spritzer RL, Williams JB, Berry JT, Mokdad AH. The
and more illness. This suggests that data based only on those who PHQ-8 as a measure of current depression in the general population. J Affect
consent may be biased toward greater medical utilization and costs. Disord (2009) 114(1–3):163–73. doi:10.1016/j.jad.2008.06.026
The findings also suggest that asking participants to consent to 10. Idler EL, Angel RJ. Self-rated health and mortality in the NHANES-I epidemio-
match Medicare data may reduce participation in an intervention logic follow-up study. Am J Public Health (1990) 80:446–52. doi:10.2105/AJPH.
80.4.446
study. These findings have a potential to affect the use of data
11. Lorig K, Stewart A, Ritter P, González V, Laurent D, Lynch J. Outcome Measures
for policy decisions based on linking Medicare data with specific for Health Education and Other Health Care Interventions. Thousand Oaks CA:
interventions. Sage Publications (1996).
12. Morisky DE, Green LW, Levine DM. Concurrent and predictive validity of a
self-reported measure of medication adherence. Med Care (1986) 24:67–74.
ACKNOWLEDGMENTS doi:10.1097/00005650-198601000-00007
Support was provided by the Centers for Medicare and Medic- 13. Ritter PL, Stewart AL, Kaymaz H, Sobel DS, Bloch DA, Lorig KR. Self-reports
aid Services through contract # HHSM-500-2011-00088C. Nancy of health care utilization compared to provider records. J Clin Epidemiol (2001)
54:136–41. doi:10.1016/S0895-4356(00)00261-4
Whitelaw of the National Council on Aging was a principal investi- 14. Berwick DM, Nolan TW, Whittington J. The triple aim: care, health, and cost.
gator for the National Study. Mirna Sanchez of the Stanford Patient Health Aff (2008) 27:759–69. doi:10.1377/hlthaff.27.3.759
Education Research Center assisted in the management of that
study and the consenting process. Conflict of Interest Statement: The authors declare that the research was conducted
in the absence of any commercial or financial relationships that could be construed
as a potential conflict of interest.
REFERENCES
1. Brady TJ, Murphy L, O’Colmain BJ, Beauchesne D, Daniels B, Greenberg M, This paper is included in the Research Topic, “Evidence-Based Programming for Older
et al. Meta-analysis of health status, health behaviors, and health care utilization Adults.” This Research Topic received partial funding from multiple government and
outcomes of the chronic disease self-management program. Prev Chronic Dis private organizations/agencies; however, the views, findings, and conclusions in these
(2013) 10:120112. doi:10.5888/pcd10.120112 articles are those of the authors and do not necessarily represent the official position
2. Lorig KR, Sobel DS, Stewart AL, Brown BW, Bandura A, Ritter P, et al. Evidence of these organizations/agencies. All papers published in the Research Topic received
suggesting that a chronic disease self-management program can improve health peer review from members of the Frontiers in Public Health (Public Health Education
status while reducing hospitalization: a randomized trial. Med Care (1999) and Promotion section) panel of Review Editors. Because this Research Topic repre-
37(1):5–14. doi:10.1097/00005650-199901000-00003 sents work closely associated with a nationwide evidence-based movement in the US,
3. Ory MG, Ahn S, Jiang L, Lorig K, Ritter P, Laurent DD, et al. National study of many of the authors and/or Review Editors may have worked together previously in
chronic disease self-management: six-month outcome findings. J Aging Health some fashion. Review Editors were purposively selected based on their expertise with
(2013) 25:1258. doi:10.1177/0898264313502531 evaluation and/or evidence-based programming for older adults. Review Editors were
4. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, Whitelaw N, et al. Successes independent of named authors on any given article published in this volume.
of a national study of the chronic disease self-management program: meet-
ing the triple aim of health care reform. Med Care (2013) 51(11):992–8. Received: 16 June 2014; accepted: 19 September 2014; published online: 27 April 2015.
doi:10.1097/MLR.0b013e3182a95dd1 Citation: Ritter PL, Ory MG, Smith ML, Jiang L, Alonis A, Laurent DD and Lorig
5. Ahn S, Basu R, Smith ML, Jiang L, Lorig K, Whitelaw N, et al. The impact K (2015) Linking evidence-based program participant data with Medicare data: the
of chronic disease self-management programs: healthcare savings through a consenting process and correlates of retrospective participant consents. Front. Public
community-based intervention. BMC Public Health (2013) 13:1141. doi:10. Health 2:176. doi: 10.3389/fpubh.2014.00176
1186/1471-2458-13-1141 This article was submitted to Public Health Education and Promotion, a section of the
6. Lorden AL, Radcliff TA, Jiang L, Horel SA, Smith ML, Lorig K, et al. Lever- journal Frontiers in Public Health.
aging administrative data for program evaluations: a method for linking Copyright © 2015 Ritter, Ory, Smith, Jiang , Alonis, Laurent and Lorig . This is an
datasets without unique identifiers. Eval Health Prof (2014). doi:10.1177/ open-access article distributed under the terms of the Creative Commons Attribution
0163278714547568 License (CC BY). The use, distribution or reproduction in other forums is permitted,
7. National Council on Aging. Pilot Report of the Chronic Disease Self-Management provided the original author(s) or licensor are credited and that the original publica-
Program, Contract HHSM-500-2011-00088C. Washington, DC: Report submit- tion in this journal is cited, in accordance with accepted academic practice. No use,
ted to the Centers for Medicare and Medicare Services (CMS) (2013). distribution or reproduction is permitted which does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 176 | 215

 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00258

Fall prevention in community settings: results from

implementingTai Chi: Moving for Better Balance
in three states
Marcia G. Ory 1 *, Matthew Lee Smith 2 , Erin M. Parker 3 , Luohua Jiang 4 , Shuai Chen 5 , Ashley D. Wilson 1 ,
Judy A. Stevens 3 , Heidi Ehrenreich 3 and Robin Lee 3
1
Department of Health Promotion and Community Health Sciences, Texas A&M Health Science Center School of Public Health, College Station, TX, USA
2
Department of Health Promotion and Behavior, The University of Georgia College of Public Health, Athens, GA, USA
3
National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, Atlanta, GA, USA
4
Department of Epidemiology and Biostatistics, Texas A&M Health Science Center School of Public Health, College Station, TX, USA
5
Department of Statistics, Texas A&M University, College Station, TX, USA

Edited by: Tai Chi: Moving for Better Balance (TCMBB) is an evidence-based fall prevention exer-
Michal Grivna, United Arab Emirates
cise program being disseminated in selected communities through state injury prevention
University, United Arab Emirates
programs. This study: (1) describes the personal characteristics of TCMBB participants; (2)
Reviewed by:
Milka Dančević-Gojković, Public quantifies participants’ functional and self-reported health status at enrollment; and (3) mea-
Health Institute of Federation of sures changes in participants’ functional and self-reported health status post-intervention.
Bosnia and Herzegovina, Bosnia and There were 421 participants enrolled in 36 TCMBB programs delivered in Colorado, New
Herzegovina
York, and Oregon. Of the 209 participants who completed both baseline enrollment and
Pankaja Desai, University of Illinois at
Chicago, USA post-intervention surveys, the average age of participants was 75.3 (SD ± 8.2) years. Most
*Correspondence: participants were female (81.3%), non-Hispanic (96.1%), White (94.1%), and described
Marcia G. Ory , Department of Health themselves as in excellent or very good health (52.2%). Paired t -test and general esti-
Promotion and Community Health mating equation models assessed changes over the 3-month program period. Pre- and
Sciences, Texas A&M Health Science
post-assessment self-reported surveys and objective functional data [Timed Up and Go
Center, 1266, College Station,
TX 77843–1266, USA (TUG) test] were collected. On average, TUG test scores decreased (p < 0.001) for all par-
e-mail: [email protected] ticipants; however, the decrease was most noticeable among high-risk participants (mean
decreased from 18.5 to 15.7 s). The adjusted odds ratio of reporting feeling confident that
a participant could keep themselves from falling was five times greater after completing
the program. TCMBB, which addresses gait and balance problems, can be an effective
way to reduce falls among the older adult population. By helping older adults maintain
their functional abilities, TCMBB can help community-dwelling older adults continue to live
independently.
Keywords: Tai Chi: Moving for Better Balance, fall prevention, fall prevention program, community setting, older
adults

INTRODUCTION shown in a randomized controlled trial to be effective in reduc-

Tai Chi is a Chinese form of exercise that uses slow, flowing body ing falls (11, 12). TCMBB consists of eight forms that progress
movements. It had been practiced for centuries in Asia before being from easy to difficult to improve older adults’ postural stability,
introduced to the United States in the early twentieth century (1). balance, and coordination (13). Classes consist of 10–15 par-
The physical and mental health benefits of Tai Chi are well doc- ticipants led by a trained instructor. One-hour classes are held
umented (2–4), and in the 1990s, Tai Chi was rigorously tested twice a week for 12 weeks (24 total classes) (13). Feasibility test-
by the National Institute on Aging as a fall prevention interven- ing has demonstrated that this program is well accepted by older
tion (5, 6). A Cochrane review and meta-analysis concluded that adults and can be implemented with fidelity in community settings
Tai Chi reduced the risk of falling 28%, with greater effectiveness (14, 15).
among those with lower initial fall risk (7). Today Tai Chi is widely In 2011, the CDC launched a 5-year project to implement
recognized as an effective fall intervention (8–10). TCMBB in selected communities in Oregon, Colorado, and New
The Tai Chi: Moving for Better Balance (TCMBB) program York. This was part of a larger project to reduce falls and fall-related
is an evidence-based fall prevention exercise program that was injuries by engaging fall prevention coalitions, healthcare organi-
developed by researchers at the Oregon Research Institute with zations, and other partners to integrate clinical and evidence-based
partial funding from the Centers for Disease Control and Pre- community fall prevention programs in selected communities
vention (CDC). The original 26-week intervention used 24 Tai (16). TCMBB is intended for relatively healthy older adults with
Chi forms or sequences of controlled movements, and it was few functional limitations.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 258 | 216
Ory et al. Implementing Tai Chi: Moving for Better Balance

This study describes the results of implementing TCMBB dur- The Timed Up and Go (TUG) test was used to measure func-
ing the first 2 years of the project. The purposes of the study tional status at enrollment and completion. This test has been
were to: (1) describe the personal characteristics of TCMBB par- widely used to assess functional mobility and predict fall risk and
ticipants; (2) quantify participants’ functional and self-reported has been validated among community-dwelling older adults (18–
health status at enrollment; and (3) measure changes in partici- 20). The test measures the time in seconds required for participants
pants’ functional and self-reported health status after completing to “rise from a standard arm chair, walk at [their] typical or normal
the program. pace to a line on the floor 3 meters away, turn, return, and sit down
again” (21). Participants who completed the TUG in <12 seconds
MATERIALS AND METHODS were classified as low risk and those who took 12 or more seconds
TCMBB IMPLEMENTATION were classified as high risk (22).
The three states offered TCMBB in a variety of settings includ-
ing YMCAs, healthcare organizations, residential facilities, faith- STATISTICAL ANALYSES
based organizations, recreational facilities, and senior centers. Baseline characteristics (demographic characteristics, class atten-
State grantees hosted 30 TCMBB trainings from 2011 to 2013 dance, and TUG results) were compared for those who completed
at which Master Trainers from the Oregon Research Institute both the baseline and post-intervention questionnaires to those
trained 400 instructors. In addition, the YMCA of the USA (Y- who completed only the baseline questionnaire using chi-square
USA) engaged the Oregon Research Institute to train 10 YMCA tests to identify potential biases from loss to follow-up. Changes in
faculty trainers to be TCMBB instructors. TUG test times between baseline and post-intervention were com-
The target audience for TCMBB is community-dwelling older pared using two-tailed paired t -tests; results were examined for all
adults aged 60 and older who can walk easily with or without assis- participants combined and stratified by baseline risk level. Gen-
tive devices. In each state, participants were recruited by staff at eral estimating equations (GEE) models using a logit link function
member organizations, through family and friends, and through were used to compare differences in self-reported functional and
advertisements aimed at older adults. Methods of recruitment and health status at baseline and post-intervention; models were run
referral varied across states and were based on existing partner- using SAS version 9.3 GENMOD procedure (SAS Institute Inc.,
ships. For example, Colorado and New York were most likely to Cary, NC, USA) and adjusted for gender, age, race, and state. GEE
recruit at YMCAs whereas Oregon recruited through senior centers models are longitudinal data models that use all available data
and health care organizations since there were no YMCAs in their in model estimation (i.e., do not require paired data) and can
service delivery areas. As a program implemented through existing account for the correlation among repeated measures from the
traditional community settings, there were limited exclusionary same participant (23).
criteria and medical clearance was not required for participation. The Texas A&M University Institutional Review Board granted
While no age restrictions were placed on enrollment, our analy- approval to analyze data on program participants and outcomes
ses were restricted to people aged 60 years and older to reflect the collected using survey instruments and functional assessments.
study target population. RESULTS
PARTICIPANT CHARACTERISTICS AND COURSE ATTENDANCE
DATA COLLECTION Between September 1, 2011 and December 31, 2013, the three
Data for this project were collected from multiple sources. Atten- states offered 36 TCMBB programs and enrolled 537 people aged
dance was obtained from attendance logs collected at each class. 60 and older. Of these enrollees, baseline data were collected from
A 20-question self-administered survey was used to collect pre- 421 (78.4%); 20.2% of participants were in Oregon, 39.9% in
and post-TCMBB program data. The first was administered at Colorado, and 39.9% in New York (Table 1). Of the 421 partici-
the initial TCMBB class (enrollment or baseline survey) and the pants who provided a baseline questionnaire, 209 also completed
second at the final class (course completion or post-intervention a post-intervention questionnaire (Table 1).
survey). The surveys took approximately 15 minutes to com- The average age of participants was 75.3 (SD ± 8.2) years. Most
plete and assistance was provided to participants who needed participants were female, non-Hispanic, and White. About half of
help filling out the forms. Questions included socio-demographic the participants attended at least 70% of classes (17 out of 24),
characteristics (e.g., age, sex, race, and ethnicity), whether the with participants attending on average 13.6 (SD ± 8.0) of the 24
participant had been referred to the program by a health- possible classes. Only 16 participants (8.5%) reported they were
care provider, self-reported health status (excellent, very good, referred to TCMBB by a healthcare provider.
good, fair, or poor), satisfaction with their current activity lev- The 212 participants who “dropped out” or were lost to follow-
els (very, mostly, somewhat, or not at all satisfied), and confi- up were not significantly different from those who completed the
dence in their ability to keep themselves from falling (four-point program in terms of gender, race, ethnicity, self-reported health
scale ranging from strongly agree to strongly disagree). Self- status, or provider referral to class. However, dropouts were sig-
reported functional ability was assessed by the reported level nificantly older (average age 76.1 vs. 74.1) and more likely to have
of difficulty in performing various activities (e.g., climbing one been classified as high risk based their TUG time at baseline.
flight of stairs) on a four-point scale ranging from no difficulty
(scored 1) to unable to do (scored 4) (17). Class completion was PARTICIPANT FUNCTIONAL PERFORMANCE
defined as attending at least 70% of the classes (i.e., 17 out of Of 421 participants with baseline data, 199 (47.3%) completed
24 classes). the TUG test at both baseline and post-intervention (Table 2).

www.frontiersin.org April 2015 | Volume 2 | Article 258 | 217

Ory et al. Implementing Tai Chi: Moving for Better Balance

Table 1 | Characteristics of Tai Chi: Moving for Better Balance (TCMBB) participants.

All enrolled Participants who Participants who X2 P -value

participantsa completed both the completed only the
baseline enrollment baseline enrollment
and post-intervention survey
surveys

N = 421 N = 209 N = 212

n (%) n (%) n (%)

Location 3.24 0.197

Oregon 85 (20.2) 35 (16.8) 50 (23.6)
Colorado 168 (39.9) 85 (40.7) 83 (39.2)
New York 168 (39.9) 89 (42.6) 79 (37.3)
Age group 8.96 0.011
60–69 115 (27.3) 61 (29.2) 54 (25.5)
70–79 177 (42.0) 98 (46.9) 79 (37.3)
80+ 129 (30.6) 50 (23.9) 79 (37.3)
Gender 0.22 0.639
Female 335 (80.3) 169 (81.3) 166 (79.4)
Male 82 (19.7) 39 (18.8) 43 (20.6)
Missing 4 1 3
Race 0.11 0.742
White 388 (93.7) 192 (94.1) 196 (93.3)
Non-White 26 (6.3) 12 (5.9) 14 (6.7)
Missing 7 5 2
Ethnicity (Hispanic/Latino) 0.06 0.807
Yes 15 (3.7) 8 (3.9) 7 (3.4)
No 395 (96.3) 198 (96.1) 197 (96.6)
Missing 11 3 8
Self-reported health status 3.86 0.145
Excellent/very good 211 (50.7) 108 (52.2) 103 (49.3)
Good 165 (39.7) 85 (41.1) 80 (38.3)
Fair/poor 40 (9.6) 14 (6.8) 26 (12.4)
Missing 5 2 3
Referred by healthcare provider
Yes 35 (8.5) 18 (8.7) 17 (8.3) 0.03 0.872
No 378 (91.5) 189 (91.3) 189 (91.7)
Missing 8 2 6
Timed up and go (TUG) time at enrollment 5.65 0.017
Low risk (baseline TUG < 12 s) 279 (71.7) 154 (77.0) 125 (66.1)
High risk (baseline TUG ≥ 12 s) 110 (28.3) 46 (23.0) 64 (33.9)
Missing 32 9 23
Participants who completed 70%+ classes 209 (49.6) 163 (78.0) 46 (21.7) 133.41 <0.001

a
Enrolled participants include all persons 60 years and older who filled out the baseline enrollment survey on the first day of the program (421/537 participants).
Individual survey questions may have had missing data.

Of these, 45 (22.6%) were categorized as high risk. After com- SELF-REPORTED OUTCOME IMPROVEMENTS
pleting TCMBB, the proportion of participants categorized as Table 3 compares self-reported outcome measures at baseline and
high risk decreased significantly to 14% (n = 28; data not shown). post-intervention. Results are presented as percentages and as
On average, TUG test scores decreased significantly for all par- odds ratios adjusted for gender, age, race, and state. Significant
ticipants but the change was most evident among high risk improvements from baseline to post-intervention were observed
participants where the average TUG time decreased from 18.5 for all outcomes except self-reported difficulty in walking across
to 15.7 seconds. the room.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 258 | 218
Ory et al. Implementing Tai Chi: Moving for Better Balance

Table 2 | Changes in Tai Chi: Moving for Better Balance (TCMBB) participants’ timed up and go (TUG) times from baseline enrollment to
post-interventiona .

Changes in TUG times (in seconds) TUG at baseline TUG at post- Change in TUG from baseline
intervention to post-interventionb

N Mean (±SD) N Mean (±SD) N Mean (±SD) P -value

TUG times for all participants 199 11.2 (±6.7) 199 9.9 (±6.0) 199 −1.3 (±2.7) <0.001
Low risk (enrollment TUG time < 12 s) 154 9.1 (±1.4) 154 8.2 (±1.6) 154 −0.8 (±1.3) <0.001
High risk (enrollment TUG time ≥ 12 s) 45 18.5 (±11.2) 45 15.7 (±10.5) 45 −2.7 (±4.9) 0.001

SD, standard deviation.

a
While 389 participants completed the TUG at enrollment, this table highlights the 199 participants who completed the TUG at both baseline and post-intervention.
b
Paired t-tests with an alpha of 0.05 were used to compare changes in participants’ TUG times between baseline and post-intervention. A reduction in time indicates
a positive functional improvement.

Table 3 | Changes in Tai Chi: Moving for Better Balance (TCMBB) participants’ self-reported health and functional outcomes from baseline to
post-interventiona .

Health outcome Baseline Post-intervention Adjusted change from baseline

(N = 209)b (N = 209)b to post-interventionc

N (%) N (%) Odds ratios from P -value

logistic models

Health status, satisfaction, and confidence

Excellent or very good health status 108 (52.2) 123 (58.9) 1.35 (1.03, 1.77) 0.031
Very/mostly satisfied with physical activity levels 126 (60.9) 160 (76.9) 2.21 (1.60, 3.05) <0.001
Feel confident not falling (strongly agree or agree) 149 (74.9) 196 (93.8) 6.16 (3.48, 10.89) <0.001
Self-reported functional status
No difficulty in walking across room 178 (86.4) 183 (88.4) 1.30 (0.83, 2.02) 0.249
No difficulty in walking one block 149 (77.2) 166 (83.0) 1.60 (1.19, 2.17) 0.002
No difficulty in stooping, crouching, and kneeling 71 (34.8) 85 (41.3) 1.32 (1.04, 1.68) 0.023
No difficulty in getting out of a straight back chair 141 (73.4) 163 (81.1) 1.67 (1.14, 2.44) 0.008
No difficulty in climbing one flight of stairs 133 (64.6) 144 (72.0) 1.42 (1.03, 1.68) 0.034

SD, standard deviation.

a
Data are reported for n = 209 participants who completed both the baseline and post-intervention surveys.
b
The sample size is slightly smaller than 209 for some health outcomes due to missing data. The amount of missing data ranges from 0 to 8% for different outcomes.
c
Adjusted odds ratios from GEE logistic regression modeling the probability of response = 1 at an alpha of 0.05. All models account for repeated measures from the
same participant and are adjusted for gender, age, race, and state. An odds ratio >1 represents a positive improvement in self-reported health.

The GEE model results showed that the adjusted odds ratio DISCUSSION
(aOR) of reporting excellent or very good health status increased This study examined 2 years of evaluation data collected from
by 35% (aOR = 1.35, 95% CI 1.03–1.77). The odds of being very older adults age 60+ who participated in TCMBB programs
or mostly satisfied with physical activity levels also increased sig- offered in selected communities across three states. Comparing
nificantly (aOR = 2.21, 95% CI 1.60–3.05). The odds of feeling data collected at enrollment and course completion, TCMBB was
confident that a participant could keep themselves from falling associated with significant improvements in self-reported health
was five times greater after completing TCMBB (aOR = 6.16 95% status, satisfaction with physical activity levels, fall-related confi-
CI 3.48–10.89). dence, ability to perform basic functional tasks (e.g., walking one
Among the five items assessing functional status, the aORs for block, climbing a flight of stairs), and in the TUG test. Similar
participants who reported “no difficulty” significantly increased positive results have been seen in earlier studies of Tai Chi (14,
for walking one block (aOR = 1.60, 95% CI 1.19–2.17); stoop- 24), and provide additional evidence that Tai Chi is a useful fall
ing, crouching, kneeling (aOR = 1.32, 95% CI 1.04–1.68); getting prevention program for older adults.
out of a straight back chair (aOR = 1.67, 95% CI 1.14–2.44); Recruitment of participants is a concern for most fall pre-
and climbing one flight of stairs (aOR = 1.42, 95% CI 1.03– vention programs. While the distribution of TCMBB partici-
1.68). About 86% of participants reported no difficulty walking pants’ race and ethnicity was similar to the populations from
across the room at baseline, and this proportion did not increase which they were recruited, the percentage of male participants
significantly at post-intervention. was low. Retaining TCMBB participants was also challenging.

www.frontiersin.org April 2015 | Volume 2 | Article 258 | 219

Ory et al. Implementing Tai Chi: Moving for Better Balance

Participants attended on average 57% of the 24 classes. While post-intervention assessments, we can be confident we are com-
it was not possible in this study to monitor falls, those who paring the same population before and after the intervention. We
did not attend regularly may not have received an adequate did not take into account differences in total attendance among
intervention dose for reducing their fall risk. The reasons for people who provided baseline and post-intervention assessments,
low attendance are unknown. However, anecdotal reports from so the effectiveness of the full intervention may be underestimated.
the state health departments implementing TCMBB suggest that Second, the program was delivered in a multitude of settings,
some older adults may have considered the Tai Chi program a and outcomes may have been influenced by variability in instruc-
“drop-in” activity instead of an ongoing program. Those who tor and site. Although all instructors were certified trained instruc-
did not complete the course were somewhat older and took tors, we recommend more attention be given to treatment fidelity
longer to complete the TUG at enrollment, which suggests that monitoring in future research and practice. The CDC Guide for
health issues may have contributed to their not finishing the Program Implementation (13) has examples of a class observa-
program. tion form for monitoring instructor adherence to core program
Barriers to the success of TCMBB, as for other community- elements.
based fall prevention programs, include maintaining regular atten- Third, in order to limit the reporting burden on the program
dance and encouraging participants to continue activities after the delivery personnel, we used a limited number of self-reported
program ends. Although the participants in this study demon- outcomes and one timed functional assessment (i.e., the TUG).
strated positive outcomes, one 12-week program is unlikely to Although there was training provided for conducting the TUG,
provide long-term benefits without booster classes. Tai Chi, like including available step-by-step online videos, this training was
other strength and balance exercises, is most effective when it is limited. Therefore, results may not be comparable to standardized
practiced for 50 hours or more (11). Therefore, older adults would TUG tests administered by trained professionals. Finally, while
benefit from having an ongoing Tai Chi program in their commu- TCMBB participants reported improved functional status and
nity, if they attended regularly. Some participating sites are now demonstrated better TUG scores, we do not know if this led to
offering an introductory 12-week TCMBB followed by an ongoing a reduction in falls, since falls were not monitored during or after
program. the program.
Another challenge has been the limited availability of commu-
nity Tai Chi classes. State health departments have been able to CONCLUSION
address this by developing public–private partnerships with orga- Tai Chi: Moving for Better Balance, which addresses gait and bal-
nizations that have existing infrastructure to offer classes to older ance problems, can be an effective way to reduce falls among the
adults. For example, the Y-USA now endorses a modified version older adult population. Various forms of Tai Chi have been shown
of TCMBB called Y-Moving for Better Balance (Y-MFBB) that is to be most appropriate for younger and healthier older adults who
being offered in local YMCAs (25). State health departments are are at relatively low risk of falling. By helping older adults maintain
also beginning to implement and support other Tai Chi programs their functional abilities, TCMBB can help community-dwelling
(e.g., Tai Chi for Arthritis) that have been shown to be effective for older adults continue to live independently.
fall prevention (26). In this study, TCMBB participants reported positive effects on
Ideally all older adults would have access to a wide range of their functional and health status. However, the high dropout
evidence-based fall prevention programs that could meet their rates among program participants highlight a major challenge
varied needs. Thus, in the larger fall prevention project, TCMBB to implementing effective community-based fall prevention pro-
was offered along with Stepping On (27) and Otago (28), which grams. Community-based programs are a promising approach
are designed for older adults with some functional limitations who for older adult fall prevention, but there are ongoing challenges
are at moderate and high fall risk, respectively. As the availability to ensuring that high quality programs are available for – and
of Tai Chi and other fall prevention programs expands, it will be attended by – older adults who can benefit from such programs.
important to ensure that fidelity to the key elements of the original
interventions is maintained so that the programs remain effective ACKNOWLEDGMENTS
in preventing falls. We thank the Colorado, New York, and Oregon State Injury
Prevention Centers for coordinating Tai Chi: Moving for Better
LIMITATIONS Balance, a Community-Based Older Adult Fall Prevention Pro-
This study has number of limitations. First, sampling and dropout gram and collecting the study data and Margaret Kaniewski with
issues limit the generalizability of the results. Participants were the Centers for Disease Control and Prevention for her program-
self-selected from participating communities and may not be rep- matic support and critical review of the manuscript. Funding
resentative of the older adult population either in those communi- Source: This research was supported in part by the Centers for Dis-
ties or in the participating states. Program effectiveness was based ease Control and Prevention, Prevention Research Centers Special
on comparing assessments from participants who attended both Initiative Project funding for the State Fall Prevention Program
the first and last class. These participants were slightly younger under 5U48DP001924, and the Research Participation Program
and had fewer functional limitations, as measured by better TUG at the Centers for Disease Control and Prevention administered
times at baseline, compared to participants who were not avail- by the Oak Ridge Institute for Science and Education through an
able for the post-intervention assessment. However, because the interagency agreement between the U.S. Department of Energy
results are for those who provided both baseline enrollment and and CDC.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 258 | 220
Ory et al. Implementing Tai Chi: Moving for Better Balance

REFERENCES 20. Shumway-Cook A, Brauer S, Woollacott M. Predicting the probability for falls
1. National Center for Complementary and Alternative Medicine (NCCAM). Tai in community-dwelling older adults using the timed up & go test. Phys Ther
Chi: An Introduction. National Institute of Health (2010) [cited 2014 Aug 11]. (2000) 80(9):896–903.
Available from: http://nccam.nih.gov/health/taichi/introduction.htm#overview 21. Podsiadlo D, Richardson S. The timed “up & go”: a test of basic functional
2. Leung DP, Chan CK, Tsang HW, Tsang WW, Jones AY. Tai chi as an intervention mobility for frail elderly persons. J Am Geriatr Soc (1991) 39(2):142–8.
to improve balance and reduce falls in older adults: a systematic and meta- 22. Bischoff HA, Stahelin HB, Monsch AU, Iversen MD, Weyh A, von Dechend M,
analytical review. Altern Ther Health Med (2011) 17(1):40–8. et al. Identifying a cut-off point for normal mobility: a comparison of the timed
3. Lee MS, Ernst E. Systematic reviews of t’ai chi: an overview. Br J Sports Med ‘up and go’ test in community-dwelling and institutionalised elderly women.
(2012) 46(10):713–8. doi:10.1136/bjsm.2010.080622 Age Ageing (2003) 32(3):315–20. doi:10.1093/ageing/32.3.315
4. Kuramoto AM. Therapeutic benefits of Tai Chi exercise: research review. WMJ 23. Hardin JW. Generalized Estimating Equations (GEE). Encyclopedia of Sta-
(2006) 105(7):42–6. tistics in Behavioral Science. Hoboken, NJ: John Wiley & Sons, Ltd. (2005).
5. Ory MG, Schechtman KB, Miller JP, Hadley EC, Fiatarone MA, Province MA, doi:10.1002/0470013192.bsa250
et al. Frailty and injuries in later life: the FICSIT trials. J Am Geriatr Soc (1993) 24. Rogers CE, Larkey LK, Keller C. A review of clinical trials of tai chi and
41(3):283–96. qigong in older adults. West J Nurs Res (2009) 31(2):245–79. doi:10.1177/
6. Province MA, Hadley EC, Hornbrook MC, Lipsitz LA, Miller JP, Mulrow CD, 0193945908327529
et al. The effects of exercise on falls in elderly patients. A preplanned meta- 25. Ehrenreich H, Pike M, Hohman K, Kaniewski M, Longjohn M, Myers G, et al.
analysis of the FICSIT trials. Frailty and injuries: cooperative studies of interven- CDC and YMCA: a promising partnership for delivering fall prevention pro-
tion techniques. JAMA (1995) 273(17):1341–7. doi:10.1001/jama.273.17.1341 gramming. Front Public Health (2015) 2:235. doi:10.3389/fpubh.2014.00235
7. Cameron ID, Murray GR, Gillespie LD, Robertson MC, Hill KD, Cumming 26. Stevens JA, Voukelatos A, Ehrenreich H. Preventing falls with Tai Ji Quan: a
RG, et al. Interventions for preventing falls in older people in nursing care public health perspective. J Sport Health Sci (2014) 3(1):21–6. doi:10.1016/j.
facilities and hospitals. Cochrane Database Syst Rev (2010) 20(1):CD005465. jshs.2013.10.002
doi:10.1002/14651858.CD005465.pub2 27. Ory MG, Smith ML, Jiang L, Lee R, Chen S, Wilson AD, et al. Fall prevention
8. Michael YL, Whitlock EP, Lin JS, Fu R, O’Connor EA, Gold R, et al. Primary in community settings: results from implementing Stepping On in three states.
care-relevant interventions to prevent falling in older adults: a systematic evi- Front Public Health (2015) 2:232. doi:10.3389/fpubh.2014.00232
dence review for the U.S. preventive services task force. Ann Intern Med (2010) 28. Shubert TE, Smith ML, Ory MG, Clarke CB, Bomberger SA, Roberts E, et al.
153(12):815–25. doi:10.7326/0003-4819-153-12-201012210-00008 Translation of the Otago Exercise Program for adoption and implementation
9. Stevens JA. A CDC Compendium of Effective Fall Interventions: What Works for in the United States. Front Public Health (2015) 2:152. doi:10.3389/fpubh.2014.
Community-Dwelling Older Adults. Atlanta, GA: Centers for Disease Control 00152
and Prevention. National Center for Injury Prevention and Control. Division of
Unintentional Injury Prevention (2010).
Conflict of Interest Statement: The findings and conclusions in this report are those
10. National Council on Aging. Falls Free: Promoting a National Falls Preventions
of the authors and do not necessarily represent the official position of the Centers
Action Plan (2005) [cited 2014 Aug 11]:[1-51 p.]. Available from: http://www.
for Disease Control and Prevention.
ncoa.org/improve-health/center-for-healthy-aging/content-library/FallsFree_
NationalActionPlan_Final.pdf
This paper is included in the Research Topic, “Evidence-Based Programming for Older
11. Sherrington C, Tiedemann A, Fairhall N, Close JCT, Lord S. Exercise to prevent
Adults.” This Research Topic received partial funding from multiple government and
falls in older adults: an updated meta-analysis and best practice recommenda-
private organizations/agencies; however, the views, findings, and conclusions in these
tions. N S W Public Health Bull (2011) 22(3–4):78–83. doi:10.1071/NB10056
articles are those of the authors and do not necessarily represent the official position
12. Li F, Harmer P, McAuley E, Duncan TE, Duncan SC, Chaumeton N, et al. An
of these organizations/agencies. All papers published in the Research Topic received
evaluation of the effects of Tai Chi exercise on physical function among older
peer review from members of the Frontiers in Public Health (Public Health Educa-
persons: a randomized controlled trial. Ann Behav Med (2001) 23(2):139–46.
tion and Promotion section) panel of Review Editors. Because this Research Topic
doi:10.1207/S15324796ABM2302_9
represents work closely associated with a nationwide evidence-based movement in
13. National Center for Injury Prevention and Control. Tai chi: Moving for Better
the US, many of the authors and/or Review Editors may have worked together pre-
Balance A guide for program Implementation (2011) [cited 2014 Aug 6]:[1-146
viously in some fashion. Review Editors were purposively selected based on their
p.]. Available from: http://api.ning.com/files/lpWX79eu*NgqhrxYqum3lrPOm
expertise with evaluation and/or evidence-based programming for older adults. Review
6Dp4GRQCK9J1qwtJylSBPaJ2W38NVNJNcBqszLkRplWB9ee-lwV4PIN-hUo
Editors were independent of named authors on any given article published in this
Pujtn*zU9vRw/TCManual_Compiled__v19.pdf
volume.
14. Li F, Harmer P, Glasgow R, Mack KA, Sleet D, Fisher KJ, et al. Translation of an
effective tai chi intervention into a community-based falls-prevention program.
Am J Public Health (2008) 98(7):1195–8. doi:10.2105/AJPH.2007.120402 Received: 27 August 2014; paper pending published: 19 October 2014; accepted: 10
15. Li F, Harmer P, Mack KA, Sleet D, Fisher KJ, Kohn MA, et al. Tai Chi: moving for November 2014; published online: 27 April 2015.
better balance – development of a community-based falls prevention program. Citation: Ory MG, Smith ML, Parker EM, Jiang L, Chen S, Wilson AD, Stevens JA,
J Phys Act Health (2008) 5(3):445–55. Ehrenreich H and Lee R (2015) Fall prevention in community settings: results from
16. Kaniewski M, Stevens JA, Parker EM, Lee R. An introduction to the Centers for implementing Tai Chi: Moving for Better Balance in three states. Front. Public Health
Disease Control and Prevention’s efforts to prevent older adult falls. Front Public 2:258. doi: 10.3389/fpubh.2014.00258
Health (2015) 2:119. doi:10.3389/fpubh.2014.00119 This article was submitted to Public Health Education and Promotion, a section of the
17. Wallace RB, Herzog AR. Overview of the health measures in the health and journal Frontiers in Public Health.
retirement study. J Hum Resour (1995) 30:S84–107. doi:10.2307/146279 Copyright © 2015 Ory, Smith, Parker, Jiang , Chen, Wilson, Stevens, Ehrenreich and
18. Bohannon RW. Reference values for the timed up and go test: a descriptive Lee. This is an open-access article distributed under the terms of the Creative Commons
meta-analysis. J Geriatr Phys Ther (2006) 29(2):64–8. doi:10.1519/00139143- Attribution License (CC BY). The use, distribution or reproduction in other forums is
200608000-00004 permitted, provided the original author(s) or licensor are credited and that the origi-
19. Hutton I, Gamble G, McLean G, Butcher H, Gow P, Dalbeth N. Obstacles to nal publication in this journal is cited, in accordance with accepted academic practice.
action in arthritis: a community case-control study. Int J Rheum Dis (2009) No use, distribution or reproduction is permitted which does not comply with these
12(2):107–17. doi:10.1111/j.1756-185X.2009.01392.x terms.

www.frontiersin.org April 2015 | Volume 2 | Article 258 | 221

 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00232

Fall prevention in community settings: results from

implementing Stepping On in three states
Marcia G. Ory 1 *, Matthew Lee Smith 2 , Luohua Jiang 3,5 , Robin Lee 4 , Shuai Chen 5 , Ashley D. Wilson 1 ,
Judy A. Stevens 4 and Erin M. Parker 4
1
Department of Health Promotion and Community Health Sciences, Texas A&M Health Science Center, College Station, TX, USA
2
Department of Health Promotion and Behavior, The University of Georgia College of Public Health, Athens, GA, USA
3
Department of Epidemiology, University of California, Irvine, CA, USA
4
National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, Atlanta, GA, USA
5
Department of Statistics, Texas A&M University, College Station, TX, USA

Edited by: Stepping On is a community-based intervention that has been shown in a randomized
Michal Grivna, United Arab Emirates
controlled trial to reduce fall risk. The Wisconsin Institute for Healthy Aging adapted Step-
University, United Arab Emirates
ping On for use in the United States and developed a training infrastructure to enable
Reviewed by:
Wilma Alvarado-Little, AlvaradoLittle dissemination. The purpose of this study is to: (1) describe the personal characteristics
Consulting LLC, USA of Stepping On participants; (2) quantify participants’ functional and self-reported health
Katherine Henrietta Leith, University status at enrollment, and (3) measure changes in participants’ functional and self-reported
of South Carolina, USA
health status after completing the program. Both survey and observed functional status
*Correspondence:
[timed up and go (TUG) test] data were collected between September 2011 and December
Marcia G. Ory , Department of Health
Promotion and Community Health 2013 for 366 participants enrolled in 32 Stepping On programs delivered in Colorado, New
Sciences, Texas A&M Health Science York, and Oregon. Paired t -tests and general estimating equations models adjusted for
Center, 1266, College Station, TX socio-demographic factors were performed to assess changes over the program period.
77843–1266, USA
Among the 266 participants with pre–post survey data, the average participant age was
e-mail: [email protected]
78.7 (SD ± 8.0) years. Most participants were female (83.4%), white (96.9%), and in good
health (49.4%). The TUG test scores decreased significantly (p < 0.001) for all 254 partic-
ipants with pre–post data. The change was most noticeable among high risk participants
where TUG time decreased from 17.6 to 14.4 s. The adjusted odds ratio of feeling confident
about keeping from falling was more than three times greater after completing Stepping
On. Further, the adjusted odds ratios of reporting “no difficulty” for getting out of a straight
back chair increased by 89%. Intended for older adults who have fallen in the past or are
afraid of falling, Stepping On has the potential to reduce the frequency and burden of older
adult falls.
Keywords: fall prevention, evidence-based program, Stepping On, older adult

INTRODUCTION showed a positive return on investment of 59% (J. Stevens, CDC,

Although older adults fall more frequently than younger people, personal communication. 8/1/2014).
falls are not a normal part of aging (1). Over the past three decades, As described in the WIHA Implementation Manual (6), the
researchers have identified the major modifiable fall risk factors as program is delivered by a trained leader and a peer leader, who
well as effective fall interventions (2–4). Some interventions shown apply adult education and social learning principles to teach older
to be effective in randomized control trials have been translated adults about fall risk factors and strategies to reduce their fall risk.
into programs and implemented in community settings. One such The traditional program consists of a group of 10–14 participants
program is Stepping On, which was developed in Australia (5) and attending a 2 hour session held once a week for seven consecutive
later adapted for use in the United States by the Wisconsin Insti- weeks. Content is provided by the program leaders and by invited
tute for Healthy Aging (WIHA). The WIHA now provides training health professional “guest experts.” During the program, older
for Stepping On leaders as well as an implementation manual and adults learn how to improve their balance and strength, increase
evaluation plan (6, 7). their safety at home and in the community, and the importance of
Stepping On is a group program proven to reduce falls and vision assessment and medication reviews.
build confidence in ambulatory older adults who have fallen pre- In 2011, the Centers for Disease Control and Prevention (CDC)
viously or are afraid of falling (8). A randomized trial of Stepping launched a 5-year project funding State Departments of Health to
On found that participants’ risk of falling was approximately 30% implement Stepping On in selected communities in Oregon, Col-
lower than those who did not receive the intervention (5). Stevens orado, and New York. This was part of a larger project in which the
(2014) noted that a recent analysis also found that Stepping On CDC funded these states to reduce falls and fall-related injuries by

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 232 | 222
Ory et al. Implementing Stepping On

engaging fall prevention coalitions, healthcare organizations, and assistance was provided to participants who needed help filling
other partners to implement evidence-based fall prevention pro- out the forms. Survey questions included participant characteris-
grams in clinical and community settings. Stepping On is intended tics (e.g., age group, gender, race, ethnicity), general health status
for older adults with moderate fall risk, such as an older adult who (excellent, very good, good, fair, and poor), and whether the par-
fell in the past year or is afraid of falling. Additional information ticipant had been referred to the program by a healthcare provider.
about CDC’s fall prevention initiative can be found elsewhere (9). Also measured were satisfaction with their current physical activity
This manuscript describes the results of implementing Stepping level (very, mostly, somewhat, or not at all satisfied) and confidence
On during the first 2 years of the project. The purposes were to: in their ability to keep themselves from falling (five-point scale
(1) describe the personal characteristics and session attendance ranging from strongly agree to strongly disagree). Self-reported
of Stepping On participants; (2) quantify participants’ functional functional ability was assessed by the reported level of difficulty
and self-reported health status at enrollment, and (3) measure in performing various activities (e.g., climbing one flight of stairs)
changes in participants’ functional and self-reported health status on a four-point scale ranging from (1) no difficulty to (4) unable
after completing the program. to do (10).
The Timed Up and Go (TUG) test was used to measure func-
MATERIALS AND METHODS tional status at the first and last Stepping On sessions. This test has
PROGRAM PLANNING AND PARTICIPANT RECRUITMENT been widely used to assess functional mobility and predict fall risk
The WIHA offered training for master trainers who, in turn, (11, 12) and has been validated among community-dwelling older
trained local group facilitators. State program leads (i.e., desig- adults (13). The test measures the time in seconds for a participant
nated contacts at the State Departments of Health) recommended to “stand up from a standard arm chair, walk at [his or her] typical
facilitators who were part of local public health or aging ser- or normal pace to a line on the floor 3 m away, turn, return, and sit
vices delivery systems. Following the Implementation Guide (6), down again” (14). Participants who completed the TUG in <12 s
state program leads and facilitators worked together to identify were classified as having low fall risk and those who took 12 or
appropriate sites for Stepping On programs. more seconds were classified as high risk (15).
Program participants were recruited through a variety of chan- The Texas A&M University Institutional Review Board granted
nels, including distributing flyers, conducting informational pre- approval to analyze secondary data on program participants
sentations, making personal contact in places where older adults and outcomes collected using survey instruments and functional
congregated such as senior centers, recreation centers, or senior assessments.
housing or retirement homes, as well as through contacts with
their health care providers and television, newspaper, and radio STATISTICAL ANALYSES
advertisements. To identify potential biases from loss to follow-up, we used the
Stepping On staff used a standardized admission form and chi-square test to compare participant characteristics, number of
screening questions to identify appropriate participants. To be sessions attended, and TUG results from participants who com-
eligible, a participant needed to be 60 years of age or older, live pleted both the baseline and post-intervention surveys to those
independently in the community, and be able to walk without the who only completed the baseline survey (were lost to follow-up).
help of another person or with an assistive device (e.g., walker, Two-tailed paired t -tests were used to compare participant’s TUG
scooter). Although some information about chronic illnesses was results at the start and end of the program. General estimating
obtained during the screening process, information on the num- equation (GEE) models using a logit link function were used to
ber and type of chronic conditions was not systematically collected compare self-reported health status, satisfaction with activity lev-
as part of the evaluation survey. els, confidence in not falling, and self-reported functional status
In preparation for program delivery, each state conducted indicators at the start and end of the program. GEE models are
training sessions for Stepping On program leaders. The program longitudinal data models that use all available data in model esti-
was delivered in multiple settings, including healthcare organi- mation (i.e., do not require paired data) and can account for the
zations, senior housing or assisted living facilities, faith-based correlation among repeated measures from the same participant.
organizations, recreational facilities, and senior centers. Addi- Each GEE model controlled for age group, gender, race, and pro-
tional information about program preparation, implementation, gram location. All models were run using SAS version 9.3 (SAS
and evaluation can be found in the Stepping On Implementation Institute Inc., Cary, NC, USA).
Guide (6).
RESULTS
DATA COLLECTION PROGRAM IMPLEMENTATION AND PARTICIPANT CHARACTERISTICS
Data were collected from multiple sources. Attendance was Between September 1, 2011 and December 31, 2013, the three
recorded at each session and these records were used to describe states hosted four Stepping On training sessions. There were
participant retention over the 7 week program. Program comple- 64 leaders trained and 32, 7-week Stepping On programs deliv-
tion was defined as attending five of the seven sessions. A 20- ered. Four hundred nineteen participants aged 60 years and older
question self-administered questionnaire was used to collect data enrolled and 336 participants (80.2%) completed the enroll-
at the initial Stepping On session (enrollment or baseline survey) ment or baseline survey. Of these, 274 (81.5%) participants
and at the last session (program completion or post-intervention attended five or more sessions and 138 (41.1%) attended all seven
survey). Each survey took about 15 minutes to complete and sessions.

www.frontiersin.org April 2015 | Volume 2 | Article 232 | 223

Ory et al. Implementing Stepping On

Table 1 | Characteristics of Stepping On participants.

All enrolled Participants who Participants who

participantsa completed both the completed only the
baseline enrollment baseline enrollment
and post-intervention survey
surveys

N = 336 n = 266 n = 70

N (%) N (%) N (%) X2 p

Program location 0.90 0.639

Oregon 60 (17.9) 45 (16.9) 15 (21.4)
Colorado 91 (27.1) 74 (27.8) 17 (24.3)
New York 185 (55.1) 147 (55.3) 38 (54.3)
Age group 0.58 0.749
60–69 53 (15.8) 44 (16.5) 9 (12.9)
70–79 119 (35.4) 93 (35.0) 26 (37.1)
80+ 164 (48.8) 129 (48.5) 35 (50.0)

Gender 0.01 0.914

Female 279 (83.3) 221 (83.4) 58 (82.9)
Male 56 (16.7) 44 (16.6) 12 (17.1)
Missing 1 1 0

Race 2.61 0.106

White 316 (96.0) 253 (96.9) 63 (92.7)
Non-white 13 (4.0) 8 (3.1) 5 (7.4)
Missing 7 5 2

Ethnicity (Hispanic/Latino) 2.07 0.151

Hispanic 7 (2.1) 4 (1.5) 3 (4.4)
Non-Hispanic 322 (97.9) 256 (98.5) 66 (95.7)
Missing 7 6 1

General health status 3.09 0.214

Excellent or Very Good 114 (34.1) 96 (36.2) 18 (26.1)
Good 168 (50.3) 131 (49.4) 37 (53.6)
Fair/Poor 52 (15.6) 38 (14.3) 14 (20.3)
Missing 2 1 1

Referred to program by healthcare provider 22 (6.7) 19 (7.2) 3 (4.6) 0.57 0.452

Missing 8 3 5

Timed up and go (tug) time at enrollment 0.78 0.378

Low risk (enrollment TUG <12 s) 165 (50.3%) 135 (51.5%) 30 (45.5%)
High risk (enrollment TUG ≥12 s) 163 (49.7%) 127 (48.5%) 36 (54.6%)
Missing 8 4 4
Participants who completed 70%+ of sessions 274 (81.6%) 252 (94.7%) 22 (31.4%) 147.60 <0.001

a
Enrolled participants include all persons 60 years and older who filled out the baseline survey on the first day of the program (336/419 Stepping On participants).
Individual survey questions may have had missing data.

As indicated in Table 1, of the 336 participants who completed good (50.3%) or excellent to very good health (34.1%). Only 22
the baseline survey, 60 (17.9%) attended programs in Oregon, 91 (6.7%) participants were referred to Stepping On by a healthcare
(27.1%) in Colorado, and 185 (55.1%) in New York (Table 1). provider.
The age distribution was similar among participants in each state. There were 266 (63.5%) participants who completed both the
The mean age was 78.7 (SD ± 8.0) years. Overall, the major- baseline and post-intervention surveys; 70 completed only the
ity of people who enrolled were female (83.3%), white (96.0%), baseline survey and were considered drop outs. Among the 266
and non-Hispanic (97.9%). The majority of participants reported participants with pre-post survey data, the average participant age

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 232 | 224
Ory et al. Implementing Stepping On

was 78.7 (SD ± 8.0) years. Most participants were female (83.4%), SELF-REPORTED HEALTH AND FUNCTIONAL OUTCOMES
white (96.9%), and in at least good health (85.6%). The majority Table 3 compares self-reported health and functional outcomes
of the participants with baseline and post-intervention surveys at baseline enrollment and post-intervention. Odds ratios were
(94.7%) completed 70% of the seven session program. There were adjusted for gender, age, race, and state. The adjusted odds ratio
no statistically significant differences between those who com- (aOR) of reporting excellent or very good health status increased
pleted both surveys (the analytical sample) and those who only by 56% (aOR = 1.56, 95% CI 1.22–2.00). The odds of being very
completed the baseline survey except in terms of class completion or mostly satisfied with their physical activity levels increased sig-
(Table 1). nificantly (aOR = 1.74, 95% CI 1.36–2.23) as did their confidence
that a fall could be avoided (aOR = 4.60, 95% CI 2.94–7.22). Three
PARTICIPANT FUNCTIONAL PERFORMANCE of the five items assessing functional status indicated improvement
Of 336 participants with baseline data, 254 (75.6%) completed (Table 3). Participants were more likely to report no difficulty in
the TUG test at both baseline and post-intervention (Table 2). walking one block (aOR = 1.36, 95% CI 1.09, 1.69); getting out
Of these, 123 (48.4%) were classified as high risk. After complet- of a straight backed chair (aOR = 1.89, 95% CI 1.43–2.50); and
ing Stepping On, overall TUG scores significantly decreased 2.1 s climbing one flight of stairs (aOR = 1.42, 95% CI 1.11–1.82). Con-
(SD ± 3.1). The change was greatest among high risk participants trolling for the number of sessions attended did not substantially
whose TUG scores decreased an average of 3.2 s (SD ± 3.9). affect our results (data not shown).

Table 2 | Changes in Stepping On participants’ timed up and go (TUG) times in seconds from baseline to post-interventiona .

Baseline TUG Post-intervention TUG Change in TUG from baseline to

post-interventionb

Changes in timed up and go (TUG) times (in seconds) N Mean (±SD) N Mean (±SD) N Mean (±SD) p-value

TUG times for all participants 254 13.5 (±5.7) 254 11.4 (±4.7) 254 −2.1 (±3.1) <0.001
High risk (enrollment TUG time ≥12 s) 123 17.6 (±5.6) 123 14.4 (±4.9) 123 −3.2 (±3.9) <0.001
Low risk (enrollment TUG time < 12 s) 131 9.6 (±1.4) 131 8.6 (±1.8) 131 −1.0 (±1.5) <0.001

SD, standard deviation.

a
While 329 participants completed the TUG at enrollment, this table highlights the 254 participants who completed the TUG at both baseline and post-intervention.
b
Paired t-tests with an alpha of 0.05 were used to compare changes in participant’s TUG time between baseline and post-intervention. A reduction in time indicates
a positive functional improvement.

Table 3 | Changes in Stepping On participants’ self-reported health and functional outcomes from baseline to post-interventiona .

Self-reported health and Baseline Post-intervention Adjusted change from baseline to

functional outcome measures (N = 266)b (N = 266)b post-interventionc

N (%) N (%) Odds ratios from p-value

logistic models

Health status, satisfaction, and confidence

Excellent or very good health status 96 (36.2%) 123 (46.8%) 1.56 (1.22, 2.00) <0.001
Very/mostly satisfied with physical activity levels 123 (46.8%) 155 (59.4%) 1.74 (1.36, 2.23) <0.001
Feel confident not falling (strongly agree or agree) 180 (69.8%) 237 (91.2%) 4.60 (2.94, 7.22) <0.001
Self-reported functional status
No difficulty in walking across room 195 (75.0%) 204 (79.4%) 1.23 (0.95, 1.59) 0.121
No difficulty in walking one block 144 (55.8%) 161 (62.4%) 1.36 (1.09, 1.69) 0.007
No difficulty in stooping, crouching, kneeling 59 (23.0%) 66(25.8%) 1.12 (0.86, 1.46) 0.403
No difficulty in getting out of a straight back chair 154 (59.7%) 189 (73.3%) 1.89 (1.43, 2.50) <0.001
No difficulty in climbing one flight of stairs 102 (40.2%) 125 (48.6%) 1.42 (1.11, 1.82) 0.006

a
Data are reported for the n = 266 participants who completed both the baseline and post-intervention surveys.
b
The sample size is slightly smaller than 266 for some health outcomes due to missing data on individual outcome measures. The amount of missing data ranges
from 0 to 5% for different outcomes.
c
Adjusted odds ratios from GEE logistic regression modeling the probability of response = 1 at an alpha of 0.05. All models account for repeated measures from the
same participant and are adjusted for gender, age, race, and program location. An odds ratio >1 represents a positive improvement in self-reported health.

www.frontiersin.org April 2015 | Volume 2 | Article 232 | 225

Ory et al. Implementing Stepping On

DISCUSSION In order to minimize the reporting burden on the program

This study examined 2 years of evaluation data collected from implementation staff, we used a limited number of self-reported
older adults aged 60 years and older who participated in the Step- outcomes and one timed functional assessment (i.e., the TUG test).
ping On community-based fall prevention program. We observed Although there was training provided for conducting the TUG
improvements in both the observed and self-reported functional (24), including available step-by-step online videos, this training
abilities of program participants. Comparing data collected at was limited. Therefore, results may not be comparable to standard-
baseline enrollment and program completion, Stepping On was ized TUG tests administered by trained professionals and some
associated with significant improvements in TUG scores and in misclassification of a participant’s fall risk may have occurred.
self-reported measures of health status, satisfaction with their While participants reported improvements in self-reported func-
physical activity levels, and fall-related confidence. This suggests tional ability and demonstrated better TUG scores, we do not know
that Stepping On contributes to functional improvements and if there was a reduction in falls. Data about falls were not collected
may also contribute to participants’ general sense of well-being. because of anticipated problems with recall bias.
The largest improvement was seen in feeling confident that falls Although we did not assess fidelity directly, we believe that pro-
could be avoided, which increased from approximately 70% at gram fidelity was maintained by training and certifying group
enrollment to over 90% after completion of the Stepping On facilitators and using the detailed Implementation Guide that
program. Given that fear of falling is a fall risk factor (16–19), emphasized the importance of program fidelity.
reduced fear coupled with increased functional ability is important
components of an effective fall prevention program. CONCLUSION
Recruitment and retention of participants is a concern for most Stepping On was previously shown to be effective at reducing fall
fall prevention programs. While the race and ethnicity of Stepping risk in a randomized controlled trial. Intended for older adults who
On participants reflected the population from which they were have fallen in the past or are afraid of falling, Stepping On applies
recruited, there was a low percentage of male participants. There adult education and social learning principles to teach older adults
were limited numbers of referrals from health care providers, strategies that they can use to reduce their risk of falling. Step-
which suggest the need for better linkages between clinical and ping On participants practice balance and strength exercises, learn
community approaches to fall prevention (20). Involvement of how to increase their safety at home and in the community,
health care professionals can be critical for motivating older and learn about the importance of vision assessment and med-
patients at risk of falling to enroll in and complete evidence-based ication reviews. This study confirms that the program provides
fall prevention programs. positive benefits and reduces fall risk factors among participants
In regards to participant retention, we observed some attrition; when implemented in multiple community-based settings in three
however, the majority of the 366 enrolled participants (81.5%) states.
completed at least 70% of the sessions. Stepping On runs only
7 weeks, so program attrition may be less of a problem than for ACKNOWLEDGMENTS
longer running programs. For example, the fall prevention pro- We thank the Colorado, New York, and Oregon State Injury Pre-
gram, Tai Chi Moving for Better Balance (TCMBB), requires two vention Centers for coordinating The Stepping On programs and
1 hour sessions over the course of 12 weeks (21). For TCMBB, collecting the study data and Margaret Kaniewski and Heidi Ehren-
only about half of the participants completed at least 70% of reich with the Centers for Disease Control and Prevention for their
the program sessions (22). It also may have helped that Step- programmatic support and critical review of the manuscript. This
ping On includes a social component, a break halfway through the research was supported in part by the Centers for Disease Con-
2 hour session, when participants can mingle and share refresh- trol and Prevention, Prevention Research Centers Special Initiative
ments. Further, it is possible that using the TUG test may have Project, with funding from the National Center for Injury Preven-
helped retain participants. While no data were systematically col- tion and Control for the State Fall Prevention Program under
lected on participants like or dislike of the TUG test, multiple 5U48DP001924, and the Research Participation Program at the
participants told their leaders that they enjoyed receiving their Centers for Disease Control and Prevention administered by the
TUG times. The importance of timely performance feedback has Oak Ridge Institute for Science and Education through an inter-
been documented previously as a motivating factor for program agency agreement between the U.S. Department of Energy and
participation (23). CDC. The findings and conclusions in this report are those of the
authors and do not necessarily represent the official position of
LIMITATIONS the Centers for Disease Control and Prevention.
This study has a number of limitations that must be acknowledged.
Participants were self-selected and this may limit the generaliz- REFERENCES
ability of the results to the broader older adult population in those 1. Stevens J, Noonan R, Rubenstein L. Older adult fall prevention: perceptions,
communities. As we did not collect data on co-morbid conditions, beliefs, and behaviors. Am J Lifestyle Med (2010) 4(1):16–20. doi:10.1177/
we could not determine if our participants were similar to the 1559827609348350
broader population of older adults who were fearful of falling or 2. Tinetti M, Baker D, Garrett P, Gottschalk M, Koch M, Horwitz R. Yale FIC-
SIT: risk factor abatement strategy for fall prevention. J Am Geriatr Soc (1993)
had experienced prior falls. Similarly, we were unable to assess the 41(3):315–20.
extent to which co-morbid conditions were related to our study 3. Rubenstein LZ. Falls in older people: epidemiology, risk factors and strategies for
outcomes. prevention. Age Aging (2006) 35(Suppl 2):ii37–41. doi:10.1093/ageing/afl084

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 232 | 226
Ory et al. Implementing Stepping On

4. Stevens JA, Baldwin GT, Ballesteros MF, Noonan RK, Sleet DA. An older adult 20. Stevens JA, Phelan EA. Development of STEADI: a fall prevention resource for
falls research agenda from a public health perspective. Clin Geriatr Med (2010) health care providers. Health Promot Pract [Internet]. (2012); [1-10 pp.]. Avail-
26(4):767–79. doi:10.1016/j.cger.2010.06.006 able from: http://www.ncbi.nlm.nih.gov/pubmed/23159993
5. Clemson L, Cumming RG, Kendig H, Swann M, Heard R, Taylor K. The effec- 21. National Center for Injury Prevention and Control. Tai chi: Moving for Better
tiveness of a community-based program for reducing the incidence of falls Balance a Guide for Program Implementation. (2011). p. 1–146. Available from:
in the elderly: a randomized trial. J Am Geriatr Soc (2004) 52(9):1487–94. http://api.ning.com/files/lpWX79eu*NgqhrxYqum3lrPOm6Dp4GRQCK9J1qw
doi:10.1111/j.1532-5415.2004.52411.x tJylSBPaJ2W38NVNJNcBqszLkRplWB9ee-lwV4PIN-hUoPujtn*zU9vRw/TC
6. Wisconsin Institute for Healthy Aging. Stepping on an Implementation Guide: Manual_Compiled__v19.pdf
How to Prepare for, Implement, and Evaluate Stepping on in Community 22. Ory MG, Smith ML, Parker EM, Jiang L, Chen S, Wilson AD, et al. Fall pre-
Settings. Centers for Disease Control and Prevention (2012). Available from: vention in community settings: results from implementing Tai Chi: Mov-
https://wihealthyaging.org/_data/files/SO_materials/Stepping-On-Manual_10- ing for Better Balance in three states. Front Public Health (2015) 2:258.
17-2013.pdf. doi:10.3389/fpubh.2014.00258
7. Mahoney JE. “Stepping On”: stepping over the chasm from research to practice. 23. American College of Sports Medicine. ACSM’s Guidelines for Exercise Testing
Front Public Health (2015) 2:148. doi:10.3389/fpubh.2014.00148 and Prescription. 9 ed. Baltimore, MD: Lippincott Williams & Wilkins (2013).
8. Stevens JA. A CDC Compendium of Effective Fall Interventions: What Works for 480 p.
Community-Dwelling Older Adults. Atlanta, GA: Centers for Disease Control 24. Centers for Disease Control and Prevention (CDC). STEADI (Stopping Elderly
and Prevention. National Center for Injury Prevention and Control. Division of Accidents, Deaths & Injuries) Tool Kit for Health Care Providers Atlanta, GA,
Unintentional Injury Prevention (2010). (2014). Available from: http://www.cdc.gov/homeandrecreationalsafety/Falls/
9. Kaniewski M, Stevens JA, Parker EM, Lee R. An introduction to the Centers for steadi/index.html?s_cid=tw_injdir154.
Disease Control and Prevention’s efforts to prevent older adult falls. Front Public
Health (2015) 2:119. doi:10.3389/fpubh.2014.00119
Conflict of Interest Statement: The authors declare that the research was conducted
10. Wallace RB, Herzog AR. Overview of the health measures in the health and
in the absence of any commercial or financial relationships that could be construed
retirement study. J Hum Resour (1995) 30:S84–107. doi:10.2307/146279
as a potential conflict of interest.
11. Hutton I, Gamble G, McLean G, Butcher H, Gow P, Dalbeth N. Obstacles to
action in arthritis: a community case-control study. Int J Rheum Dis (2009)
This paper is included in the Research Topic, “Evidence-Based Programming for Older
12(2):107–17. doi:10.1111/j.1756-185X.2009.01392.x
Adults.” This Research Topic received partial funding from multiple government and
12. Shumway-Cook A, Brauer S, Woollacott M. Predicting the probability for falls
private organizations/agencies; however, the views, findings, and conclusions in these
in community-dwelling older adults using the timed up & go test. Phys Ther
articles are those of the authors and do not necessarily represent the official position
(2000) 80(9):896–903.
of these organizations/agencies. All papers published in the Research Topic received
13. Bohannon RW. Reference values for the timed up and go test: a descriptive
peer review from members of the Frontiers in Public Health (Public Health Educa-
meta-analysis. J Geriatr Phys Ther (2006) 29(2):64–8. doi:10.1519/00139143-
tion and Promotion section) panel of Review Editors. Because this Research Topic
200608000-00004
represents work closely associated with a nationwide evidence-based movement in
14. Podsiadlo D, Richardson S. The timed “up & go”: a test of basic functional
the US, many of the authors and/or Review Editors may have worked together pre-
mobility for frail elderly persons. J Am Geriatr Soc (1991) 39(2):142–8.
viously in some fashion. Review Editors were purposively selected based on their
15. Bischoff HA, Stahelin HB, Monsch AU, Iversen MD, Weyh A, von Dechend M,
expertise with evaluation and/or evidence-based programming for older adults. Review
et al. Identifying a cut-off point for normal mobility: a comparison of the timed
Editors were independent of named authors on any given article published in this
‘up and go’ test in community-dwelling and institutionalised elderly women.
volume.
Age Ageing (2003) 32(3):315–20. doi:10.1093/ageing/32.3.315
16. Tennstedt S, Howland J, Lachman M, Peterson E, Kasten L, Jette A. A random-
ized, controlled trial of a group intervention to reduce fear of falling and asso- Received: 27 August 2014; accepted: 27 October 2014; published online: 27 April 2015.
ciated activity restriction in older adults. J Gerontol B Psychol Sci Soc Sci (1998) Citation: Ory MG, Smith ML, Jiang L, Lee R, Chen S, Wilson AD, Stevens JA and
53(6):384–92. doi:10.1093/geronb/53B.6.P384 Parker EM (2015) Fall prevention in community settings: results from implementing
17. Tinetti ME, Baker DI, McAvay G, Claus EB, Garrett P, Gottschalk M, et al. A Stepping On in three states. Front. Public Health 2:232. doi: 10.3389/fpubh.2014.00232
multifactorial intervention to reduce the risk of falling among elderly people This article was submitted to Public Health Education and Promotion, a section of the
living in the community. N Engl J Med (1994) 331(13):821–7. doi:10.1056/ journal Frontiers in Public Health.
NEJM199409293311301 Copyright © 2015 Ory, Smith, Jiang , Lee, Chen, Wilson, Stevens and Parker. This is an
18. Tinetti ME, Richman D, Powell L. Falls efficacy as a measure of fear of falling. open-access article distributed under the terms of the Creative Commons Attribution
J Gerontol (1990) 45(6):239–43. doi:10.1093/geronj/45.6.P239 License (CC BY). The use, distribution or reproduction in other forums is permitted,
19. Vellas BJ, Wayne SJ, Romero LJ, Baumgartner RN, Garry PJ. Fear of falling provided the original author(s) or licensor are credited and that the original publica-
and restriction of mobility in elderly fallers. Age Ageing (1997) 26(3):189–93. tion in this journal is cited, in accordance with accepted academic practice. No use,
doi:10.1093/ageing/26.3.189 distribution or reproduction is permitted which does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 232 | 227

 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00152

Translation of The Otago Exercise Program for adoption

and implementation in the United States
Tiffany E. Shubert 1 *, Matthew Lee Smith 2 , Marcia G. Ory 3 , Cristine B. Clarke 1 , Stephanie A. Bomberger 4 ,
Ellen Roberts 5 and Jan Busby-Whitehead 5
1
Center for Aging and Health, University of North Carolina, School of Medicine, Chapel Hill, NC, USA
2
College of Public Health, The University of Georgia, Athens, GA, USA
3
Department of Health Promotion and Community Health Sciences, Texas A&M School of Public Health, College Station, TX, USA
4
University of North Carolina Center for Health Promotion and Disease Prevention, Chapel Hill, NC, USA
5
Division of Geriatrics, University of North Carolina, School of Medicine, Chapel Hill, NC, USA

Edited by: Background: The Otago Exercise Program (OEP) is an evidence-based fall prevention pro-
Sanjay P. Zodpey, Public Health
gram developed, evaluated, and disseminated in New Zealand. The program was designed
Foundation of India, India
for delivery in the home by physical therapists (PTs). It was not known if American PTs
Reviewed by:
Xiaoguang Ma, University of South would require additional training and resources to adopt the OEP. This article describes the
Carolina, USA process of translating the OEP for dissemination in the US. Processes included review-
Nilesh Chandrakant Gawde, Tata ing and piloting the New Zealand training materials to identify implementation challenges,
Institute of Social Sciences, India
updating training materials to be consistent with American physical therapy practices, pilot-
*Correspondence:
ing the updated training materials in an online format, and determining if the online format
Tiffany E. Shubert , Center for Aging
and Health, University of North reached the target PT audience.
Carolina, School of Medicine, Old
Clinic, Campus Box 7550, Chapel Hill,
Methods – Process Activities:The New Zealand manual was reviewed by expert American
NC 27599-7550, USA PTs and a training webinar was piloted with 56 American PTs. Feedback suggested that the
e-mail: [email protected] program itself was understood by PTs, but training materials required modification related
to documentation and reimbursement policies. Additional content was developed and inte-
grated into an online training module. The online training was piloted and then deemed
adequate by seven PT subject matter experts. The online training was launched in March
2013. Demographic and practice data were collected to characterize the PTs attending the
online training as well as perceived barriers and facilitators to implementation (n = 522).
Perceived facilitators include the effectiveness of the OEP to facilitate adoption, but the
lack of agency support, billing and reimbursement challenges pose a significant barrier to
OEP implementation.
Conclusion: The OEP required additional information to facilitate adoption by American
PTs. Online training that specifically targets PTs appears to effectively reach the target
audience and be well received by participants. More research is required to determine the
impact of online training on a PT’s adoption and implementation of this material into their
practice.
Keywords: fall prevention, health promotion, physical therapy, balance, aging, policy

INTRODUCTION Given the extensive and complex nature of falls among older
Older adult falls are a significant public health problem (1). The adults, interventions to prevent falls and related injuries have
reasons why older adults fall are complex and typically a result of been studied for over two decades. Several fall prevention pro-
multiple, interacting risk factors unique to the individual as they grams have been developed, tested, and proven effective to reduce
interact with their physical environment (2). The most common falls among community-dwelling older adults (6). To facilitate
risk factors for falling are leg muscle weakness, difficulty walking, the dissemination and implementation of these programs, the
polypharmacy (too much or the wrong type of medications), cog- Centers for Disease Control and Prevention (CDC) published
nitive impairment, vision impairment, and challenges within the “Compendium of Effective Fall Interventions: What Works for
environment (3). Of greatest concern are the falls experienced by Community-Dwelling Older Adults” in 2008 (7), with a second
those aged 75 and over. It is estimated that 50% of adults in this edition in 2010 (6). The second edition of the Compendium lists
age group fall annually (4). These falls result in the greatest num- 22 interventions that have effectively reduced the rate of falls or
ber of visits to healthcare providers and significant morbidity and fall-related injury. Each intervention includes a summary of the
mortality (5). outcomes, program setting, target audience, content (key elements,

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 152 | 228
Shubert et al. The Otago Exercise Program

frequency, and duration), and delivery system (who is qualified to materials with PTs to identify implementation challenges, updat-
deliver, level of training required). Of the 22 interventions, only ing the OEP training materials to be more consistent with Ameri-
three have incorporated and expanded the key elements into an can physical therapy practices, piloting the updated training mate-
implementation manual and training system to ensure program rials in an online format, and then determining if the online format
delivery with fidelity across community (Tai Chi: Moving For Bet- reached the target audience of PTs who work with frail older adults.
ter Balance and Stepping On) (8, 9) and home-based (the Otago
Exercise Program – OEP) settings (10). METHODS – PROCESS ACTIVITIES
These three programs target older adults with the physical and TRANSLATION OF THE OTAGO EXERCISE PROGRAM FOR
mental abilities to live in non-institutional settings. Tai Chi is most DISSEMINATION IN THE UNITED STATES
appropriate for those older adults with the greatest mobility skills The OEP was developed and tested for dissemination and imple-
(11), Stepping On is for those older adults who are transitioning to mentation in a country with a nationalized healthcare system. A
be less mobile (12), and the OEP is the most effective for those older manual to describe the implementation process was published in
adults who are the least mobile and at the highest risk of falling (6). 2003 by the program developers (10). Before dissemination in
The OEP target audience may have limited mobility and access to the United States, it was deemed necessary to review all train-
group exercise settings, which differs from the other two programs ing materials and make modifications to support adoption and
in that it was designed to be delivered in the home (10). implementation by an American audience. Part of translation
The OEP was developed and evaluated in New Zealand in the plan developed by the American team responsible for translat-
late 1990s and proven effective in randomized controlled trials at ing the OEP was to create and integrate a centralized system to
reducing falls in high-risk older adults by 35% (13, 14). Due to offer education and training to PTs.
the complex medical conditions inherent in the target audience, The following plan was deployed to review and revise the OEP
the OEP was delivered by healthcare professionals. The creators manual and training materials for dissemination in the United
of the New Zealand OEP deemed that physical therapists (PTs), States:
who receive extensive training in musculoskeletal rehabilitation,
should at a minimum supervise, and ideally implement, the OEP 1. PTs with expertise in fall prevention and implementation of
(10). PTs have the training and expertise to evaluate an individ- the OEP were to review the materials and identify any revisions
ual’s risk of falling; identify additional medical risk factors such as necessary to support program adoptions
orthostatic hypotension, polypharmacy, arrhythmia; refer to other 2. Pilot a real-time webinar based on the revised manual for Amer-
healthcare providers to manage risk; and prescribe and progress ican audiences amongst a small group of PTs from three states –
an older adult through a fall prevention program (15). Oregon, Colorado, and New York that were participating in the
The OEP is an innovative model of low frequency of physical Centers for Disease Control Fall Prevention Pilot Project
therapy sessions over a long duration. The original program was 3. Identify “lessons learned” from Otago implementation based
delivered in six visits over a year. The first four visits are in the on feedback from the webinars
first 2 months of the program (i.e., the initial visit, a visit a week 4. Revise training materials based on lessons learned
later, then a visit 2 weeks later, then 4 weeks later); then follow-up 5. Develop an online training program for broad dissemination
visits are conducted at 6 and 12 months with monthly “check-in” in the United States
phone calls between (13, 16). This type of model sets the stage for 6. Pilot training with a small group of practicing PTs for feedback
the patient engagement and ownership of their exercise program. 7. Revise and deploy online training
The program only works if the patient does the exercises. The OEP 8. Determine if online training was reaching the target audience
achieves that goal with over 35% of participants stating they per- of PTs most likely to adopt and implement the OEP in their
form the exercises three times a week 1 year after the start of the practice settings.
program (13).
Given the robust results of the OEP, and the above average REVISIONS SPECIFIC TO AMERICAN PTs
adherence and compliance rates, the CDC selected the OEP as one Expert PTs (T. Shea and T. Shubert) who had extensive knowl-
of three evidence-based fall prevention programs for dissemina- edge of the OEP implementation both in the United States and
tion in the United States. The implementation and dissemination in New Zealand worked with one of the OEP program devel-
materials for the OEP were developed in New Zealand. These opers (C. Robertson) to review the Otago Exercise Programme
materials offered a concise summary of the research supporting Manual (10). Revisions were made to the original manual. A
the OEP and step-by-step instructions about how the program United States version of the OEP Manual was released and
was prescribed (10). However, the New Zealand manual did not made available in early 2012 (http://www.med.unc.edu/aging/
account for policies and practices unique to the American health- cgec/exercise-program). The content of this manual was presented
care system, nor did it provide any guidance about how to integrate in a 1-hour training webinar offered four times in 2012 to 56 PTs.
the OEP into the workflow of a PT. It was not known if American The 56 PTs who attended had been recruited by their respective
PTs would require additional training and resources to adopt the State Division of Public Health Units (OR, CO, NY) to participate
OEP and implement it as intended. in a project to implement the OEP as part of the Fall Prevention
The purpose of this article is to describe the process of trans- Pilot Project. Attendance at the webinars was the first step in that
lating the OEP for dissemination in the United States. Processes process, and they were recruited via personal invitation from their
included reviewing and piloting the New Zealand OEP training state partners.

www.frontiersin.org April 2015 | Volume 2 | Article 152 | 229

Shubert et al. The Otago Exercise Program

The training webinars were designed to pilot the mater- variations in clinical care around falls. Standardizing practice
ial. Throughout the course of each webinar, therapists were was appealing to some PTs and distasteful to others (18, 19).
encouraged to ask questions either by telephone or using the 6. The OEP was delivered in the home; however, the policies
online chat function. We anticipated that many of the questions for billing and reimbursement for Physical Therapy under
would be about how to actually prescribe the program; however, Medicare Part A (Home Health) make it virtually impossible to
questions and discussions were more about implementation dif- implement Otago over a year period.
ferences between New Zealand and America and how to address 7. A new model of PT delivery of care, which has emerged, allows
these differences. The following themes were identified as common for delivering physical therapy in the home but billing under
challenges to implementation throughout the webinars: Medicare Part B (outpatient). Though this model allows for
greater opportunity to deliver the OEP over the year-long
period, the paperwork burden on the PT was still sizeable.
1. The theory and implementation of evidence-based health 8. Webinars and online training were deemed as an acceptable
promotion programs were not common knowledge for PTs. mode of training by PTs.
2. In the original OEP research studies, subjects were at risk of
falls but not actually seeing a PT for a diagnosed impairment. Given the feedback from the webinars, the content from the
In order for Medicare to reimburse a PT for an episode of care, New Zealand manual was deemed appropriate for teaching thera-
there needs to be a diagnosed impairment that requires “skilled pists exercise program specifics. However, they believed that imple-
and necessary” physical therapy (17). mentation in the United States would require additional informa-
3. Subjects in the original OEP research scored at risk of falls. This tion about how to integrate the program into the workflow, given
criterion was used for PTs to implement the OEP as part of the documentation, billing, and reimbursement requirements. It was
plan of care. Given that patients required skilled therapy, they also identified that therapists would benefit from additional back-
were often weak and required a dose of physical therapy before ground about the theory behind evidence-based programs and the
starting the OEP. This dose of physical therapy was necessary research behind the OEP.
to improve their strength and mobility so that they would be The feedback from the PTs was then incorporated into the train-
able to participate at the appropriate frequency and duration. ing manual. The PTs who attended the webinar agreed that the
4. The OEP exercises were not unique to physical therapy, but content of the OEP did not need to be presented in a face-to-
the low frequency of PT visits and long duration of the OEP face setting because much of the actual program was common
was deemed be an innovative practice model. Typical courses to both PT practice and education. It was deemed that an online
of physical therapy follow a model for 2–3 times a week for a medium would be acceptable to disseminate the training to PTs in
period of 4–8 weeks. There was concern from therapists that the the United States.
OEP model with its low frequency and long duration would be
considered outside of the acceptable course of therapy. Being DEVELOPMENT OF ONLINE TRAINING PROGRAM
outside of normative values may result in a “red flag” to be The online curriculum was an adaption of the webinar and devel-
audited by Medicare. oped by the same authors. The curriculum incorporated “adult
5. The OEP offered an opportunity to standardize practice learning theory”using video, interactive assignments, and required
around fall prevention. The literature demonstrates significant posting to external discussion boards (Figure 1). The online

FIGURE 1 | Otago online training program and activities.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 152 | 230
Shubert et al. The Otago Exercise Program

version was developed into a power point and piloted by seven The SMEs reported that the course was acceptable and engag-
subject matter expert (SME) PTs – three knowledgeable about the ing. The training was deemed adequate in length (2–3 h) and
OEP and four without prior experience of the OEP. All clinicians appropriately priced ($25). The curriculum was easy to navigate.
had at least 5 years of clinical experience. The content was acceptable and clinically relevant. The exercise
The SMEs were invited to review the content from a select videos and case studies were well received; however, clinicians with
group of PTs who had received advanced certification in geriatrics, several years of experience (>5) felt that the video cases were too
and who had contacted the researchers independent of the online contrived and not realistic.
training to learn more about the efforts to implement the OEP. Subject matter experts listed the following additional concerns:
The SMEs were instructed to complete the course including
outside assignments, quizzes, and a final exam. The SMEs then 1. “The OEP is appropriate for clinical use, but I have con-
evaluated the following with open-ended questions (Table 1): cerns about billing, reimbursement, and program fidelity”(four
(1) Course logistics – was it easy to find, navigate, complete? SMEs made this statement)
(2) Course content – was the information interesting, helpful, 2. “The OEP may be challenging to implement and deliver for
presented with fidelity to the original program? (3) Clinical therapists who are not in-home Part B providers” (two SMEs).
usability/feasibility – could they apply this in their clinical set-
ting? (4) Research – was it presented in a meaningful way? The DEPLOYMENT OF ONLINE TRAINING PROGRAM
responses were summarized and reviewed by course creators and The feedback from the SMEs was collated and revisions to the
independent external consultant. course were made. The course was deployed in March 2013. The
course was advertised via the University of North Carolina at
Chapel Hill’s School of Medicine website, national listserv for
PTs, and word-of-mouth. Key partners such as the American
Table 1 | Otago online pilot evaluation open-ended questions. Physical Therapy Association and CDC informed various groups
interested in balance and fall prevention. The online training pro-
Logistics gram continues to be advertised through monthly postings on
1. Describe how you found the course navigation to be. Was it easy to national listservs for PTs, quarterly webinars for the National Falls
get around? Free Coalition, and at national and international meetings and
2. Were the directions clear to access phConnect? conferences.
3. How well did the quizzes cover the content in your opinion? To minimize cost barriers, the course was priced at $25. Upon
4. What was your opinion on the usefulness of the case studies completion of the course, the registrants received 2 Continuing
presented in the videos? Education Units (CEUs). Many states require PTs to attend and
5. How realistic did the case studies feel to you? report a minimum amount of continuing competence training
6. How easy was it to post questions on PH Connect? annually to renew their license. These courses are often expensive.
Content We felt offering low-cost CEUs would add an additional incentive
1. Tell me three things you remember from the content? to therapists interested in completing the training.
2. Was there anything that felt incompletely explained? Participants enrolled in the course via the Area Health Edu-
3. Was there anything that seemed too elementary? cation Center Connect website. The course was described as a
4. . . ..or too advanced? 3-hour experience, which could be started and stopped at any
Your motivation
time. After registration, participants completed a demographic
1. How different does the Otago Exercise Program feel from your
form including key characteristics about their clinical practice
customary PT practice?
(e.g., number of years in practice, percent of caseload over the age
2. How likely are you to use some of what you learned in the online
of 65), a pre-assessment of confidence in skills, and a baseline test
course?
about falls knowledge. The course had three other mini-quizzes
3. How motivated do you feel as a result of this experience to start
embedded into the content throughout the course: (1) knowl-
using the Otago Exercise Program with your patients?
edge assessment of standardized protocols for functional tests; (2)
an assessment of ability to evaluate functional tests and prescribe
Research
appropriate exercises from the OEP; and (3) an evaluation to assess
1. How convincing did you find the research we presented?
the mastery of the concept of fidelity. Participants were not allowed
2. Is the push toward using evidenced-based programs in PT more
to proceed to the next course section until they had demonstrated
important to you now than it was at the beginning of the course?
mastery of the content per the quiz score. Upon completion of
Fidelity the course, but before participants were awarded CEUs, they had
1. How strictly do you think PTs have to stick to the Otago Exercise to pass a final exam of 10 questions with a minimum score of
Program? 80% and complete a post-assessment about confidence in skills; an
What unanswered questions do you still have regarding intention to implement survey that included items about perceived
1. The assessment tests barriers and facilitators; and an evaluation of the course presen-
2. Billing for Otago tation and content. All participants received a follow-up survey
3. Choosing the exercises via email 6 weeks after completing the course to assess level of
4. On the Otago schedule and continuum? program implementation. All participant data were collected with

www.frontiersin.org April 2015 | Volume 2 | Article 152 | 231

Shubert et al. The Otago Exercise Program

tools embedded in the training. Data were exported, de-identified, PERCEIVED FACILITATORS AND BARRIERS TO PROGRAM
cleaned, and analyzed at 6 and 13 months post deployment. IMPLEMENTATION (TABLE 4)
Trainees were instructed to “Please estimate the degree to which
RESULTS each of the following items will facilitate your ability to imple-
It was unknown if the PTs who registered for the course would ment Otago,” and were given a list of 11 potential facilitators.
be the target audience for adoption. The goal was for PTs who Facilitators ranged from administrative support (i.e., would super-
worked primarily with older adults to complete the course. It visors pay for copying of materials and help support docu-
was also unknown if the perceived facilitators and barriers by a mentation) to payor policies (i.e., what were the local Medicare
larger audience would be consistent with the pilot results from the polices toward longer duration of treatments with low fre-
webinars. To ensure that the target audience was actually reached, quencies) to compliance issues (i.e., would patients actually
frequencies for trainee demographics and perceived facilitators do the exercises on their own?). Trainees were also asked to
and barriers were calculated using data of the first 552 PTs enrolled “Please estimate the degree to which each of the following items
in the course. will be a barrier to your ability to implement Otago,” and
were given a list of 14 barriers. Barriers ranged from getting
CHARACTERISTICS OF ONLINE OEP TRAINEES weights for patients to co-pays to paperwork issues. Table 5
The characteristics of all the OEP trainees who completed the lists the top three perceived facilitators and the top 3 perceived
training in the first 11 months of deployment are described in barriers.
Table 2. During that time frame, 552 PTs, physical therapy assis-
tants, and students enrolled in physical therapy programs enrolled DISCUSSION
in the course, and 398 completed the training. Table 3 describes This study described the process of translating a research-based
the characteristics of the trainees practice settings. Of the 398, intervention developed in a country with nationalized healthcare
30% were not in practice. These individuals were either students, for use in clinical practice within the United States. This article
researchers, or from other professions. The remaining 279 were described the process of translating the OEP to facilitate adoption
predominately therapists with significant experience in geriatrics
(211 had over 8 years of experience working with older adults) and
worked primarily in geriatric settings (75% of sample stated more Table 3 | Characteristics of therapist practice (N = 279).
than 75% of their caseload was over the age of 65).
%

Years in practice
Table 2 | Demographics of therapists who completed the online
≤3 14
program (N = 398).
4–7 10
% Sample ≥8 76
Years working with older adults
Age ≤3 14
20–29 23 4–7 16
30–39 21 ≥8 70
40–49 20
Average # visits/week
50–59 29
0–9 17
60+ 7
10–19 27
Gender 20–39 46
Male 25 >40 90
Female 75
% of caseload age 65 or older?
Race <25% 4
White 89 25–49% 6
African-American 1 50–74% 15
Asian 5 >75% 75
Native American 1 Experience with evidence-based health promotion programs (EBHP)
Other 4 Ever referred?
Practice setting Yes 36
Rural 30 No 61
Suburban/urban 66 I do not know 3
Other 4 Which program? (Select all that apply)
Patient care Matter of balance (n = 25) 9
Full time 45 Stepping on (n = 15) 5
Part time 25 Tai Chi (n = 75) 27
Not in practice 30 Other (n = 21) 8

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 152 | 232
Shubert et al. The Otago Exercise Program

Table 4 | List of facilitators and barriers.

Facilitators Barriers

I have active support from my Agency’s administration My agency does not have reimbursement or billing policies in place
I have an internal “champion” or key leader who is supportive of Otago Current Medicare reimbursement practices do not support delivery of
the program
My agency has enough staff member, skills, resources to support the work Poor patient compliance
and phone calls
My agency is/will be able to modify reimbursement and billing practices to fit My agency is not set up keep patients on caseload over an extended
Otago guidelines period of time
The program is low cost and does not need substantial resources to continue My agency does not have a system for follow-up phone calls
The research data helped convince my Agency of the value It is difficult to get weights for patients
The research data helped convince referral partners (physicians, accountable Patients will not continue with a different Part B provider
care organizations) of value
The research data and program structure helped convince me of the value Patients unable or do not want to pay co-pays
My patients like the program Medicare C payors will not cover Otago
The program is supported by community and state-based fall coalitions No way to transition patient from home health to Part B
I am able to bill as a Part B provider Agency does not have enough trained staff members, skills, resources
to support the work
Other facilitators (please specify) Agency leadership does not support the work.
Turnover among therapists implementing Otago
Other barriers (please specify)

Table 5 | Top three facilitators and top three barriers. The process of translating an intervention developed and tested
in another county was innovative, and our experience indicates
Not at all Somewhat A lot that it may be challenging to overcome barriers imposed by imple-
menting programs under different healthcare systems. Two unan-
Facilitators
ticipated challenges unique to the American healthcare system
I have active support from my Agency’s 23 95 155
became apparent during the translation process: (1) reimburse-
administration
ment issues and (2) current policies regarding frequency and
The program is low cost and does not 18 119 131
duration of physical therapy treatment.
need substantial resources to continue
Significant changes in Medicare Home Health Payment Poli-
The research data helped convince my 19 118 130
Agency of the value
cies were implemented during the time period of 2010–2013 (20,
21). When the OEP was first selected by the CDC to disseminate,
Barriers
it was assumed that PTs in the home health setting would be able
My agency is not set up to keep patients 66 127 65
to deliver Otago as intended and be reimbursed for their services.
on caseload over an extended period of
However, in October 2011, CMS released “The Final Rule” for
time
implementation in 2012 (20, 21). The “Final Rule” significantly
Patients unable or do not want to pay 38 160 59
changed reimbursement for home health rehabilitation services
co-pays
with the goal of assuring equal access to services and reduce finan-
My agency does not have a system for 94 114 50
cial gaming. In essence, the final rule limited an episode of home
follow-up phone calls
health to no more than 60 days (it can be extended but with much
paperwork) and reimbursed therapists at lower rates as more ther-
apy was utilized. The 60-day limitation, in conjunction with an
in the United States. Inherent in this process was identifying the increase in acuity of home health patients and a 3–8% reduction
barriers to adoption presented by implementing a program devel- in reimbursements depending on the patient’s acuity, effectively
oped in a different healthcare system as well as identifying and made it impossible for home health therapists to deliver Otago
implementing solutions to these barriers. In addition to translating with fidelity.
the intervention materials, this process included the development Alternative models proposed by the American translation team
of an efficient and effective system to disseminate training to PTs. leveraged PTs that treat patients in outpatient settings and have the
A secondary purpose of this project was to determine if online ability to keep their patients on caseload for a longer period of time;
training was an acceptable and feasible mechanism to reach our however, this poses a significant challenge to the fidelity of the pro-
target audience of PTs. gram. Innovative models that have therapists work with patients

www.frontiersin.org April 2015 | Volume 2 | Article 152 | 233

Shubert et al. The Otago Exercise Program

in the home, but bill as an outpatient have been investigated and should be made toward demonstrating the value of the OEP at the
demonstrate promise. However, this model for delivering therapy agency level.
is relatively new and does not have widespread penetration. Barriers included system-based challenges to maintaining a
Despite the popularity of evidence-based programs among patient on caseload, concerns about costs to the patients in the
public health professionals serving older adult populations (22), form of co-pays, and the inability to perform follow-up phone
clinicians such as PTs are often not familiar with such evidence- calls. Interestingly, the therapists who completed the training did
based programs. The concept of fidelity, or delivering a program not perceive the billing and reimbursement challenges to be a
as intended, was not familiar to the majority of learners. More barrier to program implementation. This may be because the
than 64% of those who took the training had never referred or therapists were being asked to rate these items immediately upon
incorporated an evidence-based health promotion program into completing the online training and before actually implementing
their treatment plan. Many therapists felt that a standardized pro- the program.
gram was not flexible enough to meet the needs of their patients.
The gaps identified through the work with the SMEs and the pilot CONCLUSION
testing with the PTs indicated that OEP content would be easy to The implementation of standardized fall prevention programs into
convey to PTs, but the implementation of the program with fidelity physical therapy practice is not as simple or as straightforward as
would prove to be a challenge. anticipated. PTs are well versed in the content of the OEP but
In recognition of these challenges, the online training was were not familiar with the frequency, duration, and standardiza-
revised to include several case studies to demonstrate different tion of the program. In general, PTs appreciated the effectiveness
implementation models including a home health to outpatient of the program, but there are challenges inherent to reimburse-
and an outpatient only case. Additionally, we believe the online ment for providing the OEP with fidelity to appropriate patients.
model afforded several advantages over the traditional face-to- Online training appears to be an effective way to disseminate the
face model: (1) cost-effectiveness – participants were charged $25 OEP to PTs who work with older adults; however, we anticipate
to attend versus a face-to-face course, which is typically $100–200; that additional support and resources will be necessary for PTs to
(2) reach – in the first 9 months of deployment, we had participants implement the OEP with fidelity to impact the nature of falls.
from all 50 states take the training; (3) community – participants
were invited to other opportunities to support their work; and
(4) convenience – participants could start and stop the training ACKNOWLEDGMENTS
whenever they liked. This work was supported in part by the Bureau of Health Pro-
In the first 9 months, the online training appears to be an effec- fessions (BHPr), Health Resources and Services Administration
tive mechanism to target PTs who work primarily with aging (HRSA), Department of Health and Human Services (DHHS)
patients. The program itself was advertised through a word- under grant #UB4HP19053, Carolina Geriatric Education Center.
of-mouth, website, and a few physical therapy-based listserv. This information, content, and conclusions are those of the author
The “early adopters” who completed the program were those and should not be construed as the official position or policy of,
who would be considered “senior” therapists (in practice 8 or nor should any endorsements be inferred by the BHPr, HRSA,
more years) and spent the majority of their clinical practice DHHS, or the U.S. Government. The Centers for Disease Control
time working with older adults. This supports that our tar- Prevention Research Center at Texas A&M Health Science Cen-
get audience was reached. One concerning item was that only ter, School of Rural Public Health Center for Community Health
13% of the sample were categorized as “new” therapists (3 years Development also provided funding support from the Centers for
or less of clinical practice). The low number of new graduates Disease Control and Prevention Cooperative Agreement Number
may reflect the demographics within the greater practice setting 1U48 DP001924, and the Centers for Disease Control Preven-
and that the majority of PTs in geriatrics are older and more tion Research Center at the University of North Carolina, Chapel
seasoned (23). Hill Prevention Research Center, supported by Cooperative Agree-
The perceived barriers and facilitators to program implementa- ment Number U48-DP001944. The findings and conclusions in
tion provided significant insights about the challenges of the OEP this paper are those of the author(s) and do not necessarily rep-
adoption and implementation. At the end of the online training, resent the official position of the Centers for Disease Control and
therapists were asked to rate the extent an item was considered Prevention.
to be a facilitator or a barrier to implementation. The top facil-
itator was support from Agency administration. Therapists who REFERENCES
implement the OEP without agency support are responsible for 1. Stevens J. Falls Among Older Adults: An Overview (2013) [cited 2013
August 5]. Available from: http://www.cdc.gov/HomeandRecreationalSafety/
procuring weights, copying home exercise program handouts, and Falls/adultfalls.html
ensuring all paperwork is completed correctly and in a timely 2. Tinetti ME, Kumar C. The patient who falls: “It’s always a trade-off ”. JAMA
manner. One therapist estimated the personal cost of implement- (2010) 303:258–66. doi:10.1001/jama.2009.2024
ing the OEP at about $50/patient. Agencies that supported the 3. Prevention of Falls in Older Persons AGS/BGS Clinical Practice Guidelines (2010)
OEP created systems to procure ankle weights for patients to use [cited 2012 October 15]. Available from: http://www.medcats.com/FALLS/
frameset.htm
as part of the exercise program and ensured that all photocopying 4. CDC. Self-Reported Falls and Fall-Related Injuries Among Persons Aged >65
costs were absorbed by the agency as opposed to the therapist. Years – United States, 2006 (2008) [cited 2014 June 1]. Available from: http:
Agency support is critical for program success, and more efforts //www.cdc.gov/mmwr/preview/mmwrhtml/mm5709a1.htm

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 152 | 234
Shubert et al. The Otago Exercise Program

5. Stevens J. The Cost of Falls Among Older Adults (2012) [cited 2012 December 1]. 19. Mangione KK, Lopopolo RB, Neff NP, Craik RL, Palombaro KM. Interventions
Available from: http://www.cdc.gov/HomeandRecreationalSafety/Falls/fallcost. used by physical therapists in home care for people after hip fracture. Phys Ther
html (2008) 88(2):199–210. doi:10.2522/ptj.20070023
6. Stevens J. A CDC Compendium of Effective Fall Inteventions: What Works for 20. APTA. Highlights of the 2012 Physician Fee Schedule Rule. Alexandria, VA: Amer-
Community-Dwelling Older Adults. 2nd ed. Atlanta, GA: Centers for Disease ican Physical Therapy Association (2012).
Control and Prevention, National Center for Injury Prevention and Control 21. APTA. Highlights of the 2014 Physician Fee Schedule Rule. Alexandria, VA: Amer-
(2010). ican Physical Therapy Association (2014).
7. Stevens J. Preventing Falls: What Works a CDC Compendium of Effective Com- 22. NCOA. Evidence-Based Health Promotion Programs for Older Adults: Key Factors
munity Based Interventions from Around the World. Atlanta, GA: National Center and Strategies Contributing to Program Sustainability. Washington, DC: NCOA
for Injury Prevention and Control (2008). (2012).
8. Li F, Harmer P, Glasgow RB, Mack K, Sleet D, Fisher KJ, et al. Tai Chi: moving for 23. APTA. APTA Physical Therapist Member Demographic Profile. Alexandria, VA:
better balance – development of a community-based falls prevention program. American Physical Therapy Association (2013).
Am J Public Health (2008) 98(7):1195–8.
9. Clemson L, Cumming RG, Kendig H, Swann M, Heard R, Taylor K. The
effectiveness of a community-based program for reducing the incidence of Conflict of Interest Statement: The authors declare that the research was conducted
falls in the elderly: a randomized trial. J Am Geriatr Soc (2004) 52:1487–94. in the absence of any commercial or financial relationships that could be construed
doi:10.1111/j.1532-5415.2004.52411.x as a potential conflict of interest.
10. Campbell AJ, Robertson MC. Otago Exercise Programme to Prevent Falls in Older
People: A Home-Based, Individually Tailored Strength and Balance Retraining
This paper is included in the Research Topic, “Evidence-Based Programming for Older
Program. Otago, NZ: Accident Compensation Corporation (2003).
Adults.” This Research Topic received partial funding from multiple government and
11. Cho J, Smith ML, Shubert TE, Jiang L, Ahn S, Ory MG. Gait speed among older
private organizations/agencies; however, the views, findings, and conclusions in these
participants enrolled in an evidence-based fall risk reduction program: a sub-
articles are those of the authors and do not necessarily represent the official position
group analysis. Front Public Health (2015) 3:26. doi:10.3389/fpubh.2015.00026
of these organizations/agencies. All papers published in the Research Topic received
12. Ory MG, Smith ML, Jiang L, Lee R, Chen S, Wilson AD, et al. Fall prevention
peer review from members of the Frontiers in Public Health (Public Health Education
in community settings: results from implementing Stepping On in three states.
and Promotion section) panel of Review Editors. Because this Research Topic repre-
Front Public Health (2015) 2:232. doi:10.3389/fpubh.2014.00232.
sents work closely associated with a nationwide evidence-based movement in the US,
13. Campbell AJ, Robertson MC, Gardner MM, Norton RN, Buchner DM. Falls
many of the authors and/or Review Editors may have worked together previously in
prevention over 2 years: a randomized controlled trial in women 80 years and
some fashion. Review Editors were purposively selected based on their expertise with
older. Age Ageing (1999) 28:513–8. doi:10.1093/ageing/28.6.513
evaluation and/or evidence-based programming for older adults. Review Editors were
14. Thomas S, Mackintosh S, Halbert J. Does the “Otago Exercise Programme”
independent of named authors on any given article published in this volume.
reduce mortality and falls in older adults? A systematic review and meta-analysis.
Age Ageing (2010) 39:681–7. doi:10.1093/ageing/afq102
15. American Physical Therapy Association. Guide to physical therapist practice. Received: 08 July 2014; accepted: 06 September 2014; published online: 27 April 2015.
Second edition. American Physical Therapy Association. Phys Ther (2001) Citation: Shubert TE, Smith ML, Ory MG, Clarke CB, Bomberger SA, Roberts
81:9–746. E and Busby-Whitehead J (2015) Translation of The Otago Exercise Program for
16. Campbell AJ, Robertson MC, Gardner MM, Norton RN, Tilyard MW, Buch- adoption and implementation in the United States. Front. Public Health 2:152. doi:
ner DM. Randomised controlled trial of a general practice programme of 10.3389/fpubh.2014.00152
home based exercise to prevent falls in elderly women. BMJ (1997) 315:1065–9. This article was submitted to Public Health Education and Promotion, a section of the
doi:10.1136/bmj.315.7115.1065 journal Frontiers in Public Health.
17. CMS. Medicare Benefit Policy Manual 002 – Chapter 15 Covered Medical Copyright © 2015 Shubert , Smith, Ory, Clarke, Bomberger, Roberts and Busby-
and Other Health Services. Baltimore, MD: Centers for Medicare & Medicaid Whitehead. This is an open-access article distributed under the terms of the Creative
Services (2014). Commons Attribution License (CC BY). The use, distribution or reproduction in other
18. Mahoney JE, Shea TA, Przybelski R, Jaros L, Gangnon R, Cech S, et al. Kenosha forums is permitted, provided the original author(s) or licensor are credited and that
county falls prevention study: a randomized, controlled trial of an intermediate- the original publication in this journal is cited, in accordance with accepted academic
intensity, community-based multifactorial falls intervention. J Am Geriatr Soc practice. No use, distribution or reproduction is permitted which does not comply with
(2007) 55:489–98. doi:10.1111/j.1532-5415.2007.01144.x these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 152 | 235

 PERSPECTIVE ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00209

Developing an evidence-based fall prevention curriculum

for community health workers
Julie A. St. John 1 *, Tiffany E. Shubert 2 , Matthew Lee Smith 3 , Cherie A. Rosemond 4 , Doris A. Howell 5 ,
Christopher E. Beaudoin 6 and Marcia G. Ory 7
1
Department of Public Health, Graduate School of Biomedical Sciences, Texas Tech University Health Sciences Center, Abilene, TX, USA
2
Division of Geriatrics, School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
3
College of Public Health, The University of Georgia, Athens, GA, USA
4
Center for Health Promotion and Disease Prevention, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
5
Program on Healthy Aging, Texas A&M Health Science Center School of Public Health, College Station, TX, USA
6
Department of Communication, Texas A&M University, College Station, TX, USA
7
Department of Health Promotion and Community Health Sciences, Texas A&M Health Science Center School of Public Health, College Station, TX, USA

Edited by: This perspective paper describes processes in the development of an evidence-based fall
Sanjay P. Zodpey, Public Health
prevention curriculum for community health workers/promotores (CHW/P) that highlights
Foundation of India, India
the development of the curriculum and addresses: (1) the need and rationale for involving
Reviewed by:
Milka Dancevic Gojkovic, Public CHW/P in fall prevention; (2) involvement of CHW/P and content experts in the curriculum
Health Institute of Federation of development; (3) best practices utilized in the curriculum development and training imple-
Bosnia and Herzegovina, Bosnia and mentation; and (4) next steps for dissemination and utilization of the CHW/P fall prevention
Herzegovina
curriculum. The project team of CHW/P and content experts developed, pilot tested, and
Shridhar Murlidharrao Kadam, Public
Health Foundation of India, India revised bilingual in-person training modules about fall prevention among older adults. The
*Correspondence: curriculum incorporated the following major themes: (1) fall risk factors and strategies to
Julie A. St. John, Department of reduce/prevent falls; (2) communication strategies to reduce risk of falling and strategies for
Public Health, Graduate School of developing fall prevention plans; and (3) health behavior change theories utilized to prevent
Biomedical Sciences, Texas Tech
and reduce falls. Three separate fall prevention modules were developed for CHW/P and
University Health Sciences Center,
1718 Pine Street, Abilene, TX CHW/P Instructors to be used during in-person trainings. Module development incorpo-
79601-3044, USA rated a five-step process: (1) conduct informal focus groups with CHW/P to inform content
e-mail: [email protected] development; (2) develop three in-person modules in English and Spanish with input from
content experts; (3) pilot-test the modules with CHW/P; (4) refine and finalize modules
based on pilot-test feedback; and (5) submit modules for approval of continuing education
units. This project contributes to the existing evidence-based literature by examining the
role of CHW/P in fall prevention among older adults. By including evidence-based com-
munication strategies such as message tailoring, the curriculum design allows CHW/P to
personalize the information for individuals, which can result in an effective dissemination
of a curriculum that is evidence-based and culturally appropriate.
Keywords: community health workers, promotores, curriculum development, training, fall prevention, older adults

INTRODUCTION are investigating how to effectively prevent and reduce falls among
“I’ve fallen and I can’t get up,” was a phrase made popular by older adults.
LifeCall in 1989. This commercial was a dramatized version of an Research has demonstrated that a large proportion of falls
older adult’s fall. However, this situation is the reality for numer- among community-dwelling older adults are preventable (9, 10).
ous older adults in the United States. Falls are a threat to the lives, Numerous documented strategies address fall prevention among
independence, and health of adults – especially those aged 65 and older adults – community programing, primary care practice
older. Every 18 s an older adult visits an emergency department as guidelines, and integration of physical therapists into models of
a result a fall, and every 35 min an older adult dies due to injuries care (11). Despite the growth of evidence-based fall prevention
from such a fall (1–7). programs and the emergence of state-wide fall prevention policy
Other significant consequences are associated with falls. For efforts, there continues to be a gap in community adoption of
example, one in three adults aged 65 and older fall each year, fall prevention interventions among underserved, rural, minority,
costing the U.S. healthcare system more than $30 billion dollars and low-income populations (12). Literature is lacking regard-
annually (1). This problem is even more significant due to the ing: (1) the reasons why fall prevention policies and programs
rapidly expanding aging population (8). In light of the rate of falls are or are not adopted and spread in community settings; (2)
among older adults, physical and fiscal costs, severity of falls, and the most efficient practices for creating a trained workforce for
population growth among adults ages 65 and older, researchers delivering interventions; (3) the best strategies for reaching out

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 209 | 236
St. John et al. Training CHWs in fall prevention

to underserved populations in terms of recruitment, geographic, evidence-based curriculum to train nurse assistants in fall pre-
and needs-based challenges; and (4) how to connect community vention in home health settings, which has been available since
and clinical care settings (12). A lack of infrastructure for dis- 2007. Based on the best practice strategies discussed in detail
seminating and implementing interventions to community-based in the following section, along with fall prevention strategies,
programs has contributed significantly to this gap. we developed a series of CHW/P curriculum entitled, “How can
The Policies, Programs, and Partners for Fall Prevention CHW/Promotores help older adults stay safe from falls and related
(PPPFP) study incorporated multi-level intervention strategies to injuries?” The curriculum was developed by the study team from
develop several dissemination approaches (13). This paper focuses October 2012 to May 2013 and piloted in McAllen, Texas, on June
on a training Community Health Workers/Promotores (CHW/P) 4–5, 2013, in English and Spanish with 49 CHW/P. Revisions were
to deliver fall prevention messages to older adults. CHW/P are made from July 2013 to December 2013, and the revised curricu-
described as frontline public health workers, serving as liaisons lum was deployed nationally in face to face and virtual formats in
between health and social services and the community. They facil- April 2014. This project was approved by the Institutional Review
itate access and improve the quality and cultural competence of Board at Texas A&M University.
service delivery by utilizing a wide array of skill sets (14–17).
CHW/P are trusted members of the target community, work for EMPLOYMENT OF BEST PRACTICE APPROACHES
pay or as volunteers, and typically share ethnicity, language, socioe- To effectively develop and train CHW/P, we employed three best
conomic status, and life experiences with the community members practices: (1) utilization of CHW/P to deliver health education
served. As such, CHW/P can communicate with other members messages; (2) adult learning theory; and (3) tailored messag-
of the healthcare system to ensure that community members’ ing. Studies have demonstrated CHW/P are effective in delivering
care is sensitive to cultural and community issues. Research has health education to community residents due to shared ethnic-
demonstrated the effectiveness of CHW/P among targeted His- ity, language, socioeconomic status, and life experiences with
panic populations in achieving positive health outcomes through the community members they serve (18–31). The project team
health education, case management, service coordination, and included CHW/P throughout the development of processes to
referrals (18–31). Specifically, CHW/P are effective due to their train and engage CHW/P to deliver fall prevention health edu-
cultural similarity and understanding of the population they serve, cation messages. Specifically, CHW/P identified the needs of their
as well as the subsequent trust clients have in them. These CHW/P communities, reviewed the training materials, and identified gaps
characteristics are largely due to their residing in the same com- in information and service.
munities. Further, studies have demonstrated the effectiveness Second, an adult learner-centered training approach that con-
of CHW/P in providing social support to help Hispanics adopt siders characteristics of the target audience was utilized. CHW/P
behavior change (24–30). are typically between the ages of 20 and 65, have lower educa-
One of the greatest challenges in effective fall prevention is tional attainment with reading and math skills ranging between
ensuring the population at risk actually receives recommended 4th and 8th grade levels, and are non-native English speak-
interventions. CHW/P can serve as liaisons for older adults at ers. Adult learner-centered educational strategies engage learners
risk for falling, making sure they are referred to accessible ser- in problem-based learning and teaching. Rather than a “lec-
vices and helping make sure interventions at the community or ture,” learner-centered approaches engage learners in hands-on,
healthcare level are supported to maximize program adherence. interactive activities based upon discussion and skill-building
The literature supports CHW/P as a conduit to increase fall pre- exercises (32–35).
vention awareness, but there is a gap in fall prevention training Third, tailored messaging was incorporated into the CHW/P
for CHW/P. This perspective paper will describe the development fall prevention curriculum. Message tailoring deploys informa-
and implementation of the fall prevention curriculum for CHW/P tion and change strategies to reach one specific person based
with a focus on: (1) making the case for developing evidence-based on the individual characteristics (36, 37). Tailoring differs from
CHW fall prevention training; and (2) explaining programmatic targeting of general audiences and segmenting of subgroups by
activity, including informal focus groups, module development, customizing (or personalizing) educational approaches and mes-
pilot testing, curriculum refinement, continuing education units sages to the individual. CHW/P were trained to employ tailoring
(CEUs) approval, and next steps in dissemination. to effectively educate clients. Instead of providing general edu-
cation to their overall audience – or more refined education to
MAKING THE CASE FOR DEVELOPING AN EVIDENCE-BASED certain subgroups within that overall audience – CHW/P made
CHW FALL PREVENTION TRAINING assessments about and delivered education based on the char-
A preparatory national scan of CHW/P curriculum, trainings, and acteristics of individuals in their constituency, including culture,
resources about fall prevention among older adults was conducted language, health literacy, education, gender, age, and pertinent
in October–December 2011, revealing a lack of a comprehensive experiences, beliefs, and attitudes. Tailoring-based approaches
fall prevention curricula specifically designed for CHW/P. The acknowledge how individuals differentially use, learn, and benefit
scan included searches conducted on web-based search engines from varied educational and messaging approaches. The pur-
and phone calls and emails to CHW/P organizations, networks, pose of developing training with integrated tailored messaging
associations, state CHW program offices, employers, and acade- taught via adult learning strategies was to support CHW/P to
mic institutions. No curricula on fall prevention for older adults utilize their strengths to ensure optimization of fall prevention
specifically for CHW/P were located. A wider search revealed an interventions.

www.frontiersin.org April 2015 | Volume 2 | Article 209 | 237

St. John et al. Training CHWs in fall prevention

EXPLAINING PROGRAMMATIC ACTIVITY Prior to material development, the project team conducted
INFORMAL FOCUS GROUPS TO IDENTIFY NEED informal conversations with CHW/P about fall prevention (Step
The lead partner was a Texas Department of State Health Services 1). These conversations identified potential gaps in knowledge to
(DSHS) Certified CHW/P Training Center. The curriculum incor- inform curriculum development and identify cultural influences
porated the eight competencies recognized by the Texas DSHS and attitudes critical for message tailoring. This procedure ensured
CHW/P Training and Certification Program: (1) communication; the training modules integrated input from the CHW/P and cre-
(2) teaching; (3) advocacy; (4) interpersonal skills; (5) service ated a framework to integrate characteristics unique to CHW/P
coordination; (6) capacity building; (7) organization; and (8) (e.g., culture, language, and gender) to best tailor messages.
knowledge-based skills. Texas certifies CHW/P and requires a min-
imum of 20 CEUs (i.e., 10 DSHS-certified and 10 non-certified MODULE DEVELOPMENT
CEUs) every 2 years for recertification. To maximize adoption, the The feedback from the informal focus groups and input by con-
project team developed the curriculum in English and Spanish and tent experts in healthy aging, fall prevention, health messaging,
met DSHS requirements for CEUs for CHW/P in Texas. Figure 1 and CHW/P training contributed to developing and refining
depicts the five-step training development module process, which tailoring-based training modules in English and Spanish. The
is discussed in detail in the ensuing paragraphs. curriculum was developed for any CHW/P instructor (whether

• Conducted informal focus groups with CHW/P on fall preven!on among

older adults.
• Assessed what cultural influences, a#tudes, and individual
Informal focus characteris!cs (e.g., culture, language, gender) should be used to best
groups tailor messages for CHW/P on fall preven!on.

• Developed three in-person modules in English and Spanish with input

by content experts in healthy aging, fall preven!on, tailored messaging,
and CHW/P training.
Module • Incorporated best prac!ce strategies: CHW/P, tailored messaging, adult
Development learning theory.

• Tested the three modules with CHW/P and CHW/P Instructors via in-
person training.
• CHW/P par!cipants completed pre/post assessments, evalua!ons, and
Pilot Test a six month follow-up online survey.

• Incorporated feedback from the pilot trainings.

• Finalized curriculum modules.
Curriculum
Refinement

• Submi%ed three modules for CEU approval by the Texas Department of

State Health Services CHW/P Training and Cer!fica!on Program.
• Two modules for 4 DSHS cer!fied CHW/P CEUS and one module for 5
Curriculum DSHS cer!fied CHW/P Instructor CEUs were approved.
CEU approval

FIGURE 1 | Process of CHW/P fall prevention training module development.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 209 | 238
St. John et al. Training CHWs in fall prevention

DSHS-certified or not) and those with varying knowledge about after training to assess knowledge and implementation in their
fall prevention (whether a novice or expert). The intent was for roles as CHW/P.
any CHW/P to use the curriculum and training guide to train
other CHW/P about fall prevention among older adults. Specific CURRICULUM REFINEMENT
objectives included: The team refined the curriculum based on feedback from the
pilot test. Specific feedback and revisions to the final curriculum
included:
• Explain that falls are not a normal part of aging and the majority
of falls are preventable. (1) Standardized wording for CHW/P and fall prevention termi-
• Articulate why preventing and reducing falls and related injuries nology.
among older people is especially important. (2) Refined case studies.
• Increase awareness about risk factors for falling inside and (3) Added detailed physical activity exercises.
outside the home. (4) Added a handout on local, state, and national resources; a glos-
• Develop and augment observation, reporting, and communi- sary of terms; and a fall-related Frequently Asked Questions
cation skills to improve communication with older adults and (FAQ) handout.
their families around fall prevention.
• Describe ways to help prevent or manage falls in older adults. CONTINUING EDUCATION UNITS APPROVAL
• Develop fall prevention plans. The final step of the curriculum development process included
• Explain and discuss different approaches to changing health submitting the three modules in English and Spanish for CEU
behaviors. approval by the Texas DSHS CHW/P Training and Certifica-
• Apply behavior change strategies to fall prevention and reduc- tion Program. Each of the three models was approved. Infor-
tion. mation regarding the CEU approval process can be found at
• Teach health behavior change strategies to CHW/P and older http://www.dshs.state.tx.us/mch/chw.shtm; information regard-
adults. ing the CHW/P fall prevention modules may be found at http:
//nchwtc.tamhsc.edu/.
As shown in Table 1, the three in-person training modules
addressed the following topics: (1) fall risk factors and strategies NEXT STEPS IN DISSEMINATION
to reduce/prevent falls (4 h in length); (2) communication strate- The curriculum was converted into an online format for
gies to reduce risk of falling and develop fall prevention plans broader dissemination (http://nchwtc.tamhsc.edu/fall-prevention-
(4 h in length); and (3) using health behavior change theories to curriculum). The online formats include two courses:
prevent and reduce falls (5 h in length). The first two modules
were designed to be completed independently but were linked 1. CHW/P Course: Preparing CHWs/Promotores to prevent and
in theme and content. The third module focuses on equipping reduce falls among older.
CHW/P Instructors to apply health behavior theory to fall pre- 2. CHW/P Instructor Course: Helping older adults change their
vention and reinforces fall risk factors and strategies to prevent health behaviors to prevent falls and related injuries: health
falls and related injuries. behavior change theories.
Course materials included: an introduction for trainers, facil-
For the online format, CHW/P complete pre/post-assessments,
itator’s guides, participant handouts, case studies, pre/post-
an evaluation, and a 6-month follow-up survey. The goal of this
assessments, and evaluations. Pre/post-assessments measured
approach is to create a feasible and sustainable training method
knowledge and confidence related to fall prevention and commu-
that minimizes resources while maximizing dissemination – par-
nication strategies. The evaluations gathered participant demo-
ticularly in rural and remote communities that have CHW/P
graphic information and satisfaction with the training. The cur-
but do not have local CHW/P training programs. This strat-
riculum incorporated teaching methods focusing on increasing
egy is designed to support implementation sustainability because
self-awareness and skill building through practical application,
CHW/P who received training from the fall prevention modules
including case scenarios, role play, group work, and interactive
can continue to revisit these modules at no incurred cost in their
presentations. Refer to Table 1 for detailed content.
future health outreach, education, and promotion strategies.

PILOT TESTING DISCUSSION

After developing the modules, each module was pilot tested with This perspective paper describes the development of a fall pre-
a group of CHW/P and CHW/P Instructors (i.e., 44 participants vention curriculum for CHW/P. Given the access of CHW/P to
for Module 1; 41 for Module 2; and 18 for Module 3). CHW/P at-risk older adults and their effectiveness to educate and promote
and CHW/P Instructors were recruited via emails and word-of- behavior change, CHW/P are logical partners in promoting fall
mouth. Two DSHS-certified, experienced, bilingual, and bicultural prevention strategies. However, to date, little has been attempted
CHW/P Instructors conducted the in-person trainings in English to engage CHW/P in fall prevention interventions, despite the
and Spanish – with all materials provided in both languages. The scope of the problem. More specifically, there has not been another
pilot included evaluation and assessment onsite after completion evidence-based curriculum on fall prevention among older adults
of each training module and an online survey deployed 6 months specifically designed for CHW/P.

www.frontiersin.org April 2015 | Volume 2 | Article 209 | 239

St. John et al. Training CHWs in fall prevention

Table 1 | Contents of CHW/P fall prevention training modules.

Fall prevention: curriculum for community health workers/promotores

How can CHW/promotores help older adults stay safe from falls and related injuries?

Session title Content outline Target audience Session length

Session 1: ways to prevent 1. Statistics on falls among older adults CHW/P 4 h/4 CEUs
falls and related injuries in a. Why talk about fall prevention?
older adults b. Goals of the session
c. The facts on falls
1) U.S. falls statistics
2) Local falls statistics
2. Common risk factors for falling
a. Individual risk factors
1) Physical mobility
2) Medications
3) Transitioning home from hospital
4) Fear of falling
5) Traumatic brain injury (TBI)
6) Cognitive impairment
a. Environmental risk factors
1) Home safety
2) Safety factors outside the home
3. How to identify and address risk factors
a. Assessments
1) Home fall prevention assessment for older adults
2) Check your risk for falling – self assessment
b. Communication strategies
Closed and open-ended questions
4. How to prevent and reduce falls
a. Prevention tips
1) Regular exercise program
2) Have healthcare provider review medications
3) Have vision checked
4) Make home safer
b. Putting information into action
1) Case studies & role play
2) Identification of resources

Session 2: learning skills to 1. Risk factors for falls CHW/P 4 h/4 CEUs
reduce falls and related a. Individual risk factors
injuries 1) Physical mobility
2) Medications
3) Transitioning home from hospital
4) Fear of falling
5) Traumatic brain injury (TBI)
6) Cognitive impairment
b. Environmental risk factors
1) Home safety
2) Safety factors outside the home
2. Enhance communication skills to tailor messages and ask open-ended questions
a. Strategies to communicate effectively about falls prevention
1) Closed and open-ended questions
2) Observe, record, and report
3) Communicating with older adults
a) What to do if an older adult falls
b) How to get up from a fall

(Continued)

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 209 | 240
St. John et al. Training CHWs in fall prevention

Table 1 | Continued

Session title Content outline Target audience Session length

4) Communicating with care givers

5) Communicating with health professionals
b. Tailored messaging
1) Assessment
a) Stage of behavioral change of the individual
b) Personal characteristics of the individual
2) Message creation and delivery
a) Analyze falls prevention strategies
b) Identify characteristics
c) Develop relevant messages
3. Apply communication skills to case scenarios and identify
risk factors to reduce risk of falling
a. Case studies
b. Role play
4. Develop and implement a fall prevention plan
a. Role play
b. Interview an older adult
5. Identify resources for fall prevention
a. Group and individual activities

Session 3: helping older 1. Theories of health behavior change CHW/P instructors 5 h/5 CEUs
adults change their health a. Why talk about behavior change
behaviors to prevent falls and b. Health belief model
related injuries: health 1) Constructs
behavior change theories a) Perceived susceptibility
b) Perceived severity
c) Perceived benefits
d) Perceived barriers
e) Perceived self-efficacy
f) Cues to action
2) Scenarios
c. Trans-theoretical model
1) Stages of change
a) Pre-contemplation
b) Contemplation
c) Preparation
d) Action
e) Maintenance
2) Processes of change
3) Scenarios
d. Theory of reasoned action/theory of planned behavior
1) Constructs
a) Attitude
b) Norm
c) Intention
d) Perceived behavioral control
2) Scenarios
2. Fall prevention risk factors
a. Individual risk factors
1) Physical mobility
2) Medications
3) Transitioning home from hospital
4) Fear of falling

(Continued)

www.frontiersin.org April 2015 | Volume 2 | Article 209 | 241

St. John et al. Training CHWs in fall prevention

Table 1 | Continued

Session title Content outline Target audience Session length

5) Traumatic brain injury (TBI)

6) Cognitive impairment
b. Environmental risk factors
1) Home safety
2) Safety factors outside the home
3. Strategies for managing falls
a. Regular exercise program
b. Have healthcare provider review meds
c. Have vision checked
d. Make home safer
4.Tailored communication
a. Communication approaches
1) General messaging
2) Targeted messaging
3) Segmented messaging
4) Tailored messaging
b. Reasons to tailor messages
c. Research on tailoring
d. Steps in tailoring
1) Analyze a health problem
2) Identify pertinent characteristics
3) Develop pertinent messaging
e. Key fall prevention messages
5. Application of behavior change concepts to fall prevention
and reduction
a. Review game
b. Case studies
6. Skills to work with older adults/CHW/P to implement
behavioral changes to prevent falls
a. Interviews with older adults
b. Practice assessments

Numerous studies have highlighted the utilization of CHW/P practice strategies of CHW/P, adult learning theory, and tailored
and their effectiveness in helping their target populations achieve messaging, as well as evidence-based fall prevention strategies.
positive health outcomes through health education, promotion, The structured engagement of CHW/P during the development
and outreach (17–31). CHW/P are effective in these roles due to process had many benefits. First, the included messages, content,
their cultural similarity and understanding of the population they and format of the curricula were relevant, acceptable, and com-
serve and the subsequent trust that residents have in them (17– prehendible for the CHW/P and CHW/P Instructors. Second, the
31). Specifically, CHW/P-led educational interventions have led messages, content, and format were appropriate for older adults
to increased participant self-efficacy, knowledge, and adoption of who would be reached by the fall prevention activities. Third, by
preventive behaviors (38–43). Further, studies have demonstrated vetting the modules among the intended community, the goal of
the effectiveness of CHW/P in providing social support to help widespread adoption, dissemination, and sustainability was more
participants adopt behavior change (38–40). One intent of this realistic and obtainable. A potential limitation of the CHW/P cur-
project was to build on the literature that has demonstrated the riculum on fall prevention among older adults is that the impact
effectiveness of CHW/P in educational interventions to improve of the curriculum on reducing falls and injuries caused by falls
knowledge and adoption of behavior changes by the target pop- by older adults could depend on the actual implementation of
ulation to include fall prevention education and promotion for the curriculum – relying on the capacity of CHW/P programs,
older adults, which previously was a gap in the literature, given the trainers, and employers – rather than on the actual curriculum.
significant burden and cost of falls by older adults in the U.S. The project team anticipated this limitation and attempted to
An innovative aspect of the study was to actively engage CHW/P address this potential limitation through detailed instructions,
in all stages of the curriculum development, which included pilot tools, and resources within the curriculum and through providing
testing, refining, and implementation. The curricula utilized best additional technical assistance upon request to CHW/P, CHW/P

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 209 | 242
St. John et al. Training CHWs in fall prevention

instructors or trainers, and CHW/P employers on how to utilize Katharine Nimmons who provided administrative support during
and implement the curriculum with CHW/P and older adults. this project.
The future directions of this project as a translated curriculum
in an online format has great potential to broaden its reach and fur- REFERENCES
ther disseminate the modules – particularly in rural areas where 1. Stevens JA, Corso PS, Finkelstein EA, Miller TR. The costs of fatal and nonfatal
CHW/P training programs are scarce and communities have a falls among older adults. Inj Prev (2006) 12:290–5. doi:10.1136/ip.2005.011015
2. Alexander BH, Rivara FP, Wolf ME. The cost and frequency of hospitalization
greater population of older adults. Key will be to assess the ease of
for fall–related injuries in older adults. Am J Public Health (1992) 82(7):1020–3.
use, esthetics, and platform selected for online content to ensure doi:10.2105/AJPH.82.7.1020
CHW/P can successfully access and complete the online modules 3. Stevens JA, Dellinger AM. Motor vehicle and fall related deaths among older
and revisit resources and training materials. Careful and strategic Americans 1990–98: sex, race, and ethnic disparities. Inj Prev (2002) 8:272–5.
selection of partners for promoting and supporting this internet- doi:10.1136/ip.8.4.272
4. Jager TE, Weiss HB, Coben JH, Pepe PE. Traumatic brain injuries evaluated in
based curricula can substantially impact uptake and utilization. U.S. emergency departments, 1992–1994. Acad Emerg Med (2000) 7(2):134–40.
The project team members will rely on existing partners and col- doi:10.1111/j.1553-2712.2000.tb00515.x
laborators to broadly disseminate the self-paced, online training 5. Faul M, Xu L, Wald MM, Coronado VG. Traumatic Brain Injury in the United
modules available for CHW/P and CHW/P Instructors in English States: Emergency Department Visits, Hospitalizations, and Deaths. Atlanta, GA:
or Spanish. Centers for Disease Control and Prevention, National Center for Injury Preven-
tion and Control (2010).
6. Thomas KE, Stevens JA, Sarmiento K, Wald MM. Fall-related traumatic brain
CONCLUSION injury deaths and hospitalizations among older adults – United States, 2005.
This perspective paper describes the process of evaluating the J Safety Res (2008) 39(3):269–72. doi:10.1016/j.jsr.2008.05.001
CHW/P training curriculums, identifying a gap in the curricu- 7. Scott JC. Osteoporosis and hip fractures. Rheum Dis Clin North Am (1990)
16(3):717–40.
lum, and then the process of developing a CHW/P curriculum
8. U.S. Department of Health and Human Services Health Resources and Ser-
that specifically targets management of fall prevention. The goal vices Administration on Aging. Aging Statistics. (2014). Available from: http:
of the curriculum is to equip and mobilize CHW/P so they can //www.aoa.gov/AoARoot/Aging_Statistics/index.aspx
play a key role in joining other public health professionals in the 9. Gillespie LD, Robertson MC, Gillespie WH, Sherrington C, Gates S, Clemson
fight to prevent and reduce injuries among older adults due to LM, et al. Interventions for preventing falls in older people living in the com-
munity. Cochrane Database Syst Rev (2012) 9:CD007146.
falling. In other areas, CHW/Ps have proven effective in helping
10. Moyer VA. Prevention of falls in community-dwelling older adults: U.S. Preven-
individuals adopt preventive behaviors through educational inter- tive Services Task Force Recommendation Statement. Ann Intern Med (2012)
ventions (38–43). The curriculum development was the first step 157(3):197–204. doi:10.7326/0003-4819-157-8-201210160-00533
in determining if a similar model could be used for fall prevention. 11. Michael YL, Lin JS, Whitlock EP, Gold R, Fu R, O’Connor E, et al. Interven-
The process we undertook validated that CHW/Ps had a need for tions to Prevent Falls in Older Adults: An Updated Systematic Review. Rockville,
MD: Agency for Healthcare Research and Quality (US) (2010). Available from:
this type of curriculum, were interested in the curriculum, and http://www.ncbi.nlm.nih.gov/books/NBK51683/
could improve their knowledge by participating in the curricu- 12. Noonan RK, Sleet DA, Stevens JA. Closing the gap: a research agenda to acceler-
lum. Engaging CHW/P remains especially vital in that CHW/P ate the adoption and effective use of proven older adult fall prevention strategies.
have a unique opportunity to reach out to older adults who may J Safety Res (2011) 42(6):427–30. doi:10.1016/j.jsr.2010.12.002
otherwise be neglected and have an increased risk for falling. 13. Kaniewski M, Stevens JA, Parker EM, Lee R. An introduction to the Centers for
Disease Control and Prevention’s efforts to prevent older adult falls. Front Public
Health (2015) 2:119. doi:10.3389/fpubh.2014.00119
ACKNOWLEDGMENTS 14. U.S. Department of Health and Human Services Health Resources and
The authors would like to acknowledge the funder of this project – Services Administration Bureau of Health Professions. Community Health
the Centers for Disease Control Prevention Research Centers. The Worker National Workforce Study. (2007). Available from: http://bhpr.hrsa.gov/
healthworkforce.chw/
research was supported with funding support from the Centers for
15. American Public Health Association. Community Health Worker Section. (2013).
Disease Control and Prevention under Cooperative Agreement Available from: http://www.apha.org/membergroups/sections/aphasections/
Number 1U48 DP001924 with the Texas A&M Health Science chw/
Center School of Rural Public Health Center for Community 16. Texas Department of State Health Services and health and Human Services
Health Development and Cooperative Agreement Number U48- Commission. Report to the Texas Legislature, as Required by H.B. 2610, 82 nd
Legislature, Regular Session. (2012). Available from: http://www.dshs.state.tx.
DP001944 with the University of North Carolina at Chapel Hill
us/mch/chw.shtm
Prevention Research Center. The funding for the fall prevention 17. Viswanathan M, Kraschnewski J, Nishikawa B, Morgan LC, Thieda P, Hon-
CHW/P curriculum was provided by the Centers for Disease Con- eycutt A, et al. Outcomes of community health worker interventions. Evi-
trol and Prevention Special Interest Project 12-058. The findings dence Report/Technology Assessment, Number 181, Agency for Healthcare Research
and conclusions in article are those of the author(s) and do not and Quality. (2009). Available from: http://www.ahrq.gov/downloads/pub/
evidence/pdf/comhealthwork/comhwork.pdf
necessarily represent the official position of the Centers for Dis- 18. Staten LK, Scheu LL, Bronson D, Peña V, Elenas J. Pasos Adelante: the effective-
ease Control and Prevention. The authors would also like to thank ness of a community-based chronic disease prevention program. Prev Chronic
the CHW/P who provided input into the curriculum development Disease (2005) 2(1):A18.
and participated in the pilot-test and to acknowledge the following 19. Crowe JL, Keifer MC, Salazar MK. Striving to provide opportunities for
CHW/P and instructors for their contribution in the development farm worker community participation in research. J Agric Saf Health (2008)
14(2):205–19. doi:10.13031/2013.24351
modules: Denise Adame, Aracely Garibay, Lupita Garza, Leticia 20. Larkey LK, Gonzalez JA, Mar LE, Glantz N. Latina recruitment for cancer preven-
Gutierrez, Dinorah Martinez, and Paula Saldaña. The authors also tion education via community based participatory research strategies. Contemp
want to express their appreciation to Stephanie Bomberger and Clin Trials (2009) 30(1):47–54. doi:10.1016/j.cct.2008.08.003

www.frontiersin.org April 2015 | Volume 2 | Article 209 | 243

St. John et al. Training CHWs in fall prevention

21. Ramos IN, May M, Ramos KS. Environmental health training of promo- 38. Hansen LK, Feigl P, Modiano MR, Lopez JA, Sluder SE, Moinpour CM, et al. An
toras in colonias along the Texas-Mexico border. Am J Public Health (2001) educational program to increase cervical and breast cancer screening in Hispanic
91(4):568–70. doi:10.2105/AJPH.91.4.568 women. Cancer Nurs (2005) 28(1):47–53. doi:10.1097/00002820-200501000-
22. Levine DM, Becker DM, Bone LR. Narrowing the gap in health status of minor- 00007
ity populations: a community-academic medical center partnership. Am J Prev 39. Larkey L. Las Mujeres Saludables: reaching Latinas for breast, cervical and col-
Med (1992) 8(5):319–23. orectal cancer. J Community Health (2006) 31(1):69–77. doi:10.1007/s10900-
23. Witmer A, Seifer SD, Finocchio L, Leslie J, O’Neil EH. Community health work- 005-8190-2
ers: integral members of the health care work force. Am J Public Health (1995) 40. Livaudais JC, Coronado GD, Espinoza N, Islas I, Ibarra G, Thompson B.
85(8 Pt 1):1055–8. doi:10.2105/AJPH.85.8_Pt_1.1055 Educating Hispanic women about breast cancer prevention: evaluation of a
24. Balcázar H, Alvarado M, Cantu F, Pedregon V, Fulwood R. A promotora de home-based promotora-led intervention. J Womens Health (Larchmt) (2010)
salud model for addressing cardiovascular disease risk factors in the US-Mexico 19(11):2049–56. doi:10.1089/jwh.2009.1733
border region. Prev Chronic Dis (2009) 6(1):A02. 41. Luque JS, Mason M, Reyes-Garcia C, Hinojosa A, Meade CD. Salud es Vida:
25. Baquero B, Ayala GX, Arredondo EM, Campbell NR, Slymen DJ, Gallo L, et al. development of a cervical cancer education curriculum for promotora outreach
Secretos de la Buena Vida: processes of dietary change via a tailored nutrition with Latina farmworkers in rural southern Georgia. Am J Public Health (2010)
communication intervention for Latinas. Health Educ Res (2009) 24(5):855–66. 101(12):2233–5. doi:10.2105/AJPH.2011.300324
doi:10.1093/her/cyp022 42. Moralzes EA, Rao SP, Livaudais JC, Thompson B. Improving knowledge and
26. Elder JP, Ayala GX, Slymen DJ, Arredondo EM, Campbell NR. Evaluating screening for colorectal cancer among Hispanics: overcoming barriers through
psychosocial and behavioral mechanisms of change in a tailored commu- a promotora-led home-based educational intervention. J Cancer Edu (2012)
nication intervention. Health Edu Behav (2009) 36(2):366–80. doi:10.1177/ 27:533–9. doi:10.1007/s13187-012-0357-9
1090198107308373 43. O’Brien MJ, Halbert CH, Bixby R, Pimentel S, Shea JA. Community health
27. Heisler M, Spencer M, Forman J, Robinson C, Shultz C, Palmisano G, et al. Par- worker intervention to decrease cervical cancer disparities in Hispanic women.
ticipants’ assessments of the effects of a community health worker intervention J Gen Intern Med (2010) 25(11):1186–92. doi:10.1007/s11606-010-1434-6
on their diabetes self-management and interactions with healthcare providers.
Am J Prev Med (2009) 37(6 Suppl 1):S270–9. doi:10.1016/j.amepre.2009.08.016
28. McCloskey J. Promotores as partners in a community-based diabetes interven- Conflict of Interest Statement: The authors declare that the research was conducted
tion program targeting Hispanics. Fam Community Health (2009) 32(1):48–57. in the absence of any commercial or financial relationships that could be construed
doi:10.1097/01.FCH.0000342816.87767.e6 as a potential conflict of interest.
29. Medina A, Balcazar H, Hollen ML, Nkhoma A, Mas FS. Promotores de Salud:
educating Hispanic communities on heart-healthy living. Am J Health Edu This paper is included in the Research Topic, “Evidence-Based Programming for Older
(2007) 38(4):194–202. doi:10.1080/19325037.2007.10598970 Adults.” This Research Topic received partial funding from multiple government and
30. Postma J, Karr C, Kieckhefer G. Community health workers and environmental private organizations/agencies; however, the views, findings, and conclusions in these
interventions for children with asthma: a systematic review. J Asthma (2009) articles are those of the authors and do not necessarily represent the official position
46(6):564–76. doi:10.1080/02770900902912638 of these organizations/agencies. All papers published in the Research Topic received
31. Deitrick LM, Paxton HD, Rivera A, Gertner EJ, Biery N, Letcher AS, et al. Under- peer review from members of the Frontiers in Public Health (Public Health Education
standing the role of the promotora in a Latino diabetes education program. Qual and Promotion section) panel of Review Editors. Because this Research Topic repre-
Health Res (2010) 20(3):386–99. doi:10.1177/1049732309354281 sents work closely associated with a nationwide evidence-based movement in the US,
32. Yannacci J, Roberts K, Ganju V. Principles from Adult Learning Theory, Evidence- many of the authors and/or Review Editors may have worked together previously in
Based Teaching, and Visual Marketing: What are the Implications for Toolkit some fashion. Review Editors were purposively selected based on their expertise with
Development?. Center for Mental Health Quality and Accountability, NRI, Inc. evaluation and/or evidence-based programming for older adults. Review Editors were
(2006). Available from: http://ebp.networkofcare.org/uploads/Adult_Learning_ independent of named authors on any given article published in this volume.
Thoery_2497281.pdf
33. Green ML, Ellis PJ. Impact of an evidence-based medicine curriculum based on
adult learning theory. J Gen Intern Med (1997) 12(12):742–50. doi:10.1046/j. Received: 08 September 2014; accepted: 10 October 2014; published online: 27 April
1525-1497.1997.07159.x 2015.
34. Neame RLB, Powis DA. Toward independent learning: curricular design for Citation: St. John JA, Shubert TE, Smith ML, Rosemond CA, Howell DA, Beaudoin
assisting students to learn how to learn. J Med Educ (1991) 56:886–93. CE and Ory MG (2015) Developing an evidence-based fall prevention curriculum for
35. Comings JP, Soricone L, Santos M. An Evidence-Based Adult Education Program community health workers. Front. Public Health 2:209. doi: 10.3389/fpubh.2014.00209
Model Appropriate for Research. NSCALL Occasional Paper. Cambridge, MA: This article was submitted to Public Health Education and Promotion, a section of the
National Center for the Study of Adult Learning and Literacy (2006). Available journal Frontiers in Public Health.
from: http://www.nscall.net/fileadmin/resources/research/op_comings4.pdf Copyright © 2015 St. John, Shubert , Smith, Rosemond, Howell, Beaudoin and Ory.
36. Kreuter MW, Farrell D, Olevitch L, Brennan L. Tailored Health Messages: Cus- This is an open-access article distributed under the terms of the Creative Commons
tomizing Communication with Computer Technology. Mahwah, NJ: Lawrence Attribution License (CC BY). The use, distribution or reproduction in other forums is
Erlbaum (1999). permitted, provided the original author(s) or licensor are credited and that the origi-
37. Foerster SB, Kizer KW, Disogra LK, Bal DG, Krieg BF, Bunch KL. California’s “5 nal publication in this journal is cited, in accordance with accepted academic practice.
a day – for better health!” campaign: an innovative population-based effort to No use, distribution or reproduction is permitted which does not comply with these
effect large-scale dietary change. Am J Prev Med (1995) 11(2):124–31. terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 209 | 244
 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00182

Effects of an evidence-based falls risk-reduction program

on physical activity and falls efficacy among oldest-old
adults
Jinmyoung Cho 1,2 *, Matthew Lee Smith 3 , SangNam Ahn 4 , Keonyeop Kim 5 , Bernard Appiah 6 and
Marcia G. Ory 2
1
Center for Applied Health Research, Baylor Scott and White Health, Temple, TX, USA
2
Department of Health Promotion and Community Health Science, Texas A&M Health Science Center, School of Public Health, College Station, TX, USA
3
Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA, USA
4
Division of Health Systems Management and Policy, School of Public Health, The University of Memphis, Memphis, TN, USA
5
Department of Preventive Medicine, Graduate School of Public Health, Kyungpook National University, Daegu, South Korea
6
Department of Public Health Studies, Texas A&M Health Science Center, School of Public Health, College Station, TX, USA

Edited by: Purpose of the study:The current study was designed to examine changes in falls efficacy
Sanjay P. Zodpey, Public Health
and physical activities among oldest-old and young-old participants in a falls risk-reduction
Foundation of India, India
program called a matter of balance/volunteer lay leader model.
Reviewed by:
Suresh Narayanrao Ughade, Design and methods: An oldest-old group (aged 85 years and older; n = 260) and a young-
Government Medical College, India
Muni Rubens, Florida International
old group (aged between 65 and 84 years old; n = 1,139) in Texas with both baseline and
University, USA post-intervention measures were included. Changes in Falls Efficacy Scale scores and
*Correspondence: weekly physical activity levels were examined from baseline to post-intervention. Repeated
Jinmyoung Cho, Center for Applied measures analysis of covariance were employed to assess program effects on falls efficacy.
Health Research, Baylor Scott and
White Health, Department of Health Results: Results showed significant changes in falls efficacy from baseline to post-
Promotion and Community Health intervention, as well as a significant interaction effect between time (baseline and
Science, Texas A&M Health Science
post-intervention) and physical activity on falls efficacy.
Center, School of Public Health, 2401
S 31st Street, MS-01-501, Temple, TX Implications: Findings from this study imply the effectiveness of evidence-based programs
76508, USA
e-mail: jinmyoung.cho@
for increasing falls efficacy in oldest-old participants. Future implications for enhancing
sph.tamhsc.edu; [email protected] physical activities and reducing fear of falling for oldest-old adults are discussed.
Keywords: oldest-old adults, falls efficacy, falls risk-reduction program

INTRODUCTION had fallen at least once during the previous 6 months or restricted
The population of oldest-old adults – or those 85 years and older – their daily activities or both because of a fear of falling (13).
is one of the fastest growing segments of the American population Heterogeneity exists in levels of physical activity among oldest-
and is estimated to increase from 5.7 million to 19 million by old adults, despite the lower overall physical activity levels, sug-
2050 (1). Despite growth among this population segment, rela- gesting the value in identifying modifiable factors associated with
tively less attention is given to the oldest-old population compared higher activity levels. A sense of efficacy, particularly falls efficacy –
to people aged younger than 85 years old (2). Attention to health “the degree of confidence in performing common daily activities
status among oldest-old adults is critical because approximately without falling” (14) (p. M141) – has been found as a significant
half of those in this age group experience limitations in function- factor for physical activity among older adults. Higher levels of
ing, which not only impacts their health and independence (3) but efficacy have been related to faster gait speeds (15, 16) as well as
also has societal implications on escalating health care utilization lower levels of fear of falling (17); furthermore, physical activity
and costs (4, 5). interventions have shown significant positive effects on physical
Substantial research has identified functional and behavioral performance related to efficacy (18–20).
factors associated with loss of independence among the aging pop-
ulation (6, 7). Less physical activity, for example, is increasingly A MATTER OF BALANCE (AMOB) FALLS RISK-REDUCTION PROGRAM
seen as a major contributor to health deterioration and mortal- Previous research indicates that falls risks can be ameliorated, espe-
ity, even among oldest-old adults (8, 9). Lower levels of physical cially through increases in physical activities, which are combined
activity contribute to increased number of medical comorbidities with behavioral strategies to help older adults prevent or man-
in oldest-old populations (10, 11). In addition, anxiety or fear of age falls (21–25). Behavioral interventions have been identified to
falling is related to risk for subsequent falls and limited physical improve falls efficacy (12, 26, 27). A matter of balance (AMOB) is
activity among older adults (12, 13). For instance, in a sample of an evidence-based program to reduce falls risk among older adults
adults aged 70 years and older living in a community, over half based on cognitive restructuring methods (28). Established at the

www.frontiersin.org April 2015 | Volume 2 | Article 182 | 245

Cho et al. Falls efficacy among oldest-old adults

Roybal Center for Enhancement of Late-Life Function at Boston

University, the original AMOB program was tested through a ran- Covariates:
domized clinical trial (RCT) (22). The major outcome variables Age
Sex
of the program included significant improvements of perceived Living status
capacity to manage the risk of falling and confidence in everyday Ethnicity
Education
activities without falling. Two primary aims of the AMOB included Number of chronic conditions Physical activity
(a) reducing fear of falling and (b) increasing physical, social, and
functional activity (22).
Because the goal of AMOB/volunteer lay leader (VLL) is
AMOB/VLL falls
to build falls efficacy and encourage physical activities, many risk-reduction
Falls efficacy
researchers have examined the effects of AMOB/VLL and found program
improvement in overall health status, as well as falls efficacy among
older adults (29–33). For example, Ory and colleagues (29) found FIGURE 1 | Conceptual model.
that Texas AMOB/VLL participants showed significant improve-
ments in falls efficacy, physical activity, and normal everyday rou-
tines. These results are consistent with other studies that included MATERIALS AND METHODS
participants from South Florida and South Carolina (34, 35). Ull- INTERVENTION ELEMENTS
mann and colleagues (35) found that South Carolina participants A matter of balance/volunteer lay leader entails a lay leader model
showed greater confidence in managing falls and performing activ- and is widely disseminated in the health and aging services sec-
ities of daily living, as well as improvements in functional mobility. tors (44). The intervention is typically delivered by a pair of
In addition to short-term benefits in behavioral outcomes from trained lay leaders, known as coaches (30, 32, 44). AMOB/VLL
the program, Smith et al. (36) observed significant yet modest was designed to modify fall-related factors, such as behaviors, atti-
improvements in falls efficacy were maintained 6 months after tudes, and environmental aspects that increase falling risk among
intervention. Furthermore, improvements in falls efficacy and older adults (45). Standardized AMOB/VLL workshops take place
physical activity have been identified in studies examining the at licensed delivery sites and are facilitated by certified coaches
rurality of participants’ residence, participant ethnicity, and the to ensure program fidelity (44). As facilitators, these lay leaders
influence of class size and session attendance on health outcomes. use an extensively detailed training manual and two instructional
Rural residents, Hispanic participants, and participants in smaller videos (32). The AMOB/VLL intervention consists of eight 2-h
size classes with higher attendances showed significant improve- sessions either once a week for 8 weeks or twice a week for 4 weeks
ments in falls efficacy and physical activity compared to their own (30). Early sessions focus on individual behavior and mindsets
counterparts (30, 37, 38). with an emphasis on decreasing the fear of falling and increas-
The extant studies documenting improvements associated with ing participants’ confidence to prevent falls; later sessions focus
AMOB/VLL typically include a full range of older participants on environmental aspects, so leaders assist participants to change
(e.g., all those 65 years and older). Scant research has examined their physical surroundings to reduce risk factors for falling and
benefits in falls efficacy and physical activities uniquely among learn exercises to increase balance and strength (30, 46, 47).
oldest-old adults (39). A call has been raised to examine those Because the intervention processes focus on building fall-
aged 85 years and older as a separate age group (e.g., a forth age) related self-efficacy and setting realistic goals for increasing activ-
because of the unique nature and challenges faced by those in ity, the intervention includes a variety of components, such as lec-
this subgroup (40–42). Age-related stereotypes about the benefits tures, group discussions, mutual problem solving, role-play activ-
of health promotion programs for seniors (43), however, might ities, exercise training, assertiveness training, and home assign-
be a barrier to examining physical activities programs among the ments. A standard definition of a “successful” class completion
oldest-old adults (39). Despite current knowledge of the poten- (i.e., attending five or more of the eight sessions) and an ideal class
tial effectiveness of behavioral interventions across the life span size (i.e., 8–12 older adults) has been established (38).
(43), few studies have focused specifically on examining the joint
influence of falls efficacy and physical activities in the oldest-old PARTICIPANTS
population. As noted in our previous research (31, 37), a total of 3,276 partici-
The purposes of this study were, therefore, to (a) assess the pants enrolled in the Texas AMOB/VLL Falls-Prevention Interven-
changes in falls efficacy and physical activity from baseline to post- tion between September 2007 and April 2009 through local area
intervention among oldest-old adults and (b) examine the effect agencies on aging (AAA) and other community-based organiza-
of the interaction between improvement of physical activity from tions. Eighteen AAA regions offered 243 classes during that period.
baseline to post-intervention on falls efficacy, with a targeted focus The authors obtained Institutional Review Board (IRB) approval
on oldest-old participants. A conceptual model for this study is at Texas A&M University to analyze secondary data on program
shown in Figure 1. This model depicts the AMOB/VLL falls risk- participants and the effectiveness of the program.
reduction program as a predictor for changes in physical activity
and falls efficacy. In addition, improvement of physical activity PROCEDURES AND INSTRUMENTS
acts as a moderator between falls efficacy and falls risk-reduction The same instruments were used at baseline and after completion
program. of an intervention. A paper-based questionnaire included 28 items.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 182 | 246
Cho et al. Falls efficacy among oldest-old adults

The survey items included four different formats (i.e., Likert-type DATA ANALYSIS
scales, yes/no, closed response, and open ended). Public health Three different analyses were performed. In univariate analy-
and aging research experts who established a common database ses, frequencies were calculated for personal characteristics, falls
for evaluation of program effectiveness in a national consortium efficacy, and physical activity. In bivariate analyses, Pearson’s chi-
of studies helped guide the selection of the measures (48). Par- square tests were conducted to examine the goodness of fit for
ticipants took approximately 15 min to complete the baseline and frequency distributions and the independence between categorical
post-intervention instruments, respectively. participants’ characteristics (e.g., sex, living status) (49). Multivari-
ate analyses were also performed to obtain adjusted estimates. SAS
MEASURES (ver. 9.2, 2010) Proc Mixed (50) procedures were used when con-
Personal characteristics ducting two repeated measures analysis of covariance (ANCOVA)
Six personal characteristic variables were used: age, sex, to calculate the adjusted mean changes in falls efficacy scale scores
race/ethnicity, education, living status, and number of chronic by physical activity groups (i.e., those who showed improvement
conditions. Age was treated as a continuous variable and was vs. those with no-improvement). As far as the measurement of
based on a participant’s birth date. Sex was scored 1 if the par- physical activity, many previous studies on the program have
ticipant is female. Race or ethnicity was scored 0 if the par- showed the intervention effects on physical activities (29–33).
ticipant is non-Hispanic White, and 1 if non-White. Educa- When we tested a model including physical activity in this study
tion was scored 0 if a participant’s highest level of education (not shown in this study), this independent variable showed sig-
was less than high school graduation, 1 if graduated from high nificant effects on falls efficacy in both age groups after controlling
school, and 2 if more than a high school education. Living sta- for covariates (i.e., slope β = 0.30, p < 0.001 for oldest-old group;
tus was scored 0 if participants lived alone and 1 if they lived slope β = 0.28, p < 0.001 for young-old group). Assuming sig-
with others. The self-reported number of chronic conditions nificant effects of physical activity on falls efficacy, the physical
ranged from zero to seven and was considered as a continuous activity levels were purposively categorized into two groups to
variable. see the interaction effect between levels of physical activity (i.e.,
improvement group vs. no-improvement) and the intervention.
Falls efficacy scale In other words, independent variables included time (two time
Falls efficacy was assessed with the scale developed by Tennstedt points: baseline and post-intervention) and two levels of physical
et al. (22). The scale consists of five items that measure partic- activity (improvement vs. no-improvement) worked as a moder-
ipants’ perceived ability to manage risk of falls or actual falls ator. Age, sex, race/ethnicity, education, living status, and number
(22). Participants were asked to rate the following statements: of chronic conditions at baseline were also included as covari-
(1) you can find a way to get up if you fall, (2) you can find a ates. Many methodological experts of longitudinal studies have
way to reduce falls, (3) you can protect yourself if you fall, (4) advised centering time-varying covariates (51, 52); therefore, we
you can increase your physical strength, and (5) you can become centered age and the number of chronic conditions before con-
more steady on your feet. Ratings were used with a four-point ducting advanced analyses. Specifically, we examined whether time
Likert scale: 1 = not sure at all, 2 = not very sure, 3 = somewhat (baseline and post-intervention) and two levels of physical activity
sure, and 4 = absolutely sure. Cronbach’s α was 0.87 for the five influence the changes in falls efficacy. In addition, we examined the
items of falls efficacy. Scores ranged from 5 to 20 with higher interaction effect between time (baseline and post-intervention)
scores indicating higher levels of managing risk of falls. These falls and physical activity groups (improvement vs. no-improvement)
efficacy scores were collected from participants at baseline and to detect the difference in change of falls efficacy. Covariates
post-intervention. and one of the independent variables, time (baseline and post-
intervention), were included in the first model. Two levels of
Physical activity physical activity (improvement vs. no-improvement) and inter-
Physical activity was measured using one item that asked partic- action term between time (baseline and post-intervention) and
ipants to report the number of days they were physically active physical activity groups were included in the second model.
in the previous 7 days (i.e., scores could range from 0 to 7 days).
Participants were given examples of physical activities (e.g., brisk RESULTS
walking, bicycling, vacuuming, gardening, or anything else that SAMPLE DESCRIPTION
causes one to breathe faster); however, the actual physical activ- As shown in Figure 2, a total of 3,276 participants enrolled in
ities in which the participant engaged were not independently the Texas AMOB/VLL fall risk-reduction program. About 30% of
documented. Physical activity was measured at baseline and post- the total participants (n = 978) who were younger than 65 years
intervention. Furthermore, the change in the number of days from old were excluded. Among those who met our inclusion cri-
baseline to post-intervention was assessed. Improvement indi- teria (n = 2,298), 899 participants (39.1%) did not complete
cates a greater number of days at post-intervention than baseline; post-intervention survey instruments. Only those who completed
no-improvement indicates a same or less number of days at post- both baseline and post-intervention assessments (n = 1,399) were
intervention when compared with baseline. Based on the change included in this study. Those aged 85 years and older were cate-
in number of days from baseline to post-intervention, the authors gorized into the oldest-old group as a target group for this study
defined two groups for physical activity: improvement (scored 1) (n = 260); those aged between 65 and 84 years represent young-old
and no-improvement (scored 0). group as a comparison age group (n = 1,139).

www.frontiersin.org April 2015 | Volume 2 | Article 182 | 247

Cho et al. Falls efficacy among oldest-old adults

Table 1 | Participant characteristics at baseline.

Assessed for eligibility
(n = 3,276) Oldest-old Young-old F /χ2
group group
Excluded (n = 978)
(n = 260) (n = 1,139)
Not meeting inclusion criteria:
younger than 65 yrs.(n= 2,804)
Agea 87.84 (± 2.84) 76.43 (± 5.24) 1,156.67***
Sex 1.59
Meeting inclusion criteria (n = 2,298)
Male 59 (23.6) 216 (20.0)
Female 191 (76.4) 863 (80.0)
Living status 26.53***
Not completed post- Completed post-
intervention (n = 899) intervention (n = 1,399) Living alone 177 (70.2) 575 (52.4)
Compared: (n = 1,139) Living with one or more 75 (29.8) 523 (47.6)
Young-old group: others
aged 65–84 yrs.
Ethnicity 21.21***
Target Group (n = 260) White not Hispanic 212 (86.5) 775 (72.5)
Oldest-old group: African-American 21 (8.6) 202 (18.9)
aged 85+ yrs.
Hispanic 12 (4.9) 92 (8.6)
Education levels 0.21
FIGURE 2 | Diagram for participants inclusion. No statistical differences Less than high school 45 (17.6) 196 (17.6)
found in sex, age, and the number of chronic conditions. Statistical
High school graduate 69 (27.0) 285 (25.6)
differences found between those who completed and did not complete
post-intervention for African-Americans, Hispanics, and those with less More than high school 142 (55.5) 632 (56.8)
than high school education. Number of sessions
attended
Less than 5 sessions 14 (5.4) 47 (4.1) 0.78
In addition, we examined characteristics associated with pro- 5–8 sessions 246 (94.6) 1, 087 (95.9)
gram completion. Although no significant differences were appar- Number of chronic 1.64 (±1.14) 1.75 (±1.20) 1.58
ent by sex, age, or the number of chronic conditions at base- conditionsa
line (not shown in tables), a significant race/ethnicity difference Ave. days of physically 3.55 (±2.56) 3.46 (±2.29) 0.23
(p < 0.05) was identified between those who completed post- activea (0–7)
intervention (inclusion group; n = 1,399) and those who did not Ave. score of falls efficacy 13.58 (±3.92) 14.42 (±3.65) 9.37**
complete post-intervention (exclusion group; n = 899). African- scalea (5–20)
American participants represented over 25% and Hispanic par-
ticipants represented 4.6% among those who did not complete a
Means (±SD) reported for continuous variables.
the baseline and post-intervention assessment (exclusion group); **p < 0.01, ***p < 0.001.
African-American participants constituted 17.0% and Hispanic
participants constituted 7.9% among those who completed the 80% were female, and about half of the comparison group (52.4%)
baseline and post-intervention assessment (inclusion group). Fur- lived alone. The majority of the group (72.5%) was non-Hispanic
thermore, a significant education difference (p < 0.05) occurred. White, and about half the group (56.8%) had an education above
Those who had less than high school graduation constituted 17.6% high school. Over 90% of the group (95.9%) attended five and
among those who completed both baseline and post-intervention more sessions. The average number of chronic conditions was
(inclusion group), but those who had less than high school 1.75 (SD = 1.20). Participants in the young-old group engaged
graduation constituted 26.4% among the exclusion group. in slightly less physical activities (M = 3.46, SD = 2.29). Further-
more, the average falls efficacy score of this group was 14.42
BASELINE CHARACTERISTICS (SD = 3.65).
Table 1 shows study participants’ characteristics at baseline for
those having both baseline and post-assessment data between CHANGE IN FALLS EFFICACY
oldest-old and young-old group. For the oldest-old group, the Table 2 presents the results of repeated measures ANCOVA in
mean age was 87.84 (SD = 2.84) years old; 76.4% were female, and the oldest-old group and the young-old group. Two models were
70.2% of the group lived alone. The majority of the group (86.5%) compared for each group in Table 2. For the oldest-old group, time
was non-Hispanic White, and about half the group (55.5%) had was statistically significant for change of falls efficacy from base-
an education above high school. Over 90% of the group (94.6%) line to post-intervention in Model 1. In other words, the mean
attended five or more workshop sessions. The average number scores of falls efficacy scores significantly increased from base-
of self-reported chronic conditions was 1.64 (SD = 1.14). Partic- line to post-intervention (slope: β = 1.98, p < 0.001). In addition,
ipants in the oldest-old group engaged in physical activities on improvement of days of physical activities and the interaction term
three or more days on average (M = 3.55, SD = 2.56). In addition, between time (baseline and post-intervention) and physical activ-
their average falls efficacy score was 13.58 (SD = 3.92). For the ities were included in Model 2. Both physical improvement and
young-old group, the mean age was 76.43 (SD = 5.24) years old; the interaction term were significant (slopes: β = 1.32, p < 0.05,

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 182 | 248
Cho et al. Falls efficacy among oldest-old adults

Table 2 | Models for changes in falls efficacy among oldest-old group and young-old group.

Predictors Oldest-old group (n = 190) Young-old group (n = 1,015)

Model 1 Model 2 Model 1 Model 2

Intercept 15.50 14.98 11.96 12.40

Covariates
Age −0.17 (0.07)* −0.16 (0.07)* −0.01 (0.01)*** −0.07 (0.01)***
Sex (female = 1) −0.99 (0.51) −0.93 (0.51) −0.62 (0.22)** −0.57 (0.51)*
Living status (living alone = 0) −0.68 (0.48) −0.63 (0.49) 0.16 (0.17) 0.19 (0.18)
Ethnicity (White not Hispanic = 1) −0.78 (0.69) −0.66 (0.69) 0.24 (0.22) 0.08 (0.24)
Education (less than HS = 1) 0.33 (0.29) 0.29 (0.29) 0.28 (0.12)* 0.34 (0.14)*
Number of chronic condition −0.25 (0.19) −0.22 (0.19) −0.52 (0.07)*** −0.52 (0.08)***
Time (baseline = 0) 1.98 (0.30)*** 1.33 (0.39)*** 2.03 (0.12)*** 1.71 (0.16)***
Improvement of physically active (Improved = 1) 1.32 (0.52)* 1.05 (0.42)*
Time*improvement of physically active 1.43 (0.58)* 0.73 (0.25)**

AIC (Akaike’s information criteria) 1818.3 1814.6 9640.2 8359.1

Figures shown in the table are metric coefficients and standard errors (in parentheses).
*p < 0.05, **p < 0.01, ***p < 0.001.

β = 1.43, p < 0.05, respectively). These results indicate that the

mean score of falls efficacy in improvement group was higher than
in the no-improvement group and mean scores of falls efficacy in
both improvement group and no-improvement group at baseline
were different from those at post-intervention.
Similar results of changes in falls efficacy were found in the
young-old group. Time was statistically significant for change of
falls efficacy from baseline to post-intervention in Model 1 (slope:
β = 2.03, p < 0.001); both physical improvement and the interac-
tion term were statistically significant (slopes: β = 1.05, p < 0.05,
β = 0.73, p < 0.01, respectively) in Model 2.

RELATIONSHIP BETWEEN IMPROVEMENT OF PHYSICAL ACTIVITIES

AND FALLS EFFICACY
As shown in Model 2 (Table 2), the interaction term between
time (baseline and post-intervention) and physical activity had
significant effects on falls efficacy in both oldest-old and young-
old groups. This indicates mean scores of falls efficacy in both
improvement group and no-improvement group at baseline were
different from those at post-intervention. To examine interactions,
methodologists have advised plotting the figure (53). As shown in FIGURE 3 | Falls efficacy at baseline and post-intervention by physical
Figure 3 for oldest-old group, the improvement group in physical activity groups (improvement vs. no-improvement) in oldest-old
activity had lower score of falls efficacy at baseline than the no- group.
improvement group, but after they participated in the AMOB/VLL
program, their falls efficacy score improved greater than the partic-
ipants in the no-improvement group. In other words, the improve- shown the effectiveness of the program. Most studies, however, did
ment in falls efficacy was associated with increased physical activity not differentiate oldest-old participants from general old adults.
as well as program participation among oldest-old participants. Specification of age group may contribute to a closer look at the
The young-old group also showed same trends; the improvement effectiveness of evidence-based program in falls efficacy and phys-
group in physical activity had lower score of falls efficacy at base- ical activities. The aim of this study was to examine physical and
line, but their score improved greater than the participants in the psychological benefits among oldest-old adults enrolled in the
no-improvement group (Figure 4). Texas AMOB/VLL falls risk-reduction program. First, this study
contributes to understanding of falls efficacy among oldest-old
DISCUSSION adults by extending the evidence base of the AMOB/VLL falls
Many previous studies have assessed falls efficacy and physical risk-reduction program. Our findings confirm the increased falls
activities among participants in the AMOB/VLL program and efficacy among oldest-old adults in Texas similar to that reported

www.frontiersin.org April 2015 | Volume 2 | Article 182 | 249

Cho et al. Falls efficacy among oldest-old adults

(SD = 4.71) and 15.50 (SD = 4.18), respectively. There was no

significant difference between the two falls efficacy scores at
post-intervention, t (305) = 0.198, p = 0.579.
This finding provides another significant benefit of evidence-
based programs in improving the quality of life among oldest-old
population. Most studies related to the effectiveness or benefits of
evidence-based program have focused on separate health-related
outcomes, such as health behaviors, self-efficacy, or falling or
injury rates (29, 58, 59). The results of this study, however, pro-
vide critical evidence suggesting that the AMOB/VLL program can
positively affect psychological beliefs (i.e., falls efficacy), as well
as physical activities among oldest-old participants at the same
time. One possible explanation of the synergy/doubled/combined
effect of physical and psychological improvement from the falls
risk-reduction program may be that oldest-old adults had more
barriers for physical activities than younger counterparts. Through
a systemic review of physical activity in oldest-old adults, Baert
and colleagues (60) have reported many different types of barriers,
such as physical impairment (61), weakness of physical strength
(62), being too tired (59), fear of injury or pain (63), or the belief
that older people cannot change (64). Our oldest-old participants
enrolled in the AMOB/VLL intervention may, however, overcome
FIGURE 4 | Falls efficacy at baseline and post-intervention by physical those barriers. In particular, the group that improved their physi-
activity groups (improvement vs. no-improvement) in young-old cal activities may enhance ability or strategy of prevention of falls
group.
risks and this, in turn, contribute to improve falls efficacy.

LIMITATIONS
previously for the general population of older adults (29, 31, 38, Some limitations were related to this study, despite noteworthy
44) From baseline to post-intervention, oldest-old participants findings. First, the study variables collected at baseline and post-
who enrolled in the AMOB/VLL intervention showed significant intervention were self-reported. We should consider the possibility
improvement in falls efficacy as shown in the young-old group. of recall bias because participants were asked to recall occurrences
This finding indicates that entry into the program may have a within the previous week or month (31). Second, the partici-
significant effect on changes in confidence of managing falls- pants in this study were recruited from only one geographic region
related risks from baseline to post-intervention. A few studies of the United States (i.e., Texas). Participants from more demo-
showed that the effect of psychological variables is attenuated graphically diverse states of United States or other countries might
for those over 75 years old (54, 55). Our results, however, indi- demonstrate different patterns in the change in falls efficacy. More
cate that older adults, especially those aged 85 years and older, can studies from other states and in diverse settings could contribute
improve their own self-beliefs related to risks of falling through to generalization of the results. Third, participants were not ran-
intervention (56, 57). domly assigned into the intervention, nor were a true comparison
Second, this study suggests a mechanism for overcoming psy- group included in the study design (i.e., older adults who did not
chological barriers. Our results suggest that increased physical receive the AMOB/VLL intervention). With translational research
activities contributed to improving falls efficacy among oldest- studies, the main objective is to replicate outcomes previously
old adults enrolled in an evidence-based falls risk-reduction obtained in more controlled intervention designs across different
program. From baseline to post-intervention, as noted above, groups. Hence, such translational studies are often not designed as
participants showed significant improvement in falls efficacy; RCTs (65); nevertheless, our use of a one group design in this study
however, 44% of participants who enrolled in the intervention limits our ability to definitively confirm the presence of significant
showed significant improvement in days of physical activities intervention effects between baseline and post-intervention on
in the improvement group while 56% of participants indicated falls efficacy and physical activity. As such, we recommended that
decline or same days of physical activities in the no-improvement future studies include both intervention and comparison groups to
group. At baseline, the falls efficacy scores of the improvement detect true intervention effects (e.g., RCT by specified age groups).
group (M = 12.84, SD = 4.78) were lower than those of the no- Admittedly, self-selection bias may be another limitation for this
improvement group (M = 14.17, SD = 4.35). There was a signifi- study because participants chose to enroll in the AMOB/VLL pro-
cant difference between the two falls efficacy scores, t (306) = 2.57, gram. However, our findings are similar to those reported in other
p < 0.01. This indicates that the AMOB/VLL intervention con- studies in which no treatment comparison group was used (22, 44).
tributed to differential improvement in the falls efficacy between Fourth, the single item used to measure physical activity asked
the two groups. At post-interventions, falls efficacy scores of participants to report the number of days they were physically
the improvement group and no-improvement group were 15.60 active in the previous 7 days. We were limited in our ability to

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 182 | 250
Cho et al. Falls efficacy among oldest-old adults

perform more complex analyses with this variable or weight spe- function/mood, remove barriers for physical activities, or improve
cific physical activities. In addition, because this variable simply those with specific conditions, such as dementia, are foci for future
asks the number of days physically active, not the number of min- research. This study examined an interaction effect between phys-
utes, the ability to detect change is less because the item is not very ical activity and time (from baseline to post-intervention) on
sensitive to change. Thus, we elected to measure changes in phys- falls efficacy among oldest-old adults. As a couple of differences
ical activity from baseline to post-intervention as “improved/not between those included and excluded were identified (i.e., edu-
improved.” We acknowledge that this decision may be influenced cation and ethnicity) and additional interaction effects were not
by a potential ceiling effect among those who were physically investigated in this study, we acknowledge there may have been
active upon entering the program. This may have accounted for other factors that influenced program participation and falls effi-
fewer significant improvements in physical activity to be observed cacy among these participants. More specifically, future studies
at post-intervention. If a more specified scale (e.g., measuring should explore confounding effects among participant samples
duration such as minutes per activity) or more specific items with diverse racial/ethnic backgrounds (e.g., African-American,
related to falling (exercise vs. daily living) were used, the effective- Hispanic) and differing education levels (e.g., 17.6% did not
ness of fall-reduction programs may become more pronounced. complete high school) to assess their influences on intervention
Fifth, 899 participants who did not complete post-intervention effects.
assessments were excluded from study analyses. The target group
of this study was participants aged 85 years and older; thus, we IMPLICATIONS FOR PRACTITIONERS AND POLICY MAKERS
believe a majority of missing data from participants aged from The results of this study suggest that more practical and pol-
65 to 84 years did not strongly impact our intervention findings. icy applications are needed, especially for oldest-old population.
Nevertheless, strategies are needed to raise participant retention Although the oldest-old group (i.e., over 85 years old) will form a
and assessment response rate, which can reduce a selection bias large proportion of global population in the next couple of decades
in future implementation efforts. Sixth, Figures 3 and 4 show (1, 39), few studies have been conducted on the effectiveness of
that intervention effects influenced changes in falls efficacy levels evidence-based programs for oldest-old population compared to
differently for physical activity groups, regardless of age groups. younger groups (i.e., younger than 85 years old) (67). In contrast
Because baseline levels of falls efficacy were substantially lower in to misconceptions and age-related stereotypes (39), the results of
the improvement group compared to the no-improvement group, this study suggest that systematic strategies must be employed to
the effect on improvement group participants would be expected develop falls risk-reduction programs for oldest-old adults. We
to be larger than the no-improvement group participants. Again, a recommend that falls risk-reduction programs be developed or
ceiling effect may account for the less change in falls efficacy for the modified, specifically targeting different age groups (e.g., younger
no-improvement group relative to the improvement group. Levels than 85 years old vs. 85 years old and over) and different levels of
of falls efficacy at post-intervention were similar in both groups. physical activities. This may allow oldest-old adults to gain a more
Because, regardless of age, participants in this study showed signif- powerful intervention effect and to enhance their physical activi-
icant improvements after the intervention, we acknowledge there ties and falls prevention and, in turn, may contribute to reducing
may be other extraneous effects that were not captured in this medical expenses for falls; furthermore, staff from nursing homes
study. Future researchers may elect to collect a more encom- or senior centers as well as health professionals could be trained
passing set of measures to assess the complex factors associated to develop appropriate ways to make environments more physical
with falls efficacy improvement among participants. Finally, only activity friendly for oldest-old residents in long-term care facilities.
a short-term assessment of this intervention program was con- To summarize, findings from the present study are unique
ducted (e.g., at 8 weeks post-intervention initiation). The study in that they show simultaneous physical and psychological ben-
outcomes may be more robust if participants were studied for efits of AMOB/VLL among oldest-old participants. This study
6 months or 1 year (66). re-emphasizes the critical effectiveness of an evidence-based fall
risk-reduction program on oldest-old participants by increasing
IMPLICATIONS FOR RESEARCH their levels of physical activity and falls efficacy. Identifying char-
The findings from this study have considerable implications for acteristics of oldest-old participants who benefit from this inter-
future research on aging studies. Most notably, the inclusion of vention has the potential to enhance its effectiveness and inform
the young-old group in this study emphasizes the magnitude of the development of systematic strategies to encourage enrollment
intervention benefits for the oldest-old population. Although it and participation among oldest-old adults.
is expected that younger seniors may benefit from the interven-
tion more than their older counterparts, findings of this study ACKNOWLEDGMENTS
indicate both groups’ improvement in physical activity was asso- A matter of balance/voluntary lay leader (AMOB/VLL) is a major
ciated with improvements in falls efficacy. Moreover, oldest-old program activity in the Aging Texas Well’s Texas Healthy Lifestyles
participants showed larger rate of improvement when compared Initiative. State-wide implementation is supported by the Depart-
to the younger-old participants. Future studies should focus on ment of Aging and Disability Services and administered through
participants aged 85 years and older to examine what charac- the Texas Association of Area Agencies on Aging. The evaluation
teristics are associated with the effectiveness of evidence-based is conducted by the Texas A&M Health Science Center, School
programs, such as AMOB/VLL. Detailed examination of whether of Public Health. We recognize faculty support from The Center
physical activities from the AMOB/VLL could influence cognitive for Community Health Development, which is a member of the

www.frontiersin.org April 2015 | Volume 2 | Article 182 | 251

Cho et al. Falls efficacy among oldest-old adults

Prevention Research Centers Program, supported by the Centers 21. Clemson L, Cumming RG, Kendig H, Swann M, Heard R, Taylor K. The
for Disease Control and Prevention cooperative agreement num- effectiveness of a community-based program for reducing the incidence of
falls in the elderly: a randomized trial. J Am Geriatr Soc (2004) 52:1487–94.
ber 5U48 DP000045. The findings and conclusions in this article
doi:10.1111/j.1532-5415.2004.52411.x
are those of the author(s) and do not necessarily represent the 22. Tennstedt S, Howland J, Lachman M, Peterson E, Kasten L, Jette A. A random-
official position of Department of Aging and Disability Services ized, controlled trial of a group intervention to reduce fear of falling and asso-
or the Centers for Disease Control and Prevention. ciated activity restriction in older adults. J Gerontol B Psychol Sci Soc Sci (1998)
53:384–92. doi:10.1093/geronb/53B.6.P384
23. Tennstedt S, Lawrence R, Kasten L. An intervention to reduce fear of falling and
REFERENCES enhance activity: who is most likely to benefit? Educ Gerontol (2001) 27:227–40.
1. Federal Interagency Forum on Aging-Related Statistics. Older Americans 2010: doi:10.1080/036012701750194969
Key Indicators of Well-Being. Washington, DC: U.S. Government Printing Office 24. Tinetti ME. Prevention of falls and fall injuries in elderly persons: a research
(2010). agenda. Prev Med (1994) 23:756–62. doi:10.1006/pmed.1994.1130
2. Zhao J, Barclay S, Farquhar M, Kinmonth AL, Brayne C, Fleming J. The old- 25. Tinetti ME. Clinical practice: preventing falls in elderly persons. N Engl J Med
est old in the last year of life: population-based findings from Cambridge city (2003) 348:42–9. doi:10.1056/NEJMcp020719
over-75s cohort study participants aged 85 and older at death. J Am Geriatr Soc 26. Cumming RG, Salkeld G, Thomas M, Szonyi G. Prospective study of the
(2010) 58:1–11. doi:10.1111/j.1532-5415.2009.02622.x impact of fear of falling on activities of daily living, SF-36 scores, and nurs-
3. Edlund BJ, Lufkin SR, Franklin B. Long-term care planning for baby boomers: ing home admission. J Gerontol (2000) 55:M299–305. doi:10.1093/gerona/55.5.
addressing an uncertain future. Online J Issues Nurs (2003) 8(2):3. M299
4. Strunk BC, Ginsburg PB, Banker MI. The effect of population aging on future 27. Lachman ME, Howland J, Tennstedt S, Jette A, Assmann S, Peterson EW.
hospital demand. Health Aff (Millwood) (2006) 25:w141–9. doi:10.1377/hlthaff. Fear of falling and activity restriction: the survey of activities and fear of
25.w141 falling in the elderly (SAFE). J Gerontol B Psychol Sci Soc Sci (1998) 53:43–50.
5. Yeaworth RC. Long-term care and insurance. J Gerontol Nurs (2002) 28:45–51. doi:10.1093/geronb/53B.1.P43
doi:10.3928/0098-9134-20021101-10 28. Lachman ME, Weaver SL, Bandura M, Elliott E, Lewkowicz C. Improving mem-
6. Boyle PA, Buchman AS, Wilson RS, Bienias JL, Bennett DA. Physical activity ory and control beliefs through cognitive restructuring and self-generated strate-
is associated with incident disability in community-based older persons. J Am gies. J Gerontol (1992) 47:293–9. doi:10.1093/geronj/47.5.P293
Geriatr Soc (2007) 55:195–201. doi:10.1111/j.1532-5415.2007.01038.x 29. Ory MG, Smith ML, Wade AF, Wright JC, Parrish R. Addressing falls in Texas:
7. Paterson DH, Warburton DE. Physical activity and functional limitations in evidence-based fall prevention programming for older adults. Tex Public Health
older adults: a systematic review related to Canada’s physical activity guidelines. Assoc J (2010) 62:15–20.
Int J Behav Nutr Phys Act (2010) 7:38. doi:10.1186/1479-5868-7-38 30. Smith ML, Ahn S, Sharkey JR, Horel S, Mier N, Ory MG. Successful falls pre-
8. Fernández-Ballesteros R, Zamarrón MD, Díez-Nicolás J, Lopez-Bravo MD, vention programming for older adults in Texas: rural-urban variations. J Appl
Molina MA, Schettini R. Mortality and refusal in the longitudinal 90+ project. Gerontol (2012) 31(1):3–27. doi:10.1177/0733464810378407
Arch Gerontol Geriatr (2011) 53:e203–8. doi:10.1016/j.archger.2010.09.007 31. Smith ML, Ory MG, Larsen R. Older women in a state-wide, evidence-based
9. Loland N. Exercise, health, and aging. J Aging Phys Act (2004) 12:170–84. falls prevention program: who enrolls and what benefits are obtained? Womens
10. Bijnen FC, Feskens EJ, Caspersen CJ, Mosterd WL, Kromhout D. Age, period, and Health Issues (2010) 20:427–34. doi:10.1016/j.whi.2010.07.003
cohort effects on physical activity among elderly men during 10 years of follow- 32. Smith ML, Ory MG, Ahn S, Bazzarre TL, Resnick B. Older adult’s participation
up: the Zutphen Elderly Study. J Gerontol A Biol Sci Med Sci (1998) 53:M235–41. in a community-based falls prevention exercise program: relationships between
doi:10.1093/gerona/53A.3.M235 the EASY tool, program attendance, and health outcomes. Gerontologist (2011)
11. von Bonsdorff MB, Rantanen T. Progression of functional limitations in rela- 51:809–21. doi:10.1093/geront/gnr084
tion to physical activity: a life course approach. Eur Rev Aging Phys Act (2011) 33. Smith ML, Quinn C, Gipson R, Wilson AD, Ory MG. Serving rural communities
8:23–30. doi:10.1007/s11556-010-0070-9 for falls prevention: the dissemination of a matter of balance in the Brazos valley
12. Friedman SM, Munoz B, West SK, Rubin GS, Fried LP. Falls and fear of region of Texas. Tex Public Health Assoc J (2011) 63:54–8.
falling: which comes first? A longitudinal prediction model suggests strategies 34. Batra A, Melchior M, Seff L, Frederick N, Palmer RC. Evaluation of a community-
for primary and secondary prevention. J Am Geriatr Soc (2002) 50:1329–35. based falls prevention program in South Florida, 2008–2009. Prev Chronic Dis
doi:10.1046/j.1532-5415.2002.50352.x (2012) 9:110057. doi:10.5888/pcd9.110057
13. van Haastregt JC, Zijlstra GA, van Rossum E, van Eijk JT, Kempen GI. Feelings 35. Ullmann G, Williams HG, Plass CF. Dissemination of an evidence-based pro-
of anxiety and symptoms of depression in community-living older persons who gram to reduce fear of falling, South Carolina, 2006–2009. Prev Chronic Dis
avoid activity for fear of falling. Am J Geriatr Psychiatry (2008) 16(3):186–93. (2012) 9:110093. doi:10.5888/pcd9.110093
doi:10.1097/JGP.0b013e3181591c1e 36. Smith ML, Jiang L, Ory MG. Falls efficacy among older adults enrolled
14. Tinetti ME, Williams CS. The effect of falls and fall injuries on functioning in an evidence-based program to reduce fall-related risk: sustainability of
in community-dwelling older persons. J Gerontol A Biol Sci Med Sci (1998) individual benefits over time. Fam Community Health (2012) 35(3):256–63.
53:M112–9. doi:10.1093/gerona/53A.2.M112 doi:10.1097/FCH.0b013e318250bdb8
15. Rosengren KS, McAuley E, Mihalko SL. Gait adjustments in older adults: activ- 37. Smith ML, Ahn S, Mier N, Jiang L, Ory MG. An evidence-based program to
ity and efficacy influences. Psychol Aging (1998) 13:375–86. doi:10.1037//0882- reduce fall-related risk among older adults: a comparison of program efficacy
7974.13.3.375 by ethnicity. Californian J Health Promot (2012) 10(1):28–44.
16. Tinetti ME, Richman D, Powell L. Falls efficacy as a measure of fear of falling. 38. Smith ML, Hochhalter AK, Cheng Y, Wang S, Ory MG. Programmatic influ-
J Gerontol (1990) 45:239–43. doi:10.1093/geronj/45.6.P239 ences on outcomes of an evidence-based falls program for older adults: a trans-
17. McAuley E, Mihalko SL, Rosengren KS. Self-efficacy and balance correlates of lational assessment. Transl Behav Med (2011) 1(3):384–93. doi:10.1007/s13142-
fear of falling in the elderly. J Phys Act (1997) 5:329–40. 011-0058-7
18. Kwon MS. Effects of a fall prevention program on physical fitness and psy- 39. Ling CH, Taekema D, de Craen AJ, Gussekloo J, Westendorp RG, Maier AB.
chological functions in community dwelling elders. J Korean Acad Nurs (2011) Handgrip strength and mortality in the oldest old population: the Leiden 85-
41:165–74. doi:10.4040/jkan.2011.41.2.165 plus study. Can Med Assoc J (2010) 182:43. doi:10.1503/cmaj.091278
19. Li FZ, Fisher KJ, Harmer P, McAuley E. Falls self-efficacy as a mediator of fear 40. Baltes PB. On the incomplete architecture of human ontogeny: selection, opti-
of falling in an exercise intervention for older adults. J Gerontol A Biol Sci Med mization, and compensation as foundation of developmental theory. Am Psychol
Sci (2005) 60:34–40. doi:10.1093/geronb/53B.1.P43 (1997) 52:366–380. doi:10.1037/0003-066X.52.4.366
20. McAuley E, Katula J, Mihalko SL, Blissmer B, Duncan TE, Pena M, et al. Mode of 41. Baltes PB, Smith J. A systemic-holistic view of psychological functioning
physical activity and self-efficacy in older adults: a latent growth curve analysis. in very old age: introduction to a collection of articles from the Berlin
J Gerontol B Psychol Sci Soc Sci (1999) 54:283–92. doi:10.1093/geronb/54B.5. aging study. Psychol Aging (1997) 12(3):395–409. doi:10.1037/0882-7974.12.3.
P283 395

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 182 | 252
Cho et al. Falls efficacy among oldest-old adults

42. Smith J. Stress and aging: theoretical and empirical challenges for interdiscipli- 60. Baert V, Gorus E, Mets T, Geerts C, Bautmans I. Motivators and barriers for
nary research. Neurobiol Aging (2003) 24:S77–S80. doi:10.1016/S0197-4580(03) physical activity in the oldest old: a systematic review. Aging Res Rev (2011)
00049-6 10:464–74. doi:10.1016/j.arr.2011.04.001
43. Ory MG, Hoffman MK, Hawkins M, Sanner B, Mockenhaupt R. Challenging 61. Newson RS, Kemps EB. Factors that promote and prevent exercise engagement in
aging stereotypes: strategies for creating a more active society. Am J Prev Med older adults. J Aging Health (2007) 19:470–81. doi:10.1177/0898264307300169
(2003) 25:164–71. doi:10.1016/S0749-3797(03)00181-8 62. Forkan R, Pumper B, Smyth N, Wirkkala H, Ciol MA, Shumway-Cook A. Exer-
44. Healy TC, Peng C, Haynes MS, McMahon EM, Botler JL, Gross L. The feasibility cise adherence following physical therapy intervention in older adults with
and effectiveness of translating a matter of balance into a volunteer lay leader impaired balance. Phys Ther (2006) 86:401–10.
model. J Appl Gerontol (2008) 27:34–51. doi:10.1177/0733464807308620 63. Heesch KC, Brown DR, Blanton CJ. Perceived barriers to exercise and stage
45. Partnership for Healthy Aging. A Matter of Balance Volunteer Lay of exercise adoption in older women of different racial/ethnic groups. Women
Leader Model (2011). Retrieved from http://www.ncoa.org/improve-health/ Health (2000) 30:61–76. doi:10.1300/J013v30n04_05
center-for-healthy-aging/content-library/Matter-of-Balance-6-30-2011.pdf 64. Dye CJ, Wilcox S. Beliefs of low-income and rural older women regarding phys-
46. Graafmans WC, Ooms ME, Hofstee HMA, Bezemer PD, Bouter LM, Lips P. Falls ical activity: you have to want to make your life better. Women Health (2006)
in the elderly: a prospective study of risk factors and risk profiles. Am J Epidemiol 43:115–34. doi:10.1300/J013v43n01_07
(1996) 143(11):1129–36. doi:10.1093/oxfordjournals.aje.a008690 65. Wilcox S, Dowda M, Leviton LC, Bartlett-Prescott J, Bazzarre T, Campbell-
47. Stevens JA, Dellinger AM. Motor vehicle and fall related deaths among older Voytal K, et al. Active for Life® : results from the translation of two physical
Americans 1990-98: sex, race, and ethnic disparities. Inj Prev (2002) 8(4):272–5. activity programs. Am J Prev Med (2008) 35:340–51. doi:10.1016/j.amepre.2008.
doi:10.1136/ip.8.4.272 07.001
48. National Council on the Aging. Conducting Evidence-Based Programs in 66. Müller-Riemenschneider F, Reinhold T, Nocon M, Willich SN. Long-term effec-
Senior Centers (2008). Retrieved from http://www.ncoa.org/improve-health/ tiveness of interventions promoting physical activity: a systemic review. Prev
center-for-healthy-aging/content-library/Conducting_EBPs_in_Senior_ Med (2008) 47:354–68. doi:10.1016/j.ypmed.2008.07.006
Centers-FL8-08.pdf 67. He W, Muenchrath MN. U.S. Census Bureau, American Community Survey
49. Chernoff H, Lehmann EL. The use of maximum likelihood estimates in x 2 Reports, ACS-17, 90+ in the United States: 2006–2008. Washington, DC: U.S.
tests for goodness of fit. Ann Math Stat (1954) 25:579–86. doi:10.1214/aoms/ Government Printing Office (2011).
1177728726
50. Littell RC, Milliken GA, Stroup WW, Wolfinger RD, Schabenberger O. SAS for
Mixed Models. 2nd ed. Cary, NC: SAS Institute, Inc (2006). Conflict of Interest Statement: The authors declare that the research was conducted
51. Horney J, Osgood D, Marshall I. Criminal careers in the short-term: intra- in the absence of any commercial or financial relationships that could be construed
individual variability in crime and its relation to local life circumstances. Am as a potential conflict of interest.
Sociol Rev (1995) 60:655–73. doi:10.2307/2096316
52. Raudenbush SW, Bryk AS. Hierarchical Linear Models: Applications and Data This paper is included in the Research Topic, “Evidence-Based Programming for Older
Analysis Methods. 2nd ed. Thousand Oaks, CA: Sage (2002). Adults.” This Research Topic received partial funding from multiple government and
53. Frazier P, Tix A, Barron K. Testing moderator and mediator effects in counseling private organizations/agencies; however, the views, findings, and conclusions in these
psychology research. J Couns Psychol (2004) 51:115–34. doi:10.1037/0022-0167. articles are those of the authors and do not necessarily represent the official position
51.1.115 of these organizations/agencies. All papers published in the Research Topic received
54. Fleming J, Brayne C. Mental and physical factors in relation to falls amongst peer review from members of the Frontiers in Public Health (Public Health Education
very old people – prospective follow-up of over 90-year olds in the CC75C and Promotion section) panel of Review Editors. Because this Research Topic repre-
study. Poster Session Presented at the Meeting of the Congress of the International sents work closely associated with a nationwide evidence-based movement in the US,
Psychogeriatric Association, Stockholm (2005). many of the authors and/or Review Editors may have worked together previously in
55. Rapp MA, Gerstorf D, Helmchen H, Smith J. Depression predicts mortality in some fashion. Review Editors were purposively selected based on their expertise with
the young old, but not in the oldest old: results from the Berlin aging study. Am evaluation and/or evidence-based programming for older adults. Review Editors were
J Geriatr Psychiatry (2008) 16:844–52. doi:10.1097/JGP.0b013e31818254eb independent of named authors on any given article published in this volume.
56. Kraaij V, Garnefski N, Maes S. The joint effects of stress, coping and coping
resources on depressive symptoms in the elderly. Anxiety Stress Coping (2002) Received: 23 June 2014; accepted: 23 September 2014; published online: 27 April 2015.
15:163–77. doi:10.1080/10615800290028468 Citation: Cho J, Smith ML, Ahn S, Kim K, Appiah B and Ory MG (2015) Effects of
57. Penrod J, Gueldner S, Poon L. Managing multiple chronic health conditions in an evidence-based falls risk-reduction program on physical activity and falls efficacy
everyday life. In: Poon L, Gueldner S, Sprouse B, editors. Successful Aging and among oldest-old adults. Front. Public Health 2:182. doi: 10.3389/fpubh.2014.00182
Adaptation with Chronic Diseases. New York, NY: Springer (2002). p. 181–208. This article was submitted to Public Health Education and Promotion, a section of the
58. Glasgow RE, Ory MG, Klesges LM, Cifuentes M, Fernald DH, Green LA. Practical journal Frontiers in Public Health.
and relevant self-report measures of health behaviors for primary care settings. Copyright © 2015 Cho, Smith, Ahn, Kim, Appiah and Ory. This is an open-access
Ann Fam Med (2005) 3:73–81. doi:10.1370/afm.261 article distributed under the terms of the Creative Commons Attribution License (CC
59. Wilcox S, Bopp M, Oberrecht L, Kammermann SK, McElmurray CT. Psychoso- BY). The use, distribution or reproduction in other forums is permitted, provided the
cial and perceived environmental correlates of physical activity in rural and older original author(s) or licensor are credited and that the original publication in this
African American and White women. J Gerontol B Psychol Sci Soc Sci (2003) journal is cited, in accordance with accepted academic practice. No use, distribution or
58:329–37. doi:10.1093/geronb/58.6.P329 reproduction is permitted which does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 182 | 253

 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2015.00026

Gait speed among older participants enrolled in an

evidence-based fall risk reduction program: a subgroup
analysis
Jinmyoung Cho 1,2 *, Matthew Lee Smith 3,2 , Tiffany E. Shubert 4 , Luohua Jiang 5,6 , SangNam Ahn 7,2 and
Marcia G. Ory 2
1
Center for Applied Health Research, Baylor Scott & White Health, Temple, TX, USA
2
Department of Health Promotion and Community Health Science, Texas A&M Health Science Center School of Public Health, College Station, TX, USA
3
Department of Health Promotion and Behavior, The University of Georgia, Athens, GA, USA
4
School of Medicine, University of North Carolina, Chapel Hill, NC, USA
5
Department of Epidemiology, School of Medicine, University of California, Irvine, CA, USA
6
Department of Epidemiology and Biostatistics, Texas A&M Health Science Center School of Public Health, College Station, TX, USA
7
Division of Health Systems Management and Policy, The University of Memphis, Memphis, TN, USA

Edited by: Background: Functional decline is a primary risk factor for institutionalization and mortality
Sanjay P. Zodpey, Public Health
among older adults. Although community-based fall risk reduction programs have been
Foundation of India, India
widely disseminated, little is known about their impact on gait speed, a key indicator of
Reviewed by:
Cheryll Diann Lesneski, The University functional performance. Changes in functional performance between baseline and post-
of North Carolina at Chapel Hill, USA intervention were examined by means of timed up and go (TUG), a standardized functional
Preeti Negandhi, Indian Institute of assessment test administered to participants enrolled in A Matter of Balance/Volunteer
Public Health Delhi, India
Lay Leader (AMOB/VLL) model, an evidence-based fall risk reduction program.
*Correspondence:
Jinmyoung Cho, Department of Methods: This study included 71 participants enrolled in an AMOB/VLL program in the
Health Promotion and Community Brazos Valley and South Plain regions of Texas. Paired t -tests were employed to assess
Health Science, Texas A&M Health
Science Center, School of Public
program effects on gait speed at baseline and post-intervention for all participants and by
Health, Center for Applied Health subgroups of age, sex, living status, delivery sites, and self-rated health.The Bonferroni cor-
Research, Baylor Scott and White rection was applied to adjust inflated Type I error rate associated with performing multiple
Health, MS-01-501, 2401 S 31st t -tests, for which p-values <0.0042 (i.e., 0.5/12 comparisons) were deemed statistically
Street, Temple, TX 76508, USA
e-mail: [email protected],
significant.
[email protected]
Results: Overall, gait speed of enrolled participants improved from baseline to post-
intervention (t = 3.22, p = 0.002). Significant changes inTUG scores were observed among
participants who lived with others (t = 4.45, p < 0.001), rated their health as excellent, very
good, or good (t = 3.05, p = 0.003), and attended program workshops at senior centers
(t = 3.52, p = 0.003).
Conclusion: Findings suggest community-based fall risk reduction programs can improve
gait speed for older adults. More translational research is needed to understand factors
related to the effectiveness of fall risk reduction programs in various populations and
settings.
Keywords: older adults, A Matter of Balance/Volunteer Lay Leader model, timed up and go

FALLS AMONG OLDER ADULTS include age (2, 4), being female (9, 10), a prior history of falls (2,
Falls among older adults are a serious public health problem 4), gait and mobility deficits (2, 4, 9, 11), and poor self-reported
in America (1). Approximately one-fourth of older adults aged health status (9, 10). In addition to personal characteristics, partic-
80 years and older experience at least two falls per year (2–4). As ular attention has been paid to the environmental circumstances
the risk of falling increases with advanced age, dramatic escalations surrounding falls, such neighborhood environments or program
in fall-related morbidity, hospitalization, institutionalization, and delivery settings (12).
mortality can be expected to accompany the aging of the popu- Fall-prevention programs and integration of prevention ser-
lation (5). Direct annual medical care costs related to falls have vices have been shown to decrease fall recurrence (13) and reduce
been estimated at almost $20 billion and are projected to increase health-care costs (14). However, literature about the effective-
sharply in the coming decades (6, 7). ness of evidence-based fall-prevention programs for improving
Various demographic, functional, and health factors are known objectively measured functional performance has been limited
to increase the risk for falling among older adults (8). These factors for programs delivered in the community by lay facilitators.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 26 | 254
Cho et al. TUG and AMOB old participants

Given its ease of use and standardization, gait speed, often called of intervention targets behavior change and emphasizes building
“timed up and go (TUG),” has been frequently used to assess fall self-efficacy and setting goals for increasing physical activ-
functional performance as an outcome measurement for effec- ity through lectures, group discussions, various problem-solving
tive interventions (15, 16). Many studies have demonstrated a and role-playing activities, exercise and assertiveness training, and
strong relationship between gait speed and fall-related risk, health individual assignments (24).
and functional status, institutionalization, and mortality among Since 2006, a well-established infrastructure has facilitated the
older adults (17–19). To address the existing research gaps, the delivery of the intervention to older adults in Texas (24). The net-
overall goal of this study was to examine improvement in func- work for aging services arranged a signed agreement with the Texas
tional performance among older participants enrolled in A Mat- Association of Area Agencies on Aging (AAA) for implementation
ter of Balance/Volunteer Lay Leader (AMOB/VLL) model, an of the AMOB/VLL program in many AAA regions. The program
evidence-based fall risk reduction program. targets low-income older minority adults and their caregivers and
focuses on reaching a diverse population in a large geographic
A MATTER OF BALANCE/VOLUNTEER LAY LEADER FALL RISK area. Residential facilities, health-care institutions, public health
REDUCTION PROGRAM departments, faith-based organizations, business sectors, and local
A Matter of Balance (AMOB), established at the Roybal Center government were included as partners with the Texas AAA sites to
for Enhancement of Late-Life Function at Boston University, is build fall-prevention capacity (24, 25).
an evidence-based program to reduce risk of falls among older
adults (20). The effectiveness of the AMOB program was orig- PURPOSE OF STUDY
inally tested through a randomized clinical trial, which showed Although community-based health promotion programs result
positive improvements in physical activity and mobility control in for improvement in falls efficacy, overall health status, and
(21). Derived from the original program, the AMOB/VLL model increased physical activities (11, 26), less is known about their
has been adapted for widespread community dissemination in var- impact on physical performance (i.e., TUG) among older par-
ious health and aging sectors (22, 23). Delivered by trained lay-led ticipants. The purposes of this study were to (a) assess the
facilitators in 38 states, it is presented in 2-h sessions for 8 weeks. changes in physical performance measured by the TUG test
One hour is taught by a physical therapist. This hour focuses on from baseline to post-intervention and (b) compare the improve-
the role of exercise in fall prevention. It is not meant to be a stand- ment in physical performance by personal characteristics and
alone session, but rather an introduction for older adults to build delivery sites. A conceptual model for this study is shown in
upon. At the end of AMOB, participants are more likely to exer- Figure 1. This model depicts the fall risk reduction program
cise and intended activity (21). Each session includes specific goals as an intervention that can have positive effects on changes
for older adults to reduce the risk of falling and continue remain- in physical performance. If participation in AMOB results in
ing active and independent (24). The major goals of the program improved efficacy as well as improved physical performance,
are as follows: to make participants perceive control, to increase physical therapists may want to include AMOB as a program
their confidence, and to learn falls are controllable. The design for appropriate patients. To better understand who may benefit

Personal Characteristics
(Age, Sex, Residential Status)

Health Indicators
(Chronic Conditions, Self-rated Health)

Delivery Site Setting

REDUCED
Fall Incidence
Fall Risk Functional Institutionalization
Reduction Program Performance
(AMOB/VLL) Healthcare Utilization
(Timed Up and Go)
Healthcare Cost
Premature Mortality

FIGURE 1 | Conceptual model.

www.frontiersin.org April 2015 | Volume 3 | Article 26 | 255

Cho et al. TUG and AMOB old participants

the most from this program, personal characteristics and deliv- scores from both baseline and post-intervention. Results indicated
ery sites act as moderators between the fall risk reduction pro- the presence of three outliers, who were then omitted. An addi-
gram (as illustrated by the AMOB/VLL model) and physical tional two cases reporting an “other” ethnic group were excluded
performance as measured by the TUG test. In turn, the TUG to maximize racial and ethnic homogeneity of participants for
measurements are associated with long-term improvements in this study. As a result, only non-Hispanic White participants
reduced health-care use and costs as well as enhanced health and were included in this study. Figure 2 shows the recruitment flow
well-being. diagram, indicating that 71 participants were included in study
analyses.
METHODS
PROCEDURES AND PARTICIPANTS MEASURES
A total of 301 participants enrolled in the Texas AMOB/VLL fall Timed up and go test
risk reduction program between September 2007 and April 2009 The TUG test, introduced in 1991 by Podsiadlo and Richardson
in two regional AAAs: Brazos Valley and South Plain. Although (27), has been used extensively for over a decade to predict fall risk
functional assessment was optional in the statewide AAA deliv- and to examine functional mobility among older adults (26, 28).
ery of the AMOB/VLL program, assessments were conducted in It assesses the time in seconds that participants required to “rise
some classes that taught these two regions. Workshop leaders were from a standard arm chair, walk at your typical or normal pace to
trained in assessment procedures and performed assessments in a line on the floor 3 meters away, turn, return, and sit down again”
eight of the AMOB/VLL classes, which served as the basis for this (p. 64) (27). This test was validated to test physical functioning
study. A total of 171 participants who attended classes in these two and mobility among community-dwelling older adults (26) and
regions but who were not assessed using the TUG were excluded; showed high predictive validity with the Berg Balance Scale (27).
thus, 76 participants completed the TUG test at baseline and post- Those who completed the TUG tasks in more than 14 s also showed
intervention, whereas 54 participants did not complete the test lower scores on the Berg Balance Scale, which was associated with
at both times. Boxplots were used to screen for outliers for TUG higher risk for institutionalization (26).

Enrolled in two Texas

counties delivering
AMOB/VLL (n = 301)

Excluded (n = 225)
• Not offered the TUG test (n = 171)
• Not completed the TUG test (n = 54)

Completed the TUG test

at baseline and post-intervention (n = 76)

Excluded (n = 5)
• Outliers (n = 3)
• Non-White (n = 2)

Allocated to final analysis

(n = 71)

FIGURE 2 | Diagram for study participants inclusion.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 26 | 256
Cho et al. TUG and AMOB old participants

Personal characteristics Table 1 | Characteristic of the study participants.

Age was coded as a continuous variable based on a participant’s
birth date and ranged from 56 to 95. The age was then categorized Characteristics Frequency (%)
into three groups for the purpose of the study: young-old (up Age, mean (SD) (range: 56−95) 77.8 (9.3)
to 69 years), mid-old (from 70 to 79 years), and old-old (80 years
and older). Sex was scored 0 if the participant was male and 1 if Sex

the participant was female. Living status was scored 1 if partici- Male 13 (19.4)

pants lived with others and 0 if they lived alone. Self-rated health Female 54 (80.6)

was included. Self-rated health, a single item measuring in which Residential status
participants rate current status of their overall health, has been Living alone 29 (43.9)
widely used as a significant predictor of physical and psychological Living with others 37 (56.1)
health such as mortality or functionality among various popu- Self-rated health
lations (29–32). Many studies have shown that the single item Excellent 5 (7.7)
is a reliable and valid measure reflecting objective health status Very good 17 (26.2)
(e.g., cardio-cerebral vascular diseases, visual impairment) (32– Good 27 (41.5)
34). At baseline, participants were also asked to self-report their Fair 15 (23.1)
perceived health status: “Would you say that in general your health Poor 1 (1.5)
is poor, fair, good, very good, or excellent?” For comparisons of
Numbers of chronic condition
self-rated health, the responses were divided into two categories
None 11 (15.5)
(i.e., poor/fair vs. good/very good/excellent).
1–2 44 (61.9)
3+ 16 (22.5)
Delivery sites
To compare outcomes at the various settings in which the Delivery sites
AMOB/VLL program was conducted in the Brazos Valley and Senior centers 15 (21.1)
South Plain regions, delivery site types were obtained from admin- Community centers 2 (2.8)
istrative data. Delivery site categories included senior centers, Residential facilities 37 (52.1)
community centers, faith-based organizations, residential facili- Faith-based organizations 9 (12.7)
ties, and other Parks Department facilities. For comparisons of Other-parks department facilities 8 (11.3)
delivery sites, five sites were categorized into three groups: senior
Different numbers of missing cases were observed for each variable. Missing
centers and community centers, residential facilities, and others.
cases were excluded from calculations and analyses.

Data analysis
The paired t -test was employed to compare mean TUG scores centers (21.1%), faith-based organizations (12.7%), other Parks
for all participants pre- and post-intervention. Statistical signif- Department facilities (11.3%), and community centers (2.8%).
icance was examined at the level of 0.05 for this test. Then, a
series of paired t -tests were employed to compare the TUG scores CHANGES IN TIMED UP AND GO TEST
by subgroups: age groups, sex, residential status, delivery sites, and Before the paired t -test for the TUG score was conducted, the TUG
self-rated health. Bonferroni’s correction was applied for subgroup scores at baseline and post-intervention were observed. Almost a
(12 groups) comparisons to adjust the inflated Type I error rate third of participants (28.2%) at baseline and 22.5% of partic-
associated with performing multiple t -tests, for which p-values ipants at post-intervention performed slower than 14 s, which
<0.0042 (i.e., 0.5/12 comparisons) were deemed statistically sig- represents a critical value on the TUG test. Table 2 presents
nificant. Statistical analyses were conducted with SPSS statistical results of the paired t -tests for TUG scores among all AMOB/VLL
software (version 20.0). As an indicator of practical significance, program participants and by subgroups from baseline to post-
Cohen’s d standardized effect sizes were calculated to compare intervention. Among all participants, the average TUG score at
intervention effects from baseline to post-intervention within each baseline was 12.89 (SD = 5.08) and changed to 11.95 (SD = 4.30)
group. at post-intervention (t = 3.22, p = 0.002). When comparing TUG
score changes by subgroup, three significant improvements were
RESULTS found. First, participants who lived with others showed signifi-
SAMPLE CHARACTERISTICS cant changes in TUG scores from baseline (M = 12.61, SD = 5.92)
Table 1 summarizes characteristics of study participants. The aver- to post-intervention (M = 11.32, SD = 5.04), t = 4.45, p < 0.001.
age age of the study participants was 77.8 (SD = 9.3) years old. The The effect size (Cohen’s d) was 0.23. Second, participants who
majority of participants was female (80.6%), and more than half attended the AMOB/VLL program at senior centers or commu-
the participants lived with others (56.1%). Over three quarters nity centers showed statistically significant improvement in TUG
of the participants rated their health good, very good, or excel- scores from 14.96 (SD = 7.20) at baseline to 13.30 (SD = 6.21) at
lent (75.4%). Most participants had at least one chronic health post-intervention, t = 3.52, p = 0.003. Cohen’s d was 0.25. Third,
problem (84.4%). Within the two regional AAAs, the AMOB/VLL those who perceived their health good, very good, or excellent
program was implemented in residential facilities (52.1%), senior showed significant improvement in TUG scores: 12.77 (SD = 5.41)

www.frontiersin.org April 2015 | Volume 3 | Article 26 | 257

Cho et al. TUG and AMOB old participants

Table 2 | Average TUG scores in pre- and post-test by groups.

Pre (SD) Post (SD) t -Value p Cohen’s d

Total participants 12.89 (5.08) 11.95 (4.30) 3.22 0.002 0.25

Age groups
Young-old (n = 14) 9.74 (2.21) 8.89 (2.06) 2.60 0.018 0.40
Mid-old (n = 16) 11.67 (3.43) 10.83 (2.80) 2.25 0.040 0.27
Old-old (n = 34) 14.85 (6.06) 13.82 (5.00) 2.20 0.035 0.19
Sex
Male (n = 13) 12.41 (4.00) 11.53 (3.59) 1.33 0.208 0.23
Female (n = 54) 12.95 (5.45) 12.08 (4.62) 2.77 0.008 0.17
Living status
Living alone (n = 29) 12.93 (4.05) 12.66 (3.44) 0.52 0.605 0.07
Living with others (n = 37) 12.61 (5.92) 11.32 (5.04) 4.45 <0.001 0.23
Delivery sites
Senior/community centers (n = 15) 14.96 (7.20) 13.30 (6.21) 3.52 0.003 0.25
Residential facilities (n = 38) 13.58 (4.51) 12.79 (3.69) 1.56 0.128 0.19
Others (n = 19) 9.90 (2.27) 9.24 (2.02) 2.54 0.020 0.31
Self-rated health groups
Excellent/VG/good (n = 57) 12.77 (5.41) 11.87 (4.56) 3.05 0.003 0.18
Fair/poor (n = 11) 13.08 (2.78) 11.86 (3.36) 1.25 0.240 0.40

Different numbers of missing cases were observed for each variable. Missing cases were excluded from calculations and analyses.

at baseline and 11.87 (SD = 4.56) at post-intervention, t = 3.05, improvements in the ability to perform important social and role
p = 0.003. The effect size was 0.18. functions (23, 36).
Findings of the current study also highlight the importance of
DISCUSSION physical health, social, and environmental correlates to enhance
The primary objective in this study was to examine changes in the effectiveness of the evidence-based program. First, the analy-
functional performance between baseline and post-intervention ses revealed that better perception of health was associated with
among participants enrolled in the Texas AMOB/VLL fall risk significant improvement on the TUG test. It is obvious that those
reduction program. Several important findings emerged from this who perceived their health to be of better status showed signifi-
study. First, the average score for all participants’ walking speed cant improvement because these individuals may be more likely to
assessed with the standardized TUG test improved from base- have fewer chronic conditions and may be less influenced by daily
line to post-intervention. These findings demonstrate that this activity limitations. However, the largest standardized Cohen’s
fall risk reduction program can improve gait speed among old effect size (Cohen’s d = 0.40) was notably observed among par-
participants in addition to its previously reported benefits for ticipants who self-reported their health to be fair/poor despite the
falls efficacy and fear of falling (22). Second, subgroup compar- lack of statistical significance of the TUG score change. In other
isons showed significant improvements among those who rated words, those with worse health status may show larger changes in
their health more positively, lived with others, and attended pro- functional assessments because they have greater opportunity for
gram workshops in senior centers or community centers. These improvement, whereas those healthier participants who score high
findings reveal that improvement in functional performance (i.e., at baseline have little room for improvement (36). This finding
TUG) may be directly associated with participating in a fall risk points to the need for future research to increase understanding of
reduction program for these subgroups. the functional improvements of individuals of different health sta-
The most significant aspect of this study was the incor- tus levels and detect underlying statistical effects, such as regression
poration of an objectively measured functional assessment to to the mean.
compare participant improvement based on self-reported mea- Second, the significant improvement in gait speed based on res-
sures. Because most measures from evidence-based programs idential status emphasizes the importance of social correlates on
have been based on self-reported information, such as health- the effectiveness of the evidence-based program. Results showed
related quality of life, number of falls, and number of chronic significant improvement in functional performance from baseline
conditions within the previous week or month, a couple of to post-intervention among participants who live with others. This
other researchers noted that self-reported measures might pro- finding is consistent with previous studies that has shown the sig-
duce recall bias as a data collection limitation (23, 35, 36). nificant relationship between physical activities and support from
Using a standardized functional assessment test (i.e., TUG) can family or friends (37, 38). Living with others is likely to prevent
contribute to the validation of previous findings that reported older adults from social isolation, which has been identified as a

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 26 | 258
Cho et al. TUG and AMOB old participants

barrier to physical activity (38). This finding may also indicate that for lay leaders should include a training session about objective
participants who lived with others had social support mechanisms functional measurement. Such provider training is important for
that may have encouraged them to attend more AMOB/VLL pro- maintaining measurement necessary for research assessment. In
gram sessions (i.e., received more intervention dose) and engage recognition of the importance of objective measurements for pur-
in recommended physical activities outside class time. poses of both research and programing, this type of training has
Third, findings of this study suggest delivery settings in which been built into CDC’s State Fall Prevention Program (39). Fur-
evidence-based programs that are implemented can enhance thermore, future studies should focus on participants’ degree of
physical performance among old participants. Participants who disability to examine more comprehensively the effectiveness of
attended workshops in senior centers or community centers evidence-based fall risk reduction programs in different popula-
showed significant improvement in TUG scores from baseline to tions. Considering the extent of the disability or investigating the
post-intervention, which may highlight an environmental benefit difference in physical performance between fallers and non-fallers
for delivering evidence-based programs to older adults in these may suggest detailed strategies to promote physical activity for
group settings. This finding may be associated with the notion older adults with various baseline functional levels. Also, more
that these participants were healthier upon program enrollment translational research is needed to understand potential modi-
or that the location of the delivery site was more accessible, which fiable and non-modifiable correlates related to effectiveness of
increased their attendance (i.e., intervention does) and led to fall risk reduction programs on functional performance within
significant improvement. various populations and settings.
In an attempt to disseminate widely fall-prevention programs,
the Texas AAA sites have continued to build fall-prevention capac- ACKNOWLEDGMENTS
ity by partnering with the public health network and others to A Matter of Balance/Voluntary Lay Leader (AMOB/VLL) is a major
establish programs in various settings, such as residential facilities, program activity in the Aging Texas Well’s Texas Healthy Lifestyles
faith-based organizations, workplace setting, health-care institu- Initiative. Statewide implementation is supported by the Depart-
tions, public health departments, and governmental facilities (25). ment of Aging and Disability Services and administered through
Although other studies have identified differences in program the Texas Association of Area Agencies on Aging. The evaluation
outcomes by delivery site types (35), further investigation is war- is conducted by the Texas A&M Health Science Center School of
ranted to understand the influence of delivery site on functional Public Health. We recognize faculty support from The Center for
assessment measures among lay-led fall-prevention programs. Community Health Development, which is a member of the Pre-
vention Research Centers Program, supported by the Centers for
LIMITATIONS AND IMPLICATIONS Disease Control and Prevention cooperative agreement number
The findings of this study showed significant TUG score changes 5U48 DP000045. The findings and conclusions in this article are
associated with this fall risk reduction program; however, a few those of the author(s) and do not necessarily represent the official
notable limitations were associated with this study. First, this study position of Department of Aging and Disability Services or the
included only 71 study participants. The small sample size may Centers for Disease Control and Prevention.
limit generalization of our findings to other populations. Second,
as stated in the procedures and participant section, older minority REFERENCES
1. Sleet DA, Moffett DB, Stevens J. CDC’s research portfolio in older adult fall pre-
adults were excluded from this study because too few participated vention: a review of progress, 1985-2005, and future research directions. J Safety
for meaningful analyses. Although Texas is a geographically large Res (2008) 39:259–67. doi:10.1016/j.jsr.2008.05.003
and demographically diverse state, the two minority group cases 2. Rubenstein LZ. Falls in older people: epidemiology, risk factors and strategies
were intentionally excluded to yield a homogeneous sample of for prevention. Age Ageing (2006) 35:37–41. doi:10.1093/ageing/afl084
3. Tinetti M. Preventing falls in elderly persons. N Engl J Med (2003) 348:42–9.
participants. If enough minorities had been available for mean-
doi:10.1056/NEJMcp020719
ingful comparisons, we may find ethnic difference in functional 4. American Geriatrics Society, British Geriatrics Society, American Academy Of
performance among more diverse groups of participants. Finally, Orthopaedic Surgeons Panel on Falls Prevention. Guideline for the prevention
if objective method of rating current health status (e.g., biomark- of falls in older persons. J Am Geriatr Soc (2001) 49:664–72. doi:10.1046/j.1532-
ers) was used in the fall risk reduction program instead of self-rated 5415.2001.49115.x
5. Persad CC, Cook S, Giordani B. Assessing falls in the elderly: should we use
health, the result may provide an association between health status
simple screening tests or a comprehensive fall risk evaluation? Eur J Phys Rehabil
and functional capacity among old participants. Med (2010) 46:249–59.
The findings from the current study have considerable impli- 6. Stevens JA, Corso PS, Finkelstein EA, Miller TR. The costs of fatal and non-fatal
cations for future research on translational studies. Although falls among older adults. Inj Prev (2006) 12:290–5. doi:10.1136/ip.2005.011015
this study provides an important view of the use of TUG tests 7. Roudsari BS, Ebel BE, Corso PS, Molinari NA, Koepsell TD. The acute med-
ical care costs of fall-related injuries among the U.S. older adults. Injury (2005)
in a community-based fall risk reduction program, additional 36:1316–22. doi:10.1016/j.injury.2005.05.024
research is needed to link functional assessment scores to the actual 8. Ullmann G, Williams HG, Plass CF. Dissemination of an evidence-based pro-
fall experience, subsequent health-care use, and the availability gram to reduce fear of falling, south Carolina, 2006-2009. Prev Chronic Dis
of supportive environments illustrated in our conceptual model (2012) 9:E103.
of fall risk behaviors, interventions, and long-term outcomes. 9. Kato R, Takagi C, Sakurai N, Hoshi T. Risk factors for falls and survival after
falling in elderly people in a community. Nihon Koshu Eisei Zasshi (2012)
First, the capacity for objective functional measurement among 59:305–14.
community-dwelling older adults should be built into evidence- 10. Muraki S, Akune T, Ishimoto Y, Nagata K, Yoshida M, Tanaka S, et al. Risk fac-
based fall-prevention programs. For example, instructor manuals tors for falls in a longitudinal population-based cohort study of Japanese men

www.frontiersin.org April 2015 | Volume 3 | Article 26 | 259

Cho et al. TUG and AMOB old participants

and women: the ROAD study. Bone (2013) 52:516–23. doi:10.1016/j.bone.2012. 31. Appels A, Bosma H, Grabauskas V, Gostautas A, Sturmans F. Self-rated health
10.020 and mortality in a Lithuanian and a Dutch population. Soc Sci Med (1996)
11. Viccaro LJ, Perera S, Studenski SA. Is timed up and go better than gait speed in 42:681–9. doi:10.1016/0277-9536(95)00195-6
predicting health, function, and falls in older adults? J Am Geriatr Soc (2011) 32. Murata C, Kondo T, Tamakoshi K, Yatsuya H, Toyoshima H. Determinants of
59:887–92. doi:10.1111/j.1532-5415.2011.03336.x self-rated health: could health status explain the association between self-rated
12. Landy DC, Mintzer MJ, Silva AK, Dearwater SR, Schulman CI. Hispanic ethnic- health and mortality? Arch Gerontol Geriatr (2006) 43:369–80. doi:10.1016/j.
ity and fatal fall risk: do age, gender, and community modify the relationship? archger.2006.01.002
J Surg Res (2012) 175:113–7. doi:10.1016/j.jss.2011.02.035 33. Molarius A, Janson S. Self-rated health, chronic diseases, and symptoms among
13. Tinetti ME. Prevention of falls and fall injuries in elderly persons: a research middle-aged and elderly men and women. J Clin Epidemiol (2002) 55:364–70.
agenda. Prev Med (1994) 23:756–62. doi:10.1006/pmed.1994.1130 doi:10.1016/S0895-4356(01)00491-7
14. Rizzo J, Baker D, McAvay G, Tinetti M. The cost-effectiveness of a multifactorial 34. Wu S, Wang R, Zhao Y, Ma X, Wu M, Yan X, et al. The relationship between self-
targeted prevention program for falls among community elderly persons. Med rated health and objective health status: a population-based study. BMC Public
Care (1996) 34:954–69. doi:10.1097/00005650-199609000-00007 Health (2013) 13:320. doi:10.1186/1471-2458-13-320
15. Cesari M, Kritchevsky SB, Newman AB, Simonsick EM, Harris TB, Penninx 35. Smith ML, Ory MG, Belza B, Altpeter M. Personal and delivery site char-
BW, et al. Added value of physical performance measures in predicting adverse acteristics associated with intervention dosage in an evidence-based fall risk
health-related events: results from the health, aging and body composition study. reduction program for older adults. Transl Behav Med (2012) 2:188–98.
J Am Geriatr Soc (2009) 57:251–9. doi:10.1111/j.1532-5415.2008.02126.x doi:10.1007/s13142-012-0133-8
16. Abellan van Kan G, Rolland Y, Andrieu S, Bauer J, Beauchet O, Bonnefoy 36. Smith ML, Jiang L, Ory MG. Falls efficacy among older adults enrolled in an
M, et al. Gait speed at usual pace as a predictor of adverse outcomes in evidence-based program to reduce fall-related risk sustainability of individual
community-dwelling older people an international academy on nutrition and benefits over time. Fam Community Health (2012) 35:256–63. doi:10.1097/FCH.
aging (IANA) task force. J Nutr Health Aging (2009) 13:881–9. doi:10.1007/ 0b013e318250bdb8
s12603-009-0246-z 37. Wilcox S, Der Ananian C, Sharpe PA, Robbins J, Brady T. Correlates of physi-
17. Beauchet O, Fantino B, Allali G, Muir SW, Montero-Odasso M, Annweiler C. cal activity in persons with arthritis: review and recommendations. J Phys Act
Timed up and go test and risk of falls in older adults: a systematic review. J Nutr Health (2005) 2:230–52.
Health Aging (2011) 15:933–8. doi:10.1007/s12603-011-0062-0 38. Gordon NF, Gulanick M, Costa F, Fletcher G, Franklin BA, Roth EJ, et al. Physical
18. Studenski S, Perera S, Patel K, Rosano C, Faulkner K, Inzitari M, et al. Gait activity and exercise recommendations for stroke survivors: an American heart
speed and survival in older adults. JAMA (2011) 305:50–8. doi:10.1001/jama. association scientific statement from the council on clinical cardiology, subcom-
2010.1923 mittee on exercise, cardiac rehabilitation, and prevention; the council on car-
19. Huang WW, Perera S, VanSwearingen J, Studenski S. Performance measures diovascular nursing; the council on nutrition, physical activity, and metabolism;
predict onset of activity of daily living difficulty in community-dwelling and the stroke council. Stroke (2004) 35:1230–40.
older adults. J Am Geriatr Soc (2010) 58:844–52. doi:10.1111/j.1532-5415.2010. 39. Center for Disease Control and Prevention, National Center for Injury Pre-
02820.x vention and Control, Stopping Elderly Accidents, Dealths, & Injuries. STEADI
20. Lachman M, Weaver S, Bandura M, Elliott E, Lewkowicz C. Improving memory Implementation Guide: For Health Care Providers. Washington, DC: Center for
and control beliefs through cognitive restructuring and self-generated strategies. Disease Control and Prevention (2012).
J Gerontol (1992) 47:293–9. doi:10.1093/geronj/47.5.P293
21. Tennstedt S, Howland J, Lachman M, Peterson E, Kasten L, Jette A. A random-
ized, controlled trial of a group intervention to reduce fear of falling and asso- Conflict of Interest Statement: The authors declare that the research was conducted
ciated activity restriction in older adults. J Gerontol B Psychol Sci Soc Sci (1998) in the absence of any commercial or financial relationships that could be construed
53:384–92. doi:10.1093/geronb/53B.6.P384 as a potential conflict of interest.
22. Healy TC, Peng C, Haynes MS, McMahon EM, Botler JL, Gross L. The feasibility
and effectiveness of translating A Matter of Balance into a Volunteer Lay Leader This paper is included in the Research Topic, “Evidence-Based Programming for Older
model. J Appl Gerontol (2008) 27:34–51. doi:10.1177/0733464807308620 Adults.” This Research Topic received partial funding from multiple government and
23. Smith ML, Ahn SN, Sharkey JR, Horel S, Mier N, Ory MG. Successful falls pre- private organizations/agencies; however, the views, findings, and conclusions in these
vention programming for older adults in Texas: rural-urban variations. J Appl articles are those of the authors and do not necessarily represent the official position
Gerontol (2012) 31:3–27. doi:10.1177/0733464810378407 of these organizations/agencies. All papers published in the Research Topic received
24. Ory MG, Smith ML, Wade AF, Wright JC, Parrish R. Addressing falls in Texas: peer review from members of the Frontiers in Public Health (Public Health Education
evidence-based fall prevention programming for older adults. Tex Public Health and Promotion section) panel of Review Editors. Because this Research Topic repre-
J (2010) 62:15–20. sents work closely associated with a nationwide evidence-based movement in the US,
25. Cho J, Smith ML, Ahn S, Kim K, Appiah B, Ory MG. Effects of an evidence-based many of the authors and/or Review Editors may have worked together previously in
falls risk-reduction program on physical activity and falls efficacy among oldest- some fashion. Review Editors were purposively selected based on their expertise with
old adults. Front Public Health (2015) 2:182. doi:10.3389/fpubh.2014.00182 evaluation and/or evidence-based programming for older adults. Review Editors were
26. Shumway-Cook A, Brauer S, Woollacott M. Predicting the probability for falls independent of named authors on any given article published in this volume.
in community-dwelling older adults using the timed up & go test. Phys Ther
(2000) 80:896–903. Received: 23 June 2014; accepted: 31 January 2015; published online: 27 April 2015.
27. Podsiadlo D, Richardson S. The timed up and go – a test of basic functional Citation: Cho J, Smith ML, Shubert TE, Jiang L, Ahn S and Ory MG (2015) Gait speed
mobility for frail elderly persons. J Am Geriatr Soc (1991) 39:142–8. among older participants enrolled in an evidence-based fall risk reduction program: a
28. Hutton I, Gamble G, McLean G, Butcher H, Gow P, Dalbeth N. Obstacles to subgroup analysis. Front. Public Health 3:26. doi: 10.3389/fpubh.2015.00026
action in arthritis: a community case-control study. Int J Rheum Dis (2009) This article was submitted to Public Health Education and Promotion, a section of the
12:107–17. doi:10.1111/j.1756-185X.2009.01392.x journal Frontiers in Public Health.
29. Benjamins MR, Hummer RA, Eberstein IW, Nam CB. Self-reported health and Copyright © 2015 Cho, Smith, Shubert , Jiang , Ahn and Ory. This is an open-access
adult mortality risk: an analysis of cause-specific mortality. Soc Sci Med (2004) article distributed under the terms of the Creative Commons Attribution License (CC
59:1297–306. doi:10.1016/j.socscimed.2003.01.001 BY). The use, distribution or reproduction in other forums is permitted, provided the
30. Idler EL, Angel RJ. Self-rated health and mortality in the NHANES-I epidemio- original author(s) or licensor are credited and that the original publication in this
logic follow-up study. Am J Public Health (1990) 80:446–52. doi:10.2105/AJPH. journal is cited, in accordance with accepted academic practice. No use, distribution or
80.4.446 reproduction is permitted which does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 26 | 260
 PERSPECTIVE ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00291

The conversion of a practice-based lifestyle enhancement

program into a formalized, testable program: from
Texercise Classic toTexercise Select
Marcia G. Ory 1 *, Matthew Lee Smith 2 , Doris Howell 1 , Alyson Zollinger 1 , Cindy Quinn 1 , Suzanne M. Swierc 3
and Alan B. Stevens 4
1
Department of Health Promotion and Community Health Sciences, Texas A&M Health Science Center School of Public Health, College Station, TX, USA
2
Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA, USA
3
Big Brothers Big Sisters of South Texas, San Antonio, TX, USA
4
Baylor Scott & White Health, Temple, TX, USA

Edited by: Little is known about the structure, content, and benefits of practice-based or grass roots
Sanjay P. Zodpey, Public Health
health programs that have been widely delivered by a variety of community organizations
Foundation of India, India
and stakeholders. This perspective will document the natural history of Texercise Classic, a
Reviewed by:
Mary Altpeter, University of North state-endorsed but previously untested lifestyle health promotion program. It will: (1) dis-
Carolina at Chapel Hill, USA cuss Texercise Classic’s participant reach and adoption over time; (2) describe the rationale
Robert Otto Valdez, University of New and processes employed to formalize Texercise Classic into a more structured program
Mexico, USA
known as Texercise Select ; (3) outline the essential elements and activities included in Tex-
*Correspondence:
ercise Select and contrast them with those included in Texercise Classic; and (4) highlight
Marcia G. Ory , Department of Health
Promotion and Community Health key components for uniform facilitator training. The discussion will reflect upon the evolu-
Sciences, School of Public Health, tion of Texercise, compare and contrast the benefits and challenges of each program, and
Texas A&M Health Science Center, review the “next steps” for Texercise Select. In contrasting Texercise Classic and Select,
TAMU 1266, College Station, TX
it is important to understand the benefits and challenges of both programs. Preliminary
77843, USA
e-mail: [email protected] results indicate that Texercise Select is effective, yet its ability to sustain the same reach
as Texercise Classic remains unknown and an area for future study.
Keywords: translational research, program implementation, program evaluation, older adults, physical activity

INTRODUCTION developed and tested by academics in controlled settings. This

With greater recognition of the value of health promotion for movement reflects the assumption that EBPs are preferable
adults across the life-course (1), a multitude of programs now because they are“assumedly”more efficient and cost-effective than
exist to improve health and wellness among older adults (2, 3). programs that are not theoretically based and rigorously evaluated
These programs recognize that older adults are able and willing (13). With the initiation of the Administration on Aging (AoA)
to engage in health promotion programs and can derive substan- Evidence-based Disease Prevention Initiative (14, 15), substantial
tial health benefit from those programs (4). There have been two knowledge has been gained about the nature and effectiveness of
programmatic streams to meet the needs of the rapidly grow- EBPs for older adults, especially those related to fall prevention,
ing population of older adults: (1) practice-based or grass roots chronic disease self-management, and specific lifestyle behaviors
programs promoted by community-based organizations; and such as physical activity or healthy nutrition (16–18).
(2) research-tested programs developed and tested in academic Less is known, however, about the structure, content, and ben-
research centers. efits of grass roots health programs that have been delivered by
Included in the first programmatic stream, health promo- a variety of community organizations across vast geographic dis-
tion/disease prevention programs have traditionally been deliv- tances (i.e., regional or state-wide health promotion campaigns
ered by non-academic community-based practitioners with the or community walking programs). Further, little attention has
generic goals of maintaining or improving health (5–8). Because been given to understand how these programs might contribute
these programs are delivered in real-world settings, they have a to practice-based evidence. Given the recent emphasis on admin-
greater potential for large population reach and long-term sus- istrative policies within the Administration on Community Living
tainability; however, they are typically unstructured and hence (ACL), which give funding preference to EBPs (19), it is especially
not easily replicable or testable. Further, even when based on important to understand how these long-standing, community-
“best practices,” such programs may have minimal attention to based health promotion programs have functioned in the past and
behavioral change theories and little to no formal evaluation how they might be adapted to permit formal evaluation and be
to document intervention effectiveness or characterize essential eligible for governmental funding streams.
program elements (9–12). This perspective article presents a case study of Texercise, a
A second programmatic stream involves the recent nation- community-based health promotion movement established in
wide movement toward the widespread dissemination of 1999 to help Texans ages 45 and older live healthier lives. Based on
evidence-based programs (EBPs) for older adults that were a historical review of existing Texercise materials, supplemented

www.frontiersin.org April 2015 | Volume 2 | Article 291 | 261

Ory et al. From Texercise Classic to Texercise Select

by information provided by key Texercise staff, this article will in 2012. Further, since 2003, over 160,000 Texercise handbooks
examine the processes employed to structure the original Texer- (available at www.texercise.com) about how to initiate an exercise
cise program (referred to as Texercise Classic) so that it could program have been distributed to individuals wanting to exer-
be formally evaluated. Specifically, this article will: (1) document cise on their own (22). Despite its widespread reach across Texas
the natural history of Texercise Classic, including its programmatic and national recognition (e.g., the International Council on Active
reach and adoption across Texas over the past decade and a half; (2) Aging Industry Innovator Award; President’s Council on Fitness,
describe the rationale and processes employed to formalize Tex- Sports, and Nutrition Community Leadership Award; and the
ercise Classic into a more structured program known as Texercise Texas Cardiovascular Health Promotions Award), Texercise Classic
Select ; (3) outline the essential elements and activities included in had never been formally evaluated.
Texercise Select and contrast them with those included in Texercise In 2012, a contract was awarded to the Texas A&M Health Sci-
Classic; and (4) highlight key components for uniform facilitator ence Center in collaboration with Baylor Scott & White Health
training. The discussion will reflect upon the evolution of Texer- to review and evaluate the program. The primary aims of this
cise, the comparisons across the two program types, and the “Next contract were to formalize the processes and procedures (includ-
Steps” for Texercise Select. ing materials and facilitator training) and collect more detailed
information from participants with the hopes of establishing a rig-
NATURAL HISTORY orous, scientific evidence base for this program. This evaluation
Texercise Classic emerged from a vision by the state public health has generated a new phase of activity.
and aging services leadership to help the growing number of Tex-
ans age well. We highlight some salient events in the development FORMALIZATION PROCESSES AND PROCEDURES
and evolution of Texercise. Starting as a state-wide public health PROCESS OVERVIEW
campaign in the late 1990s, this grass roots program was officially We identified in the beginning stages of the evaluation project
launched in 2002 as part of the Governor’s Challenge Walk for that the loosely structured nature of the existing Texercise Clas-
Wellness. Under the Governor’s Office, Texercise was envisioned sic program would make program evaluation difficult. Initially,
as a state-wide health promotion program to encourage individ- Texercise Classic was designed as a participant-driven grass roots
uals and communities to adopt healthy lifestyle habits such as program in which participants, in collaboration with group lead-
physical activity and good nutrition. In 2005, strong endorsement ers, decide upon the nature and amount of group exercises. While
was received from the Governor’s Office through Executive Order this strategy offered substantial choice, the lack of uniformity
(20), which stated that “The Department of Aging and Disability between workshops offered made it difficult to examine effective-
Services, Department of State Health Services, Governor’s Advi- ness and generalize to all Texercise programs. The Texas A&M
sory Council on Physical Fitness, and other appropriate state and research team and DADS staff jointly decided to utilize exist-
community organizations shall continue to promote and expand ing program materials and activities to create a more formally
the internationally recognized Texercise program as a means to designed program. As indicated in Table 1, the resulting “struc-
ensure healthy lifestyles in older Texans.” In 2006, under the aus- tured” program is known as Texercise Select. Texercise Select is
pices of the Texas Department of Aging and Disability Services implemented in 12 weeks, which includes 2 weeks for partici-
(DADS), Texercise became more formalized with the creation of pant recruitment and 10 weeks of 1.5-hour sessions conducted
a 12-week face-to-face fitness program, with the tag line “Fit for twice a week. Utilizing evidence-based skills and tools, each ses-
the Health of It!”, and the identification of community-based vol- sion incorporates interactive educational discussions, interactive
unteers or “program champions” to promote Texercise. In 2009, activities related to physical activity and/or nutrition topics, and
with input from experts such as Dr. Kenneth Cooper from the 30–45 minutes of actual exercise.
internationally known Cooper Clinic (21), the 12-week program
and materials were updated to include attention to both physical ESSENTIAL INTERVENTION ELEMENTS
activity and nutrition. The basis for the nutritional content was Pulling from foundational concepts in evidence-based health
the existing information sheets developed by nutrition experts at and wellness programs (23–26), the research team developed the
DADS. We also utilized some standard nutrition items that other “structured” Texercise Select program, manual, and training that
EBPs employ as well, such as the USDA my plate, and had all operationalized essential intervention elements and processes. To
materials reviewed by a nutrition expert at the Texas A&M Health accomplish this task, the research team reviewed the literature as
Science Center School of Public Health. well as comparable EBPs. This review enabled the team to iden-
Thanks to the collaboration between DADS and its partners, tify key exercise and behavior change elements that would work
what has become known as Texercise Classic is available free-of- best in the Texercise context (e.g., ideal session length, ideal class
charge and includes resources and incentives such as pedometers, duration, and types of effective exercises). Drawing on social cog-
resistance bands, pledge sheets, and 12-week daily fitness and nitive learning principles (27), the entire program was designed
nutrition logs distributed to participants during the program. heavily around the concept of self-efficacy with a goal of having
Using a lay-leader facilitator model, Texercise Classic has been participants take a more active role in their health through health
delivered through a variety of settings including worksites, senior choices and behaviors. The underlying programmatic intent was
centers, faith-based organizations, and long-term care facilities. to increase self-efficacy and behavioral skills so that participants
Texercise Classic has reached more than 15,000 Texans starting would continue to engage in healthy aging activities after the
with 794 participants in 2006 and growing to 3,400 participants program ended.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 291 | 262
Ory et al. From Texercise Classic to Texercise Select

Table 1 | Texercise Select topics, objectives, and resources by week.

Week Sessions and topics Objectives

1&2 None: participant recruitment

3 Session 1: Ready, Set, Get Active: Launching Describe the 11 principles of physical activity success
an Active Lifestyle Identify their personal exercise levels
Understand the importance of a warm-up and cool-down
Set realistic goals related to physical activity
Session 2: Ready, Set, Eat Healthy! Healthy Understand the benefits of healthy eating as well as the nutritional components of a healthy diet
Eating for a Healthy Lifestyle Make an achievable and realistic nutrition goal
Describe the purpose of a nutrition log

4 Session 3: Ready, Set, Get Moving! Getting Recognize the essential components of being and staying physically active
& Staying Physically Active Practice endurance, strength, balance, and stretching exercises safely and correctly
Create an action plan
Session 4: Ready, Set, Eat Healthy! Eating a Recognize the essential components of a balanced and healthy diet
Balanced and Healthy Diet Practice new exercises safely and correctly
Select and incorporate five sources of fruits and vegetables into their diets

5 Session 5: Ready, Set, Hydrate! Hydration for Explain the basis and requirements of proper hydration
Health Practice previous exercises safely and correctly
Identify barriers and apply problem-solve skills when action planning
Session 6: Ready, Set, Eat Proper Portions! Understand healthy portion sizes for most types of foods
Establishing a Sense of Portion Control Practice new exercises safely and correctly
Identify ways to eat sensible portions

6 Session 7: Ready, Set, Go Endurance! A Identify ways to safely increase endurance

Focus on Endurance Practice exercises safely and correctly
Evaluate previous action plan and apply strategies to overcome challenges with personal action
plans
Session 8: Ready, Set, Decode Food Labels! Explain the fundamental components of a food label
Understanding Food Labels Practice new exercises safely and correctly
Evaluate dietary logs and action plans and apply strategies to overcome challenges

7 Session 9: Ready, Set, Prevent Injury! Injury Identify and apply injury prevention methods before, during, and after physical activity
Prevention for Better Health & Safety Practice safe and correct exercises
Evaluate previous action plans and apply strategies to overcome challenges
Session 10: Ready, Set, Cook Healthy! Identify and apply healthy cooking modifications to maximize nutritional intake
Cooking Healthy for Improved Nutrition Practice new exercises safely and correctly
Recognize challenges and apply strategies to improve dietary behaviors

8 Session 11: Ready, Set, Get Strength Training Understand and apply the fundamentals of strength training introduced in class
Practice exercises safely and correctly
Evaluate previous action plans and challenges
Identify and apply strategies to overcome challenges with personal action plans
Session 12: Ready, Set, Eat Out Healthy: Eat Identify and apply strategies to make healthy choices when eating outside the home
Healthy When Dining Out Practice exercises safely and correctly
Create a healthy meal or menu

9 Session 13: Ready, Set, Don’t Stress! Stress Recognize and discuss healthy behaviors that reduce stress
Management & Mental Health Practice exercises safely and correctly
Evaluate previous actions plans
Session 14: Ready, Set, Prevent Chronic Identify and select strategies to overcome challenges with action plans/goals
Illness! Healthy Preventive Behaviors Recognize and discuss healthy behaviors
Practice exercises safely and correctly
Identify ways to prevent and better manage chronic illnesses

(Continued)

www.frontiersin.org April 2015 | Volume 2 | Article 291 | 263

Ory et al. From Texercise Classic to Texercise Select

Table 1 | Continued

Week Sessions and topics Objectives

10 Session 15: Ready, Set, Keep Fitness Fun! Practice exercises safely and correctly
Keeping Fitness Fun Identify ways to make long-term fitness enjoyable
Identify and select strategies to overcome challenges with action plans/goals
Session 16: Ready, Set, Eat Healthy! Eating Identify healthy eating alternatives during the holidays
Healthy During the Holidays Practice exercises (safely and correctly)
Recognize unhealthy eating habits

11 Session 17: Ready, Set, Stay Committed! Practice exercises safely and effectively
Staying Committed to Fitness & Review Apply the two-step approach to creating an action plan
Identify and apply strategies to overcome challenges
Identify and apply safe ways to stay physically active (review)
Session 18: Ready, Set, Stay Nutritious! Identify and apply ways to stay committed to nutritional goals and healthy eating
Keeping Nutrition a Lifestyle & Review Practice new exercises safely and correctly
Identify and apply skills maintaining a healthy lifestyle

12 Session 19: Ready, Set, GO! Moving Identify and apply ways to stay physically active and eat nutritiously
Forward Successfully Practice exercises safely and correctly
Identify and apply strategies to overcome barriers of physical activity and healthy eating
Session 20: Ready, Set, CELEBRATE! Identify and apply ways to stay committed to physical activity nutritional goals and healthy eating
Demonstrate exercises safely and correctly

Texercise Select sessions were organized around a “Ready, Set, Given the high probability of behavioral relapse in achieving
Go, Stay” rubric developed by our our Texas A&M program one’s desired lifestyle behavioral goals (24), the curriculum was
designers to help participants know what they needed to do to ini- designed for 10 weeks to increase the likelihood that behaviors
tiate healthier behaviors, engage in those behaviors, and to make would be adopted and become habit through ongoing reinforce-
them part of their everyday routines. Typically, each session has a ment. It included attention to explicit strategies for helping partici-
physical activity and nutrition educational component, with the pants stay committed. This involved hands-on practice of different
exception of the weekly session that focused more generally on behavioral skills (e.g., goal setting) combined with discussion
managing emotional issues and general lifestyle behaviors. Using about ways to overcome barriers and meet physical activity and
handout materials that had already been developed by experts for healthy eating goals. Additionally, participants were encouraged to
Texercise Classic (22), the course activities were intended to help incorporate more walking into their daily routines and use the Tex-
participants apply strategies for enhancing healthy lifestyle behav- ercise workout DVD at least 1 day each week outside of class (rein-
iors. For example, group brainstroming was utilized to help partic- forced by program facilitators at the conclusion of each session).
ipants identify solutions to common barriers. Class instructors also
taught participants the essence of action planning – identifying, COMPARISON OF TEXERCISE CLASSIC AND SELECT
setting, and implementing realistic goals. Table 2 compares the elements of Texercise Classic with those
Although the program was structured (e.g., class length, dis- of Texercise Select. When compared to Texercise Classic, Texercise
cussion topics, and types of exercise specified), it was designed to Select has some similarities and many substantial differences. We
be highly participatory and interactive with participants learning draw upon Schulz and colleague’s (33) taxonomy of interventions
to actively engage in behavior change principles such as goal set- to describe some of the most prominent similarities and differ-
ting, problem solving, tracking behaviors, and providing support ences. When Texercise Classic was first designed, less was known
to fellow class participants. In addition to a peer-to-peer learning about best practices for exercise training for older adults, and the
approach, Texercise Select was built around a lay-leader model, program concepts were more implicitly related to best practices
which has proven highly successful in the delivery of other EBPs (rather than explicitly related to best practices). In contrast, Tex-
for older adults (17, 28–30). This is consistent with the new exer- ercise Select was designed by individuals with formal training in
cise guidelines for older adults that stress the importance of risk exercise science and behavioral science as related to older adults.
management in the delivery of physical activity programs (25, 26, As such, this version of the program has benefited from an emerg-
31). Formal training sessions (i.e., a 6-hour group training) were ing science and practice base in both of these disciplines (26).
hosted by the research team to provide lay-leaders (known in Tex- Additionally, when creating Texercise Select, the developers drew
ercise Select as facilitators) with information about how to safely upon the RE-AIM and other public health frameworks (34–36), for
introduce exercises. These training sessions were supplemented understanding the importance of key implementation and dissem-
with course material that included screening questions and safety ination elements such as maximizing population reach, adoption,
tips for participants (32). implementation, and sustainability.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 291 | 264
Ory et al. From Texercise Classic to Texercise Select

Table 2 | Comparison of features in Texercise Classic and Select.

Texercise Classic Texercise Select

THEORETICAL UNDERPINNINGS
Built around best practices for Not explicit, but implicit through endorsement of Yes, using ACSM for older adult guidelines
exercise training Dr. Kenneth Cooper
Built upon best practices for Actual theoretical basis not clearly stated Social cognitive theory (self-efficacy and other behavior
behavioral change change principles)
RE-AIM framework
Diffusion of innovation
PROGRAM STRUCTURE, APPROACH, AND POPULATION TARGET
Total program duration 12 weeks of chosen exercise/activity 12 weeks: 2 weeks of recruitment plus 10 weeks, 2×/week
of actual sessions equals 20 sessions total
Number of weeks of active 12 weeks of chosen exercise 10 weeks, 2×/week of actual classes equals 20 sessions
intervention total
Amount of time per class Variable Structured – 90 minutes
Sensitivity to participant Can be lay lead, leaders can be representative of Can be lay led, leaders can be representative of the
characteristics the participants participants
Participants do not need to be cognitively able to Participants must be cognitively able to understand the
understand possible educational components educational component including action planning
Spanish materials are available Not yet translated to Spanish
PROGRAM DESIGN
Intervention manual General guidance with class instructors are only Structured manual for program facilitators with detailed
given the Texercise pink packet that includes: session outlines
Promotional DVD
Texercise handbook
Pledge sheets
Incentives (pedometers, t-shirts, etc.) and have
access to the online resources
Adaptability All aspects can be adapted, except involving some Exercises can be adapted to participant level of PA
sort of PA
Anyone can make an adaptation to the program Field coordinators/class facilitators cannot make adaptation
including sites and class leaders to essential features of the program
Adaptations can be made at any time
PROGRAM CONTENT
Attention to physical activity and Possibility with the fact sheets Built into the program
nutrition Not necessary or monitored to see if the info First session of the week deals with a physical activity topic
provided to participants is factual Second session of the week deals with a nutrition topic
Use of information sheets Optional Integrated into class curriculum
Opportunity for engaging in in-class Yes Yes
exercises
Recommended exercises Variable Drawn from prescribed list with goal of 30–45 minutes of
exercise per session that must include flexibility, strength,
balance, and endurance
Opportunity for interactive class Due to the variability of the classes this is unknown Utilizes action planning and brainstorming
discussions on goal setting and Goal setting and problem solving are not Physical activity and dietary logs are kept through the first
problem solving specifically addressed in the classic class half of the sessions
Uses incentives for behavior change
Tracking and monitoring behavior (logs)
Teaches problems solving
Provides skill building (i.e., learning exercises)
Provides social support
TRAINING AND EVALUATION
Training of instructors Variable Structured – 1 day 6 hour training
Pre-post assessment None Part of curriculum
Fidelity monitoring None On-site class fidelity checklist
Post class survey for participants

www.frontiersin.org April 2015 | Volume 2 | Article 291 | 265

Ory et al. From Texercise Classic to Texercise Select

Although both programs are viewed as 12-week programs (with Facilitators were then critiqued by trainers and any issues discussed
10 weeks of active programing), the structure varies across the two and clarified.
programs, which makes the actual program duration and class Twenty-nine facilitators were trained during this pilot study.
time likely to vary as well. Texercise Classic and Select both provide The curriculum was originally developed with two Texas A&M
time for participants to engage in group exercises and demonstrate trained facilitators per class in mind. Once the research team began
some sensitivity to individual participant’s needs, preferences, and working with organizations to identify implementation sites and
level of physical functioning. One primary difference relates to facilitators, it became apparent the class would most likely be
program flexibility in that Texercise Classic operates under gen- led by one trained facilitator with assistance from another per-
eral guidance in contrast to Texercise Select that has a detailed son who had not gone through the formal A&M training session.
implementation manual, which limits any adaptation to essen- Given the pilot nature of this demonstration study, the Texas A&M
tial program characteristics or general program flow. Given the class trainers were available by telephone and email to provide
flexibility in structure and lack of detailed facilitator manual, it additional assistance to newly trained facilitators.
is assumed that Texercise Classic attends less consistently to both
physical activity and nutrition aspects of healthy living. Texer- SUMMARY OF EVIDENCE-BASED STEPS
cise Classic was also assumed less likely to utilize the information We employed several steps in transforming Texercise Select into a
sheets, demonstrate specific exercises, and promote interactive testable and replicable EBP. These involved: (1) inventorying the
class discussions about goal setting and problem solving. However, current literature to identify foundational concepts in evidence-
the extent to which this is true is unknown because Texercise Clas- based health and wellness programs, with special emphasis on
sic has never been formally evaluated. Finally, the two programs strategies for promoting participant’s self-efficacy for engaging
differ in instructor training, evaluation, and fidelity monitoring in physical activity and peer-to-peer learning; (2) evaluating the
with only Texercise Select including a pre- and post-assessments match between existing programmatic elements and anticipated
and a fidelity checklist as part of the program. While Texercise delivery capacity and structure to ensure program adoptability and
Select was evaluated as part of a research study, it should be noted maximal population coverage; (3) organizing the sessions around
that the capacity for on-site fidelity check monitoring in grand a “ready, set, go, stay” framework for ease of implementation;
scale dissemination efforts may be limited. (4) developing a standardized manual and training protocol; (5)
incorporating fidelity checks and quality assurance into the imple-
TEXERCISE FACILITATOR TRAINING mentation and evaluation processes; and (6) identifying a practical
When developing Texercise Select, the research team decided to uti- measurement battery to assess pre–post intervention outcomes.
lize the term “facilitator” for program “lay” or “peer” leaders who
are typically community volunteers versus health professionals. DISCUSSION
This decision was intended to emphasize that facilitators are not As indicated in our brief historical review, the evolution of Tex-
experts; rather, their role was to “facilitate” participants’ ability to ercise mirrors many of the critical steps taken during the devel-
influence their health and functioning by presenting them with the opment and evaluation of an evidence-based health promotion
concepts and exercises included within the program. The facilita- program. This review also demonstrates the interaction between
tor training was seen as essential for maintaining treatment fidelity state-wide policy priorities, community practice, and research.
(37, 38). Total facilitator training time consisted of one 6-hour day, Unlike research-based programs that often struggle for scalability,
with a structured training manual to which facilitators could refer its state governmental sponsorship made Texercise Classic widely
after training. As specified in the training manual (39), the facili- available and disseminated through existing community partners
tator training was divided into five main topic blocks, each lasting and delivery systems even before it was formally evaluated. Fur-
between 30 minutes and 1.5 hours. Topics were delivered by the ther, participation in Texercise Classic has grown state-wide for
Texas A&M trainers through an interactive lecture style, including over 15 years, confirming the importance of high level endorse-
activities that allowed facilitators to apply presented information ments and community buy-in for achieving long-term program
and elicit group feedback. sustainability (40–42). Its relatively low cost and use of volun-
The training included a brief program overview of Texercise teer networks have also been probable factors in its successful
Select and an introduction to the format of each session in the dissemination (43).
curriculum. The training provided an opportunity for facilitators Although there have not been systematic studies, it is likely that
to observe and practice selected exercises, as well as to observe Texercise Classic gained steam, in part, because it was endorsed by
and demonstrate their ability to engage in group facilitation roles. the Governor, codified in executive orders, supported for imple-
During the demonstration session, trainers played the roles of mentation as part of state services, and stimulated through active
the facilitators during a session and facilitators played the role of encouragement of public–private partnerships. Texercise Classic
participants. grew from a public health campaign with community-friendly
To demonstrate competency to lead a Texercise class, facilitators handouts to a face-to-face group program based on best prac-
were assigned a Texercise session and tasked with four activities: tices and expert opinion. As evolved by science, Texercise Classic
(1) identifying the session topic; (2) identifying session materials progressed from an exercise program focused on physical activ-
needed; (3) identifying session activities; and (4) choosing one ity to a behaviorally based program including attention to both
exercise from each exercise category (i.e., one warm-up, one upper physical activity and nutrition. Additionally, over time, with the
body, and one lower body strength activity) and demonstrating it. movement toward evidence-based programing with replicable and

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 291 | 266
Ory et al. From Texercise Classic to Texercise Select

demonstrated effects, the program was redesigned as Texercise attempting to simultaneously achieve public health reach and
Select to include explicit attention to best practices about exercise effectiveness (48), and remains an area for future study.
training and behavioral change found in other successful lifestyle The research team was able to redesign Texercise and con-
programs that meet the highest tier criteria Evidence-based duct standardized training within two months. This accelerated
Disease Prevention and Health Promotion Program (19). timeframe was possible because of the research team’s familiarity
The development of new programs or the formalization of with EBPs and leader training as well as the insights provided by
existing programs can expand the evidence base as it pertains the original state-based developers of Texercise. Working together,
to older adult health and wellness. While community-generated opportunities, challenges, and potential solutions were identified.
programs such as Texercise Classic may have already demon- Texercise has a brand that is already established throughout the
strated success based on their reach and adoption, new policies state with an existing network of partners. This brand was capital-
from federal funders in the U.S. aging services sector are restrict- ized upon during the transition from Texercise Classic to Texercise
ing reimbursement to reproducible health promotion programs Select. While a new name was considered for Texercise Select,
with proven benefits (16, 19). As such, this case study illustrates the Texercise name was kept to ensure recognition and consis-
the processes and procedures involved in the formalization of tency. Both programs will continue to be promoted and supported
this community-generated program to advance its sophistication, in Texas because DADS sees value in allowing their partners to
replicability, and likelihood of evoking health benefits among its choose the appropriate program to offer based on their settings
participants. To mirror requirements for EBP status, Texercise and participants.
Select is now characterized by a set of essential features including
formal manual and training infrastructure for widespread delivery NEXT STEPS
(44, 45). Thus, the process described in this review represents the The purpose of this article was to illustrate the evolution of a grass
first steps toward formalizing Texercise Select, which is undergoing roots program to become a theoretically derived and research-
systematic evaluation to examine program effectiveness. Looking tested program. The further expansion of Texercise Select is depen-
forward, the RE-AIM framework (46) will be used as a guide for dent, in part, upon demonstrating positive outcomes comparable
examining strategies for demonstrating Texercise Select program to those found in similar EBPs for seniors. An initial pilot test
effectiveness and public health impact, especially around program of the feasibility of implementation and outcomes was conducted
implementation, scalability, and sustainability issues. in 2013. Preliminary results are promising (49, 50), with signif-
In contrasting Texercise Classic and Select, it is important to icant pre-post improvements (P < 0.05) seen in positive health
understand the benefits and challenges of both programs. Texercise behaviors (i.e., increased aerobic activity, weekly fruit/vegetable
Classic has demonstrated its widespread appeal and sustainability consumption, and daily water consumption) with large effect sizes
by continual delivery for more than ten years by volunteer facili- for physical activity and smaller ones for nutrition behaviors.
tators who do so without external financial support. As indicated Additionally, enhanced dissemination of Texercise Select requires
in interviews with stakeholders (47), it is a program that has name infrastructure resources such as the widespread availability of stan-
recognition, is easy to implement, and is well-liked by program dardized training. Toward this end, DADS is updating the training
facilitators and participants alike. However, its greatest strength – and implementation manual so it will be web-based and eas-
flexibility in the content and type of delivery – is also its greatest ily accessible by community partners and potential facilitators.
potential weakness as an EBP. Such flexibility makes it difficult to Although initial outcome results are promising, further study is
replicate consistently, know exactly what program components are needed to understand factors associated with the ability of Tex-
being implemented, and measure the extent to which participants ercise Select to be widely disseminated and sustained over time.
are benefiting (or in what ways). Specific outcomes are unknown Once the results of initial pilot testing from 2013 are fully pub-
but are likely to be quite variable and affected by individual delivery lished, we recommend a state-wide campaign with DADS’s current
settings, facilitators, and participant populations. Texercise partners to help spread the word about the benefits of
Conversely, Texercise Select provides structured training for implementing evidence-based programing for seniors and how
facilitators and a scripted curriculum that, if followed, should Texercise Select might be broadly disseminated through existing
result in positive health outcomes similar to those of other community channels.
evidence-based lifestyle programs. Yet, some existing partners who
primarily offer exercise programs might not like (or be able to ACKNOWLEDGMENTS
implement) the reconfigured program with fidelity. For example, We thank study participants, research, and administrative staff.
some park and recreation programs might easily adapt the exer- This study was supported by a grant from the Texas Depart-
cise training part but not be as comfortable with facilitating the ment of Aging and Disability Studies. The research presented was
behavioral lifestyle educational aspects. also supported in part through the Community Research Cen-
However, in terms of program impact, Texercise Select is likely ter for Senior Health, a joint partnership between Scott & White
to be more effective than Texercise Classic in changing lifestyle Healthcare, Texas A&M Health Science Center, and the Central
behaviors because of its standardized incorporation of evidence- Texas Aging & Disability Resource Center. The Center is funded
based behavioral change principles. Yet, it is also unknown whether through a grant from the National Institute on Aging (Award
Texercise Select will be as appealing to community organizations Number RC4AG038183-01). The contents of this work are solely
and able to sustain the same reach and adoption as Texercise Clas- the responsibility of the authors and do not necessarily represent
sic. This issue illustrates a potential trade-off often seen when the official views of DADS or NIH.

www.frontiersin.org April 2015 | Volume 2 | Article 291 | 267

Ory et al. From Texercise Classic to Texercise Select

REFERENCES 24. Ory MG, Smith ML, Mier N, Wernicke MM. The science of sustaining health
1. Ory MG, Smith ML, Resnick B. Changing behavior throughout the life-course: behavior change: the health maintenance consortium. Am J Health Behav (2010)
translating the success of aging research. Transl Behav Med (2012) 2:1–4. 34(6):647–59. doi:10.5993/AJHB.34.6.2
doi:10.1007/s13142-012-0129-4 25. Cress M, Buchner D, Prohaska T, Rimmer J, Brown M, Macera C, et al. Best
2. Prohaska TR, Anderson LA, Binstock RH. Public Health for an Aging Society. practices for physical activity programs and behavior counseling in older adult
Baltimore: JHU Press (2012). 456 p. populations. Eur Rev Aging Phys Act (2006) 3(1):34–42. doi:10.1007/s11556-
3. Haber D. Health Promotion and Aging: Practical Applications for Health Profes- 006-0003-9
sionals. Sixth ed. New York: Springer Publishing Company (2013). 513 p. 26. Chodzko-Zajko W, American College of Sports Medicine. ACSM’s Exer-
4. Ory M, Kinney Hoffman M, Hawkins M, Sanner B, Mockenhaupt R. Challeng- cise for Older Adults. Philadelphia: Lippincott Williams & Wilkins (2013).
ing aging stereotypes: strategies for creating a more active society. Am J Prev Med 256 p.
(2003) 25(3):164–71. doi:10.1016/S0749-3797(03)00181-8 27. Bandura A. Social learning theory. Englewood Cliffs, NJ: Prentice Hall (1977).
5. National Prevention Council. National Prevention Strategy. Washington, DC: 247 p.
U.S. Department of Health and Human Services, Office of the Surgeon General 28. Bodenheimer T, Chen E, Bennett HD. Confronting the growing burden of
(2011). chronic disease: can the US health care workforce do the job? Health Aff (2009)
6. National Center for Chronic Disease Prevention and Health Promotion. Helping 28(1):64–74. doi:10.1377/hlthaff.28.1.64
People to Live Long and Productive Lives and Enjoy a Good Quality of Life. Atlanta, 29. Brady TJ, Murphy L, O’Colmain BJ, Beauchesne D, Daniels B, Greenberg M,
GA: Centers for Disease Control and Prevention (2009). et al. A meta-analysis of health status, health behaviors, and health care utiliza-
7. Task Force on Community Preventive Services. Recommendations to increase tion outcomes of the chronic disease self-management program. Prev Chronic
physical activity in communities. Am J Prev Med (2002) 22(4 Suppl):67–72. Dis (2013) 10:120112. doi:10.5888/pcd10.120112
doi:10.1016/S0749-3797(02)00433-6 30. Wilcox S, Dowda M, Griffin SF, Rheaume C, Ory MG, Leviton L, et al. Results of
8. National Center for Injury Prevention and Control. Preventing Falls: How to the first year of active for life: translation of 2 evidence-based physical activity
Develop Community-Based Fall Prevention Programs for Older Adults. Atlanta, programs for older adults into community settings. Am J Public Health (2006)
GA: Centers for Disease Control and Prevention (2008). 96(7):1201–9. doi:10.2105/AJPH.2005.074690
9. Bryant LL, Altpeter M, Whitelaw NA. Evaluation of health promotion pro- 31. Ory M, Resnick B, Jordan PJ, Coday M, Riebe D, Garber CE, et al. Screen-
grams for older adults: an introduction. J Appl Gerontol (2006) 25(3):197–213. ing, safety, and adverse events in physical activity interventions: collaborative
doi:10.1177/0733464806288562 experiences from the behavior change consortium. Annals Behav Med (2005)
10. Glanz K, Bishop DB. The role of behavioral science theory in development and 29(2):20–8. doi:10.1207/s15324796abm2902s_5
implementation of public health interventions. Annu Rev Public Health (2010) 32. Program on Healthy Aging. EASY Exercise and Screening for You. Texas A&M
31:399–418. doi:10.1146/annurev.publhealth.012809.103604 Health Science Center School of Public Health (2008). Available from: http:
11. Harris JR, Cheadle A, Hannon PA, Lichiello P, Forehand M, Mahoney E, et al. A //easyforyou.info/
framework for disseminating evidence-based health promotion practices. Prev 33. Schulz R, Czaja SJ, McKay JR, Ory MG, Belle SH. Intervention taxonomy (ITAX):
Chronic Dis (2012) 9:E22. describing essential features of interventions (HMC). Am J Health Behav (2010)
12. Merzel C, D’afflitti J. Reconsidering community-based health promotion: 34(6):811–21. doi:10.5993/AJHB.34.6.15
promise, performance, and potential. Am J Public Health (2003) 93(4):557–74. 34. Gaglio B, Shoup JA, Glasgow RE. The RE-AIM framework: a systematic review
doi:10.2105/AJPH.93.4.557 of use over time. Am J Public Health (2013) 103(6):e38–46. doi:10.2105/AJPH.
13. Brownson RC, Fielding JE, Maylahn CM. Evidence-based public health: a fun- 2013.301299
damental concept for public health practice. Annu Rev Public Health (2009) 35. Lobb R, Colditz GA. Implementation science and its application to popula-
30:175–201. doi:10.1146/annurev.publhealth.031308.100134 tion health. Annu Rev Public Health (2013) 34:235–51. doi:10.1146/annurev-
14. Administration on Aging. AoA Evidence-Based Prevention Program. Washington, publhealth-031912-114444
DC: Department of Health and Human Services (2013). 36. Brownson RC, Colditz GA, Proctor EK. Dissemination and Implementation
15. Administration for Community Living. ARRA – Communities Putting Prevention Research in Health: Translating Science to Practice. New York: Oxford University
to Work: Chronic Disease Self-Management Program. Atlanta, GA: Department Press (2012). 560 p.
of Health and Human Services (2012). 37. Bellg AJ, Borrelli B, Resnick B, Hecht J, Minicucci DS, Ory M, et al. Enhancing
16. National Council on Aging. Where to Find Evidence-Based Programs. Wash- treatment fidelity in health behavior change studies: best practices and recom-
ington, DC (2013). Available from: http://www.ncoa.org/improve-health/ mendations from the NIH behavior change consortium. Health Psychol (2004)
center-for-healthy-aging/where-to-find-evidence-based.html 23(5):443–51. doi:10.1037/0278-6133.23.5.443
17. Ory MG, Smith ML, Wade A, Mounce C, Wilson A, Parrish R. Implementing 38. Frank JC, Coviak CP, Healy TC, Belza B, Casado BL. Addressing fidelity in
and disseminating an evidence-based program to Prevent falls in older adults, evidence-based health promotion programs for older adults. J Appl Gerontol
Texas, 2007-2009. Prev Chronic Dis (2010) 7(6):A130. (2008) 27(1):4–33. doi:10.1177/1090198114543007
18. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, Whitelaw N, et al. Successes 39. Texas Department of Aging and Disability Services. Texercise Training Manual.
of a national study of the chronic disease self-management program: meet- Available from: http://www.dads.state.tx.us/texercise/resources/index.html
ing the triple aim of health care reform. Med Care (2013) 51(11):992–8. 40. Evashwick C, Ory M. Organizational characteristics of successful innovative
doi:10.1097/MLR.0b013e3182a95dd1 health care programs sustained over time. Fam Community Health (2003)
19. Administration on Aging. Disease Prevention and Health Promotion Services 26(3):177–93. doi:10.1097/00003727-200307000-00003
(OAA Title IIID). Washington, DC: Department of Health and Human Services 41. Estabrooks PA, Smith-Ray RL, Dzewaltowski DA, Dowdy D, Lattimore D,
(2014). Rheaume C, et al. Sustainability of evidence-based community-based physical
20. Office of the Governor Rick Perry. Executive Order RP42 – Relating to the Cre- activity programs for older adults: lessons from active for life. Transl Behav Med
ation of the Aging TexasWell Advisory Committee and Plan. (2005). Available (2011) 1(2):208–15. doi:10.1007/s13142-011-0039-x
from: http://governor.state.tx.us/news/executive-order/3687/ 42. Scheirer MA, Dearing JW. An agenda for research on the sustainability of pub-
21. Cooper KH. Texans Urged to Follow Kids’ Lead; Shed “Fattest” Label. Texas lic health programs. Am J Public Health (2011) 101(11):2059–67. doi:10.2105/
Department of Aging and Disability Services (2013). Available from: http: AJPH.2011.300193
//www.dads.state.tx.us/texercise/partners/cooper.html 43. Rogers EM. Diffusion of Innovations. 5th ed. New York: Free Press (2003).
22. Texas Department of Aging and Disability Services. Texercise: A How-to Hand- 551 p.
book for Lifelong Health and Wellness. (2014). Available from: http://www.dads. 44. Altpeter M, Bryant L, Schneider E, Whitelaw N. Evidence-based health practice:
state.tx.us/texercise/resources/handbook/english/index.html knowing and using what works for older adults. Home Health Care Serv Q (2006)
23. Ory M, Jordan P, Bazzarre T. The behavior change consortium: setting the stage 25(1–2):1–11. doi:10.1300/J027v25n01_01
for a new century of health behavior-change research. Health Educ Res (2002) 45. Center for Healthy Aging. Module 2: What is EBHP?. Washington, DC: National
17(5):500–11. doi:10.1093/her/17.5.500 Council on Aging (2014).

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 291 | 268
Ory et al. From Texercise Classic to Texercise Select

46. Human Nutrition Foods and Exercise. Reach Effectiveness Adoption Implemen- of these organizations/agencies. All papers published in the Research Topic received
tation Maintenance. Blacksburg, VA: Virginia Tech College of Agriculture and peer review from members of the Frontiers in Public Health (Public Health Education
Life Sciences (2014). and Promotion section) panel of Review Editors. Because this Research Topic repre-
47. Stevens Ab, Thiel S, Thorud JL, Smith ML, Howell D, Cargill J, et al. Increasing sents work closely associated with a nationwide evidence-based movement in the US,
the availability of physical activity programs for older adults: lessons learned many of the authors and/or Review Editors may have worked together previously in
from Texercise stakeholders. J Aging Phys Activ (in press). some fashion. Review Editors were purposively selected based on their expertise with
48. National Collaborating Centre for Methods and Tools. Assessing the Public Health evaluation and/or evidence-based programming for older adults. Review Editors were
Impact of Health Promotion Initiatives. Hamilton, ON: McMaster University independent of named authors on any given article published in this volume.
(2010).
49. Ory MG, Smith ML, Jiang J, Howell D, Chen S, Pulczinski J, et al. Texercise effec-
tiveness: an examination of physical function and quality of life. J Aging Phys Received: 11 September 2014; accepted: 24 December 2014; published online: 27 April
Activ. (2015) (in press). 2015.
50. Smith ML, Ory MG, Jiang L, Howell D, Chen S, Pulczinski J, et al. Texercise select Citation: Ory MG, Smith ML, Howell D, Zollinger A, Quinn C, Swierc SM and
effectiveness: an examination of physical activity and nutrition outcomes. Transl Stevens AB (2015) The conversion of a practice-based lifestyle enhancement program
Behav Med (2015) (in press). into a formalized, testable program: from Texercise Classic to Texercise Select. Front.
Public Health 2:291. doi: 10.3389/fpubh.2014.00291
Conflict of Interest Statement: The authors declare that the research was conducted This article was submitted to Public Health Education and Promotion, a section of the
in the absence of any commercial or financial relationships that could be construed journal Frontiers in Public Health.
as a potential conflict of interest. Copyright © 2015 Ory, Smith, Howell, Zollinger, Quinn, Swierc and Stevens. This is
an open-access article distributed under the terms of the Creative Commons Attribution
This paper is included in the Research Topic, “Evidence-Based Programming for Older License (CC BY). The use, distribution or reproduction in other forums is permitted,
Adults.” This Research Topic received partial funding from multiple government and provided the original author(s) or licensor are credited and that the original publica-
private organizations/agencies; however, the views, findings, and conclusions in these tion in this journal is cited, in accordance with accepted academic practice. No use,
articles are those of the authors and do not necessarily represent the official position distribution or reproduction is permitted which does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 291 | 269

 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00187

Translation of Fit & Strong! for middle-aged and older

adults: examining implementation and effectiveness of a
lay-led model in CentralTexas
Marcia G. Ory 1 , Shinduk Lee 1 , Alyson Zollinger 1 , Kiran Bhurtyal 2 , Luohua Jiang 3 and Matthew Lee Smith 4 *
1
Department of Health Promotion and Community Health Sciences, Texas A&M Health Science Center, School of Public Health, College Station, TX, USA
2
Office of Surveillance, Evaluation, and Research, Texas Department of State Health Services, Austin, TX, USA
3
Department of Epidemiology and Biostatistics, Texas A&M Health Science Center, School of Public Health, College Station, TX, USA
4
Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA, USA

Edited by: The Fit & Strong! program is an evidence-based, multi-component program promoting
Kerry Stephen Kuehl, Oregon Health
physical activity among older adults, particularly those suffering from lower-extremity
& Science University, USA
osteoarthritis. The primary purpose of the study is to examine if the Fit & Strong! pro-
Reviewed by:
Cheryl Lynn Addy, University of South gram translated into a lay-leader model can produce comparable outcomes to the original
Carolina, USA program taught by physical therapists and/or certified exercise instructors. A single-group,
Charles D. Treser, University of pre–post study design was employed, and data were collected at the baseline (n = 136 par-
Washington School of Public Health,
ticipants) and the intervention conclusion (n = 71) with both baseline and post-intervention
USA
data. The measurements included socio-demographic information, health- and behavior-
*Correspondence:
Matthew Lee Smith, Department of related information, and health-related quality of life. Various statistical tests were used
Health Promotion and Behavior, for the program impact analysis and examination of the association between participant
College of Public Health, The characteristics and program completion. As in the original study, there were statistically
University of Georgia, 330 River
significant (p < 0.05) improvements in self-efficacy for exercise, aerobic capacity, joint stiff-
Road, 315 Ramsey Center, Athens,
GA 30602, USA ness, level of energy, and amount and intensity of physical activities.The odds of completing
e-mail: [email protected] the program were significantly lower for the participants from rural areas and those hav-
ing multiple chronic conditions. Successful adaptation of the Fit & Strong! program to a
lay-leader model can increase the likelihood of program dissemination by broadening the
selection pool of instructors and, hence, reducing the potential issue of resource limita-
tion. However, high program attrition rates (54.1%) emphasize the importance of adopting
evidence-based strategies for improving the retention of the participants from rural areas
and those with multiple chronic conditions.
Keywords: evidence-based program, aging, exercise, implementation research

INTRODUCTION disseminate Fit & Strong! to more diverse populations and settings
In recent years, there has been growing interest in evidence- (12). While originally developed for older adults with lower-
based disease prevention programs that help middle-aged and extremity osteoarthritis, it is now being marketed more broadly
older adults improve their health and quality of life through as an evidence-based physical activity/behavioral change program
self-management strategies. This greater attention has resulted, in that can be delivered to sedentary, older adults through aging
part, from an emerging recognition that adults of all ages includ- services, and public health networks (13, 14).
ing older adults can benefit from health promotion programs Despite the potential advantages of widely disseminating Fit
(1, 2) and a larger appreciation of the value of evidence-based & Strong! in community settings, some challenges were antici-
approaches (3–5). Many of these programs incorporate elements pated in the actual delivery through the aging services network in
to increase physical activity among participants (6–8) because of Central Texas. One identified translational research problem was
the importance of mobility for sustained independent living (9, resource limitation related to the inadequate availability of trained
10). Fit & Strong! is an example of one such multi-component instructors (15, 16). Prior to its translation in Central Texas, eligi-
physical activity program that combines guided aerobic, strength, ble instructors for Fit & Strong! were limited to physical therapists
and flexibility training with health education (7, 11). Previously (PTs) and certified exercise instructors (CEIs) as a means to ensure
tested in a randomized clinical trial, Fit & Strong! has demon- safety and effectiveness in conducting the program (7, 11, 17).
strated efficacy to improve participants’: (1) self-efficacy (SE), or However, this narrow pool of eligible instructors limits possibilities
confidence, for exercise; (2) physical activity adherence; (3) aerobic for grand-scale uptake and dissemination. Therefore, in collabo-
capacity; and (4) lower-extremity joint pain and stiffness (7, 11). ration with the program developers, efforts were taken to modify
After a series of successful efficacy trials, the program the instructor criteria and expand the types of instructors deemed
developers have proactively partnered with multiple agencies to appropriate to deliver the Fit & Strong! program.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 187 | 270
Ory et al. Translation of Fit & Strong! for older adults

Given the growth of other evidence-based programs deliv- to lead Fit & Strong! classes. For lay individuals to be qualified
ered using train-the-trainer, lay-leader models through a variety to lead Fit & Strong! classes, the lay-leaders needed to meet the
of aging, public health, and health care organizations, questions following criteria: (1) be certified in another evidence-based class
arose about whether Fit & Strong! might similarly be translated to (e.g., A Matter of Balance, Chronic Disease Self-Management Pro-
a lay-led model and, thus, broaden the selection pool of instruc- gram, or Diabetes Self-Management Program) and have experience
tors and minimize the anticipated resource limitation problem. and comfort leading group classes, if not already a PT or a CEI;
Of primary concern was whether instructors trained in other (2) participate in the full instructor and supplemental lay-leader
evidence-based programs, without exercise training experience, trainings conducted by the Master Trainers; and (3) adhere to
could safely deliver Fit & Strong! to seniors with non-specific fidelity standards by following the training manual in conducting
chronic conditions while maintaining program effectiveness. In the program (18, 19).
response to such questions, this study examined the adaptation of Recruitment efforts for lay-leaders consisted of collaboration
Fit & Strong! to a lay-leader model in Central Texas using a quasi- with community stakeholders who were instrumental in: (1) refer-
experimental study design. The specific purposes of this study were ring and identifying qualified/capable individuals; and (2) dissem-
to: (1) describe the characteristics of participants enrolled in the inating information (e.g., flyers and emails) about the lay-leader
translated Fit & Strong! program; (2) examine factors associated training.
with program attendance; and (3) assess changes in health-related Training for the original Fit & Strong! program was conducted
outcomes among participants. by Fit & Strong! staff, Master, and T-Trainers (e.g., the most
experienced trainers who are able to train and certify Master train-
MATERIALS AND METHODS ers). Lay individuals as well as CEIs completed the mandatory
FIT & STRONG! INTERVENTION instructor training, which lasted 8 h (in 1 day) and covered: (1)
The Fit & Strong! program is a multi-component physical activ- program background & development; (2) importance of fidelity;
ity and behavioral change program that is structured around two (3) roles/responsibilities of instructors in relation to other Fit
key components: (1) participation in group-facilitated (or guided) & Strong! team members (e.g., developers); (4) Fit & Strong!
physical activity; and (2) group-based health education/problem- exercise components (description and demonstration of vari-
solving. Over an 8-week period, individuals participate in 24 ous types of exercises used throughout the program); (5) Fit &
total sessions, meeting 3 days each week for 90-min each ses- Strong! group discussion/problem-solving component (including
sion. Each session begins with 60-min of structured physi- role plays, facilitator management roles); and (6) data collection,
cal activity, which is then followed by a 30-min group-based evaluation, and fidelity responsibilities of instructors (18, 19). Lay
discussion/problem-solving period (7, 11, 17). individuals then completed an additional day of training (half-
The physical activity component includes: (1) warm-up exer- day, 4 h) tailored to lay-leaders that emphasized basic exercise
cises (5–10 min); (2) low-impact aerobic conditioning (e.g., walk- principles and safety as they applied to the Fit & Strong! program.
ing and step aerobics) where participants begin with 10 min of
activity and gradually work up to 30 min by the end of the pro- IMPLEMENTATION AND FIDELITY OF THE TRANSLATED FIT & STRONG!
gram; (3) strength exercises (primarily lower-extremity) using PROGRAM
ankle weights and resistance bands (15–20 min); and (4) cool- During the implementation of the adapted Fit & Strong! program
down and flexibility exercises (5–10 min) (7, 11, 17). During the in Central Texas, the Texas A&M Fit & Strong! evaluation team
group-based discussion/problem-solving component, instructors along with program developers engaged in best practice quality
lead participants in discussions of various health-related topics assurance strategies to assure that the adapted program would be
relying on a program curriculum guide. These interactive sessions delivered with fidelity (20). This included: (1) fidelity assessments
are intended to help participants make healthy changes that sustain (using a specified fidelity checklist) through observations at the
long-term healthy lifestyle management (e.g., improved arthri- delivery sites; (2) setting up a mechanism for frequent communi-
tis symptom management and physical activity engagement). cation with the lay-leaders; and (3) conducting process evaluations
Toward the end of the program, participants are also encour- of program implementation and participant experiences. The pro-
aged to create an individualized physical activity plan to enable gram evaluations assessed participants in terms of: (a) attendance;
and promote continued physical activity after the 8-week program (b) experiences with the program and instructors; and (c) pro-
concludes (7, 11, 17). gram impact. Evaluations also included instructors’ experiences
and assessment of the program (instructor manual, group discus-
ADAPTATION OF THE FIT & STRONG! PROGRAM sion, and exercise components) as well as the effectiveness of the
The proposed adaptation of Fit & Strong! in Central Texas involved instructor training. The fidelity assessments and program evalu-
two modifications: (1) a shift in the required instructor qualifica- ations provided further guidance and support for instructors in
tion from exercise-experts (i.e., PTs or CEIs) to lay-leaders; and (2) conducting classes more effectively and correctly. These quality
a modification in the training protocol. In response to a shortage assurance strategies helped reinforce adherence to the curricula
of qualified instructors in the targeted communities, especially in material presented through the original program manuals.
the rural sites, program implementers at the Texas A&M Program
on Healthy Aging collaborated with the original program devel- PROGRAM SETTING AND DELIVERY
opers at the University of Illinois – Chicago to modify the existing Five intervention sites were selected from Central Texas, and 12
qualification requirement for instructors and allow lay individuals different Fit & Strong! classes were offered across the various sites

www.frontiersin.org April 2015 | Volume 2 | Article 187 | 271

Ory et al. Translation of Fit & Strong! for older adults

from September 2012 through June 2013. Site selection was based during data collection to assist older adults when filling out the
on three interrelated criteria: (1) community support for hosting forms, as needed.
Fit & Strong! classes (community “buy-in” was seen as a critical
factor for both recruitment and sustainability), (2) facility avail- Self-efficacy for exercise
ability for Fit & Strong! classes, and (3) the presence of a sufficient Self-efficacy for exercise was measured using four items. The items
number of older adults who could benefit from Fit & Strong! and asked how confident participants are in performing different types
who were interested in participating in the program. The number of exercise (e.g., strength and flexibility), performing vigorous
of participants in each class ranged from 16 to 25, which roughly exercises, and performing exercise despite pain or symptoms. Each
paralleled the recommended 20–25 participant maximum ideal item is based on a 10-point scale ranging from“not at all confident”
(21). Institutional review board approval was obtained at Texas (score = 1) to “totally confident” (score = 10). The score for SE for
A&M University. exercise was the mean of the four items. Higher SE scores indicated
Local senior centers, community centers, and health resource higher self-efficacy. The scale value was set to“missing” if more than
centers served as host agencies for the delivery of the Fit & Strong! one item was missing (23); based on the criteria, seven total miss-
classes. These host agencies also assisted with program promotion ing cases were omitted from the analyses. If only one item was
and participant recruitment. For example, a couple of agencies missing, the mean of the remaining three items was used. Internal
hosted promotional meetings for the program as well as volun- reliability was high for this composite scale (Cronbach’s α = 0.96).
tarily conducted on-site program enrollment while coordinating
Aerobic capacity, flexibility, and strength
these efforts with Texas A&M program implementers. Many of
A slight adaptation of the rapid assessment of physical activity
these agencies expressed appreciation for the offering of a new
(RAPA) was used to measure the amount and intensity of par-
program at their sites and, thus, were more willing to volunteer
ticipants’ physical activity (24). The adapted RAPA consisted of
their services to assist with promotion and recruitment endeavors.
eight items, and each item had a “yes” and “no” option. The first
six items, which measured the intensity and frequency of physical
PARTICIPANTS
activity were used to assess aerobic capacity.
Study participants included adults aged 47–94 years who enrolled
The six items were: (1) I rarely or never do any physical activity;
in Fit & Strong! in rural and urban counties in Central Texas
(2) I do some light or moderate physical activities but not every
between 2012 and 2013. All middle-aged and older adults resid-
week; (3) I do some light physical activity every week; (4) I do
ing in the area were eligible to enroll in the program; how-
moderate physical activity every week; (5) I do 30 min or more
ever, only those who had never previously participated in a
per day of moderate physical activity, five or more days per week;
Fit & Strong! class and attended the first or second class ses-
and (6) I do 20 min or more per day of vigorous physical activities,
sion and also completed a baseline survey were included in the
three or more days per week. Each of the six items reflected a spe-
study analyses (n = 136). As previously mentioned, participants
cific level of aerobic capacity. For example, affirmative response to
were recruited by host agency members as well as Texas A&M
the item “(1)” represents “sedentary” and was scored 1; affirmative
program implementers. Participants were recruited through var-
response to the item“(2)”represents“under-active”and was scored
ious sources, including print materials (e.g., program guides,
2; affirmative response to the item “(3)” represents “under-active
brochures/flyers, and newspaper postings), community resources
regular – light activities” and was scored 3; affirmative response
(e.g., senior clubs/classes and promotional meetings), family or
to the item “(4)” represents “under-active regular” and was scored
friends (word of mouth), and health care providers. The major-
4; and affirmative response to items “(5)” and/or “(6)” represents
ity of participants were recruited through print materials (43.4%)
“active” and was scored 5. The highest score among the six items
and family or friends (28.7%).
was selected for the aerobic capacity score (25). The remaining two
items assessed strength and flexibility, and affirmative response to
MEASURES
each item was scored 1. The strength and flexibility items were
Data sources included a baseline survey at the beginning (first
summed for descriptive purposes. The summed scale ranged from
and second sessions of each class), a post-test survey at the end of
0 to 2 (0 = none, 1 = either, and 2 = both).
the 8-week program (final week), and an attendance log. Demo-
graphic data that was drawn from the baseline survey included Self-rated health
age, gender, race/ethnicity, education, marital status, employment A single item was used to assess self-rated health (26), which
status, and annual household income. Several outcome measures has been identified as an outstanding predictor of future health
were extracted and analyzed based upon these baseline and post- (27). This item was a five-point scale with lower values indicating
test surveys. Primary outcomes included: (1) SE for exercise and worse health (poor = 1) and higher values indicating better health
(2) level of physical activity related to aerobic capacity, flexibility, (excellent = 5).
and strength. Secondary outcomes included: (1) self-rated health
status; (2) joint pain and stiffness; and (3) level of energy (fatigue). Joint pain and stiffness
Paralleling assessment protocols being utilized by the original pro- The Western Ontario and McMasters University Osteoarthritis
gram developers in their program dissemination phase (22), the Index (WOMAC) was used to measure lower-extremity pain and
measurement battery was designed to be administered to older stiffness (28). The adopted WOMAC consisted of seven items:
adults in community settings. The surveys were designed to be five pain and two stiffness items. All seven items were in a five-
completed on average in <20 min. Program staff was available point Likert scale structure ranging from “none” (score = 0) to

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 187 | 272
Ory et al. Translation of Fit & Strong! for older adults

“extreme” (score = 4). Scores for each section were summed to

produce composite scales for pain and stiffness. The pain-scale
ranged from 0 to 20 with higher values indicating greater pain; and
the stiffness-scale ranged from 0 to 8 with higher values indicating
greater stiffness. Internal reliabilities were high for both composite
scales (Cronbach’s α = 0.89 for pain; 0.86 for stiffness) (29).

Level of energy and fatigue

The level of energy and fatigue was measured using five items
(30). Each item was a six-point scale ranging from “none of the
time” (score = 0) to “all of the time” (score = 5). Some scores were
recoded to have an equal direction of answers among the five
items (i.e., higher scores indicate worse health). The mean of the
five items was used as the composite scale for the level of energy
and fatigue. The scale ranged from 0 to 5 with higher values indi-
cating a lower level of energy and a greater level of fatigue. Internal
reliability was high for this composite scale (Cronbach’s α = 0.90).

Successful class completion

Attendance was tracked via attendance logs for each session, and
the attendance data were used to calculate the attendance and
completion rates. “Completion” was defined as attending at least
18 out of the 24 total Fit & Strong! sessions per class offering.

RECRUITMENT FLOW FIGURE 1 | Recruitment flow diagram for Fit & Strong! participants.
The recruitment flow from initial program enrollment is presented
in Figure 1 as a consort type diagram. This figure begins with all
“participant enrollees” and concludes with eligible participants RESULTS
with linked baseline and post-test data who were treated as the OBJECTIVE 1: STUDY CHARACTERISTICS
analytic sample for outcome analyses. This flow documents rea- As shown in Table 1, the average age of eligible Fit & Strong!
sons for exclusion (e.g., those who took the class previously were participants (including all enrollees with baseline surveys) was
not part of the analytical survey) and those lost to follow-up at the 73.02 (SD = 9.16) years (49.3% were age 75 and older, and 35.0%
end of the program. were between the ages of 65 and 74). The majority of participants
A total of 234 participants were enrolled in the program. Among were female (80.2%) and were of non-Hispanic White ethnic-
this initial group, 181 (77.4%) individuals were potentially eligible ity (82.8%). Over 75% had more than a high school degree, and
for the outcomes study, 21 (9.0%) individuals did not meet study 62.2% were married. Among the four chronic conditions reported
criteria (e.g., to be considered active, participants needed to attend (diabetes, hypertension, heart disease, and respiratory problems),
either the first or second training session), and 32 (13.7%) individ- hypertension was most frequently reported among the partici-
uals were repeaters (previous Fit & Strong! participants). Among pants (47.3%). Of the 136 eligible participants who completed the
the 181 potentially eligible participants, however, only 136 (75.1%) baseline survey, 33.8% were from rural counties and 66.2% were
completed the baseline survey and were, therefore, eligible to be from an urban county.
part of the initial participant comparison analyses. Only 71 partic- Compared to eligible participants omitted from the impact
ipants (39.2%) of the 136 eligible participants completed both pre study because of lack of matched data (completed baseline
and post-test surveys and served as “impact study participants.” and post-tests) (n = 65), a significantly larger proportion of
impact study participants (n = 71) were female (87.9 vs. 72.3%,
DATA ANALYSIS p = 0.025). On average, SE at baseline for impact study participants
Characteristics of those who completed both baseline and post-test (p = 0.033) was significantly higher relative to eligible participants
surveys (matched surveys) were compared to the other partici- omitted from the impact study; whereas, average self-rated health
pants (non-matched surveys; those who only completed a baseline (p = 0.028) at baseline for participants included in the impact
survey) using χ2 tests for categorical variables and two-sample study was significantly higher than participants who were not
t -tests for continuous variables. Next, association between par- included in the impact study. There were no significant differences
ticipant characteristics and program completion status for the by other socio-demographic characteristics and baseline levels of
analytical sample was identified using logistic regression with physical activity and illness symptomatology.
odd ratios. The impact of Fit & Strong! was then evaluated by
comparing the outcome measures using various methods (paired- OBJECTIVE 2: CLASS COMPLETION
t -test for continuous scales, Wilcoxon Signed Rank Test for ordinal As shown in Table 1, the average number of classes attended
scales, and McNemar test for two-level categorical scales). for all eligible participants who completed baseline surveys was

www.frontiersin.org April 2015 | Volume 2 | Article 187 | 273

Ory et al. Translation of Fit & Strong! for older adults

Table 1 | Baseline characteristics of eligible participants by data availability (i.e., presence of both baseline and post-test surveys).

Baseline Categories Eligible participants Eligible participants Impact study p-value*

characteristics with baseline excluded from the participants
survey (n = 136)a impact study (n = 65)b (n = 71)c

Age <75 68 (50.7%) 29 (44.6%) 39 (56.5%) 0.168

≥75 66 (49.3%) 36 (55.4%) 30 (43.5%)
Mean (SD) 73.02 (±9.16) 74.25 (±9.82) 71.87 (±8.39) 0.134
Gender Male 26 (19.8%) 18 (27.7%) 8 (12.1%) 0.025
Female 105 (80.2%) 47 (72.3%) 58 (87.9%)
Race/ethnicity White (not Hispanic origin) 106 (82.8%) 52 (85.2%) 54 (80.6%) 0.486
Non-White 22 (17.2%) 9 (14.8%) 13 (19.4%)
Education ≤High school graduate 30 (23.3%) 17 (26.6%) 13 (20.0%) 0.378
>High school graduate 99 (76.7%) 47 (73.4%) 52 (80.0%)
Marital status Married 84 (62.2%) 38 (58.5%) 46 (65.7%) 0.385
Not married 51 (37.8%) 27 (41.5%) 24 (34.3%)
Site Rural 46 (33.8%) 19 (29.2%) 27 (38.0%) 0.279
Urban 90 (66.2%) 46 (70.8%) 44 (62.0%)
Reported number of chronic Mean (SD)e 0.83 (±0.82) 0.94 (±0.90) 0.73 (±0.74) 0.168
conditionsd
Median 1 1 1
Mode 0 1 0
Chronic conditions Diabetes 18 (13.5%) 11 (17.5%) 7 (10.0%) 0.209
Hypertension 61 (46.6%) 28 (44.4%) 33 (48.5%) 0.640
Heart disease 23 (17.3%) 13 (20.6%) 10 (14.3%) 0.334
Respiratory problems 11 (8.3%) 7 (11.1%) 4 (5.8%) 0.270
Self-efficacy Mean (SD)e 6.95 (±2.44) 6.48 (±2.47) 7.39 (±2.34) 0.033
RAPA (aerobic capacity) Mean (SD)e 3.86 (±1.11) 3.86 (±1.12) 3.85 (±1.11) 0.991
RAPA (strength/flexibility) None 73 (60.3%) 35 (60.3%) 38 (60.3%) 0.395
Either 33 (27.3%) 18 (31.0%) 15 (23.8%)
Both 15 (12.4%) 5 (8.6%) 10 (15.9%)
Strength 22 (18.0%) 8 (13.6%) 14 (22.2%) 0.214
Flexibility 44 (35.8%) 22 (37.3%) 22 (34.4%) 0.736
Self-rated health Mean (SD)e 3.30 (±0.82) 3.14 (±0.85) 3.45 (±0.78) 0.028
Joint pain Mean (SD)e 4.06 (±3.60) 4.21 (±3.75) 3.90 (±3.48) 0.636
Joint stiffness Mean (SD)e 2.49 (±1.78) 2.59 (±1.86) 2.40 (±1.71) 0.555
Level of energy Mean (SD)e 2.13 (±0.99) 2.18 (±0.98) 2.07 (±1.00) 0.527
Fit & Strong! attendance
Completion status Completed 76 (55.9%) 21 (32.3%) 55 (77.5%) 0.000
Not completed 60 (44.1%) 44 (67.7%) 16 (22.5%)
Total number of classes Mean (SD)e 15.96 (±7.16) 11.40 (±7.58) 20.14 (±3.04) 0.000
attended (MAX = 24)

a
n = 136 With a slight variation for each variable.
b
n = 65 With a slight variation for each variable.
c
n = 71 With a slight variation for each variable.
d
Chronic conditions: heart diseases, diabetes, hypertension, and respiratory problems.
e
SD, standard deviation.
*p-value for statistical analyses (i.e., χ2 or t-tests) for comparing the enrollees with only baseline surveys and the enrollees with both baseline and follow-up surveys.
Eligibility criteria for the baseline analysis, or the initial participant comparison analyses, included: (1) attendance of the first or second class sessions, (2) first-time
participants (no previous participation in a Fit & Strong! class); and (3) completion of a baseline survey. Eligibility criteria for the impact study analysis included: (1)
fulfillment of the aforementioned baseline analysis criteria, and (2) completion of a post-test survey. Participants who fulfilled the baseline analysis criteria and did not
complete a post-test survey were excluded from the impact study analysis.

approximately 16 (SD = 7.16) out of 24. The program completion completion rate and the number of classes attended between the
rate was 55.9% (i.e., attending 18 or more of the 24 ses- two groups of eligible participants (impact analysis participants vs.
sions). Significant differences were observed when comparing the non-impact analysis participants). On average, participants in the

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 187 | 274
Ory et al. Translation of Fit & Strong! for older adults

impact analysis group attended more sessions (average number for exercise (p = 0.020, d = 0.30) and aerobic capacity (p = 0.022,
of sessions attended = 20.14 vs. 11.40, p < 0.001) and had higher d = 0.34) from baseline to post-test. In terms of the magnitude of
completion rates (77.5 vs. 32.3%, p < 0.001). improvement at the individual level, there was an 8.1% improve-
As further seen in Table 2, there were a few variables that ment in SE for exercise and an 11.9% improvement in aerobic
differentiated the participants who did and did not complete capacity. Furthermore, 54.8% of the sample reported an improve-
the program. Participants from the rural sites were less likely to ment in confidence to exercise and a 29.8% improvement in
complete 18 or more classes than the participants from urban aerobic capacity. Additionally, there was a shift in the propor-
sites (OR = 0.41, p = 0.015). Those without any chronic condi- tion of participants who met the Surgeon General’s recommended
tions were also more likely to complete the program (OR = 2.34, physical activity guidelines (31). At baseline, 38.7% of the partic-
p = 0.022); for every increase in number of chronic conditions, the ipants were determined to be “active” according to the Surgeon
odds of completing the class drops by 46.4%. There were no sig- General’s guidelines; whereas, by the end of the program, 59.4%
nificant differences by other socio-demographic characteristics or of participants were determined to be “active.”
baseline levels of physical activity, general health status, or illness In terms of secondary outcomes, there were significant changes
symptomatology. observed for joint stiffness, level of energy, and amount and
intensity of physical activities related to strength and flexibility
OBJECTIVE 3: IMPACT OF FIT & STRONG! (p < 0.05). The effect sizes for all secondary outcomes ranged
According to the results illustrated in Table 3, in terms of primary from 0.05 to 0.59. The strongest effect sizes were observed for
outcomes, there were significant improvements in participants’ SE strength and flexibility scales (d = 0.59), then for the level of
energy (d = 0.33), and then joint stiffness (d = 0.31). At the indi-
vidual participant level, there was a 19.2% improvement in the
Table 2 | Comparison of participant baseline characteristics by their degree of joint stiffness and an 11.7% improvement in the level of
program completion status (i.e., attended at least 18 out of 24 energy. Furthermore, 17.5% of the participants reported improve-
sessions). ments in joint stiffness and 27.0% reported improvements in the
level of energy. Over one-third of participants reported improve-
na Program completion ments in the degree of physical activities related to strength. 35.6%
reported improvements in the degree of physical activities related
Odds ratio p-valueb
to flexibility, and 48.3% reported improvements in the degree of
Age 134 (98.5%) 1.008 0.685 physical activities related to both strength and flexibility.
Sex (female) 131 (96.3%) 0.433 0.063
Race/ethnicity (non-Hispanic 128 (94.1%) 1.562 0.357
DISCUSSION
As with many evidence-based programs, the randomized trials
White)
often use a higher level of interventionists to provide a best case
Education (>high school 129 (94.9%) 1.882 0.147
scenario (32, 33). Alternatively, translated models frequently use
graduate)
lay-leaders to expand dissemination efforts while minimizing costs
Marital status (married) 135 (99.3%) 1.089 0.812
(34, 35). The same is true of the original Fit & Strong! program,
Site (urban) 136 (100.0%) 0.407 0.015
which originally used PTs or CEIs as class instructors as a means
Reported number of chronic 131 (96.3%)
of minimizing harm to participants (7, 17).
conditions
The current study examined a lay-leader model of the Fit &
Indicative/binaryc (≥1) 2.337 0.022
Strong! program adapted to overcome common challenges to
Countd 0.536 0.006
program implementation such as instructor availability (15, 16).
Baseline self-efficacy 129 (94.9%) 1.113 0.145
Consistent with other findings showing successful applicability of
Baseline physical activity 118 (86.8%) 0.724 0.400
lay-leaders with a variety of physical activity programs in diverse
(aerobic capacity) (active)
settings (34, 36–38), we saw many positive outcomes and recom-
Baseline physical activity 121 (89.0%) 0.111
mend the implementation of a lay-led model. Our program fidelity
(strength/flexibility)
observations (data not reported here) indicated that group facili-
None 73 (60.3%) 0.521 0.301
tators with more experience in evidence-based programing tended
Either 33 (27.3%) 0.268 0.053
to adhere more closely to program guidelines than instructors with
Both (ref) 15 (12.4%)
no or limited prior experience adhering to scripted programs.
Baseline self-rated health 134 (98.5%) 1.309 0.213
Our study resonates with previous literature that shows the
Baseline joint pain 124 (91.2%) 0.942 0.234
value of lay-led programs for seniors, especially those with arthri-
Baseline joint stiffness 128 (94.1%) 0.933 0.488
tis, which was the original target group for Fit & Strong! classes.
Baseline level of energy 129 (94.9%) 0.882 0.487
Cohen et al. (39) compared a lay-led arthritis self-management
a
Number of cases included in the analysis (maximum possible n = 136). course and professional-led arthritis self-management course and
b
p-value from bivariate logistic regression model. identified no significant differences for participant outcomes
c
Reported number of chronic conditions (0 = no chronic conditions; 1 = at least by leader type (although, it should be noted that the courses
one chronic condition). compared differed slightly in course content). Similarly, Lorig
d
Reported number of chronic conditions (count variable ranging from 0 to 4). et al. (40) compared a lay-led and a professional-led arthritis

www.frontiersin.org April 2015 | Volume 2 | Article 187 | 275

Ory et al. Translation of Fit & Strong! for older adults

Table 3 | Baseline and post-test comparisons for assessing the impact of Fit & Strong! program.

Outcome na Mean (SD)b p-value Effect Improvement (%) Percentage of

sizec improved
participants (%)

Pretest Post-test

Primary outcomes
Self-efficacy for adherence to exercise 62 (87.3%) 7.40 (±2.28) 8.00 (±1.99) 0.020 0.30 8.1 54.8
Aerobic physical activity leveld 57 (80.3%) 3.85 (±1.11) 4.31 (±1.02) 0.022 0.34 11.9 29.8
Secondary outcomes
Self-rated healthd 3.45 (±0.78) 3.49 (±0.73) 0.491 0.08 1.2 11.6
Weight 62 (87.3%) 172.02 (±33.87) 168.66 (±36.64) 0.056 0.25 2.0 14.5
Joint pain 62 (87.3%) 3.89 (±3.50) 3.73 (±3.62) 0.714 0.05 4.1 35.5
Joint stiffness 63 (88.7%) 2.38 (±1.73) 1.94 (±1.48) 0.017 0.31 18.7 17.5
Level of energy 63 (88.7%) 2.06 (±0.98) 1.82 (±0.94) 0.010 0.33 11.9 27.0
Strengthe 59 (83.1%) 14 (22.2%)f 34 (52.3%)f 0.002 33.9
Flexibilitye 59 (83.1%) 22 (34.4%)f 42 (65.6%)f 0.001 35.6
Strength and flexibilityd 58 (81.7%) 0.56 (±0.76) 1.19 (±0.85) 0.000 0.59 48.3

a
Number of cases included in the analysis (maximum possible n = 71).
b
Mean and standard deviation (SD), unless otherwise indicated.
c
Cohen’s d [effect sizes of d ≈ 0.2 (small), d ≈ 0.5 (medium), and d ≥ 0.8 (large)].
d
Wilcoxon’s paired sign rank test was used.
e
McNemar test was used (No t-statistic).
f
Frequency and valid percentages.

self-management course, and both courses showed a significant flexibility and strength training. These findings are consistent with
increase in participant knowledge. Participants in the professional- those of Hughes and colleagues (7, 11), who reported Fit & Strong!
led courses showed a greater gain in knowledge than those in participant improvement for exercise efficacy, exercise adherence,
the lay-led courses; however, participants from the lay-led model joint stiffness, physical functioning, and exercise capacity. Hughes
showed greater improvement in relaxation practice and higher reported a 15.6% reduction in participants’ stiffness scores at
attendance rates (40). These studies utilizing lay-leaders for phys- post-test (7), which is consistent with the 19.2% reduction in stiff-
ical activity programs confirm the feasibility of using a lay-leader ness scores for participants in the current study. Other measure
model for increasing the availability and adoption of the Fit & comparisons could not be made because the current study used
Strong! program. different outcome measures than those used by Hughes.
The completion rate for those in the impact study (77.5%) was
comparable with that found in other research studies using differ- STUDY LIMITATIONS
ent time-bound evidence-based programs (41). It is not surprising There are several limitations to this pilot study that should be noted
that those in rural areas vs. those in more urban areas were less but are acceptable considering this was the initial investigation of
likely to complete the program given the previously documented a translated intervention. A major limitation for the generalizabil-
challenges to bringing health services or health promotion pro- ity of study findings is the small sample size for the final impact
grams to rural areas (42, 43). Additionally, the fact that those with study. Additionally, compared to the original Fit & Strong! stud-
one or more comorbidities were less likely to complete classes can ies (7, 11), there was a relatively high attrition rate (47.9%) from
be attributed to the challenges reported by those facing multi- pre to post-test, as commonly found in more community-oriented
ple chronic conditions (44); although, more research is needed exercise programs (45). As documented by local program admin-
to understand how different conditions might affect completion istrators, this high attrition rate was attributed to “loose program
rates. These findings suggest that additional efforts are needed to adherence/commitment” as some participants preferred “drop-
attract and retain participants from rural areas and those with ping into classes” (i.e., attend at their leisure) as opposed to fully
multiple chronic conditions. committing to the 8-week program. Others, especially in the rural
The current study also examined the impact of lay-led Fit & areas, had limited transportation and, therefore, had difficulties
Strong! classes on various outcome measures. Participants showed with program attendance. In the current study, we assessed out-
a significant improvement in their aerobic capacity, joint stiff- comes only for those with complete data, and thus were not able to
ness, level of energy/fatigue, and SE for exercise. Participants assess whether those who lacked complete data might have biased
also reported greater participation in exercise types (flexibil- study results. However, when we compared the baseline charac-
ity, strength, or both) such that more individuals met the Sur- teristics of the eligible participants included and excluded from
geon General’s recommendations of including exercises targeting the impact analysis, we only found a few significant differences

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 187 | 276
Ory et al. Translation of Fit & Strong! for older adults

between those two groups, indicating the potential similarity of are seen as critical in allowing the Fit & Strong! program to be
the two groups. disseminated more broadly as a lay-led model.
This study only included a post-test that was administered dur- There is now a growing literature on factors affecting recruit-
ing the last week of the program. No follow-up assessments were ment and strategies for boosting program retention (50). Given the
administered after the last session. The lack of follow-up mea- reported attrition levels in attendance from entry into the study till
surements after the program limited our ability to observe any class completion, efforts to retain participants from start to finish
potential long-term effects of utilizing the lay-led Fit & Strong! should focus on committing and motivating participants to fully
program. However, this study enabled the primary question to be complete the program. This is often accomplished during enroll-
addressed regarding applicability to a broader population of older ment of participants or during the first session, or orientation, of
adults and also the potential value of a lay-led approach for this the program (51). Furthermore, instructors should emphasize to
program in other communities. participants early on the benefits gained from full participation
Other study limitations can be attributed to program design and should strive to interact and engage participants during ses-
and evaluation issues. The participants were self-selected into the sions and outside of class where necessary (e.g., follow-up phone
program from different delivery sites, creating a potential self- calls if a participant misses a class).
selection, or delivery site bias. Also, participants included in the This pilot study also highlights the need for additional research.
impact analysis had higher SE and self-rated health than those Future research should compare lay-led and professional-led Fit
who were not included in the impact analysis, potentially influ- & Strong! classes in terms of the magnitude of program impact
encing the program impact analysis. This is not surprising given and program fidelity. Also, lay-led Fit & Strong! classes should be
the literature to date suggesting that older adults with better evaluated/assessed in other settings to draw a more generalizable
health are more likely to attend and complete a health promotion conclusion about the utility and effectiveness of varying levels of
program (46–48). Such relationships pose a potential interven- instructor expertise and training components.
tion bias, which must be considered when interpreting study
results. CONCLUSION
Finally, participants from this study differed somewhat from Overall, utilizing a lay-led model was successfully adapted from
participants for which the program was originally intended. Older the original Fit & Strong! program that relied on professional
adults in various physical capacities, including those who were and experienced leaders (PTs and CEIs). The lay-led Fit & Strong!
more sedentary or suffered from “achy joints” were recruited model produced outcomes that are consistent with the previous
for this iteration. In contrast, participants in the original ran- findings from the original intervention. Specifically, the program
domized control trials were originally selected based upon the showed improvement in participants’ SE for exercise, aerobic
presence of lower-extremity joint stiffness and pain associated capacity, engagement in strength, and flexibility exercises, while
with osteoarthritis and related symptomatology. Consequently, it increasing energy levels and decreasing joint stiffness. The mag-
is not possible to do a direct comparison with the earlier studies by nitude of program attrition in community-based exercise pro-
Hughes and colleagues (7, 11, 17) since the extent to which partici- grams can be large; hence, creative strategies are needed to boost
pants in the current study had arthritis and specifically osteoarthri- participant retention throughout the entire intervention period.
tis is unknown. Thus, outcomes for arthritis-related symptoma-
tology may have been attenuated in this more generalized study ACKNOWLEDGMENTS
population. We acknowledge support of Baylor Scott and White HealthCare,
St. Joseph’s HealthCare System, and Brazos Valley Area Aging on
PROGRAM IMPLICATIONS AND FUTURE RESEARCH DIRECTIONS Aging and are appreciative of the delivery sites and older adults
An important implication of the study is that the Fit & Strong! pro- who made this translational research possible. We extend special
gram may benefit the general older adult population and not just thanks to Doris Howell, Rachel Foster, and Cindy Quinn who
those with lower-extremity osteoarthritis. This may be because assisted in training.
a substantial proportion of older adults experience some type
of joint pain and/or stiffness, not just specifically in the lower- REFERENCES
1. Ory MG, Resnick B, Smith ML. Changing behavior throughout the life-course:
extremity (49) Furthermore, as indicated from program facilita-
translating the success of aging research. Trans Behav Med (2012) 2(2):159–62.
tors, Fit & Strong! can benefit sedentary older adults who want doi:10.1007/s13142-012-0129-4
a beginning level and less intimidating means to start a physical 2. Ory MG, Ahn S, Smith ML, Jiang L, Lorig K, Whitelaw N. National study of
activity regiment. Thus, this program has universal benefits. chronic disease self-management: age comparison of outcome findings. Health
Although it is impossible to draw a definitive conclusion, these Educ Behav (2014) 41(1 Suppl):34s–42s. doi:10.1177/1090198114543008
3. Jacobs JA, Jones E, Gabella BA, Spring B, Brownson RC. Tools for implementing
findings strongly suggest that the Fit & Strong! program can be an evidence-based approach in public health practice. Prev Chronic Dis (2012)
instructed by lay-leaders with standardized training and contin- 9:110324. doi:10.5888/pcd9.110324
ued support from the developers and/or on-site Master Trainers. 4. Task Force on Community Preventive Services. Preventive Services: What Works
This is important because the training provides lay-leaders (i.e., to Promote Health?. New York, NY: Oxford University Press (2005). 36 p.
those without formal exercise or professional training) with guided 5. Brownson RC, Fielding JE, Maylahn CM. Evidence-based public health: a fun-
damental concept for public health practice. Annu Rev Public Health (2009)
instruction and ongoing feedback related to program admin- 30:175–201. doi:10.1146/annurev.publhealth.031308.100134
istration as well as proper techniques for exercise progression. 6. de Jong ORW, Hopman-Rock M, Tak ECMP, Klazinga NS. An implementa-
These modifications are essential for conducting the program and tion study of two evidence-based exercise and health education programmes

www.frontiersin.org April 2015 | Volume 2 | Article 187 | 277

Ory et al. Translation of Fit & Strong! for older adults

for older adults with osteoarthritis of the knee and hip. Health Educ Res (2004) 27. Stanford Patient Education Research Center, Stanford University School of Med-
19(3):316–25. doi:10.1093/her/cyg028 icine. Chronic Disease Self-Management Program Questionnaire Code Book. Palo
7. Hughes SL, Seymour RB, Campbell R, Pollak N, Huber G, Sharma L. Impact of Alto, CA: Stanford University (2007). Available from: http://patienteducation.
the Fit & Strong intervention on older adults with osteoarthritis. Gerontologist stanford.edu/research/cdCodeBook.pdf
(2004) 44(2):217–28. doi:10.1093/geront/44.2.217 28. Ackerman I. Western Ontario and McMaster Universities Osteoarthritis index
8. King AC, Baumann K, O’Sullivan P, Wilcox S, Castro C. Effects of moderate- (WOMAC). Aust J Physiother (2009) 55(3):213. doi:10.1016/S0004-9514(09)
intensity exercise on physiological, behavioral, and emotional responses to fam- 70088-1
ily caregiving: a randomized trial. J Gerontol A Biol Sci (2002) 57(1):M26–36. 29. Olsen IC, Kvien TK, Uhlig T. Consequences of handling missing data for treat-
doi:10.1093/gerona/57.1.M26 ment response in osteoarthritis: a simulation study. Osteoarthritis Cartilage.
9. Thibaud M, Bloch F, Tournoux-Facon C, Breque C, Rigaud AS, Dugue B, et al. (2012) 20(8):822–8. doi:10.1016/j.joca.2012.03.005
Impact of physical activity and sedentary behavior on fall risks in older people: 30. Stanford Patient Education Research Center. Energy/Fatigue. Palo Alto,
a systematic review and meta-analysis of observational studies. Eur Rev Aging CA (2014). Available from: http://patienteducation.stanford.edu/research/
Phys Activ (2012) 9(1):5–15. doi:10.1007/s11556-011-0081-1 energyfatigue.pdf
10. Landi F, Abbatecola AM, Provinciali M, Corsonello A, Bustacchini S, Manigrasso 31. US Department of Health and Human Services. 2008 Physical Activity Guide-
L, et al. Moving against frailty: does physical activity matter? Biogerontology lines for Americans. ODPHP Publication No. U0036 (2008). 76 p. Available from:
(2010) 11(5):537–45. doi:10.1007/s10522-010-9296-1 http://www.health.gov/paguidelines/pdf/paguide.pdf
11. Hughes SL, Seymour RB, Campbell RT, Huber G, Pollak N, Sharma L, et al. Long- 32. Tennstedt S, Howland J, Lachman M, Peterson E, Kasten L, Jette A. A random-
term impact of Fit & Strong! on older adults with osteoarthritis. Gerontologist ized, controlled trial of a group intervention to reduce fear of falling and asso-
(2006) 46(6):807–14. doi:10.1093/geront/46.6.801 ciated activity restriction in older adults. J Gerontol B Psychol Sci Soc Sci (1998)
12. Hughes SL, Seymour R, Cheryl D, Desai P. Translation of Fit and Strong!: Focus 53B(6):384–92. doi:10.1093/geronb/53B.6.P384
Group Findings from Providers and Instructors in Two States. 61st Annual Scien- 33. Clemson L, Cumming RG, Kendig H, Swann M, Heard R, Taylor K. The effec-
tific Meeting; Maryland (US). UK: The Gerontologist (2008). 509 p. tiveness of a community-based program for reducing the incidence of falls
13. Center for Healthy Aging, National Council on Aging. Title III-D Highest Tier in the elderly: a randomized trial. J Am Geriatr Soc (2004) 52(9):1487–94.
Evidence-Based Disease Prevention & Health Promotion Programs Chart. (2014). doi:10.1111/j.1532-5415.2004.52411.x
32 p. Available from: http://www.ncoa.org/improve-health/center-for-healthy- 34. Healy TC, Peng C, Haynes MS, McMahon EM, Botler JL, Gross L. The feasibility
aging/content-library/Title-IIID-Highest-Tier-Evidence-FINAL.pdf and effectiveness of translating a matter of balance into a volunteer lay leader
14. Centers for Disease Control and Prevention. Arthritis: Physical Activity Program. model. J Appl Gerontol (2008) 27(1):34–51. doi:10.1177/0733464807308620
(2011). Available from: http://www.cdc.gov/arthritis/interventions/physical_ 35. Clemson L, Swann M, Mahoney J. Stepping On: Building Confidence and Reduc-
activity.htm ing Falls in Older Adults. 3rd ed. Cedar Falls, IA: Freiburg Press Inc (2011).
15. DerAnanian CA, Desai P, Smith-Ray R, Seymour RB, Hughes SL. Perceived ver- 36. Etkin CD, Prohaska TR, Harris BA, Latham N, Jette A. Feasibility of imple-
sus actual factors associated with adoption and maintenance of an evidence- menting the strong for life program in community setting. Gerontologist (2006)
based physical activity program. Trans Behav Med (2012) 2(2):209–17. doi:10. 46(2):284–92. doi:10.1093/geront/46.2.284
1007/s13142-012-0131-x 37. Krukowki RA, Lensing S, Love S, Prewitt TE, Adams B, Cornell CE, et al. Train-
16. Brownson RC, Ballew P, Dieffenderfer B, Haire-Joshu D, Heath GW, Kreuter ing of lay health educators to implement an evidence-based behavioral weight
MW, et al. Evidence-based interventions to promote physical activity: what con- loss intervention in rural senior centers. Gerontologist (2013) 53(1):162–71.
tributes to dissemination by state health departments. Am J Prev Med (2007) doi:10.1093/geront/gns094
33(1S):S66–78. doi:10.1016/j.amepre.2007.03.011 38. Werner D, Teufel J, Brown SL. Evaluation of a peer-led, low-intensity physi-
17. Seymour RB, Hughes SL, Campbell RT, Huber GM, Desai P. Comparison of two cal activity program for older adults. Am J Health Educ (2014) 45(3):133–41.
methods of conducting the Fit and Strong! program. Arthritis Rheum (2009) doi:10.1080/19325037.2014.893851
61(7):879–84. doi:10.1002/art.24517 39. Cohen JL, Sauter SH, DeVellis RF, DeVellis BM. Evaluation of arthritis self-
18. Hughes S, Seymour R, Huber G, Desai P, Der Ananian C, Kunkel C. Fit & Strong! management courses led by laypersons and by professionals. Arthritis Rheum
Instructor Manual 4 th Version [Booklet]. Chicago, IL: Center for Research on (1986) 29(3):388–93. doi:10.1002/art.1780290312
Health and Aging, University of Illinois at Chicago (2009). 40. Lorig K, Feigenbaum P, Regan C, Ung E, Chastain R, Holman HR. A compar-
19. Hughes S, Huber G, Desai P, Smith-Ray R, Heide D. Fit & Strong! Instructor Train- ison of lay-taught and professional-taught arthritis self-management courses.
ing [Booklet]. Chicago, IL: Center for Research on Health and Aging, University J Rheumatol (1986) 13(4):763–7.
of Illinois at Chicago (2012). 41. Woodcock C, Korda H, Erdem E, Pedersen S, Kloc M, Tollefson E. Admin-
20. Center for Healthy Aging, National Council on Aging. Module 4: Assur- istration on aging: Chronic Disease Self-Management Program (CDSMP)
ing Program Quality: Fidelity. (2014). Available from: http://www.ncoa. Process Evaluation. Report No.: HHSP233201100492G (2013). 142 p. Avail-
org/improve-health/center-for-healthy-aging/online-training-modules/ able from: http://www.aoa.gov/aoaroot/Program_Results/docs/CDSMPProcess
module-4-assuring-program.html EvaluationReportFINAL062713.pdf
21. Institute for Health Research and Policy, School of Public Health, Univer- 42. DiBartolo MC, McCrone S. Recruitment of rural community-dwelling older
sity of Illinois at Chicago. Resource Requirements. (2014). Available from: adults: barriers, challenges, and strategies. Aging Ment Health (2003) 7(2):75–82.
http://www.fitandstrong.org/providers/resource_requirements.html doi:10.1080/1360786031000072295
22. Hughes S, Smith-Ray RL, Shah A, Huber G. Translating Fit and Strong!: Lessons 43. Morgan LL, Fahs PS, Klesh J. Barriers to research participation identified by
learned and next steps. Front Public Health (2015) 2:131. doi:10.3389/fpubh. rural people. J Agric Saf Health (2005) 11(4):407–14. doi:10.13031/2013.19719
2014.00131 44. US Department of Health and Human Services. US Department of Health and
23. Stanford Patient Education Research Center. Chronic Disease Self-Efficacy Human Services Inventory of Programs, Activities, and Initiatives Focused on
Scales. (2014). Available from: http://patienteducation.stanford.edu/research/ Improving the Health of Individuals with Multiple Chronic Conditions (MCC).
secd32.html Washington, DC: U.S. Department of Health and Human Services Interagency
24. Topolski TD, LoGerfo J, Patrick DL, Williams B, Walwick J, Patrick MB. The Workgroup on Multiple Chronic Conditions (2011). 103 p. Available from:
rapid assessment of physical activity (RAPA) among older adults. Prev Chronic http://www.hhs.gov/ash/initiatives/mcc/mcc-inventory-20111018.pdf
Dis (2006) 3(4):A118. 45. Dishman R, Sallis J. Determinants and interventions for physical activity and
25. University of Washington Health Promotion Research Center. How Physically exercise. In: Bouchard C, Shepard R, Stephens T editors. Physical Activity, Fitness
Active are You? – An Assessment of Level and Intensity of Physical Activity. (2014). and Health: International Preceding and Consensus Statement. Champaign, IL:
Available from: http://depts.washington.edu/hprc/docs/rapa_03_06.pdf Human Kinetics (1994). p. 214–38.
26. Centers for Disease Control and Prevention. CDC HRQOL-14 “Healthy Days 46. Flegal KE, Kishiyama S, Zajdel D, Haas M, Oken BS. Adherence to yoga and
Measure”. (2011). Available from: http://www.cdc.gov/hrqol/hrqol14_measure. exercise interventions in a 6-month clinical trial. BMC Complement Altern Med
htm#1 (2007) 7:37. doi:10.1186/1472-6882-7-37

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 187 | 278
Ory et al. Translation of Fit & Strong! for older adults

47. Dattalo M, Giovannetti ER, Scharfstein D, Boult C, Wegener S, Wolff JL, et al. private organizations/agencies; however, the views, findings, and conclusions in these
Who participates in Chronic Disease Self-management (CDSM) programs? Dif- articles are those of the authors and do not necessarily represent the official position
ferences between participants and nonparticipants in a population of mul- of these organizations/agencies. All papers published in the Research Topic received
timorbid older adults. Med Care (2012) 50(12):1071–5. doi:10.1097/MLR. peer review from members of the Frontiers in Public Health (Public Health Education
0b013e318268abe7 and Promotion section) panel of Review Editors. Because this Research Topic repre-
48. Biedenweg K, Meischke H, Hammerback K, Williams B, Poe P, Phelan EA. sents work closely associated with a nationwide evidence-based movement in the US,
Understanding older adults’ motivators and barriers to participating in orga- many of the authors and/or Review Editors may have worked together previously in
nized programs supporting exercise behaviors. J Prim Prev (2014) 35(1):1–11. some fashion. Review Editors were purposively selected based on their expertise with
doi:10.1007/s10935-013-0331-2 evaluation and/or evidence-based programming for older adults. Review Editors were
49. Centers for Disease Prevention and Control. Prevalence of doctor-diagnosed independent of named authors on any given article published in this volume.
arthritis and arthritis-attributable activity limitation – United States, 2007 –
2009. MMWR Morb Mortal Wkly Rep (2010) 59(39):1261–5. Available from:
http://www.cdc.gov/mmwr/pdf/wk/mm5939.pdf Received: 15 July 2014; paper pending published: 19 August 2014; accepted: 26
50. Ory MG, Lipman PD, Karlen PL, Gerety MB, Stevens VJ, Singh MAF, et al. September 2014; published online: 27 April 2015.
Recruitment of older participants in frailty/injury prevention studies. Prev Sci Citation: Ory MG, Lee S, Zollinger A, Bhurtyal K, Jiang L and Smith ML (2015)
(2002) 3(1):1–21. doi:10.1023/A:1014610325059 Translation of Fit & Strong! for middle-aged and older adults: examining implementa-
51. Jiang L, Smith ML, Chen S, Ahn S, Kulinski KP, Lorig K, et al. The role of Session tion and effectiveness of a lay-led model in Central Texas. Front. Public Health 2:187.
Zero in successful completion of Chronic Disease Self-Management Program doi: 10.3389/fpubh.2014.00187
workshops. Front Public Health (2015) 2:205. doi:10.3389/fpubh.2014.00205 This article was submitted to Public Health Education and Promotion, a section of the
journal Frontiers in Public Health.
Conflict of Interest Statement: The authors declare that the research was conducted Copyright © 2015 Ory, Lee, Zollinger, Bhurtyal, Jiang and Smith. This is an open-
in the absence of any commercial or financial relationships that could be construed access article distributed under the terms of the Creative Commons Attribution License
as a potential conflict of interest. (CC BY). The use, distribution or reproduction in other forums is permitted, provided
the original author(s) or licensor are credited and that the original publication in this
This paper is included in the Research Topic, “Evidence-Based Programming for Older journal is cited, in accordance with accepted academic practice. No use, distribution or
Adults.” This Research Topic received partial funding from multiple government and reproduction is permitted which does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 187 | 279

 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00171

Fit & Strong! promotes physical activity and well-being in

older cancer survivors
Jana Reynolds 1,2 *, Lorie Thibodeaux 1,3 , Luohua Jiang 4 , Kevin Francis 1 and Angie Hochhalter 1
1
Baylor Scott & White Health, Temple, TX, USA
2
Baylor Charles A. Sammons Cancer Center, Baylor University Medical Center at Dallas, Dallas, TX, USA
3
Seton Medical Center, Austin, TX, USA
4
School of Rural Public Health, Texas A&M Health Science Center, College Station, TX, USA

Edited by: Introduction: Physical activity reduces fatigue and depression while improving quality
Matthew Lee Smith, University of
of life in cancer survivors. Exercise is generally considered safe and is recommended
Georgia, USA
to survivors of all ages. Despite the high prevalence of cancer in the elderly, few stud-
Reviewed by:
Emily Joy Nicklett, University of ies address physical activity interventions targeting this older population. Fit & Strong! is
Michigan, USA an evidence-based physical activity program shown to improve level of physical activity,
Samuel D. Towne, University of South exercise-self-efficacy, and mood in older adults with osteoarthritis. This study tests the
Carolina, USA
feasibility and short-term impact of the Fit & Strong! exercise program adapted for older
*Correspondence:
cancer survivors.
Jana Reynolds, Departments of
Internal Medicine and Oncology and Methods: Participants were cancer survivors at least 50 years of age who were not on
Hematology, Baylor Charles A.
Sammons Cancer Center, Baylor
active treatment with intravenous chemotherapy or radiation. They participated in the
University Medical Center at Dallas, 8-week Fit & Strong! program, which included three 90-min sessions per week; 60 min of
3410 Worth Street, Dallas, TX 75246, group physical activity and 30 min of education. Education on osteoarthritis was removed
USA from the Fit & Strong! program and replaced with relevant topics on cancer survivorship
e-mail: jana.reynolds@
baylorhealth.edu
issues. Feasibility was measured by the ability to recruit and retain older cancer survivors.
Pre and post-intervention surveys evaluated the effect of the intervention on physical
activity and quality of life.
Results: The study enrolled 72 cancer survivors to participate in an 8-week exercise
program. The mean age of participants was 70. Over two-thirds (68%) of participants com-
pleted the program and with a mean attendance rate of 75% (18 of 24 sessions). No
safety issues occurred. Improvements from baseline to post-intervention were observed
for self-reported minutes of physical activity per week, self-efficacy for aerobic exercise,
and symptoms related to depression and anxiety.
Conclusion: This study was successful in recruiting and retaining a population of older
cancer survivors to participate in a group exercise program. Significant improvement in
level of physical activity and mood suggests this evidence-based physical activity interven-
tion can be adapted to promote health benefits in cancer survivors. Additional studies are
necessary to confirm efficacy and assess long-term benefits.
Keywords: evidenced based intervention, older cancer survivors, physical activity, exercise, cancer survivorship

INTRODUCTION in cancer survivors reduces fatigue and depression while improv-

With early cancer detection and greater availability of curative ing quality of life (8–12), and at the same time has been shown to
therapy, 64% of cancer survivors in the United States are living be safe in this population (13). It is recommended that cancer sur-
five or more years after cancer diagnosis (1, 2). As the number of vivors of all ages participate in a combination of strength training
long-term survivors continues to increase, so has the recognition and moderate aerobic exercise (such as brisk walking) for at least
of negative late and long-term health effects of cancer and cancer 150 min per week, or to the best of their physical ability. These
treatment (2, 3). It is well documented that once cancer survivors guidelines are similar to those recommended for the general pop-
complete their initial treatment, many face persistent fatigue, ulation (13, 14). Additionally, results from observational studies
depression, fear of recurrence, and long-term physical effects of suggest that participation in physical activity before and/or after
treatment (3–7). Thus, finding ways to combat these long-term diagnoses of certain cancers may serve as a potential preventive
health effects in cancer survivors is of paramount importance. measure against recurrence and mortality (4, 15–19).
One way to address these long-term effects of cancer and cancer In the United States, nearly 90% of cancer survivors are aged 50
treatment is through increased physical activity. Physical activity and older (88%), with persons aged 70 and older accounting for

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 171 | 280
Reynolds et al. Exercise for older cancer survivors

almost half (46%) of all cancer survivors (2). Despite the frequency by phone survey to determine whether they met eligibility criteria.
of cancer in the elderly, the majority of studies targeting physi- Those who met enrollment criteria were invited to participate in
cal activity in cancer survivors either exclude or do not achieve the study by enrolling in one of five courses. Participants gave their
high levels of participation from older cancer survivors (20, 21). consent and official enrollment occurred during the first session
Given the prevalence of cancer in the older population and lack of of each course.
evidence-based physical activity programs engaging this popula-
tion, this study aims to test a group exercise intervention targeting COURSE SETTING
older cancer survivors. The intervention was offered as an 8-week exercise course with
This study chose Fit & Strong!, an evidence-based physical three 90-min sessions per week, for a total of 24 sessions. Approx-
activity program for older adults with osteoarthritis, to adapt to imately 60 min of each session was dedicated to physical activity,
a population of older cancer survivors. The program is a combi- and 30 min was dedicated to education. Each course was con-
nation of group exercise and education/support. In randomized ducted in a group setting with a goal class size of 8–20 participants.
controlled trials, Fit & Strong! significantly increased participa- A total of five courses were offered between January 2013 and
tion in physical activity while decreasing levels of anxiety and August 2013. The first course was conducted in a large conference
depression and reducing symptoms of osteoarthritis in adults room in a medical office building. Due a higher than anticipated
older than 60 (22–24). We chose to use the Fit & Strong! pro- number of participants, the subsequent four courses were offered
gram because of its relative low cost and ease of reproducibility. in a larger aerobics room at a local health center.
Additionally, the program adapts to the abilities of individual
participants and thus would be reasonable to implement in a SAFETY
population of older cancer survivors with differing capacities All participants were encouraged to consider consulting with a
for exercise. The program includes basic education on exercise physician prior to beginning of the program. During the eligibility
with the goal of sustainability through a continued home-based screening phone calls, participants were screened for the presence
program (22–24). Our adaptation replaces the osteoarthritis spe- of specific medical conditions including recent joint surgery or
cific educational curriculum with education addressing important current rehabilitation for joint surgery, known cancer metastases
cancer survivorship issues. to bone (indicating higher risk of fracture), or history of cardiac
The primary aim of the study was to evaluate the feasibility disease. Interested participants with these or any other health-
of recruiting and retaining older survivors to participate in an related concerns were required to contact their physician to discuss
8-week group exercise intervention and education program. The participation prior to enrollment. They were prompted to describe
secondary aim was to test the short-term impact of the Fit & the course as “mild to moderate physical activity that includes
Strong! intervention on self-reported physical activity, self-efficacy walking and light weight lifting,” and ask if there were particular
for exercise, and quality of life. types of activities they should avoid. All course instructors were
certified in Basic Life Support.
MATERIALS AND METHODS
PARTICIPANTS ADAPTATION OF FIT & STRONG! EXERCISE INTERVENTION
Participants were eligible for this study if they (1) were 50 years of Prior to enrolling participants, a license to conduct the Fit &
age or older, (2) had a previous diagnosis of cancer, (3) were not Strong! program was obtained through the Fit & Strong! program
on active cancer treatment such as chemotherapy or radiation, office (Institute for Health Research and Policy at the University
and (4) were self-reported able to engage in light-to-moderate of Illinois at Chicago). Additionally, our two instructors com-
physical activity. Although the intention was to target older cancer pleted a Fit & Strong! Master Training Program. Fit & Strong!
survivors (i.e., 65 years of age and up), we chose to allow partic- Master Training instruction provided 8 h of education on top-
ipants 50 years and older. This age allowance was in recognition ics including appropriate types of exercises for older adults and
that some younger patients with lower functional status, either how to implement Fit & Strong! in the community setting. The
at baseline or due to cancer or treatment effects, similarly might program supplied instructional manuals for the instructors to fol-
benefit from the intervention. There was no restriction on type low when facilitating Fit & Strong! courses. In addition to this
of cancer or years since cancer diagnosis for patient eligibility. training, our two instructors held certifications in Chronic Dis-
Individuals taking oral hormonal or biologic treatments for their ease Self-Management (Stanford CDSMP). They were experienced
cancer were allowed to participate at the discretion of the physi- in leading group discussion of health behaviors among adults,
cian investigator (Jana Reynolds). This study was approved by the but our instructors had limited experience leading group exer-
Institutional Review Board of Scott & White Healthcare. cise activities. At least one of our two trained instructors and one
assistant facilitated each 90 min session.
RECRUITMENT Participants had exercise equipment available as recommended
Participants were recruited by advertisement put in the local news- by Fit & Strong! This equipment included resistance bands for arm
paper and flyers placed at senior centers, community cancer sur- exercises, 10 pound adjustable ankle weights for leg exercises, and
vivorship events, and local oncology clinics. The mode of recruit- mats for floor-based exercises. Chairs were available for sitting
ment that generated the most interest in our program, located exercises or for those who required modification to their exercise
in rural Central Texas, was newspaper advertisement. Interested program. Unique to our study, those participating in courses at the
individuals contacted the program coordinator and were screened local health center had the option to use exercise machines, such

www.frontiersin.org April 2015 | Volume 2 | Article 171 | 281

Reynolds et al. Exercise for older cancer survivors

as treadmills and stationary bikes, for the aerobics portion of the made adaptations of exercises as needed to match participant
class. abilities.
Fifty to 60 min of each 90-min class session was dedicated to Thirty minutes of each 90-min class session was dedicated to
aerobic and strength-training activities. The complete instructor- education designed to increase self-efficacy for exercise and exer-
led exercise routine consisted of a 5- to 15-min warm-up with cise adherence (23, 24). The exercise education curriculum was
stretching, 15–20 min of an aerobic activity, 15–20 min of resis- taught by our instructors and a Fit & Strong! manual was provided
tance training, and a 5-min cool-down session. Resistance-training to each participant for reference during each class. Information
exercises followed those recommended in the Fit & Strong! on the components of an exercise program and exercise safety
instructor handbook (i.e., leg lifts while seated in a chair). The was presented (Table 1). Participants engaged in group problem-
aerobic component consisted of sustained walking and a low- solving activities and set physical activity goals. Two educational
impact aerobics routine created by our instructors. Each ses- sessions specific to osteoarthritis were removed from the original
sion used this complete exercise routine in the same sequence. Fit & Strong! curriculum for our program because they were not
Fit & Strong! trained instructors monitored participants and applicable to all cancer survivors.

Table 1 | Sample course curriculum.

Session Fit & Strong! exercise curriculum Cancer survivorship curriculum

Source: Fit & Strong participant manual Source: NCI facing forward: life after cancel treatment (25)
(http://www.fitandstrong.org/index.html)

1 Introduction, consent, and baseline survey

2 Introduction to Fit & Strong (Ch 1) Definition of survivorship (preface) and finding a new “Normal” (p. 1)

3 Benefits and barriers of exercise (Ch 2) Follow-up medical care (pp. 2–5)

4 What to wear (Ch 3)

5 Pain and exercise modifications (Ch 6) Creating a wellness plan (pp. 5–11)

6 Warm-up exercises (Ch 7) Services and community resources (pp. 12–13)

7 Stretching (Ch 8) Nutrition for cancer survivorsa

8 Aerobic exercise (Ch 9)
9 Treatment effects, Part I: fatigue, memory, and concentration (pp. 15–19)

10 Walking (Ch 10) Treatment effects, Part II: pain and physical changes (pp. 20–31)

11 Strengthening exercise (Ch 11)

12 Resistance training (Ch 12) Managing your feelings: stress, depression, anxiety (pp. 37–45)

13 Cool-down exercises (Ch 13) Finding a meaning (pp. 46–48) and making a difference after Cancerb
14 Posture and bone health (Ch 14)
15 Fall prevention (Ch 15) Social and work relationships (pp. 49–55)

16 Setting goals (Ch 16)

17 Other ways to do exercise (Ch17) Learning to relax: instructor guided relaxation exercise no. 1 (p. 60)

18 Lifestyle changes (Ch 18)

19 Exercise: a world of options (Ch 19) Support for caregiversc
20 Getting past barriers to exercise (Ch 20)
21 Diet and exercise (Ch 21)

22 Stress management (Ch 22) Learning to relax: instructor guided relaxation exercise no. 2 (pp. 60–61)
23 Maintaining an active lifestyle (Ch 23) Feedback session on survivorship component
24 Putting it all together (Ch 24) and survey

Each 90 min session included 60 min of exercise and 30 min of education.A sample schedule for the educational component is listed above.All chapters/page
numbers refer to the source listed in heading unless otherwise noted.
a
National Cancer Institute: “Eating hints; Before, During, and After Cancer Treatment” pp.44–45 (26).
b
National Cancer Institute: “Facing Forward. Making a difference in cancer” (27).
c
National Cancer Institute: “Facing Forward: when someone you love has completed cancer treatment” (28).

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 171 | 282
Reynolds et al. Exercise for older cancer survivors

SURVIVORSHIP EDUCATION COMPONENT OF THE FIT & STRONG! exercise, and confidence to participate in exercise without making
PROGRAM their symptoms (of chronic disease) worse. This was reported on
The adaptation of the Fit & Strong! program tested in this study a 10 point scale of “not at all confident” (score of 1) to “totally
replaced the original education on osteoarthritis with cancer- confident” (score of 10). A calculation of the mean rating across
related topics. The content for cancer survivorship education came the three questions determined the score on this measure. This
from the National Cancer Institute’s Facing Forward series along measure showed improvement in exercise-self-efficacy at 2, 6, and
with additional materials from the National Cancer Institute (25– 12 months for participants with osteoarthritis in the original Fit &
29). Topics included the long-term effects of cancer treatment, self- Strong! intervention (22–24).
management of the long-term physical and psychosocial effects of
cancer and cancer treatment, nutrition for cancer survivors, sup- Cancer-related quality of life
port for the caregiver, seeking follow-up medical care, and ways Participants completed the quality of life in adult cancer survivors
to make a difference after cancer (Table 1). Participants were pro- (QLACS) survey, a 47-item questionnaire with five cancer-specific
vided copies of the printed materials to reference in class, and if and seven generic domains. This survey captures issues affect-
desired, to keep for future reference. ing long-term cancer survivors rather than acute cancer or cancer
Trained Fit & Strong! instructors incorporated the cancer sur- treatment-related effects. Cancer-related domains of the survey
vivorship materials during the 30 min educational sessions. They include concerns with appearance, financial problems, distress
presented information from the handouts and then facilitated the over recurrence, family-related distress, and benefits of cancer.
group discussion. Clinicians specializing in the cancer care field Generic domains include negative feelings, positive feelings, cog-
were invited to teach the cancer-specific curriculum in two to nitive problems, physical pain, fatigue, and social avoidance. The
four sessions for each course (Jana Reynolds, Kevin Francis). Par- scores of each domain and a summary score of the cancer-related
ticipants with cancer-specific questions beyond the scope of the (seven items) and generic domains (four items, benefits of cancer
course materials were encouraged to ask their oncologist or health not included in the summary score) were reported (31).
care provider.
At course completion, a course evaluation survey captured the
OUTCOME MEASURES participant’s satisfaction with the exercise and cancer-specific por-
The primary aim of feasibility of Fit & Strong! for older cancer tions of the program. It allowed participants to provide suggestions
survivors was measured by the course completion rate. Partici- for improvement. The intention of this survey was to provide
pants were considered to have successfully completed the study if feedback for future studies.
they filled out a survey at baseline and course completion, pre-
defined as within 1 week of the 24th (final) session. Instructors STATISTICAL ANALYSIS
documented attendance and calculated the total number of ses- Participant characteristics at baseline and the study completion
sions attended by each participant. Self-reported demographics rate used descriptive statistics. The impact of the intervention on
and disease characteristics were obtained at baseline to describe exercise efficacy, physical activity, and quality of life was assessed
the population and identified potential characteristics of likely using paired t -tests. Significance was defined as p ≤ 0.05.
participants for similar studies. These data included gender, age,
weight, height, ethnicity, race, marital status, employment status, RESULTS
type of cancer, time since treatment completion, and whether one Seventy-two (72) cancer survivors participated in one of five
considered if they are living with active cancer (yes/no). courses offered as part of this study, with an average of 14 par-
The secondary aim of the study was to test the short-term ticipants per 8-week course. The mean age of participants was
impact of the intervention on exercise and quality of life. This 70.4 (±13.3) years. Forty-nine of the 72 participants completed
aim was measured by changes in baseline and post-intervention the course, for a 68% retention rate. The mean number of ses-
surveys comparing minutes of physical activity, self-efficacy for sions attended by those who completed the course was 18 out of
exercise, and cancer-related quality of life. The surveys are as 24 (75%).
follows: Participant characteristics are illustrated in Table 2. The major-
ity of participants were female (82%). The average BMI at baseline
Minutes of physical activity was 29.08 (±6.79), with 40% of participants considered obese
Participants reported the number of days in the past 7 days and (BMI 30 or greater). Patients with 18 types of cancer were rep-
they did moderate to strenuous exercise. They also reported how resented in the study, with the majority (52%) of participants
many minutes, on average, they exercised per day. Physical activity reporting a prior breast cancer diagnosis. Almost half the partici-
time per week was calculated by multiplying the reported days by pants (46%) had been diagnosed and completed cancer treatment
the reported minutes, similar to the original Fit & Strong! study at least 5 years prior, with a median time since treatment of 7 years.
(22–24). Though not on active intravenous chemotherapy or radiation per
study protocol, six participants (8%) considered themselves to
Self-efficacy for exercise have active cancer during the study.
Self-efficacy for exercise was measured on a three item scale devel- Participants significantly increased their weekly total min-
oped Lorig and colleagues (30). Participants reported their con- utes of moderate to strenuous exercise from baseline to
fidence to do frequent aerobic exercise, frequent strengthening post-intervention (94.1 vs. 131.5 min, p = 0.0005). Their overall

www.frontiersin.org April 2015 | Volume 2 | Article 171 | 283

Reynolds et al. Exercise for older cancer survivors

Table 2 | Participant demographics and cancer history. was observed from baseline to post-intervention (p = 0.0198)
(Table 4).
N (%) In a post-intervention survey of the utility of the course and
the cancer-specific education, 79% reported learning something
Age
helpful about cancer survivorship that they did not know before
<70 37 (52.11)
starting our course. Sixty-eight percent of participants reported
≥70 34 (47.89)
sharing the information on support for caregivers with a friend or
Sex a family member. Some enjoyed meeting others with similar expe-
Male 13 (18.06) riences and took pleasure in sharing information. All respondents
Female 59 (81.94) stated that they would recommend the course to another cancer
Race survivor.
White 61 (89.71) One of the survivors thought the survivorship discussions were
Other 7 (10.29) too emotional. Another reported feeling uncomfortable during
Martial Status
voluntary group discussion, preferring more privacy regarding
Married 37 (54.41)
cancer survivorship issues. Some desired a longer exercise course
Not married 31 (45.59)
or to be allowed to repeat the course again. Though participants
were encouraged to set goals and create a plan for sustained exer-
Employment Status
cise beyond the program, many wanted to continue within the
Employed 8 (12.12)
current group setting.
Not employed 58 (87.88)

BMI DISCUSSION
<20 3 (4.41) Despite the known physical and emotional benefits of exercise in
20–25 20 (29.41) cancer survivors, the majority of studies targeting physical activ-
25–30 18 (26.47) ity in this population either exclude or do not achieve high levels
30–35 17 (25.00) of participation from older cancer survivors (20, 21). Our study
35+ 10 (14.71) was successful in recruiting a population of older cancer survivors
Type of Cancer with a mean age of 70. The 68% retention rate and 75% session
Breast 37 (52.11) attendance rate is indicative of an intervention individuals were
Colon 5 (7.04) willing to engage in over time. These results support the feasibility
Prostate 5 (7.04) of recruiting and retaining older cancer survivors to participate in
Lung 4 (5.63) an 8-week group exercise intervention and education program.
Othera 20 (28.17) Our study utilized Fit & Strong!, an evidence-based physical
activity intervention for older adults with osteoarthritis, as it pre-
Time since completion of cancer treatment
viously showed long-term physical activity benefits in older adults
<1 year 14 (20.90)
with a mean age of 73 (23, 24). Hughes and colleagues observed
1–5 years 22 (32.84)
similar retention rate (72%) and attendance (79% of sessions)
5+ years 31 (46.27)
in their original Fit & Strong! for osteoarthritis study (23). Our
Participants who consider themselves to study kept the same physical activity content and adapted the
have active cancer educational component of Fit & Strong! by replacing osteoarthri-
No 66 (91.67) tis education with education on common issues facing cancer
Yes 6 (8.33) survivors. No major safety issues were reported.
a
Participants in our study showed improvement in level of phys-
Hodgkin’s lymphoma, non-Hodgkin’s lymphoma, chronic leukemia, multiple
ical activity and mood, supporting the short-term efficacy of Fit &
myeloma, adenoid cystic cancer, gastrointestinal stromal tumor, bladder, kidney,
Strong! when adapted to a population of older cancer survivors.
pancreatic, thyroid, ovarian, uterine, cervical, and vulvar cancer.
Participants successfully increased their self-reported weekly min-
utes of physical activity from 94.1 minutes at the beginning of
self-efficacy for exercise (i.e., average of the three self-efficacy for the study to 131.5 minutes at the end of the 8-week interven-
exercise items) did not differ from baseline to post-intervention tion (Table 3). Participants showed significant improvement in the
(p = 0.0964). However, in the measure of self-efficacy for doing negative feeling domain of the cancer-related quality of life assess-
for aerobic exercise regularly, participants showed significant ment (QLACS), though not in overall quality of life (Table 4). The
improvement in their ratings from baseline to post-intervention questions in the negative feeling domain address depression and
(M Baseline = 7.94, M Post-intervention = 8.85; p = 0.05) (Table 3). anxiety, which are reported more commonly in cancer survivors
Participants’ scores on the generic and cancer-specific and should be a specific measure in future studies (3, 4).
subscale of the QLACS survey did not differ from base- Participants also improved exercise-self-efficacy specific to
line to post-intervention (p = 0.0770 and p = 0.9303, respec- aerobic activity; however, no changes were observed in overall
tively). Nonetheless, an improvement in scores on the negative exercise-self-efficacy (Table 3). This is in contrast to the findings
feeling domain (questions related to anxiety and depression) of the original Fit & Strong! intervention in which participants

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 171 | 284
Reynolds et al. Exercise for older cancer survivors

Table 3 | Intervention impact on exercise efficacy and total minutes of physical activity by paired t -test.

Baseline (n = 72) Post (n = 49) Paired change (n = 49) t p

Mean (±SD) Mean (±SD) Mean (±SD)

1. Confidence to do strength and flexibility 8.68 (±1.75) 9.19 (±1.50) 0.45 (±1.86) 1.64 0.1069
exercises 3–4 times a week
2. Confidence to do aerobic exercises 3–4 times a 7.94 (±2.45) 8.85 (±1.96) 0.69 (±2.37) 2.01 0.0503
week
3. Confidence to exercise without making 8.51 (±1.98) 8.60 (±2.20) 0.10 (±2.26) 0.32 0.7511
symptoms (of chronic disease) worse
Overall self-efficacy for exercise (mean of 1–3) 8.39 (±1.73) 8.88 (±1.67) 0.41 (±1.69) 1.70 0.0964
Total minutes of physical activity 94.10 (±87.02) 131.51 (±91.01) 42.22 (±73.80) 3.79 0.0005

Bold text indicates statistically significant values (p ≤ 0.05).

Table 4 | Intervention impact on Quality of Life (QLACS scale) by paired t -test.

Baseline (n=72) Post (n=49) Paired change (n=49) t p-value

Mean (±SD) Mean (±SD) Mean (±SD)

GENERIC QOL
Negative feelingsa 10.18 (±4.40) 9.22 (±3.62) −1.27 (±3.17) −2.44 0.0198
Positive feelings 10.26 (±4.69) 10.32 (±4.72) −0.72 (±3.85) −1.17 0.2509
Cognitive problems 10.47 (±4.02) 10.43 (±3.53) 0.04 (±3.11) 0.10 0.9241
Sexual problems 11.83 (±6.49) 10.29 (±5.55) −1.59 (±4.81) −1.78 0.0864
Energy/fatigue 14.81 (±2.81) 14.67 (±4.21) 0.15 (±4.83) 0.21 0.8319
Pain 11.66 (±5.74) 11.00 (±5.52) −0.54 (±4.95) −0.69 0.4912
Social avoidance 9.11 (±5.19) 8.57 (±3.73) 0.12 (±3.14) 0.25 0.8049
Generic summary score 77.42 (±24.99) 73.03 (±18.56) −8.52 (±22.03) −1.86 0.0770
CANCER-SPECIFIC QOL
Financial problems 7.00 (±4.75) 6.48 (±4.78) 0.30 (±2.46) 0.80 0.4308
Benefits 19.24 (±6.24) 19.49 (±6.36) 0.51 (±4.38) 0.77 0.4483
Distress-family 9.43 (±5.61) 8.80 (±5.00) −0.18 (±3.92) −0.30 0.7621
Appearance 8.80 (±5.70) 8.48 (±4.82) 0.04 (±3.91) 0.08 0.9395
Distress-recurrence 12.79 (±6.33) 12.91 (±6.09) −0.21 (±3.26) −0.42 0.6755
Cancer-specific summary score 37.95 (±16.88) 36.89 (±14.81) −0.12 (±8.87) −0.09 0.9303

a
Questions in the negative feeling domain (1) bothered by mood swings, (2) felt blue or depressed, (3) worried about little things, and (4) felt anxious.
Bold text indicates statistically significant values (p ≤ 0.05).

with osteoarthritis showed improvement on the overall exercise- on outcomes for cancer survivors beyond the 8-week intervention.
self-efficacy scale at 2, 6, and 12 months (22, 23). One explana- Second, the study did not measure the effects of the intervention
tion of the variation between the studies is that the self-efficacy on actual physical health or function; outcomes were limited to
scale is more specific to persons with symptoms of osteoarthritis. self-reported measures. Future studies should consider tests of the
The item “confidence to do exercise without making symptoms intervention effects using direct measures of physical health and
of chronic disease worse” may be more relevant to osteoarthri- function.
tis pain symptoms rather than a population of cancer survivors Additionally, participants did not meet the 150 minutes of phys-
with a wide variation of chronic symptoms. A self-efficacy scale ical activity per week as recommended in the guidelines (13, 14).
examining perceived ability to do exercise without a focus on Though it is reasonable for capable participants to strive to this
symptomatology of chronic disease may be more appropriate for goal, it may not be necessary to gain benefits of exercise. In a
cancer survivors. separate study of older cancer survivors, an increase in minutes of
This study had several limitations. First, this study was not physical activity over baseline but to less than a total of 150 minutes
designed to test long-term effects of the intervention on physi- per week still showed measurable functional and health-related
cal activity, self-efficacy, or cancer-related quality of life. Studies benefits (32).
of long-term efficacy and sustained benefits will be necessary to Despite the limitations of this study, the majority of the feed-
establish whether this program is likely to have meaningful impact back on the program was positive. Most participants indicated

www.frontiersin.org April 2015 | Volume 2 | Article 171 | 285

Reynolds et al. Exercise for older cancer survivors

that they would recommend a similar course to other survivors. 9. Daley AJ, Crank H, Saxton JM, Mutrie N, Coleman R, Roalfe A. Randomized
Many participants expressed appreciation for meeting other can- trial of exercise therapy in women treated for breast cancer. J Clin Oncol (2007)
25:1713–21. doi:10.1200/JCO.2006.09.5083
cer survivors and sharing experiences. Most participants in our
10. Alfano CM, Smith AW, Irwin ML, Bowen DJ, Sorensen B, Reeve BB, et al. Physical
study were female (82%), suggesting this type of group interven- activity, long-term symptoms, and physical health-related quality of life among
tion may be of particular interest to females. Almost half of the breast cancer survivors: a prospective analysis. J Cancer Surviv (2007) 1:116–28.
participants were at least 5 years post-cancer treatment, indicating doi:10.1007/s11764-007-0014-1
older cancer survivors are interested in a cancer-related exercise 11. Peddle CJ, Au HJ, Courneya KS. Associations between exercise, quality of life,
and fatigue in colorectal cancer survivors. Dis Colon Rectum (2008) 51:1242–8.
program long after they finish their treatment (Table 2). Cancer
doi:10.1007/s10350-008-9324-2
survivors undergoing therapy were excluded from the protocol 12. Haas BK, Kimmel G, Hermanns M, Deal B. Community-based FitSTEPS for
for the purpose of keeping baseline characteristics similar and life exercise program for persons with cancer: 5-year evaluation. J Oncol Pract
to minimize conflicts between class times and cancer treatment (2012) 8:320–4. doi:10.1200/JOP.2012.000555
schedules. Given that exercise is safe for most patients while under- 13. Schmitz KH, Courneya KS, Matthews C, Demark-Wahnefried W, Galvão DA,
Pinto BM, et al. American college of sports medicine roundtable on exer-
going treatment, it would be reasonable to include them in future cise guidelines for cancer survivors. Med Sci Sports Exerc (2010) 42:1409–26.
programs (20). doi:10.1249/MSS.0b013e3181e0c112
Anecdotally, many participants reported a desire to continue 14. 2008 Physical Activity Guidelines for Americans. U.S. Department of Health and
the course indefinitely as their primary exercise program. The Fit Human Services (2008). Available from: http://www.health.gov/paguidelines/
& Strong! intervention focuses primarily on initiating an exercise pdf/paguide.pdf
15. Holmes MD, Chen WY, Feskanich D, Kroenke CH, Colditz GA. Physical activ-
routine that could be sustained in one’s home after course com- ity and survival after breast cancer diagnosis. JAMA (2005) 293:2479–86.
pletion. The original version was not designed to continue in a doi:10.1001/jama.293.20.2479
group setting. Cancer survivors may benefit from an additional 16. Meyerhardt JA, Giovannucci EL, Holmes MD, Chan AT, Chan JA, Colditz GA,
adaptation that helps participants find appropriate community- et al. Physical activity and survival after colorectal cancer diagnosis. J Clin Oncol
(2006) 24:3527–34. doi:10.1200/JCO.2006.06.0855
based group exercise programs with social support similar to the
17. Holick CN, Newcomb PA, Trentham-Dietz A, Titus-Ernstoff L, Bersch AJ,
Fit & Strong! program. It would also be reasonable to consider Stampfer MJ, et al. Physical activity and survival after diagnosis of invasive breast
monthly maintenance classes open to all graduates to help inspire cancer. Cancer Epidemiol Biomarkers Prev (2008) 17:379–86. doi:10.1158/1055-
and refocus exercise goals for long-term sustainability of benefits. 9965.EPI-07-0771
18. Meyerhardt JA, Giovannucci EL, Ogino S, Kirkner GJ, Chan AT, Willett W, et al.
Physical activity and male colorectal cancer survival. Arch Intern Med (2009)
CONCLUSION 169:2102–8. doi:10.1001/archinternmed.2009.412
Results of this pilot study support the feasibility of implement- 19. Je Y, Jeon JY, Giovannucci EL, Meyerhardt JA. Association between physical
ing an 8-week exercise intervention for older cancer survivors. activity and mortality in colorectal cancer: a meta-analysis of prospective cohort
Short-term efficacy of the Fit & Strong! program was noted from studies. Int J Cancer (2013) 133:1905–13. doi:10.1002/ijc.28208
baseline to the end of the 8-week intervention by increases in min- 20. Courneya KS, Vallance JK, McNeely ML, Karvinen KH, Peddle CJ, Mackey
JR. Exercise issues in older cancer survivors. Crit Rev Oncol Hematol (2004)
utes of physical activity, increased self-efficacy for aerobic exercise, 51:249–61. doi:10.1016/j.critrevonc.2004.05.001
and decreased negative feelings in the quality of life (QLACS) 21. Pekmezi DW, Demark-Wahnefried W. Updated evidence in support of diet
scale. Tests of efficacy and effectiveness over time are needed to and exercise interventions in cancer survivors. Acta Oncol (2011) 50:167–78.
determine the utility of this intervention as a program to pro- doi:10.3109/0284186X.2010.529822
mote sustained physical activity among older cancer survivors and 22. Hughes SL, Seymour RB, Campbell R, Pollak N, Huber G, Sharma L. Impact of
the fit and strong intervention on older adults with osteoarthritis. Gerontologist
support long-term health outcomes. (2004) 44:217–28. doi:10.1093/geront/44.2.217
23. Hughes SL, Seymour RB, Campbell RT, Huber G, Pollak N, Sharma L, et al. Long-
REFERENCES term impact of fit and strong! on older adults with osteoarthritis. Gerontologist
1. Frieden TR, Jaffe HW, Stephens JW, Thacker SB, Zaza S, Centers for Disease (2006) 46:801–14. doi:10.1093/geront/46.6.801
Control and Prevention (CDC). Cancer Survivors – United States, 2007. MMWR 24. Hughes SL, Seymour RB, Campbell RT, Desai P, Huber G, Chang HJ. Fit
Morb Mortal Wkly Rep (2011) 60(9):269–72. and Strong!: bolstering maintenance of physical activity among older adults
2. American Cancer Society. Cancer Treatment and Survivorship Facts & Figures with lower-extremity osteoarthritis. Am J Health Behav (2010) 34:750–63.
2014 – 2015. Atlanta: American Cancer Society (2014). doi:10.5993/AJHB.34.6.10
3. Stein KD, Syrjala KL, Andrykowski MA. Physical and psychological long-term 25. Facing Forward: Life After Cancer Treatment. National Cancer Insti-
and late effects of cancer. Cancer (2008) 112:2577–92. doi:10.1002/cncr.23448 tute (2014). Available from: http://www.cancer.gov/cancertopics/coping/
4. Polsky D, Doshi JA, Marcus S, Oslin D, Rothbard A, Thomas N, et al. Long-term life-after-treatment.pdf
risk for depressive symptoms after a medical diagnosis. Arch Intern Med (2005) 26. Eating Hints: Before, During, and After Cancer Treatment. National Cancer
165:1260–6. doi:10.1001/archinte.165.11.1260 Institute (2014). Available from: http://www.cancer.gov/cancertopics/coping/
5. Ganz PA. Monitoring the physical health of cancer survivors: a survivorship- eatinghints.pdf
focused medical history. J Clin Oncol (2006) 24:5105–11. doi:10.1200/JCO.2006. 27. Facing Forward: Making a Difference in Cancer. National Cancer Institute (2014).
06.0541 Available from: http://www.cancer.gov/cancertopics/facing-forward.pdf
6. Hofman M, Ryan JL, Figueroa-Moseley CD, Jean-Pierre P, Morrow GR. Cancer- 28. Facing Forward: When Someone You Love Have Completed Cancer
related fatigue: the scale of the problem. Oncologist (2007) 12(Suppl 1):4–10. Treatment. National Cancer Institute (2014). Available from: http://www.
doi:10.1634/theoncologist.12-S1-4 cancer.gov/cancertopics/coping/someone-you-love-completed-cancer-
7. Koch L, Jansen L, Brenner H, Arndt V. Fear of recurrence and disease progression treatment.pdf
in long-term (>/= 5 years) cancer survivors – a systematic review of quantitative 29. Follow-up Care After Cancer Treatment. National Cancer Institute
studies. Psychooncology (2013) 22:1–11. doi:10.1002/pon.3022 (2014). Available from: http://www.cancer.gov/cancertopics/factsheet/Therapy/
8. Burnham TR, Wilcox A. Effects of exercise on physiological and psycholog- followup
ical variables in cancer survivors. Med Sci Sports Exerc (2002) 34:1863–7. 30. Lorig K, Stewart A, Ritter P, Gonzalez V, Laurent D, Lynch J. Chapter 2. Construc-
doi:10.1097/00005768-200212000-00001 tion of Measures of Behaviors, Self-Efficacy, and Outcomes, Outcome Measures for

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 171 | 286
Reynolds et al. Exercise for older cancer survivors

Health Education and Other Health Care Interventions. Thousand Oaks, CA: Sage Education and Promotion section) panel of Review Editors. Because this Research
Publications, Inc (1996). p. 10–45. Topic represents work closely associated with a nationwide evidence-based movement
31. Avis NE, Smith KW, McGraw S, Smith RG, Petronis VM, Carver CS. Assess- in the US, many of the authors and/or Review Editors may have worked together
ing quality of life in adult cancer survivors (QLACS). Qual Life Res (2005) previously in some fashion. Review Editors were purposively selected based on their
14:1007–23. doi:10.1007/s11136-004-2147-2 expertise with evaluation and/or evidence-based programming for older adults. Review
32. Morey MC, Snyder DC, Sloane R, Cohen HJ, Peterson B, Hartman TJ, et al. Editors were independent of named authors on any given article published in this
Effects of home-based diet and exercise on functional outcomes among older, volume.
overweight long-term cancer survivors: renew: a randomized controlled trial.
JAMA (2009) 301:1883–91. doi:10.1001/jama.2009.643 Received: 18 June 2014; accepted: 17 September 2014; published online: 27 April 2015.
Citation: Reynolds J, Thibodeaux L, Jiang L, Francis K and Hochhalter A (2015) Fit
Conflict of Interest Statement: The authors declare that the research was conducted & Strong! promotes physical activity and well-being in older cancer survivors. Front.
in the absence of any commercial or financial relationships that could be construed Public Health 2:171. doi: 10.3389/fpubh.2014.00171
as a potential conflict of interest. This article was submitted to Public Health Education and Promotion, a section of the
journal Frontiers in Public Health.
This paper is included in the Research Topic, “Evidence-Based Programming for Copyright © 2015 Reynolds, Thibodeaux, Jiang , Francis and Hochhalter. This is an
Older Adults.” This Research Topic received partial funding from multiple govern- open-access article distributed under the terms of the Creative Commons Attribution
ment and private organizations/agencies; however, the views, findings, and conclusions License (CC BY). The use, distribution or reproduction in other forums is permitted,
in these articles are those of the authors and do not necessarily represent the official provided the original author(s) or licensor are credited and that the original publica-
position of these organizations/agencies. All papers published in the Research Topic tion in this journal is cited, in accordance with accepted academic practice. No use,
received peer review from members of the Frontiers in Public Health (Public Health distribution or reproduction is permitted which does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 171 | 287

 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00164

Adoption of evidence-based health promotion programs:

perspectives of early adopters of Enhance®Fitness in
YMCA-affiliated sites
Basia Belza 1,2 *, Miruna Petrescu-Prahova 1 , Marlana Kohn 1 , Christina E. Miyawaki 1,3 , Laura Farren 1 ,
Grace Kline 1 and Ann-Hilary Heston 4
1
Health Promotion Research Center, University of Washington, Seattle, WA, USA
2
School of Nursing, University of Washington, Seattle, WA, USA
3
School of Social Work, University of Washington, Seattle, WA, USA
4
Healthy Living Department, YMCA of the USA, Chicago, IL, USA

Edited by: Purpose: To identify facilitators and barriers among early adopters of Enhance® Fitness
Matthew Lee Smith, The University of
(EF), in Young Men’s Christian Association-affiliated (Y-affiliated) sites from the perspective
Georgia, USA
of program staff. EF is an evidence-based group exercise program for seniors.
Reviewed by:
Mary Odum, Texas A&M University, Methods:This qualitative study used semi-structured phone interviews with 15 staff mem-
USA
Elisa Beth Hutson McNeill, Texas
bers representing 14 Y-affiliated sites. Interviews were digitally recorded, transcribed, and
A&M University, USA analyzed using qualitative content analysis informed by the RE-AIM framework.
*Correspondence:
Findings: Staff were, on average, 48.7 years old (SD 13.5) and had been involved with EF
Basia Belza, Health Promotion
Research Center, University of for 5.2 years (SD 3.1). Key themes related to facilitating adoption of EF were: match with
Washington, 1107 NE 45th Street, theY mission, support from different organizational levels, match between the target popu-
Suite 200, Seattle, WA 98195, USA lation need and EF, initial and on-going financial support, presence of champions, novelty of
e-mail: [email protected]
EF, an invitation to partner with a community-based organization to offer EF, and program-
specific characteristics of EF. Key themes related to barriers interfering with EF adoption
included competing organizational programs and space limitations, limited resources and
expertise, and costs of offering the program.
Implications: Our findings identify the types of organizational support needed for adop-
tion of evidence-based health promotion programs like EF. Recommendations for practice,
research, and policy based on the findings, including assessing organizational readiness,
researching late adopters, and developing revenue streams, may help facilitate program
adoption. Packaging and sharing these practical recommendations could help community-
based agencies and nationally networked organizations facilitate adoption of EF and other
evidence-based programs.
Keywords: older adults, RE-AIM, physical activity, dissemination, adoption, evidence-based programs, community
intervention, dissemination framework

INTRODUCTION and include exercises for cardiovascular endurance, balance, flex-

Since the development of evidence-based medicine (EBM) over ibility, and strength. Research has demonstrated that EF improves
two decades ago (1), the Centers for Disease Control and Pre- upper and lower body muscle strength and flexibility (9, 10). EF
vention (CDC) has promoted healthy aging with an emphasis on participants report improved over-all health (9). Participation in
reliable, efficient, and cost-effective care models with measurable EF has been associated with healthcare costs saving. The average
outcomes (2). The goal of evidence-based programs is to create increase in annual total healthcare costs was less among EF par-
healthier communities and prevent chronic diseases (3). Translat- ticipants compared to non-participating controls in a managed
ing evidence-based programs from research studies to community health care plan ($642 vs. 1175) (6). Among Medicare beneficia-
practice and sustaining them is a top priority for public health ries, enrollment in EF was associated with per person medical
researchers and practitioners. savings of $945/year after enrollment (8). Since 2001, there have
Evidence-based health promotion programs for older adults been a total of 42,560 EF participants (unduplicated) served at a
have been adopted and implemented by organizations in commu- total of 689 sites in 33 states and Washington DC (Susan Snyder,
nities throughout the United States (4, 5). Enhance®Fitness (EF) personal communication, Senior Services, 2014 April 14).
is an evidence-based group exercise program for frail and active Senior Services (Seattle, WA, USA) licenses and disseminates EF
older adults (6–8). The one hour classes meet three-times a week to multipurpose social service agencies, faith-based organizations,

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 164 | 288
Belza et al. Adoption of Enhance® Fitness

retirement communities, and recreational organizations. With MEASURE

more than 10,000 community locations, Young Men’s Christian We developed a structured interview guide for staff with ques-
Association (Y) is a leading non-profit committed to improving tions that were informed by the RE-AIM framework (Table 1). We
the nation’s health and well-being by offering programs that nur- piloted our guide, which adds rigor to our data collection process
ture the mind, body, and spirit. From 2005 to 2012, 116 Y-affiliated (19). The interview guide contained a total of 39 questions and
community sites adopted EF. These included both classes on-site probes about benefits of EF, fitness checks, facilitators and barri-
at Y brick-and-mortar buildings, and classes licensed by Ys but ers/challenges to offering EF, staff responsibilities, support from
offered in community settings such as churches or retirement com- the Y management for offering EF, and strategies to recruit EF
munities. These sites represent a first stage in the adoption of EF participants and instructors. Seven of the 39 questions included
by a nationally networked organization; as such, their experience skip patterns (#5, 9, 12, 15, 16, 30, and 39). For example, item #9
is likely to inform efforts to scale-up adoption of EF and other asked: were you involved in the original adoption of EF? If the staff
evidence-based programs. In this paper, we refer to these sites as answered in the affirmative, then the follow on question was asked:
early adopters (11). if yes, how did your YMCA come about offering EF? If the staff
Various models have been used to evaluate evidence-based answered not in the affirmative the interviewer immediately went
health promotion programs. One of these program planning onto the next question. Ten of the 39 items were close-ended or
and evaluation models is RE-AIM. RE-AIM, an acronym for demographic items requiring short responses (#1, 2, 3, 7, 8, 17, 20,
Reach, Effectiveness, Adoption, Implementation, and Mainte- 36, 37, and 38). Additionally, we asked staff their age, duration of
nance, is a systematic process that researchers, practitioners, and time involved with EF, educational level, and title of their position.
policy makers use to evaluate the dissemination of health pro-
motion programs (12). Within the RE-AIM framework, adop- PROCEDURES
tion refers to the “proportion and representativeness of settings Our study was determined to be exempt by the University of Wash-
willing to initiate a given program” (13). Understanding how ington Institutional Review Board. We obtained administrative
adoption of interventions plays out in different organizational program records from Senior Services of all Y-affiliated program
settings is critical to the current and potential impact of an sites that had offered EF between January 2005 and June 2012.
intervention (14). Inclusion criteria for this study were EF program management staff
Previous research has identified a number of motivating fac- listed in the administrative program records from 2005 to 2012 in
tors for program adoption: participant interest and/or demand, 116 Y-affiliated sites and who had complete contact information.
proven safety and effectiveness for older adults, cost, and a well- Exclusion criteria were: (1) staff on this list without complete con-
rounded program structure that attracts multiple groups (13). tact information, and (2) staff during the pre-enrollment screening
Additionally, the availability of resources is important for pro- call were determined that they did not have experience with EF. The
gram adoption (15). Facilitating factors for program adoption list included 94 names of EF program management staff; of which
identified in previous studies include having a curriculum, avail- 75 has complete contact information. Recruiting letters were sent
ability of training, space, and equipment (12), awareness of the to those 75 staff. Additionally, a Y-USA staff member (AHH) sent
importance of promoting physical activity in the community and emails to the staff employed by Ys inviting them to participate in
internal support for physical activity interventions (15), sufficient our study. Two reminder recruiting postcards were sent to staff
funds, leadership support, capable staff, and successful partner- who had not responded to the initial recruiting letter; reminders
ships and collaborations (16). Barriers include scheduling issues, were sent at 2 and 4 weeks after the initial mailing. Interested
lack of space, and insufficient participant recruitment efforts (13), staff called a study phone line or sent an email to the study email
program cost (16, 17), and lack of leadership, and time and train- account. Twenty-five out of 75 (33%) responded to the recruit-
ing (16). These findings reinforce the idea that adoption can be ing letter, and 8 were determined ineligible for the study. Two
improved by developing organizational support and capacity to staff were placed on a wait list. The study recruiting coordina-
deliver a program (18). tor (LF) tracked and responded to all phone and email messages,
The research question for this study was: what are the facilita- determined eligibility, and scheduled interviews. At the beginning
tors and barriers among early adopters of EF in Y-affiliated sites of each phone interview, verbal informed consent was obtained.
from the perspective of program staff. Based on our findings, we Interviews were conducted by two members of the research team
provide practice, research, and policy recommendations that may (BB and MPP). Each study participant received a gift card for
help inform the adoption of other evidence-based programs in $20. Interviews ranged from 30 to 71 min (average duration was
community settings. Increasing the number of community orga- 47 min) and were digitally recorded. In 2008, Gill et al. (20) note
nizations that adopt evidence-based health promotion programs that when conducting interviews the length varies depending on
for older adults will contribute to the goal of creating healthier the topic, researcher and participant. However, on average, the
communities and preventing chronic disease. duration of interviews about health care topics is 20–60 min (20).
The average length of our interview is within this range.
MATERIALS AND METHODS The interviews were transcribed by a professional transcrip-
DESIGN tionist using a “lightly edited verbatim” style for readability with
This qualitative study used semi-structured individual phone an emphasis on sentence structure. This is a more frequently used
interviews with 15 staff from 14 Y-affiliated sites that had style over “strictly verbatim” since it is executed without compro-
responsibility for the oversight of EF. mising the actual content or altering the intended expression (21).

www.frontiersin.org April 2015 | Volume 2 | Article 164 | 289

Belza et al. Adoption of Enhance® Fitness

Table 1 | Interview guide for Y administrative staff.

INFORMATION ABOUT THE INTERVIEWEE AND ENHANCE® FITNESS OVERSIGHT

1. What is your title with the YMCA? How long have you held this position?

2. Approximately what year did your YMCA first become involved with EF?

3. Approximately what year did you become involved with EF?

4. What are your responsibilities related to EF?

5. Are you responsible for the oversight of EF with your YMCA?

a. If not, what is the title of the person responsible for oversight?
6. What are the responsibilities related to oversight of the operations of EF with your YMCA?

7. Is your YMCA still offering EF classes? Yes No

8. What, if any, other exercise classes and/or programs does/did your YMCA offer specifically for seniors (defined as older than 62 years) besides EF?
ENHANCE® FITNESS READINESS AND ADOPTION
9. *Were you involved in the original adoption of EF?
a. If yes, how did your YMCA come about offering EF?

10. *What was the primary motivation for your YMCA getting involved in EF?

11. *What were some of the other motivations for your YMCA getting involved in EF?

12. *Does/did your YMCA receive funding to provide EF classes?

a. If yes, what kind of funding?
b. Source of funding?
c. Use of funds? (Probe: staff support, participant fee offset, marketing, etc.)

13. Do you think your YMCA has/had the right and adequate number of . . . to support EF classes?
a. Staff
b. Instructors
c. Class materials (chairs, weights, etc.)
d. Space/room

14. Are there any resources you wish you had, or had more of, to support EF classes at your YMCA?

15. *Is there someone at your YMCA that is a champion for EF? (A champion can be paid staff or a volunteer who helps keep classes going, recruits new
participants, or works to expand the program, for example)
a. If YES: can you describe some of the things this person does to champion EF?

16. Are there paid and/or volunteer staff that manage or oversee EF operations with your YMCA? (This would include scheduling classes, managing
instructors, and/or answering EF questions for current or potential participants.)
a. If YES: how was staff recruited or selected to manage EF with your YMCA?

17. Is the management of EF:

a. Centralized and occurs at the association-level, or
b. De-centralized and managed at the branch level?

18. What are/were the methods used to recruit EF instructors to teach classes with your YMCA?

19. What are/were the methods used to recruit participants to EF classes with your YMCA?

20. Location of EF classes:

a. Are EF classes held at: a) your YMCA, b) off-site locations, or c) both?
b. If off-site, what type of locations? If more than one site, list all sites and be specific as to type of site. (churches, senior centers, community
centers, Parks and Recreation facilities, schools, and retirement communities)?

21. What is/was the primary motivation for you personally getting involved in EF? (Probe: part of my job, personal interest, etc.)

22. What are/were some of your other motivations for you getting personally involved in EF? (Probe: saw the benefits, have aging family members that
had benefited from this or similar programs.)

(Continued)

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 164 | 290
Belza et al. Adoption of Enhance® Fitness

Table 1 | Continued

EXPERIENCE WITH ENHANCE® FITNESS

23. What benefits do you think your YMCA gets/got from participating in EF? (Probe: member engagement, community outreach, and serving a need).
a. Is/was your YMCA reimbursed in any way for offering EF classes?
i. If yes, by whom?
b. Do/did your participants pay a fee to participate in EF classes?
i. If yes, how much?
ii. If the fee changed over time, provide a range but note the current amount
iii. Does your Y ever offer EF:
1. On a sliding scale?
2. For reduced cost?
iv. How reasonable do you think this fee is?
v. How does this fee compare to other like classes or programs? (Probe: More, same, less than?)
c. Are EF classes through your YMCA available to both Y members and non-members?
i. Does your Y track conversion rates of non-member program participants to members? (e.g., For 10 non-member participants, 1 becomes a
member.)
1. If YES: what is the conversion rate?
d. Do you know if any of your EF participants are/were reimbursed for participating in EF, such as through health insurance?
i. If yes, by what plan or program?

24. What do/did you like most about EF? (Probe: full classes, positive reports from participants, outcomes tracking)

25. What do/did you like least about EF? (Probe: time of the day the class is offered, instructor, room the class was held, conducting the fitness checks)

26. What are/were characteristics of EF classes that you think make them successful? (Probes: instructor ownership of the class, size of class,
well-ventilated room, good time of day, room easily accessible, and class offered immediately before or after another event like a meal program)

27. What are/were characteristics of EF classes that you think are barriers to success (probe: too small of a room, mismatch between instructor and
participant characteristics, high fees, time of day)?

28. If the class is no longer offered: why do you think EF is not/no longer offered at your YMCA? (Probe: cost of running EF, no instructor, instructor not a
good match with participants, not enough participants, not large enough space, and other concurrent exercise classes)

29. What are/have been your challenges with offering EF? (Probe: finding instructors, frailty of participants, and for the participant transportation to and
from class)

30. Are/were there particular issues you have faced in offering EF?
a. If YES: can you describe these issues, and how you have handled them?

31. What does/did your YMCA do that you think makes EF appealing to: (Probes: advertising, offering at prime time, and offer it at no or low cost)
a. Participants and members at your YMCA?
b. The greater community that your YMCA serves?
c. If your Y receives funding to offer EF classes (separate from member due or class fees), who are the other funders (probe: specific organization
and individual)?

32. What does/did your YMCA do that you think makes EF not appealing?

33. What are the reasons you believe your YMCA has/not been able to implement/maintain EF classes?

34. Do you have any recommendations for changes to EF based on your experiences?

35. How was your experience with EF been similar to or different from other classes at the YMCA, particularly fitness classes or classes for seniors?

36. Regarding your Y members:

a. About how many members does your YMCA serve?
b. About what percent of your members are 65 and older?
c. About what percent of your members are in the 50–64-year-old age range?
DEMOGRAPHIC ITEMS

37. What is your age?

38. In what education category do you fall: less than college degree, some college or college degree, and more than college degree?
CLOSING
39. Is there anything else you would like us to know about your experience with EF with your YMCA?

Items with high relevance to the findings presented in this paper are noted with an asterisk.

www.frontiersin.org April 2015 | Volume 2 | Article 164 | 291

Belza et al. Adoption of Enhance® Fitness

“Lightly edited”is simply used to refer to the reduction of superflu- organization to offer EF, and program-specific characteristics of
ous words such as “hmm, ah, mm, you know, well, yeah, uh-huh.” EF, such as being evidence-based and having a recognizable name.
Transcripts are still considered full and complete and do not in
any way deviate from representing the full intent and thoughts as Match with the Y mission
expressed by each individual respondent. Transcripts were then The Y mission is to put Christian principles into practice through
entered and analyzed in ATLAS.ti version 7. programs that build a healthy spirit, mind, and body for all (25).
Staff employed by Ys and community organizations with whom
DATA ANALYSIS the Y partners noted that EF’s evidence-based curriculum was a
Our research team had content expertise in dissemina- good match with the Y mission since it is proven to improve older
tion and implementation science, administrative management, adults’ physical health in a fun and engaging atmosphere that pro-
community-based participatory research, gerontology, public motes social interaction. Staff expressed strong commitment to the
health, and healthy aging. All members of our team that were Y mission and to addressing unmet demographic needs:“Adopting
involved in the analysis had previous experience in conducting an evidence-based program fit well with the Y goals and standards
qualitative analysis. Our study used qualitative content analysis and was congruent with the Y mission.” The staff saw it as their
(22) to identify facilitators and barriers to the adoption of EF. A responsibility to keep seniors socially and mentally involved so
codebook was developed using a combination of a priori themes they were not isolated at home. Another staff member mentioned,
addressed in the interview guide and additional themes identified “Our strategic initiatives and our strategies roadmap for our asso-
through the initial review of interview transcripts. The team met ciation states very clearly that we will have a growing focus on
weekly for 3 months to discuss and come to agreement on coding expanding our senior membership and increasing our programing
rules. A dyad (BB and GK) double-coded a subset of five transcripts to meet the needs of the aging population.” The staff were com-
until agreement reached 85%. Remaining transcripts were divided mitted to providing programs that seniors enjoyed and needed.
between both team members (BB and GK) and coded indepen- The Ys’ values promote inclusiveness, which is operationalized by
dently. After the initial coding, we used a deductive approach (23) staff providing programing to improve the health and well-being
to review the coded text within the RE-AIM framework compo- of older adults.
nent of adoption (24). Descriptive statistics were calculated for
demographic items including staff age, duration of involvement Organizational support
with EF, and education level. Support from different organizational levels of the Y also served to
facilitate adoption. For example, staff were consulted when EF
RESULTS marketing materials were developed. This helped the staff feel
Study participants were, on average, 48.7 years old (SD 13.5) and like they had a say in how and to whom the program was being
had been involved with EF for 5.2 years (SD 3.1). Three staff (20%) promoted. Staff also felt that adoption of EF was facilitated with
had some college education, seven (47%) had a college degree, support from junior management, Wellness Directors, Executive
and five (33%) had more than a college degree. Ten staff mem- Directors, and CEOs.
bers (67%) were employed by a Y while the remaining five staff
(33%) were employed by other organizations (faith-based orga- Match with the target population
nization, senior center, social service organization, and residential Enhance®Fitness was more easily adopted when the staff perceived
facility). Titles and levels of responsibility varied: five were Health there was a good fit between the needs of the target population
and Wellness staff, four had roles specific to older adults, six were (older adults) and the program itself (EF). Older adults were
at the program coordinator/instructor level, and eight were at the considered to be in an age bracket that had previously been under-
program manager/director level or above. served by the Y. A perceived gap in programing led staff to look for
Staff interviewed represented 10 different YMCA associations exercise programs like EF because it was a valuable addition for
in six states. Fourteen out of 15 staff worked in locations where the senior community. Because EF exercises can be adapted to suit
classes were conducted; one staff worked in an administrative participant abilities, the staff felt it was a very inclusive program in
office. Ten sites currently offered EF classes; the remaining five which older adults at varying levels of function could participate.
had previously offered EF but did not have active classes at the In addition, some of the EF classes were offered off-site in set-
time of the interview. tings that catered to older adults and with which the Y branch had
In the remainder of this paper, we summarize our findings been partnering. One such setting was a retirement community.
regarding facilitators for adoption of EF and barriers that inter- One staff commented: “the residents [in assisted living] were good
fered with adoption of EF, with representative quotes presented in candidates because they were at a point where they had not been
Tables 2 and 3, respectively. exercising. We could start at the beginning and see where their
progress was which would not have been if we brought it into our
FACILITATORS FOR ADOPTION OF EF Y and tried to offer it to our regular seniors”.
Key facilitators identified by staff, which contributed to EF adop-
tion, were match with the Y mission, support from different orga- Financial support
nizational levels, match between the target population need and Initial financial support was an important factor in the adop-
EF, initial and on-going financial support, presence of champions, tion of EF. Several staff members reported receiving funding
novelty of EF, an invitation to partner with a community-based through grants at the time of EF adoption to cover EF training

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 164 | 292
Belza et al. Adoption of Enhance® Fitness

Table 2 | Facilitator themes for adoption of Enhance® Fitness (EF): perceptions of Y staff.

Theme Transcripts Exemplar quotations

represented

1 EF matches well 8 of 15 (53%) Adopting an evidence-based program fit well with the Y goals and standards and was congruent with the
with Y mission Y mission. The Y mission includes providing programing that helps to improve the health and fitness of
older adults (Age 30, 11 years with the Y).
It’s back to the spirit of mind and body of what the YMCA does (Age 33, 11 years with the Y).
We knew it was an evidence-based program, one that fit well with the Y goals and standards (Age 30,
11 years with the Y).
I think it falls into our focus areas i.e., healthy living for seniors and social responsibility as well. We’re
being responsible when we provide those types of exercise programs (Age 29, 5 years with the Y).

2 Organizational 5 of 15 (33%) The association office asked us if we would be interested. Because of the clientele we have, we have lots
support of seniors, of course I stepped up to the plate and said: “Yes, definitely let’s try this for our seniors.”
That’s how we got involved (Age 41, 14 years with the Y).
I think having a focus and support from junior management is important (Age 38, 8 years with the Y).
I think you need to have an Executive Director or CEO really understanding what it means to deliver
evidence-based programs (Age 38, 8 years with the Y).
I think it’s always good thing to bring something new in. It was driven by the Y of the USA. And then also I
was asked to do this by our Health and Wellness director of the main branch (Age 51, 15 years with the Y).

3 Match with the 10 of 15 (67%) We had been looking for an older adult program because we have a large aging population in our
target population community. It has been an age bracket that had been underserved at our Y (Age 30, 11 years with the Y).
We are known for the fact that we offer programing that is valuable to the community and to the seniors
in the community. EF is one of those programs (Age 63, 2 years with Y-affiliated site).
We are in close proximity to (low income housing) and so this is a very easy place for them to come. It’s
convenient for them. If we’re talking about people that are low income and don’t have money for public
transportation, it makes it very easy for them to do something to take care of their healthy living (Age 63,
2 years with Y-affiliated site).
It is completely appropriate for many health seekers and people who struggle with becoming more active
or staying active (Age 38, 8 years with the Y).

4 Financial support 5 of 15 (33%) When we started it, we started with the [state department of health]. . . . they gave us a grant basically
along with other YMCAs in [the state] with all of those being downstate. They basically paid for my staff’s
training and they sent us. I think that they also paid for all of our equipment. They were a huge, huge
partner in this and for us being able to start EF when we did (Age 33, 11 years with the Y).
The [state] contacted us and we’ve been working with them for some other programs. They offered to
help with the initial training, and that’s where we learned about the program (Age 30, 11 years with the Y).
We offer financial assistance. Based upon income I can give participants a certain percentage off the price
of the class. And then based upon some of the grants that we have been given, I can give them even a
higher percentage off. We do the best that we can to really make it happen for them. I don’t like saying no
to anybody (Age 29, 5 years with the Y).

5 Champions 10 of 15 (67%) I am a go-getter and if I hear something, I go after it because it is beneficial to our residents (Age 65,
6 years with the Y-affiliated site).
For me, personally, it was something else for me to offer to the seniors. I absolutely love working with the
active older adults (Age 41, 14 years with the Y).
The more and more I learned about it, the more I loved it. I didn’t really know of any other like
evidence-based programs for older adults. I really liked the pre and post-tests that they did. It just seemed
like a great program (Age 33, 11 years with the Y).
I told my boss about it and how I thought it would be beneficial. I told our members about it because I
wanted to get them on board and get them excited. I did anything I could when the (grant sponsor)
people came over. I did everything I could to promote our space (Age 60, 7 years with Y-affiliated site).

(Continued)

www.frontiersin.org April 2015 | Volume 2 | Article 164 | 293

Belza et al. Adoption of Enhance® Fitness

Table 2 | Continued

Theme Transcripts Exemplar quotations

represented

6 Novelty of EF 5 of 15 (33%) I thought it would be something different, you know? I thought it would be more different and something
that we could offer to our seniors (Age 41, 14 years with the Y).
It was just something new and exciting, evidence-based. It was everything we wanted (Age 51, 11 years
with the Y).
I think it’s always good thing to bring something new in (Age 51, 16 years with the Y).
I just wanted to have a varied program offering, and I thought this would fit. . .I wanted to keep the people
who come here happy with our center. I want to give them a variety of things, and so I don’t want
anything to be stagnant (Age 60, 7 years with Y-affiliated site).

7 Invitation to partner 8 of 15 (53%) I think that I thought it looked like a great program. Our partnership at the [state department of health]
with another was so strong. They really wanted to help the YMCAs start it (Age 33, 11 years with the Y).
organization to Sometimes the senior centers request us to do a program. That is kind of how it happened. It was just
offer EF really good timing when we started EF because they were requesting that we come and do some
different things. We thought it would be perfect and so it just kind of fell into place (Age 33, 11 years with
the Y).
I was working with a grant writer at [a university]. I was looking for something that we could get through a
grant. This is the something she came up with (Age 60, 7 years with Y-affiliated site).
Someone [YMCA staff member] called [my manager] and said: “We have this EF class and would you
want to be our pilot program?” And she said “Absolutely, yes! That’s how it all started” (Age 51, 11 years
with the Y).

8 Program-specific 10 of 15 (67%) It is evidence-based and has got solid backing. It has a proven track record and can meet the needs that
characteristics of EF are out there (Age 30, 11 years with the Y).
such as being I won’t touch anything that does not have data or an evidence-based curriculum, especially as related to
evidence-based and chronic disease management (Age 38, 8 years with the Y).
with name It is an incentive to bring people in when they know that you have a program that is known throughout the
recognition country. It’s a recognizable name. You are branded already (Age 63, 2 years with Y-affiliated site).

Table 3 | Barrier themes to adoption of Enhance® Fitness (EF): perceptions by Y staff.

Theme Transcripts Supporting quotations

represented

1 Competing 5 of 15 (33%) For group classes we have dance, water aerobics, step aerobics, spinning, and the range of movement class
programing from [another exercise program]. We have additional programs that are available at a cost, and those include
our nutritional services; the EF classes; swimming lessons; different sports programs, and then small group
training types of classes (Age 30, 11 years with the Y).
We acknowledge that space is an issue . . . They [wellness directors] see it as oh we already have [another
program], our program for active older adults. Why would we want to do this one? (Age 38, 8 years with the Y).
. . . We [offer EF] off-site. We are not in our own building anymore. It was to save on rent . . . The big room is
often taken up with children’s camps and things like that (Age 60, 7 years with Y-affiliated site).

2 Limited 7 of 15 (47%) The staff did not see the benefit or the value to their people (Age 45, 10 years).
resources and . . . And getting our health and wellness directors to understand and not condemn it, like “What’s in it for us?”
expertise (Age 38, 8 years with the Y).
. . . Where are we going to put it; who is the instructor going to be; who’s going to pay for this, or where are
the funds coming from (Age 51, 16 years with the Y).
I know the whole issue is that people don’t have time. There is a lack of staff. We have it here, too, and so I
know some of the issues (Age 65, 6 years with Y-affiliated site).

for their instructors, weights and other equipment, licensing fees, Champions
and/or instructor salaries. These funds also allowed Ys to offer the Champions for both EF and older adult programing facilitated
program at no charge to participants in some sites. adoption of EF. When asked to identify champions for EF, staff

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 164 | 294
Belza et al. Adoption of Enhance® Fitness

identified both paid staff and volunteers who fully embraced Competing programing and space limitations
EF, and passionately and frequently promoted the program both Staff noted one of the barriers to offering EF was that the Y Asso-
within and outside the Y. Staff champions advocated for and ciation was already offering a number of other programs for active
secured resources to launch the EF program. Champions described older adults, and the health and wellness staff did not see the need
themselves as “go-getters,” extolled the virtues of EF, and commu- or benefit to offering another one. Additionally, in some Y sites
nicated often with managers, staff, and site members about the staff reported there was a “space crunch” with rooms that would
benefits of EF. One staff champion expressed adopting the “we be appropriate for offering EF being allocated to other programs
will make it work” attitude when it came to rolling out EF for the such as camps for children. One staff member said: “we no longer
first time. Volunteer champions welcomed new comers, brought had the luxury of having two senior programs running because of
guests to class, and took on other tasks such as setting up fitness space limitations.” One staff that was trained to offer EF but never
check areas. “[Champions] just do it on their own. Nobody asked did said: “we were having a time and space crunch. It wasn’t any-
them to do that (in reference to setting up fitness check areas). thing wrong with the program per se but we’re not going to take
They just love the community that EF provides and obviously the away our already very strong programs and try something new.”
physical benefits. They want to capture anybody that comes into Off-site locations like retirement communities also had issues with
class and really helps them feel that same way.” finding adequate space.

Novelty Limited resources, expertise, and program costs

Staff were looking for new and exciting programs to offer older Another factor that interfered with the adoption of EF was the lack
adults. They viewed adopting EF as an opportunity to keep their of staff resources, both in terms of time availability and the need
programing fresh, and valued being an early adopter when the pro- to find instructors with appropriate skills for working with older
gram was just getting started: “Back then [when EF was adopted] adults. Staff noted that costs of the program were a potential bar-
EF was kind of an experiment. There were only a few sites in the rier to adoption: rent, materials, and instructor costs, on the one
country offering it, I believe, and so I thought that it would be nice hand, and affordability for participants, on the other, were taken
to be part of that group.” into account before deciding to adopt EF.

Invitation to partner DISCUSSION

The initial adoption of EF by Y-affiliated sites was often trig- In this study, we examined the facilitators and barriers to the adop-
gered by an invitation from an established community partner, tion on EF in early adopter Y-affiliated sites. The Health Promotion
providing motivation to adopt EF. The Y has close links with Research Center (HPRC) dissemination framework provides a
community partners and values their suggestions. Being part context for interpreting the findings from this study and informs
of an active community-based network that also provides ser- the translation of our findings to other community-delivered,
vices to older adults positions the Y to be on the cutting-edge evidence-based programs (26). The HPRC dissemination frame-
of learning when new programs are launched. Established rela- work incorporates the terminology of the RE-AIM framework
tionships with state departments of health and affiliations with such that the definition of adoption is consistent between the two
academic and philanthropic organizations were often key to adop- frameworks.
tion. These relationships afforded access to financial resources The HPRC dissemination framework (Figure 1) identifies
that provided initial start-up and on-going funding for EF. three main actors involved in the dissemination of an evidence-
One staff member noted: “the opportunity to work with and based program: researchers, disseminating organizations, and
partner with an outside agency to help address another por- user organizations. Researchers and disseminating organiza-
tion of our population definitely interested me.” There were tions partner to develop a dissemination approach that is suit-
also examples of invitations to partner with new organiza- able for the targeted user organizations. The approach is built
tions, such as assisted living communities that had never had on learnings about the user organization’s characteristics and
a program like EF. An invitation from a new partner “opened readiness for adoption and implementation, and is continu-
doors.” ally refined through the collaboration of the three main actors.
At the same time, all actors operate within a broader context
Program-specific facilitators that includes both modifiable and unmodifiable components,
There were program-specific facilitators that helped with adop- such as funding and partnerships, and economic conditions,
tion. Staff described EF as being unlike other programs they had respectively.
offered. Most frequently mentioned was EF being evidence-based, Study participants identified several facilitators and barriers
branded, and having name recognition. Staff reported that EF was, related to the characteristics of Y-affiliated sites, the user orga-
“. . . an easy sale as it was proven to improve things,” and “it has nizations in this study. Ys have made major strides in recent
solid backing.” years to become leaders in community-based health promotion
programs. It was readily apparent that there was a strong fit
BARRIERS TO THE ADOPTION OF ENHANCE® FITNESS between EF and the Y mission to offer older adults an envi-
A number of factors that interfered with the adoption of EF were ronment that promotes physical and emotional health, and the
noted such as competing senior programs and space limitations, availability of resources to ensure adequate programing. This rich
limited staff resources, and costs of the program. environment also included champions for both EF and older

www.frontiersin.org April 2015 | Volume 2 | Article 164 | 295

Belza et al. Adoption of Enhance® Fitness

FIGURE 1 | The HPRC dissemination framework [taken from Harris et al. (26)].

adult programs. Programs with this type of fit have an enhanced Associations committed to Activate America and built their capac-
ability to sustain themselves. On the other hand, our findings sug- ity to better align their programs, practices, and policies with the
gest a need for improved evaluation of organizational readiness needs of health seekers. This was the Y climate when EF was being
prior to the adoption of any evidence-based program. Structured brought into the organization by early adopters. The organiza-
and rigorous determination of organizational readiness would tion was primed for embracing evidence-based health promotion
help organizations avoid known barriers to successful imple- programs and working with a community of adults with chronic
mentation, such as competing programs, and limited space and conditions. At this time, chronic disease prevention programs were
resources. Organizational readiness for change is an important introduced as well, including LIVESTRONG at the YMCA, a can-
precursor to the successful implementation of health promotion cer survivorship program, and the YMCA’s Diabetes Prevention
programs (27). Program (DPP).
Our study also identified facilitators related to the modifiable Ackermann and colleagues tested offering DPP at Y sites (11,
context in which user organizations operate. Staff noted the sig- 28–30). Ys were able to increase the number of participants and
nificant role grant funding played in the initial adoption of the offer DPP at a lower cost compared to other community settings,
program, as well as the importance of building partnerships at the making Ys an ideal community partner. They noted a variety of
local and state level. factors contributing to the success of DPP offered at Ys: the Y is a
Finally, we would like to highlight the role of the disseminating nationwide, community-based organization reaching diverse U.S.
organization, Y-USA, the national office of the Y. Between 2007 communities; it has a successful history of adopting and imple-
and 2011, Y-USA significantly increased its national effort, Acti- menting health promotion program for all age groups; it takes a
vate America®, to engage Ys in organizational and community group-delivery approach, which uses minimal over-all personnel
change focused on supporting health seekers, those who struggle cost; and it has a national policy to accept all participants regard-
to adopt and maintain a healthy lifestyle. Nearly two-thirds of Y less of their ability to pay the membership fee. Y-USA and its

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 164 | 296
Belza et al. Adoption of Enhance® Fitness

associations and branches nationwide are well-positioned to suc- Based on the findings in this study, we propose practice,
cessfully adopt new evidence-based programs, and more broadly research, and policy recommendations for the adoption of
disseminate existing programs based on the program experience evidence-based health promotion programs by community orga-
noted above, and the organizational infrastructure already in nizations. They are summarized in Table 4, which also includes
place. the facilitator/barrier themes they address.
There were several limitations in this study. First, we used a Our practice recommendations focus on organizational readi-
convenience sample. Convenience samples can introduce response ness for adopting new programs. Assessing organizational readi-
bias, with those having positive experiences being more likely ness, formally or informally, identifying gaps in readiness, and
to participate than those with less favorable experience. How- addressing any gaps may improve adoption (11). Organizational
ever, staff who volunteered to participate in this study shared a readiness can be evaluated through the following activities: assess-
range of experiences, both positive and negative, related to the ing fit of the program with organizational mission; assessing
adoption of EF. Second, our sample size of 15 may have lim- overlap with other programing, and physical space and time
ited our ability to reach conceptual saturation with regard to the constraints; identifying potential partners and funding/revenue
research questions. After completing 15 interviews, the research models; identifying capable instructors and securing training and
team determined that additional interviews were unlikely to return technical assistance for staff and potential instructors; and assess-
substantively new information, and that we had reached concep- ing cultural and demographic needs of the target population.
tual saturation. Third, inherent in qualitative methodologies is the Addressing these aspects of readiness may provide opportuni-
potential issue of transferability. To minimize issues with transfer- ties to open dialog with partners and stakeholders, or develop
ability and assure our findings would have applicability in other supporting resources like a business plan to support successful
contexts, a partner from the Y (AHH) was actively involved in adoption.
all phases of our study and another Y staff member served on The practice recommendations outlined above are based on the
our Project Advisory Group. Fourth, some of the staff had started experience of early adopters. However, the adoption experience of
overseeing EF up to 8 years prior to the phone interviews, result- majority adopters and late adopters/laggards may face different
ing in potential recall bias with respect to the initial decisions facilitators and barriers (11). Future research is needed to better
and process of adopting EF. Last, Y staff who were the target understand the spectrum of adoption in all phases, and how early
of our interviews were in positions in which they had respon- adopters’ experience may influence later adopters. In addition,
sibilities for program management and also could adopt new successful adoption should lead to program implementation and,
initiatives like EF. When further exploring the concept of adop- ultimately, maintenance. Additional research on how adoption
tion, it would be valuable to also include senior leadership who strategies influence subsequent implementation and maintenance
are likely to be key decision makers and have an influence in of evidenced-based health promotion programs could contribute
whether a new program would be considered, paid for and/or to development of best practices for the translation of research
adopted. into practice.

Table 4 | Recommendations for practice, research, and policy in the adoption of evidence-based health promotion programs.

Facilitators addressed Barriers addressed

Recommendations for practice

Assess fit with the organizational mission Match with mission
Assess fit with other programing Novelty, match with target population Competing programing
Identify existing community partners and new potential partners Invitation to partner, financial support Limited resources and expertise
Identify capable staff and instructors Champions Limited resources and expertise
Identify training and technical assistance for staff and potential Limited resources and expertise
instructors
Assess cultural and demographic needs of the target population Match with target population, match
with mission
Assess physical space and time constraints Limited resources and expertise
Assess start-up and on-going costs and offsetting funding/revenue Financial support Limited resources and expertise
Recommendations for research
Explore adoption among majority and laggard adopters, and
compare to early adopters
Explore influence of adoption on implementation and maintenance
Recommendations for policy
Explore policy approaches to revenue development Financial support Limited resources and expertise

www.frontiersin.org April 2015 | Volume 2 | Article 164 | 297

Belza et al. Adoption of Enhance® Fitness

Finally, broad policy support for evidence-based programs may adults: the EnhanceFitness program. J Appl Gerontol (2006) 25(4):291–306.
create an environment more primed for successful adoption. Pol- doi:10.1177/0733464806290934
8. Centers for Medicare and Medicaid Services. Report to Congress: The Cen-
icy support may include establishing a revenue stream to offset
ters for Medicare & Medicaid Services’ Evaluation of Community-based Well-
program costs, an approach that has been seen in other pro- ness and Prevention Programs under Section 4202 (b) of the Affordable Care
grams (30, 31). Policy approaches to developing revenue streams Act. Washington, DC (2013). Available from: http://innovation.cms.gov/Files/
may be fruitful among evidence-based programs demonstrat- reports/CommunityWellnessRTC.pdf
ing health improvements and reductions in health care costs 9. Belza B, Snyder S, Thompson M, LoGerfo J. From research to practice:
EnhanceFitness, an innovative community-based senior exercise program. Top
(6, 8).
Geriatr Rehabil (2010) 26(4):299–309. doi:10.1097/TGR.0b013e3181fee69e
10. Wallace J, Buchner D, Grothaus L, Leveille S, LaCroix A, Wagner E. Implemen-
CONCLUSION tation and effectiveness of a community-based health promotion program for
While these recommendations are based on our study of the expe- older adults. J Gerontol A Biol Sci Med Sci (1998) 53a(4):M301–6.
rience of early adopters of EF at Y-affiliated sites, they are likely 11. Rogers EM. Diffusion of Innovations. 5th ed. New York: Simon and Schuster
(2003).
applicable to other evidence-based programs conducted in com-
12. Glasgow RE, Klesges LM, Dzewaltowski DA, Bull SS, Estabrooks P. The future
munity settings. Facilitators and barriers to adoption apply across of health behavior change research: what is needed to improve translation of
programs and settings (8). Furthermore, facilitators, barriers, and research into health promotion practice? Ann Behav Med (2004) 27(1):3–12.
recommendations address modifiable aspects of adoption that doi:10.1207/s15324796abm2701_2
may improve success, including support of the organizational mis- 13. Derananian CA, Desai P, Smith-Ray R, Seymour RB, Hughes SL. Perceived versus
actual factors associated with adoption and maintenance of an evidence-based
sion, available resources, and options for offsetting costs. Y-USA
physical activity program. Transl Behav Med (2012) 2(2):209–17. doi:10.1007/
has successfully adopted a variety of evidence-based programs, s13142-012-0131-x
and can serve as a model for other regionally and nationally net- 14. Virginia Tech. RE-AIM: A Model for the Planning, Evaluation, Reporting and
worked community organizations. Organizations looking to adopt Review of Translational Research and Practice. [Internet]. Blacksburg, VA (2014)
new programs may increase their likelihood of success by apply- [cited 2014 April 14]. Available from: http://re-aim.org/.
15. Brownson RC, Ballew P, Dieffenderfer B, Haire-Joshu D, Heath GW, Kreuter
ing the recommendations appropriate to their organization and MW, et al. Evidence-based interventions to promote physical activity: what con-
program. tributes to dissemination by state health departments. Am J Prev Med (2007)
33(1 Suppl):S66–73. doi:10.1016/j.amepre.2007.03.011
ACKNOWLEDGMENTS 16. Ballew P, Brownson RC, Haire-Joshu D, Heath GW, Kreuter MW. Dissemination
The authors wish to thank Susan Snyder and Meghan Thompson at of effective physical activity interventions: are we applying the evidence? Health
Educ Res (2010) 25(2):185–98. doi:10.1093/her/cyq003
Senior Services, Seattle, WA. This study was funded in part by the
17. Pagoto SL, Kantor L, Bodenlos JS, Gitkind M, Ma Y. Translating the diabetes
Centers for Disease Control and Prevention (CDC), Prevention prevention program into a hospital-based weight loss program. Health Psychol
Research Centers Program, through the University of Washing- (2008) 27(1 Suppl):S91–8. doi:10.1037/0278-6133.27.1.S91
ton Health Promotion Research Centers Cooperative Agreement 18. Gale BVP, Schaffer MA. Organizational readiness for evidence-based practice.
U48DP001911. The contents of this work are solely the responsi- J Nurs Adm (2009) 39(2):91–7. doi:10.1097/NNA.0b013e318195a48d
19. Kohlbacher F. The use of qualitative content analysis in case study research.
bility of the authors and do not necessarily represent the official
Forum Qual Soc Res (2006) 7(1):Art.21. Available from: http://nbn-resolving.
views of the CDC. de/urn:nbn:de:0114-fqs0601211
20. Gill P, Stewart K, Treasure E, Chadwick B. Methods of data collection in qual-
REFERENCES itative research: interviews and focus groups. Br Dent J (2008) 204(6):291–5.
1. Guyatt G, Cairns J, Churchill D, Cook D, Haynes B, Hirsh J, et al. Evidence-based doi:10.1038/bdj.2008.192
medicine: a new approach to teaching the practice of medicine. JAMA (1992) 21. Powers WR. Transcription Techniques for the Spoken Word. Lanham, MD:
268(17):2420–5. doi:10.1001/jama.1992.03490170092032 Altamira Press (2005).
2. Pope ND, Rollins L, Chaumba J, Risler E. Evidence-based practice knowledge 22. Graneheim UH, Lundman B. Qualitative content analysis in nursing research:
and utilization among social workers. J Evid Based Soc Work (2011) 8(4):349–68. concepts, procedures and measures to achieve trustworthiness. Nurse Educ Today
doi:10.1080/15433710903269149 (2004) 24(2):105–12. doi:10.1016/j.nedt.2003.10.001
3. United States Senate Democrats, Democratic Policy Committee. The patient pro- 23. Hsieh HF, Shannon SE. Three approaches to qualitative content analysis. Qual
tection and affordable care act detailed summary. Washington, DC (2010) [cited Health Res (2005) 15(9):1277–88. doi:10.1177/1049732305276687
2014 April 15]. Available from: http://www.dpc.senate.gov/healthreformbill/ 24. Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of health
healthbill04.pdf . promotion interventions: the RE-AIM framework. Am J Public Health (1999)
4. National Council on Aging. Using the Evidence Base to Promote Healthy 89(9):1322–7. doi:10.2105/AJPH.89.9.1322
Aging [Issue Brief]. Washington, DC (2006) [cited 2014 April 15]. 25. The YMCA. About Us. [Internet]. Chicago, IL: YMCA of the USA (2014) [cited
Available from: http://www.ncoa.org/news-ncoa-publications/publications/ 2014 April 14]. Available from: http://www.ymca.net/about-us.
issuebrief_1-r_usingeb.pdf . 26. Harris JR, Cheadle A, Hannon M, Forehand M, Lichiello P, Mahoney E, et al. A
5. National Council on Aging. Evidence-Based Health Promotion Programs for framework for disseminating evidence-based health promotion practices. Prev
Older Adults: Key Factors and Strategies Contributing to Program Sustainabil- Chronic Dis (2012) 9:110081. doi:10.5888/pcd9.110081
ity [Issue Brief]. Washington, DC (2012) [cited 2014 April 15]. Available from: 27. Weiner B. A theory of organizational readiness for change. Implement Sci (2009)
http://www.ncoa.org/improve-health/NCOA-Health-Promo-Issue-Brief.pdf . 4(67):1–9. doi:10.1186/1748-5908-4-67
6. Ackermann RT, Williams B, Nguyen HQ, Berke EM, Maciejewski ML, 28. Ackermann RT. Working with the YMCA to implement the diabetes prevention
LoGerfo JP. Healthcare cost differences with participation in a community- program. Am J Prev Med (2013) 44(4 Suppl 4):S352–6. doi:10.1016/j.amepre.
based group physical activity benefit for medicare managed care health plan 2012.12.010
members. J Am Geriatr Soc (2008) 56(8):1459–65. doi:10.1111/j.1532-5415. 29. Ackermann RT, Finch EA, Caffrey HM, Lipscomb ER, Hays LM, Saha C. Long-
2008.01804.x term effects of a community-based lifestyle intervention to prevent type 2
7. Belza B, Shumway-Cook A, Phelan E, Williams B, Snyder S, LoGerfo J. The diabetes: the DEPLOY extension pilot study. Chronic Illn (2011) 7(4):279–90.
effects of a community-based exercise program on function and health in older doi:10.1177/1742395311407532

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 164 | 298
Belza et al. Adoption of Enhance® Fitness

30. Ackermann RT, Marrero DG. Adapting the diabetes prevention program lifestyle many of the authors and/or Review Editors may have worked together previously in
intervention for delivery in the community: the YMCA model. Diabetes Educ some fashion. Review Editors were purposively selected based on their expertise with
(2007) 33(1):69. doi:10.1177/0145721706297743 evaluation and/or evidence-based programming for older adults. Review Editors were
31. Nguyen HQ, Ackermann RT, Maciejewski M, Berke E, Patrick M, Williams independent of named authors on any given article published in this volume.
B, et al. Managed-medicare health club benefit and reduced health care costs
among older adults. Prev Chronic Dis (2008) 5(1):A14. Received: 15 June 2014; accepted: 12 September 2014; published online: 27 April 2015.
Citation: Belza B, Petrescu-Prahova M, Kohn M, Miyawaki CE, Farren L, Kline G
Conflict of Interest Statement: The authors declare that the research was conducted and Heston A-H (2015) Adoption of evidence-based health promotion programs:
in the absence of any commercial or financial relationships that could be construed perspectives of early adopters of Enhance® Fitness in YMCA-affiliated sites. Front. Public
as a potential conflict of interest. Health 2:164. doi: 10.3389/fpubh.2014.00164
This article was submitted to Public Health Education and Promotion, a section of the
This paper is included in the Research Topic, “Evidence-Based Programming for Older journal Frontiers in Public Health.
Adults.” This Research Topic received partial funding from multiple government and Copyright © 2015 Belza, Petrescu-Prahova, Kohn, Miyawaki, Farren, Kline and Hes-
private organizations/agencies; however, the views, findings, and conclusions in these ton. This is an open-access article distributed under the terms of the Creative Commons
articles are those of the authors and do not necessarily represent the official position Attribution License (CC BY). The use, distribution or reproduction in other forums is
of these organizations/agencies. All papers published in the Research Topic received permitted, provided the original author(s) or licensor are credited and that the origi-
peer review from members of the Frontiers in Public Health (Public Health Education nal publication in this journal is cited, in accordance with accepted academic practice.
and Promotion section) panel of Review Editors. Because this Research Topic repre- No use, distribution or reproduction is permitted which does not comply with these
sents work closely associated with a nationwide evidence-based movement in the US, terms.

www.frontiersin.org April 2015 | Volume 2 | Article 164 | 299

 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00200

Development and evaluation of a fidelity instrument

for PEARLS
Laura Farren 1 , Mark Snowden 2 *, Lesley Steinman 1 and Maria Monroe-DeVita 3
1
Health Promotion Research Center, University of Washington, Seattle, WA, USA
2
Department of Psychiatry and Behavioral Sciences, Harborview Medical Center, University of Washington, Seattle, WA, USA
3
Division of Public Behavioral Health and Justice Policy, Department of Psychiatry and Behavioral Sciences, University of Washington, Seattle, WA, USA

Edited by: Purpose: This manuscript describes the development and the preliminary evaluation of a
Matthew Lee Smith, The University of
fidelity instrument for the Program for Encouraging Active and Rewarding Lives (PEARLS),
Georgia, USA
an evidence-based depression care management (DCM) program. The objective of the
Reviewed by:
Rosemary M. Caron, University of study was to find an effective, practical, multidimensional approach to measure fidelity
New Hampshire, USA of PEARLS programs to the original, research-driven PEARLS protocol in order to inform
Samuel D. Towne, University of South program implementation at various settings nationwide.
Carolina, USA
*Correspondence: Methods: We conducted key informant interviews with PEARLS stakeholders, and held
Mark Snowden, Harborview Medical focus groups with former PEARLS clients, to identify core program components. These
Center, 325 Ninth Avenue, Room components were then ranked using a Q-sort process, and incorporated into a brief instru-
PS-5020, Box 359911, Seattle, WA
98104, USA
ment. We tested the instrument at two time points with PEARLS counselors, other DCM
e-mail: [email protected] program counselors, and non-DCM program counselors (n = 56) in six states. Known-
groups method was used to compare findings from PEARLS programs, other DCM pro-
grams, and non-DCM programs. We asked supervisors of the counselors to complete the
fidelity instrument on behalf of their counselors to affirm the validity of the results. We
examined the association of PEARLS program fidelity with individual client outcomes.
Results: Program for Encouraging Active and Rewarding Lives providers reported the high-
est fidelity scores compared to DCM program providers and non-DCM program providers.
The sample size was too small to yield significant results on the comparison between
counselor experience and fidelity. Scores varied between PEARLS counselors and their
supervisors. PEARLS program fidelity was not significantly correlated with client outcomes,
suggesting that other implementation factors may have influenced the outcomes and/or
that the instrument needs refinement.
Conclusion: Our findings suggest that providers may be able to use the instrument to
assess PEARLS program fidelity in various settings across the country. However, more
rigorous research is needed to evaluate instrument effectiveness.
Keywords: fidelity, implementation, depression, evidence-based, older adults

INTRODUCTION service agencies are trained as PEARLS counselors and work with
The Program to Encourage Active and Rewarding Lives (PEARLS) participants, teaching them problem-solving treatment methods
was developed in the late 1990s by the University of Washing- in combination with behavioral activation techniques.
ton Health Promotion Research Center, in collaboration with In a randomized controlled trial (RCT), the PEARLS program
our local Area Agency on Aging and a network of senior cen- was shown to significantly improve depression treatment out-
ters. PEARLS is a depression care management (DCM) program comes for frail, socially isolated elders with minor depression and
using the Chronic Care model (1). PEARLS includes active screen- dysthymia when compared to the typical care that these clients
ing for depression, using a trained depression care manager, a received (2). Fifty-four percent of PEARLS clients showed at least a
team approach, stepped care, and built-in follow-up. Depression 50% decrease in the 20-item Hopkins Symptoms Checklist (HSCL-
care managers deliver brief, evidence-based interventions and pro- 20) (3) depression score from baseline to 6-months versus 8% of
vide education and self-management support, proactive outcome clients in usual care at the same time interval. Forty-four percent
measurement, and tracking using the nine-item Patient Health of PEARLS clients showed complete remission from depression
Questionnaire (PHQ-9). PEARLS is a participant-driven program, after 6-months versus 10% of clients in usual care. Compared to
aimed at teaching individuals effective skills to manage their lives usual care, PEARLS participants also had greater health-related
when they get overwhelmed. Case managers, nurses, social work- quality-of-life improvements in functional well-being (p = 0.001)
ers, and other front-line staff at community-based aging and social and emotional well-being (p = 0.048). Thirty-four percent of

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 200 | 300
Farren et al. PEARLS fidelity instrument

individuals in usual care reported any hospitalization during the items: we presented the list of possible items to the interview par-
first 6 months of treatment compared to only 22% of PEARLS ticipants on a spreadsheet, along with ranking chips (four 1s, nine
clients, suggesting potential health care cost savings (2). A sub- 2s, sixteen 3s, nine 4s, and four 5s). The interview participants
sequent RCT (4, 5) demonstrated the effectiveness of PEARLS in ranked the items in order of priority (“5” = most core elements of
treating adults of all ages with epilepsy and major depression. PEARLS to “1” = least core elements of PEARLS), and only used
Currently, PEARLS is offered to older adults and adults with the ranking chips allotted to create a normal distribution. We cal-
epilepsy in approximately 45 sites across 14 states. As it is dissemi- culated the numbers of 1s and 2s (“low” ranking) and 4s and 5s
nated nationally, and implemented in various community settings, (“high” ranking) for each of the items. We then subtracted the
fidelity becomes increasingly important. Implementation, as it number of high rankings from the number of low rankings to
is described in the RE-AIM framework, “refers to the extent to come up with a “high-low” score for each item. Those items that
which a program is delivered as intended” both at the site-level received a high-low score >0 were presented back to the interview
and individual level (6). Fidelity is the adherence of a scientifi- and focus group participants to confirm that no items were miss-
cally developed program to the original, research-based protocol, ing. The participants were asked to then identify any additional
and is necessary for implementing evidence-based programs (7). items. The research team then created multiple-choice questions
The literature around evidence-based program implementation from each of the items, with five anchor points and a scale of one
emphasizes the importance of maintaining program fidelity to to five. The interview participants also reviewed the anchor points,
ensure positive outcomes (8–10). Low program fidelity has been provided feedback, and we modified the anchor points based on
shown to negatively impact outcomes (8, 11). Similarly, high pro- that feedback. A copy of the resulting PEARLS Fidelity instrument
gram fidelity has been linked to more positive outcomes across is provided in Data Sheet 1 in Supplementary Material.
a range of evidence-based practices serving a variety of popula-
tions (12–14). Dissemination and implementation models such as PRELIMINARY EVALUATION
the Fixsen Implementation framework (10), the RE-AIM frame- We used known-groups method data analysis (18) to conduct a
work (15), and the dissemination framework for evidence-based preliminary evaluation of the reliability and validity of the PEARLS
health promotion practices (16) assume that program adaptation fidelity instrument. We compared PEARLS, other DCM programs,
is necessary, expected, and must inform program evolution. How- and non-DCM programs across six states (CA, FL, GA, IL, MD,
ever, measuring program adaptations, and how those adaptations WA). The DCM programs included IMPACT (19) and Healthy
relate to client outcomes, is an essential step in determining the IDEAS (20). IMPACT is a primary care-based DCM program
extent to which an evidence-based program may be modified, that has demonstrated effectiveness with diverse populations in
while continuing to remain evidence-based. a range of clinical settings. Healthy IDEAS is an evidence-based
DCM program that integrates depression awareness and man-
MATERIALS AND METHODS agement into existing case management services for older adults.
The study was completed in three phases. The first phase involved The non-DCM programs included other psychotherapy and case
developing a brief, multidimensional instrument for measuring management program models. We collected the following data:
PEARLS fidelity across sites. The second phase involved con- PEARLS counselors completed (1) the fidelity instrument at two
ducting a preliminary evaluation of that fidelity instrument. The time points, and (2) a survey about their clinical experience at
third phase evaluated the association of PEARLS program fidelity one time point. Clinical supervisors for the PEARLS programs
instrument scores with individual client outcomes. This study was completed the fidelity instrument on behalf of each counselor
approved by the University of Washington Institutional Review that they supervised at one time point. We gave the counselors
Board. and supervisors the option of completing the fidelity instru-
ment and clinical experience survey online or with a paper and
INSTRUMENT DEVELOPMENT pencil.
We used purposive convenience sampling to identify study partici- The preliminary analysis involved evaluating mean scores and
pants with experience in PEARLS program development, training, SDs at the site-level and the individual level using paired t -tests.
delivery, and receipt. First, we recruited interview participants. We compared the mean scores for PEARLS sites against the mean
The participants included program developers and researchers, scores for the other DCM programs and the non-DCM programs.
program administrators, and PEARLS counselors. Using qual- We also compared site-level mean scores by experience level. We
itative methods, we conducted 30–60 min interviews (MS, LS) compared individual-level scores by education level (up to 4 years
with these individuals to identify the core, programmatic com- of college and master’s degree), by counselor experience (<1 year,
ponents of PEARLS from their perspective. Next, we recruited 1–3 years, 4–7 years, and 8 or more years), and by PEARLS-specific
former PEARLS clients who had completed the program within experience (<1 year, 1–3 years, and 4 or more years). Instrument
the previous 12 months, and held focus groups with these individ- validity was assessed using sensitivity and specificity, and we used
uals to identify the core, programmatic components of PEARLS this information to calculate optimal cut-off scores. We also con-
from their perspective. We provided a $25 incentive to focus ducted ROC curve analysis, calculating weighted and unweighted
group participants. Incentives were not provided to interview areas under the curve (AUC) as a quantitative method for com-
participants. bining sensitivity and specificity into a single metric. Reliability
After the interview and focus group participants identified pro- was evaluated using inter-class coefficients (ICC) between the two
gram components, we used Q-sort ranking (17) to prioritize these survey administration time points.

www.frontiersin.org April 2015 | Volume 2 | Article 200 | 301

Farren et al. PEARLS fidelity instrument

ASSOCIATION WITH CLIENT OUTCOMES fidelity instrument to examine what adaptations are being made
For this study, we worked with eight community partners in four in PEARLS implementation.
states around the U.S. (CA, NY, WA, VT) to examine the relation-
ship between PEARLS program fidelity and PEARLS client out- RESULTS
comes. These PEARLS programs were based on aging, social ser- INSTRUMENT DEVELOPMENT
vices, and mental health agencies and represent diverse geographic Seventeen informants provided input on the initial development
settings (urban, rural, suburban), and racial/ethnic minority com- of the PEARLS fidelity instrument. Ten people participated in the
munities (including African-Americans, Filipinos, and other Asian interviews to identify key components: four program developers
immigrant communities). Many of these programs serve persons and researchers, three program administrators, and three PEARLS
with limited income (as indicated by their eligibility for Medic- counselors. Seven former PEARLS clients participated in focus
aid and other assistance programs with less than a high school groups to identify the core, programmatic components of PEARLS
or college education. Each PEARLS program included one to five from their perspective.
PEARLS counselors. The four WA PEARLS programs work with The interview and focus group participants identified 42 pro-
one clinical supervisor. The four PEARLS programs outside of gram components, including items related to training, supervision,
Seattle have their own clinical supervisor. Each PEARLS program treatment, and eligibility. Eighteen items received a high-low score
graduates up to 25 clients over a 6-month period. Agencies were during the Q-sort ranking process, suggesting higher priority for
selected based on their current implementation of PEARLS for inclusion in the fidelity instrument. After these 18 items were pre-
at least 1 year and their participation on regular PEARLS tech- sented back to the interview and focus group participants, two
nical assistance conference calls. We obtained memorandums of additional items were added, resulting in a total of 20 multiple-
understanding from each participating PEARLS program. choice items. Each item had five possible text anchor points (score
We assessed PEARLS program fidelity by administering the of “1” = least fidelity to “5” = highest fidelity), with the total pos-
PEARLS fidelity instrument to PEARLS counselors and clinical sible score ranging from 20 to 100. The interview participants
supervisors. PEARLS client outcomes were obtained from exist- reviewed the anchor points, and provided feedback. We modified
ing de-identified PEARLS client outcome data from clients of the anchor points based on that feedback.
participating PEARLS counselors. All PEARLS counselors at par- Questions are divided into two sections: Program Design and
ticipating PEARLS programs were invited to participate in this Program Delivery. The Design section includes questions about
study. Each counselor was asked to complete the fidelity instru- how the organization implements PEARLS, including training,
ment at two time points over the course of the study. In addition, clinical supervision, client recruitment and referrals, and eligibil-
each counselor provided basic information about her or his clini- ity criteria. The Delivery section focuses on how the counselor
cal experience and demographics. No identifiable information was implements PEARLS with their clients, such as the average num-
collected about the counselor. We also invited clinical supervisors ber of sessions that are delivered at home or that identify and
at each participating agency to complete a PEARLS fidelity instru- discuss social activities. A copy of the resulting PEARLS Fidelity
ment on each participating counselor. We linked the PEARLS clin- instrument is provided in Data Sheet 1 in Supplementary Material.
ical supervisor fidelity instrument data to the PEARLS counselor
data using a unique code assigned for study purposes only. Each PRELIMINARY EVALUATION
clinical supervisor also provided basic information about their We used known-groups method data analysis to compare 12
clinical experience and demographics. No identifiable information depression programs in six states: six PEARLS programs, four
was collected about the clinical supervisor. other DCM programs, and two non-DCM programs. Fifty-
We obtained de-identified depression outcome data from the two PEARLS counselors and seven clinical supervisors provided
PEARLS clients of each participating PEARLS counselor 6-months responses to both the fidelity instrument and the experience sur-
following the fidelity survey administration. The outcome data vey: 16 respondents from six PEARLS programs, 23 respondents
include baseline and final overall and item PHQ-9 depression from four DCM programs (one IMPACT program and three
scores, as had been done in our prior research (2, 4, 5) examin- Healthy IDEAS programs), and 20 respondents from two non-
ing outcomes from treating clients with major depression. We also DCM programs. One practitioner from a DCM site was excluded,
obtained data on client age, gender, race/ethnicity, and language due to missing data. Seventy-three percent of the participants com-
spoken. We worked with each participating PEARLS agency to pleted the fidelity instrument online. It took an average of 14 min
ensure that appropriate human subjects protections were in place to complete. Participants averaged 48 days between the first and
before obtaining client data. We analyzed the relationship between second time point for taking the survey.
PEARLS program fidelity total and item scores with the mean
change in PHQ-9 scores between participants’ baseline and final Individual level (PEARLS)
PEARLS session. We used Spearman’s rank correlation to measure The range of scores was 40–89 for the PEARLS counselors (n = 16).
the degree of association between the PEARLS program fidelity Five counselors attended up to 4 years of college and 10 counselors
scores and the mean change in PHQ-9. We also dichotomized held Master’s degrees. Six practiced as a counselor for <1 year,
the total PEARLS fidelity score and looked at whether falling five practiced for 1–3 years, and five practiced for over 8 years.
above or below the cut-off predicted significant differences in No counselors practiced from 4–7 years. Seven counselors imple-
the mean change in the PHQ-9, using independent t -tests to mented PEARLS for <1 year, five implemented PEARLS for 1–
evaluate this difference. Lastly, we summarized responses to the 3 years, and two implemented PEARLS for over 4 years. The sample

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 200 | 302
Farren et al. PEARLS fidelity instrument

size was too small to yield significant results on the comparison score of 70 (15) for PEARLS programs in the preliminary eval-
between education level, experience as a counselor, and experience uation. In contrast, the PEARLS counselor and PEARLS clinical
with implementing the PEARLS program. supervisor scores differed by an average of 12.17 points (8.09),
with some counselors reporting lower scores and others reporting
Site-level (PEARLS, DCM, non-DCM) higher scores than their clinical supervisor, even within the same
Program for Encouraging Active and Rewarding Lives sites agency.
reported the highest fidelity score [Mean (SD) 70 (15.5)] com- Program for Encouraging Active and Rewarding Lives partic-
pared to sites delivering other DCM programs [55.2 (19.1)] and ipant data were obtained for 127 persons with a mean age of
non-DCM programs [58.0 (13.0)] (p < 0.05). Average item scores 69 years (8.87). 38.2% identified as White, 26% identified as His-
were 3.56 (0.77) for PEARLS sites compared to 2.9 (0.8) for other panic (59% as Mexican), 21.1% as Asian (mainly Filipino with
DCM sites and non-DCM sites. PEARLS sites with more years some Vietnames and Korean participants), 11.4% as African-
of experience reported higher scores (mean 81, range 74–89) American, and as other races. Only 21% (26 clients) provided
than newer programs (mean 59, range 26–81). PEARLS supervi- data on their income, with all but three of these clients reporting
sors (n = 7) from three PEARLS programs completed the fidelity very low income as defined by the Federal Poverty Level, Median
instrument. Mean fidelity scores were comparable between the Income, or Housing and Urban Development criteria depending
supervisor and counselor for all three programs [83 (9) for super- on the PEARLS program. Almost half (46.7%) of the respondents
visors and 82 (7) for counselors, p = 0.87 (NS)]. Individual item that provided information on language spoken at home (n = 96)
scores were similar, with an average difference of 0.04 between reported speaking a language other than English. Half of those
items. Unweighted scoring yielded an AUC in ROC analyses of reporting who they lived with stated that they currently lived alone.
0.77. Weighting the overall score improved ROC scores yielding an We did not collect data on education. The mean change in PHQ-9
AUC of 0.81. Optimal cut-off scores for weighted PEARLS fidelity was 8.79 (5.50). There was little correlation between PEARLS pro-
score is 77, yielding a sensitivity of 77% and specificity of 67% gram fidelity and participant outcomes. The correlation between
for identifying PEARLS counselors and non-PEARLS counselors, the overall fidelity score and mean change in PHQ-9 was −0.069
respectively (Figure 1). Inter-rater reliability was satisfactory, with (p = 0.444). Several fidelity items were significantly correlated
an overall ICC of 0.77. with the mean change in the PHQ-9, but all suggested weak
correlations. The most strongly correlated items were those that
ASSOCIATION WITH CLIENT OUTCOMES involved the administration of the PHQ-9 (r = 0.231, p = 0.009),
Twenty-six PEARLS counselors and six clinical supervisors com- the use of problem-solving treatment (r = 0.227, p = 0.010), and
pleted the PEARLS fidelity instrument. The mean PEARLS fidelity the use of homework in between in-person sessions (r = 0.227,
score was 79.75 (8.33), which was similar to the average total p = 0.010).
We conducted additional analyses removing outliers-data for
those counselors and clinical supervisor pairings that had a differ-
ence in total fidelity score that was 13 or greater (higher than
the mean difference between counselors and clinical supervi-
sors of 12.17). Eighty-four PEARLS participants and 16 coun-
selors were included in this revised dataset. The mean change
in PHQ-9 was slightly higher than in the original group [9.00
(5.25)]. The mean (SD) total PEARLS fidelity score was also higher
[83.42 (5.29)].
We dichotomized the full dataset to look at whether falling
above or below a cut-off for the PEARLS fidelity total score pre-
dicted a difference in the mean change in PHQ-9 score. Using the
cut-off of 70 (as identified in the preliminary evaluation described
above), there was not a significant difference between mean change
in PHQ-9 between those falling above or at 70 [n = 106, 9.16
(5.42)] and those falling below 70 [n = 21, 6.90 (5.69), p = 0.086].
The difference was also not significant when the cut point was
set at 80 (the mean total PEARLS fidelity score in this correlation
study), with a mean PHQ-9 change of 8.94 (5.38) (n = 81) for
those at or above the cut point and 8.52 (5.78) (n = 46) for those
below the cut point.
We summarized the responses to the fidelity instrument in
Table 1. The table provides a snapshot of how PEARLS programs
are implementing PEARLS compared to the original research
model. Differences exist for clinical supervision, counselor assess-
FIGURE 1 | ROC curves for comparisons. ment, client eligibility, and the content and format of PEARLS
sessions.

www.frontiersin.org April 2015 | Volume 2 | Article 200 | 303

Farren et al. PEARLS fidelity instrument

Table 1 | Research to practice: a summary of responses to the PEARLS with regularly providing supervision that helps guide counselors
fidelity instrument as compared to the original PEARLS model. in adhering closely to the PEARLS model, while others are brought
in less frequently and advise more on co-occurring chronic con-
Original PEARLS model Practice model ditions and medication use than on the PEARLS model. More
rigorous research is needed to confirm the effectiveness of the
CLINICAL SUPERVISION
tool (e.g., by comparing the tool to the current gold standard of
Formal contract with 89% have formal supervision in place
an in-person, full-day, external evaluation of fidelity) before it is
supervisor
disseminated widely.
Bi-monthly supervision 40% meet at least monthly Research suggests that translation of evidence-based programs
Each client discussed at each Range from “as needed” (16%) to must be completed systematically, both at the site-level and indi-
session “weekly” (24%) vidual level, to assure effectiveness (6). The RE-AIM framework
COUNSELOR ASSESSMENT is an important tool that is used by researchers and public health
Audiotapes of PEARLS 56% assessed during formal clinical practitioners to inform that dissemination. The framework defines
sessions supervision; 28% during job supervision or translation across five areas – Reach, Effectiveness, Adoption,
self-assessment Implementation, and Maintenance (6). Assuring the fidelity of
ELIGIBILITY a particular program to its original evidence-based design is an
Home-based program 42% deliver PEARLS outside of the home important aspect of the implementation phase of dissemination
(7–14).
Older adults (60+) 40% include younger adults (<age 60 years)
It is important to evaluate PEARLS counselors at the site-level
PEARLS SESSIONS
to assure that clients are receiving the best possible treatment. It
6–8 in-person sessions 71% average ≥6 sessions per client
is also imperative to look at the PEARLS program at the site-level
PHQ-9 80% PHQ-9 at ≥6 sessions to prove to funding agencies that it is effective (6–14). This is par-
Education about depression 64% counselors use both written and ticularly important as many funding sources now require some
using both written and verbal verbal materials measure of quality assurance or fidelity for supporting evidence-
materials based programs (e.g., Administration on Aging Title III-D fund-
ing for evidence-based disease prevention programing). However,
Sessions focus on the 42% ≤2 sessions focus on the past
many sites lack adequate funding and staff capacity to conduct
present
in-depth, programmatic assessments; therefore, it is important to
Client chooses problems and 42% ≥6 sessions develop fidelity tools that are effective, user-friendly, and operate
solutions at low-cost.
Homework completed 17% ≥6 sessions, 75% ≥4 sessions The PEARLS mean score on the fidelity instrument was lower
than expected (70.5 out of a possible 100), and Table 1 illustrates
Behavioral activation 46–58% ≥6 sessions
some of the changes that agencies implementing PEARLS are mak-
Written PST worksheet 46% ≥6 sessions ing. From our work providing technical assistance to PEARLS sites,
we believe that a couple of factors may be at play. Many sites
have made programmatic changes to address the needs of their
DISCUSSION staff and the local population, or due to funding and staffing con-
The PEARLS fidelity instrument is a brief, easy-to-use, low-cost straints. For instance, while programs would like to meet weekly
option for PEARLS program staff to assess fidelity during pro- with their clinical supervisor, they may meet less frequently due
gram implementation. The tool takes an average of 14 min to to supervisor availability, limited funds to support the supervisor
complete, allowing for routine assessments by clinical supervi- time, and/or smaller PEARLS client caseloads, which make more
sors and counselors. For example, a PEARLS program may use frequent supervision unnecessary. Another example is the case
the fidelity instrument periodically as a way to assess whether a of programs delivering PEARLS outside the home. This shift has
counselor is continuing to maintain fidelity to the original pro- occurred driven by client preferences to meet at places where they
gram model, and to identify where adaptations have been made. are already congregating (e.g., after a nutrition program at a com-
Clinical supervisors may use the instrument to guide ongoing munity center) or preferences to meet elsewhere when a spouse or
clinical supervision sessions and activities. Funders of PEARLS caregiver is at home due to privacy concerns. Some of the changes
programs and agency administrators may be interested in the are to be expected given that the original PEARLS model was based
PEARLS fidelity instrument as a quality assurance tool to guide on the research protocol during an RCT. Elements such as having
ongoing implementation. the clinical supervisor review audiotapes of each PEARLS session
It is surprising that fidelity to the PEARLS program and client are not feasible for community-based agencies that do not have
outcomes are not more strongly correlated. Also, the discrepancy the funds, staffing, or resources. Nor is this level of supervision
between clinical supervisor and counselor ratings suggests that required for the program to be implemented successfully.
there is variation in how each party completes the instrument, and Another example is expanding participant eligibility criteria to
perhaps in how clinical supervisors are involved in the program. better align with the multifaceted, complex clients that agencies see
We know, for instance, from PEARLS technical assistance activ- every day. Previous research (21) suggested that the strict PEARLS
ities that some clinical supervisors are more intimately involved eligibility criteria was screening out more clients than screening

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 200 | 304
Farren et al. PEARLS fidelity instrument

them in, frustrating providers who repeatedly refer clients who are the responsibility of the authors and do not necessarily represent
ultimately not treated. During our work with PEARLS providers, the official views of CDC.
we have focused their assessment of client eligibility more on client
function, whether they are able to attend PEARLS sessions and to SUPPLEMENTARY MATERIAL
do the work during and in between the sessions. In addition, we The Supplementary Material for this article can be found
know that PEARLS programs are making adaptations to fit their online at http://www.frontiersin.org/Journal/10.3389/fpubh.2014.
local populations. Sites do not use written information with every 00200/abstract
participant as some are illiterate or have low literacy, or speak a
language other than English and materials are interpreted versus REFERENCES
translated. In addition, some sites allow younger older adult par- 1. Coleman K, Austin BT, Brach C, Wagner EH. Evidence on the chronic care
ticipants in the program (typically age 50–59 years) as they have model in the new millennium. Health Aff (2009) 28(1):75–85. doi:10.1377/
hlthaff.28.1.75
no other treatment options, are seen in similar settings, and are as 2. Ciechanowski P, Wagner E, Schmaling K, Schwartz S, Williams B, Diehr P, et al.
successful as the 60+ population. Community-integrated home-based depression treatment in older adults: a ran-
These findings also point to the fact that fidelity is only a domized controlled trial. JAMA (2004) 291(13):1569–77. doi:10.1001/jama.291.
piece of the implementation of PEARLS. The weak, generally non- 13.1569
significant correlations between fidelity score and client outcomes 3. Derogatis LR, Lipman RS, Rickels K, Uhlenhuth EH, Covi L. The Hopkins
symptom checklist (HSCL): a self-report symptom inventory. Behav Sci (1974)
may suggest that other factors are at play, which impact whether 19(1):1–15. doi:10.1002/bs.3830190102
a client improves their depressive symptoms. From our work pro- 4. Chaytor N, Ciechanowski P, Miller JW, Fraser R, Russo J, Unutzer J, et al.
viding technical assistance to PEARLS programs, we know that at Long-term outcomes from the PEARLS randomized trial for the treatment
the individual level, motivation, stigma about depression, physi- of depression in patients with epilepsy. Epilepsy Behav (2011) 20(3):545–9.
doi:10.1016/j.yebeh.2011.01.017
cal health and management of other comorbidities, informal and
5. Ciechanowski P, Chaytor N, Miller J, Fraser R, Russo J, Unutzer J, et al.
formal support, and mobility and function are all factors that influ- PEARLS depression treatment for individuals with epilepsy: a randomized
ence whether someone successfully participates in PEARLS. At the controlled trial. Epilepsy Behav (2010) 19(3):225–31. doi:10.1016/j.yebeh.2010.
site-level, other approaches for improving EBP implementation 06.003
include outcome monitoring, regular and structured supervision, 6. Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of health
promotion interventions: the RE-AIM framework. Am J Public Health (1999)
effective organization and climate, rigorous selection and reten-
89(9):1322–7. doi:10.2105/AJPH.89.9.1322
tion of team members, and ongoing consultation and technical 7. Teague GB, Mueser KT, Rapp CA. Advances in fidelity measurement for men-
assistance (21). tal health services research: four measures. Psychiatr Serv (2012) 63(8):765–71.
After measuring fidelity, it is important to then provide fidelity doi:10.1176/appi.ps.201100430
feedback to EBP providing agencies so that they can modify their 8. Derzon JH, Sale E, Springer JF, Brounstein P. Estimating intervention effec-
tiveness: synthetic projection of field evaluation results. J Prim Prev (2005)
services based on feedback from their fidelity reviews (22). One 26(4):321–43. doi:10.1007/s10935-005-5391-5
example of a systematic approach for providing fidelity feedback is 9. Durlak JA, DuPre EP. Implementation matters: a review of research on the
from the National Implementing Evidence-Based Practices Project influence of implementation on program outcomes and the factors affecting
(23). Most of the successfully implementing sites had altered their implementation. Am J Community Psychol (2008) 41(3–4):327–50. doi:10.1007/
services based on feedback received during their fidelity reviews, s10464-008-9165-0
10. Fixsen DL, Naoom SF, Blase KA, Friedman RM, Wallace F. Implementation
suggesting that the fidelity review process can be effective (22). Research: A Synthesis of the Literature. Tampa, FL: The National Implementation
With PEARLS, we have shared findings from each phase of this Research Network, Louis de la Parte Florida Mental Health Institute, University
fidelity study with participants on our monthly conference calls to of South Florida (2005). Contract No.: FMHI Publication #231.
provide technical assistance. 11. Botvin GJ, Baker E, Dusenbury L, Tortu S, Botvin EM. Preventing adolescent
drug abuse through a multimodal cognitive-behavioral approach: results of a 3-
The fidelity instrument that we developed in this study may
year study. J Consult Clin Psychol (1990) 58(4):437–46. doi:10.1037/0022-006X.
be the most effective way to evaluate effective implementation 58.4.437
in various locations and community settings across the country. 12. Burns T, Catty J, Dash M, Roberts C, Lockwood A, Marshall M. Use of intensive
While independent, observational measures of fidelity are ideal, case management to reduce time in hospital in people with severe mental ill-
this instrument provides a practical, user-friendly tool that pro- ness: systematic review and meta-regression. Br Med J (2007) 335(7615):336–40.
doi:10.1136/bmj.39251.599259.55
grams can use internally and at minimal cost for monitoring
13. McGrew JH, Bond GR, Dietzen L, Salyers M. Measuring the fidelity of imple-
program quality. Further work is necessary to ascertain the valid- mentation of a mental health program model. J Consult Clin Psychol (1994)
ity of the instrument given the discrepancy between counselor and 62(4):670–8. doi:10.1037/0022-006X.62.4.670
supervisor ratings and it may need refinement to correlate more 14. McHugo GJ, Drake RE, Teague GB, Xie H. Fidelity to assertive community treat-
strongly with client outcomes. ment and client outcomes in the New Hampshire dual disorders study. Psychiatr
Serv (1999) 50(6):818–24.
15. Glasgow RE, Lichtenstein E, Marcus AC. Why don’t we see more transla-
ACKNOWLEDGMENTS tion of health promotion research to practice? Rethinking the efficacy-to-
We thank study participants, research, and administrative staff. effectiveness transition. Am J Public Health (2003) 93(8):1261–7. doi:10.2105/
This study was supported in part by the CDC Prevention Research AJPH.93.8.1261
Centers (PRC) Program, through cooperative agreement U48 16. Harris JR, Cheadle A, Hannon PA, Forehand M, Lichiello P, Mahoney E, et al. A
framework for disseminating evidence-based health promotion practices. Prev
DP000050. Additional PEARLS research funding was provided by Chronic Dis (2012) 9:E22. doi:10.5888/pcd9.110081
the R-18 grant Improving Public Health Practice through Trans- 17. Brown SR. Q methodology and qualitative research. Qual Health Res (1996)
lation Research, #DP001150. The contents of this work are solely 6(4):561–7. doi:10.1177/104973239600600408

www.frontiersin.org April 2015 | Volume 2 | Article 200 | 305

Farren et al. PEARLS fidelity instrument

18. Mowbray CT, Holter MC, Teague GB, Bybee D. Fidelity criteria: development, This paper is included in the Research Topic, “Evidence-Based Programming for Older
measurement, and validation. Am J Eval (2003) 24(3):315–40. doi:10.1177/ Adults.” This Research Topic received partial funding from multiple government and
109821400302400303 private organizations/agencies; however, the views, findings, and conclusions in these
19. Unutzer J, Katon W, Callahan CM, Williams JW, Hunkeler E, Harpole L, articles are those of the authors and do not necessarily represent the official position
et al. Collaborative care management of late-life depression in the primary of these organizations/agencies. All papers published in the Research Topic received
care setting – a randomized controlled trial. JAMA (2002) 288(22):2836–45. peer review from members of the Frontiers in Public Health (Public Health Education
doi:10.1001/jama.288.22.2836 and Promotion section) panel of Review Editors. Because this Research Topic repre-
20. Quijano LM, Stanley MA, Petersen NJ, Casado BL, Steinberg EH, Cully JA, sents work closely associated with a nationwide evidence-based movement in the US,
et al. Healthy IDEAS: a depression intervention delivered by community-based many of the authors and/or Review Editors may have worked together previously in
case managers serving older adults. J Appl Gerontol (2007) 26(2):139–56. some fashion. Review Editors were purposively selected based on their expertise with
doi:10.1177/0733464807299354 evaluation and/or evidence-based programming for older adults. Review Editors were
21. Steinman L, Cristofalo M, Snowden M. Implementation of an evidence-based independent of named authors on any given article published in this volume.
depression care management program (PEARLS): perspectives from staff and
former clients. Prev Chronic Dis (2012) 9:E91. doi:10.5888/pcd9.110250 Received: 17 June 2014; accepted: 03 October 2014; published online: 27 April 2015.
22. Monroe-DeVita M, Morse G, Bond GR. Program fidelity and beyond: multiple Citation: Farren L, Snowden M, Steinman L and Monroe-DeVita M (2015) Develop-
strategies and criteria for ensuring quality of assertive community treatment. ment and evaluation of a fidelity instrument for PEARLS. Front. Public Health 2:200.
Psychiatr Serv (2012) 63(8):743–50. doi:10.1176/appi.ps.201100015 doi: 10.3389/fpubh.2014.00200
23. McHugo GJ, Drake RE, Whitley R, Bond GR, Campbell K, Rapp CA, This article was submitted to Public Health Education and Promotion, a section of the
et al. Fidelity outcomes in the National Implementing Evidence-Based journal Frontiers in Public Health.
Practices Project. Psychiatr Serv (2007) 58(10):1279–84. doi:10.1176/appi.ps. Copyright © 2015 Farren, Snowden, Steinman and Monroe-DeVita. This is an open-
58.10.1279 access article distributed under the terms of the Creative Commons Attribution License
(CC BY). The use, distribution or reproduction in other forums is permitted, provided
Conflict of Interest Statement: The authors declare that the research was conducted the original author(s) or licensor are credited and that the original publication in this
in the absence of any commercial or financial relationships that could be construed journal is cited, in accordance with accepted academic practice. No use, distribution or
as a potential conflict of interest. reproduction is permitted which does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 200 | 306
 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00143

Perceived utility of the RE-AIM framework for health

promotion/disease prevention initiatives for older adults: a
case study from the U.S. evidence-based disease
prevention initiative
Marcia G. Ory 1 *, Mary Altpeter 2 , Basia Belza 3 , Janet Helduser 4 , Chen Zhang 5 and Matthew Lee Smith 6
1
Department of Health Promotion and Community Health Sciences, School of Public Health, Texas A&M Health Science Center, College Station, TX, USA
2
Center for Health Promotion and Disease Prevention, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
3
Health Promotion Research Center, School of Nursing and School of Public Health, University of Washington, Seattle, WA, USA
4
Department of Health Policy and Management, School of Public Health, Texas A&M Health Science Center, College Station, TX, USA
5
Emory Global Health Institute, Emory University, Atlanta, GA, USA
6
Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA, USA

Edited by: Dissemination and implementation (D&I) frameworks are increasingly being promoted in
Sanjay P. Zodpey, Public Health
public health research. However, less is known about their uptake in the field, especially
Foundation of India, India
for diverse sets of programs. Limited questionnaires exist to assess the ways that frame-
Reviewed by:
Himanshu Negandhi, Public Health works can be utilized in program planning and evaluation. We present a case study from
Foundation of India, India the United States that describes the implementation of the RE-AIM framework by state
Katherine Henrietta Leith, University aging services providers and public health partners and a questionnaire that can be used to
of South Carolina, USA
assess the utility of such frameworks in practice. An online questionnaire was developed
*Correspondence:
to capture community perspectives about the utility of the RE-AIM framework. Distributed
Marcia G. Ory , Department of Health
Promotion and Community Health to project leads in 27 funded states in an evidence-based disease prevention initiative for
Sciences, School of Public Health, older adults, 40 key stakeholders responded representing a 100% state-participation rate
Texas A&M Health Science Center, among the 27 funded states. Findings suggest that there is perceived utility in using the
TAMU 1266, College Station, TX
RE-AIM framework when evaluating grand-scale initiatives for older adults. The RE-AIM
77843, USA
e-mail: [email protected] framework was seen as useful for planning, implementation, and evaluation with rele-
vance for evaluators, providers, community leaders, and policy makers. Yet, the uptake
was not universal, and some respondents reported difficulties in use, especially adopt-
ing the framework as a whole. This questionnaire can serve as the basis to assess ways
the RE-AIM framework can be utilized by practitioners in state-wide D&I efforts. Maxi-
mal benefit can be derived from examining the assessment of RE-AIM-related knowledge
and confidence as part of a continual quality assurance process. We recommend such
an assessment be performed before the implementation of new funding initiatives and
throughout their course to assess RE-AIM uptake and to identify areas for technical
assistance.
Keywords: RE-AIM, program planning, program implementation, program evaluation, older adults, aging

INTRODUCTION assessing programmatic implementation, especially in multi-site

With concerns about the aging population and attendant growth intervention initiatives (11).
of multiple co-morbidities (1, 2) support has grown for national Originally conceived in the late 1990s, the RE-AIM framework
initiatives to improve the health, function, and quality of life of (12) was designed to assess the public health impact of health pro-
older adults (3, 4). Despite the growing evidence base about the motion interventions through the identification of five core evalu-
nature of public health problems among older adults and suc- ation elements (i.e., reach, efficacy/effectiveness, adoption, imple-
cessful intervention approaches for improving their health and mentation, and maintenance). In an attempt to understand better
well-being (5–7), there remains a notable gap in transferring the translation of interventions tested within controlled trials to
what we know works into practice (8, 9). Many reasons can be implementation within community settings (13), RE-AIM has
cited for the existence of a research-to-practice gap including that changed the research paradigm from one focused exclusively on
researchers are not aware of the realities of programmatic imple- controlled clinical trials with a priority on internal validity to one
mentation in real world settings and community providers lack that acknowledges the importance of pragmatic interventions that
the guidance for implementing proven programs tested in other give salience to external validity – or the degree to which interven-
settings (10). There is also a lack of quality questionnaires for tion results can be generalized across interventions, populations,

www.frontiersin.org April 2015 | Volume 2 | Article 143 | 307

Ory et al. RE-AIM framework for health promotion/disease prevention

and settings (14–18). The use of the RE-AIM framework has been effectiveness, adoption, implementation, and maintenance (24).
refined since its conception to include guidance for the planning, Each component addresses a major research question that can
implementation, maintenance, and evaluation of programs and guide program planning and evaluation.
policies by clinicians, community providers, and policy makers “Reach” is the extent to which a program attracts and retains
(19). Its utilization is appropriate for those in the fields of aging the target audience. Measures of Reach include the number, pro-
services and public health, as well as allied disciplines. portion, and representativeness of participants. It is important to
Building on early community-wide efforts to identify best prac- monitor Reach to determine if the desired audience is participating
tice programs for older adults through the aging services network, in the program, in what numbers, and whether there is program
the United States Administration on Aging (AoA), a program divi- completion or attrition. This in turn, can help gage the success of
sion within the Administration for Community Living (ACL), marketing, recruitment, and retention efforts.
has dedicated resources to the implementation and dissemina- “Effectiveness” refers to assessing the change in short- and/or
tion of state-wide evidence-based practices (20). This emphasis long-term program outcomes, such as health behaviors and
on evidence-based practices reflects the emergence of several well- lifestyles, symptom management, health status, or health care uti-
tested health promotion/disease prevention programs, which have lization outcomes. Effectiveness indicators also monitor for other
been shown to not only make a difference in older adults’ health outcomes, whether negative or unintended that result from the
but also in reduced health care utilization (21). program. It is important to monitor Effectiveness to provide the
In 2006, the Atlantic Philanthropies and the AoA funded the evidence as to whether the program is producing positive changes,
evidence-based disease prevention (EBDP) initiative with the which ultimately makes the case for the program’s value and return
intention of supporting stronger linkages between State Aging Ser- on investment.
vices and State Health Departments to address the health needs of “Adoption” activities assess organizational capacity and part-
the growing population of older adults. The overall goals of this nership support. Measures include the number, proportion, and
initiative were to (1) develop the systems necessary to support representativeness of staff and settings who adopt a program as
the ongoing implementation and sustainability of evidence-based well as tracking of the various ways partners contribute to pro-
programs for older adults; (2) develop multi-sector community gram delivery. It is important to know if the supply of delivery
partnerships to enhance program accessibility and extend pro- staff and sites matches program demand and is located in areas
gram capacity; (3) reach the maximum number of at-risk older where the target audience resides and whether there is capacity to
adults who could benefit from the programs; and (4) deliver bring the program to scale.
evidence-based programs with fidelity (22). “Implementation” is the extent to which the program is deliv-
Seen as an opportunity for fostering learning collaborative, ered consistently, as intended by the program developers, across all
the funders contracted for technical assistance to the 27 state implementation sites by all instructors. Implementation measures
grantees funded under the EBDP initiative. Since this was the first also tracks program costs. It is important to monitor Implementa-
time RE-AIM was integral to health promotion program imple- tion in order to identify areas of need for improvement in program
mentation activities for these partnerships, there was interest in delivery, assure participant results can be attributed to the program
exploring how well and in what ways the framework was being and identify return on investment for stakeholders.
adopted and applied, especially since no systematic collection of At the setting level, “Maintenance” refers to the extent to which
this information existed. As investigators from three CDC Preven- the program can be embedded within the routine organizational
tion Research Center–Healthy Aging Research Network (HAN) practice. Some factors, such as “ongoing staff support,” “partner-
campuses charged with providing technical assistance to the fun- ship with community,” “sufficient funding,” and “health market-
der and State grantees, we wanted to explore how translational ing,” are all essential elements for organizational maintenance. At
research frameworks were being implemented in the real world the individual level, “Maintenance” refers to the extent to which
settings by state-level aging services providers and their public individual participants experience long-term benefits (longer than
health partners. This paper expands upon previously reported 6-months following program completion) and better quality of life
findings (23). Its purposes are to (1) introduce the reader to the from the health promotion interventions or policies. Attention to
RE-AIM framework; (2) describe the development of a question- these elements helps inform strategies to ensure individual bene-
naire to assess the implementation processes in the field based fits are sustained over time and that the necessary infrastructure is
on elements from the RE-AIM framework; (3) using this ques- in place to ensure a program will receive ongoing institutional or
tionnaire, examine ways RE-AIM was viewed by grantees and community support.
used in their program planning, implementation, and evalua-
tion of evidence-based programs; and (4) summarize implications PROCEDURES
for future use of RE-AIM and training needs in the evaluation Data were collected using internet-delivered methodology. The
of community-based dissemination and implementation (D&I) questionnaire utilized to collect data from respondents was devel-
efforts of evidence-based programs. oped by the HAN project team using online survey software. Elec-
tronic mail-based invitations to participate in the questionnaire
MATERIALS AND METHODS were sent in January 2009 to designated project leads representing
DEFINITIONS OF RE-AIM ELEMENTS 27 states receiving funding from and participating in the EBDP
As illustrated in Figure 1, the acronym RE-AIM represents the initiative. The instructions requested that the questionnaire be
five essential components of the RE-AIM framework: reach, completed separately by one state lead (either public health or

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 143 | 308
Ory et al. RE-AIM framework for health promotion/disease prevention

REACH
How do I reach
those who need
this intervention?

MAINTENANCE
How do I EFFECTIVNESS
incorporate the How do I know my
intervention so it is intervention is
delivered over the working?
long-term?

RE-AIM

ADOPTION
IMPLEMENTATION
How do I develop
How do I ensure
organizational
the intervention is
support to deliver
delivered properly?
my intervention?

FIGURE 1 | RE-AIM elements: planning and evaluating questions (see www.re-aim.org for more information).

aging) and one state-level program evaluator. Other team mem- was designed to collect information about the respondent’s knowl-
bers who played key roles in program implementation and/or edge, attitudes, and current practices related to different aspects
evaluation (e.g., a local project coordinator and/or regional coor- of the RE-AIM framework as a whole as well as attention to its
dinator or university partner) were also welcome to complete the individual components. The questionnaire was pilot tested for
questionnaire. Some of the items (e.g., knowledge and confidence ease of understanding and face validity with local community
in applying the RE-AIM framework) were asked retrospectively. practitioners.
After completing the questionnaire, the respondents were invited The final questionnaire contained 47 multi-part items includ-
to share their responses with their state team as a way of enhancing ing close-ended and open-ended items, as well as checklists. Rec-
their planning and evaluation efforts. The initial survey requested ognizing the importance of “survey fatigue” or attrition, the HAN
that responses be returned within 2 weeks. Two follow-up emails project team was careful not to make the questionnaire too long.
were sent to state respondents to increase the survey response rate. Therefore, close-ended items with Likert-type scaling were used
This study received Institutional Review Board (IRB) approval at to make it easy for respondents to respond to questionnaire items.
Texas A&M Health Science Center where data were collected and Additionally, open-ended items were integrated into the question-
analyzed. naire to allow for additional responses to give richer detail and
context to close-ended items.
QUESTIONNAIRE AND MEASURES It was estimated that the online survey would take approxi-
Reflecting expertise in several health professions (public health, mately 10–20 minutes to complete. Individualized links were sent
nursing, and social work) and prior experience with the RE-AIM through the online survey website to state leads that were identi-
model and implementation research (25–29), the authors designed fied through the AoA’s Technical Assistance Center. Respondents
the questionnaire to address how state grantees integrated RE-AIM had unlimited access to the online questionnaire to enable them to
elements into different planning, implementation, evaluation, and complete the task at their convenience and as a means of increasing
monitoring processes (a copy of the full questionnaire is appended completion rates. The questionnaire opened with a brief defini-
to the end of this article). tion of the RE-AIM elements, with directions to the respondents
As there were no comparable questionnaires in the litera- to go to the RE-AIM website (www.re-aim.org) if they desired
ture, the authors built the questionnaire around concepts deemed more information about the rationale for and measurement of
important to reflect implementation processes. The questionnaire each element.

www.frontiersin.org April 2015 | Volume 2 | Article 143 | 309

Ory et al. RE-AIM framework for health promotion/disease prevention

RE-AIM utilization elements. Respondents were asked to respond to seven statements

Respondents were asked to rate the degree to which the RE-AIM about the RE-AIM framework as a whole as well as its component
framework was used for planning, implementation/evaluation, elements.
and maintenance. A series of 15 items were used to assess aspects of
utilization. For example, for planning, respondents were asked to Respondent characteristics
respond to how they used RE-AIM to“select community partners,” Items were included to collect information about the respon-
“select host and/or implementation sites,” and “select assess- dents’ role on the AoA/Atlantic EBDP grant (i.e., state lead, state
ment/evaluation tools.” For implementation/evaluation, respon- evaluator, regional project coordinator, local project coordinator,
dents were asked to rate the framework use for “plan or alter and other); the year that the respondent started working with
participant recruitment,” “conduct mid-course evaluations,” and evidence-based programs (i.e., from 2000 to 2008); and the type
“present/publicize program findings.” For maintenance, respon- of evidence-based programs being delivered (from a list of 16
dents rated the framework use for “secure funding for maintain- approved evidence-based programs).
ing program delivery,” “build infrastructure to maintain program
staffing,” and “build capacity for ongoing quality assurance (QA).” RESULTS
UTILITY OF AN ONLINE SURVEY FOR COLLECTING INFORMATION IN A
Self-rated knowledge MULTI-STATE INITIATIVE WITH MULTIPLE STAKEHOLDERS AND
Respondents were asked to rate their knowledge about “EBDP PROGRAM TYPES
programs” and “the RE-AIM framework” at the start of the grant As previously reported (23), 40 questionnaires were submitted
initiative (retrospectively) versus the current time. If respondents electronically representing a 100% state-participation rate among
were not present at the initial stages of program implementation, the 27 funded grantee states. Almost half (48.2%) of the states had
they were instructed to mark the “not relevant” category. two respondents. Approximately one-third of the states (37.0%)
reported not having a state-wide evaluator. State leads and state-
RE-AIM-related confidence
wide coordinators represented the majority of respondents (65%);
Self-efficacy refers to individuals’ beliefs in their ability to succeed
state-wide evaluators represented 30% of the respondents; and
in a given situation (30). These beliefs act as determinants of how
regional or local coordinators represented the remaining 5% of
individuals think, behave, and feel (31). Individuals’ sense of self-
the respondents.
efficacy determines how goals, tasks, and challenges are addressed.
In terms of when they started working with EBDP programs for
Individuals with a strong sense of self-efficacy view challenging
older adults, less than half of the respondents reported that they
problems as tasks to be mastered; develop stronger interest in the
had worked with evidence-based programs before the onset of the
activities in which they participate; and are more committed to
current initiative. Of the 16 approved evidence-based programs,
their interests. (30) We were interested in learning about grantees
15 programs were offered across the grantee states. The most
confidence in the use and application of RE-AIM and whether
commonly offered programs by grantee states included Chronic
their confidence levels changed over the course of the grant. “Con-
Disease Self-Management Program (CDSMP) (100%), Enhance-
fidence” is the term Bandura uses as synonymous to self-efficacy
Fitness (37.5%), A Matter of Balance (30.0%), and Healthy IDEAS
when measuring the construct. Respondents were asked questions
(10.0%). There were no reported problems with understanding or
to measure their confidence about applying each of the five RE-
answering any questionnaire items.
AIM at the start of the grant initiative (retrospectively) versus the
current time. Again, if respondents were not present at the initial
APPLICATION OF RE-AIM FOR PLANNING,
stages of program implementation, they were instructed to mark
IMPLEMENTATION/EVALUATION, AND MONITORING
the “not relevant” category.
Table 1 reports the extent to which respondent’s decisions
Perceptions of RE-AIM usefulness about this initiative were influenced by the RE-AIM framework
Respondents were asked to share their attitudes about the appli- in terms of planning, implementation/evaluation, and main-
cation of RE-AIM for various tasks related to their grant efforts. tenance. With respect to planning, the largest proportion of
Respondents were asked to rate the usefulness of RE-AIM applied respondents reported RE-AIM influenced their decisions about
to the following activities: “planning of this initiative,”“implemen- selecting evidence-based programs to deliver, identifying tar-
tation of this effort,” “evaluation of this effort,” “planning efforts get populations, and selecting assessment/evaluation tools. With
with our other aging programs,” and “implementation efforts with respect to implementation/evaluation, about 58% of respon-
our other aging programs.” Respondents were also asked to report dents reported RE-AIM influenced decisions about planning
how valuable they believed RE-AIM was for different audiences. or altering participant recruitment. A majority of respondents
Participants were asked to respond to the following audiences: reported RE-AIM moderately influenced decisions when con-
“providers,” “community leaders,” “policy makers,” and “evalua- ducting mid-course evaluations and structuring reports. With
tors.” Finally, respondents were asked to indicate if they would respect to maintenance, a majority of respondents reported
apply RE-AIM in their future projects. RE-AIM influenced decisions about planning for program
sustainability. A majority of respondents reported RE-AIM
Ease of RE-AIM use and application moderately influenced decisions about maintenance strategies
Respondents were asked to report how easy they believed RE-AIM related to participant improvement, securing funding, and ongo-
was to use/apply and their preferences about monitoring RE-AIM ing QA.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 143 | 310
Ory et al. RE-AIM framework for health promotion/disease prevention

Table 1 | Ways in which RE-AIM was used for planning, implementation/evaluation, and maintenance (n = 40).

Not at A little Some A lot Do not

all (%) (%) (%) (%) know (%)

Planning
Select community partners 10.5 28.9 39.5 7.9 13.2
Select evidence-based programs for implementation 17.5 17.5 30.0 20.0 12.8
Select host and/or implementation sites 7.7 25.6 35.9 15.4 15.4
Identify target populations (people who may participate in programs) 12.5 22.5 30.0 22.5 15.0
Select assessment/evaluation tools 13.2 21.1 32.5 20.2 12.5

Implementation Evaluation
Plan or alter participant recruitment 10.5 21.1 39.5 18.4 10.5
Structure agendas and/or team meetings 17.5 25.0 25.0 20.0 12.5
Conduct mid-course evaluations 10.0 25.0 30.0 22.5 12.5
Structure reports 15.0 30.0 27.5 20.0 7.5
Present/publicize program findings 12.5 22.5 22.5 27.5 15.0

Maintenance
Address strategies for maintaining participant improvement 10.3 30.8 25.6 15.8 18.4
Guide discussions and/or planning around program sustainability 10.0 20.0 30.0 30.0 10.0
Secure funding for maintaining program delivery 15.8 31.6 27.5 15.0 12.5
Building infrastructure to maintain program staffing 12.5 27.5 22.5 22.5 15.0
Build capacity for ongoing quality assurance 5.1 33.3 28.2 25.6 7.7

KNOWLEDGE AND CONFIDENCE WITH EBDP AND RE-AIM ELEMENTS Table 2 | Knowledge about and confidence applying RE-AIM elements
OVER TIME at the start of the intervention versus the current time (n = 40).
Table 2 reports respondents’ knowledge about EBDP and RE-AIM,
as well as confidence applying RE-AIM elements at the start of the At start Currently Improvement (%)
initiative versus the time in which they completed this study. On
Knowledge about evidence- 2.73 3.92 43.6
average from the start of the initiative to the time of the ques-
based disease prevention
tionnaire (approximately 2 years), fewer than half of respondents
programs
increased their knowledge about EBDP programs, yet, over two-
thirds increased their knowledge about the RE-AIM framework. In Knowledge about RE-AIM 1.98 3.33 68.2
terms of confidence applying elements of the RE-AIM framework, framework as a whole
the largest increase was reported for applying reach, adoption, Confidence applying the RE-AIM elements
and implementation, which was followed by maintenance and Reach 2.13 3.43 61.0
effectiveness. Effectiveness 2.13 3.13 46.9
Adoption 2.08 3.35 61.1
PERCEPTIONS OF RE-AIM USEFULNESS FOR VARIOUS TASKS AND
Implementation 2.10 3.38 61.0
AUDIENCES
Maintenance 2.05 3.26 59.0
Table 3 reports respondents’ attitudes about the usefulness of the
RE-AIM application for various tasks and audiences. The vast Items scored from, not at all (1) to a lot (4).
majority agreed the framework was useful for planning, for imple-
mentation, and for evaluation. When asked about the application
of RE-AIM in other aging programs, the majority also agreed RE-AIM elements. Approximately three-quarters of respondents
that the framework was useful for planning and for implementa- agreed that it was easy to understand the RE-AIM elements. Fur-
tion. Further, when asked about audiences for which the RE-AIM ther, only a small minority believed that RE-AIM was too academic
framework is most useful, the majority of respondents agreed and took too much time to implement. However, nearly half of
RE-AIM was useful for evaluators, providers, community leaders, the respondents felt special expertise was required to monitor RE-
and policy makers. AIM requirements and approximately one-third felt the successful
application of RE-AIM elements was difficult to measure. When
PERCEPTIONS OF EASE OF USING THE RE-AIM FRAMEWORK AND asked about monitoring RE-AIM elements, over half believed it
MONITORING RE-AIM ELEMENTS was best to track all of the elements, whereas a sizable proportion
Table 4 reports respondents’ perceptions about the ease of using of respondents (over one-third) believed looking at one or two
and applying RE-AIM and their preferences about monitoring elements was most useful.

www.frontiersin.org April 2015 | Volume 2 | Article 143 | 311

Ory et al. RE-AIM framework for health promotion/disease prevention

Table 3 | Perceptions of RE-AIM usefulness for various tasks and RE-AIM elements in their grant planning, implementation, and
audiences (n = 40). maintenance activities. Thus, this study adds to our understand-
ing of the general use of RE-AIM for different grant tasks, and how
Disagree or Agree or Do not the application and usefulness varies by specific users.
strongly strongly know In contrast to previous research that documents a primary focus
disagree (%) agree (%) (%) on reach and effectiveness and excludes attention to maintenance
(33), in this initiative the RE-AIM framework was used by state
Tasks
agencies for building infrastructure or capacity for ongoing QA
Planning in this initiative 5.0 90.0 5.0
and sustainability. In retrospect, this is not surprising given the
Implementation of this initiative 2.5 90.0 7.5
salience of sustainability to this initiative and targeted technical
Evaluation of this initiative 2.5 84.7 2.6
assistance from the funder and outside consultants in this area.
Planning efforts with other aging 5.0 85.0 10.0 An important issue addressed in this research was the extent to
programs which RE-AIM elements were seen as an indivisible whole versus
Implementation efforts with other 2.5 87.5 10.0 the sum of individual parts. As indicated in the Section “Results,”
aging programs only slightly more than half of the respondents endorsed the use-
Audiences fulness of tracking all of RE-AIM the elements together, while
Providers 2.5 77.5 20.0 nearly a quarter did not express an opinion. It is not known if this
Community leaders 2.5 77.5 20.0 reflects an inclination for adopting single elements over the frame-
Policy makers 5.0 72.5 22.5 work as a whole, or a lack of experience with the framework, or a
Evaluators 0.0 92.5 7.5 lack of resources to fully assess and track all of the framework com-
ponents simultaneously. Additional research is needed to identify
which RE-AIM components different types of program imple-
Table 4 | Perceived ease to use and apply RE-AIM and preferences menters will find most useful and what resources are warranted.
about monitoring RE-AIM elements (n = 40). While there was strong endorsement of the usefulness of RE-
AIM for applying various tasks, the framework was seen as most
Disagree or Agree or Do not useful for evaluators versus providers, community leaders, or pol-
strongly strongly know icy makers. This may reflect the original origins of RE-AIM as
disagree (%) agree (%) (%) an evaluation tool for public health research (12), or the fact that
about half of the respondents still felt monitoring RE-AIM ele-
The different RE-AIM elements are 10.3 74.6 5.1
ments required special expertise. Alternatively, it may be that the
easy to understand
respondents who were evaluators in this study had more public
Monitoring RE-AIM elements 43.6 48.7 7.7 health training. These findings point to the importance of commu-
requires special expertise nity providers partnering with academics, with each being aware
RE-AIM is too academic 75.0 10.0 15.0
of the language and context of the other party (34). Such part-
nering has become even more critical with the increased push for
RE-AIM takes too much time to 65.0 15.0 20.0 demonstrated outcomes, continuous quality improvement (CQI)
implement of delivery agencies, and selected funding opportunities requiring
Measuring the successful 40.0 32.5 27.5 these partnerships. In community settings, it is especially impor-
application of different RE-AIM tant to identify and implement pragmatic measures and evaluation
elements is difficult designs (14).
Consistent with the growth of literature about RE-AIM (19),
Looking at just one or two RE-AIM 52.5 35.0 12.5
large increases in knowledge about RE-AIM and confidence in
elements is what I find most useful
applying the RE-AIM framework were seen over the 2-year time
I think it is best to try to track all of 20.0 57.5 22.5 period from initial funding to the time of the questionnaire. It is
the RE-AIM elements our feeling that these large increases reflect more active dissemina-
tion versus passive diffusion of the RE-AIM framework through-
out the funded states. Such increases can be attributed, in large
DISCUSSION part, to the technical assistance provided grantees about the RE-
This study presents a unique real world application of how the AIM framework both in terms of the annual grantee conferences as
RE-AIM framework was embedded into a national effort by aging well as monthly grantee calls organized by our team. The National
services providers and their partners to expand the dissemina- Council on Aging’s Center for Healthy Aging Technical Resource
tion of evidence-based programing for older adults. The appli- Center also broadly advertised and sponsored webinars and work-
cation of RE-AIM and other implementation and dissemination shops featuring online self-instructional training modules that
frameworks can be encouraged or mandated by funding agen- were created to train providers on how to apply the RE-AIM frame-
cies as illustrated by a prior examination of the application of work to their evidence-based health promotion programs. Many
RE-AIM to funding applications (32). However, little is known of these offerings were co-presented by academics paired with
about how key state decision makers will actually employ different state and aging service provider partners. This enabled community

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 143 | 312
Ory et al. RE-AIM framework for health promotion/disease prevention

respondents to receive information from peers who often served the end of the program). For those interested in more comprehen-
as role models in the dissemination of experience-based infor- sive evaluation aspects, questions can also be added to determine
mation about best strategies for implementing different RE-AIM what types of standardized outcome measures would be feasible
elements. However, great variation in confidence improvements to collect in the dissemination of EBDPs conducted outside of a
was observed among RE-AIM elements (i.e., 46.9% for effective- research setting. Seeking such input from the field aligns with the
ness and 61.0% for reach, adoption, and implementation). This recent emphasis on person-centered research, which stresses the
finding suggests that the need for additional attention for effec- importance of including major stakeholders in research (35).
tiveness and outcome evaluation during trainings and in online As the EBDP field has matured, there are several important
resources provided to grantees in future initiatives. Thus, we offer implications for the future use of RE-AIM. The AoA’s guidelines
the questionnaire as a practical tool for collecting information for initiatives in evidence-based programing for older adults now
about program implementation and evaluation processes from key embed RE-AIM within a larger CQI approach for QA. To carry
program decision makers in a national EBDP initiative. A copy of out CQI, state agencies and their partners need to orient the
the questionnaire is located at the end of this article. team about the QA plan; agree upon RE-AIM performance indica-
A few limitations can be noted. With only 40 responses, this tors; specify designated roles, responsibilities, and timelines for all
research is best viewed as an implementation case study of the program partners; establish mechanisms for periodic review and
RE-AIM framework. While we had anticipated having two respon- standardize protocols for making corrective actions when neces-
dents per state to reflect both planning and evaluation perspectives, sary (36). We believe the questionnaire we developed is valuable
it became evident that not all states had state-wide evaluators. With for conducting initial assessments, as well as ongoing assessments
the small number of respondents, we were not able to examine of the implementation and evaluation process as it unfolds over
responses by respondent type, which in turn made us unable to the life of a funded project.
assess differences in perceptions by whether the respondent was a In 2012, the U.S. ACL/AoA funded 22 states to continue to
state lead, a state-wide coordinator, or program evaluator. How- scale the evidence-based CDSMP and establish a sustainable infra-
ever, it should be noted that there was representation from each of structure for EBDP program delivery (37). With QA as a central
the funded states and this type of data related to practitioner self- focus of the infrastructure operations, the RE-AIM framework
reported confidence levels about RE-AIM use is rarely evaluated provides the guidance for state agencies to create a comprehensive
and/or reported. With the intent of collecting data from stake- system for describing, measuring, and evaluating program deliv-
holders in 27 states quickly and inexpensively, we were restricted to ery to ensure that respondents receive effective, quality services
survey methodology. Our questionnaire reveals interesting obser- and that funding requirements are met. However, with the grow-
vations about the utility of employing the RE-AIM framework, ing expansion of community partnerships for program delivery
which points to issues that can be followed-up about through and staff turnover, ongoing training on the use of the RE-AIM
more in-depth interviews in a particular state. framework is needed.
Additionally, another potential limitation is that this current To support these efforts, the NCOA Center for Healthy Aging
study examines a community grants program implemented at one (38), building on general materials provided by the original
point in time. Requests for respondents to reflect back on their RE-AIM developers (24), offers a myriad of tools, checklists,
familiarity and knowledge about evidence-based programs and issue briefs, and 10 online training modules to inform and
the RE-AIM framework may be subject to recall bias or be affected guide providers working with older populations on the applica-
by personnel changes. Hence, we recommend that implementation tion of the RE-AIM framework. Trainings about frameworks like
assessments be ongoing from the beginning to the end of the pro- RE-AIM would be best attended by community partners along
gram period. Further, different intervention programs could have with their academic partners to help integrate evaluation strategies
been implemented over time, thus, knowledge, attitudes, and prac- and measures within the fabric of program delivery. Additional
tices about RE-AIM elements may be changing. Since this initial questionnaires are available now to help in the identification and
AoA EBDP initiative there has been a 2010–2012 ARRA initiative selection of appropriate frameworks to inform one’s work (39),
for further disseminating the CDSMP in 45 states, the District of and these questionnaires could be incorporated into trainings.
Columbia, and Puerto Rico. However, no systematic data on the Within a relatively short period of time, evidence-based health
application of RE-AIM elements were collected, and the current promotion programing for older adults has evolved into a system
study is the only national examination of the implementation and change movement with the goal of embedding these programs
adoption of the RE-AIM framework in the aging services network. into integrated community, long-term care, and health systems.
According to the AoA (37), state aging services and their public
IMPLICATIONS FOR PRACTICE health partners are developing sustainable service systems uti-
We offer our questionnaire as a pragmatic tool that can be used lizing diverse strategies including embedding programs within
to assess implementation of the RE-AIM framework as a whole, Affordable Care Act initiatives such as care transitions and med-
or its constituent parts. We recommend attention to the full con- ical homes; partnering with Medicaid and other health insurance
tinuum of implementation processes from planning, implemen- providers; pursuing accreditation and Medicare reimbursement
tation, evaluation, and sustainability considerations. Additionally, for Diabetes Self-Management Training; collaborating with Fed-
users of this questionnaire will need to consider in advance the erally Qualified Health Centers, Veterans Administration Med-
most feasible administration (e.g., by online questionnaire or ical Centers, and other healthcare organizations; and teaming
in-person or telephonic interview) and ideal assessment points up with non-traditional partners such as the State Department
(e.g., before a program starts, at a midway point, and then toward of Corrections and State and Local mental health agencies. The

www.frontiersin.org April 2015 | Volume 2 | Article 143 | 313

Ory et al. RE-AIM framework for health promotion/disease prevention

breadth and diversity of these efforts and partnerships calls for 15. Klesges LM, Estabrooks PA, Dzewaltowski DA, Bull SS, Glasgow RE. Beginning
continued attention to capacity-building through ongoing devel- with the application in mind: designing and planning health behavior change
interventions to enhance dissemination. Ann Behav Med (2005) 29(2):66–75.
opment of state-of-the-art training to address the new ways of
doi:10.1207/s15324796abm2902s_10
offering evidence-based programs within an implementation and 16. Green L, Glasgow R, Atkins D, Stange K. The slips ‘twixt cup and lip’: get-
dissemination framework. ting evidence from science to practice. Am J Prev Med (2009) 37(651):S187–91.
doi:10.1016/j.amepre.2009.08.017
ACKNOWLEDGMENTS 17. Glasgow RE, Lichtenstein E, Marcus AC. Why don’t we see more translation of
This work was supported in part by the U.S. Administration on health promotion research to practice? Rethinking the efficacy-to-effectiveness
Aging (90OP0001) and by Cooperative Agreement Number 1U48 transition. Am J Public Health (2003) 93(8):1261–7. doi:10.2105/AJPH.93.8.
1261
DP001924 from the Centers for Disease Control and Prevention
18. Green LW, Glasgow RE. Evaluating the relevance, generalization, and applicabil-
and Special Interest Project, Member Centers at Texas A&M Uni- ity of research issues in external validation and translation methodology. Eval
versity (CDC #U48 DP001924) and University of North Carolina Health Prof (2006) 29(1):126–53. doi:10.1177/0163278705284445
(CDC #U48 DP000059) and the University of Washington Coordi- 19. Gaglio B, Shoup JA, Glasgow RE. The RE-AIM framework: a systematic review
nating Center of the Healthy Aging Research Network (CDC #U48 of use over time. Am J Public Health (2013) 103(6):e38–46. doi:10.2105/AJPH.
2013.301299
DP001911). The findings and conclusions in this journal article are 20. Department of Health and Human Services. AoA Evidence-Based Prevention
those of the authors and do not necessarily represent the official Programs (2013). Available from: http://www.aoa.gov/AoARoot/Press_Room/
position of the Centers for Disease Control and Prevention. Products_Materials/pdf/fs_EvidenceBased.pdf
21. National Council on Aging. Where to Find Evidence-Based Programs (2013).
REFERENCES Available from: http://www.ncoa.org/improve-health/center-for-healthy-aging/
1. U.S. Department of Health and Human Services. U.S. Department of Health and where-to-find-evidence-based.html
Human Services (HHS) Inventory of Programs, Activities and Initiatives Focused 22. Department of Health and Human Services, Administration on Aging. Empow-
on Improving the Health of Individuals with Multiple Chronic Conditions (MCC). ering Older People to Take More Control of Their Health Through Evidence-Based
Washington, DC: Government Printing Office (2011). Prevention Programs: A Public/Private Collaboration. Washington, DC: Govern-
2. Schneider KM, O’Donnell BE, Dean D. Prevalence of multiple chronic con- ment Printing Office (2006). OMB Approval No 0985-0018; HHS-2006-AoA-
ditions in the United States’ Medicare population. Health Qual Life Outcomes BP-0611.
(2009) 7(82):82. doi:10.1186/1477-7525-7-82 23. Ory MG, Altpeter M, Belza B, Helduser J, Zhang C, Smith ML. Perceptions about
3. U.S. Department of Health and Human Services. Healthy People 2020 Top- the community application of RE-AIM in the promotion of evidence-based pro-
ics and Objectives (2012). Available from: http://www.healthypeople.gov/2020/ grams for older adults. Eval Health Prof (2014). doi:10.1177/0163278714542335
topicsobjectives2020/overview.aspx?topicid=5 24. Virginia Polytechnic Institute and State University. Reach Effectiveness Adoption
4. Centers for Disease Control and Prevention (CDC). The State of Aging & Implementation Maintenance (RE-AIM) | Virginia Tech (2013). Available from:
Health in America 2013 (2013). Available from: http://www.cdc.gov/features/ http://www.re-aim.org/
agingandhealth/state_of_aging_and_health_in_america_2013.pdf 25. Altpeter M, Gwyther LP, Kennedy SR, Patterson TR, Derence K. From evidence
5. Ory MG, Smith ML, Kulinski KP, Lorig K, Zenker W, Whitelaw N. Self- to practice: using the RE-AIM framework to adapt the REACHII caregiver inter-
management at the tipping point: reaching 100,000 Americans with evidence- vention to the community. Dementia (2013). doi:10.1177/1471301213494499
based programs. J Am Geriatr Soc (2013) 61(5):821–3. doi:10.1111/jgs.12239 26. Belza B, Toobert DJ, Glasgow RE. RE-AIM for Program Planning: Overview and
6. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, Whitelaw N, et al. Successes Applications. Washington, DC: National Council on Aging (2007).
of a national study of the chronic disease self-management program: meet- 27. Shubert TE, Altpeter M, Busby-Whitehead J. Using the RE-AIM Framework
ing the triple aim of health care reform. Med Care (2013) 51(11):992–8. to translate a research-based falls prevention intervention into a community-
doi:10.1097/MLR.0b013e3182a95dd1 based program: lessons learned. J Safety Res (2011) 42(6):509–16. doi:10.1016/
7. National Center for Chronic Disease Prevention and Health Promotion. Healthy j.jsr.2011.09.003
Aging Research Network: Putting Collective Wisdom to Work for Older Americans. 28. Ory MG, Evashwick CJ, Glasgow RB, Sharkey JR, Browning C, Thomas S. Push-
Atlanta: Centers for Disease Control and Prevention (2011). p. 1–13. CS226787. ing the boundaries of evidence-based research: enhancing the application and
8. Green LW, Ottoson J, Garcia C, Robert H. Diffusion theory and knowledge dis- sustainability of health promotion programs in diverse populations. In: Brown-
semination, utilization, and integration in public health. Annu Rev Public Health ing CJ, Thomas SA, editors. Behavioral Change: An Evidence-Based Handbook
(2009) 30:151–74. doi:10.1146/annurev.publhealth.031308.100049 for Social and Public Health. Edinburgh: Elsevier Churchill (2005). p. 267–93.
9. Brownson RC, Colditz GA, Proctor EK. Dissemination and Implementation 29. Ory MG, Smith ML, Wade A, Mounce C, Wilson A, Parrish R. Implementing
Research in Health: Translating Science to Practice. New York, NY: Oxford Uni- and disseminating an evidence-based program to prevent falls in older adults,
versity Press (2012). Texas, 2007–2009. Prev Chronic Dis (2010) 7(6):A130–5.
10. Ory MG, Smith ML, Resnick B. Changing behavior throughout the life-course: 30. Bandura A. Self-efficacy: toward a unifying theory of behavioral change. Psychol
translating the success of aging research. Transl Behav Med (2012) 2(2):159–62. Rev (1977) 84(2):191–215. doi:10.1037/0033-295X.84.2.191
doi:10.1007/s13142-012-0129-4 31. Bandura A. Self-efficacy. In: Ramachaudran VS, editor. Encyclopedia of Human
11. Chaudoir SR, Dugan AG, Barr CH. Measuring factors affecting implementa- Behavior (Vol. 4), New York, NY: Academic Press (2005). p. 71–81.
tion of health innovations: a systematic review of structural, organizational, 32. Kessler RS, Purcell EP, Glasgow RE, Klesges LM, Benkeser RM, Peek C. What does
provider, patient, and innovation level measures. Implement Sci (2013) 8:22. it mean to “employ” the RE-AIM model? Eval Health Prof (2013) 36(1):44–66.
doi:10.1186/1748-5908-8-22 doi:10.1177/0163278712446066
12. Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of health 33. Eakin EG, Bull SS, Glasgow RE, Mason M. Reaching those most in need: a review
promotion interventions: the RE-AIM framework. Am J Public Health (1999) of diabetes self-management interventions in disadvantaged populations. Dia-
89(9):1322–7. doi:10.2105/AJPH.89.9.1322 betes Metab Res Rev (2002) 18(1):26–35. doi:10.1002/dmrr.266
13. Glasgow RE, Klesges LM, Dzewaltowski DA, Bull SS, Estabrooks P. The future 34. Brownson RC, Jones E. Bridging the gap: translating research into policy and
of health behavior change research: what is needed to improve translation of practice. Prev Med (2009) 49(4):313–5. doi:10.1016/j.ypmed.2009.06.008
research into health promotion practice? Ann Behav Med (2004) 27(1):3–12. 35. Burdine JN, McLeroy K, Blakely C, Wendel ML, Felix MR. Community-based
doi:10.1207/s15324796abm2701_2 participatory research and community health development. J Prim Prev (2010)
14. Glasgow RE. What does it mean to be pragmatic? Pragmatic methods, mea- 31(1):1–7. doi:10.1007/s10935-010-0205-9
sures, and models to facilitate research translation. Health Educ Behav (2013) 36. Administration on Aging. Administration on Aging (AoA) Recommenda-
40(3):257–65. doi:10.1177/1090198113486805 tions for Grantee Quality Assurance Programs (2011). Available from: http:

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 143 | 314
Ory et al. RE-AIM framework for health promotion/disease prevention

//www.ncoa.org/improve-health/center-for-healthy-aging/content-library/ and Promotion section) panel of Review Editors. Because this Research Topic repre-
AoA_Quality_Assurance_Expectations-9-16-w-text-boxes.pdf sents work closely associated with a nationwide evidence-based movement in the US,
37. Administration on Aging. 2012 Prevention and Public Health Funds: Empow- many of the authors and/or Review Editors may have worked together previously in
ering Older Adults and Adults with Disabilities through Chronic Disease Self- some fashion. Review Editors were purposively selected based on their expertise with
Management Education Programs (2013). Available from: http://www.aoa.gov/ evaluation and/or evidence-based programming for older adults. Review Editors were
AoARoot/AoA_Programs/HPW/ARRA/PPHF.aspx independent of named authors on any given article published in this volume.
38. National Council on Aging. Center for Healthy Aging (2013). Available from:
http://www.ncoa.org/improve-health/center-for-healthy-aging/
39. Tabak RG, Khoong EC, Chambers DA, Brownson RC. Bridging research and Received: 11 July 2014; accepted: 01 September 2014; published online: 27 April 2015.
practice: models for dissemination and implementation research. Am J Prev Citation: Ory MG, Altpeter M, Belza B, Helduser J, Zhang C and Smith ML (2015)
Med (2012) 43(3):337–50. doi:10.1016/j.amepre.2012.05.024 Perceived utility of the RE-AIM framework for health promotion/disease prevention ini-
tiatives for older adults: a case study from the U.S. evidence-based disease prevention
Conflict of Interest Statement: The authors declare that the research was conducted initiative. Front. Public Health 2:143. doi: 10.3389/fpubh.2014.00143
in the absence of any commercial or financial relationships that could be construed This article was submitted to Public Health Education and Promotion, a section of the
as a potential conflict of interest. journal Frontiers in Public Health.
Copyright © 2015 Ory, Altpeter, Belza, Helduser, Zhang and Smith. This is an
This paper is included in the Research Topic, “Evidence-Based Programming for Older open-access article distributed under the terms of the Creative Commons Attribution
Adults.” This Research Topic received partial funding from multiple government and License (CC BY). The use, distribution or reproduction in other forums is permit-
private organizations/agencies; however, the views, findings, and conclusions in these ted, provided the original author(s) or licensor are credited and that the original
articles are those of the authors and do not necessarily represent the official position publication in this journal is cited, in accordance with accepted academic practice.
of these organizations/agencies. All papers published in the Research Topic received No use, distribution or reproduction is permitted which does not comply with these
peer review from members of the Frontiers in Public Health (Public Health Education terms.

www.frontiersin.org April 2015 | Volume 2 | Article 143 | 315

Ory et al. RE-AIM framework for health promotion/disease prevention

APPENDIX
SURVEY OF RE-AIM AWARENESS AND UTILIZATION
This questionnaire is being sent to all states receiving AoA or The Atlantic Philanthropies (Atlantic) funding as part of the Evidence-
Based Disease Prevention Program. We request one state lead (either public health or aging) and one program evaluator, preferably
someone who works at a state-wide level to complete the questionnaire separately. Other team members who play key roles in program
implementation and/or evaluation (e.g., a local project coordinator and/or regional coordinator or University partner) are also welcome
to complete the questionnaire.
The purposes of this questionnaire are to (1) describe grantees awareness about RE-AIM; (2) examine ways RE-AIM is used by
grantees in their program planning, implementation, and evaluation of evidence-based programs; and (3) identify useful RE-AIM
materials and training needs. Information learned from this questionnaire will help the AoA and NCOA provide better assistance to
grantees. Completing the questionnaire also allows state teams to reflect on issues related to program planning and implementation
and use these insights to improve state and local processes.
RE-AIM is a framework that has been used in the aging services field to bridge the gap between research and practice by identi-
fying key steps involved in the application of programs and policies in real-world settings. The five elements of RE-AIM are reach,
effectiveness, adoption, implementation, and maintenance.
We recognize that each of the respondents may not be familiar with the technical items about RE-AIM or details about program
implementation or assessment. If you do not know the answer to a specific question, please check the “do not know category.” Other
questions are asking for attitudes about RE-AIM, and we welcome everyone’s opinion.
The questionnaire takes about 10–20 min to complete. Please complete the questionnaire by XXX. A Word document is avail-
able should you want to see all the questions in advance. While we are asking that each state respondent fill out the questionnaire
independently, we suggest that the state teams may want to review their responses at a team meeting after submission.
Contact XXX for questions about the questionnaire or to obtain a Word document of the questionnaire. The questionnaire is set
up such that your role and responses determine specific questions you are asked to complete (so note that the computer version may
differ slightly from the word version).
Completing this questionnaire is voluntary. The responses will be confidential and reporting will occur in aggregate for the entire
group of respondents. For those willing to participate further, we will also be seeking to record some in-depth experiences and will be
documenting a few grantee stories that detail the successes and challenges in the application of RE-AIM elements. Thank you for your
time and interest.

I have read and understand the information above and wish to voluntarily participate in this survey.
q Yes
q No

Information about the Person Completing this Survey

First Name:
_______________________________________

Last Name:
_______________________________________

What is your email address:

_______________________________________

What is your primary role on the AoA/Atlantic evidence-based disease prevention grant project? (Select one)
q State lead or state-wide coordinator
q State-wide evaluator
q Regional project coordinator
q Local project coordinator
q Other

If you selected Other, please specify:

___________________________________________________________

If you selected State lead or State-wide coordinator, please specify the name of your agency:
___________________________________________________________

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 143 | 316
Ory et al. RE-AIM framework for health promotion/disease prevention

If you selected State-wide evaluator, please specify the name of your agency:
___________________________________________________________

May we contact you if any responses are unclear or elaboration is needed?

q Yes
q No
Which state are you completing this survey for?
q Arizona
q Arkansas
q California
q Colorado
q Connecticut
q Florida
q Hawaii
q Idaho
q Illinois
q Indiana
q Iowa
q Maine
q Maryland
q Massachusetts
q Michigan
q Minnesota
q New Jersey
q New York
q North Carolina
q Ohio
q Oklahoma
q Oregon
q Rhode Island
q South Carolina
q Texas
q Washington
q West Virginia
q Wisconsin
q Other, please specify

Which evidence-based programs are you currently delivering under the auspices of the AoA/Atlantic Evidence-based
Disease Prevention Program (Check all that apply)
q Chronic Disease Self-Management Program (CDSMP)
q A Matter of Balance/Volunteer Lay Leader
q Active Choices
q Active Living Every Day (ALED)
q Enhance Fitness
q Enhance Wellness
q Healthy Eating
q Healthy IDEAS
q Healthy Moves
q Medication Management
q PEARLS
q Spanish Arthritis Self-Management Program
q Step by Step
q Stepping On
q Strong for Life
q Tai Chi
q Other
q Do not know

www.frontiersin.org April 2015 | Volume 2 | Article 143 | 317

Ory et al. RE-AIM framework for health promotion/disease prevention

If Other, please specify:

When did you start working with evidence-based disease prevention programs for older adults?
q Before 2000
q 2001
q 2002
q 2003
q 2004
q 2005
q 2006
q 2007
q 2008

Familiarity of and Confidence with RE-AIM. These next set of questions pertain to your AoA/Atlantic evidence-based disease
prevention state-grant funded in 2006, 2007, 2008):

At the start of your state-grant funding (2006, 2007, 2008): Use not relevant (NR), if YOU WERE NOT present during the
initial stages of the program implementation

not at all a little some a lot NR

How familiar were you with evidence-based disease prevention programs? q q q q q

How knowledgeable were you with the RE-AIM framework as a whole? q q q q q

How confident were you at the start of your grant-funding in applying the
RE-AIM element:
Reach? q q q q q
Effectiveness? q q q q q
Adoption? q q q q q
Implementation? q q q q q
Maintenance? q q q q q

At the current time

not at all a little some a lot NR

How knowledgeable are you now about evidence-based disease prevention q q q q q

programs?

How knowledgeable are you about RE-AIM framework as a whole? q q q q q

How confident are you that you can now apply the RE-AIM element:
Reach? q q q q q
Effectiveness? q q q q q
Adoption? q q q q q
Implementation? q q q q q
Maintenance? q q q q q

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 143 | 318
Ory et al. RE-AIM framework for health promotion/disease prevention

Application of RE-AIM
We are interested in learning about ways that RE-AIM has been used by the State teams in this AoA/Atlantic initiative. To what
extent have decisions about the following been influenced by the RE-AIM framework and approach? Please mark DK for Do not
Know.

not at all a little some a lot DK

PLANNING:
Select community partners q q q q q
Select evidence-based programs for implementation q q q q q
Select host and/or implementation sites q q q q q
Identify target populations (people who may participate in your program[s]) q q q q q
Select assessment/evaluation tools q q q q q

IMPLEMENTATION/EVALUATION:
Plan or alter participant recruitment q q q q q
Structure agendas and/or team meetings q q q q q
Conduct midcourse evaluations q q q q q
Structure reports q q q q q
Present/publicize program findings q q q q q

MAINTENANCE:
Address strategies for maintaining participant improvement q q q q q
Guide discussions and/or planning around program sustainability q q q q q
Secure funding for maintaining program delivery q q q q q
Building infrastructure to maintain program staffing q q q q q
Build capacity for ongoing quality assurance q q q q q

What other decisions have been influenced by the RE-AIM framework and approach?

—————————————————————————————————————————————————–

Attitudes regarding Application of RE-AIM. Please base your responses in terms of your attitudes related to the application of
RE-AIM in your AoA/Atlantic evidence-based disease prevention project

What is your level of agreement with each of the following:

RE-AIM is useful for:

strongly Disagree agree strongly DK
disagree agree

Planning in this initiative q q q q q

Implementation of this effort q q q q q
Evaluation of this effort q q q q q
Planning efforts with our other aging programs q q q q q
Implementation efforts with our other aging programs q q q q q

How valuable do you see RE-AIM for different audiences? What is your level of agreement with each of the following?
RE-AIM is a valuable tool for:
strongly disagree agree strongly DK
disagree agree

Providers q q q q q
Community Leaders q q q q q
Policy makers q q q q q
Evaluators q q q q q

www.frontiersin.org April 2015 | Volume 2 | Article 143 | 319

Ory et al. RE-AIM framework for health promotion/disease prevention

What is your level of agreement with each of the following statements about RE-AIM?
strongly disagree agree strongly DK
disagree agree

RE-AIM is too academic q q q q q

Monitoring RE-AIM elements requires special expertise q q q q q
The different RE-AIM elements are easy to understand q q q q q
The training I have received in how to apply RE-AIM is sufficient q q q q q
RE-AIM takes too much time to implement q q q q q
Measuring the successful application of different RE-AIM elements is difficult q q q q q
Looking at just one or two RE-AIM elements is what I find most useful q q q q q
I think it is best to try to track all of the RE-AIM elements q q q q q
The training material explaining RE-AIM are easy to access q q q q q

Does your state team measure RE-AIM elements?

q Yes
q No
q Do not know

Please indicate how Reach is being measured? (Check all that apply)
q Number of enrollees
q Participant characteristics
q Other

If you selected Other, please specify:

Please indicate how Effectiveness is being measured? (Check all that apply)
q Health status
q Quality of life
q Symptomatology (e.g., pain or fatigue)
q Health behaviors (physical activity or nutrition)
q Self-efficacy
q Health care utilization
q Health care costs
q Interference with routine activities
q Medication management
q Communication with health care providers
q Physical functioning
q Other

If you selected Other, please specify:

Please indicate how Adoption is being measured? (Check all that apply)
q Number of implementation sites
q Type of sites
q Location of sites
q Other

If you selected Other, please specify:

Please indicate how Implementation is being measured? (Check all that apply)
q Checklists
q Observational data
q Regular phone calls for retraining
q Periodic face to face meetings for retraining
q Other

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 143 | 320
Ory et al. RE-AIM framework for health promotion/disease prevention

If you selected Other, please specify:

Please indicate how Maintenance is being measured? (Check all that apply)
q On-going benefits for participants
q Continuation of program delivery
q Expansion of organizational partners
q Identification of external funding
q Identification of in-kind resources
q Other

If you selected Other, please specify:

Training Feedback and Needs

There are a number of resources available to help learn about RE-AIM. How valuable have these resources been to you?

Do not know Not at all A little Some A lot

resource
Publications and Webinars
Re-aim.org website q q q q q
Moving Ahead: Strategies and Tools to Plan, Conduct, and Main- q q q q q
tain Effective Community-Based Physical Activity Programs for Older
Adults (blue cover monograph, produced by HAN)
RE-AIM for Program Planning: Overview and Applications (NCOA q q q q q
issue brief)
NCOA evidence-based online training modules q q q q q
NCOA issue briefs on EBHP q q q q q

Presentations and Centers

Presentations about RE-AIM at national meetings q q q q q
Presentations about RE-AIM at state meetings q q q q q
NCOA Center for Healthy Aging Technical Resource Center q q q q q

What other resources have you used to help learn about RE-AIM?

Is there RE-AIM training and/or technical assistance available to all of the designated geographic areas in your state
grant?
q Yes
q No
q Do not know

Is this a one-time offering?

q Yes
q No
q Do not know

About how many times do you offer this training a year?

q 2 times
q 3-5 times
q 6-10 times
q more than 10

Has training been helpful?

q Yes
q No
q Do not know

www.frontiersin.org April 2015 | Volume 2 | Article 143 | 321

Ory et al. RE-AIM framework for health promotion/disease prevention

In what ways could the training be improved? (Check all that apply)
q Training on the RE-AIM components
q Training on the application of RE-AIM components
q More in-depth training on RE-AIM components
q Announce training in advance
q Offer repeated trainings
q Make training more practical and less academic
q Provide case examples from the field
q Conduct phone webinars
q Develop on-line training
q Other

If you selected Other, please specify:

In your state, who is currently coordinating training and technical assistance on RE-AIM? (Check all that apply)
q State lead(s)
q Program evaluator
q Regional coordinator
q Local coordinator
q Do not know
q No one
q Other

If Other, please specify:

In your state, who would you recommend to coordinate training and technical assistance on RE-AIM? (Check all that
apply)
q State lead(s)
q Program evaluator
q Regional coordinator
q Local coordinator
q Do not know
q No one
q Other

If Other, please specify:

Dissemination of RE-AIM

Have you used RE-AIM in programmatic efforts other than the AoA/Atlantic evidence-based disease prevention
programs?
q Yes
q No

If yes, how many different projects? (Please indicate a number)

——————————–

Have you taught someone else in your agency how to use RE-AIM?
q Yes
q No

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 143 | 322
Ory et al. RE-AIM framework for health promotion/disease prevention

Would you apply RE-AIM in future projects?

q Definitely yes
q Probably yes
q Probably no
q Definitely no

We are interested in knowing and documenting how long it is taking sites to implement programmatic activities and develop data
collection systems. Please estimate the month/date that your site initiated the different activities listed below in terms of CDSMP

Since receiving your State funding for AoA/Atlantic evidence-based disease prevention funding (2006, 2007, or 2008).

Please enter in the form of XX/XXXX (e.g., 06/2006) or enter in NR if not yet conducted. If you are not exactly sure, please
give us your best estimate.

What month/year did your State offer the first CDSMP training for master trainers to work _______________________
on the State evidence-based grant?
What month/year did you have your first lay leader training? _______________________
What month/year did you offer your first CDSMP class? _______________________
What month/year did you start collecting outcome or data? _______________________
What month/year did you begin analyzing your data? _______________________
What month/year did you provide your first report back to your community settings? _______________________

Current Program and Evaluation Stage Outcome Assessments

An outcome assessment measures programmatic impacts on each participant, e.g., on health or health behaviors, functioning or quality of
life. We are interested in learning about your outcome assessments in your AoA/Atlantic Program.

Do you collect participant outcomes data?

q Yes
q No

On average, how long does it take a participant to complete current baseline outcome measures?
q Less than 5 minutes
q 6-10 minutes
q 11-20 minutes
q 21-30 minutes
q More than 30 minutes
q Do not know

On average, how long does it take to complete each follow up measure?

q Less than 5 minutes
q 6-10 minutes
q 11-20 minutes
q 21-30 minutes
q More than 30 minutes
q Do not know

Have you made modifications in your participant outcome assessment form in year two or three of your funding? (Check
all that apply)
q Kept the same items, no modifications
q Added new items
q Eliminated many of the items
q Changed the original items
q Do not know

www.frontiersin.org April 2015 | Volume 2 | Article 143 | 323

Ory et al. RE-AIM framework for health promotion/disease prevention

What were the reasons for making these changes? (Check all that apply)
q Questions were confusing
q Survey took too long
q We wanted to compare our findings with other states
q Data we were collecting was not useful
q Too burdensome for participants
q Too burdensome for staff
q We wanted to have data to report to our key stakeholders
q Wanted to wait until got programs up and running
q Added the common core battery recommended by Measures of Success group
q Other. Please Specify

Post Grant Data Collection Plans

Process data includes demographics characteristics, number of participants, record of their attendance, characteristics of implementation
sites, etc.

Outcome data is programmatic effects on health, health behaviors, functioning and quality of life, etc.

After your AoA/Atlantic project funding ends, will you:

This questions pertains to Process Data
q Collect approximately the same amount of process data
q Decrease amount of process data collected
q Increase amount of process data collected
q Not collect any process data
q No decision has been made yet
q Do not know

After your AoA/Atlantic project funding ends, will you:

This questions pertains to Outcome Data
q Collect approximately the same amount of outcome data
q Decrease amount of outcome data collected
q Increase amount of outcome data collected
q We do not collect any outcome data now
q No decision has been made yet
q Do not know

In future studies what is the longest/maximum amount of time you would recommend for the collection of participant
outcome data?
q Less than 5 minutes
q 6-10 minutes
q 11-20 minutes
q 21-30 minutes
q More than 30 minutes
q Do not know

What is the single most important lesson learned so far about RE-AIM, the one thing you wish you would have known
ahead of time?

You have now completed the survey. Kudos to you! Thank you for your time and interest. Click submit.

If you have a story you would like to share with us about your successes and/or challenges with RE-AIM, please indicate
here your willingness for us to contact you
q You may contact me

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 143 | 324
 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00156

National dissemination of multiple evidence-based disease

prevention programs: reach to vulnerable older adults
Samuel D. Towne Jr.1 *, Matthew Lee Smith 2 , SangNam Ahn 1,3 , Mary Altpeter 4 , Basia Belza 5 ,
Kristie Patton Kulinski 6 and Marcia G. Ory 1
1
Department of Health Promotion and Community Health Sciences, Texas A&M Health Science Center School of Public Health, College Station, TX, USA
2
Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA, USA
3
Division of Health Systems Management and Policy, School of Public Health, The University of Memphis, Memphis, TN, USA
4
University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
5
Biobehavioral Nursing and Health Systems, School of Nursing, University of Washington, Seattle, WA, USA
6
National Council on Aging, Washington, DC, USA

Edited by: Older adults, who are racial/ethnic minorities, report multiple chronic conditions, reside in
Sue Ellen Levkoff, University of South
medically underserved rural areas, or have low incomes carry a high burden of chronic
Carolina, USA
illness but traditionally lack access to disease prevention programs. The Chronic Dis-
Reviewed by:
Cheryll Diann Lesneski, UNC-Chapel ease Self-Management Program (CDSMP), A Matter of Balance/Volunteer Lay Leader
Hill, USA (AMOB/VLL), and EnhanceFitness (EF) are widely disseminated evidence-based programs
Katherine Henrietta Leith, University (EBP), but the extent to which they are simultaneously delivered in communities to reach
of South Carolina, USA
vulnerable populations has not been documented. We conducted cross-sectional analyses
*Correspondence:
of three EBP disseminated within 27 states throughout the United States (US) (2006–
Samuel D. Towne Jr., Department of
Health Promotion and Community 2009) as part of the Administration on Aging (AoA) Evidence-Based Disease and Disability
Health Sciences, Texas A&M Health Prevention Initiative, which received co-funding from the Atlantic Philanthropies.This study
Science Center School of Public measures the extent to which CDSMP, AMOB/VLL, and EF reached vulnerable older adults.
Health, College Station, TX
It also examines characteristics of communities offering one of these programs relative to
77843-1266, USA
e-mail: [email protected] those simultaneously offering two or all three programs. Minority/ethnic participants rep-
resented 38% for CDSMP, 26% for AMOB/VLL, and 43% for EF. Rural participation was
18% for CDSMP, 17% for AMOB/VLL, and 25% for EF. Those with comorbidities included
63.2% for CDSMP, 58.7% for AMOB/VLL, and 63.6% for EF while approximately one-
quarter of participants had incomes under $15,000 for all programs. Rural areas and health
professional shortage areas (HPSA) tended to deliver fewer EBP relative to urban areas
and non-HPSA. These EBP attract diverse older adult participants. Findings highlight the
capability of communities to serve potentially vulnerable older adults by offering multiple
EBP. Because each program addresses unique issues facing this older population, further
research is needed to better understand how communities can introduce, embed, and sus-
tain multiple EBP to ensure widespread access and utilization, especially to traditionally
underserved subgroups.
Keywords: evidence-based programs, community intervention, minority adults, older adults, aging health

INTRODUCTION Self-management is seen as a critical component of clinical- and

The aging of the US population has far reaching effects on the community-based health care (11, 12). Although self-management
American health care system (1). Chronic disease is becoming strategies are widely promoted (13), individuals with multiple
endemic among older Americans (2). National statistics indicate chronic conditions experience barriers to successful self-care (14).
most adults aged 65 and older have at least one chronic condi- Given that older adults have different chronic diseases, varying
tion (91%), while nearly three-quarters have two or more chronic comorbidity combinations, and are at differing stages of disease
conditions (2). Additionally, age-related geriatric conditions are progression, there is need for multiple intervention approaches in
prevalent in this population and have stark public health conse- any given community.
quences. Each year, falling affects approximately one-third of older In concert with public health officials and policy makers’ inter-
adults in the US (3) contributing to death and serious injuries and ests to identify effective ways to lessen the impact of chronic
costing billions of dollars in healthcare expenses annually (esti- disease and other complications among the aging population [e.g.,
mated to reach $30 billion by 2020) (4–8). In addition, high blood Healthy People 2020 (15)], evidence-based programs (EBP) for
pressure, high cholesterol, heart disease, arthritis, and diabetes are older adults have emerged and proliferated in the US (16–20).
common among older adults (9), and in many cases comorbidities In recent years, multiple EBP have been disseminated through the
are also present (10). US aging services network to address different healthcare concerns

www.frontiersin.org April 2015 | Volume 2 | Article 156 | 325

Towne et al. Multiple evidence-based disease prevention programs

experienced by older adults (21). However, there is no “one size fits ratio of less than 3,500 to 1) (43). Rural areas were defined as hav-
all” EBP, which highlights the need for communities to introduce ing a UIC of ≥3 versus urban/metropolitan defined as having a
multiple programs to meet the various needs of a diverse aging UIC of 1–2. UICs take into consideration the population size and,
population. for rural areas, the relative proximity to metropolitan or microp-
While it is assumed that distinct EBP attract specific types of olitan areas (44). We used ArcGIS version 10.2 for all mapping
participants (17, 18) and certain types of participants are more of data presented in the figures (45). Chi-square tests were used
likely to attend EBP at particular types of delivery sites (16), the to compare categorical study variables and independent sample
extent to which EBP attract and retain potentially vulnerable older t -tests were used to assess differences in continuous variables. We
adults is not fully understood. Older adults deemed vulnerable can used SAS version 9.4 for all statistical analyses (46).
include those with comorbid conditions (22), in advanced age, and
of racial/ethnic minority status (23–25). Vulnerability can also be VARIABLES
defined as older adults residing in areas with limited resources, Vulnerability
which include rural areas (26–28), those with limited health care Vulnerable adults are the focus of our analysis. Acknowledging that
providers, or those with high poverty rates compared to most other vulnerability can be defined in numerous ways, the operational
areas (29, 30). As such, the purpose of this study was twofold: (1) definition of vulnerability used in this study includes participants
to measure the extent to which three widely disseminated EBP meeting one or more of the following criteria: being in advanced
reached vulnerable older adults and (2) to assess the extent to age (i.e., age 75 and older), having low income (i.e., self-reporting
which delivery areas offered multiple EBP. an annual household income <$15,000), being in a racial/ethnic
minority (non-White), having one or more chronic conditions, liv-
MATERIALS AND METHODS ing in a HPSA (47, 48), living in an area with poverty rates above
SELECTED EVIDENCE-BASED PROGRAMS the median (i.e., based on the percent Federal Poverty Rates in
For the purposes of this study, three EBP for older adults 2008 (14.1%) at the county level according to the 2013 AHRF), or
were examined. The programs included in this study include: living in a rural area (i.e., counties with UIC ≥3) (49). Only those
Stanford University’s Chronic Disease Self-Management Pro- individuals with one or more chronic conditions were included in
gram (CDSMP), A Matter of Balance/Volunteer Lay Leader our analyses.
(AMOB/VLL), and EnhanceFitness (EF). Each program was
selected because of its national dissemination spanning multi- Covariates
ple states and well-documented effectiveness for improving health Sex of the participants who attended the EBP was reported.
outcomes in community settings. Income was categorical; however, a missing category for income
These EBP have demonstrated their effectiveness in improving was included in analyses, as we did not assume this was missing at
health among older adults. CDSMP targets adults with multiple random.
chronic conditions (e.g., teaching self-management skills) and has
been shown to be effective at delaying the onset of illness and HANDLING MISSING DATA
helping participants improve the management of multiple chronic As described elsewhere (50), the AoA initiative required only a
diseases while reducing hospitalizations (31–34). AMOB/VLL tar- few participant level variables be collected, including age, sex,
gets older adults, especially those at risk of falling (35) and has living alone status, race/ethnicity, and ZIP Code. Even this lim-
been shown to reduce the fear of falling, improve long-term social ited number of variables was not collected routinely by all state
functioning, and improve long- and short-term mobility in older grantees; however, some states chose to routinely collect informa-
adults (17, 36–38). EF is a group exercise program (39) that has tion related to chronic conditions and income. Missingness (i.e.,
been shown to improve upper and lower body muscle strength, missing data) was addressed independently according to the analy-
depression (40), and lower healthcare costs (41). sis performed and variables included. Independently (i.e., only
considering each variable’s missingness exclusive of other miss-
DATA ANALYSES ing variables), our sample size (n = 48,413) was gradually reduced
We conducted a cross-sectional analysis of three EBP. Partici- when removing missing observations for race (n = 37,661), sex
pant data and information about program delivery locations were (n = 39,488), county Federal Information Processing Standard
drawn from the National Council on Aging’s database of 24 states (FIPS) (n = 36,599), age (n = 35,248), the number of chronic
implementing EBP from 2006 to 2009 as part of the Adminis- conditions (n = 22,007), and income (n = 22,956). Dependently,
tration on Aging (AoA) Evidence-Based Disease and Disability when collectively removing observations for race, sex, county FIPS,
Prevention Initiative and 3 states funded by the Atlantic Philan- and age, our sample size used in univariate and bivariate analysis
thropies (16). Only data collected between 2006 and 2009 from was 30,185 observations.
these initiatives were included in these analyses. These data were
linked with the 2013 Area Health Resource File (AHRF) to iden- RESULTS
tify Primary Care Health Professional Shortage Areas (HPSA) REACH INTO VULNERABLE POPULATIONS
and Urban Influence Codes (UIC) (42). HPSA is classified into Table 1 presents the distribution of participant characteristics in
full-HPSA, partial (only a portion of the county was classified the aggregate and by program type. Of the 30,185 participants
as a HPSA), and a non-HPSA. A HPSA is classified based on enrolled in one of three EBP in this study, the majority participated
geographic area and population size (e.g., primary care physician in CDSMP (n = 16,612), followed by AMOB/VLL (n = 8,391), and

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 156 | 326
Towne et al. Multiple evidence-based disease prevention programs

Table 1 | Distribution of participant characteristics by program.

CDSMP AMOB/VLL EF Total

(n = 16,612) (n = 8,391) (n = 5,182) (n = 30,185)

n % n % n % n %

Age group
<50 1,323a,b 8.0 55b,c 0.7 135a,c 2.6 1,513 5.0
50–64 3,635a,b 21.9 656b,c 7.8 1,043a,c 20.1 5,334 17.7
65–74 5,151a,b 31.0 2,120b,c 25.3 1,933a,c 37.3 9,204 30.5
75 and older 6,503a,b 39.2 5,560b,c 66.3 2,071a,c 34.0 14,134 46.8
Age (mean) 69.6* 77.5* 71.4* 72.1 Age (mean) 69.6* 77.5* 71.4*
(SD = 13.2) (SD = 9.1) (SD = 10.5) (SD = 12.2) (SD = 13.2) (SD = 9.1) (SD = 10.5)
Race/ethnicity
White 10,250 61.7 6,270 74.7 3,010 58.1 19,530 64.7
Black or African American 2,136 12.9 581 6.9 987 19.1 3704 12.3
American Indian/Alaska Native 147 0.9 221 2.6 180 3.5 548 1.8
Asian 882 5.3 151 1.8 265 5.1 1,298 4.3
Other 764 4.6 199 2.4 146 2.8 1,109 3.7
Hispanic 2,433 14.7 969 11.6 594 11.5 3,996 13.2
Sex
Male 3,648 22.0 1,393 16.6 756 14.6 5,797 19.2
Female 12,964 78.0 6,998 83.4 4,426 85.4 24,388 80.8
Number of chronic conditions
1 4,185 36.6 806 40.9 1,120 37.9 6,111 37.3
2 3,828 33.5 733 37.2 1,048 35.5 5,609 34.3
3 2,379 20.8 332 16.9 544 18.4 3,255 19.9
4 835 7.3 85 4.3 209 7.1 1,129 6.9
5+ 217 1.9 14 0.7 33 1.1 264 1.6
Average 2.04* (SD = 1.0) 1.87* (SD = 0.9) 1.98* (SD = 2.0) 2.01 (SD = 1.0)
Income
Missing 3,292 47.8 1,498 40.2 1,917 39.8 6,707 43.5
Less than $15,000 1,692 24.6 975 26.2 1,059 22.0 3,726 24.1
$15,000–24,999 820 11.9 479 12.9 742 15.4 2,041 13.2
$25,000–49,999 694 10.07 465 12.48 715 14.84 1,874 12.14
$50,000–75,000 251 3.64 199 5.34 254 5.27 704 4.56
More than $75,000 143 2.07 109 2.93 132 2.74 384 2.49
Rurality
Rural 2,675 16.10 1,189 14.17 1,437 27.73 5,301 17.56
Urban 13,937 83.90 7,202 85.83 3,745 72.27 24,884 82.44

*Significantly (p < 0.05) different by program for select comparisons (i.e., age and the number of chronic conditions).
a
Significantly different CDSMP versus EF, within age group.
b
Significantly different CDSMP versus AMOB/VLL, within age group.
c
Significantly different AMOB/VLL versus EF, within age group.

EF (n = 5,182). On average, participants were aged 72.09 (±12.21) aged 75 years and older (66.3%) compared to 39.2% for CDSMP
with 46.8% aged 75 and older. The majority of participants were and 40.0% for EF. Those with comorbid conditions (i.e., 2 or
female (80.8%), white (64.7%), and non-Hispanic (87.8%). The more chronic conditions) represented 63.4% for CDSMP, 59.1%
mean number of self-reported chronic conditions was 2 (±1.00). for AMOB, and 62.1% for EF. The average number of chronic
Approximately 24% of participants reported household incomes diseases was significantly (p < 0.05) different for all comparisons
less than $15,000 per year, and 17.6% resided in rural areas. across programs except CDSMP versus EF; CDSMP attracted par-
The average age of participants varied significantly (p < 0.05) ticipants with the most chronic conditions. CDSMP also attracted
across program types (i.e., 77.49 for AMOB/VLL, 71.39 for EF, the largest proportion of Hispanic participants (14.7%). Those
69.58 for CDSMP) with AMOB/VLL attracting the oldest par- residing in rural areas represented 16.0% for CDSMP, 14.2% for
ticipants. AMOB/VLL had the highest proportion of participants AMOB/VLL, and 27.7% for EF. Those reporting incomes less than

www.frontiersin.org April 2015 | Volume 2 | Article 156 | 327

Towne et al. Multiple evidence-based disease prevention programs

FIGURE 1 | The distribution of areas with a higher poverty rate than the median and a presence of evidence-based programs in 2006–2009.

$15,000 per year were 24.6% CDSMP, 26.2% AMOB/VLL, and compared to 88.8% by CDSMP and 86.5% by AMOB/VLL. Again,
22% EF. programs were delivered in high need areas, but that the extent
To graphically illustrate the extent to which programs were varied by state (also seen in Figure 1).
being delivered in areas classified as vulnerable by poverty rate Figure 3 depicts the intersection of poverty and HPSA, where
or health access, a series of three maps highlighting participat- black shading represents places where programs were offered in
ing states were constructed. Figure 1 shows where programs were areas classified as both high poverty (above the median percent
delivered in areas with higher poverty rates than the 2008 median poverty for 2008 measured at the county) and HPSA (full or par-
rate. States without shading include those states that were not tial). As seen, approximately 47.5% of the participants attended
included in the initiative. Gray shading represents where pro- programs in areas with both higher poverty rates and that were
grams (i.e., CDSMP, AMOB/VLL, EF) were offered in areas equal a HPSA. A greater proportion of participants in these areas were
to or below the 2008 median poverty rate. Black shading represents served by EF at 55.7%, compared to 48.6% by AMOB/VLL and
where programs were offered in areas higher than the 2008 median 44.3% by CDSMP.
poverty rate. As seen, approximately 49.6% of the participants
attended programs in areas with higher poverty rates. A greater AVAILABILITY OF MULTIPLE EVIDENCE-BASED PROGRAMS
proportion of participants in areas with higher poverty rates were Table 2 presents the distribution of counties that delivered one,
served by EF at 58.4%, compared to 52.2% by AMOB/VLL and two, and three of the EBP included in this study. Overall, 78.8%
45.5% by CDSMP. As can be seen, programs were delivered in of counties labeled full-HPSA delivered only one EBP, 18.5%
high need areas, but the extent varied by state. For example, larger delivered two of the EBP, and 2.6% delivered all three EBP. Approx-
portions of California and North Carolina and smaller propor- imately 84% of rural counties delivered one of these EBP, and 1.6%
tions of Oklahoma, Maine, and Washington delivered programs delivered all three programs. Nearly 75% of counties within higher
in areas with higher poverty. poverty areas delivered one EBP versus 2.3% that offered all three
Figure 2 shows where programs were delivered in areas clas- EBP.
sified as a HPSA. Gray shading represents where programs (i.e., Table 3 presents the distribution of participants by counties
AMOB/VLL, CDSMP, EF) were offered in a non-HPSA. Black that delivered one, two, and three of the EBP included in this study.
shading represents where programs were offered in a HPSA (full Overall, 43.6% of participants attended programs in areas offer-
or partial). As presented in the map (Figure 2), approximately ing only one EBP, 39.6% attended programs offering two of the
88.9% of the participants attended programs in a HPSA. A greater EBP, and 16.7% attended programs offering all three EBP. Fifty-
proportion of participants in a HPSA were served by EF at 92.9%, nine percent of participants in rural counties had only one EBP

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 156 | 328
Towne et al. Multiple evidence-based disease prevention programs

FIGURE 2 | Counties identified as a primary care health professional shortage area (HPSA) with a presence of evidence-based programs.

available to them, and 12.5% had all three programs available in through the 2006–2009 AoA Evidence-Based Disease and Dis-
their counties. Approximately 13% of participants within higher ability Prevention Initiative and 3 states funded by the Atlantic
poverty areas had all three EBP available in their counties versus Philanthropies. The findings reveal that the three EBP reached
20.8% in areas with lower poverty rates. Among areas that were a substantial percentage of adults who were aged 75 years or
designated as a full-HPSA, the majority of participants were in older and had incomes below $15,000. The proportion of minor-
areas where one or two EBP were available as compared to 6% in ity/ethnic participants in each of these three EBP was higher than
areas where all three EBP were available. the current proportion of minority/ethnic adults in US (approx-
Figure 4 shows the distribution of rural counties that delivered imately 22%) in 2012 (51). Additionally, among those with at
one versus two versus three of the EBP included in this study. As least one chronic condition, the majority of these participants
seen, there were very few areas that delivered all three programs had comorbid conditions (i.e., two or more chronic conditions)
(2.3%) and even fewer in rural counties (1.6%), and those that did overall and within each program. We note that an overwhelm-
(i.e., delivered all three programs) were concentrated in just a few ing number of women participated in these programs, which, in
states (e.g., AZ, CA, MA, NC, SC, TX). part, seems to reflect the larger proportion of women representing
Table 4 shows the distribution of counties by selected charac- the American older adult population. However, this is frequently
teristics and programs. Overall, CDSMP was located in the largest reported in other national studies of EBP for older adults (16–
number of counties at 419, followed by AMOB/VLL (253), and 18, 20). The lower reach to males and ethnic minorities raises
EF (103). In addition, the majority of counties offering EBP were questions as to whether the programs lack saliency to specific
located in a full or partial-HPSA (see Table 4). A higher propor- subpopulations or whether the providers are finding it difficult
tion of the EBP were located in metropolitan areas, as compared to find the right strategies to recruit such subpopulations. Further
to non-metropolitan areas. More counties offering these EBP were research is needed to explore and examine ways in which nationally
also located in lower poverty areas (compared to above the median coordinated intervention efforts can recruit a greater proportion
poverty rate). of diverse populations.
It is not surprising that CDSMP had the largest number of par-
DISCUSSION ticipants, given that all participating states were required to deliver
This study examines the delivery of three EBP delivered to vul- this program, but could add other EBP desired by community part-
nerable individuals (i.e., minority/ethnic individuals, those living ners. The overall distribution of programs (as seen in the figures)
in rural or HPSA areas, with low income, and those having one illustrates the limited reach within the 27 grantee states during
or more chronic conditions or advanced age) within 24 states this specific initiative. However, there has been subsequent growth

www.frontiersin.org April 2015 | Volume 2 | Article 156 | 329

Towne et al. Multiple evidence-based disease prevention programs

FIGURE 3 | Counties identified as a health professional shortage area, and having higher than the median poverty distribution with the presence of
evidence-based programs.

Table 2 | Distribution of counties by availability of multiple Table 3 | Distribution of participants by availability of multiple
evidence-based programs (CDSMP, AMOB/VLL, EF) by health evidence-based programs (CDSMP, AMOB/VLL, EF) by health profes-
professional shortage area (HPSA), rurality, and poverty status in sional shortage area (HPSA), rurality, and poverty status in 2008.
2008.
One Two Three
One Two Three program programs programs
program programs programs
n % n % n %
n % n % n %
HPSA status
HPSA status Full-HPSA 6,120a,b 43.9 6,976a,b 50.0 845b 6.1
Full-HPSA 183a 78.9 43 18.5 6 2.6 Partial-HPSA 4,477b,c 34.7 4,208b,c 32.6 4,209b 32.6
Partial-HPSA 159 71.0 57b 25.5 8 3.6 Non-HPSA 2,577a,c 76.9 773a,c 23.1 0 0
Non-HPSA 135a 82.8 28b 17.2 0 0 Rurality
Rurality Rural 3,128* 59.0 1,511* 28.5 662* 12.5
Rural 216* 84.4 36* 14.1 4 1.6 Urban 10,046* 40.4 10,446* 42.0 4,392* 17.7
Urban 261* 71.9 92 25.3 10 2.8 Poverty rate
Poverty rate Above median 5,946* 39.7 7,122* 47.6 1,894* 12.7
Above median 298* 75.6 85* 21.6 11* 2.8 At/below median 7,228* 47.5 4,835* 31.8 3,160* 20.8
At/below median 179* 79.6 43* 19.1 3* 1.3 Total 13,174 43.6 11,957 39.6 5,054 16.7
Total 477 77.1 128 20.7 14 2.3
*Significantly (p < 0.05) different by characteristic (e.g., rurality).
*Significantly (p < 0.05) different by characteristic (e.g., rurality). a
Significantly (p < 0.05) different non-HPSA versus full-HPSA.
a
Significantly (p < 0.05) different non-HPSA versus full-HPSA. b
Significantly (p < 0.05) different non-HPSA versus partial-HPSA.
b
Significantly (p < 0.05) different non-HPSA versus partial-HPSA. c
Significantly (p < 0.05) different partial-HPSA versus full-HPSA.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 156 | 330
Towne et al. Multiple evidence-based disease prevention programs

FIGURE 4 | Distribution of evidence-based programs (EnhanceFitness, A Matter of Balance/Volunteer Lay Leader Model and the Chronic Disease
Self-Management Program) by county and rurality.

in program dissemination and participant reach in recent years. recommend some practical approaches for increasing the delivery
For example, CDSMP was delivered in 27 funded states during the of multiple programs in a given area. One approach may include
2006–2009 initiative, but it was delivered in 45 states, the District building an infrastructure that can support multiple EBP (58).
of Columbia, and Puerto Rico, reaching more than 100,000 older While the co-ordination of area agency on aging (AAA) funding
adults from 2010 to 2012 (20). AMOB/VLL was offered in 24 states varies by state (i.e., either centralized or decentralized infrastruc-
during 2006–2009, but is now available in over 30 states. Further, ture), these EBP may not be capable of reaching certain geographic
EF was delivered in 22 states and is now offered in over 25 states. locations. Moreover, even when communities want to offer these
We recognize that while more states are offering these EBP since programs, they may not have the program delivery infrastructure
the 2006–2009 initiative, there has been variability in their delivery, to serve the demands in their communities. As such, more research
with some counties increasing their offerings, and others cutting is needed to better understand why states and AAA elect to offer
back due to lack of funding. only certain programs, as well as the infrastructure-related chal-
Having multiple evidence-based interventions available to lenges associated with EBP delivery (especially as it pertains to
older adult populations provides an opportunity for better tailor- multi-program implementation). Further, future research might
ing to the unique needs of seniors with a variety of chronic con- explore why vulnerable adults only choose to participate in one
ditions. Such tailoring may be especially important for the most program despite the potential benefits of participating in multiple
vulnerable participants (52, 53). Yet, the study data showed that programs. Another approach to enhance program delivery capac-
the largest proportion of participants were located in areas where ity could be offering cross-training opportunities for different lay
only one program type was offered, regardless of area characteris- leaders and healthcare professionals so they can lead workshops
tics. The data also showed that multiple programs are typically less for multiple programs. Such an approach is being implemented
likely to be offered in areas serving the most vulnerable popula- by the Stanford Patient Education Research Center, which offers
tions (e.g., those living in low income or rural areas and in a HPSA). the suite of chronic disease self-management education programs
It was not surprising to find that these areas offered the least (Retrieved from http://patienteducation.stanford.edu/training/).
number of different programs, as this confirms prior research indi- Another approach might be to address and solve transportation
cating rural residents have lower access to healthcare services than needs to and from sites offering programs that are often an issue
their urban counterparts (54–56) where there are typically fewer in rural areas.
resources and greater distances to providers (57). Drawing from There were several limitations in the current study. First, this
our collective experience implementing and disseminating EBP, we study only examined the three most prevalent EBP being delivered

www.frontiersin.org April 2015 | Volume 2 | Article 156 | 331

Towne et al. Multiple evidence-based disease prevention programs

Table 4 | Distribution of counties by availability of evidence-based programs (CDSMP, AMOB/VLL, EF) by health professional shortage area
(HPSA), rurality, and poverty status in 2008.

CDSMP AMOB/VLL EF CDSMP and CDSMP AMOB/VLL

AMOB/VLL and EF and EF

n % n % n % n % n % n %

HPSA status
Full-HPSA 165a 39.4 90 35.6 32c 31.1 37 34.9 12a 41.4 12a 34.3
Partial-HPSA 149b 35.6 97b 38.3 51b, c 49.5 45b 42.5 16b 55.2 20b 57.1
Non-HPSA 105a,b 25.1 66b 26.1 20b 19.4 24b 22.6 1a,b 3.5 3a,b 8.6
Rurality
Rural 178* 42.5 81* 32.0 41* 39.8 27* 25.5 8* 27.6 13 37.1
Urban 241* 57.5 172* 68.0 62* 60.2 79* 74.5 21* 72.4 22 62.9
Poverty rate
Above median 136* 32.5 93* 36.8 45 43.7 31* 29.3 7* 24.1 14 40.0
At/below median 283* 67.5 160* 63.2 58 56.3 75* 70.8 22* 75.9 21 60.0
Total 419 253 103 106 29 35

*Significantly (p < 0.05) different by characteristic (e.g., rurality).

a
Significantly (p < 0.05) different non-HPSA versus full-HPSA.
b
Significantly (p < 0.05) different non-HPSA versus partial-HPSA.
c
Significantly (p < 0.05) different partial-HPSA versus full-HPSA.

through the AoA Evidence-Based Disease and Disability Preven- prevents analysis of trends over time; however, the goal of this
tion Initiative and the Atlantic Philanthropies from 2006 to 2009. study was to measure the overall reach among vulnerable adults,
While these data are now over 5 years old, no other national and service delivery characteristics during the initiative period.
database exists; hence, they are particularly powerful for illumi- Future studies should also identify strategies for identifying the
nating the two research questions posed in this study. Second, dissemination of multiple EBP throughout the US and their
only the three identified programs sponsored by this initiative interactive impacts on our aging population. At the current
were included, so that the study does not account for other EBP time, there is no mechanism for doing so. However, we should
that might also have been offered by different sponsors. Third, look toward a national inventory of EBP for seniors, poten-
the type of available data and amount of missing data is also a tially linked to healthcare utilization outcomes, or community
limitation to be acknowledged. In order to reach large numbers assessments that can track county level changes in health and
of participants being offered EBP through existing community functioning.
organizations, the amount of required data for this study was lim- Study findings demonstrate that individually these three EBP
ited to a few basic demographic and programmatic factors. Even have the capacity to appeal to vulnerable populations. Going
with this streamlined data collection protocol, there was substan- forward, the challenge is to create an efficient national infra-
tial missing data due to the inability of community providers to structure that encourages widespread adoption and bundling of
systematically collect and release all requested data (e.g., in some these programs for delivery in underserved populations and areas.
healthcare systems providers were not able to release informa- Systematical engagement and meaningful involvement of vul-
tion due to institutional review board restrictions). However, large nerable populations to fine tune outreach strategies, enhancing
amounts of missing administrative or programmatic data are not linkages with the healthcare system that includes advocating for
uncommon in evidence-based community interventions (59–62). the importance of evidence-based programing, building market-
In addition, analyses that include chronic conditions were lim- ing strategies and business models, and accelerating adaptation of
ited to data for individuals with one or more chronic conditions. evidence-based programing are approaches that program admin-
Cases reporting no chronic conditions were omitted because it istrators, policy makers, and funders can use to continue outreach
was impossible to determine whether these cases had no chronic to vulnerable older adults (63).
conditions or neglected to respond to these survey items (i.e., miss- New federal initiatives (e.g., Affordable Care Act) (62) are
ing data). Our analyses do not take into consideration the level encouraging the aging services network sector to work collabo-
of social support among participants; however, future analyses ratively with public health and medical care sectors and other
should include this as a possible factor associated with participant key stakeholders responsible for improving the health and func-
outcomes. tioning of our rapidly escalating population of older adults
Finally, we could not measure the actual penetration among with multiple chronic conditions. Growing and sustaining EBP
all possible participants for these EBP. Future research should in a diversity of delivery sites that attract a broader range of
examine the extent of reach among those potential partici- participants will be critical for achieving a greater population
pants for these EBP. The cross-sectional nature of the study health impact (16).

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 156 | 332
Towne et al. Multiple evidence-based disease prevention programs

ACKNOWLEDGMENTS 14. Bayliss EA, Steiner JF, Fernald DH, Crane LA, Main DS. Descriptions of barriers
Administration for Community Living is a primary CDSME work- to self-care by persons with comorbid chronic diseases. Ann Fam Med (2003)
1(1):15–21. doi:10.1370/afm.4
shop funder. Opinions expressed do not necessarily represent
15. U.S. Department of Health and Human Services. Healthy People 2020. (2014).
official Administration for Community Living policy. The national Available from: http://www.healthypeople.gov/2020/default.aspx
dissemination of these programs was supported as part of the 16. Smith ML, Belza B, Altpeter M, Ahn S, Dickerson JB, Ory MG. Disseminating an
United States Administration on Aging’s Evidence-Based Disease evidence-based disease self-management program for older Americans: impli-
and Disability Prevention program (EBDDP) initiative. In addi- cations for diversifying participant reach through delivery site adoption. In:
Maddock J, editor. Public Health: Social and Behavioral Health. Rijeka, Croatia:
tion to the AoA, this initiative was made possible by partnerships
InTech (2012).
with the Centers for Disease Control and Prevention, Agency for 17. Smith ML, Ory MG, Belza B, Altpeter M. Personal and delivery site char-
Healthcare Research and Quality (AHRQ), Centers for Medicare acteristics associated with intervention dosage in an evidence-based fall risk
and Medicaid Services (CMMS), Health Resources and Services reduction program for older adults. Transl Behav Med (2012) 2(2):188–98.
Administration (HRSA), Substance Abuse and Mental Health Ser- doi:10.1007/s13142-012-0133-8
18. Smith ML, Belza B, Braun KL, King S, Palmer RC, Sugihara NS, et al. National
vices Administration (SAMHSA), and over 30 private foundations. reach and dissemination of EnhanceFitness. Health Behav Policy Rev (2014)
We recognize the support from the AoA with assistance from the 1(2):150–60. doi:10.14485/HBPR.1.2.7
National Council on Aging for the evaluation of the EBDDP ini- 19. Ory MG, Smith ML. Exemplifying the evidence-based for health promotion
tiative under co-operative agreement number 90OP0001/01, as programs across populations and settings. Family Commun Health (2012)
well as the Healthy Aging Research Network, funded through the 35(3):188–91. doi:10.1097/FCH.0b013e318250bbff
20. Ory MG, Smith ML, Kulinski KP, Lorig K, Zenker W, Whitelaw N. Self-
Centers for Disease Control and Prevention’s Prevention Research management at the tipping point: reaching 100,000 Americans with evidence-
Centers. The findings and conclusions in this article are those of based programs. J Am Geriatr Soc (2013) 61(5):821–3. doi:10.1111/jgs.12239
the author(s) and do not necessarily represent the official posi- 21. Frank JC, Lau CA. Empowering Older People to Take More Control of their
tion of AoA, National Council on Aging, or any other agency. Health through Evidence-Based Prevention Programs: A Capping Report. National
Council on Aging (2013). Available from: http://www.ncoa.org/improve-health/
The authors specifically acknowledge the contributions of Nancy
center-for-healthy-aging/NCOA-Capping-Report_3-26-13_FINAL.pdf
Whitelaw, Don Grantt, Wendy Zenker, Kelly Horton, Meghan 22. Anderson G, Horvath J. The growing burden of chronic disease in America.
Thompson, Ashley Wilson, and Linnae Hutchison. Public Health Rep (2004) 119(3):263–70.
23. Lasser KE, Himmelstein DU, Woolhandler S. Access to care, health status,
and health disparities in the United States and Canada: results of a cross-
REFERENCES national population-based survey. Am J Public Health (2006) 96(7):1300–7.
1. Dall TM, Gallo PD, Chakrabarti R, West T, Semilla AP, Storm MV. An aging pop- doi:10.2105/AJPH.2004.059402
ulation and growing disease burden will require a large and specialized health 24. Miranda J, Duan N, Sherbourne C, Schoenbaum M, Lagomasino I, Jackson-
care workforce by 2025. Health Aff (2013) 32(11):2013–20. doi:10.1377/hlthaff. Triche M, et al. Improving care for minorities: can quality improvement inter-
2013.0714 ventions improve care and outcomes for depressed minorities? Results of a ran-
2. Anderson G. Chronic Care: Making the Case for Ongoing Care. Princeton, NJ: domized, controlled trial. Health Serv Res (2003) 38(2):613–30. doi:10.1111/
Robert Wood Johnson Foundation (2010). Available from: http://www.rwjf.org/ 1475-6773.00136
content/dam/farm/reports/reports/2010/rwjf54583 25. Glover S, Moore CG, Samuels ME, Probst JC. Disparities in access to care among
3. Yoshida S. A Global Report on Falls Prevention: Epidemiology of Falls. World rural working-age adults. J Rural Health (2004) 20(3):193–205. doi:10.1111/j.
Health Organization (WHO) (2013). Available from: http://www.who.int/ 1748-0361.2004.tb00029.x
ageing/projects/1.Epidemiology%20of%20falls%20in%20older%20age.pdf 26. Gamm L, Hutchison L, Dabney BJ, et al. Rural Healthy People 2010: A companion
4. Carroll NV, Slattum PW, Cox FM. The cost of falls among the community- document to Healthy People 2010. College Station, TX: The Texas A&M Univer-
dwelling elderly. J Manag Care Pharm (2005) 11(4):307–16. sity System Health Science Center, School of Rural Public Health, Southwest
5. Stevens JA, Corso PS, Finkelstein EA, Miller TR. The costs of fatal and non- Rural Health Research Center (2003).
fatal falls among older adults. Inj Prev (2006) 12(5):290–5. doi:10.1136/ip.2005. 27. Norton CH, McManus MA. Background tables on demographic character-
011015 istics, health status, and health services utilization. Health Serv Res (1989)
6. Roudsari BS, Ebel BE, Corso PS, Molinari NA, Koepsell TD. The acute med- 23(6):725–56.
ical care costs of fall-related injuries among the U.S. older adults. Injury (2005) 28. Jones CA, Parker TS, Ahearn M, Mishra AK, Variyam JN. Health Status and
36(11):1316–22. doi:10.1016/j.injury.2005.05.024 Health Care Access of Farm and Rural Populations. Washington, DC: U.S. Depart-
7. Englander F, Hodson TJ, Terregrossa RA. Economic dimensions of slip and fall ment of Agriculture, Economic Research Service (2009). p. 1–64.
injuries. J Forensic Sci (1996) 41(5):733–46. 29. Aday LA. At Risk in America: The Health and Health Care Needs of Vulnerable
8. Centers for Disease Control and Prevention. Falls Among Older Adults: An Populations in the United States. 2nd ed. San Francisco: Jossey-Bass (2001).
Overview. (2014). Available from: http://www.cdc.gov/HomeandRecreational 30. Mechanic D, Tanner J. Vulnerable people, groups, and populations: societal view.
Safety/Falls/adultfalls.html Health Aff (2007) 26(5):1220–30. doi:10.1377/hlthaff.26.5.1220
9. Centers for Medicare and Medicaid Services. Chronic Conditions Among 31. Lorig KR, Sobel DS, Stewart AL, Brown BW, Bandura A, Ritter P, et al. Evidence
Medicare Beneficiaries, Chartbook. 2012 ed. Baltimore, MD: CMS (2012). suggesting that a chronic disease self-management program can improve health
10. Machlin S, Cohen J, Beauregard K. Health Care Expenses for Adults with Chronic status while reducing hospitalization: a randomized trial. Med Care (1999)
Conditions, 2005. Rockville, MD: Agency for Healthcare Research and Quality 37(1):5–14. doi:10.1097/00005650-199901000-00003
(2008). 32. Lorig KR, Sobel DS, Ritter PL, Laurent D, Hobbs M. Effect of a self-management
11. Wagner EH, Glasgow RE, Davis C, Bonomi AE, Provost L, McCulloch D, et al. program on patients with chronic disease. Eff Clin Practi (2001) 4(6):
Quality improvement in chronic illness care: a collaborative approach. Jt Comm 256–62.
J Qual Improv (2001) 27(2):63–80. 33. Lorig KR, Ritter P, Stewart AL, Sobel DS, Brown BW Jr, Bandura A, et al.
12. Barr VJ, Robinson S, Marin-Link B, Underhill L, Dotts A, Ravensdale D, et al. Chronic disease self-management program: 2-year health status and health care
The expanded chronic care model: an integration of concepts and strategies utilization outcomes. Med Care (2001) 39(11):1217–23. doi:10.1097/00005650-
from population health promotion and the chronic care model. Hosp Q (2003) 200111000-00008
7(1):73–82. 34. Ory MG, Ahn S, Jiang L, Smith ML, Ritter PL, Whitelaw N, et al. Successes
13. Bodenheimer T, Lorig K, Holman H, Grumbach K. Patient self-management of of a national study of the chronic disease self-management program: meet-
chronic disease in primary care. JAMA (2002) 288(19):2469–75. doi:10.1001/ ing the triple aim of health care reform. Med Care (2013) 51(11):992–8.
jama.288.19.2469 doi:10.1097/MLR.0b013e3182a95dd1

www.frontiersin.org April 2015 | Volume 2 | Article 156 | 333

Towne et al. Multiple evidence-based disease prevention programs

35. National Council on Aging. Center for Healthy Aging: A Matter of Balance. (2014). 58. Scheirer MA. Is sustainability possible? A review and commentary on empirical
Available from: http://www.ncoa.org/improve-health/center-for-healthy-aging/ studies of program sustainability. Am J Eval (2005) 26(3):320–47. doi:10.1177/
a-matter-of-balance.html 1098214005278752
36. Tennstedt S, Howland J, Lachman M, Peterson E, Kasten L, Jette A. A Ran- 59. Edwards KE, Gibson N, Martin J, Mitchell S, Andersson N. Impact of
domized, controlled trial of a group intervention to reduce fear of falling and community-based interventions on condom use in the Tłicho region of
associated activity restriction in older adults. J Gerontol B Psychol Sci Soc Sci Northwest Territories, Canada. BMC Health Serv Res (2011) 11(Suppl 2):S9.
(1998) 53B(6):384–92. doi:10.1093/geronb/53B.6.P384 doi:10.1186/1472-6963-11-S2-S9
37. Healy TC, Peng C, Haynes P, McMahon E, Botler J, Gross L. The feasibility and 60. Mayworm AM, Sharkey JD. Gender-specific mental health outcomes of a
effectiveness of translating a matter of balance into a volunteer lay leader model. community-based delinquency intervention. J Juv Justice (2013) 3(1).
J Appl Gerontol (2008) 27(1):34–51. doi:10.1177/0733464807308620 61. Temel S, van Voorst SF, Jack BW, Denktas S, Steegers EA. Evidence-based
38. Smith ML, Hochhalter AK, Cheng Y, Wang S, Ory MG. Programmatic influ- preconceptional lifestyle interventions. Epidemiol Rev (2014) 36(1):19–30.
ences on outcomes of an evidence-based fall prevention program for older doi:10.1093/epirev/mxt003
adults: a translational assessment. Transl Behav Med (2011) 1(3):384–93. 62. Report to Congress: The Centers for Medicare & Medicaid Services’ Evaluation of
doi:10.1007/s13142-011-0058-7 Community-based Wellness and Prevention Programs under Section 4202 (b) of
39. National Council on Aging. Center for Healthy Aging. EnhanceFitness. (2014). the Affordable Care Act. (2014). Available from: http://www.projectenhance.org/
Available from: http://www.ncoa.org/improve-health/center-for-healthy-aging/ Portals/40/Docs/RTC_CommunityBasedWellnessPrograms.pdf
enhance-fi.html 63. Altpeter M, Schneider EC, Whitelaw N. Fostering engagement and indepen-
40. Wallace JL, Buchner DM, Grothaus L, Leveille S, Tyll L, LaCroix AZ, et al. dence: opportunities and challenges for an aging society. Health Educ Behav
Implementation and effectiveness of a community-based health promotion pro- (2014) (in press).
gram for older adults. J Gerontol A Biol Sci Med Sci (1998) 53A(4):M301–6.
doi:10.1093/gerona/53A.4.M301 Conflict of Interest Statement: Samuel D. Towne Jr. has no conflicts of interest
41. Ackermann RT, Williams B, Nguyen HQ, Berke EM, Maciejewski ML, LoGerfo to disclose. Neither himself nor this institution at any time received payment or
JP. Healthcare cost differences with participation in a community-based group services from a third party for any aspect of the submitted work. Dr. Towne has no
physical activity benefit for medicare managed care health plan members. J Am financial relationships with entities that could be perceived to influence, or that give
Geriatr Soc (2008) 56:1459–65. doi:10.1111/j.1532-5415.2008.01804.x the appearance of potentially influencing, what he wrote in the submitted work. Dr.
42. Area Health Resources Files (ARF). US Department of Health and Human Ser- Towne has no patents or copyrights to declare (whether pending, issued, licensed
vices, Health Resources and Services Administration, Bureau of Health Professions. and/or receiving royalties) relevant to the work. Dr. Towne has no other relationships
Rockville, MD (2013). or activities that readers could perceive to have influenced, or that give the appear-
43. Health Resources and Services Administration. Health Professional Shortage ance of potentially influencing, what he wrote in the submitted work. Matthew Lee
Areas (HPSAs). (2014). Available from: http://bhpr.hrsa.gov/shortage/hpsas/ Smith, PhD, MPH, CHES has no conflicts of interest to disclose. Neither myself
designationcriteria/index.html nor this institution at any time received payment or services from a third party for
44. USDA Economic Research Service. Urban Influence Codes. (2014). Available any aspect of the submitted work. Dr. Smith has no financial relationships with
from: http://www.ers.usda.gov/data-products/urban-influence-codes.aspx# entities that could be perceived to influence, or that give the appearance of poten-
.U-_DNPldWa8 tially influencing, what he wrote in the submitted work. Dr. Smith has no patents
45. ESRI Inc. ArcGIS: Release 10.2 by ESRI Inc. Redlands, CA: Environmental Sys- or copyrights to declare (whether pending, issued, licensed, and/or receiving roy-
tems Research Institute. Copyright © 1999–2013. alties) relevant to the work. Dr. Smith has no other relationships or activities that
46. SAS Institute Inc. SAS 9.4 by SAS Institute Inc. Cary, NC: Copyright © 2002–2013. readers could perceive to have influenced, or that give the appearance of potentially
47. Health Resources and Services Administration. Primary Medical Care HPSA Des- influencing, what he wrote in the submitted work. SangNam Ahn, PhD, MPSA has
ignation Overview. (2014). Available from: http://bhpr.hrsa.gov/shortage/hpsas/ no conflicts of interest to disclose. Basia Belza, PhD, RN, FAAN has no conflicts of
designationcriteria/primarycarehpsaoverview.html interests and he or his institution did not receive payment or services from a third
48. Health Resources and Services Administration. Primary Medical Care HPSA Des- party. Mary Altpeter, PhD, MSW, MPA has no conflicts of interest to disclose. Kristie
ignation Criteria: Part 1 – Geographic Areas. (2014). Available from: http://bhpr. Patton Kulinski, MSW has no conflicts of interest to disclose. Marcia G. Ory, PhD,
hrsa.gov/shortage/hpsas/designationcriteria/primarycarehpsacriteria.html MPH has no conflicts of interest to disclose.
49. United States Department of Agriculture Economic Research Service. Available
from: http://www.ers.usda.gov/data-products/urban-influence-codes.aspx# This paper is included in the Research Topic, “Evidence-Based Programming for Older
.UwahVfldV8E Adults.” This Research Topic received partial funding from multiple government and
50. Kulinski KP, Boutaugh M, Smith ML, Ory MG, Lorig K. Setting the stage: mea- private organizations/agencies; however, the views, findings, and conclusions in these
sure selection, coordination, and data collection for a national self-management articles are those of the authors and do not necessarily represent the official position
initiative. Front Public Health (2015) 2:206. doi:10.3389/fpubh.2014.00206 of these organizations/agencies. All papers published in the Research Topic received
51. US Census Bureau. State and County QuickFacts. USA People Quickfacts. (2014). peer review from members of the Frontiers in Public Health (Public Health Education
Available from: http://quickfacts.census.gov/qfd/states/00000.html and Promotion section) panel of Review Editors. Because this Research Topic repre-
52. Mier N, Ory MG, Toobert D, Smith ML, Osuna D, McKay J, et al. A qualitative sents work closely associated with a nationwide evidence-based movement in the US,
case study examining intervention tailoring for minorities. Am J Health Behav many of the authors and/or Review Editors may have worked together previously in
(2010) 34(6):822–32. some fashion. Review Editors were purposively selected based on their expertise with
53. Community Research Center for Senior Health in Partnership. Toolkit evaluation and/or evidence-based programming for older adults. Review Editors were
on Evidence-Based Programming for Seniors. Available from: http://www. independent of named authors on any given article published in this volume.
evidencetoprograms.com/
54. Council on Graduate Medical Education. Physician Distribution and Health Care Received: 11 July 2014; accepted: 08 September 2014; published online: 27 April 2015.
Challenges in Rural and Inner City Areas: Tenth Report to Congress and the Depart- Citation: Towne SD Jr., Smith ML, Ahn S, Altpeter M, Belza B, Kulinski KP and
ment of Health and Human Services Secretary. Rockville, MD: Health Resources Ory MG (2015) National dissemination of multiple evidence-based disease preven-
and Services Administration, US Dept of Health and Human Services (1998). tion programs: reach to vulnerable older adults. Front. Public Health 2:156. doi:
55. Knapp KK, Paavola FG, Maine LL, Sorofman B, Politzer RM. Availability of pri- 10.3389/fpubh.2014.00156
mary care providers and pharmacists in the United States. J Am Pharm Assoc This article was submitted to Public Health Education and Promotion, a section of the
(1999) 39(2):127–35. journal Frontiers in Public Health.
56. MacDowell M, Glasser M, Fitts M, Nielsen K, Hunsaker M. A national view of Copyright © 2015 Towne, Smith, Ahn, Altpeter, Belza, Kulinski and Ory. This is an
rural health workforce issues in the USA. Rural Remote Health (2010) 10(3):1531. open-access article distributed under the terms of the Creative Commons Attribution
57. Towne SD, Smith ML, Ory MG. Geographic variations in access and utilization License (CC BY). The use, distribution or reproduction in other forums is permitted,
of cancer screening services: examining disparities among American Indian and provided the original author(s) or licensor are credited and that the original publica-
Alaska native elders. Int J Health Geogr (2014) 13:18. doi:10.1186/1476-072X- tion in this journal is cited, in accordance with accepted academic practice. No use,
13-18 distribution or reproduction is permitted which does not comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 156 | 334
 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2015.00017

Healthcare providers’ perceptions and self-reported fall

prevention practices: findings from a large New York
health system
Matthew Lee Smith 1 *, Judy A. Stevens 2 , Heidi Ehrenreich 2 , Ashley D. Wilson 3 , Richard J. Schuster 4 ,
Colleen O’Brien Cherry 4 and Marcia G. Ory 3
1
Department of Health Promotion and Behavior, College of Public Health, The University of Georgia, Athens, GA, USA
2
National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, Atlanta, GA, USA
3
Department of Health Promotion and Community Health Sciences, Texas A&M Health Science Center School of Public Health, College Station, TX, USA
4
Center for Global Health, College of Public Health, The University of Georgia, Athens, GA, USA

Edited by: Among older adults, falls are the leading cause of injury-related deaths and emergency
Michal Grivna, United Arab Emirates
department visits, and the incidence of falls in the United States is rising as the number
University, United Arab Emirates
of older Americans increases. Research has shown that falls can be reduced by mod-
Reviewed by:
Laura Rudkin, University of Texas ifying fall-risk factors using multifactorial interventions implemented in clinical settings.
Medical Branch, USA However, the literature indicates that many providers feel that they do not know how to
Koustuv Dalal, Örebro University, conduct fall-risk assessments or do not have adequate knowledge about fall prevention. To
Sweden
help healthcare providers incorporate older adult fall prevention (i.e., falls risk assessment
*Correspondence:
and treatment) into their clinical practice, the Centers for Disease Control and Prevention’s
Matthew Lee Smith, Department of
Health Promotion and Behavior, (CDC) Injury Center has developed the Stopping Elderly Accidents, Deaths, and Injuries
College of Public Health, The (STEADI) tool kit. This study was conducted to identify the practice characteristics and
University of Georgia, #345E Wright providers’ beliefs, knowledge, and fall-related activities before they received training on
Hall, Health Science Campus, Athens,
how to use the STEADI tool kit. Data were collected as part of a larger State Fall Preven-
GA 30602, USA
e-mail: [email protected] tion Project funded by CDC’s Injury Center. Completed questionnaires were returned by
38 medical providers from 11 healthcare practices within a large New York health system.
Healthcare providers ranked falls as the lowest priority of five conditions, after diabetes,
cardiovascular disease, mental health, and musculoskeletal conditions. Less than 40% of
the providers asked most or all of their older patients if they had fallen during the past
12 months. Less than a quarter referred their older patients to physical therapists for bal-
ance or gait training, and <20% referred older patients to community-based fall prevention
programs. Less than 16% reported they conducted standardized functional assessments
with their older patients at least once a year. These results suggest that implementing the
STEADI tool kit in clinical settings could address knowledge gaps and provide the nec-
essary tools to help providers incorporate fall-risk assessment and treatment into clinical
practice.
Keywords: clinical practice, fall prevention, fall screening, intervention science

INTRODUCTION However, primary care physicians have been slow to put the
Falls are the leading cause of death and emergency department vis- AGS/BGS guideline into clinical practice because many feel that
its for injury among older adults (1), and the direct medical costs they do not know how to conduct fall-risk assessments or do
for these injuries are estimated to be more than $30 billion dollars not have adequate knowledge about fall prevention (11, 12). To
annually (2). Falls are caused by a number of risk factors usu- help healthcare providers incorporate older adult fall prevention
ally classified as either intrinsic (e.g., age, sex, chronic diseases, into their clinical practice, experts at Centers for Disease Control
medication side effects, gait and/or balance problems, muscle and Prevention’s (CDC) Injury Center developed the Stopping
weakness) or extrinsic (e.g., environmental factors such as uneven Elderly Accidents, Deaths, and Injuries (STEADI) tool kit. The
surfaces, poor lighting, and lack of railings and/or grab bars) (3– tool kit is based on the AGS/BGS clinical practice guideline (10),
7). It is expected that the incidence of falls and associated injuries applies concepts from Wagner’s Chronic Care Model (CCM) (13)
will continue to rise as the nation’s population of older adults to fall risk, and includes input from healthcare providers (14).
increases. However, fall risk can be reduced through multifactorial It contains basic information about falls, standardized gait and
interventions that are implemented in clinical settings (8, 9). balance assessment tests, case studies, and conversation starters.
The American and British Geriatrics Societies (AGS/BGS) In addition, there are educational handouts about fall prevention
have published a clinical practice guideline to reduce falls (10). specifically designed for older patients and their friends and family.

www.frontiersin.org April 2015 | Volume 3 | Article 17 | 335

Smith et al. Fall prevention practices

The contents of the STEADI tool kit and supplemental resources about your older patients, please rate the level of priority given
are available online (15). to conditions in your practice.” Then, given a list of eight fall-
Data for this study were collected as part of a larger 5-year risk factors, providers were asked to, “Rate the extent to which
project begun in 2011. This project funded three state health you believe the following items are fall-risk factors for your older
departments (Colorado, New York, Oregon) to integrate clini- patients.” Each factor was rated from 1 (low) to 10 (high). Finally,
cal and evidence-based community fall prevention programs in they were asked, “In the past month, approximately what percent
selected communities. This study describes the beliefs, knowl- of your older patients have you referred to attend community fall
edge, and fall-related activities of 38 healthcare providers from prevention programs?”
11 healthcare practices within a large New York health system, Given a list of 10 intervention activities, providers were asked
prior to receiving training about implementing the STEADI tool to report the proportion of older patients who received specific
kit. This community case study describes the current attributes, fall interventions at least once a year. Examples of intervention
perceptions, and self-reported practices of healthcare providers. activities included discussing prescription medications, discussing
The results underscore the need to enhance providers’ knowledge mobility aids, assessing visual acuity, and performing standardized
about fall prevention and for clinical resources to support falls physical functioning assessments. Responses were measured using
screening, assessment, and treatment. five-point Likert scales but, based on the frequency distribution,
these were collapsed into three categories: none, a few or some,
MATERIALS AND METHODS and most or all.
At part of a cooperative agreement with the CDC, the STEADI Other items asked of providers, but not presented in tables,
evaluation and implementation teams (led by the Texas A&M included the average amount of time (in minutes) they spent with
Health Science Center and The University of Georgia, respec- an older patient during a typical visit and the average amount of
tively) developed a provider training based on academic detailing time (in minutes) they spent assessing fall risk during a typical visit
called the Clinical Engagement and Education (CEE) session (16). with an older patient. Providers were also asked their level of agree-
These teams trained the state grantees to conduct CEE sessions ment with statements including, “My older patients are reluctant
and developed, tested, and refined the evaluation materials and to tell me they have fallen;” “It is important to perform a stan-
processes. dardized fall-risk assessment with older adults;”“Gait and balance
The purpose of the CEE session was to help clinicians find inno- tests are easy to perform;” and “I have adequate time during a clin-
vative ways to incorporate STEADI into their clinical practice (16). ical visit to assess fall risk among my older patients.” Responses
The 1-h session was led by a physician fall prevention “Champion” were measured using four-point Likert scales but, based on the
who had been identified and trained by the state grantee, and was frequency distribution, these were collapsed into two categories:
open to all clinicians and office personnel in the practice. These agree and disagree.
interactive sessions were designed to bring healthcare providers
and office staff together to discuss the burden of older adult falls STATISTICAL ANALYSES
and to foster collective decisions about fall prevention activities Given the limited number of participants in this study, data are
that they could implement during clinical visits with older adult described but no tests for statistical significance were performed.
patients (16). Some data are presented in tabular form and others are described
DATA COLLECTION in the text.
Clinical Engagement and Education session data about the char-
acteristics of the practice, provider characteristics, and provider RESULTS
beliefs, knowledge, and fall-related activities were collected from PRACTICE CHARACTERISTICS
two sources. First, office personnel completed a registration form Between September 2012 and June 2013, 11 New York based prac-
after the practice agreed to participate in the CEE session. This tices within United Healthcare, a managed healthcare company,
form provided general information about the healthcare group hosted 11 CEE sessions. These practices had existed for an average
(e.g., number of years the practice has been in business, the number of 20 years (range: 5–30 years). Each practice served an average
of employees, size of the patient base). of 6,365 patients (range: 320–12,000 patients) and, on average,
Second, each CEE session participant was asked to complete a 43% of these patients (range: 20–70%) were aged 65 years or
two-page questionnaire at the beginning of the CEE Session. The older. Each practice employed an average of 15 medical person-
35-item questionnaire took approximately 15 min to complete and nel (range: 4–30 employees) that included between one and six
asked for the participant’s characteristics (i.e., job title, gender), physicians.
opinions about fall-risk factors, practice priorities, and activities
conducted during clinical visits with older patients. Responses PROVIDER CHARACTERISTICS
consisted of Likert scales and closed-response formats. Institu- Forty-nine persons attended the CEE sessions. For this study, office
tional Review Board approval was obtained from Texas A&M personnel (n = 5) and those with missing socio-demographic data
University to conduct descriptive analyses using de-identified data. (n = 6) were excluded. Therefore, data are presented for 38 medical
providers. Of these, 34% were nurses, 26% physicians, 18% nurse
MEASURES practitioners, 8% physician assistants, 8% medical assistants, and
Providers were asked to rate each of five health conditions from 3% specialty care providers. The median age was 38 years (range:
1 (low) to 10 (high) in response to the question, “When thinking 23–69 years), and 84% were female.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 17 | 336
Smith et al. Fall prevention practices

PROVIDER BELIEFS AND KNOWLEDGE Thirty-one providers reported that they referred on average
Table 1 shows the participants’ level of priority given to specific 20% (±18.5%) of their older adult patients to community fall
health conditions and beliefs about fall-risk factors among older prevention programs (range: 10–100%). Similarly, 32 providers
adults. Of the five health conditions, diabetes received the high- reported that they referred on average 22% (±18.8%) of their
est average score (8.4) while falls received the lowest (7.1). Of older patients to physical therapy for gait and/or balance retraining
eight fall-risk factors, a history of falling received the highest aver- (range: 10–100%).
age score (8.1). Postural hypotension received the lowest average
score (6.1). DISCUSSION
This study examined the beliefs, knowledge, and fall-related activ-
PROVIDER FALL-RELATED ACTIVITIES ities conducted among 38 healthcare providers. These data, col-
Providers reported that a typical office visit with an older patient lected at the beginning of the CEE session, showed that the
lasted on average 20.7 (±9.9) minutes (range: 0–60 min). The providers considered all five specified health conditions were high
time spent assessing fall-risk factors averaged 3.8 (±2.5) minutes priority. However, falls were considered a lower priority than
(range: 0–10 min). Approximately, 66% of respondents agreed or chronic conditions such as diabetes and cardiovascular disease.
strongly agreed with the statement, “I have adequate time during Despite clinical guidelines (10), few providers routinely asked
a clinical visit to assess fall risk among my older patients.” their older patients if they had fallen in the past year. This is espe-
Table 2 shows the proportion of respondents who delivered cially troubling since providers reported that their patients were
specific fall interventions to their older patients at least once a reluctant to tell them that they had fallen. These data further indi-
year. Over 81% of providers discussed details about prescribed cated that few providers actually conducted standardized tests to
medications with most or all of their older patients. About 47% assess gait and balance, although these tests are seen as both impor-
conducted a cognitive screening with most or all of their older tant and easy to perform. The low assessment rate by providers was
patients, and 37% asked most or all of their older patients about partially counterbalanced by patient referrals to physical therapy
falls during the past 12 months. to address gait or balance problems.
All providers reported that it was important to perform a stan- Prior research suggests that primary care providers feel that
dardized fall-risk assessment and said gait and balance tests were they do not know how to conduct fall-risk assessments and this
easy to perform. Just over one-third of the providers routinely study found that providers were not conducting multifactorial
asked their older patients if they had fallen in the past year. Yet, risk assessments on every patient (11, 12, 17). These are missed
about 61% of providers agreed or strongly agreed with the state- opportunities for prevention that are likely to result in higher
ment, “My older patients are reluctant to tell me they have fallen.” fall rates. An important next step is to make fall prevention a
As shown in Table 2, fewer than 16% reported that they con- routine part of clinical care. This requires educating providers
ducted either the Timed Up and Go test, 30-s Chair Stand, or about how to conduct fall-risk assessments and providing them
4-Stage Balance Test with most or all of their older patients at with the necessary tools to streamline the process. Promising
least once a year. approaches include educating providers about the STEADI tool

Table 1 | Healthcare providers’ priorities and beliefs about the importance of issues facing older adult patients.

N Median Mean SD Range

Minimum Maximum

Priority given to health conditionsa

Diabetes 37 9 8.35 1.69 3 10
Cardiovascular disease, including stroke 37 8 8.08 1.99 3 10
Mental health, including depression 34 8 7.44 2.22 3 10
Musculoskeletal conditions 37 8 7.35 1.93 3 10
Falls 37 7 7.05 2.15 3 10
Beliefs about fall-risk factors for older patientsa
History of falling 38 9 8.11 2.35 3 10
Balance issues 38 9 7.95 2.27 3 10
Gait issues 37 9 7.68 2.46 2 10
Environmental issues within the home 38 8 7.16 2.09 3 10
Medication issues 38 8 7.13 2.47 2 10
Neurological issues 38 7 6.68 2.35 2 10
Vision issues 35 7 6.66 2.35 2 10
Postural hypotension 36 6 6.11 2.51 2 10

a
All items measured on a scale from 1 = low to 10 = high.

www.frontiersin.org April 2015 | Volume 3 | Article 17 | 337

Smith et al. Fall prevention practices

Table 2 | Proportion of older patients for whom activities are creating a software module to integrate STEADI’s fall prevention
performed at least once a year. processes into EHR.
Lastly, the study relied on self-reported estimates of fall preven-
N None (%) A few/ Most/ tion activities that could not be confirmed by objective measures
some (%) all (%) such as medical chart reviews. These estimates may not accu-
Discuss details about their 37 0.0 18.9 81.1 rately reflect the true frequencies of these activities in primary
prescribed medications (e.g., care settings.
number, type, dose, side effects)
Conduct a cognitive screening 34 11.8 41.2 47.1 CONCLUSION
Discuss their use of mobility aids 37 5.4 56.8 37.8 This study found that most healthcare providers did not consider
Collect fall history over the past 38 18.4 44.7 36.8 falls as high a priority as other chronic conditions among older
12 months patients, and did not routinely assess and address these patients’
Educate about their specific fall-risk 36 16.7 52.8 30.6 fall-risk factors. The STEADI tool kit may be a valuable resource to
factors help providers incorporate fall-risk assessment, treatment, and/or
Follow-up with patients who are at 36 16.7 55.6 27.8 referral into clinical practice. However, providers must first be con-
risk for falling within 30 days of their vinced that falls are a priority issue among their older patients, and
clinical visit devote as much, or more, time to assessing falls risks and educat-
Assess their visual acuity 35 5.7 77.1 17.1 ing patients about appropriate programs to reduce fall risks. Future
Conduct the Timed Up and Go test 32 53.1 31.3 15.6 studies will focus on educating providers about the STEADI tool
Conduct the 30-s Chair Stand test 34 58.8 32.4 8.8 kit, their adoption of STEADI, and STEADI’s impact on fall-risk
Conduct the 4-Stage Balance test 35 65.7 25.7 8.6 screening, assessment, and treatment.

ACKNOWLEDGMENTS
kit as well as providing them with additional resources such as
The authors would like to thank Margaret Kaniewski, MPH, Pub-
online provider education and clinical decision support modules
lic Health Advisor at the CDC, and the staff at the New York State
that are integrated into the provider’s electronic health records
Department of Health, Bureau of Occupational Health and Injury
(EHR) system.
Prevention, for their significant contributions to this work. We
thank Dr. Frank Floyd, and UHS Physician Practices for collect-
LIMITATIONS
ing the study data. Additional thanks are given to Natalie Martin
A limitation of this study is that provider data were available
for her contribution to the STEADI implementation and training
for only 38 healthcare providers from one healthcare organiza-
process. This research was supported by the CDC, National Center
tion in one state, so findings must be considered preliminary.
for Injury Prevention and Control, under Cooperative Agreement
While study data were obtained from a diverse set of healthcare
Number 1U48 DP001924 with the Texas A&M Health Science Cen-
providers, the small number of respondents made it impossible
ter School of Rural Public Health Center for Community Health
to examine how the knowledge, beliefs, and activities differed by
Development, and by an appointment to the Research Participa-
provider or practice type. Further investigation is warranted to
tion Program at the CDC, administered by the Oak Ridge Institute
assess such differences. Additionally, these data were collected pre-
for Science and Education, through an interagency agreement
intervention, before the healthcare providers were introduced to
between the U.S. Department of Energy and CDC. Disclaimer: The
the tool kit. Because insufficient follow-up data were collected
findings and conclusions in this report are those of the authors and
post-intervention, changes in healthcare providers’ beliefs and
do not necessarily represent the official position of the Centers for
behaviors could not be assessed. Further investigation is war-
Disease Control and Prevention.
ranted to examine the impact of the STEADI tool kit on healthcare
providers’ perceptions and clinical practice.
The approach used for recruiting healthcare practices to receive REFERENCES
1. Web-Based Injury Statistics Query and Reporting System (WISQARS) [Database
training in using the STEADI tool kit may have limited partic-
on the Internet]. National Center for Injury Prevention and Control (2012).
ipation; it may have selected those participants who were espe- Available from: http://www.cdc.gov/injury/wisqars/index.html
cially interested in fall prevention. This would suggest that the 2. Stevens JA, Corso PS, Finkelstein EA, Miller TR. The costs of fatal and non-
frequency of fall-risk assessments (e.g., collecting fall history, con- fatal falls among older adults. Inj Prev (2006) 12(5):290–5. doi:10.1136/ip.2005.
ducting standardized gait, and balance tests) actually might be 011015
3. Tinetti ME, Speechley M, Ginter SF. Risk factors for falls among elderly persons
lower among healthcare providers. living in the community. N Engl J Med (1988) 319(26):1701–7.
Another limitation was associated with hosting CEE sessions 4. Tinetti M, Doucette J, Claus E, Marottoli R. Risk factors for serious injury during
in healthcare practices. State health department grantees were falls by older persons in the community. J Am Geriatr Soc (1995) 43(11):1214–21.
required to first identify and then train a highly motivated provider 5. Tinetti ME, McAvay G, Claus E. Does multiple risk factor reduction explain
Champion. This was a difficult and labor-intensive process because the reduction in fall rate in the Yale FICSIT trial? Am J Epidemiol (1996)
144(4):389–99. doi:10.1093/oxfordjournals.aje.a008940
the grantees had to first establish new partnerships with healthcare 6. Kannus P, Sievänen H, Palvanen M, Järvinen T, Parkkari J. Prevention of
provider groups. Future efforts will include developing an online falls and consequent injuries in elderly people. Lancet (2005) 366:1885–93.
training for STEADI with continuing education (CE) credits, and doi:10.1016/S0140-6736(05)67604-0

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 17 | 338
Smith et al. Fall prevention practices

7. Stevens M, Holman CAJ, Bennett N. Preventing falls in older people: impact of by community physicians. J Am Geriatr Soc (2004) 52(9):1527–31. doi:10.1111/
an intervention to reduce environmental hazards in the home. J Am Geriatr Soc j.1532-5415.2004.52417.x
(2001) 49(11):1442–7. doi:10.1046/j.1532-5415.2001.4911235.x
8. Haines TP, Bennell KL, Osborne RH, Hill KD. Effectiveness of targeted falls pre-
vention programme in subacute hospital setting: randomised controlled trial. Conflict of Interest Statement: The authors declare that the research was conducted
BMJ (2004) 328(7441):676. doi:10.1136/bmj.328.7441.676 in the absence of any commercial or financial relationships that could be construed
9. Cameron ID, Murray GR, Gillespie LD, Robertson MC, Hill KD, Cumming as a potential conflict of interest.
RG, et al. Interventions for preventing falls in older people in nursing care
facilities and hospitals. Cochrane Database Syst Rev (2010) 1(1):CD005465. This paper is included in the Research Topic, “Evidence-Based Programming for Older
doi:10.1002/14651858.CD005465.pub2 Adults.” This Research Topic received partial funding from multiple government and
10. American Geriatrics Society, British Geriatrics Society. AGS/BGS Clinical Prac- private organizations/agencies; however, the views, findings, and conclusions in these
tice Guideline: Prevention of Falls in Older Persons. New York, NY: American articles are those of the authors and do not necessarily represent the official position
Geriatrics Society (2010). of these organizations/agencies. All papers published in the Research Topic received
11. Chou WC, Tinetti ME, King MB, Irwin K, Fortinsky RH. Perceptions of peer review from members of the Frontiers in Public Health (Public Health Education
physicians on the barriers and facilitators to integrating fall risk evalua- and Promotion section) panel of Review Editors. Because this Research Topic repre-
tion and management into practice. J Gen Intern Med (2006) 21(2):117–22. sents work closely associated with a nationwide evidence-based movement in the US,
doi:10.1007/s11606-006-0244-3 many of the authors and/or Review Editors may have worked together previously in
12. Fortinsky RH, Iannuzzi-Sucich M, Baker DI, Gottschalk M, King MB, Brown some fashion. Review Editors were purposively selected based on their expertise with
CJ, et al. Fall-risk assessment and management in clinical practice: views from evaluation and/or evidence-based programming for older adults. Review Editors were
healthcare providers. J Am Geriatr Soc (2004) 52(9):1522–6. doi:10.1111/j.1532- independent of named authors on any given article published in this volume.
5415.2004.52416.x
13. Wagner EH. Chronic disease management: what will it take to improve care for
chronic illness? Eff Clin Pract (1998) 1:2–4. Received: 27 August 2014; accepted: 15 January 2015; published online: 27 April 2015.
14. Stevens JA, Phelan EA. Development of STEADI: a fall prevention resource for Citation: Smith ML, Stevens JA, Ehrenreich H, Wilson AD, Schuster RJ, Cherry CO
health care providers. Health Promot Pract (2012) 14(5):706–14. doi:10.1177/ and Ory MG (2015) Healthcare providers’ perceptions and self-reported fall prevention
1524839912463576 practices: findings from a large New York health system. Front. Public Health 3:17. doi:
15. Centers for Disease Control and Prevention (CDC). STEADI (Stopping 10.3389/fpubh.2015.00017
Elderly Accidents, Deaths & Injuries) Tool Kit for Health Care Providers. This article was submitted to Public Health Education and Promotion, a section of the
Atlanta, GA (2013) [Cited 2014 July 14]. Available from: http://www.cdc.gov/ journal Frontiers in Public Health.
homeandrecreationalsafety/Falls/steadi/index.html?s_cid=tw_injdir154 Copyright © 2015 Smith, Stevens, Ehrenreich, Wilson, Schuster, Cherry and Ory.
16. Schuster RJ, Cherry COB, Smith ML. The clinician engagement and education This is an open-access article distributed under the terms of the Creative Commons
session modernizing “academic detailing. Am J Med Qual (2013) 28(6):533–5. Attribution License (CC BY). The use, distribution or reproduction in other forums is
doi:10.1177/1062860613491976 permitted, provided the original author(s) or licensor are credited and that the original
17. Rubenstein LZ, Solomon DH, Roth CP, Young RT, Shekelle PG, Chang JT, publication in this journal is cited, in accordance with accepted academic practice. No
et al. Detection and management of falls and instability in vulnerable elders use, distribution or reproduction is permitted which does not comply with these terms.

www.frontiersin.org April 2015 | Volume 3 | Article 17 | 339

 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00235

CDC andYMCA: a promising partnership for delivering fall

prevention programing
Heidi Ehrenreich 1 *, Maureen Pike 2 , Katherine Hohman 2 , Margaret Kaniewski 1 , Matt Longjohn 2 ,
Gaya Myers 1 and Robin Lee 1
1
Division of Unintentional Injury Prevention, National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, Atlanta, GA, USA
2
YMCA of the USA, Chicago, IL, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: fall prevention, Tai Chi, community-based programs, partnerships, dissemination

OLDER ADULT FALLS not always reach the intended audience (5). disseminate EnhanceFitness, an evidence-
Falls threaten the independence of adults This is largely due to the lack of local infra- based exercise program for older adults.
aged 65 years and older. In the U.S., one in structure needed to deliver community- In the program, trained YMCA fitness staff
three older adults fall annually, causing sig- based programs (6). Developing and main- and volunteers lead a comprehensive exer-
nificant disability and reduced quality of taining the necessary organizational infra- cise routine shown to increase physical,
life (1). The high prevalence of falls, cou- structure can be time consuming and mental, and social functioning in older
pled with more than $30 billion in direct costly, limiting program sustainability. A adults (11). In the first year of this part-
medical costs (2) has created a critical need recent systematic review revealed that sta- nership, the program reached 2,000 older
for effective older adult fall prevention pro- ble financial program support, integrated adults at 41 YMCAs. In another initia-
grams. As the nation’s public health agency, programing, and the ability to make pro- tive, 157 YMCAs partnered with the LIVE-
the Centers for Disease Control and Pre- gram adaptations were major factors that STRONG Foundation for extensive train-
vention (CDC) is committed to identifying sustained successful fall prevention pro- ing as hubs of support for cancer survivors.
ways to reduce the burden of older adult grams (7). To date, over 21,000 survivors have been
falls. In this commentary, we describe a YMCA’s robust infrastructure for pro- served by this initiative. Finally, CDC part-
promising approach to reach older adults gram delivery, large membership base, nered with Y-USA to reach over 19,000
with effective interventions by partnering and local credibility offer strong potential participants at 128 YMCAs to expand its
with the YMCA to deliver community- for building successful and wide-reaching evidence-based National Diabetes Preven-
based fall prevention programs. public health programing. The marketing tion Program to participating communi-
literature supports the use of these types ties (12). As part of this CDC-led pro-
IMPLEMENTATION OF EFFECTIVE of distribution channels to improve the gram, YMCAs have trained their wellness
COMMUNITY FALL PREVENTION adoption and implementation of evidence- instructors as “lifestyle coaches” to imple-
PROGRAMS based programs (8). For these reasons, ment lifestyle-change programs focused on
Centers for Disease Control and Preven- CDC is pursuing partnerships with organi- participants losing weight, being physically
tion has produced several guides dedicated zations such as YMCAs to help implement active, and coping with stress.
to fall prevention programing and deliv- effective fall prevention programs.
ery. The CDC Compendium of Effective Fall YMCA ADAPTS CDC’S OLDER ADULT
Interventions is intended to help public PARTNERING WITH THE YMCA TO FALL PREVENTION PROGRAM
health practitioners use the best scientific SCALE-UP SUCCESSFUL PUBLIC Motivated by the success of the National
evidence to effectively address falls among HEALTH PROGRAMS Diabetes Prevention Program, CDC initi-
older adults in the community (3). YMCAs are independent but federated ated a similar strategy to implement an
The Compendium describes 22 scien- organizations working to spread health evidence-based older adult fall preven-
tifically tested interventions for use by and wellness in their communities. YMCAs tion program using the YMCA infrastruc-
public health practitioners, aging service offer classes for all ages, all skill levels, and ture. With funding from the CDC, Y-
providers, and others. In addition, CDC all interests. As a national resource cen- USA licensed the rights to the Tai Chi
developed a how-to guide for community- ter, YMCA of the USA (Y-USA) supports Moving for Better Balance fall prevention
based organizations seeking to develop, over 2,600 YMCAs located across 10,000 program (13) and adapted the program
implement, and evaluate their own effec- U.S. neighborhoods and with 20.6 million to fit the YMCA training system. Y-USA
tive fall prevention programs (4). members (http://www.ymca.net). reintroduced the program under the name,
While federal and state public health Y-USA has a history of collaborating on Y-Moving for Better Balance (Y-MFBB) and
agencies have used CDC’s guide to national public health initiatives (9, 10). Y- contracted with the program’s creator to
implement fall prevention programs, the USA partnered with CDC and the National train YMCA Faculty Trainers as Y-MFBB
information about effective programs does Association of Chronic Disease Directors to instructors. As of September 2013, 287

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 235 | 340
Ehrenreich et al. Promising CDC and YMCA partnership

interagency agreement between the U.S.

Department of Energy and CDC. The
findings and conclusions in this report
are those of the authors and do not
necessarily represent the official position
of the Centers for Disease Control and
Prevention.

REFERENCES
1. Tromp AM, Pluijm SMF, Smit JH, Deeg DJH,
Bouter LM, Lips P. Fall-risk screening test:
a prospective study on predictors for falls
in community-dwelling elderly. J Clin Epi-
demiol (2001) 54(8):837–44. doi:10.1016/S0895-
4356(01)00349-3
2. Stevens JA, Corso PS, Finkelstein EA, Miller TR.
The costs of fatal and non-fatal falls among older
adults. Inj Prev (2006) 12(5):290–5. doi:10.1136/
FIGURE 1 | Y-MFBB Class in Broome County, New York. ip.2005.011015
3. Stevens JA. Compendium of Effective Fall Interven-
tions: What Works for Community-Dwelling Older
Adults. Atlanta, GA: Centers for Disease Control
Y-MFBB instructors have been trained. programing at their sites. CDC is currently
and Prevention, National Center for Injury Pre-
To encourage implementation of Y-MFBB supporting the development of a national vention (2010).
locally, Y-USA awarded 131 YMCAs grants Y-MFBB rollout plan based on further 4. Centers for Disease Control and Prevention
of $1000 each to hold instructor trainings research into program and implementation (CDC). National Center for Injury Prevention
and initiate Y-MFBB classes. effectiveness. and Control. Preventing Falls: How to Develop
Community-based Fall Prevention Programs for
In fall 2013, the Y-USA interviewed staff Older Adults. Atlanta, GA: Centers for Disease Con-
directors from 8 of the 75 YMCAs offering NEXT STEPS trol and Prevention (2008).
Y-MFBB in the last year to gather lessons While CDC and state public health agencies 5. Durlak J, DuPre E. Implementation matters: a
learned about local Y-MFBB implemen- have the tools to assess and address public review of research on the influence of imple-
tation. After approximately 17 months, health problems, community-based orga- mentation on program outcomes and the factors
affecting implementation. Am J Community Psy-
these 8 YMCAs had reached 706 partic- nizations are often tasked with delivering chol (2008) 41(3–4):327–50. doi:10.1007/s10464-
ipants. Most participants were women, programs. The YMCA has been an impor- 008-9165-0
aged 65 years and older, and YMCA mem- tant partner for scaling up the CDC fall 6. Fixsen D, Scott V, Blase K, Naoom S, Wagar L.
bers (see Figure 1 depicting Y-MFBB par- prevention program to a wider and more When evidence is not enough: the challenge of
ticipants). Participants reported discover- diverse audience. This example of lever- implementing fall prevention strategies. J Safety
Res (2011) 42(6):419–22. doi:10.1016/j.jsr.2011.
ing the Y-MFBB program through YMCA aging partnerships with organizations that 10.002
advertising, local community organiza- already have a robust infrastructure in place 7. Lovarini M, Clemson L, Dean C. Sustainability
tions, and from medical professionals. Cost for large-scale program delivery is critical of community-based fall prevention programs: a
to participate ranged from no additional for population-level gains. CDC will con- systematic review. J Safety Res (2013) 47(0):9–17.
charge for members to $70 per 12-week tinue to work with organizations such as doi:10.1016/j.jsr.2013.08.004
8. Maibach E, Van Duyn MAS, Bloodgood B. A
session, or approximately $3 per class, the YMCA to increase the availability of marketing perspective on disseminating evidence-
for a YMCA offering the program at an Y-MFBB and other evidence-based fall pre- based approaches to disease prevention and health
off-site facility. Four YMCAs reported an vention programs to reduce fall risk among promotion. Prev Chronic Dis (2006) 3(3):1–11.
implementation cost of $386 per 12-week older adults. doi:10.5888/pcd9.110081
session, based mainly on instructor time. 9. Adamson K, Shepard D, Easton A, Jones ES.
The YMCA/steps community collaboratives, 2004-
Overall, staff directors determined that ACKNOWLEDGMENTS
2008. Prev Chronic Dis (2009) 6(3):1–5.
fall prevention programing fit well with We sincerely thank the Program Direc- 10. Ackermann RT, Marrero DG. Adapting the dia-
YMCA’s health and wellness mission, tors from the local YMCAs who engaged betes prevention program lifestyle intervention for
citing two main reasons for offering in in-depth interviews to describe their delivery in the community: the YMCA model.
Y-MFBB: (1) the growing number of older offerings of the Y-Moving for Better Bal- Diabetes Educ (2007) 33(1):69–78. doi:10.1177/
0145721706297743
adult members; and (2) the high incidence ance program. We also thank Dr. Erin 11. Belza B, Shumway Cook A, Phelan E, Williams
of falls among them. However, Y-MFBB Parker for her thoughtful review and sug- B, Snyder S. The effects of a community-
instructors also need the opportunity to gestions. This work was supported in based exercise program on function and health
check program fidelity and offer pro- part by an appointment to the Research in older adults: the EnhanceFitness program.
gressively more challenging classes. With Participation Program at the Centers for J Appl Gerontol (2006) 25(4):291–306. doi:10.
1177/0733464806290934
more Y-MFBB instructors and training, the Disease Control and Prevention (CDC) 12. Murphy D, Chapel T, Clark C. Moving diabetes
directors are considering placing Y-MFBB administered by the Oak Ridge Institute care from science to practice: the evolution of the
into a larger portfolio of falls prevention for Science and Education through an national diabetes prevention and control program.

www.frontiersin.org April 2015 | Volume 2 | Article 235 | 341

Ehrenreich et al. Promising CDC and YMCA partnership

Ann Intern Med (2004) 140(11):978–84. doi:10. findings, and conclusions in these articles are those of Citation: Ehrenreich H, Pike M, Hohman K, Kaniewski
7326/0003-4819-140-11-200406010-00040 the authors and do not necessarily represent the official M, Longjohn M, Myers G and Lee R (2015) CDC and
13. Li F, Harmer P, Mack KA, Sleet D, Fisher KJ, position of these organizations/agencies. All papers pub- YMCA: a promising partnership for delivering fall pre-
Kohn MA, et al. Tai Chi: moving for better bal- lished in the Research Topic received peer review from vention programing. Front. Public Health 2:235. doi:
ance: development of a community-based falls pre- members of the Frontiers in Public Health (Public Health 10.3389/fpubh.2014.00235
vention program. J Phys Act Health (2008) 5(3): Education and Promotion section) panel of Review Edi- This article was submitted to Public Health Education
445–55. tors. Because this Research Topic represents work closely and Promotion, a section of the journal Frontiers in
associated with a nationwide evidence-based movement Public Health.
in the US, many of the authors and/or Review Editors Copyright © 2015 Ehrenreich, Pike, Hohman,
Conflict of Interest Statement: The authors declare may have worked together previously in some fash- Kaniewski, Longjohn, Myers and Lee. This is an open-
that the research was conducted in the absence of any ion. Review Editors were purposively selected based on access article distributed under the terms of the Creative
commercial or financial relationships that could be their expertise with evaluation and/or evidence-based Commons Attribution License (CC BY). The use, dis-
construed as a potential conflict of interest. programming for older adults. Review Editors were inde- tribution or reproduction in other forums is permitted,
pendent of named authors on any given article published provided the original author(s) or licensor are credited
This paper is included in the Research Topic, “Evidence- in this volume. and that the original publication in this journal is cited,
Based Programming for Older Adults.” This Research in accordance with accepted academic practice. No use,
Topic received partial funding from multiple government Received: 16 June 2014; accepted: 28 October 2014; distribution or reproduction is permitted which does not
and private organizations/agencies; however, the views, published online: 27 April 2015. comply with these terms.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 235 | 342
 OPINION ARTICLE
PUBLIC HEALTH
published: 09 January 2015
doi: 10.3389/fpubh.2014.00287

A missing piece in the infrastructure to promote healthy

aging programs: education and work force development
Janet Christine Frank *
University of California at Los Angeles (UCLA) Center for Health Policy Research, The University of California at Los Angeles (UCLA) Fielding School of Public Health,
Los Angeles, CA, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: gerontology education, career technical education, workforce development, competency-based education, faculty development, educational
evaluation

There are compelling data available, both “aging” of the aging services workforce, concerns for major service delivery systems
for the rapidly expanding older adult pop- with impending mass retirement of long- and higher education. A number of impor-
ulation, and for the value of evidence-based time leaders and service providers. In fact, tant recommendations were made in this
health promotion and disease management the California labor force study noted report. Unfortunately, slow and incremen-
programs (EBHPs). The systems approach above documented that 52% of the aging tal progress has been achieved in address-
to transforming our aging services deliv- services workforce is age 50 or over (1). ing them, especially in the area of public
ery system has been brilliant, but there is And, the national study of aging ser- health and aging. The documented gaps
an important system missing – our educa- vices personnel echoed concerns about in preparedness from the 1995 report were
tional system. Building the infrastructure the “aging” of the aging services work- resounded in the 2008 Institute of Medi-
to create embedded and accessible healthy force, noting that about 20% of current cine Report,“Retooling for An Aging Amer-
aging programs must take into account staff is projected to retire within the next ica”(5). The 2008 IOM Report summarized
workforce preparation. Most of the peo- 5 years (by 2015) (2, 3). This means that critical workforce preparedness deficits and
ple currently working in the aging services workers nearing retirement age may have called for increased competencies in every
delivery system are doing so without the been entering college in the late 1960s type of health and social services personnel
benefit of any formal education or an orga- and early 1970s. The first gerontology edu- at every level.
nized course of study about older adults cation programs at colleges began about The interesting thing is that geriatric
and aging services. For the state of Califor- 1972, and there were very few available and gerontologic competencies do exist for
nia, 61% of aging services agencies reported until the 1980s. These anticipated high many health and social service disciplines,
zero current staff with formal gerontology rates of retirement will soon lead to rapid including medicine, nursing, social work,
education, defined as having had even one turnover and the opportunity for new per- pharmacy, gerontology, and others (6–11).
academic course in aging content (1). In sonnel replacements, perhaps with geron- Public health currently does not have com-
a national study, less than half (46.6%) of tology education backgrounds, at all levels. petencies specific to addressing the needs
responding Area Agencies on Aging (AAAs) However, this “opportunity” assumes that of older adults (12).
had at least one staff with either a certificate current health and aging services leader- In spite of the call for action, the exis-
or degree in gerontology and almost 27% ship often without formal “aging” educa- tence of professional competencies, and
have an Evidence-Based Program (EBP) tion will deem it a priority to hire avail- the estimated 600 current gerontology pro-
Coordinator position (2, 3). There was no able individuals with gerontology degrees grams in higher education, are we grad-
data reported on aging services workforce or aging specialty education. uating a sufficient number of people to
preparedness in program planning, imple- The second reason for aging services fill positions vacated by retirement? The
mentation, and evaluation of EBHP, even workforce preparation deficits is also his- answer is no. We are losing ground, and
for the EBP Coordinator positions. torical. Workforce preparedness for our we did not have much “ground” to lose. A
There are several reasons, even with aging society has been an important topic recent article in the Chronicles of Higher
the availability of over 600 gerontology for decades – beginning with the land- Education discussed an 11% reduction in
higher education programs nationwide, mark publication of the U.S. Health Ser- the number of gerontology degree pro-
that our current aging services workforce vices Resource Administration, Bureau of grams between 2000 and 2010 (13). The
lacks needed academic preparation. The Health Professions (HRSA BHPr), enti- reasons cited were low enrollments, budget
first reason is a historical one: beginning tled,“A National Agenda for Geriatric Edu- cuts for higher education programs, and
employment in aging services may have cation: White Papers” (4). The National few student incentives, such as availability
pre-dated the widespread availability of Agenda documented the lack of training of scholarships. It is clear that reductions in
gerontology education programs. Recent and preparedness for the many needed state budgets for higher education, and lack
labor force studies have documented the health and social service professions and of funding at the federal level, have taken

www.frontiersin.org January 2015 | Volume 2 | Article 287 | 343

Frank Education can promote healthy aging programs

a toll on gerontology education at the very about 60% of AAAs have an established development of current aging services
time the programs should be robust and relationship with a local college or uni- employees. It is competency-based and
productive. versity for the purpose of securing well- tailored to deliver content on healthy
The Eldercare Workforce Alliance has trained personnel as positions become aging, behavior change, EBHP program
documented the geriatric workforce short- available (2, 3). Involving students early implementation, and management. It also
fall (14). Simply stated, there are not on in practical training through intern- involves a service learning internship
enough people specializing in geriatrics ships at agencies can lead to a pipeline of course that places students into agencies
and gerontology education to provide opti- well-prepared graduates. These graduates that are doing EBHPs.
mal care and services to the impending are then available for employment as jobs SHARP has been delivered a total of six
“boom” of older adults. This is primarily open up – this is a win-win for all con- times in two California community col-
because outside of the degree and certifi- cerned. National and state conferences for leges, with impressive evaluation results
cate specializations in gerontology, there aging services providers should include ses- and much higher than average college
are few courses offered. In addition, courses sions on success stories and best practices retention (17). As a tested model educa-
offered are typically elective, not required. in establishing and managing such local tional program, it has been packaged for
For example, in 2009–2010, only 2.8% of network opportunities. By showcasing suc- replication (curriculum, manual of pro-
BSW graduates and 6.7% of MSW grad- cessful models, perhaps we can move from cedures, evaluation tools, faculty develop-
uates completed a specialization in aging. 60% to more than 90% of agencies working ment), so it can be adopted at other higher
This is an average of 5% across all social with higher education in this manner. education institutions. A number of aging
work graduates (15). In accredited Schools The provision of student incentives is services providers have completed SHARP
and Programs in Public Health, the num- a second key activity to promote geri- and brought its resources back into their
bers are even lower. National data for the atrics and gerontology education to sup- agencies. Graduates of SHARP have been
academic year 2004–2005 show that less port the healthy aging movement. Incen- hired into agency positions, and agencies
than 3% of public health students enrolled tives can be the type of traditional training have even begun offering EBHP because of
in even one aging-related course (16). grants with payment for tuition and stu- available SHARP graduates. Further infor-
How can the educational system be dent stipends that long ago were a compo- mation about SHARP can be requested
engaged in national systems change efforts nent of the Older Americans Act. Incen- from the author.
to promote healthy aging and provide ser- tives may also be less tangible, in the form The education system may move slowly,
vices to those most in need? First, there is a of better branding of gerontology educa- but it can be responsive to workforce
need to bring them to the table. National tion as a central support for sustaining imperatives and addressing societal needs
and state policy and planning meetings the healthy aging workforce. Increasing the and opportunities. As the national move-
must include representatives from higher number, strength, and purpose of collab- ment for EBHP expansion and systems
education systems (e.g., community college orative relationships between educational development was underway, the national
system) and professional schools (e.g., pub- institutions and aging services organiza- education system was virtually ignored as
lic health). They have been left out; and as tions is necessary. If aging services orga- a resource. Readying current agency per-
a result, the exciting promise of EBHPs is a nizations could provide meaningful (and sonnel and recruiting volunteers to manage
well-kept secret from academic programs. perhaps paid) internships for students to and lead EBHP was the focus of infra-
This new content is upbeat, engaging, and gain practical experience in healthy aging structure capacity building to support pro-
a perfect way to entice students to enroll programs, this would definitely incentivize grams. This may have been, by necessity,
in their first aging-related class. In addi- students to enroll into classes. the first priority. However, to truly cre-
tion, involving educational systems’ lead- A third way to strengthen educational ate national delivery systems and embed
ership in policy and planning meetings for system involvement into the healthy aging healthy aging programs into the fabric of
systematic expansion of healthy aging pro- movement is to assure the relevance of how agencies and healthcare systems do
grams will enhance the administrative sup- educational programs by developing new their work, a steady supply of well-trained
port for aging related classes and programs tailored curricula for EBHP and healthy personnel are needed.
at all levels of higher education and profes- aging. An established and tested model Looking to the future, it is impera-
sional training. The “national movement” is the Skills for Healthy Aging Resources tive that content in gerontology, including
for healthy aging programs will be seen and Programs (SHARP) Career Technical EBHPs, is readily available within all lev-
as an opportunity for increased demand Education Certificate Program. SHARP© els of higher education programs. Utiliz-
in gerontology and geriatric education and was developed in 2009 with funding ing social marketing principles to “brand”
training. from the U.S. Department of Education’s healthy aging curricula as essential and
The national and state inclusion of Funds for Improving Post-Secondary Edu- appealing may increase enrollments of stu-
systems level higher education leaders in cation (FIPSE). It is a curriculum pack- dents in a variety of disciplines. I can envi-
policy discussions is needed. In addi- age that includes four first-year under- sion a future where all students graduating
tion, efforts to build relationships at the graduate courses. SHARP can be deliv- from any relevant program (health profes-
local level between healthy aging program ered as a component of community col- sionals, gerontology, public health, social
providers with colleges and universities lege or undergraduate programs, or as services, business, public administration,
must be strengthened. Currently, only a stand-alone program for professional etc.) are required to have coursework in

Frontiers in Public Health | Public Health Education and Promotion January 2015 | Volume 2 | Article 287 | 344
Frank Education can promote healthy aging programs

healthy aging and EBHPs – it is just that Baccalaureate Competencies and Curricular 16. Molina LC, Wallace SP. Health and Aging Education
important. Guidelines for Geriatric Nursing Care. Washington, in Accredited Public Health Programs. Washington,
DC: American Association of Colleges of Nursing DC: American Public Health Association Annual
REFERENCES (2000). Meeting (2007).
1. California Social Work Education Center (Cal- 8. American Society of Consultant Pharmacists. Geri- 17. Frank JC, Altpeter M, Damron-Rodriguez J, Drig-
SWEC). Aging Initiative Labor Force Survey II : atric Pharmacy Curriculum Guide, Second Edition gers J, Lachenmayr S, Manning C, et al. Prepar-
Community Based Services for Older Adults (2009). (2007). Available from: http://www.ascp.com/sites/ ing the workforce for healthy aging programs:
Available from: http://www.ncoa.org/assets/files/ default/files/2ndEd-ASCP-CurriculumGuide.pdf the skills for healthy aging resources and pro-
pdf/Labor-Force-Survey-II-Progress-Brief-3-25- 9. California Social Work Education Center (Cal- grams (SHARP) model. Health Educ Behav (2014)
09-Final.pdf SWEC). Aging Initiative: Aging Competencies 41:19S. doi:10.1177/1090198114543007
2. Morgan A, Markwood S, Eltzeroth H, Reed J. Build- (2006). Available from: http://calswec.berkeley.
ing Capacity Through Our Workforce: Workforce edu/CalSWEC/AgingCompetencies_All_Feb2006.
Survey Report on Area Agencies on Aging. Washing- pdf Conflict of Interest Statement: The University of Cal-
ton, DC: National Association of Area Agencies on 10. Wendt PF, Peterson DA, Douglass EB. Core Prin- ifornia at Los Angeles received grant funding from the
Aging (2010). ciples and Outcomes of Gerontology, Geriatrics U.S. Department of Education for the Skills for Healthy
3. Lackmeyer A, Straker JK, Kunkel SR. National and Aging Studies Instruction. Washington, DC: Aging Resources and Programs (SHARP) project from
Association of Area Agencies on Aging (n4a). AAA Association for Gerontology in Higher Education 2010 to 2013; SHARP is copyrighted by the University
Workforce Survey. Brief Report. Scripps Gerontol- (AGHE) and the University of Southern California of California Regents.
ogy Center (2010). Available from: http://sc.lib. (1993).
muohio.edu/bitstream/handle/2374.MIA/4386/ 11. American Geriatrics Society. Partnership of
Workforce%20Survey%20Research%20Record% Health and Aging: Multidisciplinary Competen- Received: 16 June 2014; accepted: 08 December 2014;
20-%20Scripps%20-%20FINAL.pdf?sequence=1 cies in the Care of Older Adults (2010). Avail- published online: 09 January 2015.
4. U.S. Department of Health and Human Ser- able from: http://www.americangeriatrics.org/ Citation: Frank JC (2015) A missing piece in the infra-
vices, Health Resources, Services Administration files/documents/pha/PHAMultidiscComps.pdf structure to promote healthy aging programs: education
(HRSA). A National Agenda for Geriatric Educa- 12. Frank JC, Weiss J. Public health workforce: prepa- and work force development. Front. Public Health 2:287.
tion. Rockville, MD: White Papers (1995). ration for an aging society. In: Prohaska T, Ander- doi: 10.3389/fpubh.2014.00287
5. Institute of Medicine of the National Academies. son L, Binstock R, editors. Public Health and an This article was submitted to Public Health Education
Retooling for an Aging America. Washington, DC: Aging Society. (Chap. 13), Baltimore, MD: Johns and Promotion, a section of the journal Frontiers in
National Academy Press (2008). Hopkins University Press (2012). 2012 p. Public Health.
6. American Association of Medical Colleges, John 13. Anft M. Gerontologists in Demand, but Degree Pro- Copyright © 2015 Frank. This is an open-access article
A. Hartford Foundation, Inc. Consensus Confer- grams Languish. Washington, DC: Chronicles of distributed under the terms of the Creative Commons
ence on Competencies in Geriatric Education. Min- Higher Education (2014). Attribution License (CC BY). The use, distribution or
imum Geriatric Competencies for Medical Stu- 14. Alliance. Geriatrics Workforce Shortage: A Loom- reproduction in other forums is permitted, provided the
dents (2008). Available from: http://www.pogoe. ing Crisis for Our Families (2014). Available from: original author(s) or licensor are credited and that the
org/Minimum_Geriatric_Competencies eldercareworkforce.org original publication in this journal is cited, in accordance
7. American Association of Colleges of Nursing, 15. Council on Social Work Education. 2011 Annual with accepted academic practice. No use, distribution or
The John A. Hartford Foundation Institute for Statistics on Social Work Education in the United reproduction is permitted which does not comply with
Geriatric Nursing. Older Adults: Recommended States (2011). Available from: www.cswe.org these terms.

www.frontiersin.org January 2015 | Volume 2 | Article 287 | 345

 ORIGINAL RESEARCH ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2014.00213

Effect of physical activity, social support, and skills training

on late-life emotional health: a systematic literature review
and implications for public health research
Mark B. Snowden 1 , Lesley E. Steinman 2 *, Whitney L. Carlson 1 , Kara N. Mochan 3,4 , Ana F. Abraido-Lanza 5 ,
Lucinda L. Bryant 6 , Michael Duffy 7 , Bob G. Knight 8 , Dilip V. Jeste 9 , Katherine H. Leith 10 , Eric J. Lenze 11 ,
Rebecca G. Logsdon 12 , William A. Satariano 13 , Damita J. Zweiback 14,15 and Lynda A. Anderson 16,17
1
Department of Psychiatry and Behavioral Sciences, University of Washington School of Medicine, Seattle, WA, USA
2
Health Promotion Research Center, University of Washington, Seattle, WA, USA
3
University of Washington School of Nursing with Environmental Health Focus, Seattle, WA, USA
4
Adolescent Medicine, Seattle Children’s, Seattle, WA, USA
5
Department of Sociomedical Sciences, Columbia University Mailman School of Public Health, New York, NY, USA
6
Department of Community and Behavioral Health, Colorado School of Public Health, University of Colorado Denver, Aurora, CO, USA
7
Department of Educational Psychology, Counseling Psychology Program, Texas A&M University, College Station, TX, USA
8
Davis School of Gerontology and Department of Psychology, University of Southern California, Los Angeles, CA, USA
9
Sam and Rose Stein Institute for Research on Aging and Department of Psychiatry, University of California San Diego, San Diego, CA, USA
10
College of Social Work, University of South Carolina, Columbia, SC, USA
11
Department of Psychiatry, Washington University School of Medicine, St Louis, MO, USA
12
Department of Psychosocial and Community Health, University of Washington School of Nursing, Seattle, WA, USA
13
School of Public Health, University of California Berkeley, Berkeley, CA, USA
14
Division of Chronic Disease and Injury Prevention, Michigan Department of Community Health, Lansing, MI, USA
15
Healthy Aging Council and Health Equity Council, National Association of Chronic Disease Directors, Atlanta, GA, USA
16
National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta, GA, USA
17
Rollins School of Public Health, Emory University, Atlanta, GA, USA

Edited by: Purpose: Given that emotional health is a critical component of healthy aging, we under-
Matthew Lee Smith, The University of
took a systematic literature review to assess whether current interventions can positively
Georgia, USA
affect older adults’ emotional health.
Reviewed by:
Eli Carmeli, Haifa University, Israel Methods: A national panel of health services and mental health researchers guided the
Brittany Rosen, University of
Cincinnati, USA
review. Eligibility criteria included community-dwelling older adult (aged ≥ 50 years) sam-
*Correspondence:
ples, reproducible interventions, and emotional health outcomes, which included multiple
Lesley E. Steinman, MSW, MPH, domains and both positive (well-being) and illness-related (anxiety) dimensions.This review
Health Promotion Research Center, focused on three types of interventions – physical activity, social support, and skills train-
University of Washington, Seattle, ing – given their public health significance and large number of studies identified. Panel
WA. 1107 NE 45th Street, Suite 200,
Seattle, WA 98105, USA
members evaluated the strength of evidence (quality and effectiveness).
e-mail: [email protected]
Results: In all, 292 articles met inclusion criteria. These included 83 exercise/physical activ-
ity, 25 social support, and 40 skills training interventions. For evidence rating, these 148
interventions were categorized into 64 pairings by intervention type and emotional health
outcome, e.g., strength training targeting loneliness or social support to address mood.
83% of these pairings were rated at least fair quality. Expert panelists found sufficient
evidence of effectiveness only for skills training interventions with health outcomes of
decreasing anxiety and improving quality of life and self-efficacy. Due to limitations in
reviewed studies, many intervention–outcome pairings yielded insufficient evidence.
Conclusion: Skills training interventions improved several aspects of emotional health in
community-dwelling older adults, while the effects for other outcomes and interventions
lacked clear evidence. We discuss the implications and challenges in moving forward in
this important area.
Keywords: mental health, aged, health promotion, review

INTRODUCTION dimensions. Hendrie et al. (1) characterized emotional health as

Emotional health is increasingly viewed as a multidimen- self-efficacy, depression, hostility and anger, anxiety, psychological
sional construct that includes both positive and illness-related stress, optimism, self-esteem, quality of life, and other domains

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 213 | 346
Snowden et al. Late-life emotional health review

assessed by multidimensional measures. A report (2) using data framework and definition of emotional health (Figure 1). Inter-
from the Behavioral Risk Factor Surveillance System (BRFSS) ventions to promote emotional health can influence various deter-
(3) identified six indicators reflecting positive and illness-related minants of emotional health. These determinants include sub-
emotional health outcomes in older adults: social and emotional stance use and other behaviors, cognitive factors, psychosocial
support; life satisfaction; frequent mental distress; current depres- factors, emotional factors, and chronic conditions. Risk and pro-
sion; lifetime diagnosis of depression; and lifetime diagnosis of tective factors for emotional health also included less modifiable
anxiety disorders. biological and genetic factors and demographics. For the purpose
Mental health is increasingly viewed as part of public health’s of this review, we focused on interventions aimed at modifiable
mission, as important as physical health in contributing to over- determinants.
all health and well being (2). Epidemiologic data links a range Borrowing from Hendrie and colleagues, we defined emotional
of health outcomes, particularly mortality and cardiovascular dis- health comprehensively as including both emotion regulation
ease, to emotions (1). Despite the public health importance, little concepts (e.g., the ability to control/regulate emotions) and emo-
is currently known about the effectiveness of interventions to pro- tion intelligence (e.g., the ability to recognize and use emotions
mote emotional health in community-dwelling older adults. One constructively). Most importantly, emotional health is multidi-
of the few available reports (4) reviews studies from UK, finding mensional, involving positive mental health constructs, such as
some evidence to support significant small-to-moderate improve- life satisfaction as well as illness-related domains such as anxi-
ments in emotional health from select exercise programs including ety. We used Hendrie and colleagues’ emotional health domains
mixed exercise programs, strength and resistance, aerobic, walk- (1) and added “general well being” and “social support,” given
ing, and individually targeted health promotion interventions. research describing the relevance of these constructs to emotional
However, it also indicated a clear shortage of robust evidence for health (6–8). The emotional health constructs used in this review
effective programs to improve late-life emotional health. are provided in the first row of Table 1. Finally, based on the litera-
Although this review (4) addressed several important questions, ture, the conceptual model included longer term health outcomes
a more rigorous review of the scientific literature is warranted. The associated with emotional health, including reductions in mortal-
primary objective of this systematic literature review was to iden- ity and improvements in functional ability, morbidity of chronic
tify interventions to promote emotional health of older adults aged conditions, and overall quality of life (entailing both physical and
50 years and older. We sought to expand Windle and colleagues emotional well being).
work by encompassing a wider range of community-based inter-
ventions, including more than UK-based studies, examining mul- Expert panel and review methods
tiple domains of emotional health incorporating both positive and This review was guided by an eight-member expert panel of
illness-related dimensions, and addressing community-dwelling health services and mental health researchers from around the
older adults. United States representing psychology, psychiatry, geriatrics, pub-
lic health, and social work. The systematic review methods were
MATERIALS AND METHODS derived from the Guide to Community Preventive Services (“The
DATA SOURCES Guide”) (9, 10) and the systematic literature review of strate-
Conceptual framework and definition of emotional health gies to address late-life depression (11), using a formal process
This review used the NIH’s Cognitive and Emotional Health to identify relevant studies, assess their quality, and summarize
Project (1, 5) to guide the development of our conceptual the evidence. We searched the peer-reviewed literature through

Substances
and Behaviors
Self-Efficacy
Cognitive Depression
Factors Anxiety
Hostility / Anger
Quality of Life
Psychosocial Psychological Stress
Mortality
Interventions Factors Optimism
Functional Ability
Self-Esteem
Chronic Conditions
General Well-Being
Chronic disease
Social Support
and risk factors
Other domains

Biological and
Genetic Factors
Interventions

Non-Modifiable Determinants
Demographics
Determinants Intermediate Outcomes
Health Outcomes

FIGURE 1 | Conceptual framework.

www.frontiersin.org April 2015 | Volume 2 | Article 213 | 347

Snowden et al. Late-life emotional health review

Table 1 | Search terms used in electronic searches. June 2008 and updated the search in June 2012 using PubMed
(www.ncbi.nlm.nih.gov), CINAHL (http://www.ebscohost.com/
Construct Search terms academic/cinahl-plus-with-full-text/), and PsycINFO (www.apa.
Emotional Emotional health Interpersonal trust
org/pubs/databases/psycinfo/index.aspx) databases. Subject head-
health
ings and text words reflected our study aims, including key con-
Self-efficacy Positive Energy
cepts of “emotional health,” “older adults,” “community based,”
Locus of control Happiness
and “intervention”; specific terms are provided in Table 1. Refer-
Personal control Contentment
ences to meta-analyses and review papers were also examined, and
Personal mastery Hardiness
expert panelists reviewed the citations of included articles.
Powerlessness Resilience
Sense of coherence Emotional vitality
Depression Shame STUDY SELECTION
Hopelessness Guilt Study inclusion criteria were (1) published data on populations
Hostility Regret aged 50 years and older, (2) community-based sample and set-
Anger Emotion regulation ting, (3) clearly described intervention; and (4) “emotional health”
Type A behavior Emotional control operationalized using the list of constructs determined by the
Anxiety Well being expert panel (see Table 1). There were no restrictions on sam-
Environmental demands Altruism ple size or study design. Articles were excluded if they: were
Life events Sadness not available in English; reported only a review of the literature,
Stress Fear meta-analysis, or commentary; focused exclusively on inpatient or
Mood states Neuroticism institutionalized persons. We included articles from any country
Positive affect Boredom as long as they were published in English. We excluded interven-
Negative affect Capacity to care tions that targeted depression given the overlap with a previously
Optimism Life satisfaction conducted review focusing on late-life depression (11). The emo-
Self-esteem Spirituality tional component of quality of life measures was included [e.g., the
Quality of life Caregiver burden role emotional subscale of the SF-36 (12)]; however, physical sub-
Loneliness Acculturation scales were excluded. For studies aimed at addressing outcomes
Social support Discrimination not strictly emotional in nature (e.g., spirituality, caregiver bur-
den), we required the inclusion of at least one other emotional
Intervention Intervention Reminiscence therapy
health outcome from the list of constructs.
Treatment Assertiveness training
We used a two-step screening process evaluating abstracts and
Prevention Strengths-based
where necessary full text to assess whether articles met inclusion
Exercise Positive psychology
criteria. A standardized form was used to systematically collect key
Physical activity Social support
data from each article, including study design, sample size, inter-
CBT Spirituality
vention setting, outcome measures, results, and indicators of study
Psychotherapy Complementary and
quality. Data were compiled in summary tables that the expert
alternative medicine
panel used for the evidence rating. As employed in our prior review
Life review Integrated medicine
(11), we grouped articles into intervention type-emotional health
stress management
outcome pairings to categorically rate the evidence. For example,
Meditation Anger management
skills training interventions aimed at reducing anxiety were paired
Mindfulness Coping skills
together.
Community Community Primary care Expert panel members rated the quality and effectiveness of
based Home Community health each intervention–outcome pairing (Table 2). For quality rating,
Neighborhood center panel members independently rated the set of studies for each
intervention–outcome pairing as Good, Fair, or Limited. Because
Older Older adults Middle-Aged few pairings received a vote of “good,” the good and fair categories
adults Aged Limits of 40 and older were collapsed into a single category labeled “at least fair” quality.
Elderly (to include 50 and older) For effectiveness ratings, the panel members independently rated
Study Clinical trial Experimental replication each intervention–outcome pairing as Strong, Sufficient, or Insuf-
design Multicenter study Follow-up study ficient. For any pairing rated as insufficient, panel members were
Randomized controlled trial Field study asked to record whether the rating was due to (1) an insufficient
Randomized clinical trial Non-clinical case study number of available studies or (2) a sufficient number of available
Evaluation studies Qualitative study studies but an insufficient amount of data to determine effective-
Clinical case study Quantitative study ness. As established at the start of the review process, final determi-
Empirical study nation of quality and effectiveness was based on 80% agreement
among panel members. The panel met to discuss areas of disagree-
Note: We did not find any physical activity, social support, or skills training ment and panel members were allowed to change their votes after
intervention studies that targeted the emotional health outcomes in italics. the discussion; however, they were not required to reach consensus.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 213 | 348
Snowden et al. Late-life emotional health review

RESULTS Asia. Thirty-nine percent of the articles specified that a theoretical

A total of 3,926 articles were identified in the initial search (1,250 framework that was used to inform the development of the inter-
from PubMed, 1,025 from PsycINFO, 1,631 from CINAHL, and 20 vention – one-third of the studies that evaluate an exercise or
from reference lists of review articles or meta-analyses). 553 arti- a social support intervention used a theoretical framework, while
cles were duplicates and were eliminated (Figure 2). Two hundred two-thirds of skills training interventions used a theoretical frame-
ninety-two articles were eligible for inclusion, with the major- work. Across interventions, the most common frameworks used
ity of the ineligible being excluded due to having too young of across interventions were social cognitive theory, self-efficacy, and
a sample size, not being an intervention study, or not having an social learning theory. Other theories include the progressively
emotional health outcome. Of the 292 eligible articles, the expert lowered stress threshold model, the self-care deficit nursing the-
panel focused on three types of interventions relevant to public ory of Orem, mindfulness meditation, self-management model
health practice and with ample studies for rating the evidence. of illness behavior, stress and coping theoretical framework, stress
These comprised a total of 148 of the 292 found articles: physical process models of caregiving, the transtheoretical model of behav-
activity and/or exercise (n = 83), skills training (n = 40), and social ior change, stages of change, negotiated adherence model, motiva-
support (n = 25) (Table 3). More than half of the studies (57%) tional interviewing, transforming hope theory, and Yalom group
were from the US or Canada, 19% were from European studies, theory.
12% were from Australia or New Zealand, and 11% were from The physical activity and/or exercise interventions included
aerobic activity, strength training, balance and flexibility inter-
ventions, motivational strategies, and a combination of exer-
cise types. The skills training group included self-management
Table 2 | Indicators of quality and effectiveness for rating the evidence.
[e.g., Chronic Disease Self-Management Program (CDSMP)],
Quality indicators Effectiveness indicators psycho-education, anger management, and stress management
interventions. The social support group included interven-
Well-described study Study quality tions targeting direct or indirect provision of social sup-
population and intervention port (e.g., interventions designed to improve ability to obtain
Sampling Study design support).
Inclusion/exclusion criteria Number of studies The 148 studies were subsequently grouped into 64 interven-
Data analysis Consistency across studies tion type–outcome pairings, or categories, for rating the evidence,
Interpretation of results Statistical results such as social support interventions aimed at elevating mood

3,926 citations (records) identified:

1,250 PubMed
1,025 PsycINFO
1,631 CINAHL
20 from reference lists
553 duplicates removed
3,373 citations screened

Level 1 screening of abstracts Level 2 screening of articles

2,610 excluded 471 excluded
163 no older adults 192 no older adults
250 not community -based 14 not community -based
1,261 intervention not described 110 intervention not described
936 no emotional health outcome 155 no emotional health outcome
703 to Level 2 screening
60 met inclusion criteria 232 met inclusion criteria

292 studies included in the review

148 intervention studies in this article:

83 physical activity/exercise
40 skills training
25 social support

Summary table for evidence review of

64 intervention-outcome pairings

FIGURE 2 | Literature review flow chart.

www.frontiersin.org April 2015 | Volume 2 | Article 213 | 349

Snowden et al. Late-life emotional health review

Table 3 | Intervention–outcome pairings for skills training, social support + skills training, and physical activity interventions.

Intervention Emotional health outcome # Of studies (n)a Quality rating Effectiveness rating

Skills training Anger 3 (258) (13–15) At least fair Insufficient (no consensus)

Skills training Anxiety 11 (1,346) (13, 16–25) At least fair Sufficient

Skills training Mood 5 (988) (13, 18, 26, 27, 76) At least fair Insufficient (no consensus)

Skills training Other positive outcomes 2 (99) (29, 145) At least fair Insufficient (not enough studies)

Skills training Psychological well 4 (1,449) (31, 32, 124, 142) At least fair Insufficient (multiple studies,
being/distress inconclusive data)

Skills training Quality of life 11 (1,417) (17, 22, 29, 31, At least fair Sufficient
35–41)

Skills training Self-efficacy 16b (3,735) (14, 15, 18, 20, 24, At least fair Sufficient
26, 27, 30, 35, 39, 41–46, 175)

Skills training Spirituality 3b (283) (23, 27, 65, 148) Limited Insufficient (not enough studies)

Skills training Stress 4b (500) (39, 45, 46, 98, 142) At least fair Insufficient (multiple studies,
inconclusive data)

Social support Anxiety 3b (502) (34, 93, 135, 138) At least fair Insufficient (no consensus)

Social support Loneliness 2 (313) (72, 108) Limited Insufficient (not enough studies)

Social support Mood 2b (144) (72, 109, 113) Limited Insufficient (not enough studies)

Social support Other positive outcomes 1 (39) (83) Limited Insufficient (not enough studies)

Social support Psychological well 5b (704) (31, 34, 89, 128, 135, At least fair Insufficient (multiple studies,
being/distress 139) inconclusive data)

Social support Quality of life 3b (450) (31, 34, 135, 138) At least fair Insufficient (no consensus)

Social support Self-efficacy/locus of control 1 (39) (83) Limited Insufficient (not enough studies)

Social support + skills Anxiety 5 (580) (54, 63, 70, 100, 143) At least fair Insufficient (multiple studies,
training inconclusive data)

Social support + skills Mood 1 (144) (70) At least fair Insufficient (not enough studies)
training

Social support + skills Other negative outcomes 2 (415) (47, 82) At least fair Insufficient (not enough studies)
training

Social support + skills Other positive outcomes 3c (58) (33, 66) At least fair Insufficient (no consensus)
training

Social support + skills Psychological well 6 (1,041) (14, 47, 70, 82, 144, Limited Insufficient (multiple studies,
training being/distress 174) inconclusive data)

Social support + skills Quality of life 3b,c (393) (66, 109, 113, 121) At least fair Insufficient (no consensus)
training

Social support + skills Self-efficacy/locus of control 3 (408) (65, 70, 121) At least fair Insufficient (no consensus)
training

Motivation/counseling Mood 1 (86) (103) At least fair Insufficient (not enough studies)

Motivation/counseling Other positive outcomes 2 (969) (71, 79) At least fair Insufficient (No consensus)

Motivation/counseling Quality of life 4 (850) (52, 64, 71, 120) At least fair Insufficient (Multiple studies,
inconclusive data)

Motivation/counseling Self-efficacy/mastery 5 (567) (71, 79, 92, 112, 176) At least fair Insufficient (multiple studies,
inconclusive data)

(Continued)

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 213 | 350
Snowden et al. Late-life emotional health review

Table 3 | Continued

Intervention Emotional health outcome # Of studies (n)a Quality rating Effectiveness rating

Motivation/counseling Stress 2 (1,712) (79, 118) At least fair Insufficient (no consensus)

Aerobic: walking Anxiety 3 (507) (59, 102, 146) At least fair No Consensus (btw sufficient and
insufficient, multiple studies)

Aerobic: other aerobic Anxiety 4 (361) (57, 73, 114, 136) At least fair Insufficient (multiple studies,
activities inconclusive data)

Aerobic: walking Caregiver burden 1b (100) (60, 102) At least fair Insufficient (not enough studies)

Aerobic: walking Mood 2 (170) (107, 147) At least fair Insufficient (no consensus)

Aerobic: walking Other positive outcomes 1 (582) (101) At least fair Insufficient (not enough studies)

Aerobic: other aerobic Other positive outcomes 2 (150) (57, 114) At least fair Insufficient (not enough studies)
activities

Aerobic: walking Quality of life 6 (1,273) (56, 101, 104, 123, 130, At least fair Insufficient (multiple studies,
147) inconclusive data)

Aerobic: other aerobic Quality of life 6 (823) (51, 57, 117, 134, 151, 179) At least fair Insufficient (multiple studies,
activities inconclusive data)

Aerobic: walking Psychological distress and 91 (28) At least fair Insufficient (not enough studies)
well-being

Aerobic: other aerobic Psychological distress and 101 (136) At last fair Insufficient (not enough studies)
activities well being

Aerobic: walking Self-efficacy/mastery/locus of 1 (32) (62) NC Insufficient (not enough studies)

control

Aerobic: Other aerobic Self-efficacy/mastery/locus of 3 (231) (56, 106, 114) NC Insufficient (no consensus)
activities control

Aerobic: walking Stress 2b (457) (59, 60, 102) At least fair No consensus (btw sufficient and
insufficient, not enough studies)

Strength/resistance Anxiety 1 (42) (129) At least fair Insufficient (not enough studies)

Strength/resistance Fear of falling 2 (94) (48, 150) At least fair Insufficient (no consensus)

Strength/resistance Loneliness 1b (32) (84, 129) At least fair Insufficient (not enough studies)

Strength/resistance Mood 2 (144) (69, 153) At least fair Insufficient (no consensus)

Strength/resistance Psychological well 2 (124) ( 134, 153) At least fair Insufficient (not enough studies)
being/distress

Strength/resistance Quality of life 13b (1,000) (28, 68, 75, 84, 115, 119, At least fair Insufficient (multiple studies,
122, 126, 132–134, 137, 153, 177) inconclusive data)

Strength/resistance Self-efficacy/locus of control 7b (442) (75, 115, 126, 129, 132, At least fair Insufficient (multiple studies,
137, 153, 177) inconclusive data)

Stretch/flexibility/ Anxiety 1 (88) (96) NC Insufficient (not enough studies)

balance/agility

Stretch/flexibility Fear of falling 2b (422) (53, 90, 181) At least fair No consensus (btw sufficient and
/balance/agility insufficient)

Stretch/flexibility/ Mood 5 (307) (49, 87, 95, 116, 147) At least fair Insufficient (no consensus)
balance/agility

Stretch/flexibility/ Other positive outcomes 1b (200) (53, 182) At least fair Insufficient (not enough studies)
balance/agility

(Continued)

www.frontiersin.org April 2015 | Volume 2 | Article 213 | 351

Snowden et al. Late-life emotional health review

Table 3 | Continued

Intervention Emotional health outcome # Of studies (n)a Quality rating Effectiveness rating

Stretch/flexibility/ Psychological well 1b (200) (53, 182) At least fair Insufficient (not enough studies)
balance/agility being/distress

Stretch/flexibility/ Quality of life 8b (853) (48, 51, 53, 87, At least fair Insufficient (multiple studies,
balance/agility 94, 96, 132, 147, 181) inconclusive data)

Stretch/flexibility/ Self-efficacy/mastery/locus of 5 (465) (48, 90, 94, 95, 132) At least fair No consensus (btw strong,
balance/agility control sufficient, insufficient)

Stretch/flexibility/ Stress 1 (39) (95) NC Insufficient (not enough studies)

balance/agility

Combination Anxiety 3 (485) (91, 180, 182) At least fair Insufficient (no consensus)

Combination Fear of falling 2 (200) (85, 88) At least fair Insufficient (no consensus)

Combination Mood 3 (257) (81, 97, 173) At least fair Insufficient (no consensus)

Combination Other positive outcomes 3 (459) (91, 131, 178) At least fair Insufficient (multiple studies,
inconclusive data)

Combination Psychological well 6 (748) (97, 131, 133, 180, At least fair Insufficient (multiple studies,
being/distress 182, 184) inconclusive data)

Combination Quality of life 16 (7,492) (55, 61, 78, 80, At least fair Insufficient (multiple studies,
81, 85, 86, 88, 97, 99, 110, inconclusive data)
111, 149, 152, 182, 183)

Combination Self-efficacy/mastery/locus of 5b (654) (77, 92, 105, 125, At least fair Insufficient (multiple studies,
control 127, 183) inconclusive data)

Combination Stress 1 (187) (180) NC Insufficient (not enough studies)

NC, no consensus.
a
Article citations for each intervention–outcome pairing are provided in this column. Some of the 148 studies are listed in more than one intervention–outcome pairing.
b
Several studies are reported in more than one article (e.g., article #40 and article #41 describe the same study using different analyses).
c
Article #62 reported on two different positive outcomes, self-esteem and life satisfaction.

(Table 3). For quality, 53 (83%) of the intervention–outcome dropout rates, and in two of these, that rate was below 20%. Study
pairings were rated as having “at least fair” quality; only 11% duration varied from 2 to 12 months, although generally the active
of these had good quality. For effectiveness, a majority of pair- phase ranged from 6 to 8 weeks.
ings (89%) were deemed to have insufficient evidence, due to The report by López et al. (16) focused on caregivers in which
lack of studies (two or fewer) or inconclusive evidence (mixed the majority of care was provided to persons living with demen-
results within or across studies). Herein, we will report findings tia (80%).They found a 38% decrease in mean anxiety score in
for the three intervention–outcome pairings for which sufficient the Hospital Anxiety and Depression Scale (HADS) (154) for
evidence was found. For further information about categories not traditional format skills training (60 min weekly over a period
presented or on detailed summary data tables, please contact the of 8 weeks) involving cognitive behavioral approaches, assertive-
corresponding author. ness training, self-esteem building exercises, and problem-solving
skills training. The other studies using the HADS found a 10–
INTERVENTION–OUTCOME PAIRINGS WITH SUFFICIENT EVIDENCE 20% decrease in anxiety scores after intervention (17, 18). The
Skills training Williams (19) study of 71 women with breast cancer found no
Sufficient evidence was found for effectiveness of skills training effect for a 20-min audiotape to teach skills for decreasing sleep,
interventions to reduce anxiety and to promote quality of life and anxiety, and fatigue problems encountered during chemotherapy.
self-efficacy (from a total of 38 studies). These studies were rated Two non-randomized, controlled trials did not show a signifi-
as having “at least fair” quality. Of these studies, 11 were aimed at cant effect. One focused on asthma self-management and another
reducing anxiety, of which four involved randomized controlled focused on Chinese older adults with history of depression or
trials (RCT). They involved 1,346 participants and represented a anxiety, although there was a non-significant trend toward effec-
diverse subject population (e.g., caregivers and people with breast tiveness (p < 0.10) (20, 21). Five single-group studies revealed
cancer, heart disease, or arthritis). Only three studies reported mixed results (13, 22–25).

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 213 | 352
Snowden et al. Late-life emotional health review

Eleven additional skills training studies aimed at emotional Exercise and social support
health as measured by the subscales of a quality of life measure The expert panel did not find sufficient evidence for either exercise
such as the SF-36. There were eight RCTs, two quasi-experimental or social support interventions to improve emotional health.
studies, and one single-group study. A total of 1,417 partici-
pants were included in these studies, with sample sizes ranging OTHER INTERVENTION–OUTCOME PAIRINGS
from 35 to 320, averaging between 75 and 100 participants. The Skills training
duration of the interventions ranged from 1 week to 8 months, The expert panel found insufficient evidence for 20 other skills
averaging between 6 and 8 weeks. Interventions included both training interventions that focused on other emotional health out-
group and individual-level activities. Dropout rates of less than comes such as mood and stress. Most of these pairings were of at
20% were reported for all but two studies. Seven studies [five least fair quality. In addition, 82 studies were found that reported
RCTs (17, 35–39) and one non-RCT (40)] reported statistically on the effects of physical activity and/or exercise on emotional
significant improvements in at least one emotional health sub- health outcomes, and 25 studies looked at social support interven-
scale of the SF-36 Quality of Life measure. Specifically, statistically tions. There was insufficient evidence of effectiveness for most of
significant improvements were reported for the vitality and role these intervention–outcome pairings and the panel rated most of
limitations emotional SF-36 subscales for Barnason et al.’s (35) the pairings as at least fair quality.
phone-based home communication intervention for older adults
with ischemic heart failure (p < 0.01). Similarly, Grant et al.’s Exercise and physical activity
(36) social problem-solving phone partnership for adult care- The expert panel did not reach consensus for several physical activ-
givers of stroke survivors improved quality of life subdomains ity and exercise intervention–outcome pairings. First, the panel
(p = 0.013). McHugh et al.’s (17) share care health education and was split between ratings of sufficient and insufficient for stretch-
motivational interviewing program for adults waiting for elec- ing, flexibility, balance, or agility interventions to decrease fear
tive CABG (p = 0.000), and Wallace et al.’s (37) nurse visit to of falling. Second, panel members did not agree on whether there
develop a customized health plan for older adults exercising at was sufficient evidence that stretching, flexibility, balance or agility
a local senior center were found to be effective (p = 0.02). No interventions improved self-efficacy, mastery, or locus of control.
significant improvements in vitality were found for Markle-Reid Panel members raised concerns about limited numbers of studies
et al.’s (26, 38) individual-level program to bolster personal and for any single outcome and about mixed results observed across
environmental resources of frail, older home care clients although the study outcomes. Finally, the expert panelists were split between
this study did find improvement using the role limitation emo- evidence ratings of sufficient and insufficient for walking inter-
tional subscale. In addition to Grant et al. (36), Markle-Reid et al. ventions that targeted anxiety or stress. Insufficient evidence was
(38), McHugh et al. (17), and Wallace et al. (37) studies, Hughes found for all other exercise and physical activity interventions.
et al. (39) study of a workshop intervention for women with self-
reported disabilities all reported significant improvements in the Social support
SF-36 mental health subscale. Furthermore, two studies (38, 40) The expert panel found insufficient evidence that the reviewed
found significant improvements in the mental health composite SF- social support interventions improved emotional health.
36 measure (including vitality, mental health, and role limitation
emotional). Significant improvements were demonstrated in two DISCUSSION
studies using emotional health subscales of quality of life-specific This review examined three broad types of interventions designed
measures for older adults with heart failure (13, 22–25, 31, 35–38). to promote emotional health: physical activity and/or exercise,
The remaining two studies (29, 41) did not find improvements in skills training, and social support. Among the interventions rated
emotional health subscales of different quality of life measures. as having at least fair quality and sufficient evidence, we found
Sixteen skills training intervention studies were directed at that skills training interventions reduced anxiety; enhanced self-
improving self-efficacy. These studies included 11 RCTs, two efficacy; and improved vitality, role functioning related to emo-
observational studies, and three single-group studies. Seven of tional limitations, and emotional health as measured in quality
the studies were of interventions using the CDSMP. A total of of life subscales. Skills training interventions are theorized to
3,735 participants received skills training interventions, with sam- promote positive domains of emotional health through cogni-
ple sizes ranging from 33 to 728. Study duration averaged 6 to tive reframing, strengthening coping resources, and increasing the
8 weeks. Dropout rates, reported in half the studies, were less than amount of support (or quality of support). We acknowledge that
20%. The frequency of the skills training interventions was rarely skills training may improve emotional health through improved
reported. When reported, adherence to the intervention was typ- self-efficacy, though the panel chose to view self-efficacy as its
ically less than 80%. The interventions were delivered most often own emotional health domain. These interventions are designed
in a group format and the control groups were generally usual care for older adults with chronic conditions (e.g., arthritis, heart dis-
and wait-list control conditions. Eight of the 11 RCTs (14, 15, 26, ease, physical disabilities) or informal caregivers (e.g., spouses,
27, 35, 42–46) reported significant improvements in self-efficacy; adult children) of older adults coping with dementia, stroke sur-
three of the significant studies used CDSMP (15, 42, 45). Four of vivors, or mental illness making them quite generalizable. These
the five non-RCT studies (15, 20, 24, 32) also demonstrated signif- populations were targeted by these interventions because chronic
icant improvements in self-efficacy. All but Smith et al. (20) study conditions or caregiving responsibilities increase the need for skills
were single-group designs with 20–32% dropout rates. training, support, information, and resources.

www.frontiersin.org April 2015 | Volume 2 | Article 213 | 353

Snowden et al. Late-life emotional health review

The CDSMP was used as an intervention in seven of the skills health constructs given the tendency (up until recently) to focus on
training studies that showed sufficient evidence for improving disease prevention over health promotion. We anticipate that more
quality of life or self-efficacy or decreasing anxiety. CDSMP has research will include emotional health outcomes as models such
been shown to enhance stress management techniques, improve as the socio-ecological model (67, 172) and guidelines such as the
communication with physicians, increase confidence in ability Public Health Action Plan to Integrate Mental Health Promotion
to manage the condition, and improve role function (32, 42, and Mental Illness Prevention with Chronic Disease Prevention,
155–159). Improving self-management skills has been shown to 2011–2015 (74) emphasize the importance of emotional health in
impact other aspects of participants’ lives, such as their ability the larger public health goals.
to manage their emotions, choose healthy foods and exercise Future research needs to address these quality concerns by
activities, and activate their social network (158). This review attending to limitations with both internal and external validity.
is limited by its end date of June 2008. While it is beyond the One way to do so is to use the RE-AIM framework, a conceptual
scope of this project to conduct an updated systematic litera- approach for evaluating the translation of research into practice
ture review, we recently searched for other review papers on in “real-world” settings (141). RE-AIM stands for reach, effective-
skills training, exercise and/or physical activity, and social sup- ness, adoption, implementation fidelity, and maintenance – five
port interventions to promote emotional health. We found two areas, which, if addressed, ensure that essential program goals are
review papers (160, 161) that reported similar findings as we report retained during the implementation process, resulting in greater
above, namely, sufficient evidence for skills training interventions external validity. More research is also needed to investigate the
impact on self-efficacy and quality of life and insufficient evi- longer term, maintenance effects of interventions to promote pos-
dence for other emotional health outcomes. We also searched for itive emotional health, and address illness-related domains in older
intervention studies for those areas where sufficient evidence was adults as most of the studies here were of short-term effectiveness.
found. Our search yielded 10 recently published articles (162– The prominence of theories such as social cognitive theory, social
171), none of which reported different findings than reported learning theory, and self-efficacy theory in those interventions with
above. sufficient evidence may also be helpful to consider in future inter-
We defined “insufficient evidence of effectiveness” in two ways: vention design and development as they may have contributed to
either there were not enough studies of at least fair quality, or the optimization of participants’ quality of life and self-efficacy
there were multiple studies with inconclusive data. Insufficient and minimization of anxiety symptoms.
evidence did not mean that interventions were clearly ineffec- Despite the gaps in the current research, our systematic review
tive. Very few intervention–outcome pairings were rated as at least provides important information about interventions that can pro-
fair quality. The expert panel identified the following common mote emotional health outcomes in community-dwelling older
quality limitations: lack of descriptive information about the inter- adults. Specifically, we found that skills training interventions
ventions, limited information about the statistical methods and resulted in improvements in both illness-related (anxiety) and
analyses, and small sample sizes or underpowered studies. Addi- positive (quality of life and self-efficacy) domains of emotional
tionally, features of some of the study designs made it difficult health. Given that more than one in four Americans lives with
to detect changes in emotional health. For example, many stud- two or more concurrent chronic conditions, the challenges of
ies included emotional health outcome measures that may not be managing multiple chronic conditions among the growing num-
responsive to small changes from programs of limited intensity bers of older persons are significant (50). One of the overarch-
and duration, and sampling “emotionally healthy” subjects that ing goals of the U.S. Department of Health and Human Ser-
created ceiling effects. In fact, many of the reviewed aerobic physi- vices’ Strategic Framework (58), Optimum Health and Quality
cal activity interventions did not meet current national guidelines of Life for Individuals with Multiple Chronic Conditions, is to
(140) for 150 min per week of moderate-intensity activity (though “maximize the use of proven self-care management and other
all reviewed strength/resistance interventions did meet existing services by individuals with multiple chronic conditions.” As
criteria of 2 days per week). shown in this review, skills training interventions can offer impor-
Our review included a wide range of emotional health con- tant benefits in the realm of promoting emotional health in
structs. Some outcomes were entirely emotional (e.g., anxiety), older adults. Given the expanding proportion of older adults in
whereas others included a mix of cognitive, emotional, and the US and globally, we hope this review will help in address-
behavioral domains (e.g., self-efficacy). In addition, some studies ing some of challenges identified in this important area of
included emotional health outcomes as their primary outcomes, study.
whereas others included emotional health as intermediate out-
comes or mediators of other health outcomes. Finally, there was ACKNOWLEDGMENTS
a dearth of intervention studies on certain emotional health con- This research was made possible through a contract with the
structs, such as hopelessness, shame, guilt, regret, fear, neuroticism, National Association of Chronic Disease Directors (NACDD) to
boredom, positive energy, contentment, hardiness, resilience, emo- the University of Washington Health Promotion Research Cen-
tional stability, emotional regulation/control, altruism, capacity to ter and funded by the CDC Healthy Aging Program’s Healthy
care, and happiness. In particular, positive constructs were under- Brain Initiative (U48-DP000050). The findings and conclusions
represented in the available literature. We were not surprised that in this article are those of the authors and do not necessarily rep-
there was limited evidence on interventions to promote emotional resent the official position of the Centers for Disease Control and
health, and particularly any studies lacking in positive emotional Prevention.

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 213 | 354
Snowden et al. Late-life emotional health review

REFERENCES 20. Smith C, Hancock H, Blake-Mortimer J, Eckert K. A randomised comparative

1. Hendrie HC, Albert MS, Butters MA, Gao S, Knopman DS, Launer LJ, et al. The trial of yoga and relaxation to reduce stress and anxiety. Complement Ther Med
NIH cognitive and emotional health project: report of the critical evaluation (2007) 15(2):77–83. doi:10.1016/j.ctim.2006.05.001
study committee. Alzheimers Dement (2006) 2:12–32. doi:10.1016/j.jalz.2005. 21. Dai Y, Zhang S, Yamamoto J, Ao M, Belin TR, Cheung F, et al. Cogni-
11.004 tive behavioral therapy of minor depressive symptoms in elderly Chinese
2. Centers for Disease Control and Prevention (CDC) and National Association Americans: a pilot study. Community Ment Health J (1999) 35(6):537–42.
of Chronic Disease Directors (NACDD). The State of Mental Health and Aging doi:10.1023/A:1018763302198
in America Issue Brief 1: What Do the Data Tell Us?. National Association of 22. Hui E, Lee PS, Woo J. Management of urinary incontinence in older
Chronic Disease Directors (2008). Available from: http://www.cdc.gov/aging/ women using videoconferencing versus conventional management: a ran-
pdf/mental_health.pdf domized controlled trial. J Telemed Telecare (2006) 12(7):343–7. doi:10.1258/
3. Centers for Disease Control and Prevention (CDC). Behavioral Risk Factor 135763306778682413
Surveillance System Survey Data. Atlanta, GA: U.S. Department of Health and 23. Kretzer K, Davis J, Easa D, Johnson J, Harrigan R. Self identity through
Human Services, Centers for Disease Control and Prevention (2006). Ho’oponopono as adjunctive therapy for hypertension management. Ethn Dis
4. Windle G, Linck P, Morgan R, Hughes D, Burholt V, Reeves C, et al. Public Health (2007) 17(4):624–8.
Interventions to Promote Mental Well-Being in People Aged 65 and Over: Sys- 24. Wright CC, Barlow JH, Turner AP, Bancroft GV. Self-management training for
tematic Review of Effectiveness and Costs-Effectiveness. (2008). Available from: people with chronic disease: an exploratory study. Br J Health Psychol (2003)
http://www.nice.org.uk/guidance/index.jsp?action=folder&o=36395 8(Pt 4):465–76. doi:10.1348/135910703770238310
5. National Institutes of Health (NIH). Cognitive and Emotional Health 25. Ishizaki J, Meguro K, Ohe K, Kimura E, Tsuchiya E, Ishii H, et al. Therapeu-
Project: The Healthy Brain. (2001). Available from: http://trans.nih.gov/cehp/ tic psychosocial intervention for elderly subjects with very mild Alzheimer
HBPemot.htm disease in a community: the tajiri project. Alzheimer Dis Assoc Disord (2002)
6. Stevens A, Coon D, Wisniewski S, Vance D, Arguelles S, Belle S, et al. Measure- 16(4):261–9. doi:10.1097/00002093-200210000-00008
ment of leisure time satisfaction in family caregivers. Aging Ment Health (2004) 26. Brody BL, Roch-Levecq AC, Gamst AC, Maclean K, Kaplan RM, Brown SI.
8:450–9. doi:10.1080/13607860410001709737 Self-management of age-related macular degeneration and quality of life:
7. Dickinson P, Neilson G, Agee M. The sustainability of mentally healthy a randomized controlled trial. Arch Ophthalmol (2002) 120(11):1477–83.
schools initiatives: insights from the experiences of a co-educational secondary doi:10.1001/archopht.120.11.1477
school in Aotearoa/New Zealand. Int J Ment Health Promot (2004) 6(2):34–9. 27. Cunningham AJ. Integrating spirituality into a group psychological ther-
doi:10.1080/14623730.2004.9721929 apy program for cancer patients. Integr Cancer Ther (2005) 4(2):178–86.
8. Haley W, Roth D, Coleton M, Ford G, West C, Collins R, et al. Appraisal, cop- doi:10.1177/1534735405275648
ing, and social support as mediators of well-being in black and white family 28. Evcik D, Sonel B. Effectiveness of a home-based exercise therapy and walk-
caregivers of patients with Alzheimer’s disease. J Consult Clin Psychol (1996) ing program on osteoarthritis of the knee. Rheumatol Int (2002) 22(3):103–6.
64:121–9. doi:10.1037/0022-006X.64.1.121 doi:10.1007/s00296-002-0198-7
9. Briss P, Zaza S, Pappaioanou M, Fielding J, Wright-De Agüero L, Truman BI, 29. Duggleby WD, Degner L, Williams A, Wright K, Cooper D, Popkin D, et al. Liv-
et al. Developing an evidence-based guide to community preventive services- ing with hope: initial evaluation of a psychosocial hope intervention for older
methods. Am J Prev Med (2000) 18(1S):35–43. doi:10.1016/S0749-3797(99) palliative home care patients. J Pain Symptom Manage (2007) 33(3):247–57.
00119-1 doi:10.1016/j.jpainsymman.2006.09.013
10. Norris S, Nichols PJ, Caspersen CJ, Glasgow RE, Engelgau MM, Jack L, et al. 30. Kline KS, Scott LD, Britton AS. The use of supportive-educative and mutual
The effectiveness of disease and case management for people with diabetes: goal-setting strategies to improve self-management for patients with heart
a systematic review. Am J Prev Med (2002) 22(4S):15–38. doi:10.1016/S0749- failure. Home Healthc Nurse (2007) 25(8):502–10. doi:10.1097/01.NHH.
3797(02)00423-3 0000289104.60043.7d
11. Frederick JT, Steinman LE, Prohaska T, Satariano WA, Bruce M, Bryant L, et al. 31. Scott LD, Setter-Kline K, Britton AS. The effects of nursing interventions to
Community-based treatment of late life depression—an expert panel informed enhance mental health and quality of life among individuals with heart failure.
literature review. Am J Prev Med (2007) 33(3):222–49. doi:10.1016/j.amepre. Appl Nurs Res (2004) 17(4):248–56. doi:10.1016/j.apnr.2004.09.004
2007.04.035 32. Lorig KR, Sobel DS, Stewart AL, Brown BW Jr, Bandura A, Ritter P, et al. Evi-
12. Ware JE, Sherbourne CD. The MOS 36-item short form health survey (SF- dence suggesting that a chronic disease self-management program can improve
36). I. Conceptual framework and item selection. Med Care (1992) 30:473–83. health status while reducing hospitalization: a randomized trial. Med Care
doi:10.1097/00005650-199206000-00002 (1999) 37(1):5–14. doi:10.1097/00005650-199901000-00003
13. Mizuno E, Hosak T, Ogihara R, Higano H, Mano Y. Effectiveness of a stress 33. Davis LL. Telephone-based interventions with family caregivers: a feasibility
management program for family caregivers of the elderly at home. J Med Dent study. J Fam Nurs (1998) 4(3):255–70. doi:10.1177/107484079800400303
Sci (1999) 46(4):145–53. 34. Mant J, Winner S, Roche J, Wade DT. Family support for stroke: one year fol-
14. Coon DW, Thompson L, Steffen A, Sorocco K, Gallagher-Thompson D. low up of a randomised controlled trial. J Neurol Neurosurg Psychiatry (2005)
Anger and depression management: psychoeducational skill training interven- 76(7):1006–8. doi:10.1136/jnnp.2004.048991
tions for women caregivers of a relative with dementia. Gerontologist (2003) 35. Barnason SL, Zimmerman L, Nieveen J, Schmaderer M, Carranza B, Reilly S.
43(5):678–89. doi:10.1093/geront/43.5.678 Impact of a home communication intervention for coronary artery bypass
15. Steffen AM. Anger management for dementia caregivers: a preliminary study graft patients with ischemic heart failure on self-efficacy, coronary disease
using video and telephone interventions. Behav Ther (2000) 31(2):281–99. risk factor modification, and functioning. Heart Lung (2003) 32(3):147–58.
doi:10.1080/13607860903420989 doi:10.1016/S0147-9563(03)00036-0
16. López J, Crespo M, Zarit SH. Assessment of the efficacy of a stress manage- 36. Grant JS, Elliott TR, Weaver M, Bartolucci AA, Giger JN. Telephone interven-
ment program for informal caregivers of dependent older adults. Gerontologist tion with family caregivers of stroke survivors after rehabilitation. Stroke (2002)
(2007) 47(2):205–14. doi:10.1093/geront/47.2.205 33(8):2060–5. doi:10.1161/01.STR.0000020711.38824.E3
17. McHugh F, Lindsay GM, Hanlon P, Hutton I, Brown MR, Morrison C, et al. 37. Wallace JI, Buchner DM, Grothaus L, Leveille S, Tyll L, LaCroix AZ, et al.
Nurse led shared care for patients on the waiting list for coronary artery Implementation and effectiveness of a community-based health promotion
bypass surgery: a randomised controlled trial. Heart (2001) 86(3):317–23. program for older adults. J Gerontol A Biol Sci Med Sci (1998) 53(4):M301–6.
doi:10.1136/heart.86.3.317 doi:10.1093/gerona/53A.4.M301
18. Barlow JH, Turner AP, Wright CC. A randomized controlled study of the 38. Markle-Reid M, Weir R, Browne G, Roberts J, Gafni A, Henderson S. Health
arthritis self-management programme in the UK. Health Educ Res (2000) promotion for frail older home care clients. J Adv Nurs (2006) 54(3):381–95.
15(6):665–80. doi:10.1093/her/15.6.665 doi:10.1111/j.1365-2648.2006.03817.x
19. Williams SA, Schreier AM. The role of education in managing fatigue, anxi- 39. Hughes RB, Robinson-Whelen S, Taylor HB, Hall JW. Stress self-management:
ety, and sleep disorders in women undergoing chemotherapy for breast cancer. an intervention for women with physical disabilities. Womens Health Issues
Appl Nurs Res (2005) 18(3):138–47. doi:10.1016/j.apnr.2004.08.005 (2006) 16(6):389–99. doi:10.1016/j.whi.2006.08.003

www.frontiersin.org April 2015 | Volume 2 | Article 213 | 355

Snowden et al. Late-life emotional health review

40. Pischke CR, Weidner G, Elliott-Eller M, Ornish D. Lifestyle changes and clinical 60. King AC, Baumann K, O’Sullivan P, Wilcox S, Castro C. Effects of moderate-
profile in coronary heart disease patients with an ejection fraction of <or=40% intensity exercise on physiological, behavioral, and emotional responses to
or >40% in the multicenter lifestyle demonstration project. Eur J Heart Fail family caregiving: a randomized controlled trial. J Gerontol A Biol Sci Med Sci
(2007) 9(9):928–34. doi:10.1016/j.ejheart.2007.05.009 (2002) 57(1):M26–36. doi:10.1093/gerona/57.1.M26
41. Kocken PL, Voorham AJJ. Effects of a peer-led senior health education 61. Focht BC, Brawley LR, Rejeski WJ, Ambrosius WT. Group-mediated activity
program. Patient Educ Couns (1998) 34(1):15–23. doi:10.1016/S0738-3991(98) counseling and traditional exercise therapy programs: effects on health-related
00038-X quality of life among older adults in cardiac rehabilitation. Ann Behav Med
42. Swerissen H, Belfrage J, Weeks A, Jordan L, Walker C, Furler J, et al. A ran- (2004) 28(1):52–61. doi:10.1207/s15324796abm2801_7
domised control trial of a self-management program for people with a chronic 62. Gary R. Exercise self-efficacy in older women with diastolic heart failure: results
illness from Vietnamese, Chinese, Italian and Greek backgrounds. Patient Educ of a walking program and education intervention. J Gerontol Nurs (2006)
Couns (2006) 64(1–3):360–8. doi:10.1016/j.pec.2006.04.003 32(7):31–9.
43. Barlow JH, Williams B, Wright CC. Instilling the strength to fight the 63. Gendron CE, Poitras LR, Engels ML, Dastoor DP, Sirota SE, Barza SL, et al.
pain and get on with life’: learning to become an arthritis self-manager Skills training with supporters of the demented. J Am Geriatr Soc (1986)
through an adult education programme. Health Educ Res (1999) 14(4):533–44. 34(12):875–80.
doi:10.1093/her/14.4.533 64. Halbert JA, Silagy CA, Finucane PM, Withers RT, Hamdorf PA. Physical activity
44. Lorig K, Gonzalez VM, Ritter P. Community-based Spanish language arthri- and cardiovascular risk factors: effect of advice from an exercise specialist in
tis education program: a randomized trial. Med Care (1999) 37(9):957–63. Australian general practice. Med J Aust (2000) 173(2):84–7.
doi:10.1097/00005650-199909000-00011 65. Slivinske LR, Fitch VL. The effect of control enhancing interventions on the
45. Solomon P, Draine J, Mannion E, Meisel M. Impact of brief family psychoedu- well-being of elderly individuals living in retirement communities. Gerontolo-
cation on self-efficacy. Schizophr Bull (1996) 22(1):41–50. doi:10.1093/schbul/ gist (1987) 27(2):176–81. doi:10.1093/geront/27.2.176
22.1.41 66. Acorn S. Assisting families of head-injured survivors through a family support
46. Solomon P, Draine J, Mannion E, Meisel M. Effectiveness of two mod- programme. J Adv Nurs (1995) 21(5):872–7. doi:10.1046/j.1365-2648.1995.
els of brief family education: retention of gains by family members of 21050872.x
adults with serious mental illness. Am J Orthopsychiatry (1997) 67(2):177–86. 67. Promoting Positive Mental Health Through A Socio-Ecological Approach. Centers
doi:10.1037/h0080221 for Addictions Research of BC (2011). Available from: http://www.heretohelp.
47. Forster A, Young J. Specialist nurse support for patients with stroke in the bc.ca/sites/default/files/CARBC-campusdoc1.pdf
community: a randomised controlled trial. BMJ (1996) 312(7047):1642–6. 68. Chien MY, Yang RS, Tsauo JY. Home-based trunk-strengthening exercise for
doi:10.1136/bmj.312.7047.1642 osteoporotic and osteopenic postmenopausal women without fracture – a pilot
48. Liu-Ambrose T, Khan KM, Eng JJ, Lord SR, McKay HA. Balance confidence study. Clin Rehabil (2005) 19(1):28–36. doi:10.1191/0269215505cr844oa
improves with resistance or agility training. Gerontology (2004) 50(6):373–82. 69. Etkin CD, Prohaska TR, Harris BA, Latham N, Jette A. Feasibility of implement-
doi:10.1159/000080175 ing the strong for life program in community settings. Gerontologist (2006)
49. Ross MC, Bohannon AS, Davis DC, Gurchiek L. The effects of a short-term 46(2):284–92. doi:10.1093/geront/46.2.284
exercise program on movement, pain, and mood in the elderly: results of a pilot 70. Hébert R, Lévesque L, Vézina J, Lavoie JP, Ducharme F, Gendron C, et al. Effi-
study. J Holist Nurs (1999) 17(2):139–47. doi:10.1177/089801019901700203 cacy of a psychoeducative group program for caregivers of demented persons
50. Anderson G. Chronic Care: Making the Case for Ongoing Care. Robert Wood living at home: a randomized controlled trial. J Gerontol B Psychol Sci Soc Sci
Johnson Foundation: (2010). Available from: http://www.rwjf.org/content/ (2003) 58B(1):S58–67. doi:10.1093/geronb/58.1.S58
dam/farm/reports/reports/2010/rwjf54583 71. Stewart AL, Mills KM, Sepsis PG, King AC, McLellan BY, Roitz K, et al. Evalua-
51. Daley AJ, Crank H, Saxton JM, Mutrie N, Coleman R, Roalfe A. Randomized tion of CHAMPS, a physical activity promotion program for older adults. Ann
trial of exercise therapy in women treated for breast cancer. J Clin Oncol (2007) Behav Med (1997) 19(4):353–61. doi:10.1007/BF02895154
25(13):1713–21. doi:10.1200/JCO.2006.09.5083 72. Stewart M, Craig D, MacPherson K, Alexander S. Promoting positive
52. Kolt GS, Schofield GM, Kerse N, Garrett N, Oliver M. Effect of telephone affect and diminishing loneliness of widowed seniors through a support
counseling on physical activity for low-active older people in primary care: a intervention. Public Health Nurs (2001) 18(1):54–63. doi:10.1046/j.1525-1446.
randomized, controlled trial. J Am Geriatr Soc (2007) 55(7):986–92. doi:10. 2001.00054.x
1111/j.1532-5415.2007.01203.x 73. Molloy GJ, Johnston DW, Gao C, Witham MD, Gray JM, Argo IS. Effects of
53. Kutner NG, Barnhart H, Wolf SL, McNeely E, Xu T. Self-report benefits of Tai an exercise intervention for older heart failure patients on caregiver burden
Chi practice by older adults. J Gerontol B Psychol Sci Soc Sci (1997) 52(5):242–6. and emotional distress. Eur J Cardiovasc Prev Rehabil (2006) 13(3):381–7.
doi:10.1093/geronb/52B.5.P242 doi:10.1097/01.hjr.0000198916.60363.85
54. Hartford K, Wong C, Zakaria D. Randomized controlled trial of a telephone 74. Centers for Disease Control and Prevention. Public Health Action Plan to Inte-
intervention by nurses to provide information and support to patients and their grate Mental Health Promotion and Mental Illness Prevention with Chronic Dis-
partners after elective coronary artery bypass graft surgery: effects of anxiety. ease Prevention, 2011–2015. Atlanta: U.S. Department of Health and Human
Heart Lung (2002) 31(3):199–206. doi:10.1067/mhl.2002.122942 Services (2011).
55. Smart N, Haluska B, Jeffriess L, Marwick TH. Predictors of a sustained response 75. Singh NA, Clements KM, Singh MA. The efficacy of exercise as a
to exercise training in patients with chronic heart failure: a telemonitoring long-term antidepressant in elderly subjects: a randomized, controlled trial.
study. Am Heart J (2005) 150(6):1240–7. doi:10.1016/j.ahj.2005.01.035 J Gerontol A Biol Sci Med Sci (2001) 56(8):M497–504. doi:10.1093/gerona/56.
56. Clark DO, Stump TE, Damush TM. Outcomes of an exercise program for older 8.M497
women recruited through primary care. J Aging Health (2003) 15(3):567–85. 76. Garand L, Buckwalter KC, Lubaroff D, Tripp-Reimer T, Frantz RA, Ansley TN.
doi:10.1177/0898264303253772 A pilot study of immune and mood outcomes of a community-based inter-
57. Courneya KS, Friedenreich CM, Quinney HA, Fields AL, Jones LW, Fairey vention for dementia caregivers: the PLST intervention. Arch Psychiatr Nurs
AS. A randomized trial of exercise and quality of life in colorectal cancer (2002) 16(4):156–67. doi:10.1053/apnu.2002.34392
survivors. Eur J Cancer Care (2003) 12(4):347–57. doi:10.1046/j.1365-2354. 77. Inokuchi S, Matsusaka N, Hayashi T, Shindo H. Feasibility and effective-
2003.00437.x ness of a nurse-led community exercise programme for prevention of falls
58. U.S. Department of Health and Human Services. Multiple Chronic among frail elderly people: a multi-centre controlled trial. J Rehabil Med (2007)
Conditions—A Strategic Framework: Optimum Health and Quality of Life for 39(6):479–85. doi:10.2340/16501977-0080
Individuals with Multiple Chronic Conditions. Washington, DC: (2010). Avail- 78. Losito J, Murphy S, Thomas M. The effects of group exercise on fatigue and
able from: http://www.hhs.gov/ash/initiatives/mcc/mcc_framework.pdf quality of life during cancer treatment. Oncol Nurs Forum (2006) 33(4):821–5.
59. King AC, Taylor CB, Haskell WL. Effects of differing intensities and doi:10.1188/06.ONF.821-825
formats of 12 months of exercise training on psychological outcomes in 79. Anderson RT, King A, Stewart AL, Camacho F, Rejeski WJ. Physical activity
older adults. Health Psychol (1993) 12(4):292–300. doi:10.1037/0278-6133.12. counseling in primary care and patient well-being: do patients benefit? Ann
5.405 Behav Med (2005) 30(2):146–54. doi:10.1207/s15324796abm3002_7

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 213 | 356
Snowden et al. Late-life emotional health review

80. Munro JF, Nicholl JP, Brazier JE, Davey R, Cochrane T. Cost effectiveness 101. Fisher K, Li FA. Community-based walking trial to improve neighborhood
of a community based exercise programme in over 65 year olds: cluster quality of life in older adults: a multilevel analysis. Ann Behav Med (2004)
randomised trial. J Epidemiol Community Health (2004) 58(12):1004–10. 28(3):186–94. doi:10.1207/s15324796abm2803_7
doi:10.1136/jech.2003.014225 102. Castro CM, Wilcox S, O’Sullivan P, Baumann K, King AC. An exercise program
81. Mutrie N, Campbell AM, Whyte F, McConnachie A, Emslie C, Lee L, et al. for women who are caring for relatives with dementia. Psychosom Med (2002)
Benefits of supervised group exercise programme for women being treated for 64(3):458–68. doi:10.1097/00006842-200205000-00010
early stage breast cancer: pragmatic randomised controlled trial. BMJ (2007) 103. Pinto BM, Frierson GM, Rabin C, Trunzo JJ, Marcus BH. Home-based phys-
334(7592):517. doi:10.1136/bmj.39094.648553.AE ical activity intervention for breast cancer patients. J Clin Oncol (2005)
82. Toseland RW, Rossiter CM, Peak T, Smith GC. Comparative effectiveness of 23(15):3577–87. doi:10.1200/JCO.2005.03.080
individual and group interventions to support family caregivers. Soc Work 104. Stewart AL, King AC, Haskell WL. Endurance exercise and health-related
(1990) 35(3):209–17. quality of life in 50-65 year-old adults. Gerontologist (1993) 33(6):782–9.
83. Bogat G, Jason LA. An evaluation of two visiting programs for elderly commu- doi:10.1093/geront/33.6.782
nity residents. Int J Aging Hum Dev (1983) 17(4):267–80. doi:10.2190/6AQX- 105. Hughes SL, Seymour RB, Campbell RT, Huber G, Pollak N, Sharma L, et al.
1TDM-DEL4-3T6L Long-term impact of fit and strong! on older adults with osteoarthritis. Geron-
84. Jette AM, Harris BA, Sleeper L, Lachman ME, Heislein D, Giorgetti M, et al. A tologist (2006) 46(6):801–14. doi:10.1093/geront/46.6.801
home-based exercise program for nondisabled older adults. J Amer Geriatr Soc 106. Senuzun F, Fadiloglu C, Burke LE, Payzin S. Effects of home-based car-
(1996) 44(6):644–9. diac exercise program on the exercise tolerance, serum lipid values and self-
85. Lin MR, Wolf SL, Hwang HF, Gong SY, Chen CY. A randomized, controlled efficacy of coronary patients. Eur J Cardiovasc Prev Rehabil (2006) 13(4):640–5.
trial of fall prevention programs and quality of life in older fallers. J Am Geriatr doi:10.1097/01.hjr.0000198445.41860.ec
Soc (2007) 55(4):499–506. doi:10.1111/j.1532-5415.2007.01146.x 107. Shin Y. The effects of a walking exercise program on physical function and emo-
86. Austin EN, Johnston YA, Morgan LL. Community gardening in a senior center: tional state of elderly Korean women. Public Health Nurs (1999) 16(2):146–54.
a therapeutic intervention to improve the health of older adults. Ther Recreation doi:10.1046/j.1525-1446.1999.00146.x
J (2006) 40(1):48–56. 108. Heller K, Thompson MG, Trueba PE, Hogg JR, Vlachos-Weber I. Peer support
87. Chen KM, Snyder M, Kirchbaum K. Tai chi and well-being of Taiwanese telephone dyads for elderly women: was this the wrong intervention? Am J
community-dwelling elders. Clin Gerontol (2001) 24(3–4):137–56. doi:10. Community Psychol (1991) 19(1):53–74. doi:10.1007/BF00942253
1300/J018v24n03_12 109. Dröes RM, Breebaart E, Ettema TP, van Tilburg W, Mellenbergh GJ. Effect
88. Devereux K, Robertson D, Briffa NK. Effects of a water-based program on of integrated family support versus day care only on behavior and mood of
women 65 years and over: a randomised controlled trial. Aust J Physiother patients with dementia. Int Psychogeriatr (2000) 12(1):99–115. doi:10.1017/
(2005) 51(2):102–8. doi:10.1016/S0004-9514(05)70038-6 S1041610200006232
89. Logiudice D, Waltrowicz W, Brown K, Burrows C, Ames D, Flicker L. Do 110. Ruhland JL, Shields RK. The effects of a home exercise program on impair-
memory clinics improve the quality of life of carers? A randomized pilot ment and health-related quality of life in persons with chronic peripheral neu-
trial. Int J Geriatr Psychiatry (1999) 14(8):626–32. doi:10.1002/(SICI)1099- ropathies. Phys Ther (1997) 77(10):1026–39.
1166(199908)14:8<626::AID-GPS990>3.0.CO;2-5 111. Alexander MJL, Butcher JE, MacDonald PB. Effect of a water exercise pro-
90. Li F, Fisher KJ, Harmer P, McAuley E. Falls self-efficacy as a mediator of fear of gram on walking gait, flexibility, strength, self-reported disability and other
falling in an exercise intervention for older adults. J Gerontol B Psychol Sci Soc psycho-social measures of older individuals with arthritis. Physiother Can
Sci (2005) 60(1):34–40. doi:10.1093/geronb/60.1.P34 (2001) 53(3):203–11.
91. Means KM, O’Sullivan PS, Rodell DE. Psychosocial effects of an exercise 112. Bennett JA, Young HM, Nail LM, Winters-Stone K, Hanson G. A telephone-
program in older persons who fall. J Rehabil Res Dev (2003) 40(1):49–58. only motivational intervention to increase physical activity in rural adults:
doi:10.1682/JRRD.2003.01.0049 a randomized controlled trial. Nurs Res (2008) 57(1):24–32. doi:10.1097/01.
92. Resnick B, Orwig D, Yu-Yahiro J, Hawkes W, Shardell M, Hebel JR, et al. Testing NNR.0000280661.34502.c1
the effectiveness of the exercise plus program in older women post-hip fracture. 113. Dröes RM, Meiland F, Schmitz M, van Tilburg W. Effect of combined support
Ann Behav Med (2007) 34(1):67–76. doi:10.1007/BF02879922 for people with dementia and carers versus regular day care on behaviour and
93. Drummond MF, Mohide EA, Tew M, Streiner DL, Pringle DM, Gilbert mood of persons with dementia: results from a multi-centre implementation
JR. Economic evaluation of a support program for caregivers of demented study. Int J Geriatr Psychiatry (2004) 19(7):673–84. doi:10.1002/gps.1142
elderly. Int J Technol Assess Health Care (1991) 7(2):209–19. doi:10.1017/ 114. Emery CF, Gatz M. Psychological and cognitive effects of an exercise pro-
S0266462300005109 gram for community-residing older adults. Gerontologist (1990) 30(2):184–8.
94. Hartman CA, Manos TM, Winter C, Hartman DM, Li B, Smith JC. Effects of doi:10.1093/geront/30.2.184
T’ai Chi training on function and quality of life indicators in older adults with 115. Foley A, Halbert J, Hewitt T, Crotty M. Does hydrotherapy improve strength
osteoarthritis. J Am Geriatr Soc (2000) 48(12):1553–9. and physical function in patients with osteoarthritis – a randomised controlled
95. Taylor-Piliae RE, Haskell WL, Waters CM, Froelicher ES. Change in perceived trial comparing a gym based and a hydrotherapy based strengthening pro-
psychosocial status following a 12-week Tai Chi exercise programme. J Adv gramme. Ann Rheum Dis (2003) 62(12):1162–7. doi:10.1136/ard.2002.005272
Nurs (2006) 54(3):313–29. doi:10.1111/j.1365-2648.2006.03809.x 116. Jeong S, Kim MT. Effects of a theory-driven music and movement program for
96. Cheung BM, Lo JL, Fong DY, Chan MY, Wong SH, Wong VC, et al. Ran- stroke survivors in a community setting. Appl Nurs Res (2007) 20(3):125–31.
domised controlled trial of qigong in the treatment of mild essential doi:10.1016/j.apnr.2007.04.005
hypertension. J Hum Hypertens (2005) 19(9):697–704. doi:10.1038/sj.jhh. 117. Savage P, Ricci MA, Lynn M, Gardner A, Knight S, Brochu M, et al. Effects of
1001884 home versus supervised exercise for patients with intermittent claudication.
97. Christopher KA, Morrow LL. Evaluating a community-based exercise program J Cardiopulm Rehabil (2001) 21(3):152–7. doi:10.1097/00008483-200105000-
for women cancer survivors. Appl Nurs Res (2004) 17(2):100–8. doi:10.1016/ 00006
S0897-1897(04)00024-2 118. Wilcox S, Dowda M, Griffin SF, Rheaume C, Ory MG, Leviton L, et al. Results
98. Gross CR, Kreitzer MJ, Russas V, Treesak C, Frazier PA, Hertz MI. Mindfulness of the first year of active for life: translation of 2 evidence-based physical activ-
meditation to reduce symptoms after organ transplant: a pilot study. Adv Mind ity programs for older adults into community settings. Am J Health (2006)
Body Med (2004) 20(2):20–9. 96(7):1201–9. doi:10.2105/AJPH.2005.074690
99. Papaioannou A, Adachi JD, Winegard K, Ferko N, Parkinson W, Cook RJ, et al. 119. Beaupre LA, Lier D, Davies DM, Johnston DB. The effect of a preoperative
Efficacy of home-based exercise for improving quality of life among elderly exercise and education program on functional recovery, health related quality
women with symptomatic osteoporosis-related vertebral fractures. Osteoporos of life, and health service utilization following primary total knee arthroplasty.
Int (2003) 14(8):677–82. doi:10.1007/s00198-003-1423-2 J Rheumatol (2004) 31(6):1166–73.
100. Smith TL, Toseland RW. The effectiveness of a telephone support pro- 120. Kerse N, Elley C, Robinson E, Arroll B. Is physical activity counseling effective
gram for caregivers of frail older adults. Gerontologist (2006) 46(5):620–9. for older people? A cluster randomized, controlled trial in primary care. J Am
doi:10.1093/geront/46.5.620 Geriatr Soc (2005) 53(11):1951–6. doi:10.1111/j.1532-5415.2005.00466.x

www.frontiersin.org April 2015 | Volume 2 | Article 213 | 357

Snowden et al. Late-life emotional health review

121. van den Heuvel ET, Witte LP, Stewart RE, Schure LM, Sanderman R, Meyboom- 140. Centers for Disease Control and Prevention. Physical Activity Guide-
de Jong B. Long-term effects of a group support program and an individual lines for Older Adults. Division of Nutrition, Physical Activity, and Obe-
support program for informal caregivers of stroke patients: which caregivers sity, National Center for Chronic Disease Prevention and Health Promo-
benefit the most? Patient Educ Couns (2002) 47(4):291–9. doi:10.1016/S0738- tion (2011). Available from: http://www.cdc.gov/physicalactivity/everyone/
3991(01)00230-0 guidelines/olderadults.html
122. Barrett CJ, Smerdely P. A comparison of community-based resistance exer- 141. Glasgow RE, McKay HG, Piette JD, Reynolds KD. The re-aim framework for
cise and flexibility exercise for seniors. Aust J Physiother (2002) 48(3):215–9. evaluating interventions: what can it tell us about approaches to chronic ill-
doi:10.1016/S0004-9514(14)60226-9 ness management? Patient Educ Couns (2001) 44:119–27. doi:10.1016/S0738-
123. Hernández MT, Rubio TM, Ruiz FO, Riera HS, Gil RS, Gómez JC. Results 3991(00)00186-5
of a home-based training program for patients with COPD. Chest (2000) 142. Ducharme F, Trudeau D. Qualitative evaluation of a stress management inter-
118(1):106–14. doi:10.1378/chest.118.1.106 vention for elderly caregivers at home: a constructivist approach. Issues Ment
124. Murray J, Manela M, Shuttleworth A, Livingston G. An intervention study with Health Nurs (2002) 23(7):691–713. doi:10.1080/01612840290052820
husband and wife carers of older people with a psychiatric illness. J Affect Disord 143. Harding R, Higginson IJ, Leam C, Donaldson N, Pearce A, George R, et al.
(1997) 46(3):279–84. doi:10.1016/S0165-0327(97)00098-0 Evaluation of a short-term group intervention for informal carers of patients
125. Stenstrom CH. Home exercise in rheumatoid arthritis functional class II: goal attending a home palliative care service. J Pain Symptom Manage (2004)
setting versus pain attention. J Rheumatol (1994) 21(4):627–34. 27(5):396–408. doi:10.1016/j.jpainsymman.2003.09.012
126. Greendale GA, Salem GJ, Young JT, Damesyn M, Marion M, Wang MY, et al. 144. Hunkeler EM, Meresman JF, Hargreaves WA, Fireman B, Berman WH, Kirsch
A randomized trial of weighted vest use in ambulatory older adults: strength, AJ, et al. Efficacy of nurse telehealth care and peer support in augmenting
performance, and quality of life outcomes. J Am Geriatr Soc (2000) 48(3): treatment of depression in primary care. Arch Fam Med (2000) 9(8):700–8.
305–11. doi:10.1001/archfami.9.8.700
127. Hughes SL, Seymour RB, Campbell R, Pollak N, Huber G, Sharma L. Impact of 145. Malde S. Guided autobiography: a counseling tool for older adults. J Couns
the fit and strong intervention on older adults with osteoarthritis. Gerontologist Dev (1988) 66(6):290–3. doi:10.1002/j.1556-6676.1988.tb00872.x
(2004) 44(2):217–28. doi:10.1093/geront/44.2.217 146. Mock V, Dow KH, Meares CJ, Grimm PM, Dienemann JA, Haisfield-Wolfe
128. Larson J, Franzén-Dahlin A, Billing E, Arbin M, Murray V, Wredling R. The ME, et al. Effects of exercise on fatigue, physical functioning, and emotional
impact of a nurse-led support and education programme for spouses of stroke distress during radiation therapy for breast cancer. Oncol Nurs Forum (1997)
patients: a randomized controlled trial. J Clin Nurs (2005) 14(8):995–1003. 24(6):991–1000.
doi:10.1111/j.1365-2702.2005.01206.x 147. Oken BS, Zajdel D, Kishiyama S, Flegal K, Dehen C, Haas M, et al. Randomized,
129. Tsutsumi T, Don BM, Zaichkowsky LD, Delizonna LL. Physical fitness and psy- controlled, six-month trial of yoga in healthy seniors: effects on cognition and
chological benefits of strength training in community dwelling older adults. quality of life. Altern Ther Health Med (2006) 12(1):40–7.
Appl Human Sci (1997) 16(6):257–66. doi:10.2114/jpa.16.257 148. Slivinske LR, Kosberg JI. Assessing the effect of a personal health management
130. van Uffelen JG, Chin A, Paw MJ, Hopman-Rock M, van Mechelen W. The effect system within retirement communities: a preliminary investigation. Gerontol-
of walking and vitamin B supplementation on quality of life in community- ogist (1984) 24(3):280–5. doi:10.1093/geront/24.3.280
dwelling adults with mild cognitive impairment: a randomized, controlled trial. 149. Villaverde-Gutiérrez C, Araújo E, Cruz F, Roa JM, Barbosa W, Ruíz-Villaverde
Qual Life Res (2007) 16(7):1137–46. doi:10.1007/s11136-007-9219-z G. Quality of life of rural menopausal women in response to a customized
131. Chin A, Paw MJ, de Jong N, Schouten EG, van Staveren WA, Kok FJ. Phys- exercise programme. J Adv Nurs (2006) 54(1):11–9. doi:10.1111/j.1365-2648.
ical exercise or micronutrient supplementation for the wellbeing of the frail 2006.03784.x
elderly? A randomised controlled trial. Br J Sports Med (2002) 36(2):126–31. 150. Hauer K, Specht N, Schuler M, Bärtsch P, Oster P. Intensive physical train-
doi:10.1136/bjsm.36.2.126 ing in geriatric patients after severe falls and hip surgery. Age Ageing (2002)
132. King AC, Pruitt LA, Phillips W, Oka R, Rodenburg A, Haskell WL. Com- 31(1):49–57. doi:10.1093/ageing/31.1.49
parative effects of two physical activity programs on measured and per- 151. Headley JA, Ownby KK, John LD. The effect of seated exercise on fatigue and
ceived physical functioning and other health-related quality of life out- quality of life in women with advanced breast cancer. Oncol Nurs Forum (2004)
comes in older adults. J Gerontol A Biol Sci Med Sci (2000) 55(2):M74–83. 31(5):977–83. doi:10.1188/04.ONF.977-983
doi:10.1093/gerona/55.2.M74 152. Helbostad JL, Sletvold O, Moe-Nilssen R. Home training with and with-
133. Bravo G, Gauthier P, Roy PM, Payette H, Gaulin P. A weight-bearing, water- out additional group training in physically frail old people living at home:
based exercise program for osteopenic women: its impact on bone, func- effect on health-related quality of life and ambulation. Clin Rehabil (2004)
tional fitness, and well-being. Arch Phys Med Rehabil (1997) 78(12):1375–80. 18(5):498–508. doi:10.1191/0269215504cr761oa
doi:10.1016/S0003-9993(97)90313-0 153. Sims J, Hill K, Davidson S, Gunn J, Huang N. Exploring the feasibility of a
134. de Vreede PL, van Meeteren NL, Samson MM, Wittink HM, Duursma SA, community-based strength training program for older people with depressive
Verhaar HJ. The effect of functional tasks exercise and resistance exercise on symptoms and its impact on depressive symptoms. BMC Geriatr (2006) 6:18.
health-related quality of life and physical activity. A randomised controlled doi:10.1186/1471-2318-6-18
trial. Gerontology (2007) 53(1):12–20. doi:10.1159/000095387 154. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psy-
135. Mant J, Carter J, Wade DT, Winner S. Family support for stroke: a randomised chiatr Scand (1983) 67:361–70. doi:10.1111/j.1600-0447.1983.tb09716.x
controlled trial. Lancet (2000) 356(9232):808–13. doi:10.1016/S0140-6736(05) 155. Kennedy A, Reeves D, Bower P, Middleton E, Richardson G, Gardner C, et al.
67276-5 The effectiveness and cost effectiveness of a national lay-led self care sup-
136. Blumenthal JA, Emery CF, Madden DJ, Schniebolk S, Walsh-Riddle M, George port programme for patients with long-term conditions: a pragmatic ran-
LK. Long-term effects of exercise on psychological functioning in older men domised controlled trial. J Epidemiol Community Health (2007) 61(3):254–61.
and women. J Gerontol (1991) 46(6):352–61. doi:10.1093/geronj/46.6.P352 doi:10.1136/jech.2006.053538
137. Singh NA, Stavrinos TM, Scarbek Y, Galambos G, Liber C, Fiatarone Singh MA. 156. Lorig KR, Ritter PL, Gonzalez VM. Hispanic chronic disease self-management:
A randomized controlled trial of high versus low intensity weight training ver- a randomized community-based outcome trial. Nurs Res (2003) 52(6):361–9.
sus general practitioner care for clinical depression in older adults. J Gerontol doi:10.1097/00006199-200311000-00003
A Biol Sci Med Sci (2005) 60(6):768–76. doi:10.1093/gerona/60.6.768 157. Dongbo F, Fu H, McGowan P, Yi-E S, Lizhen Z, Huiqin Y, et al. Implementation
138. Barker C. The value of home support for cancer patients: a study. Nurs Stand and quantitative evaluation of chronic disease self-management programme in
(1997) 11(32):34–7. doi:10.7748/ns1997.04.11.32.34.c2453 Shanghai, China: randomized controlled trial. Bull World Health Organ (2003)
139. Walsh K, Jones L, Tookman A, Mason C, McLoughlin J, Blizard R, et al. 81(3):174–82.
Reducing emotional distress in people caring for patients receiving spe- 158. Lorig KR, Ritter PL, Laurent DD, Plant K. Internet-based chronic dis-
cialist palliative care. Randomised trial. Br J Psychiatry (2007) 190:142–7. ease self-management: a randomized trial. Med Care (2006) 44(11):964–71.
doi:10.1192/bjp.bp.106.023960 doi:10.1097/01.mlr.0000233678.80203.c1

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 2 | Article 213 | 358
Snowden et al. Late-life emotional health review

159. Griffiths C, Motlib J, Azad A, Ramsay J, Eldridge S, Feder G, et al. Randomised 177. Singh NA, Clements KM, Fiatarone MA. A randomized controlled trial of pro-
controlled trial of a lay-led self-management programme for Bangladeshi gressive resistance training in depressed elders. J Gerontol A Biol Sci Med Sci
patients with chronic disease. Br J Gen Pract (2005) 55(52O):831–7. (1997) 52(1):M27–35. doi:10.1093/gerona/52A.1.M27
160. Berkman ND, Sheridan SL, Donahue KE, Halpern DJ, Viera A, Crotty K, et al. 178. Stevenson JS, Topp R. Effects of moderate and low intensity long-term exer-
Health literacy interventions and outcomes: an updated systematic review. Evid cise by older adults. Res Nurs Health (1990) 13(4):209–18. doi:10.1002/nur.
Rep Technol Assess (Full Rep) (2011) 1999:1–941. 4770130403
161. Coberley C, Rula EY, Pope JE. Effectiveness of health and wellness initiatives 179. Stiggelbout M, Popkema DY, Hopman-Rock M, de Greef M, van Mechelen
for seniors. Popul Health Manag (2011) 14(Suppl 1):S45–50. doi:10.1089/pop. W. Once a week is not enough: effects of a widely implemented group based
2010.0072 exercise programme for older adults: a randomised controlled trial. J Epidemiol
162. Gitlin LN, Chernett NL, Fields Harris L, Palmer D, Hopkins P, Dennis MP. Har- Community Health (2004) 58(2):83–8. doi:10.1136/jech.58.2.83
vest health: translation of the chronic disease self-management program for 180. Williams P, Lord SR. Effects of group exercise on cognitive functioning
older African Americans in a senior setting. Gerontologist (2008) 48:698–705. and mood in older women. Aust N Z J Public Health (1997) 21(1):45–52.
doi:10.1093/geront/48.5.698 doi:10.1111/j.1467-842X.1997.tb01653.x
163. Hochhaltera AK, Song J, Rush J, Sklar L, Stevens A. Making the most of your 181. Wolf SL, Barnhart HX, Kutner NG, McNeely E, Coogler C, Xu T, et al. Selected
healthcare intervention for older adults with multiple chronic illnesses. Patient as the best paper in the 1990s: reducing frailty and falls in older persons: an
Educ Couns (2010) 81:207–13. doi:10.1016/j.pec.2010.01.018 investigation of tai chi and computerized balance training. J Am Geriatr Soc
164. Jerant A, Moore-Hill M, Franks P. Home-based, peer-led chronic illness self- (2003) 51(12):1794–803. doi:10.1046/j.1532-5415.2003.51566.x
management training: findings from a 1-year randomized controlled trial. Ann 182. Yu DS, Lee DT, Woo J, Hui E. Non-pharmacological interventions in older
Fam Med (2009) 7:319–27. doi:10.1370/afm.996 people with heart failure: effects of exercise training and relaxation therapy.
165. Jonker AAGC, Comijs HC, Knipscheer KCPM, Deeg DJH. Promotion of self- Gerontology (2007) 53(2):74–81. doi:10.1159/000096427
management in vulnerable older people: a narrative literature review of out- 183. Greendale GA, Hirsch SH, Hahn TJ. The effect of a weighted vest on per-
comes of the chronic disease self-management program (CDSMP). Eur J Ageing ceived health status and bone density in older persons. Qual Life Res (1993)
(2009) 6:303–14. doi:10.1007/s10433-009-0131-y 2(2):141–52. doi:10.1007/BF00435733
166. Ritter PL, Lee J, Lorig K. Moderators of chronic disease self-management 184. Bravo G, Gauthier P, Roy PM, Payette H, Gaulin P, Harvey M, et al. Impact
programs: who benefits? Chronic Illn (2011) 7:162–72. doi:10.1177/ of a 12-month exercise program on the physical and psychological health of
1742395311399127 osteopenic women. J Am Geriatr Soc (1996) 44(7):756–62.
167. Rose MA, Arenson C, Harrod P, Salkey R, Santana A, Diamond J. Eval-
uation of the chronic disease self-management program with low-income,
Urban, African American older adults. J Community Health Nurs (2008) Conflict of Interest Statement: The authors declare that the research was conducted
25(4):193–202. doi:10.1080/07370010802421178 in the absence of any commercial or financial relationships that could be construed
168. Smeulders ESTF, van Haastregt JCM, Janssen-Boyne JJJ, Stoffers HEJH, van as a potential conflict of interest.
Eijk JTM, Kempen GIJM. Feasibility of a group-based self-management pro-
gram among congestive heart failure patients. Heart Lung (2009) 38:499–512. This paper is included in the Research Topic, “Evidence-Based Programming for Older
doi:10.1016/j.hrtlng.2009.01.007 Adults.” This Research Topic received partial funding from multiple government and
169. Smeulders ESTF, van Haastregt JCM, Ambergen T, Stoffers HEJH, Janssen- private organizations/agencies; however, the views, findings, and conclusions in these
Boyne JJJ, Uszko-Lencer NHKM, et al. Heart failure patients with a lower edu- articles are those of the authors and do not necessarily represent the official position
cational level and better cognitive status benefit most from a self-management of these organizations/agencies. All papers published in the Research Topic received
group programme. Patient Educ Couns (2010) 81:214–21. doi:10.1016/j.pec. peer review from members of the Frontiers in Public Health (Public Health Education
2010.01.003 and Promotion section) panel of Review Editors. Because this Research Topic repre-
170. Tomioka M, Braun KL, Compton M, Tanoue L. Adapting Stanford’s chronic sents work closely associated with a nationwide evidence-based movement in the US,
disease self-management program to Hawaii’s multicultural population. many of the authors and/or Review Editors may have worked together previously in
Gerontologist (2012) 52(1):121–32. doi:10.1093/geront/gnr054 some fashion. Review Editors were purposively selected based on their expertise with
171. Frosch DL, Rincon D, Ochoa S, Mangione CM. Activating seniors to improve evaluation and/or evidence-based programming for older adults. Review Editors were
chronic disease care: results from a pilot intervention study. J Am Geriatr Soc independent of named authors on any given article published in this volume.
(2010) 58(8):1496–503. doi:10.1111/j.1532-5415.2010.02980.x
172. McLeroy KR, Bibeau D, Steckler A, Glanz K. An ecological perspective on Received: 19 June 2014; accepted: 13 October 2014; published online: 27 April 2015.
health promotion programs. Health Educ Q (1988) 15(4):351–77. doi:10.1177/ Citation: Snowden MB, Steinman LE, Carlson WL, Mochan KN, Abraido-Lanza
109019818801500401 AF, Bryant LL, Duffy M, Knight BG, Jeste DV, Leith KH, Lenze EJ, Logsdon RG,
173. Engels HJ, Drouin J, Zhu W, Kazmierski JF. Effects of low-impact, moderate- Satariano WA, Zweiback DJ and Anderson LA (2015) Effect of physical activity,
intensity exercise training with and without wrist weights on functional social support, and skills training on late-life emotional health: a systematic litera-
capacities and mood states in older adults. Gerontology (1998) 44(4):239–44. ture review and implications for public health research. Front. Public Health 2:213.
doi:10.1159/000022018 doi: 10.3389/fpubh.2014.00213
174. Friedland JF, McColl M. Social support intervention after stroke: results of a This article was submitted to Public Health Education and Promotion, a section of the
randomized trial. Arch Phys Med Rehabil (1992) 73(6):573–81. journal Frontiers in Public Health.
175. Groessl EJ, Cronan TA. A cost analysis of self-management programs for Copyright © 2015 Snowden, Steinman, Carlson, Mochan, Abraido-Lanza, Bryant ,
people with chronic illness. Am J Community Psychol (2000) 28(4):455–80. Duffy, Knight , Jeste, Leith, Lenze, Logsdon, Satariano, Zweiback and Anderson. This
doi:10.1023/A:1005184414241 is an open-access article distributed under the terms of the Creative Commons Attri-
176. Richeson NE, Croteau KA, Jones DA, Farmer BC. Effects of a pedometer-based bution License (CC BY). The use, distribution or reproduction in other forums is
intervention on the physical performance and mobility-related self-efficacy of permitted, provided the original author(s) or licensor are credited and that the original
community-dwelling older adults: an interdisciplinary preventive health care publication in this journal is cited, in accordance with accepted academic practice. No
intervention. Ther Recreation J (2006) 40(1):18–32. use, distribution or reproduction is permitted which does not comply with these terms.

www.frontiersin.org April 2015 | Volume 2 | Article 213 | 359

 OPINION ARTICLE
PUBLIC HEALTH
published: 27 April 2015
doi: 10.3389/fpubh.2015.00018

EvidenceToPrograms.com: a toolkit to support

evidence-based programming for seniors
Alan B. Stevens 1 *, Shannon B. Coleman 1 , Richard McGhee 2 and Marcia G. Ory 3
1
Baylor Scott & White Health, Temple, TX, USA
2
Central Texas Area Agency on Aging, Belton, TX, USA
3
Program on Healthy Aging, School of Public Health, Texas A&M Health Science Center, College Station, TX, USA
*Correspondence: [email protected]
Edited by:
Matthew Lee Smith, The University of Georgia, USA
Reviewed by:
Heather Honoré Goltz, University of Houston-Downtown, USA

Keywords: toolkit, evidence-based programming, older adults, community-based organizations, senior citizens

Community-based organizations (CBOs) Rather, it is intended to guide a user determining which topics to address in the
are moving rapidly to provide new health through a series of steps and decisions Toolkit; (2) scanning existing materials to
services proven to be health promot- to facilitate an organization’s ability to identify resource gaps and materials to ref-
ing, and importantly, desired by seniors. provide programming that is desired and erence in the Toolkit; (3) developing the
In particular, evidence-based programs beneficial to the seniors being served. Toolkit content; (4) refining the Toolkit
(EBPs) that promote healthy behaviors and with adjustments suggested by a team of
proper self-management of health condi- WHO IS THE INTENDED TOOLKIT expert reviewers; and (5) creating a user-
tions have become a valuable resource to AUDIENCE? friendly website to feature the Toolkit. The
CBOs by complementing formal health- The Toolkit is designed for organizations diverse, expert advisory panel representing
care services. Yet, the level of support that are motivated to implement EBPs for community health and health promotion
needed for community adoption and seniors. Whether organizations are new researchers, state-level aging services, local
implementation of current EBPs is insuf- to evidence-based programming or have community-based service providers, and
ficient to maximize CBOs’ ability to been implementing EBPs for years, they national leaders in evidence-based pro-
enhance the health of the senior popula- will find useful materials in the Toolkit. gramming reviewed the Toolkit content
tion. Decision-making support is a critical For organizations with minimal experi- and provided feedback during develop-
factor in facilitating the use of EBPs such ence, the Toolkit will function as a guide ment.
as those listed on national clearinghouses that provides the basic information needed
(e.g., National Council on Aging, NCOA) to select, implement, and evaluate an EBP.
that address a local need or the need of WHAT IS THE FORMAT OF THE
For organizations with significant experi-
a specific senior population. The Com- TOOLKIT?
ence, the Toolkit will function as a primer
munity Research Center for Senior Health The format was designed based on feed-
that is useful in evaluating their approaches
(CRC-Senior Health) developed a web- back received during its development.
to program selection, implementation, and
based Toolkit (EvidenceToPrograms.com) The Toolkit is featured on an interac-
evaluation. Several ways are described in
to support CBOs in the selection, imple- tive, user-friendly website, EvidenceToPro-
which organizations that serve seniors can
mentation, and evaluation of EBPs1 . This grams.com. Users can explore paths for
optimize their use of the Toolkit. The
Commentary will describe the rationale learning how to select an EBP as well as
Toolkit is also a useful and user-friendly
behind the tool, its development and basic how to implement a selected program.
online resource for healthcare professionals
structure and content. Additionally, plans and students interested in learning about
for future development will be shared. evidence-based programming. The Toolkit WHAT CONTENT IS IN THE TOOLKIT?
is designed to be a resource for organiza- The Toolkit provides a comprehen-
WHAT IS THE PURPOSE OF THIS tions throughout the US who are interested sive overview of program selection,
TOOLKIT? in the promotion of health, regardless of implementation, and evaluation. The con-
The purpose of this Toolkit is to build the organization’s target clientele. tent of the Toolkit is divided into two
the capacity of CBOs to promote senior sections: (1) selecting a suitable EBP and
health and well-being through evidence- HOW WAS THE TOOLKIT DEVELOPED? (2) implementing EBPs with fidelity. This
based programming. The toolkit does not Toolkit development spanned 18 months includes information not commonly found
promote or advocate for any specific EBP. and consisted of five major tasks: (1) on EBPs.

1 TheCommunity Research Center for Senior Health is a multi-institutional, multi-disciplinary research center created to develop, implement, evaluate, and disseminate
evidence-based interventions that address multiple social and behavioral determinants of senior health (http://seniorhealth.sw.org).

Frontiers in Public Health | Public Health Education and Promotion April 2015 | Volume 3 | Article 18 | 360
Stevens et al. EvidenceToPrograms.com: support for evidence-based programming

Each section is further divided into WHAT ARE PLANS FOR FUTURE Conflict of Interest Statement: The authors declare
subsections offering questions, examples, DEVELOPMENT? that the research was conducted in the absence of any
commercial or financial relationships that could be
and resources to help organizations antici- EvidenceToPrograms.com was developed
construed as a potential conflict of interest.
pate and address barriers to implementing and is maintained as part of the CRC-
or maintaining a program. For example, Senior Health’s mission to engage indi- This paper is included in the Research Topic, “Evidence-
Section 1 has three subsections: (a) what viduals and their communities in pro- Based Programming for Older Adults.” This Research
does it mean for a program to be evidence- grams that improve senior health and Topic received partial funding from multiple government
based; (b) choosing which program to well-being. It is a critical element in a and private organizations/agencies; however, the views,
findings, and conclusions in these articles are those of the
implement; and (c) how to evaluate the greater movement to improve population
authors and do not necessarily represent the official posi-
impact of the program on clients served. health via the availability and accessibil- tion of these organizations/agencies. All papers published
Each subsection is further organized into ity of community-based health supports in the Research Topic received peer review from members
a series of more detailed questions to and programs. Evidence-based program- of the Frontiers in Public Health (Public Health Edu-
guide CBOs in the program selection and ming provides CBOs an opportunity to cation and Promotion section) panel of Review Editors.
Because this Research Topic represents work closely asso-
implementation process. take the lead in health promotion and
ciated with a nationwide evidence-based movement in
Several unique features are integrated to support seniors in self-management of the US, many of the authors and/or Review Editors may
into the Toolkit. Narrative text, lists, dia- chronic health conditions in settings that have worked together previously in some fashion. Review
grams, and tables serve as a guide for com- are associated with health, wellness, and Editors were purposively selected based on their expertise
munity organizations through the process leisure. with evaluation and/or evidence-based programming
for older adults. Review Editors were independent of
of selecting, implementing, and evaluat- The CRC-Senior Health is dedicated named authors on any given article published in this
ing EBPs. Throughout the Toolkit, links to further development of EvidenceToPro- volume.
have also been provided to useful materi- gram.com and is receptive to recommen-
als from other organizations. Additionally, dations from Toolkit users as well as policy Received: 25 November 2014; accepted: 16 January 2015;
the Toolkit features an interactive flow- makers in the evidence-based program- published online: 27 April 2015.
Citation: Stevens AB, Coleman SB, McGhee R and Ory
chart that helps organizations estimate and ming arena. Broader use and dissemination
MG (2015) EvidenceToPrograms.com: a toolkit to sup-
increase their readiness to implement EBPs, of the Toolkit will allow CBOs through- port evidence-based programming for seniors. Front.
regardless of their previous experience. out the US to improve the selection, Public Health 3:18. doi: 10.3389/fpubh.2015.00018
A section about sustaining the imple- implementation, and evaluation of EBPs, This article was submitted to Public Health Education
mented program is also included. It thus, enabling CBOs to effectively access and Promotion, a section of the journal Frontiers in
Public Health.
acknowledges a major element, the diffi- and implement programs that match their Copyright © 2015 Stevens, Coleman, McGhee and Ory.
culty of ensuring that the implemented clients’ needs. This is an open-access article distributed under the terms
program is maintained in the face of of the Creative Commons Attribution License (CC BY).
changes in funding, resource availability, ACKNOWLEDGMENTS The use, distribution or reproduction in other forums is
and audience characteristics. The Toolkit Initial funding for the Community permitted, provided the original author(s) or licensor are
credited and that the original publication in this journal
offers strategies that can help organiza- Research Center for Senior Health was is cited, in accordance with accepted academic practice.
tions increase the sustainability of their provided by the NIA/NIH (Award Number No use, distribution or reproduction is permitted which
implemented program. RC4AG038183-01) does not comply with these terms.

www.frontiersin.org April 2015 | Volume 3 | Article 18 | 361

 ADVANTAGES OF PUBLISHING IN FRONTIERS

FAST PUBLICATION COLLABORATIVE RESEARCH NETWORK

Average 90 days PEER-REVIEW Our network
from submission Designed to be rigorous – increases readership
to publication yet also collaborative, fair and for your article
constructive

OPEN ACCESS TRANSPARENT GLOBAL SPREAD

Articles are free to read, Editors and reviewers Six million monthly
for greatest visibility acknowledged by name page views worldwide
on published articles

COPYRIGHT TO AUTHORS IMPACT METRICS SUPPORT

No limit to Advanced metrics By our Swiss-based
article distribution track your editorial team
and re-use article’s impact

EPFL Innovation Park · Building I · 1015 Lausanne · Switzerland

T +41 21 510 17 00 · [email protected] · frontiersin.org