University of Southern California Dissertations and Theses

Translating research into practice: a community-based medication management intervention

(USC Thesis Other)

Translating research into practice: a community-based medication management intervention

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

TRANSLATING RESEARCH INTO PRACTICE:
A COMMUNITY-BASED MEDICATION MANAGEMENT INTERVENTION

by

Gretchen Elizabeth Alkema
________________________________________________________

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(GERONTOLOGY)

May 2007

Copyright 2007 Gretchen Elizabeth Alkema

ii
DEDICATION

To Jody
The love of my life and still the most amazing woman I have ever met

To Paul
True ambition is not what we thought it was…

To Rosie
Whatsoever things are true…

iii
ACKNOWLEDGEMENTS
Financial support for this dissertation was provided by four main
sources. First I am grateful to have received a John A. Hartford Foundation
Doctoral Fellowship in Geriatric Social Work that generously sponsored my
dissertation work. Additional funding was provided in part by the AARP
Scholars Program and the Administration on Aging’s Evidence-Based
Prevention Initiative by funding the Community-Based Medication
Management Intervention project. I also received substantial financial and
material support from the Davis School of Gerontology through the Al and
Madelyne Hanson Scholarship, the Sharon Wallace O'Loughlin Memorial
Fund, the Ed and Rita Polusky Scholarship, and the Ron Stever
Scholarship.
I am deeply indebted to Dr. Kathleen Wilber who served as my
dissertation chair and mentor for five years in the doctoral program. Kate’s
willingness to take a chance on guiding this direct service social worker
through the world of research has been the greatest turning point in my
professional career. Her encouragement, patience, direction, leadership,
and generosity of spirit are profound gifts that I hope to embody someday
and pass on to others. I thank you Kate from the bottom of my heart.
The two other members of my dissertation committee were Dr. Eileen
Crimmins and Dr. Robert Myrtle. I asked them to participate on my
committee because I have the utmost respect for their unique skills and

iv
expertise that directly related to core issues discussed in my dissertation.
They have challenged me to expand my vision of the dissertation and my
post-PhD future. I am extremely grateful for their wisdom, candor, and
guidance throughout my academic development.
I would like to thank the faculty of the Davis School of Gerontology who
contributed to my professional growth in myriad ways that allowed me to
pursue this dissertation. Most especially, I am grateful to Dr. Phoebe Liebig
whose endless support and “unofficial” mentorship supported me in
challenging times more than she will ever know.
So many people contributed to the development of the CBM Intervention
as part of the AoA Evidence-Based Prevention Initiative: Donald Grantt at
the AoA; Dr. Nancy Whitelaw, Lynn Beattie, Serena Sanker, and Binod
Suwal at the National Council on Aging, and at the Partners in Care
Foundation, the facilitation team that included June Simmons, Dr. Dennee
Frey, Dr. Elaine Der, Mira Trufasiu, Jennifer Wieckowski, Dr. Susan
Enguidanos, and Paula Jamison. However, this project would not have
been possible without the vision and leadership of Assistant Secretary for
Aging, Josefina Carbonell, who dedicated fiscal and human resources to
the idea of implementing evidence-based practices in the “aging network.” I
would also like to thank the MSSP administrative staff that believed in the
need for medication management and supported its translation. Most of all,
I wish to thank the many MSSP nurse and social work care managers that

v
work tirelessly on behalf of low-income, chronically ill older adults who
desire to live as fully as possible in their communities of choice.
A source of constant support has been the friendships made with
student and post-doctoral colleagues at the Davis School of Gerontology. I
give special thanks to my dear friend, research consultant, and carpool
buddy Dr. Rob Kennison who indulged my every question about research
methodology known to man. I am also thankful to Dr. Judy Yip who
reminded me stay focused on completion at all costs. However, it was the
other two members of the “Witches of Eastwick” whose laughter, service,
and camaraderie will stay with me forever: Dr. Dawn Elizabeth Alley and Dr.
J. Alexis Abramson.
None of this would have been possible without the love and support of
my family. This dissertation is dedicated to three fabulous people in my life:
my loving partner Jody whose devotion, laughter, and adoration sustains
me in ways I cannot describe; my truly amazing brother Paul whose
brilliance, courage, and resilience has been a guiding light for me as long
as I can remember; and my beautiful sister Rosie who has an incredible
future ahead of her that I cannot wait to witness. I am so thankful for my
parents, Cody and John, who love me unconditionally and always told me I
could be and do anything I desired. I am also thankful for their respective
partners, Bill and Linda, who love and care for my parents and have shown
that same gracious support of me in my endeavors. I am grateful for my

vi
grandparents Ruth and Prescott Chaplin and Janet and Richard Alkema,
my in-laws Joan and Sam Forter, and the entire Alkema / Chaplin / Forter /
Reiss clan who remind me of the importance of family bonds and the
heterogeneity of the aging experience.
So many friends have provided a listening ear and loving kindness
during the ups and downs of the doctoral experience. I would not have
entered this program if it were not for the loving encouragement of Diane
and Ron Guest, so a big thanks to you both. I am so thankful for my long-
time friends and unofficial PR agents Karin and John Gallagher, who have
walked with me on so many roads and remind me that it is always baseball
season. I offer my loving gratitude to Dawn and Bill Linko who constantly
reminded me that life really would look different after an evening in the
Jacuzzi or a meeting with like-minded friends. Thank you so much to Don
Power and Roscoe James for the many hours of tea, sympathy, and an
occasional kick in the butt at their kitchen table. An instrumental friend in
the final months was Michael Replogle, whose laughter and caring provided
me with a daily reprieve from my greatest obsession.
Finally, I am eternally grateful to the Great Spirit of the Universe that
guides my will and my life on a daily basis, and who introduced me to two
friends I have never met but who paved the way for this journey to be
possible: Bill Wilson and Dr. Bob Smith.

vii
TABLE OF CONTENTS

Dedication

ii
Acknowledgements

iii
List of Tables

ix
List of Figures

xi
Abstract

xii
Preface

xiv
Chapter I: Introduction

1
Chapter II: Translational Research Study: A Community-
Based Medication Management Intervention

26
Chapter III: Prevalence of Potential Medication Problems
for Community-Dwelling, Dually Eligible Older Adults

48
Chapter IV: Improving Medication Use Among Community-
Dwelling, Dually Eligible Older Adults

98
Chapter V: Staff Perceptions of Implementing an Evidence-
Based Medication Management Intervention in Care
Management

129
Chapter VI: Discussion

160
Bibliography

181
Appendices 203
Appendix A: Procedure for Medication Problem
Screening
Appendix B: Medications Errors Data Collection
Form
Appendix C: Confirmation Procedures of Medication
Problems by Category
Appendix D: Terminated Participants Prior to
Medication Problem Confirmation and
Sample
204

205

206

207

viii
Appendix E: Reason for Termination Prior to
Medication Problem Confirmation
Appendix F: Terminated Participants Prior to
Intervention and Intervention Sample
Appendix G: Reason for Termination Prior to
Intervention
Appendix H: Medication Follow-Up Form
Appendix I: Pharmacist Contacts and Medication
Change
Appendix J: MSSP Staff Questionnaire
Appendix K: Responses to MSSP Staff
Questionnaire
Appendix L: Qualitative Responses to MSSP Staff
Questionnaire
Appendix M: Three Medication Management
Protocols Created by Site #3
Appendix N: Disclaimer
208

209

210

211
212

213
216

218

221

224

ix
LIST OF TABLES

Table 2.1: Differences between Home Healthcare and
MSSP Programs

26
Table 2.2: Context and Facilitation Differences across
MSSP Sites

44
Table 3.1: Characteristics of the CBM Intervention Sample

65
Table 3.2: Characteristics of the CBM Intervention Sample
by Site

67
Table 3.3: Prevalence of Potential and Confirmed
Medication Problems for Total Sample and Sites

71
Table 3.4: Differences between Participants with Potential
and Confirmed Medication Problems

72
Table 3.5: Differences between Sample and Participants
with Potential Medication Problems

73
Table 3.6: Differences between Sample and Participants
with Confirmed Medication Problems

74
Table 3.7: Pearson Product-Moment Correlation Results of
Potential Medication Problems

77
Table 3.8: Odds Ratios of Potential Medication Problems

80
Table 3.9: Pearson Product-Moment Correlation Results of
Confirmed Medication Problems

82
Table 3.10: Odds Ratios of Confirmed Medication Problems

85
Table 4.1: Characteristics of the Intervention Sample

113
Table 4.2: Characteristics of the Intervention Sample by
Site

114
Table 4.3: Medication Problems and Change Rates at 3-
Month Follow Up

115

x
Table 4.4: Pharmacist Contacts and Medication Change

118
Table 4.5: Medication Problems and Change for Those
Who Lived Alone with Risk

120
Table 4.6: Pearson Product-Moment Correlation Results for
Medication Change

121
Table 5.1: Measures of Care Manager Characteristics

140
Table 5.2: Characteristics of Care Manager Sample

144
Table 5.3: Spearman’s Rho Results of MSSP Staff
Questionnaire Responses

147
Table 5.4: Mann-Whitney U Statistics of MSSP
Questionnaire Responses

149

xi
LIST OF FIGURES

Figure 1.1: Translational Research Process

3
Figure 1.2: AHRQ Cycle of Research

15
Figure 1.3: Theoretical Model of PARIHS Framework on
Implementation Success

21
Figure 2.1: Translational Design for CBM Intervention

38
Figure 2.2: Implementation of CBM Intervention

39
Figure 3.1: Prevalence of Potential and Confirmed
Medication Problems

70
Figure 3.2: Potential Medication Problems by Number of
Medications

78
Figure 3.3: Confirmed Medication Problems by Number of
Medications

83
Figure 4.1: Flowchart of CBM Intervention Sample

111
Figure 4.2: Number of Medication Problems by Percentage
of Medication Change

116
Figure 6.1: Revised Theoretical Model of PARIHS
Framework on Implementation Success

173
Figure 6.2: Translational Research Process of CBM
Intervention

174

xii
ABSTRACT
Translational research in the health sciences involves three key phases
linking basic research findings to application through interventions
promoting the health and welfare of aging individuals. Type II translation
disseminates efficacious guidelines and interventions proven through
randomized clinical trials into applied settings and evaluates their continued
effectiveness in the new environment. This dissertation evaluated a Type II
translation project, the Community-Based Medication Management
Intervention (CBM Intervention), funded through the Administration on
Aging’s Evidence-Based Prevention Programs Initiative. The CBM
Intervention was a collaborative effort to implement a medication
management screening and intervention protocol, originating from a home
healthcare randomized clinical trial, in a Medi-Cal waiver care management
program serving dually-eligible, functionally impaired older adults (N=615).
This dissertation analyzed the sample’s prevalence of medication problems,
effectiveness of the intervention to resolve identified problems, and staff
perceptions of implementing an evidence-based practice into care
management. The guiding conceptual framework was the Promoting Action
on Research Implementation in Health Services (PARIHS) framework.
A four problem criteria was used to screen for medication problems;
nearly half of the sample (N=299) had a potential problem. The highest
problem prevalence was for those with inappropriate therapeutic

xiii
duplication. A focused pharmacy review confirmed medication problems for
29% of the sample. Predictors of an identified problem included advancing
age, Caucasian, newly enrolled in care management, health service
utilization, and number of medications. Participants with a confirmed
problem engaged in a pharmacist-driven intervention (N=162) and 61% had
the problem resolved.
A questionnaire developed for this dissertation assessed care manager
perspectives on implementing an evidence-based practice into care
management. Nurse and social work care managers (N=29) had differing
perspectives on their scope of practice concerning medication
management, and those working at the same site as the project’s facilitation
team expressed more positive responses to the implementation process
and benefits of medication management.
This dissertation demonstrated the complexities of Type II translation
and highlighted lessons learned for the CBM Intervention and translational
research as a whole. Key issues included challenges between
implementation fidelity and adaptation, the impact of professional
differences, and importance of facilitation activities and leadership on
implementation success.

xiv
PREFACE
The overall purpose of this dissertation was to investigate the process
and outcomes of translating research findings into applied health service
settings. Translational research in the health sciences involves three key
phases that link basic research findings to their application through
interventions that promote the health and welfare of aging individuals. Type
II translation takes efficacious guidelines and interventions proven through
randomized clinical trials, disseminates these practices into applied
settings, and evaluates their continued effectiveness in the new “real world”
environment. This dissertation evaluated a Type II translation project, the
Community-Based Medication Management Intervention (CBM
Intervention), funded through the Administration on Aging’s Evidence-
Based Prevention Programs Initiative. The CBM Intervention was a
collaborative effort to implement a medication management screening and
intervention protocol, originating from a home healthcare randomized
clinical trial, in a Medi-Cal (California’s Medicaid) waiver care management
program serving dually-eligible, functionally impaired older adults (N=615).
This dissertation analyzed the prevalence of medication problems for the
sample, effectiveness of the intervention to resolve identified problems in
the new setting, and staff perceptions of implementing an evidence-based
practice into an existing care management program.

xv
Lessons learned from this work expand the current knowledge base on
translational research, advance successful implementation of evidence-
based practices in community settings, and inform future research on next
steps to improve the lives of functionally impaired older adults.

1
I: INTRODUCTION
A. Translating Research into Applied Settings
The practice of translating findings from research into usable tools for
daily life is an ancient enterprise and occurs in every area of scientific
achievement. In health sciences research, the dynamic partnership
between research and practice has long been established, from
Hippocrates’ rejection of “evil spirits” that caused disease based on studies
of human organ systems to Marie Curie’s discovery of radium that opened
pathways for cancer treatment (Porter & Ogilvie, 2000). The deliberate
movement of knowledge gained from basic research to application of that
knowledge in human lives through programs, services, and treatments is
the core of translational research in the health sciences. Sussman and
colleagues (2006) defined translation in the health professions as “an
extended process of how research knowledge that is directly or indirectly
relevant to health behavior eventually serves the public” (p. 9).
Although not a linear process (Ginexi & Hilton, 2006), translation is
generally categorized into three interrelated typologies along a continuum
(Figure 1.1). Type I translation refers to the movement of knowledge
gained from basic science into the development of clinical treatment
interventions (Sung et al., 2003; Sussman, Valente, Rohrbach, Skara, &
Ann Pentz, 2006). Type II translation then takes what is known to work
from rigorously controlled clinical trials, often called efficacy trials (Glasgow,

2
Vogt, & Boles, 1999), and applies these tested interventions or practices
into “real world” settings (Sung et al., 2003; Sussman et al., 2006). At this
stage, the focus is on dissemination and adoption of practices and decision-
making strategies emanating from empirical evidence into health and social
service systems (Glasgow, Lichtenstein, & Marcus, 2003; Rohrbach, Grana,
Sussman, & Valente, 2006; Sung et al., 2003). The third and lesser-know
type of translational research is called implementation research, which
evaluates the outcomes of widespread dissemination after a Type II
translation (Rubenstein & Pugh, 2006). The goal of implementation
research is to seek out individual, provider, and organizational level
characteristics that facilitate and impede the translation process in order to
create quality improvement pathways for successful implementation in a
given system (Eccles & Mittman, 2006). Type I translation processes have
been highly institutionalized, leading Hiss and colleagues (2004) to remark
that it has “many mothers and fathers” in the scientific community. On the
contrary, Type II translation has been described as “an orphan” given its
tendency to be forgotten or marginalized (Hiss, Green, & Ottoson, 2004).

3

B. Issues and Models of Translational Research
Researchers from a variety of human service fields have examined
the translation of innovative, evidence-based practices into applied settings
including gerontology (Pillemer, Czaja, Schulz, & Stahl, 2003),
communications (Rogers, 1995), health care professions of medicine
(Bradley, Schlesinger, Webster, Baker, & Inouye, 2004), nursing (Titler,
2004), mental health (Aarons, 2004), and substance abuse (Sussman et al.,
2006), health care management (Walshe & Rundall, 2001), psychology
(Gonzales, Ringeisen, & Chambers, 2002), public health (Glasgow et al.,
2003; National Center for HIV Prevention, 2003), and social work (Gambrill,
2006; Gira, Kessler, & Poertner, 2004). Translation of evidence-based

4
practices to address issues pertinent to older adults has occurred to some
degree across the span of aging service delivery systems from health care
institutions to home- and community-based service providers.
Deliberate translation of practices derived from empirical evidence into
community-based aging services has many benefits including the potential
to improve treatment effectiveness and reduce provider variability in
applying treatment regimens (Feldman & Kane, 2003). In addition,
community services agencies, which often have less external oversight than
health care models, can demonstrate improved quality of care by
implementing tested interventions rather than building their own models,
which may leave out important components (Institute of Medicine, 2001).
The use of empirically tested interventions can promote cost effectiveness
by focusing on efficacious interventions known to work with targeted
problems and populations (Weissert, Chernew, & Hirth, 2003). However,
there are multiple challenges to translating evidence based practices into
community settings, including professional resistance to evidence-based
practice; lack of organizational buy-in; lack of specific goals and standards
in translating the evidence; and rigidity of evidence-based practice that
cannot be molded to meet specific needs of the applied setting (Glasgow et
al., 2003; Grimshaw et al., 2001; Grol, 2001).
A critical feature of translating research into practice is the need to
resolve the tension between intervention fidelity and reasonable adaptation

5
to the setting where implementation occurs (Backer, 2001). When
implementing evidence-based practices into “real-world” programs, many
researchers believe that treatment protocols created in a research
environment must be strictly adhered to in order for the new iteration to be
accurately described as a “true” adopter (Bellg et al., 2004; Elliott & Mihalic,
2004; Titler, 2004). Some argue that the lack of evidence-based practice in
community service programs ultimately devalues social services in the eyes
of other more medically focused professions that socialize their practitioners
toward the benefit of standardized practice regimens (Murphy & McDonald,
2004). Others note that intervention fidelity purists focus so much on the
technicalities of implementing the evidence-based practice that key
characteristics of the adopting agency and target population are
disregarded, often leading to implementation failure (Castro, Barrera, &
Martinez, 2004; Dusenbury & Hansen, 2004; Leventhal & Friedman, 2004).
Given these inherent tensions, successful translation efforts must move
beyond an “all or nothing” approach and identify key factors critical to the
translation process as well as those factors that may be adapted to sustain
the intervention in the new settings.
Several conceptual frameworks have developed to structure scholarship
and measure progress, and provide tangible guidance for successful
translation. The following discussion reviews these conceptual frameworks,
highlighting core components and reporting current utilization.

6
Organizational Change and Implementation Theories
Fundamental to any effort that translates efficacious practices into
applied settings is the concept of organizational change. Organizational
change is defined as the “the establishment of a new program that
necessitates investment of resources and internal restructuring of the
division of labor” (Hasenfeld, 1983, p. 219). Efforts to facilitate change
often disrupt an organization’s culture, but when successful they can foster
the development of a new organizational identity supporting innovation
(Louis, 1983). Hasenfeld (1983) cited three factors that mediate a
successful implementation of an innovation: the quality and specific tasks of
the technology to be implemented; the relationships between intra- and
inter-organizational components; and internal power and incentives to
facilitate the implementation process (p. 240). Research on organizational
change within health care services suggests that improvement in consumer
outcomes is associated with the use of multidisciplinary teams, integrated
care structures, and computerized decision-making support (Reuben, 2002;
Wensing, Wollersheim, & Grol, 2006).
For organizations to realize change, a new operational plan, service, or
technology must be put into action. This process of implementation is
defined as the “carrying out of a basic policy decision…Ideally, that decision
identifies problems to be addressed, stipulates the objectives to be
pursued, and…structures the implementation process” (Mazmanian &

7
Sabatier, 1989, p. 20). Several conceptual frameworks guide the analysis
of implementation processes. First Mazmanian and Sabatier (1981, 1989)
suggest that implementation of any policy or program should consider three
core elements. These include: 1) the tractability of the problem for which
the program is intended to address, 2) the ability of the policy or program
components to clearly structure the implementation process, including a
clear definition of the program’s “causal theory” (Pressman & Wildavsky,
1984), and 3) variables beyond the scope of the policy or program that will
affect its implementation, such the attitudes and resources of various
stakeholders, support from regulating authorities, and the leadership skills
of implementers.
A second framework for understanding program implementation is
defined by where the intervention originates. Innovations that are derived
as “top-down” solutions try to solve problems by implementing interventions
developed from key decision makers in a particular field (Sabatier, 1986).
Top-down approaches are often driven by explicit theory, research
evidence, and/or political power. Therefore, success of a top-down
approach through a Type II translation is dependent on an intervention’s
implementation fidelity to the original evidence base that is often developed
in a randomized clinical trial. On the contrary, bottom-up approaches
address problems by using a network of local stakeholders and existing
organizations to generate and implement solutions (Hjern, 1982). Bottom-

8
up processes may or may not be developed from a theory or research
evidence, and instead focus on solutions that are adapted to the needs of
the local environment.
A third type of implementation framework looks at the role and impact of
decision makers on the success of the new innovation. Pressman and
Wildavsky (1984) emphasized that the number of decision points required
to fully implement an intervention are inversely related to implementation
success. They described the level of leadership and commitment needed
by decision makers for successful implementation with this statement:
Every program is likely to affect someone, somewhere,
sometime. Adaptation to the environment must have been
achieved; otherwise, by definition, programs would not exist.
No genius is required to make programs operative if we don’t
care how long they take, how much money they require, how
often the objectives are altered or the means for obtaining
them are changed (p. 116).

One type of decision maker that impacts implementation is what Lipsky
(1980) referred to as a “street-level bureaucrat.” These are direct service
employees in a particular program that have autonomy over their daily work
activities and hence can significantly affect the implementation process of a
new intervention. Street-level bureaucrats have the ability to both simplify
and/or distort the implementation process through a process called
discretionary experimentation (Nakamura & Smallwood, 1980), which could
ultimately affect programmatic outcomes (Lipsky, 1980).

9
Diffusion of Innovations
In the opening of his landmark book Diffusion of Innovations, Rogers
(1995) noted that a barrier and myth of translational research is the idea
that innovative ideas, practices, and technologies do not need an active
dissemination plan because they will “sell themselves” by virtue of their
perceptible capacity to improve the functioning of individuals, organizations,
and systems. He described the example the British Navy taking nearly 150
years to adopt the practice of drinking citrus juice on expeditions after a
British sea captain discovered its benefits to combat scurvy. Although
differences in the speed and quality of communication mechanisms likely
affected dissemination of the “scurvy cure”, the multiple scientific fields
listed above have all reported significant challenges with dissemination
even in this age of advanced communication technology.
Rogers (1995) delineated five characteristics of innovative practices or
technologies that affect the speed and process of diffusion.
• Relative advantage defines how well the innovation is perceived as
an improvement on existing practices or technologies;
• Compatibility relates to how potential adopters view the innovation as
complimenting to their experience, values, and future needs;
• Complexity accounts for the level of difficulty in implementing
innovation;

10
• Trialability is the feasibility to test out components of the innovation
on a “trial” basis before its full adoption; and
• Observability addresses the human interaction element of diffusion
where others are able to visibly perceive the benefits of the
innovation.
Rogers (1995) also highlighted the importance of the timing of diffusion,
change agents, communication networks, innovation decision processes,
and the characteristics of those who adopt innovations. His work built on
organizational theories to describe a five-stage adoption pathway, from
setting an agenda for change to routinizing change processes where the
new innovation becomes standard practice.
Rogers’ conceptual framework itself has been an extraordinary example
of diffusion processes at work. Since the original publication in 1962, his
framework has influenced the development of countless translation
activities in human services. A MEDLINE literature search of articles
published from 1966 to 2006 using the keyword “diffusion of innovations”
resulted in over 2,800 publications, with the first use of this keyword in
1976. Over forty years later, the concept of purposeful distribution of
efficacious innovations to a broader environmental context is highly
relevant. It has shaped the following models that are specific to translation
in health and social care systems.

11
RE-AIM Framework
Building from Rogers’ work (1995), the RE-AIM framework developed by
Glasgow and colleagues (1999) is a model for evaluating the impact of an
evidence-based practice within the public health system. RE-AIM is an
acronym that stands for Reach, Efficacy/Effectiveness, Adoption,
Implementation, and Maintenance (Glasgow, 2006; Glasgow et al., 2003;
Glasgow et al., 1999). Each dimension, described below, is rated on a
scale from 0 to 100%.
• Reach – the percentage and characteristics of those who receive or
are affected by the intervention to determine their representativeness
within the larger target population;
• Efficacy/Effectiveness – impact of the intervention on primary health
outcomes, quality of life, satisfaction, and potential negative
consequences;
• Adoption – proportion and representativeness of settings that adopt
the intervention;
• Implementation – the extent to which a program is delivered as it is
intended; and
• Maintenance – the sustainability of change at the individual and
organizational level.
RE-AIM serves as a practical guide to the most critical areas when
disseminating a new intervention into existing service systems. It

12
addresses the extent to which a specific population has access to the
intervention (Reach), how effective the intervention is in creating health
behavior change in this new setting (Efficacy/Effectiveness), how many
different types of organizations are implementing the new intervention
(Adoption), intervention fidelity to the original design (Implementation), and
the sustainability of the intervention (Maintenance).
The current reporting of randomized clinical trials in the medical
literature generally follows established criteria in an effort to improve the
quality of published research. The most prominent mechanism to
accomplish this objective is through the CONSORT statement or the
Consolidated Reporting of Standard Trials (Moher, Schulz, & Altman,
2001). The CONSORT statement includes twenty-two standards focused
mainly on the maintenance of internal validity, namely the intervention’s
success in facilitating the desired change. In translating efficacious
interventions into the public health sphere, application of the RE-AIM
Framework supports the evaluation of the intervention’s external validity, or
how generalizable the intervention is beyond the context of an original
randomized clinical trial (Glasgow et al., 2003). Beyond the original 22
CONSORT criteria, Glasgow and colleagues (2003) offered seven
additional criteria for inclusion that address external validity utilizing the
principles of the RE-AIM Framework.

13
The goal of RE-AIM is to help health services researchers and program
administrators evaluate interventions targeted for a large public health
audience. It has been utilized with several health promotion interventions
including smoking cessation (Glasgow, Whitlock, Eakin, & Lichenstein,
2000), diabetes management (Gary, Hill-Briggs, Batts-Turner, & Brancati,
2005; Glasgow et al., 2004), and physical activity (Prohaska et al., 2006;
Toobert, Strycker, Glasgow, Barrera, & Bagdade, 2002). RE-AIM is also
the guiding framework for several large-scale health promotion initiatives
such as the WISEWOMAN projects (Will, Farris, Sanders, Stockmyer, &
Finkelstein, 2004) and the Administration on Aging’s Evidence-Based
Prevention Programs Initiative (Administration on Aging, 2004; Bryant,
Altpeter, & Whitelaw, 2006).
Governmental Frameworks
Government agencies actively involved in health sciences translation
include Agency for Healthcare Research and Quality (AHRQ), the National
Institute of Health (NIH), and the Veterans Health Administration (VA). This
involvement includes not only financial support through new and existing
grant mechanisms, but also through internal research units that are
mapping out the translation components and their interrelationships. The
following section describes the various translational activities and models of
these federal entities.

14
AHRQ
Formerly known as the Agency for Health Care Policy and Research,
AHRQ is actively involved in both the “production and use of evidence to
improve policy and practice” and revised its mission in 2004 to concentrate
more directly on translational research (Clancy, 2004). Established in 1989
under the Department of Health and Human Services, AHRQ has
developed grant mechanisms to train researchers in translation and initiated
the TRIP (Translating Research into Practice) projects starting in 1999
(Clancy, Slutsky, & Patton, 2004; Farquhar, Stryer, & Slutsky, 2002).
TRIP projects based their work on the conceptual framework called the
AHRQ Cycle of Research (Figure 1.2) (Agency for Healthcare Research
and Quality, 1998). This framework addressed the iterative research
process starting the first phase with a needs assessment to determine key
issues and problems that research should address. The next phase was
knowledge development through basic and clinical research mechanisms.
Knowledge then was to be translated and disseminated to a larger
audience to maximize benefits within new evidence-based practices and
strategies. The final phase is evaluation, which considers the effectiveness
and sustainability of the dissemination process. Although a small agency,
AHRQ’s pioneering effort to systematize the phases of translational
research has paved the way for larger entities within the federal scientific
domain (e.g., NIH) to expand this research agenda.

15
Figure 1.2: AHRQ Cycle of Research

Source: Agency for Healthcare Research and Quality. (1998). AHCPR Strategic Plan.
Retrieved May 4, 2005, from http://www.ahrq.gov/about/stratpln.htm

The NIH Roadmap
Concerned about the lack of dissemination of scientific advances into
clinical outcomes, NIH leadership in 2002 realized that translating
biomedical research findings into clinical interventions to ultimately improve
health care outcomes was too large a task for any one of its agencies
(Zerhouni, 2005a). This led to the development of the “NIH Roadmap for
Medical Research” (National Institute of Health, 2003; Zerhouni, 2003).
This three-fold pathway toward reinventing the mission and purpose of the
NIH has focused on new strategies to better understand biological systems
and regulation, remove institutional barriers to facilitate more
interdisciplinary research, and most important for this discussion, stimulate

16
the clinical and translational research enterprises (Zerhouni, 2005b).
Through this work, NIH is taking a broad perspective on translation,
supporting both inductive and deductive research to build what it calls a
“robust knowledge base.”
Unlike the conceptual frameworks described above that are useful in
individual scholarship, the NIH has set a clear research agenda backed by
fiscal authority to support clinical and translational research. This effort will
likely have far reaching consequences that will fundamentally reshape
future investigations on health- and health services-related phenomena
given that bursts in innovation are often linked to federal funding
(Hasenfeld, 1983). A recent example of this shift is demonstrated in a NIH
program announcement that requested proposals for research on the
“when, how, and the why” of dissemination and implementation processes
(National Institute of Health, 2006). This announcement also clarified NIH’s
position on the debates over implementation fidelity versus adaptation by
stating, “…interventions developed in the context of efficacy and
effectiveness trials are rarely transferable without adaptations to specific
settings.” Here, NIH shifted away from the exclusive use of top-down
knowledge development processes and has encouraged learning
contributions from local settings and community stakeholders.

17
VA QUERI
The third entity that is known for moving research findings into clinical
interventions is the Veterans Heath Administration through its program
known as the Quality Enhancement Research Initiative (VA-QUERI, 2004).
The main focus of VA-QUERI is the third phase of translational research,
which seeks discoveries in implementation to promote greater and more
successful dissemination of evidence-based practices. Lessons learned
thus far include the value of formative evaluation techniques to better
understand practice settings prior to implementation (Stetler et al., 2006)
and strategies to address unforeseen barriers when they occur (Hagedorn
et al., 2006).
In an effort to disseminate its own findings, VA-QUERI researchers
developed a user-friendly guide to implementation research that describes
theories used, lessons learned, and resources available to those involved in
implementation research (VA-QUERI, 2004). One of the models used in
the effort is the Promoting Action on Research Implementation in Health
Services framework described below.
Promoting Action on Research Implementation in Health Services
(PARIHS)
The previous frameworks consider translational research from the
perspective of broad-scale organizational or system change. Taking a
different approach, the Promoting Action on Research Implementation in

18
Health Services (PARIHS) framework, offers a structured mechanism to
understand the operational complexities of translating research within
health services settings (Kitson, Harvey, & McCormack, 1998; Rycroft-
Malone, Harvey et al., 2004; Rycroft-Malone et al., 2002). The PARIHS
framework describes three critical and interrelated factors associated with a
successful translation of research to practice: 1) evidence (Rycroft-Malone,
Seers et al., 2004), 2) context (McCormack et al., 2002), and 3) facilitation
(Harvey et al., 2002).
Evidence refers to knowledge, practices, and technologies derived from
a variety of sources that will be embedded into a particular setting (Rycroft-
Malone, Seers et al., 2004). This evidence includes scientifically robust
data, protocols, or other interventions that have been highly scrutinized
through a research design, such as positive findings from a randomized
controlled trial on intervention’s efficacy to affect change. Rycroft-Malone
and colleagues (2004) also include three other areas where evidence is
gained: clinical experience of professionals, knowledge and preferences
from clients and their caregivers, and information derived from the local
context (e.g., chart audits and performance measures). The subject of what
qualifies as “evidence” is fundamental to translation, leaving many in the
health services to define it solely as the evidence derived from clinical
research (Atkins, Best, & Briss, 2004; Improved Clinical Effectiveness
through Behavioural Research Group (ICEBeRG), 2006; Rosen, Manor,

19
Engelhard, & Zucker, 2006; Timmermans & Mauck, 2005). However
growing dissent about the quality and applicability of research-based
evidence in applied settings has led health and social care professions to
utilize this broader definition (Gerrish & Clayton, 2004; Gibbs & Gambrill,
2002; Gonzales et al., 2002; Leach, 2006; Sackett, Rosenberg, Gray,
Haynes, & Richardson, 1996; Tang, Ehsani, & McQueen, 2003).
Context includes environment or organizational setting in which the
evidence will be adopted (McCormack et al., 2002). McCormack and
colleagues (2002) delineate its four elements: 1) the receptive context that
includes physical, social, financial, professional, and systemic
characteristics of the setting; 2) culture that are the values and roles
embedded in the receptive context; 3) leadership that accounts for power
and authority structures, decision-making processes, and role clarity; and 4)
evaluation, which is the feedback mechanisms to clients, staff, and systems
to improve performance. Similar to other models used by translational
researchers, the PARIHS framework acknowledges that contextual factors
have a profound impact on implementation and sustainability efforts.
Facilitation addresses the management and administrative attributes of
those in charge of directly supporting the implementation of the “evidence”
into the “context” (Harvey et al., 2002). Facilitation focuses on the
development of effective organizational processes, teamwork and
collaboration activities in order to achieve specific implementation goals.

20
Harvey and colleagues (2002) described several facilitation tasks such as
providing staff training, technical assistance, project management, and
marketing expertise, offering regular supportive contact, and sustaining
partnerships necessary for successful implementation. Although it is often
the “glue” that holds the implementation process together, facilitation as a
clear and distinct concept is rarely discussed in the translational literature.
A useful way of understanding translation via the PARIHS concepts is
through an analogy about planting a hot house tomato vine in natural soil.
Hot house tomatoes (the “evidence”) are grown in laboratories, fed through
hydroponics, and provided optimum light and fertilizing nutrients. Then a
horticulturist wishes to plant the tomato vine in a Kansas field (the
“context”). The vine may be the best quality available and the Kansas soil
may be the most nutrient-rich, holistic environment possible. However the
life of the vine depends on a Kansas farmer who is willing to dig a proper
hole for planting, water the vine regularly, and manage circumstances
beyond the farmer’s control will damage the tomatoes such as weather and
pestilence (the “facilitation”). Therefore, the activation of these three
concepts is critical for successful translation: 1) the qualities and
characteristics of the evidence being implemented; 2) the organizational
and social system that will be accepting the new evidence; and 3) the
various communication processes by which the evidence is sown and takes
root in the new setting.

21
Supporting this analogy, developers of the PARIHS framework
hypothesized that successful implementation occurs when (Figure 1.3): 1)
the evidence is scientifically robust, matching professional consensus and
client needs (“high evidence”); 2) the context is receptive to change with
strong leadership and appropriate monitoring and feedback systems (“high
context”); and 3) there is appropriate facilitation of change with input from
skilled internal and external facilitators (“high facilitation”) (Kitson et al.,
1998; Rycroft-Malone et al., 2002). Although the PARIHS framework does
not include all potential variables that may impact implementation, it
provides an orienting perspective of the implementation process itself and
can therefore shed light on intricacies and complexities of translating
research into practice. It is this reason that the PARIHS framework will be
employed as the guiding conceptual framework throughout this dissertation.

22
C. Dissertation Research
Health services professionals have long recognized the importance of
integrating research and practice to improve quality of care and health-
related quality of life for older people. Over the last several decades,
researchers and practitioners in the health and social care arenas have
joined forces to develop translational research studies that seek to
implement rigorously tested, evidence-based practices in applied settings
and evaluate the continuing effectiveness of these interventions. One such
translational study was the Community-Based Medication Management
Intervention (CBM Intervention), a collaborative effort to implement a
medication management screening and intervention protocol in community-
based care management programs.
In September 2003, the CBM Intervention was launched as part of the
Administration on Aging’s Evidence-Based Prevention Programs Initiative
(Administration on Aging, 2004). This three-year initiative was developed to
support the implementation of evidence-based practice at the community
level through the aging service provider network. Each of the thirteen
funded projects implemented an intervention initially tested in a randomized
control trial environment and found to be effective in promoting disease
prevention with community-dwelling older adults, often with multiple and
complex needs. Building on this work, the purpose of the CBM Intervention
project was to translate the evidence-based medication management

23
intervention beyond its original efficacy trial in a home healthcare setting to
a Medi-Cal (California’s Medicaid) waiver care management program.
As evidence-based practices are translated into applied settings, it is
important to evaluate both the effectiveness of the intervention in the new
locations as well as the process of implementation (Rossi, Lipsey, &
Freeman, 2004; Weiss, 1997; Wholey, Hatry, & Newcomer, 2004). This
dissertation used the PARIHS framework to organize and clarify critical
factors throughout the implementation and evaluation of the CBM
Intervention. Individual and organizational level change were assessed
through three research activities using quantitative methods: 1) describing
the prevalence and predictors of medication problems in the Medi-Cal
waiver care management sample; 2) identifying participant outcomes of
implementing the medication management intervention to reduce identified
medication problems; and 3) ascertaining care manager perspectives on
implementing an evidence-based practice into a long-standing community-
based care management program. Qualitative methods were also used
throughout the research process to gather lessons learned for the CBM
Intervention project in order to reveal the essence of translational research
beyond the project’s participant-level outcomes. Qualitative feedback was
elicited as part of the care manager questionnaire process and through
participant-observer methods at implementation training sessions, staff

24
meetings, and advisory group meetings, and ongoing contact with project
management (Weiss, 1997).
D. Contribution to the Literature
This dissertation contributes to the scientific literature in two interrelated
ways. First, as the basis for this study, medication problems and its
subsequent management are key issues in medical safety, especially for
older adults with multiple functional impairments. Results from this study
define the volume of this problem in a vulnerable population and report the
efficacy of a targeted intervention to reduce medication errors.
The larger contribution to the literature is that this dissertation
represents a Type II translational research study based on its diffusion of an
evidence-based practice into applied settings and the evaluation of its
processes and outcomes. Results from this study reveal the benefits and
challenges of translation in community-based programs highlighting
complexities of this work for individuals, organizations, and broader health
care systems. The mandate for translating research into practice is evident
as federal agencies shift more and more funding toward this activity.
E. Organization of the Dissertation
This chapter reviewed of the current knowledge on translational
research and introduced the purpose of the dissertation research. Chapter
II describes the project that serves as the basis for this dissertation, the
CBM Intervention, through the lens of the PARIHS framework. Chapter III

25
reports on the first phase of the CBM Intervention, which is identifying the
prevalence of medication problems in functionally-impaired, dually-eligible
older adults living in the community who are participating in a Medi-Cal
waiver care management program. Chapter IV reports on the outcomes of
the medication management intervention to reduce problems identified in
the previous chapter. Chapter V describes care manager perceptions of
translating research into applied settings based on a questionnaire
developed for this project. Chapters III, IV, and V are stand-alone chapters
of the dissertation and therefore incorporate relevant elements of the CBM
Intervention description and processes from Chapter II. Chapter VI
provides the discussion of the dissertation as a whole and future directions
for translational research based on findings from this work.

26
II. TRANSLATIONAL RESEARCH STUDY: A COMMUNITY-BASED
MEDICATION MANAGEMENT INTERVENTION
The Community-Based Medication Management Intervention (CBM
Intervention) was an ambitious project to translate an evidence-based
medication management intervention into a Medi-Cal waiver care
management program and then evaluate its effectiveness in identifying and
resolving medication problems for community-dwelling, dually-eligible older
adults. This chapter describes the operational and translational
components of the CBM Intervention utilizing PARIHS framework elements
of evidence, context, and facilitation. Table 2.1 captures specific
differences between these settings on several dimensions. As a Type II
translation project, this chapter details intervention components that
maintained fidelity to the original clinical trial and those adapted to the care
management environment to promote the project’s sustainability.

Table 2.1: Differences Between Home Healthcare and MSSP Programs
Original MMM trial - Home Healthcare Translation Site - MSSP
Evidence
Participant # N=259 Capacity: 615
Inclusion
criteria for
translation
Likelihood of survival; able to
understand spoken English
Community-dwelling; No
language requirement
Core element
of intervention
Structured collaboration between
consultant pharm and RN using
medication improvement protocols
Structured collaboration
via pharm & CMs (RN or
SW) using protocols

27
Table 2.1 Continued
Medication
Management
Model
protocols and
expected
variation
Study computerized screening
algorithm alerts for problems in
intervention clients and verified by
research assistant
Pharm reviewed medications for
intervention clients
Pharm informed RN of problem
and formulated care plan using
protocols; RN contacts MD and
fulfilled next steps
If changes needed, RNs helped
with medication changes and
educated patient
Pharmacist discussed complicated
medication problems with MDs and
provided education materials to
RNs
Created risk
assessment screening
and medication
database using Home
Health Criteria
algorithm
Database alerts Pharm,
who reviews w/ CM
using updated
protocols based on
Home Health Criteria
CM contacts MD re:
medication problem
If med changes
needed, CM informs
participant about
problem
Pharm contacts MD re:
complicated problems;
change in orders
relayed to CM and
participant
Enrollment
requirements
of program
Admission to home healthcare service
requires MD orders
Must meet functional and
income levels; MD order
not required
Pre and post
measurements
Pre - Initial assessment

Post – 6 weeks after intervention
Pre – Initial or re-
assessment

Post – at 3 month
quarterly visit after
intervention

28
Table 2.1 Continued
Context
Organizational
philosophy
Medical model program
Medical/social hybrid
model program
Estimated
client length of
stay
4 weeks
Ranges from 1 month to
years
Staff
participation in
intervention
HHA nurses (RNs) and Pharmacist
Care managers (RNs &
Master’s level SWs) &
Pharmacist
Staff visitation Up to several times a week
Monthly phone calls and
quarterly visits
Documentation
methods
RN progress notes, Pharmacist
consultation/progress notes, and
intervention summary sheets
MSSP sites -
Computerized records
(assessment,
reassessment and
progress notes) for CMs &
Pharmacist
consultation/progress
notes
Facilitation
Training
associated
with Home
Health Model
1-2 sessions - approximately 3-4 hrs
total
1 orientation & at least 1
clinical session per site.
Additional training planned
to reinforce Model
Facilitation
monitoring
Site visits to monitor progress.

Minimal process follow-up with staff.
Some with administrators &
consultant pharmacists
Process evaluation with
staff during & after
implementation.

Evaluators monitoring
phase-in meetings &
ongoing implementation at
sites

29
Table 2.1 Continued
Types of
partners
assisting in this
project and
their roles
Vanderbilt Researchers
Home Healthcare Agency Program
Directors/managers in 2 sites
Consensus Panel of experts
developed Home Health Criteria
Medical consultation provided by
clinical pharmacologist(N. Brown,
MD, VU)
Home Health Advisory panel for
technical assistance/dissemination
phase
Original co-
investigators is Project
Director
Original expert panel
member
Geriatric Advisory;
Original advisory panel
member. On-going
medical and protocol
consultation provided
by advisors
Local hospital as
healthcare partner
MSSP software
developer to create
screening tool
Local university
evaluation researcher
Legend:
CM=Care managers
HHA=Home Healthcare Agency nurses
MSSP=Multipurpose Senior Services Program
MD=Prescribing medical doctor
MMM=Medication Management Model
Pharm=Consultant pharmacist
RN=Registered nurses
SW=Social workers

30
A. Evidence: The Home Health Criteria for Medication Screening and
the Medication Management Model Intervention
The evidence-based practice model employed in the CBM Intervention
originated in a home healthcare setting (Brown et al., 1998) and was
empirically tested by a team led by Vanderbilt University researchers
through a randomized clinical trial (Meredith et al., 2002; Meredith et al.,
2001). The initial stage of model development involved creating screening
criteria to identify home healthcare patients whose patterns of medication
use were problematic. This screening process, called the Home Health
Criteria, was developed by an expert consensus panel who sought to
address medication problems that, combined with clinical signs and
symptoms, indicated potential risk for potential adverse events. These
criteria focused on factors that could be assessed most easily by home
health nurses and resolved as part of a plan of care (Brown et al., 1998).
The Home Health Criteria has four categories of medication problems:
1. Inappropriate therapeutic duplication of medications;
2. Cardiovascular medication problems, e.g. uncontrolled
hypertension;
3. Inappropriate psychotropic medication use with concurrent
falls or confusion; and
4. Inappropriate use of non-steroidal anti-inflammatory drugs
(NSAIDs).

31
Data used to determine a potential medication problem included the
patient’s current medication list, age, and clinical information from the home
healthcare nursing assessment. Five evidence-based clinical indicators of
risk were assessed that related specifically to the defined medication
problems in an attempt to reduce false positive screening errors: blood
pressure, pulse rate, recent falls, dizziness, and confusion (for a complete
criteria list, see Appendix A).
To determine the frequency of medication problems in a sample of
home healthcare patients, Vanderbilt researchers compared problem
prevalence using the Home Health Criteria to problem prevalence using the
Beers Criteria (Beers, 1997). The Beers Criteria is list of 33 criteria to
define potentially inappropriate medication use in older adults, and those
criteria have been widely used for prevalence studies (Curtis et al., 2004;
Piecoro, Browning, Prince, Rantz, & Schutchfield, 2000; Rigler, Jachna,
Perera, Shireman, & Eng, 2005; Simon et al., 2005; Sloane, Zimmerman,
Brown, Ives, & Walsh, 2002). Using the Home Health Criteria, Vanderbilt
researchers reported initial prevalence data that 19% of the home
healthcare sample (N=6,718) had at least one medication problem as
compared to 17% using the Beers criteria (Meredith et al., 2001).
Prevalence rates using either screening criteria were comparable, yet when
both criteria were used, medication problem rates increased to 30% of the
sample. Additionally, over 30% of those in the sample who had at least one

32
medication problem (N=1,279) were taking nine or more medications daily
(Meredith et al., 2001).
Building on these results, Vanderbilt researchers and clinical co-
investigators developed a medication use improvement program called the
Medication Management Model. The goal of this intervention model was to
resolve medication problems identified in older adults receiving home
healthcare services. The intervention centered on the role of a consultant
pharmacist within the home healthcare setting to assist nursing staff with
identifying, preventing, and resolving medication problems among
community-dwelling, high-risk older adults. To evaluate its efficacy in the
home healthcare setting, four large home healthcare agencies piloted this
medication management intervention using a randomized control trial
design.
Below are the core elements of Medication Management Model as
tested in the home healthcare randomized control trial:
• A computerized screening algorithm identified patients who had one
of four medication problems according to the Home Health Criteria;
• Those with a verified medication problem were included in the study
and randomly assigned to intervention and control conditions;
• A structured in-home assessment was conducted by a trained
research assistant to verify problem and obtain informed consent;

33
• Once verified, the pharmacist was alerted of the intervention
patients;
• The pharmacist and nurse evaluated patients’ assessment data
using study medication improvement protocols to verify the problem
and to develop recommendations;
• The nurse contacted the prescribing physician to alert him/her to the
specific medication problem using a structured template for
discussion;
• For complex cases (e.g., cases involving tapering of psychotropic
medications), the pharmacist contacted the physician directly
regarding suggested medication changes;
• The nurse informed the patient of the medication issue, assisted with
medication changes, and monitored effects of the medication
change;
• Control subjects received usual and customary services from home
healthcare nurses; and
• A research assistant collected outcome data for all study participants
at six weeks.
Utilizing this approach, medication problems were resolved for 50% of
intervention group (N=130), compared to 38% in the control group (N=129;
p<.05) (Meredith et al., 2002). Intervention participants who had
therapeutic duplication as their medication problem showed a 71%

34
reduction in medication problems after the intervention (N=17), compared to
24% in the control group (N=24; p<.003) (Meredith et al., 2002). Use of
cardiovascular medications also improved with a 37% reduction of
problems for the intervention group (Meredith et al., 2002).
In a third phase of the project, technical assistance was provided to four
additional home healthcare agencies of varying characteristics interested in
adapting what is now the Medication Management Model into daily practice.
Among the lessons learned was that the model was feasible, flexible, and
sustainable in home healthcare, and that it may be applicable to other
programs providing in-home services to functionally-impaired older adults,
such as care management (Frey & Rahman, 2003).
B. Context: Translating the Medication Management Model from
Home Healthcare to Community-Based Care Management
Home Healthcare Context
The Medication Management Model originated in a four home
healthcare sites nationwide. Home healthcare is a medical-model home-
and community-based service that provides short-term skilled nursing care
and other skilled services (e.g., physical therapy) to those living in the
community with a physician certified need (Hughes & Pittard, 2005). Given
that over 70% of home healthcare patients are aged 65 and above,
Medicare covers 52% of all home healthcare services provided nationwide
with Medicaid (20%) and private sources (17%) also contributing (National

35
Center for Health Statistics, 2004b). The overall goal of home healthcare is
physical rehabilitation by transitioning medically-fragile patients from
institutional care, such as a hospital or skilled nursing admission, to a home
environment with the skilled medical support. Based on data from the 2000
National Home and Hospice Care Survey (NHHCS), the average length of
stay in home healthcare was 41 days in 1999-2000, down from a high of 68
days in 1993-94 (Han, Remsburg, Lubitz, & Goulding, 2004). This
reduction in service days coincides with the implementation of prospective
payment system for home care as part of the Balanced Budget Act of 1997.
Although number of days has decreased, staff may still visit home-bound
patients up to twice a day as needed.
Home healthcare nurses operate from a treatment plan developed in
partnership with the older adult’s primary physician to address needs such
as wound care, intravenous medication therapy, and other skilled medical
needs. Given that a physician must provide written orders to initiate
services, home healthcare nursing staff and other specialists work
collaboratively and serve as the physician’s “eyes and ears,” observing and
reporting on the patient’s progress in the home setting. In this context,
home healthcare nurses have an open relationship with the physician upon
which to directly request modification to the older adult’s medical treatment,
including medication changes as needed.

36
Implementation into Care Management
Upon demonstrated efficacy of the Medication Management Model in
the home healthcare setting, project leadership focused on translating this
intervention beyond the Medicare-certified home healthcare to a wider
practice audience serving other older adult populations at risk for
medication problems. This new practice audience was participants in a
California community-based care management program: the Multipurpose
Senior Service Program (MSSP). By using care management systems as
the vehicle for a Type II translation study, the CBM Intervention expanded
the vision of the intervention and testing the boundary tension between
model fidelity and adaptation to a “real world” practice environment (Castro
et al., 2004; Elliott & Mihalic, 2004).
MSSP is a Medi-Cal waiver program that provides care management
and purchase of services to eligible disabled, low-income, multi-ethnic older
adults living in the community. MSSP participants are community-dwelling
adults aged 65 and over who are Medi-Cal eligible, and demonstrate
significant functional impairment as evidenced by 1) two or more activities
of daily living impairments, or 2) at least one activity of daily living deficiency
and cognitive impairment. Older adults may access MSSP through self-
referral or a referral from caregivers and/or community providers; a
physician’s order is not required for initiating services. MSSP care
managers strive to minimize fragmentation in health and social care

37
delivery systems by improving service coordination across multiple
providers (Myrtle & Wilber, 1994).
MSSP operates from a 1915(c) Medicaid waiver that allows for the
utilization of federal and state dollars for the provision of home- and
community-based services to targeted populations in an effort to decrease
inappropriate institutionalization (N. A. Miller, Ramsland, & Harrington,
1999). The 1915(c) waiver, also known as the home- and community-
based waiver, was established in 1981 under the Omnibus Budget
Reconciliation Act, and was subsequently added to Title XIX of the Social
Security Act (Harrington, Carrillo, Wellin, Miller, & LeBlanc, 2000). Section
1915 (c)(1) of the Social Security Act specifically states that:
The Secretary may by waiver provide that a State plan
approved under this title may include as “medical assistance”
under such plan payment for part or all of the cost of home or
community-based services (other than room and board)
approved by the Secretary which are provided pursuant to a
written plan of care to individuals with respect to whom there
has been a determination that but for the provision of such
services the individuals would require the level of care
provided in a hospital or a nursing facility or intermediate care
facility for the mentally retarded the cost of which could be
reimbursed under the State plan.("Social Security Act," 1981)

There are 41 MSSP sites throughout California located in both non-profit
agencies and within county-level departments. Los Angeles County has
seven MSSP sites that serve older adults living in six of the eight Service
and Planning Areas across the county.

38
Three Los Angeles County MSSP sites housed within two non-profit,
community-based agencies participated in the CBM Intervention. Figure
2.1 illustrates the translation process through the adopting MSSP agencies
and sites. MSSP A implemented the medication management intervention
first in two sites, Burbank serving the San Fernando Valley (Site #1) and
Lynwood serving the south-central Los Angeles communities (Site #2). Site
#1 began piloting the medication management intervention in June 2004
with Site #2 following three months later. In April 2005, the translation
extended to MSSP B (Site #3 in Pasadena serving the San Gabriel Valley
communities) as it began implementation. MSSP A served as the project
headquarters and initial adopting agency that guided MSSP B through the
implementation process. It was anticipated that Site #3 in MSSP B would
benefit from lessons learned at the first two sites in dealing with context and
facilitation challenges.

39
Components of the CBM Intervention
All newly enrolled and existing MSSP clients from the three MSSP
programs participated in the CBM Intervention. The medication screening
and intervention process began upon initial assessment for participants
newly enrolled in MSSP. Continuing clients were introduced at the 6-month
or annual re-assessment visit, whichever came first, in order to the gather
the most current assessment data. The intended implementation process
followed a nearly identical process as in the original home healthcare
setting, as described and represented in Figure 2.2.

3. If Yes, care plan
developed
4. Care manager &/or
pharmacist contacts
prescribing MD re:
medication problem
5. Care manager
informs participant/
caregiver of problem
Time 1 Time 2
3 Months Later
6. Care manager follows up at
quarterly visit & reviews medication
list for changes
Was med problem resolved?
If yes, what was changed?
If no, why not & follow up
plan?
Figure 2.2: Implementation of the CBM Intervention
Prevalence
1. Computer screens
participant assessment data
via medication management
algorithm
Med problem identified?
Type of problem
2. Pharmacist & care
manager verifies problem
Is problem relevant?

• Based on usual practice, participants are randomly assigned to a
primary care manager upon enrollment – either the nurse or social
worker;

40
• Nurse and social worker care managers complete usual
assessments that include a current list of medication and the five
clinical indicators of risk (blood pressure, pulse, recent falls,
confusion, and dizziness);
• Data are entered into existing computerized database updated to
include the Home Health Criteria medication screening algorithm
(see Appendix A);
• Pharmacist reviews cases with a computerized alert of a potential
medication problem and validates the problem (Prevalence in Figure
2.2);
• Pharmacist discusses cases with primary care manager, develops
care plans and determines follow-up responsibility using protocols
from the original study (see Appendix B);
• Primary care manager contacts prescribing physician to inform of the
medication problem (Time 1);
• In complex medication cases, pharmacist contacts the prescribing
physician, providing alternative medication recommendations with
written communication transmitted via fax to the physician’s office;
• Primary care manager educates participant/caregiver about the
identified medication problems, and discusses ways to improve
health status related to the specific problem; and

41
• Primary care manager follows up with participant with identified
problems at regularly scheduled three-month visit and inquires about
medication changes (Time 2).
Implementation Similarities and Differences
Several similarities and differences were present between settings of the
CBM Intervention (MSSP) and the original randomized clinical trial (home
healthcare). Similarities include a functionally-impaired sample of
community dwelling older adults; utilization of community-based agencies
that serve diverse geographic areas and racial/ethnic participants; reliance
on government funding streams for program reimbursement; and a strict
regulatory environment due to the federal funding streams.
However, the MSSP care management program differs in several ways
from the original home healthcare program in the randomized trial.
Nationwide, Medicare-certified home healthcare programs operate on a
short-term medical model using nurses and requiring physician orders and
oversight. Patients admitted to home healthcare are often discharged from
a hospital setting following an acute illness and need skilled nursing care at
home with Medicare as the primary payer of service. Although chronic
conditions are highly prevalent in older adults (Wu & Green, 2000), those
utilizing home healthcare services are being treated for an acute episode,
such as a recent surgery, and may not have an ongoing chronic condition
that would affect long-term daily functioning.

42
Conversely, MSSP as a waiver care management program provides
longer-term care and support services for their clients using a
medical/social hybrid model approach. MSSP employs both nurses and
social workers as care managers who use their specific expertise to
evaluate participant functioning and link to appropriate services through
quarterly in-home assessment visits and monthly telephone follow-up.
Community-dwelling older adults who met the income and functional
impairment guidelines were enrolled in services through community or self-
referral, which does not require a physician’s order.
C. Facilitation: AoA-Funded Project
A facilitation team was formed at the project’s headquarters (MSSP A)
to embed the CBM Intervention into the three participating sites. The
facilitation team consisted of the agency’s Chief Executive Officer (CEO), a
project director/consultant pharmacist, an additional consultant pharmacist
on contract, two project associates, and several student interns (pharmacy
and social work students) from two local universities. Facilitation team
members were primarily located at Site #1, and they worked with care
managers and the MSSP program director at each site that managed and
supervised overall operations.
Each team member was responsible for a variety of facilitation activities.
The CEO solidified commitment and negotiated the necessary contracts
with MSSP project directors whose sites participated in the project. The

43
project director fulfilled several roles given her previous work on the home
healthcare randomized clinical trial as the consultant pharmacist, and the
dual role of key leadership and consultant pharmacist in MSSP. The
project director had specific knowledge of how the program had worked in
the home healthcare setting, which allowed her to clearly articulate issues
and solutions related to implementation fidelity and adaptation. The project
director, with support from the project associates and interns, completed the
following facilitation tasks:
• Worked with MSSP program directors to facilitate implementation in
the three sites;
• Trained care managers on the medication screening and intervention
process;
• Created a data collection and communication system to track MSSP
participants through the medication screenings, intervention process
for those with an identified problem, and follow-up visits;
• Coordinated data collection across the sites;
• Attended on-site care coordination meetings with care managers
monthly;
• Provided ongoing consultation to sites as implementation challenges
arose and to gather lessons learned; and
• Recorded changes made to intervention as needed to support
implementation in the MSSP setting.

44
Inter-Site Differences in MSSP
In addition to differences between home healthcare and MSSP, there
was also variability across the three participating MSSP sites. Similarities
across sites included a functionally-impaired, Medi-Cal population, basic
staffing requirements for social workers (2/3) and registered nurses (1/3),
and structured procedures for visitation and documentation. However, each
site had unique characteristics in terms of operational style and
programmatic engagement with clients, creating another layer of complexity
to translation as clarified in Table 2.2.

Table 2.2: Context and Facilitation Differences Across MSSP Care
Management Sites
Total N=615 Site #1 (N=216) Site #2 (N=273) Site #3 (N=126)
Context
Care manager
role
Nurses & social
workers as
primary care
managers
Social workers as
primary care
managers; nurses as
consultants
Nurses & social
workers as primary
care managers
Presence of
consulting
Pharmacist
Pharmacist
regularly on site
and by phone
occasionally
Pharmacist available
by phone and
occasionally on site
Pharmacist
available by phone
and occasionally on
site
Site
Management
MSSP A MSSP A MSSP B

45
Table 2.2 Continued
Facilitation
Facilitation
process
Project lead and
first
implementation
site
Second
implementation site
Third
implementation site
Presence of
facilitation
team
Team housed at
site
Team on-site for
meetings and has
phone consults
Team on-site for
meetings and has
phone consults
Legend
MSSP = Multipurpose Senior Services Program

Sites #1 and #3 utilized both nurses and social workers as primary care
managers. Participants in these sites were randomly assigned to the care
manager type, but both a nurse and social worker completed independent
assessments of the client upon initial enrollment. In contrast, Site #2
employed nurses in a consultative role. They completed the health
assessments for all new participants and serve as consultants to social
workers who were the primary care manager for nearly all participants at
the site. Given that it is the primary care manager’s responsibility to contact
a prescribing physician in the event of a medication problem, nurses at Site
#2 (serving as consultants and not primary care managers) would not likely
be involved in this task. This situation meant that social workers in Site #2,
who generally operate from a social model approach to treatment, were

46
expected to discuss the medication problem, a medically-related issue, with
the prescribing physician.
Site # 1 was also the first implementation site and housed the facilitation
team for the entire project. This structural issue allowed care managers
from Site #1 a greater access to consultant pharmacists and facilitation
team regarding medication questions and training issues as they arose than
staff at the other sites. It was anticipated that this arrangement might
impact the implementation of the intervention. A further contextual
difference was the organizational nature where sites were located. MSSP
A (Sites #1 and #2) was within a community based organization that had
with no formal affiliation with a medical setting; MSSP B (Site #3) was
affiliated with and located at a local hospital. This proximity to health care
services, both organizationally and geographically, could impact the staff’s
ability and willingness to implement the medication management
intervention.
D. Conclusion
The CBM Intervention exemplifies the translation of an evidence-based
practice beyond its original efficacy trial in home healthcare sites to a new
long-term care practice environment in the MSSP care management
program. Although core features of original trial were in effect to maintain
fidelity, modifications were made to adapt the intervention to pre-existing
contextual elements of the care management program. The next three

47
chapters evaluate the effectiveness of the translated medication
management model to identify and resolve medication problems and care
manager perceptions of participating in the implementation process. More
specifically, Chapter III reports on the prevalence of medication problems in
the MSSP sample based on the Home Health Criteria problem list. Chapter
IV then details the success of the intervention phase to resolve medication
problems for those identified in Chapter III. Finally, Chapter V describes
perspectives and feedback from the nurse and social work care managers
on implementing an evidence-based practice into the existing MSSP
program structure.

48
III. PREVALENCE OF MEDICATION PROBLEMS IN
DUALLY-ELIGIBLE OLDER ADULTS
A. Introduction
Medication utilization and its potential problems are critical issues in
health care service delivery for adults aged 65 and above. With nearly half
of all older adults taking three or more medications on a monthly basis
(National Center for Health Statistics, 2004a) and 62% reporting two or
more chronic conditions (Wu & Green, 2000), these figures suggest that
older adults have an increased likelihood of being prescribed multiple
medications from multiple treating physicians. Increased prescribing from
multiple sources contributes to a greater potential for medication-related
problems (Hajjar et al., 2005), including adverse drug events, resulting in
negative health outcomes (Fu, Liu, & Christensen, 2004). Adverse drug
events, defined as an injury due to a drug reaction or medication error
(Silverman, Stapinski, Huber, Ghandi, & Churchill, 2004), directly account
for 7,000 deaths each year (Institute of Medicine, 2000; Phillips,
Christenfeld, & Glynn, 1998). An adverse drug event may also be
evaluated as a new medical condition, such as increased confusion
diagnosed as a dementing illness. This can lead to a “prescription
cascade” (Vinks, de Koning, de Lange, & Egberts, 2006), where additional
pharmacological treatment is initiated to address symptoms from an
adverse drug event further contributing to the problem (Rochon & Gurwitz,

49
1997, 1999). Nationwide, the average cost of medication-related problems
and their associated conditions is over $2 billion annually (Institute of
Medicine, 2000), and is expected to increase as more medications enter the
marketplace having unknown interaction effects with existing prescription
drugs. Recent studies have reported that over one-quarter of adverse drug
events in ambulatory settings could have been prevented (Gurwitz et al.,
2003), and 6-week health care cost increase related to a preventable
adverse drug event were nearly $2,000 (Field et al., 2005).
A major contributing factor to adverse drug events in older adults is
suboptimal prescribing patterns (Hanlon, Schmader, Ruby, & Weinberger,
2001), which include overuse or polypharmacy (Larson & Hoot-Martin,
1999), underuse of needed medication (Rochon & Gurwitz, 1999), and
inappropriate medication use (Liu & Christensen, 2002). Ample research
exists reporting the high prevalence of potential medication problems in
older adults due to inappropriate prescribing patterns in various health and
community settings (Aparasu & Mort, 2000; Liu & Christensen, 2002),
including those in hospitals (Hajjar et al., 2005), nursing homes (Dhalla et
al., 2002; Lau, Kasper, Potter, & Lyles, 2004), out-patient care (Curtis et al.,
2004; Goulding, 2004; Maio et al., 2006; Spiker, Emptage, Giannamore, &
Pedersen, 2001), managed care (Simon et al., 2005), community living
(Hanlon et al., 2002; Willcox, Himmelstein, & Woolhandler, 1994), and for
those who are Medicaid beneficiaries (Golden et al., 1999; Piecoro et al.,

50
2000; Rigler et al., 2005). Rates of potentially inappropriate medication use
by community-dwelling adults has varied from 12% to 40% (Zhan et al.,
2001), ranging as high as 48% in older adults at risk for nursing home
placement (Rigler et al., 2005). Much of the research has focused on older
adults in the United States, however recent studies suggest that this
problem is international in scope (Azoulay, Zargarzadeh, Salahshouri,
Oraichi, & Berard, 2005; Fialova et al., 2005; Lechevallier-Michel et al.,
2005; Niwata, Yamada, & Ikegami, 2006; Onder et al., 2003; Pitkala,
Strandberg, & Tilvis, 2002; van der Hooft et al., 2005).
Predictors for potential medication problems found in samples of dual
eligibles included advancing age, females, Caucasians, and those reporting
poor self rated health (Fick et al., 2001; Piecoro et al., 2000). Studies have
also reported that nursing home residents have higher rates of potential
medication problems compared to those living in the community (Lane et
al., 2004; Piecoro et al., 2000). However, these studies did not parcel out
community-dwelling older adults who might have similar functional abilities
as nursing home residents and there would be eligible for state Medicaid
waiver programs.
Medicaid waiver programs under the 1915(c) federal waiver offer home-
and community-based services to functionally impaired older adults living in
the community in order to avoid inappropriate institutionalization (N. A.
Miller et al., 1999). Participants enrolled in Medicaid waiver programs are

51
considered to have substantial enough physical and/or cognitive impairment
that would necessitate nursing home placement (Harrington et al., 2000; N.
A. Miller et al., 1999). However, unlike those in nursing home care, waiver
participants do not benefit from a routine medication review that could
identify potentially inappropriate medications and adverse side effects
(Eans, 2000). A recent study comparing the nursing home and Medicaid
waiver populations show that waiver clients had a 48% prevalence of
potential medication problems with nursing home residents at 38% (Rigler
et al., 2005). Waiver participants also had a significantly higher number of
medications (Rigler et al., 2005).
The purpose of this chapter is to report prevalence rates and identify
predictors of medication problems in dually-eligible older adults enrolled in a
Medi-Cal waiver care management program. Two inter-related levels of
medication problem prevalence rates are reported for this sample. The first
prevalence rate is for potential medication problems determined by using a
focused drug regimen review criteria. This measure is important because
the vast majority of prevalence studies in the literature report the type and
frequency of potential medication problems based on similar types of review
criteria, and are therefore available for comparison. The second prevalence
level is for the rate of confirmed medication problems. For those who
screened in as having a potential problem, the confirmation process
involved an in-depth clinical review by a consulting pharmacist with

52
expertise in geriatric pharmacology to verify that a medication problem was
present. The process for screening potential medication problems and
validating confirmed problems is described below.
Home Health Criteria
Various criteria have been utilized to evaluate the prevalence of
potential medication problems (Fick et al., 2003; Hanlon et al., 2002; Pugh
et al., 2005; Shelton, Fritsch, & Scott, 2000; Stuck et al., 1994;
Viswanathan, Bharmal, & Thomas III, 2005), with most studies using the
most recent version of the well-known Beers criteria (Beers, 1997). A
common element of these criteria is the exclusive use of prescription
information (e.g., specific medications and/or doses) to determine
medication appropriateness without consideration of clinical indicators of
functioning. Screening techniques that only use medication lists overlook
clinical risk factors, such as a recent fall or presence of confusion, that may
indicate an adverse drug event (Beers, 1997), which can lead to decreased
daily functioning, diminished cognitive capacity, and potential utilization of
additional health care services (e.g., hospitalization for a fall-related injury).
This is especially true for routinely-prescribed medications that are not
generally considered as problematic for older adults, such as tricyclic anti-
depressants.
One medication screening process that includes clinical information is
the Home Health Criteria (Brown et al., 1998; Meredith et al., 2002;

53
Meredith et al., 2001). Developed by an expert consensus panel, the Home
Health Criteria was initially created to identify home healthcare patients
whose patterns of medication use, combined with clinical symptoms,
indicated potential risk for adverse drug events. The four categories of
medication problems in the Home Health Criteria are:
1. Unnecessary therapeutic duplication of medications;
2. Inappropriate psychotropic medication use with concurrent falls or
confusion;
3. Cardiovascular medication problems (e.g., uncontrolled
hypertension); and
4. Inappropriate use of non-steroidal anti-inflammatory drugs (NSAIDs)
with risk for peptic ulcer complaints (Meredith et al., 2001).
Clinical data used to determine a potential medication problem included the
patient’s current medication list, age, and five evidence-based clinical
indicators of risk that relate to the defined medication problems: falls,
confusion, dizziness, blood pressure, and pulse. The Home Health Criteria
contains elements that could be assessed and resolved most easily by
nurses in the field within a standard home healthcare treatment plan (Brown
et al., 1998).
Efficacy for this medication screening criteria was established by
comparing the Home Health Criteria and the Beers Criteria using patients
enrolled in a Medicare-certified home healthcare agency. Prevalence data

54
showed that 19% of a home healthcare sample (N=6,718) had at least one
medication problem with the Home Health Criteria as compared to 17%
using the 1997 Beers criteria (Meredith et al., 2001). When both criteria
were used, medication problem rates increased to 30% of the sample
(Meredith et al., 2001). Additionally over 30% of those in the sample who
had at least one medication problem (N=1,279) were taking nine or more
medications, which was the top quintile of medication use (Meredith et al.,
2001). These prevalence data speak to the vulnerability of older adults to
medication problems and raise great concern about potential adverse
consequences.
CBM Intervention
Based on this work, these criteria were employed with a dually-eligible
population through the CBM Intervention. The CBM Intervention, funded by
the federal Administration on Aging’s Evidence-Based Prevention Initiative,
was a Los Angeles County-based project to implement an evidence-based
medication management intervention in a Medi-Cal waiver care
management program (Alkema & Frey, 2006). The first phase of the CBM
Intervention was to identify the prevalence of medication problems using
the Home Health Criteria in older adults participating in the waiver care
management program called the Multipurpose Senior Services Program
(MSSP).

55
MSSP is a slot-based, 1915(c) home- and community-based waiver
program that provides care management and purchase of services to
disabled, racially and ethnically diverse, low-income older adults. The
overall goal of MSSP is to decrease and/or avoid inappropriate
institutionalization for this population (L. S. Miller, Clark, & Clark, 1985).
MSSP participants are community-dwelling adults aged 65 and over who
are Medi-Cal eligible, and demonstrate significant functional impairment as
evidenced by 1) two or more activities of daily living impairments, or 2) at
least one of activities of daily living deficiency and cognitive impairment.
There are currently 41 MSSP providers throughout California with services
offered in 50 of the 58 counties (California Department of Aging, 2006).
The average age of MSSP participants statewide is 81 years, and nearly
70% of the sample is female. Three MSSP sites throughout Los Angeles
County implemented the Home Health Criteria into standard practice to
evaluate for medication problems.
Hypotheses
Based on the results from the original home healthcare prevalence
study (Meredith et al., 2001) and previously identified predictors of
medication problems, four hypotheses were tested in this analysis. It was
anticipated that:
1. MSSP participants will have a higher prevalence of medication
problems than the original home healthcare sample;

56
2. Participants with greater numbers of medications are more likely to
have at least one medication problem;
3. Therapeutic duplication will have the highest prevalence of the four
medication problem types in the MSSP sample; and
4. Predictors of medication problems would include advancing age,
females, Caucasians, and those with poorer health status.
B. Methods
Data Source
As part of phase one of the CBM Intervention project, baseline data
were gathered from participants enrolled in three Los Angeles County-
based MSSP sites from June 2004 to January 2006. Participating sites
were affiliated with two MSSP providers that served racial/ethnically diverse
older adults in half of the county’s service and planning areas. Site #1 and
#2 are affiliated with a community-based social service agency with a
strong research core. Site #3 is housed in the care management program
under the auspice of a local Hospital. The evaluation was conducted by a
team of researchers at the University of Southern California; protocols and
instruments were approved by the institutional review boards for all program
partners. Administrative and care management data were extracted from
the three MSSP sites with blinded, encrypted data transmitted to the
research partners for evaluation of the project.

57
Medication Review Process
Newly enrolled and existing MSSP participants in three sites were
screened for having medication problems by one of two consultant
pharmacists with extensive training in geriatric pharmacotherapy using the
Home Health Criteria screening algorithm for the drug regimen review. All
newly enrolled and existing participants at Site #1 and #2 were screened for
medication problems; Site #3 screened only newly enrolled participants due
to a shorter, six-month implementation schedule. All participants prior to
medication screening were given informed consent indicating that they may
discontinue participation in CBM Intervention at any time and continue
receiving all other MSSP services. Data needed to conduct the medication
screening were collected by MSSP care management staff during a
regularly-scheduled home visit using the standard MSSP care management
clinical assessment protocols. Care management staff consisted of both
nurses and social workers, and as part of standard agency practice,
participants were randomly assigned to one of the professional types.
However upon initial enrollment in MSSP, each participant completed a
health assessment with a nurse and a psychosocial assessment with a
social worker, regardless of which professional would be the primary case
manager. As part of the assessment process, care managers documented
medications participants were currently taking in the last seven days and

58
the five evidence-based clinical indicators of falls, confusion, dizziness,
blood pressure, and pulse.
In order to determine the presence of a potential medication problem, all
participants were screened by one of two consultant pharmacists using the
Home Health Criteria protocols. If a participant met the Home Health
criteria for a potential medication problems (e.g. presence of potential
therapeutic duplication), then the pharmacist confirmed that a medication
problem was present by clinically evaluating the following issues.
1. Verified that the participant was actually taking the medication(s) in
question;
2. Confirmed the dosage and frequency of medication regimen;
3. Identified the purpose and need for the medication in question based
on clinical symptoms; and
4. Inquired about ongoing problems (side effects, administration, etc.)
with medication in question.
Both pharmacists used the same confirmation process and consulted with
each other as needed. Each of the four medication problem types had
specific confirmation criteria related to these issues above (see Appendix C
for a detailed description of specific confirmation criteria).

59
Measures
Dependent Variables
Dependent variables measured the prevalence of potential and
confirmed medication problems in one of the four types as defined in the
Home Health Criteria (Meredith et al., 2001). The four medication problem
types, measured as dichotomous variables, included unnecessary
therapeutic duplication; inappropriate use of psychotropic medication;
cardiovascular medication problems; and inappropriate use of non-steroidal
anti-inflammatory drugs (NSAIDs).
Unnecessary therapeutic duplication was defined as concurrent use of
two or more medications from the same class based on the 100 most
frequently prescribed medications (Brown et al., 1998). Inappropriate
psychotropic medication use included use of benzodiazepines, tricyclic
antidepressants, or antipsychotics with a fall or presence of confusion in the
last three months. Cardiovascular medication problems included poorly
controlled hypertension, hypotension, orthostasis, and bradycardia with
associated medications; medication list and blood pressure and pulse
readings were used to identify this problem type. Problems with use of
NSAIDs were defined using these medications (excluding low doses of
aspirin) for participants age 80 and over, or those taking anticoagulants or
oral corticosteroids. In addition to individual problem types, the number of

60
potential and confirmed medication problems was recorded for each
participant.
If a potential medication problem was not confirmed by the consultant
pharmacist through further investigation, this condition was coded into one
of two categories: false positive and unconfirmed medication problems. A
false positive indication meant that although the drug regimen review
screening identified a potential problem, further clinical review revealed that
a medication problem was not present. There were several reasons for the
false positive identification. First, upon further review it was determined that
the participant was not consuming the alerting medications. Another
reason was that the clinical indicator of risk associated with a medication
problem was not related to a medication issue, such as a recent fall caused
by a cracked sidewalk. The third situation was when a potentially
problematic medication regimen was deemed “appropriate” based on the
presence of specific medical conditions. For example, a participant taking
Zyprexa, an antipsychotic agent, was evaluated by the care manager as
being confused. The medication was deemed appropriate by the
pharmacist because the participant had been diagnosed with Alzheimer’s
disease and bipolar disorder with psychotic features. Another example was
a participant diagnosed with rheumatoid arthritis prescribed 400mg of
ibuprofen three times a day and over the counter Advil as-needed.

61
The other indication was for unconfirmed medication problems, which
occurred when additional clinical measures needed to clearly delineate the
medication problem were not available. This category mostly captured
potential problems related to cardiovascular medication issues due to the
inability to gather additional blood pressure or pulse readings necessary to
confirm the problem. In addition to these two categories, some medication
problems could not be confirmed because the participant terminated from
the MSSP program prior to the pharmacist’s clinical review (1.6%).
Independent Variables
Independent variables included socio-demographic and health
characteristics, and clinical health indicators specific to this study. Age was
a continuous variable starting at 65 years due to MSSP eligibility
requirements. Gender was coded as female. Race/ethnicity was a
categorical variable that represented Caucasian, African-American,
Latino/a, Asian/Pacific Islander, Native American, and Unknown; Native
American and Unknown were combined into “Other/Unknown” due to small
cell sizes for each. Caucasian served as the reference group in multivariate
analyses. Primary language spoken included English, Spanish, Armenian,
and Chinese as those with the highest prevalence. Other known languages
included Arabic, Farsi, French, Korean, and Tagalog. A dichotomous
variable was created to represent primary English speakers compared to all
others. Marital status included married, widowed, divorced/separated,

62
single, and unknown/refused. Education was initially recorded as six
categories: no formal education, grade school, high school diploma or
equivalent, some college, completed college, and unknown/refused. Living
situation was a dichotomous variable defined as those who lived alone (1)
compared to those who lived with someone (0). Participants newly enrolled
in MSSP were defined as those who received less than three months of
services before engaging in the CBM Intervention (1) as compared to all
others (0). This variable was included to delineate those new to care
management services that might have had more acute care needs than
those who had received services beyond the initial assessment period.
Health characteristics focused on health status and service utilization.
Number of prescribed medications was a discrete count variable that
included prescribed medication, relevant over-the-counter medication (e.g.,
Diphenhydramine), and a single count for any vitamins if the participant
took them regularly. Although not part of the original home healthcare
study, the addition of vitamins in a medication count has been well
documented in the literature (Kaufman, Kelly, Rosenberg, Anderson, &
Mitchell, 2002; Sloane et al., 2002; Sorensen, Stokes, Purdie, Woodward, &
Roberts, 2005). Admission to the emergency room, hospital, or skilled
nursing facility within the last year was recorded as a dichotomous variable.
Evidence-based clinical indicators of risk to determine potential
medication problems included falls, confusion, dizziness, blood pressure,

63
and pulse. Falls were measured by participant self-report of an incident in
the last three months, regardless of whether an injury occurred as a result.
Although definitions vary for what constitutes a “fall” (Zecevic, Salmoni,
Speechley, & Vandervoort, 2006), this measure accounted for the
participant’s perception of a fall, which can affect both falls efficacy and fear
of falling (Tinetti & Williams, 1998). Presence of confusion was measured
by the case manager’s assessment of participant functioning using
guidelines of the Confusion Assessment Method (Inouye et al., 1990).
Dizziness was also a dichotomous measure based on participant self-
report. To evaluate those at higher risk of a negative health outcome, a
dichotomous variable was created that incorporated those who lived alone
and had a fall, dizziness, or confusion (Lived Alone x Risk). Blood pressure
and pulse, measured by the nurse care managers during a health
assessment, were clinically relevant to the specific cardiovascular
medication problem and therefore were not included in the analyses.
Sample
In the three MSSP sites, 615 participants were screened for medication
problems. A power analysis (Cohen, 1988) suggested that a sample of 237
case management participants would be required to obtain a 95%
confidence interval of +/- 5% given the medication problem prevalence of
19% found in the original home healthcare sample. Table 3.1 reports the
characteristics of the MSSP sample. Participants were an average of 81

64
years old with nearly 80% being female. The racial/ethnic make-up of the
sample reflects the diversity of Los Angeles County; however the sample
had a higher concentration of African-Americans and Latinos relative to the
county population. Forty percent of the sample was non-English speaking
as the primary language of choice, with the largest percentage speaking
Spanish. Over half of the sample was widowed and 20% were married.
Education level data were missing for 34% of the sample as this measure
was not consistently recorded at intake until 2004. However, approximately
40% reported having at least a high school education and more than 40%
of the sample reported living alone. These figures are comparable to dually
eligible older adults nationwide (Kasper, Elias, & Lyons, 2004).
Additionally, 36% of the sample was newly enrolled in MSSP.
Thirty-eight percent of participants reported having an emergency room,
hospital, or skilled nursing facility admission in the last year. Medication
use was very high, ranging as high as taking 27 different medications
regularly. Half of the sample took nine or more medications daily, and the
top quintile of use was 12 or more medications. Over 20% of the sample
reported a fall in the last three months, 27% reported dizziness, and 31%
reported experiencing confusion; 21% lived alone with one of these clinical
indicators of risk.

65
Table 3.1: Characteristics of the CBM Intervention Sample (N=615)
N % Mean (SD)
Socio-Demographics
Age (65-108) 80.80 (7.76)
Female 490 79.7%
Race/Ethnicity
Caucasian 146 23.7%
African-American 241 39.2%
Latino/a 146 23.7%
Asian/Pacific Islander 52 8.5%
Other/Unknown 30 4.9%
Primary Language
English 363 59.0%
Spanish 137 22.3%
Armenian 37 6.0%
Chinese 12 2.0%
Known Other 42 6.8%
Unknown/Refused 24 3.9%
Marital Status
Married 123 20.0%
Widowed 325 52.8%
Divorced/Separated 105 17.1%
Single 47 7.6%
Unknown/Refused 15 2.4%
Education
No Education 21 3.4%
Grade School 145 23.6%
High School 168 27.3%
Some College 45 7.3%
College Graduate 26 4.2%
Unknown/Refused 210 34.1%
Lived Alone 255 41.5%
Newly Enrolled in MSSP 223 36.3%
Health Status
ER/Hospital/SNF in Last Year 234 38.0%
# of Medications (0-27) 8.76 (4.30)
12+ Medications 137 22.3%
Falls in Past 3 Months 135 22.0%
Dizziness 168 27.3%
Confusion 190 30.9%
Lived Alone x Risk 126 20.5%
Risk defined presence of falls, dizziness, or confusion

66
Table 3.2 demonstrates the sample’s significant diversity across the
three MSSP sites on race/ethnicity, primary language, marital status,
education, living situation, percentage of new enrollees, health service
utilization, number of medications, and evidence-based clinical indicators.
The racial/ethnic composition of Site #1 nearly mirrors the make-up of Los
Angeles County for people aged 65 and above with 55% Caucasians, 11%
African-American, 19% Latino, and 7% Asian/Pacific Islander. In contrast,
Site #2 served a primarily African-American population and Site #3
participants were mostly of Latino background (p<.001). These differences
were also reflected in languages used across sites with Site #2 having the
greatest percentage of English speakers (p<.001). Site #2 also had the
highest percentage of widowed participants and those living alone (p<.001).
The distribution of education level was also highly variable, although
skewed due to substantial missing data in the sites (p<.001). Site #2 had
the highest percentage of participants living alone at nearly 50% (p=.016).
Site #1 had the significantly lowest percentage of newly enrolled
participants at 26%, compared to the other sites with over 40% (p<.001).
Site #3 reported the lowest rate of hospital, skilled nursing, or emergency
room admissions, and the highest number of medications on average
(p<.001). Site #2 reported the lowest percentages of dizziness and
confusion across sites (p<.001). However, there were no site differences
for those living alone with risk factors of falls, dizziness, or confusion.

67

Table 3.2: Characteristics of CBM Intervention Sample by Site (N=615)
N % Mean (SD) N % Mean (SD) N % Mean (SD)
Socio-demographics
Age (65-108) 80.76 (7.42) 80.83 (7.96) 80.81 (7.99)
Female 164 75.9% 223 81.7% 103 81.7%
Race/Ethnicity***
Caucasian 118 54.6% 7 2.6% 21 16.7%
African-American 23 10.6% 192 70.3% 26 20.6%
Latino/a 41 19.0% 59 21.6% 46 36.5%
Asian/Pacific Islander 14 6.5% 10 3.7% 28 22.2%
Other/Unknown 20 9.3% 5 1.8% 5 4.0%
Primary Language***
English 119 55.1% 183 67.0% 61 48.4%
Spanish 40 18.5% 61 22.3% 36 28.6%
Armenian 35 16.2% 0 0.0% 2 1.6%
Chinese 0 0.0% 3 1.1% 9 7.1%
Known Other 18 8.3% 7 2.6% 17 13.5%
Unknown/Refused 4 1.9% 19 7.0% 1 0.8%
Marital Status***
Married 52 24.1% 41 15.0% 30 23.8%
Widowed 102 47.2% 160 58.6% 63 50.0%
Divorced/Separated 49 22.7% 33 12.1% 23 18.3%
Single 8 3.7% 29 10.6% 10 7.9%
Unknown/Refused 5 2.3% 10 3.7% 0 0.0%
Education***
No Education 10 4.6% 4 1.5% 7 5.6%
Grade School 44 20.4% 43 15.8% 58 46.0%
High School 70 32.4% 58 21.2% 40 31.7%
Some College 17 7.9% 16 5.9% 12 9.5%
College Graduate 15 6.9% 2 0.7% 9 7.1%
Unknown/Refused 47 21.8% 150 54.9% 0 0.0%
Lived Alone* 75 34.7% 133 48.7% 47 37.3%
Newly Enrolled in MSSP*** 57 26.4% 114 41.8% 52 41.3%
Health Status
ER/Hospital/SNF in Last Year** 88 40.7% 125 45.8% 21 16.7%
# of Medications (0-27) 8.29 (4.43) 8.85 (4.08) 9.37 (4.46)
12+ Medications 43 19.9% 64 23.4% 30 23.8%
Falls in Past 3 Months 39 18.1% 60 22.0% 36 28.6%
Dizziness*** 77 35.6% 50 18.3% 41 32.5%
Confusion*** 82 38.0% 61 22.3% 47 37.3%
Lived Alone x Risk 40 18.5% 57 20.9% 29 23.0%
*p<.05; **p<.01; ***p<.001
Risk defined as presence of falls, dizziness, or confusion
Site #1 (N=216) Site #2 (N=273) Site #3 (N=126)

68
Analysis
Data analysis was completed using SPSS software, Version 11.
Descriptive analyses of the drug regimen screening and clinical review were
used to identify prevalence rates and associated types of the four
medication problems. Bivariate statistics report prevalence rates of
potential and confirmed medication problems for the sample including
having a medication problem, prevalence of each problem type, and
correlations of these types with socio-demographic and health
characteristics. Logistic regression examined odds ratios of socio-
demographic and health characteristics on the prevalence of medication
problems. For further specificity, psychotropic medications were delineated
by the clinical criteria for inappropriate use to due a fall or presence of
confusion.
A few variables were dropped from the analyses due to multicolinearity.
Dichotomous variables of marital status correlated highly with living
situation (r=-.372) and age (r=.331), and were therefore removed from
multivariate analyses. Also, clinical indicators of risk (falls, dizziness, and
confusion) were part of the criteria to identify psychotropic and
cardiovascular medication problems. Therefore, analyses that contained
these medication problem types did not include the variable for those living
alone with a risk factor.

69
C. Results
Figure 3.1 describes the flow of the MSSP participants through the
Home Health Criteria screening and pharmacist confirmation process.
Regarding potential medication problems, nearly 49% of the sample had at
least one identified problem. The most prevalent potential problem type
was unnecessary therapeutic duplication at 24%, followed by psychotropic
medication problems and cardiovascular medication problems at 14%.
Thirteen percent had a potential problem with NSAIDs.
Following the pharmacist clinical review to verify potential problems
identified, 181 participants (29% of the sample) had a confirmed medication
problem. Therapeutic duplication remained the most prevalent type at 14%
with over half of the potential problems being confirmed. Nearly 11% of the
sample had a confirmed psychotropic medication problems, with almost
three-fourths of those screened in as a potential problem being confirmed.
This was also the case for inappropriate use of NSAIDS, where 9% of the
sample had a confirmed problem. Only 4% of the sample, or one-third of
those who screened in with a potential cardiovascular medication problem
were confirmed upon clinical review. Of the 118 people who did not have a
confirmed medication problem, 83 were due to false positive indications, 25
were unconfirmed problems due to lack of verifying information, and 10
participants terminated from MSSP prior to the pharmacist’s clinical review.

70

Table 3.3 shows the prevalence of potential and confirmed problems for
the total sample and across the three sites. Observing potential problems,
116 participants (19%) had two or more problems with over half of those
having the problems confirmed. Inappropriate use of psychotropic
medication was further detailed into the two associated clinical symptoms,

71
confusion and falls. Although variation existed in the percentages of
potential and confirmed medication problems across sites, none of these
differences was statistically significant for any of the problem categories.

Table 3.3: Prevalence of Potential and Confirmed Medication Problems for Total Sample and Sites
N% N% N % N %
Potential Medication Problems
1+ Problems 299 48.6% 110 50.9% 125 45.8% 64 50.8%
2+ Problems 116 18.9% 43 19.9% 48 17.6% 25 19.8%
3+ Problems 33 5.4% 14 6.5% 11 4.0% 8 6.3%
Potential Problems By Type
Therapeutic Duplication 149 24.2% 55 25.5% 65 23.8% 29 23.0%
Psychotropic Medication 88 14.3% 32 14.8% 31 11.4% 25 19.8%
Confusion 49 8.0% 21 9.7% 17 6.2% 11 8.7%
Falls 52 8.5% 19 8.8% 17 6.2% 16 12.7%
Cardiovascular Medication 87 14.1% 26 12.0% 43 15.8% 18 14.3%
NSAIDs 79 12.8% 32 14.8% 29 10.6% 18 14.3%
Confirmed Medication Problems
1+ Problems 181 29.4% 62 28.7% 78 28.6% 41 32.5%
2+ Problems 64 10.4% 22 10.2% 27 9.9% 15 11.9%
3+ Problems 20 3.3% 7 3.2% 7 2.6% 6 4.8%
Confirmed Problems By Type
Therapeutic Duplication 87 14.1% 28 13.0% 39 14.3% 20 15.9%
Psychotropic Medication 65 10.6% 25 11.6% 24 8.8% 16 12.7%
Confusion 37 6.0% 17 7.9% 13 4.8% 7 5.6%
Falls 40 6.5% 16 7.4% 13 4.8% 11 8.7%
Cardiovascular Medication 27 4.4% 5 2.3% 15 5.5% 7 5.6%
NSAIDs 53 8.6% 20 9.3% 21 7.7% 12 9.5%
**No Statitstical Differences Across Sites
Potential Medication Problems = Screened in as a Problem Using Home Health Criteria
Confirmed Medication Problem = Potential Medication Problem Confirmed through Pharmacist Review
NSAIDs = Non-Steroidal Anti-Inflammatory Drugs
Sample (N=615) Site #1 (N=216) Site #2 (N=273) Site #3 (N=126)

Table 3.4 compares participants who had a potential medication
problem with those whose medication problem was confirmed through the

72
pharmacist review. There were no significant differences between the
groups on socio-demographic or health characteristics and only one
difference for medication problems. Cardiovascular medication problems
were identified as a potential problem, but did not convert to a confirmed
problem at a similar rate as the other three areas (p<.001).

Table 3.4: Differences between Participants with Potential and Confirmed Medication Problems
N % Mean (SD) N % Mean (SD) t-Test Chi
2
Socio-demographics
Age (65-108) 81.05 (7.71) 80.83 (7.88) 0.300
Female 244 81.6% 149 82.3% 0.039
Race/Ethnicity 0.821
Caucasian 76 25.4% 46 25.4%
African-American 118 39.5% 72 39.8%
Latino/a 68 22.7% 43 23.8%
Asian/Pacific Islander 22 7.4% 14 7.7%
Other/Unknown 15 5.0% 6 3.3%
English-Speaking 182 60.9% 114 63.0% 0.213
Married 51 17.1% 27 14.9% 0.379
Widowed 166 55.5% 103 56.9% 0.088
High School & Above 128 42.8% 76 42.0% 0.031
Lived Alone 131 43.8% 86 47.5% 0.624
Newly Enrolled in MSSP 120 40.1% 77 42.5% 0.270
Health Status
ER/Hospital/SNF in Last Year 121 40.5% 83 45.9% 1.339
# of Medications (0-27) 10.06 (4.45) 10.47 (4.66) -0.961
12+ Medications 96 32.1% 62 34.3% 0.235
Falls in Past 3 Months 88 29.4% 62 34.3% 1.221
Dizziness 96 32.1% 67 37.0% 1.212
Confusion 104 34.8% 64 35.4% 0.016
Lived Alone x Risk 81 27.1% 56 30.9% 0.819
Medication Problems
2+ Problems 116 38.8% 64 35.4% 0.568
Therapeutic Duplication 149 49.8% 87 48.1% 0.141
Psychotropic 88 29.4% 65 35.9% 2.180
Confusion 49 16.4% 37 20.4% 1.260
Falls 52 17.4% 40 22.1% 1.613
Cardiovascular*** 87 29.1% 27 14.9% 12.518
NSAIDs 79 26.4% 53 29.3% 0.463
*p<.05; **p<.01; ***p<.001
Risk defined as presence of falls, dizziness, or confusion
Potential (N=299) Confirmed (N=181)

73
Table 3.5 compares the differences between the total sample (N=615)
and those identified as having a potential medication problem (N=299).
Participants with a potential problem had a significantly higher number of
medications (p<.001) with a larger percentage taking 12 or more
medications (p<.001). Almost 30% of those in the potential problem group
reported a fall compared to 22% in the full sample (p=.013). Also, 27% of
those with a potential problem lived alone and had a fall, dizziness, or
confusion (p=.025).

Table 3.5: Differences between Sample and Participants with Potential Medication Problems
N % Mean (SD) N % Mean (SD) t-Test Chi
2
Socio-demographics
Age (65-108) 80.80 (7.76) 81.05 (7.71) -0.458
Female 490 79.7% 244 81.6% 0.474
Race/Ethnicity 0.630
Caucasian 146 23.7% 76 25.4%
African-American 241 39.2% 118 39.5%
Latino/a 146 23.7% 68 22.7%
Asian/Pacific Islander 52 8.5% 22 7.4%
Other/Unknown 30 4.9% 15 5.0%
English-Speaking 363 59.0% 182 60.9% 0.285
Married 123 20.0% 51 17.1% 1.131
Widowed 325 52.8% 166 55.5% 0.578
High School & Above 239 38.9% 128 42.8% 1.305
Lived Alone 255 41.5% 131 43.8% 0.455
Newly Enrolled in MSSP 223 36.3% 120 40.1% 1.288
Health Status
ER/Hospital/SNF in Last Year 234 38.0% 121 40.5% 0.496
# of Medications (0-27)*** 8.76 (4.30) 10.06 (4.45) -4.239
12+ Medications*** 137 22.3% 96 32.1% 11.039
Falls in Past 3 Months* 135 22.0% 88 29.4% 6.103
Dizziness 168 27.3% 96 32.1% 2.247
Confusion 190 30.9% 104 34.8% 1.394
Lived Alone x Risk* 126 20.5% 81 27.1% 5.006
*p<.05; **p<.01; ***p<.001
Risk defined as presence of falls, dizziness, or confusion
Sample (N=615) Potential (N=299)

74
For an additional comparison, Table 3.6 reports differences between the
total sample (N=615) and those who had a confirmed medication problem
(N=181). Similar to the table above, those with a confirmed problem took
more medications (p<.001) with 34% taking 12 or more medications
(p<.001). Over one-third of this group reported having a fall (34% vs. 22%;
p<.001) and they had higher rates of dizziness than the total sample (37%
vs. 28%; p=.011). Those with a confirmed problem also had a greater
frequency of those living alone with a risk factor (31% vs. 21%; p=.003).

Table 3.6: Differences between Sample and Participants with Confirmed Medication Problems
N % Mean (SD) N % Mean (SD) t-Test Chi
2
Socio-demographics
Age (65-108) 80.80 (7.76) 80.83 (7.88) -0.046
Female 490 79.7% 149 82.3% 0.618
Race/Ethnicity 1.018
Caucasian 146 23.7% 46 25.4%
African-American 241 39.2% 72 39.8%
Latino/a 146 23.7% 43 23.8%
Asian/Pacific Islander 52 8.5% 14 7.7%
Other/Unknown 30 4.9% 6 3.3%
English-Speaking 363 59.0% 114 63.0% 0.913
Married 123 20.0% 27 14.9% 2.362
Widowed 325 52.8% 103 56.9% 0.928
High School & Above 239 38.9% 76 42.0% 0.572
Lived Alone 255 41.5% 86 47.5% 2.091
Newly Enrolled in MSSP 223 36.3% 77 42.5% 2.349
Health Status
ER/Hospital/SNF in Last Year 234 38.0% 83 45.9% 3.557
# of Medications (0-27)*** 8.76 (4.30) 10.47 (4.66) -4.612
12+ Medications** 137 22.3% 62 34.3% 10.700
Falls in Past 3 Months*** 135 22.0% 62 34.3% 11.366
Dizziness* 168 27.3% 67 37.0% 6.323
Confusion 190 30.9% 64 35.4% 1.283
Lived Alone x Risk** 126 20.5% 56 30.9% 8.661
*p<.05; **p<.01; ***p<.001
Risk defined as presence of falls, dizziness, or confusion
Sample (N=615) Confirmed (N=181)

75
Prior to having potential medication problems confirmed through the
pharmacist review, 10 participants were terminated from MSSP (see
Appendix D and E). Known reasons for termination include death, entering
a long-term care facility, moving out of the service area, and loss of
MSSP/Medi-Cal eligibility. Compared to those who remained in the project,
those who terminated were more likely to have at least a high school
education (p=.042) and reported having a fall (p=.018). All terminated
participants had at least one potential medication problem (p=.001)
compared to 48% of the remaining sample, and cardiovascular medication
problems were the most prevalent type for those terminated (p<.001).
Differences between terminated participants and the sample on living alone
with a risk factor were marginally significant (p=.079).
Analyses of Potential Medication Problems
Table 3.7 reports Pearson product-moment correlation results
describing the relationship between potential medication problem
categories and the participant-level characteristics. Number of medications
and the top quintile of medication use (12 or more) had strong positive
correlations with several categories of potential medication problems
including having any problem, number of problems, having two or more
problems, unnecessary therapeutic duplication, and psychotropic
medication problems. Age was positively associated with potential
psychotropic medication problems related to confusion (r=.096; p=.017),

76
and NSAIDs use (r=.272; p<.001), however advanced age (80 and above)
was a possible criterion for NSAIDs problems. Age was negatively
associated with therapeutic duplication problems (r=-.084; p=.038).
Caucasian was positive associated with the number of medication
problems (r=.081; p=.045), having two or more medication problems
(r=.102; p=.011), and having a potential therapeutic duplication (r=.086;
p=.033). Being African-American was positively associated with potential
cardiovascular problems (r=.104; p=.009), whereas Latino was negatively
associated with this type (r=-.084; p=.037). Widows and married
participants were inversely associated with potential NSAIDs problems, with
widows being negatively associated (r=-.083; p=.007). Participants who
lived alone were positively associated with potential therapeutic duplication
(r=.100; p=.015) and cardiovascular problems (r=.084; p=.041). Newly
enrolled participants were positively associated with potentially
inappropriate psychotropic medication use related to falls (r=.099; p=.014)
and cardiovascular medication problems (r=.111; p=.006). Those with an
emergency room, hospital, or skilled nursing admission in the last year had
a positive correlation with potentially inappropriate psychotropic medication
use related to both confusion (r=.103; p=.010) and falls (r=.099; p=.014).
Living alone with falls, dizziness, or confusion was positively associated
with both problem types that were not derived using these risk factors:
therapeutic duplication (r=.106; p=.009) and NSAIDs problems (r=.082;

77
p=.044). Participants with therapeutic duplication problems were also
associated with potential psychotropic medication problems related to falls
(r=.087; p=.030) and NSAIDs problems (r=.135; p<.001).

Table 3.7: Pearson Product-Moment Correlation Results of Potential Medication Problems
Age -0.084 * 0.096 * 0.272 ***
Caucasian 0.081 * 0.102 * 0.086 *
African-American 0.104 **
Latino/a -0.084 *
English Speaking 0.082 *
Widow 0.110 **
Married -0.098 * -0.096 * -0.083 *
Lived Alone 0.100 * -0.114 ** 0.084 *
Newly Enrolled in MSSP 0.099 * 0.111 **
ER/Hospital/SNF in Last Year 0.103 ** 0.099 * -0.081 *
# of Medications 0.294 *** 0.361 *** 0.304 *** 0.394 *** 0.106 *** 0.138 ***
12+ Medications 0.230 *** 0.308 *** 0.251 *** 0.363 *** 0.118 ***
Lived Alone & Risk --- --- --- 0.106 ** --- --- --- 0.082 *
Therapeutic Duplication --- --- --- --- --- 0.087 * --- 0.135 ***
*p<.05; **p<.01; ***p<.001
Any = Any of the 4 Potential Medication Problems
# = Number of Potential Problems
2+ = 2 or More Potential Medication Problems
Ther. Dup. = Therapeutic Duplication
Psych-Conf. = Inappropriate Use of Psychotropic Medication with Presence of Confusion
Psych-Falls = Inappropriate Use of Psychotropic Medication with Fall in Last 3 Months
CV = Cardiovascular Medication Problems
NSAIDs = Inappropriate Use of Non-Steroidal Anti-Inflammatory Drugs
Risk defined as presence of falls, dizziness, or confusion
Psych-Conf Psych-Falls CV NSAIDs Any # 2+ Ther. Dup.

Given the strong correlations between potential medication problem
types and the number of different medications taken, Chi-square was used
to evaluate this relationship. Figure 3.2 shows that the likelihood of a
potential problem increased markedly with an increase in the number of
medications through the top quintile of medication use. The driving force
behind this relationship was multiple medication problems (Chi
2
=46.52;

78
p<.001), as well as therapeutic duplication (Chi
2
=98.19; p<.001) and
psychotropic medication problems related to falls (Chi
2
=10.44; p=.034).

Figure 3.2: Potential Medication Problems by
Number of Medications (N=615)
0%
10%
20%
30%
40%
50%
60%
70%
80%
1-3 4-6 7-9 10-11 12+
# of Medications
% of Sample
All
Problems***
2+
Problems***
Therapeutic
Duplication***
Psychotropic
w/ Falls*

To further understand the predictors of potential medication problems in
this sample, Table 3.8 reports adjusted odds ratios based on significant
socio-demographic and health characteristics identified in the bivariate
analyses. All models were significant. Adjusting for all other factors, the
number of medications was significantly associated with five of the six
problem categories with a 17% average increased likelihood of potential
problems with each medication increase. Age was also a significant factor
in three models: any potential problem, psychotropic medication problems
due to confusion, and NSAIDs problems. Each increasing year in age

79
resulted in a 3% increased chance of having any medication problems
(OR=1.027; p=.022) and a 6% increase in psychotropic medication
problems related to confusion (OR=1.058; p=.006). Age had a non-linear
relationship with NSAIDs problems showing a nearly 4.5 times increase in
prevalence rate with each increasing year (OR=4.475; p=.001). This trend
reversed for those at age 93 demonstrating a decreased propensity of
NSAIDs prevalence for those at older ages (OR=.992; p=.003). A cubed
form of age was tested to see if the curve would resume an upward trend
and was not shown to be significant.
African-American had decreased odds of having a therapeutic
duplication problem (OR=.588; p=.038) compared to Caucasian; no other
racial/ethnic differences appeared in the multivariate models. Those living
alone had a decreased risk of psychotropic medication problems related to
confusion (OR=.350; p=.006). Showing an opposite trend, those who had a
hospital, emergency room, or skilled nursing admission in the last year were
twice as likely to have a psychotropic medication problem related to
confusion (OR=1.956; p=.042) than their counterparts. Newly enrolled
participants also had twice the odds of both psychotropic medication
problems related to falls (OR=2.142; p=.014) and cardiovascular medication
problems (OR=2.043; p=.005), and greater odds of any medication problem
overall (OR=1.667; p=.006).

80
Several interaction terms were tested on relevant models to further
delineate predictors for potential medication problems: Age x Living Alone;
Age x ER/Hospital/SNF; Age x Number of Medications; Living Alone x
Number of Medications; Living Alone x ER/Hospital/SNF; and
ER/Hospital/SNF x Number of Medications. None of these terms were
significant in their respective models.

Table 3.8: Odds Ratios of Potential Medication Problems (N=615)
Age 1.027 * 1.01 0.995 1.058 ** 0.986 0.98 4.475 **
Age
2
0.992 **
Race/Ethnicity
African-American 0.786 0.588 0.558 * 0.749 0.576 1.220 0.897
Latino/a 0.898 0.536 0.665 0.729 1.148 0.607 0.951
Asian/PI 0.801 1.018 1.161 1.399 0.898 0.628 0.492
Other/Unknown 1.500 0.634 0.933 1.768 0.589 0.359 1.175
English Speaking 1.170 1.158 1.140 0.828 1.906 1.138 0.789
Lived Alone 1.031 1.184 1.249 0.350 ** 0.870 1.432 1.414
Newly Enrolled in MSSP 1.667 ** 1.435 1.092 1.844 2.142 * 2.043 ** 0.918
Hospital, ER, or SNF in Last Year 1.038 1.235 1.079 1.956 * 1.595 1.114 0.599
# of Medications 1.182 *** 1.200 *** 1.265 *** 1.144 *** 1.111 ** 1.015 1.046
Chi
2
69.16 *** 65.50 *** 104.69 *** 36.59 *** 25.30 ** 24.52 ** 69.94 ***
DF 10 10 10 10 10 10 11
*p<.05; **p<.01; ***p<.001
# = Caucasian as Referent Group
Any = Any of the 4 Confirmed Medication Problems
2+ = 2 or More Confirmed Medication Problems
Ther. Dup. = Therapeutic Duplication
Psych-Conf. = Inappropriate Use of Psychotropic Medication with Presence of Confusion
Psych-Falls = Inappropriate Use of Psychotropic Medication with Fall in Last 3 Months
CV = Cardiovascular Medication Problems
NSAIDs = Inappropriate Use of Non-Steroidal Anti-Inflammatory Drugs
Any Ther. Dup. 2+ Psych-Conf. Psych-Falls NSAIDs CV

Confirmed Medication Problems
Bivariate and multivariate analyses were also completed for those
whose medication problem was confirmed by the pharmacist in the various

81
categories. Table 3.9 reports Pearson product-moment correlation results
related to socio-demographic and health characteristics in this group.
Similar to those with a potential problem, factors associated with confirmed
medication problems included age, race/ethnicity, marital status, living
alone, emergency room, hospital, or skilled nursing facility use in the last
year, number of medications, and presence of a therapeutic duplication
problem. Age was negatively associated with confirmed therapeutic
duplication problems (r=-.084; p=.037) and positively associated with
NSAIDs (r=.193; p<.001). Caucasian was associated with therapeutic
duplication problems (r=.092; p=.023), and African-American continued to
be associated with cardiovascular problems when confirmed (r=.088;
p=.029). Married participants had negative associations with any confirmed
problem (r=-.082; p=.042), the number of problems (r=-.112; p=.005), two or
more problems (r=-.117; p=.004), psychotropic medication problems related
to falls (r=-.099; p=.014), and NSAIDs problems (r=-.096; p=.017).
Congruently, widows had a positive association with NSAIDs problems
(r=.116; p=.003), and those living alone were associated with more
medication problems (r=.092; p=.024) and therapeutic duplication problems
(r=.128; p=.002). Newly enrolled participants were positively associated
with any medication problem (r=.084; p=.036), psychotropic medications
related to confusion (r=.079; p=.049), and cardiovascular medication
problems (r=.086; p=.033). Those with an emergency room, hospital, or

82
skilled nursing facility admission were positively associated with four types
of confirmed medication problems: any problem (r=.104; p=.009), the
number of problems (r=.093; p=.020), and psychotropic medication
problems related to confusion (r=.112; p=.006) and falls (r=.106; p=.009).
Those living alone with a risk factor was only significant for therapeutic
duplication problems at the confirmation phase (r=.126; p=.002).
Therapeutic duplication problems continued its positive association with
both psychotropic medication problems related to falls (r=.101; p=.012) and
NSAIDs (r=.191; p<.001) when confirmed.

Table 3.9: Pearson Product-Moment Correlation Results of Confirmed Medication Problems
Age -0.084 * 0.193 ***
Caucasian 0.092 *
African-American 0.088 *
Widow 0.116 **
Married -0.082 * -0.112 ** -0.117 ** -0.099 * -0.096 *
Lived Alone 0.092 * 0.128 **
Newly Enrolled in MSSP 0.084 * 0.079 * 0.086 *
ER/Hospital/SNF in Last Year 0.104 ** 0.093 * 0.112 ** 0.106 **
# of Medications 0.257 *** 0.322 *** 0.298 *** 0.337 *** 0.105 ** 0.141 ***
12+ Medications 0.186 *** 0.259 *** 0.240 *** 0.276 *** 0.128 ***
Lived Alone & Risk --- --- --- 0.126 ** --- --- ---
Therapeutic Duplication 0.101 * 0.191 ***
*p<.05; **p<.01; ***p<.001
Any = Any of the 4 Confirmed Medication Problems
# = Number of Confirmed Problems
2+ = 2 or More Confirmed Medication Problems
Ther. Dup. = Therapeutic Duplication
Psych-Conf. = Inappropriate Use of Psychotropic Medication with Presence of Confusion
Psych-Falls = Inappropriate Use of Psychotropic Medication with Fall in Last 3 Months
CV = Cardiovascular Medication Problems
NSAIDs = Inappropriate Use of Non-Steroidal Anti-Inflammatory Drugs
Risk defined as presence of falls, dizziness, or confusion
Psych-Conf Psych-Falls CV NSAIDs Any # 2+ Ther. Dup.

83
Medication use continued to be a strong factor related to a confirmed
medication problem as both the number of medications and those having
12 or more medications were significant for several categories. One
difference was that psychotropic medications problems related to confusion
were associated with the number of medications (p=.009), but not for those
with 12 or more medications. Nonetheless, Figure 3.3 reports that as the
number of medications increase, the likelihood of having any confirmed
problem (Chi
2
=32.96; p<.001), two or more problems (Chi
2
=45.77; p<.001),
therapeutic duplication (Chi
2
=61.60; p<.001), and psychotropic medication
problems related to falls (Chi
2
=11.74; p=.019) also increase.

Figure 3.3: Confirmed Medication Problems by
Number of Medications (N=615)
0%
10%
20%
30%
40%
50%
60%
70%
80%
1-3 4-6 7-9 10-11 12+
# of Medications
% of Sample
All
Problems***
2+
Problems***
Therapeutic
Duplication***
Psychotropic
w/ Falls*

84
Table 3.10 reports adjusted odds ratios for those with confirmed
medication problems. All showed model significance except for
cardiovascular medication problems, which had no significant predictors
(model not shown). Number of medications continued to be a powerful
factor related to confirmed medication problems, as it was significant in
almost all models. The greatest odds were for those with a confirmed
therapeutic duplication problems, which increased 23% for each increasing
number of medications (OR=1.232; p<.001). A non-linear form of age
remained a significant factor for those with a confirmed NSAIDs problem,
reporting over twice the odds with each increasing year (OR=2.578;
p=.026); until 95 years of age where the curve trended downward
(OR=.995; p=.045). Age was also a significant factor for psychotropic
medication problems with confusion (OR=1.051; p=.033), increasing 5%
with each year.
African-Americans were the only significant racial/ethnic group
compared to Caucasians, with this group being nearly half as likely to have
a confirmed therapeutic duplication problem (OR=.501; p=.033). English
speakers were nearly three times as likely to have a psychotropic
medication problem related to falls than their counterparts (OR=3.044;
p=.026). Those living alone had lower odds of having a psychotropic
medication problem related to confusion than their counterparts (OR=.447;
p=.049). Newly enrolled participants had greater odds of any confirmed

85
medication problem (OR; 1.581; p=.022) and specifically psychotropic
medication problems related to falls (OR=2.077; p=.039). Participants who
had an emergency room, hospital, or skilled nursing facility admission in the
last year had more than twice the odds of having a psychotropic medication
problem related to confusion (OR=2.468; p=.016).

Table 3.10: Odds Ratios of Confirmed Medication Problems (N=615)
Age
1.020 0.996 0.992 1.051 * 0.976 2.578 *
Age
2
0.995 *
Race/Ethnicity#
African-American
0.761 0.659 0.501 * 0.767 0.572 0.996
Latino/a
1.153 0.599 0.809 0.631 1.875 1.037
Asian/PI
1.242 0.858 0.978 1.173 2.018 0.562
Other/Unknown
0.607 0.442 0.299 1.553 1.091 0.395
English Speaking
1.498 1.404 1.322 0.843 3.044 * 1.163
Lived Alone
1.207 1.152 1.658 0.447 * 1.302 1.110
Newly Enrolled in MSSP
1.581 * 1.283 1.067 1.911 2.077 * 1.095
ER/Hospital/SNF in Last Year
1.466 1.301 1.373 2.468 * 1.915 0.778
# of Medications
1.155 ** 1.215 *** 1.232 *** 1.143 ** 1.210 ** 1.069
Chi
2
58.39 *** 55.16 *** 77.36 *** 28.94 ** 27.74 ** 35.54 ***
DF 10 10 10 10 10 11
*p<.05; **p<.01; ***p<.001
# = Caucasian as Referent Group
Any = Any of the 4 Confirmed Medication Problems
2+ = 2 or More Confirmed Medication Problems
Ther. Dup. = Therapeutic Duplication
Psych-Conf. = Inappropriate Use of Psychotropic Medication with Presence of Confusion
Psych-Falls = Inappropriate Use of Psychotropic Medication with Fall in Last 3 Months
CV = Cardiovascular Medication Problems
NSAIDs = Inappropriate Use of Non-Steroidal Anti-Inflammatory Drugs
Psych-Falls NSAIDs Any 2+ Ther. Dup. Psych-Conf.

D. Discussion
This study identified medication problems for community-dwelling,
dually-eligible older adults enrolled in a Medi-Cal waiver case management

86
program that, if not addressed, could contribute to an adverse drug event.
The two-step review process identified the potential prevalence of
medication problems based on a focused drug regimen review protocol
(Home Health Criteria); the second step included a pharmacist clinical
review to confirm the suspected problem. Based on this process, nearly
49% of the sample had at least one potential medication problem. These
data contrast with the rates found in a home healthcare sample using the
same drug regimen review protocols where 19% were found to have a
potential problem (Meredith et al., 2001). Each of the four medication
problem types in the MSSP sample had at least double the potential
problem rate compared to the home healthcare sample, with unnecessary
therapeutic duplication showing the greatest difference at a six times higher
prevalence rate (Meredith et al., 2001).
The difference between the MSSP and home healthcare samples on
therapeutic duplication was even more pronounced given that the review
process for multiple pain medications in MSSP was measured more
conservatively than in the home healthcare study. In the home healthcare
sample, when participants were prescribed a routine pain medication
coupled with a PRN pain medication, they were screened in as having a
therapeutic duplication problem. However, the consultant pharmacists in
MSSP reviewed this procedure with an expert advisory panel and

87
concluded this prescribing pattern was appropriate for particular clinical
conditions.
Unlike the home healthcare sample, nearly 20% of those in MSSP had
two or more potential problems. For approximately 75% of this group, the
main problems were therapeutic duplication problems paired with either
inappropriate use of psychotropic medications or NSAIDs. These results
speak to the complexity of medication review for older adults with chronic
care needs prescribed multiple medications (Rochon & Gurwitz, 1999).
Clinical review from a consultant pharmacist removed participants with
false positive and/or unconfirmed readings from the potential medication
problem subset, resulting in 29% of the sample having a confirmed problem
and 10% with two or more confirmed problems. Inappropriate use of
psychotropic medications confirmed at the highest rate (74%) once
identified through the Home Health Criteria protocols. Cardiovascular
medication problems had the lowest confirmation rate at 31%. This low rate
was likely due to a systemic factor rather than a clinical one. Unfortunately,
the vast majority of potential cardiovascular medication problems could not
be confirmed due to the absence of additional blood pressure and pulse
readings needed to verify the problem. This challenge is discussed below
in the limitations section.
Similar to other research findings (Hanlon et al., 2004; Meredith et al.,
2001; Perri et al., 2005; Piecoro et al., 2000; Sorensen et al., 2005), those

88
with higher numbers of medications were at greater risk of having a
medication problem. More specifically, those taking higher numbers of
medications had greater odds of having problems with unnecessary
therapeutic duplication, inappropriate psychotropic drug use for both
confusion and falls, and inappropriate NSAIDs use when controlling for
socio-demographic and health characteristics. Although number of
medications was a powerful factor, clinical indicators of risk were also
critical in screening for medication problems. Twenty-three participants
who were prescribed four medications or less screened in with a potential
problem, and eleven of those had a confirmed problem. By including the
five key indicators of risk when screening for medication problems, clinical
staff were able to gather a deeper understanding of the participant’s
functioning in context of their medication regimen and their potential for an
adverse drug event. This is especially important for those taking
medications that may not considered problematic for all older adults, which
is often the basis for medication review protocols.
Multivariate analyses also identified participant-level characteristics
associated with medication problems that were in agreement with previous
studies including advanced age and Caucasian (Goulding, 2004; Piecoro et
al., 2000). However, due the specificity of the medication problems and
additional variables available in this data several other variables were found
to be significant predictors of having a medication problems such as

89
English-speaking, being newly enrolled in MSSP, and institutional-level
health service utilization. Adjusting for other factors, English speakers had
greater odds of psychotropic medication problems related to falls. This
result might be due to increased access to healthcare services compared to
non-English speakers, leading to polypharmacy and its negative outcomes.
Newly enrolled participants also had a greater risk for this problem type and
any of the four problems, suggesting that those entering MSSP have an
urgent need for medication review upon initial assessment.
The strong relationship between high-level health service utilization and
inappropriate psychotropic medication use is intriguing. Unfortunately due
to limitations in the data, it is unclear if this problem type contributed to the
utilization of services or if older adults with a mental health condition used
services more than their counterparts. Previous research has reported high
health service utilization among older adults with depression (Katon, Lin,
Russo, & Unutzer, 2003; Luber et al., 2001). Therefore it is plausible that
older adults taking psychotropic medications due to mental health
conditions might utilize health services more independent of a medication-
related problem. Future research is needed to clarify this relationship.
Several reasons are plausible for the higher prevalence of medication
problems in the MSSP sample compared to the home healthcare study.
First, eligibility requirements for the MSSP as a Medi-Cal waiver program
require that participants must be dually-eligible and functionally impaired

90
upon admission. Compared to Medicare-only beneficiaries, dually eligible
older adults are more likely to be older, female, living alone, have more
functional impairments, and take more medication (Moon & Shin, 2006;
Ryan & Super, 2003). These findings mirror the differences between the
MSSP and home healthcare samples on gender (80% female in MSSP vs.
67% in home healthcare) and medication utilization (median values: 8 in
MSSP; 5 in home healthcare), although the average age for both samples
was 80 years old (Meredith et al., 2001). Medicaid eligibility in the home
healthcare sample was not reported; however, it is likely that the majority of
patients were Medicare-only beneficiaries.
Patients screened in the home healthcare sample likely received some
form of acute or rehabilitative care treatment prior to entering the long-term
care system, a general requirement for accessing home healthcare
services. Recent engagement with a health care provider does offer the
possibility for medication review and amelioration of potential problems.
This is not necessarily the case for the MSSP sample, as professionals or
informal care systems (e.g., family) could refer participants to the care
management services without the consent of a health care provider. Even
with the high rate of emergency room, hospital, or skilled nursing among the
sample, 40% of these participants had been enrolled in MSSP for more
than a year and may not have had an opportunity for a comprehensive
medication review.

91
Fragmentation in the health care service delivery system may also
contribute to high problem prevalence rates. Dually-eligible older adults
with multiple impairments likely visit multiple physicians for treatment and
symptom management. Structural barriers in the health care delivery can
impede care coordination among physicians, leaving the medication
management responsibility to the patient. Responsibilities include reporting
all current medications to all physicians, working with a single pharmacy for
dispensing, and communicating all negative symptoms to all physicians
after a medication change. These may be reasonable expectations for
healthier individuals to manage, but can be quite challenging for older
adults with multiple functional and cognitive impairments.
Over 20% of the total sample lived alone and had one of the evidence-
based clinical indicators of risk (falls, dizziness, or confusion) that were
used to help identify medication problems. After the medication problem
confirmation phase, 31% of the sample lived alone with a risk factor.
Although these results were not statistically useful (risk factors were
obviously highly collinear with the two of the four problem types), having
one-third of MSSP participants with confirmed medication problems living
alone and experiencing a potentially dangerous risk factor presents clinical
challenges for both care managers and the consultant pharmacist.
Intervention strategies must attempt to resolve the medication problem and

92
maximize engagement of available support networks to minimize longer-
term risk of negative health outcomes for this group.
Policy Implications
Findings from this prevalence study suggest that dually-eligible older
adults who are at risk for institutionalization have exceedingly high rates of
potential and confirmed medication problems. This study concurs with
results from Rigler and colleagues (2005) that also found potential problems
rates at 48% using a different screening instrument. This study shows that
the dually-eligible waiver population, although similar to nursing home
residents in their functional levels, may be at greater risk for adverse drug
events due to potentially inappropriate prescribing and not having access to
routine medication review similar to the nursing home population (Golden et
al., 1999; Rigler et al., 2005).
Medication data used for this study were collected prior to the
implementation of the Medicare Part D prescription drug program, meaning
that the state of California paid for medications for the MSSP participants.
Although transfer of responsibility for prescription payment from the states
to the federally-contracted prescription drug plans has been slow and
troublesome (Smith, Gifford, Kramer, & Elam, 2006), the high prevalence of
medication problems and its relationship to high medication usage are
important findings for all entities involved with cost and quality of care
issues regarding medications for dual eligibles.

93
Medicare Part D is filled with both challenges and opportunities related
to medication management. First, a potential pitfall of the new benefit is the
likelihood of increased prescribing behavior due to greater access to
prescription drug coverage among older adults. In 2005, all states provided
some prescription drug coverage to dual eligibles (Centers for Medicare
and Medicaid Services, 2005), however this coverage had considerable
variability with California being more generous than others states (Safran et
al., 2005). As older adults, and dual eligibles in particular, have greater
access to prescription drugs, it is likely that medication problems and
adverse drug events will also increase.
On the other hand, the Medicare Prescription Drug Improvement and
Modernization Act of 2003 (MMA, 2003), which authorized Medicare Part D,
acknowledged the need for targeted medication review among activity for
“high-risk” groups. Participants of community-based waiver programs could
be defined as a priority population for this service due to their implied risk of
higher level of placement and increased chances of transitions across the
continuum of health care (Coleman, 2003). If funding or reimbursement
among the prescription drug plans for this service was available, it could
create an opportunity to use consultant pharmacists trained in geriatric
pharmacology to engage in medication review for this population. On a
state level, the current work on Olmstead plans to improve the cost and
quality of services that support functionally-impaired older adults in

94
community settings might create an opportunity to implement and finance
focused medication review for those at risk for institutionalization.
Limitations
A few limitations were present in this first phase of the CBM Intervention
project. First, data needed to determine prevalence of medication problems
were collected by different professional types. Social workers, the bulk of
the MSSP workforce, collected information on confusion and falls as part of
their standard assessment process. Nurses collected information on blood
pressure, pulse, and dizziness. Although each participant had a primary
case manager (either a social worker or nurse), s/he received a visit from
the alternate discipline at enrollment and on a yearly basis. Due to scope of
practice issues, case managers who gather screening data from
participants during a quarterly visit only gathered information within their
domain. Therefore, social work staff by and large did not collect blood
pressure and pulse readings, which negatively impacted the ability to
confirm cardiovascular medication issues.
A corollary concern was that each site had differing operational
guidelines regarding nurse or social workers accepting new participants on
their caseloads. As Chapter II described, Site #1 assigned new cases
randomly to care managers whereas nurses at Site #2 functioned in a
consultant capacity and managed a small caseload. Although similar to
Site #1, Site #3 only engaged participants who had a nursing assessment in

95
the CBM Intervention in order to gather the needed screening data. Given
this variability, care manager type was not included in the analyses.
Instead prevalence variation was evaluated based on site differences in
Table 3.3, and no differences were present across sites. To further ensure
that care manager type was not a confounding variable, a Chi-square
analysis was completed using care manager type and the prevalence of
medication problems for Site #1 given that is used random assignment at
enrollment; there were no differences for any of the medication problem
types.
Another limitation was related to data collection on some variables and
the lack of other variables. Measures such as level of education and the
clinical indicators or risk had great variability across the sites that likely
resulted from different data collection procedures. For example, level of
education was not regularly captured at intake by any site until 2004.
Although all participants in the sample were dually eligible for Medicare and
Medi-Cal, participant education level could have been useful in multivariate
analyses to assess risk for medication problems. This study did not collect
data on specific medical conditions, the number of co-morbidities, or the
number of prescribing physicians. Although participants were functionally
impaired upon admission based on programmatic guidelines, the type and
number of medical conditions would have useful to understand the
relationship between intensity of chronic conditions and medication

96
problems. The number of prescribing physicians would also have been
useful assess how prescribing patterns from multiple physicians impacts
medication problems. Both of these variables are recommended for
inclusion in future studies.
The inability to use an automated screening system to complete the
drug regimen review further complicated the study. A computerized
screening tool was in development during this project, but was not
operational during the data collection phase described in this article. This
tool was intended to merge with MSSP software used for administrative and
case management data collection. The lead agency for the project (MSSP
A) is currently working with computer vendor to resolve interface problems
identified. It is anticipated that when the technology is operating
successfully, it will streamline the screening phase of the medication
management process. Case management staff will be able identify a
participant who screens in with a potential problem based on computerized
alerts, and can then make an appropriate referral to a consultant
pharmacist for a more in-depth clinical review.
Generalizability
The data from this study were collected from three urban, ethnically-
diverse sites of the California MSSP program, a Medi-Cal waiver program
providing case management services to dual eligibles with functional
impairments. Even with this diversity, the prevalence of medication

97
problems did not differ across sites. Generalizability to other waiver
programs presents a greater challenge as criteria vary by program and
state. However, all 1915(c) waiver programs operate with the same goal of
helping people with significant health conditions and impairments avoid
higher, more costly levels of care. The MSSP sample was may be similar
to older adults in other state waiver programs, but are not likely reflective of
all dually eligible or functionally impaired older adults who are not Medicaid
beneficiaries.
Conclusion
The magnitude of these findings based on a targeted drug regimen
review highlights the need for a comparably targeted, evidence-based
intervention to reduce identified medication problems in a MSSP population.
The role of the consultant pharmacist was critical in separating out those
who screened in based on the drug regimen review from those who had a
confirmed problem. Given the organization structure and staffing of the
Medi-Cal waiver community-based case management program, a
multidisciplinary team approach with the continued involvement of the
consultant pharmacist appears to be a reasonable method for addressing
medication problems (Gurwitz, 2004; Hanlon et al., 2001; Sloane et al.,
2002). This idea was tested in the second phase of the CBM Intervention
project with results reported in Chapter IV.

98
IV: IMPROVING MEDICATION USE IN DUALLY-ELIGIBLE OLDER
ADULTS
A. Introduction
The high frequency of medication problems for older adults supported
by prevalence findings in Chapter III underscores the need for efficacious
medication management interventions. Several public and private health
care organizations and initiatives have also called for improved strategies to
address this critical public health problem (Clancy, 2004; Clancy et al.,
2004; Commonwealth Fund, 2004; Department of Health and Human
Services, 2000; Institute of Medicine, 2001; Quality Interagency
Coordination Task Force, 2000). The Healthy People 2010 Initiative
(Department of Health and Human Services, 2000) summarized the need
for increased medication review by health care professions for patients
aged 65 years and older and those with chronic illnesses or disabilities
(Objective 17.3). In addition, the development of targeted strategies to
address medication problems are timely given the passage of the Medicare
Prescription Drug, Improvement, and Modernization Act of 2003 (MMA,
2003). Section 107 of MMA 2003 mandated the Institute of Medicine (IOM)
to “carry out a comprehensive study of drug safety and quality issues in
order to provide a blueprint for system-wide change." A forthcoming IOM
publication completed this work and outlined specific policy and
implementation goals for individuals, health systems, and governmental

99
actors regarding improving patient safety associated with medication errors
(Institute of Medicine, 2007). These goals included improved collaboration
between patients and providers, greater use of information technologies,
improved medication labels and packaging, improved training, and
government funding targeted toward research on preventing medication
errors.
Responding for the private sector, the Joint Commission on
Accreditation of Healthcare Organizations (JCAHO) developed 14 National
Patient Safety Goals to improve the delivery of health care services for all
accredited facilities (JCAHO, 2006a). Goal #8-2005 states that all JCAHO-
accredited facilities should “accurately and completely reconcile
medications across the continuum of care…[which] applies to all care
settings--including ambulatory, emergency, and urgent care, long-term
care, and home care--as well as inpatient services”(JCAHO, 2006b, p. 3).
Given that medication-related problems contribute to decreased functioning
and additional care transitions to higher-level services (Coleman, 2003),
programs across the continuum of care that offer medication management
interventions have great potential to minimize negative health care
outcomes for older adults.
A number of medication management intervention studies have shown
positive results to reduce suboptimal prescribing. Interventions using a
multidisciplinary team in outpatient geriatric clinics reduced suboptimal

100
prescribing over a 12-month period (Schmader et al., 2004), and success
rates in primary care have ranged from 19% to 70% (Gilbert, Roughead,
Beilby, Mott, & Barratt, 2002; Hanlon et al., 1996; Krska et al., 2001;
Rhoads & Thai, 2003; Sellors et al., 2003; Sorensen et al., 2004).
Medication management interventions generally consist of a pharmacist
completing a medication review based on explicit criteria and then providing
recommendations to the prescribing physician or treatment team on how
best to ameliorate the identified problem. The most successful
interventions have involved direct contact between the prescribing
physician and the pharmacist with written recommendations for change
(Gilbert et al., 2002; Krska et al., 2001; Sellors et al., 2003; Sorensen et al.,
2004).
Although multiple studies describe high prevalence of medication
problems in a large variety of contexts, reports of successful interventions
are not as widespread. The development of solutions that reduce
medication problems require costly intervention studies and have generally
been context-specific. Even if an intervention study reports efficacious
results, it is unclear whether the intervention can translate to other service
delivery settings and achieve similar results. Clearly, efficacious solutions
based on a hospital setting with a built-in multidisciplinary team, a hospital-
based pharmacist, and a controlled patient environment would not likely
work for community-based models of care with less access to these

101
professional resources and less control over the participant’s environment.
Additionally, intervention studies to date have generally been implemented
and evaluated in health care settings that use a medical model of care.
One project that has attempted to translate an efficacious medication
management intervention beyond a medical setting to a Medi-Cal waiver
care management program is the CBM Intervention. In the initial phase
researchers sought to identify the prevalence of medication problems in
older adults participating in the MSSP program (see Chapter III). The
purpose of the next phase was to implement a medication management
intervention for those with a confirmed medication problem and evaluate its
effectiveness to reduce suboptimal prescribing. This chapter briefly
explains the evidence base used by the CBM Intervention, describes the
process employed in the care management programs, and reports findings
on resolution of identified medication problems.
Evidence Base for Medication Management
The intervention phase was modeled after the process developed by
Meredith and colleagues (2002) that tested the efficacy of a pharmacist-
centered intervention in home healthcare agencies to reduce medication
problems used a randomized clinical trial design. The intervention in the
home healthcare agencies used nurses collaborating with consulting
pharmacists to develop recommendations for medication change based on
the four problem types assessed from the Home Health Criteria screening

102
algorithm (Meredith et al., 2002). The nurse contacted the prescribing
physician about the problem and provided the pharmacist’s
recommendations through a structured template; in more complex cases,
the pharmacist contacted the physician directly. The nurse worked with the
patient to implement the change and monitored effects. Medication
problems were resolved for 50% of intervention group, compared to 38% in
the control group (p<.05) through this approach (Meredith et al., 2002).
Two problems with most successful resolution were therapeutic duplication
(47% reduction; p=.003) and cardiovascular medications (37%
improvement; p=.020).
CBM Intervention Process
The CBM Intervention incorporated key elements from the home
healthcare trial into a Medi-Cal waiver care management program including
the application of specific goals for problem resolution, a consultant
pharmacist to provide recommendations for resolution, and direct
communication with physicians and/or participants regarding the
recommendations. Each of the four medication problem types had a
defined goal for successful problem resolution (Meredith et al., 2002). For
therapeutic duplication, the goal was to stop the use of the medication that
was not clinically warranted. For psychotropic medication problems, the
goal was to reduce exposure to the drug to decrease potential psychomotor
effects while continuing to manage the condition for which the drug was

103
prescribed. For cardiovascular medication problems, the goal was to
modify the medication regimen according to national guidelines to promote
cardiovascular health (Chobanian et al., 2003). For NSAID problems, the
goal was to reduce exposure of peptic ulcer complaints while maintaining
adequate pain management.
Participants identified through the initial screening process as having a
confirmed medication problem (see Chapter III) entered the intervention
process. A consultant pharmacist, with support from the primary care
manager, developed an individualized treatment plan to address the
specific medication problem guided by the goal for each problem type and
created a structured written template of suggested medication changes. To
implement these changes, the pharmacist communicated with two
audiences: the prescribing physician and the participant and/or caregiver. If
the intervention was focused toward the prescribing physician,
recommendations for a medication change were faxed to the physician’s
office with a follow-up phone call. If the focus was at the participant level,
the pharmacist discussed the suggested changes with the care manager
who communicated with the participant and/or caregivers; in some complex
cases, the pharmacist contacted the participant and/or caregivers directly.
At a regularly scheduled 3-month follow-up visit, the care manager asked
the participant and/or caregiver about changes made to the medication
regimen based on the problems identified. Unlike other studies that

104
measured the physicians’ or participants’ verbal acceptance and partial
acceptance of the pharmacist recommendations (Krska et al., 2001;
Rhoads & Thai, 2003; Triller, Clause, Briceland, & Hamilton, 2003), here
the focus was on change of the medication regimen.
B. Methods
Data Source
The intervention phase of the CBM Intervention project included panel
data gathered from participants enrolled in three Los Angeles County-based
MSSP care management sites from June 2004 to January 2006.
Participating sites were affiliated with two MSSP providers that served
racial/ethnically diverse older adults in half of the county’s service and
planning areas. Site #1 and #2 are affiliated with a community-based social
service agency with a strong research core. Site #3 is housed in the care
management program under the auspice of a local hospital. The evaluation
was conducted by a team of researchers at the University of Southern
California; protocols and instruments were approved by the institutional
review boards for all program partners. Administrative and care
management data were extracted from the three MSSP sites with blinded,
encrypted data transmitted to the research partners for evaluation of the
project.

105
Sample
Participants were older adults enrolled in one of three MSSP sites in Los
Angeles County. MSSP is a slot-based, 1915(c) home- and community-
based waiver program that provides care management and purchase of
services to disabled, racially and ethnically diverse, low-income older
adults. The overall goal of MSSP is to decrease and/or avoid inappropriate
institutionalization for this population (L. S. Miller et al., 1985). MSSP
participants are community-dwelling adults aged 65 and over who are Medi-
Cal eligible, and demonstrate significant functional impairment as
evidenced by 1) two or more activities of daily living impairments, or 2) at
least one of activities of daily living deficiency and cognitive impairment.
Based on the prevalence data of confirmed medication problems
reported in Chapter III, 181 participants were eligible for the intervention
phase. Nineteen participants terminated from the MSSP program prior to
initiating the intervention process, which left an analytic sample of 162
participants. The only significant difference between those who terminated
from MSSP prior to completing the intervention phase and those who
remained in the sample was on the number of medications. Participants
who remained had a higher average number of medications (11 vs. 8), and
37% of the sample had 12 or more medications compared to 10% for those
who terminated. Over one-third of those terminated from MSSP had died;
37% were terminated due to long-term institutional placement. The

106
remainder left MSSP due to moving, change in Medi-Cal status, or change
in desire or ability to participate in services (see Appendix F and G for
detailed results).
Measures
Dependent Variables
Dependent variables measured the change of four identified medication
problem types as defined by the Home Health Criteria (Meredith et al.,
2001): unnecessary therapeutic duplication; inappropriate use of
psychotropic medication; cardiovascular medication problems; and
inappropriate use of non-steroidal anti-inflammatory drugs (NSAIDs).
Medication change was measured as a dichotomous variable (change=1).
It was recorded by the primary care manager at a regularly scheduled home
visit based on participant and/or caregiver report of a change. Prior to the
home visit, pharmacists provided care managers with specific information of
the participant’s identified medication problem and what a medication
change would actually entail given the recommendations offered (see
Appendix H).
Independent Variables
Independent variables included socio-demographic and health
characteristics and clinical health indicators specific to this study to better
understand the relationship between individual factors and medication
change. Age was a continuous variable starting at 65 years to meet MSSP

107
eligibility requirements. Gender was coded as female. Race/ethnicity was
a categorical variable that represented Caucasian, African-American,
Latino/a, Asian/Pacific Islander, Native American, and Unknown; Native
American and Unknown were combined into “Other/Unknown” due to small
cell sizes for each. Caucasian served as the reference group in multivariate
analyses. Primary language spoken included English, Spanish, Armenian,
and Chinese as those with the highest prevalence. Other known languages
included Arabic, Farsi, French, Korean, and Tagalog. A dichotomous
variable was created to represent primary English speakers compared to all
others. Marital status included married, widowed, divorced/separated,
single, and unknown/refused. Education was initially recorded as six
categories: no formal education, grade school, high school diploma or
equivalent, some college, completed college, and unknown/refused.
However, data on education were missing for 34% of the sample because it
was not consistently collected by the sites at intake until 2004. Therefore,
education was recoded for those known to have a high school education or
above. Living situation was a dichotomous variable defined as those who
lived alone (1) compared to those who lived with someone (0). Participants
newly enrolled in MSSP were defined as those who received less than
three months of services before engaging in the CBM Intervention (1) as
compared to all others (0). This variable was included to delineate those
new to care management services who might have had more acute care

108
needs than those who had received services beyond the initial assessment
period.
Health characteristics focused on health status and service utilization.
Number of prescribed medications was a discrete count variable that
included prescribed medication, relevant over-the-counter medication (e.g.,
Diphenhydramine), and a single count for any vitamins if the participant
took them regularly. Although not part of the original study, the addition of
vitamins in a medication count has been well documented in the literature
(Kaufman et al., 2002; Sloane et al., 2002; Sorensen et al., 2005).
Admission to the emergency room, hospital, or skilled nursing facility within
the last year was recorded as a dichotomous variable. Evidence-based
clinical indicators to determine potential medication problems included falls,
confusion, dizziness, blood pressure, and pulse. Falls were measured by
participant self-report of an incident in the last three months, regardless of
whether an injury occurred as a result. Although definitions vary for what
constitutes a “fall” (Zecevic et al., 2006), this measure accounted for the
participant’s perception of a fall, which can affect both falls efficacy and fear
of falling (Tinetti & Williams, 1998). Presence of confusion was measured
by the case manager’s assessment of participant functioning using
guidelines of the Confusion Assessment Method (Inouye et al., 1990).
Dizziness was also a dichotomous measure based on participant self-
report. To evaluate those at higher risk of a negative health outcome, a

109
dichotomous variable was created that incorporated those who lived alone
with having a fall, dizziness, or confusion (Lived Alone x Risk). Blood
pressure and pulse, measured by the nurse care managers during the
health assessment, were only relevant to determining specific
cardiovascular medication problems and therefore were not included data
reporting.
Analysis
Data analysis was completed using SPSS software, Version 11.
Descriptive and bivariate statistics reported the rate of medication change
for all problems and for each individual problem type. For further specificity,
psychotropic medications were delineated by the clinical criteria for
inappropriate use to due a fall or presence of confusion. Logistic regression
examined odds ratios of socio-demographic and health characteristics on
medication change given that little is know about how these characteristics
affect medication improvement.
C. Results
Figure 4.1 reports the flow for the entire MSSP sample (N=615) through
the Prevalence and Intervention processes. Two hundred ninety-nine
participants (49%) had at least one potential problem identified. Upon
clinical review by the consultant pharmacist, 181 participants (29%) had a
potential medication problem confirmed. Of the 118 people who did not
confirmed as having a medication problem, 83 were due to false positive

110
indications, 25 were unconfirmed problems due to lack of verifying
information, and 10 participants terminated from MSSP prior to the clinical
review. This article reports on the second phase of the project, which was
to complete the intervention process with those who had a confirmed
problem and evaluate the results of the intervention. Nineteen participants
left the MSSP program prior to initiating the intervention process, leaving a
sample of 162 participants. At the three month follow-up visit, 99
participants (61% of the intervention group) had a successful medication
change.

111

112
Table 4.1 reports the characteristics of the MSSP sample who
participated in the intervention process. Participants on average were 81
years old with 81% being female. Nearly 40% of the sample was African-
American with the remainder of the sample comprised mostly of Caucasian
(27%) and Latinos (25%). Thirty-nine percent of the sample was non-
English speaking as their primary language of choice, with the largest
percentage speaking Spanish (21%). Fifty-seven percent were widowed,
15% were married, and 18% were divorced or separated from a partner.
Approximately 40% reported having at least a high school education, 48%
reported living alone, and 42% were new MSSP enrollees.
Forty-eight percent of participants reported having an emergency room,
hospital, skilled nursing, or admission in the last year. Medication use was
very high, ranging as high as taking 27 different medications regularly and
37% took 12 or more medications. Thirty-six percent reported a fall in the
last three months, 38% reported dizziness, and 36% were experiencing
confusion. Additionally, 32% of the sample reported living alone and having
a clinical indicator of risk.

113
Table 4.1: Characteristics of the Intervention Sample (N=162)
N % Mean (SD)
Socio-Demographics
Age (65-108) 80.58 (7.76)
Female 131 80.9%
Race/Ethnicity
Caucasian 44 27.2%
African-American 62 38.3%
Latino/a 40 24.7%
Asian/Pacific Islander 12 7.4%
Other/Unknown 4 2.5%
Primary Language
English 99 61.1%
Spanish 35 21.6%
Armenian 11 6.8%
Chinese 2 1.2%
Known Other 9 5.6%
Unknown/Refused 6 3.7%
Marital Status
Married 25 15.4%
Widowed 93 57.4%
Divorced/Separated 29 17.9%
Single 12 7.4%
Unknown/Refused 3 1.9%
High School & Above 69 42.6%
Lived Alone 78 48.1%
Newly Enrolled in MSSP 68 42.0%
Health Status
ER/Hospital/SNF in Last Year 76 46.9%
# of Medications (3-26) 10.73 (4.70)
12+ Medications 60 37.0%
Falls in Past 3 Months 58 35.8%
Dizziness 61 37.7%
Confusion 58 35.8%
Lived Alone x Risk 51 31.5%
Risk defined as presence of falls, dizziness, or confusion

114
Table 4.2 report socio-demographic and health status characteristics
across the three MSSP sites. Significant difference across the sites
included race/ethnicity (p<.001), English-speaking (p=.004), and at least a
high school education (p=.026). Dizziness was also significantly lower for
Site #2 compared to the others (p=.037), however this might be due to data
collection issues.

Table 4.2: Characteristics of Intervention Sample by Site (N=162)
N % Mean (SD) N % Mean (SD) N % Mean (SD)
Socio-demographics
Age (65-108) 80.92 (8.47) 79.97 (6.63) 81.23 (8.73)
Female 47 79.7% 54 78.3% 30 88.2%
Race/Ethnicity***
Caucasian 31 52.5% 5 7.2% 8 23.5%
African-American 5 8.5% 50 72.5% 7 20.6%
Latino/a 17 28.8% 10 14.5% 13 38.2%
Asian/Pacific Islander 3 5.1% 3 4.3% 6 17.6%
Other/Unknown 3 5.1% 1 1.4% 0
English Speaking** 28 47.5% 52 75.4% 19 55.9%
Marital Status
Married 9 15.3% 9 13.0% 7 20.6%
Widowed 38 64.4% 37 53.6% 18 52.9%
Divorced/Separated 10 16.9% 12 17.4% 7 20.6%
Single 2 3.4% 8 11.6% 2 5.9%
High School & Above* 31 52.5% 21 30.4% 17 50.0%
Lived Alone 23 39.0% 39 56.5% 16 47.1%
Newly Enrolled in MSSP 21 35.6% 32 46.4% 15 44.1%
Health Status
ER/Hospital/SNF in Last Year 32 54.2% 34 49.3% 10 29.4%
# of Medications (3-26) 10.44 (4.67) 10.80 (4.65) 11.12 (4.94)
12+ Medications 23 39.0% 25 36.2% 12 35.3%
Falls in Past 3 Months 18 30.5% 22 31.9% 18 52.9%
Dizziness* 26 44.1% 19 27.5% 16 47.1%
Confusion 26 44.1% 18 26.1% 14 41.2%
Lived Alone x Risk 16 27.1% 23 33.3% 12 35.3%
*p<.05; **p<.01; ***p<.001
Risk defined as presence of falls, dizziness, or confusion
Site #1 (N=59) Site #2 (N=69) Site #3 (N=34)

115
Table 4.3 reports medication problem and change rates for the
intervention sample. Of the 162 participants in the intervention sample,
61% realized a medication change in one of the problem areas.
Psychotropic medication problems related to confusion showed the most
change with a 68% resolution rate for confirmed problems. Therapeutic
duplication, which had the highest problem rate at 13%, also had
substantial improvement at 62%. Although cardiovascular medication
problems were not confirmed at a high rate for the sample (4%), resolution
was comparable to the other problems at 46%. Psychotropic medication
problems related to falls had the lowest change rate at 43%.

Table 4.3: Medications Problem and Change Rates at 3-Month Follow Up
Medication Problem N % Prevalence N % ∆
Any 162 26.3% 99 61.1%
Ther. Duplication 79 12.8% 49 62.0%
Psychotropic 59 9.6% 32 54.2%
Psych-Confusion 34 5.5% 23 67.6%
Psych-Falls 37 6.0% 16 43.2%
CV 24 3.9% 11 45.8%
NSAIDs 44 7.2% 22 50.0%
Any = Any of the 4 Confirmed Medication Problems
Ther. Duplication = Therapeutic Duplication
Psychotropic = Inappropriate Use of Psychotropic Medication
Psych-Confusion = Inappropriate Use of Psychotropic Medication with Confusion
Psych-Falls = Inappropriate Use of Psychotropic Medication with Fall in Last 3 Months
CV = Cardiovascular Medication Problems
NSAIDs = Inappropriate Use of Non-Steroidal Anti-Inflammatory Drugs
∆ = Change
MSSP Sample Screened
(N=615) (N=162)
Medication ∆

116
Figure 4.2 describes the relationship between number of confirmed
medication problems and the number of problems that were resolved. For
participants with one medication problem, 59% had a medication change.
Those with two medication problems, 59% had resolution with at least one
of the problems and 43% had resolution with both. Participants who had 3
medication problems, 73% had resolution with at least one problem. All
participants with four identified problems had resolution with only one.

Figure 4.2: Number of Medication Problems by Percentage of
Medication Change (N=162)
41.2% 40.5%
26.7%
58.8%
16.7%
26.7%
100.0%
42.9%
26.7%
20.0%
0%
20%
40%
60%
80%
100%
1 Problem
(N=102)
2 Problems
(N=42)
3 Problems
(N=15)
4 Problems
(N=3)
# of Confirmed Medication Problems
% Improved
3 Improved
2 Improved
1 Improved
None
Improved

Data suggest that the consultant pharmacist was the main
communicator of medication problems and recommendations for change to
three actors: the prescribing physician, primary care manager, and
participant and/or caregiver. Communication to these actors was not
mutually exclusive, as the pharmacist could have contacted only one (e.g.,

117
the care manager) or up to all three. Communication with the physician
involved a faxed letter of recommendations for a medication change with a
follow-up phone call. Communication specifically directed at participants or
their caregivers, the pharmacist contacted with them either directly or
through the care manager who conveyed the recommendations for change.
Table 4.4 reports who the pharmacist contacted for each medication
problem and how many of those contacts were associated with a
medication change (for full information of the pharmacists’ contacts and
associated change rates, see Appendix I). For any medication problem, the
pharmacist contacted the prescribing physician for 91 participants and 58 of
those had a medication change (64%). Contacts with the primary care
manager (N=146) similarly resulted in a 62% change in medications. Only
58 participants were contacted directly by the pharmacist and the
associated change rate was a comparable 66%. The only problem where
contact with the prescribing physician resulted in a higher medication
change rate than the others was for cardiovascular medication problems
(58% vs. 50%). The highest percentage of change relative to contacts was
for participants and their caregivers who talked with the pharmacist directly
about psychotropic medication problems related to confusion (71%). These
data are descriptive in nature as none of the differences were statistically
significant.

118

Table 4.4: Pharmacist Contacts and Medication Change
# ∆ % ∆ # ∆ % ∆ # ∆ % ∆ # ∆ % ∆ # ∆ % ∆ # ∆ % ∆ # ∆ % ∆
Total # and ∆ 162 99 61.1% 79 49 62.0% 59 32 54.2% 34 23 67.6% 37 16 43.2% 24 11 45.8% 44 22 50.0%
Pharm Contacts
MD Contacts 91 58 63.7% 45 28 62.2% 37 19 51.4% 19 12 63.2% 27 12 44.4% 12 7 58.3% 21 13 61.9%
CM Contacts 146 91 62.3% 74 46 62.2% 53 29 54.7% 29 20 69.0% 34 15 44.1% 21 10 47.6% 39 20 51.3%
P/C Contacts 58 38 65.5% 30 19 63.3% 23 14 60.9% 14 10 71.4% 13 8 61.5% 4 2 50.0% 16 10 62.5%
Any = Any of the 4 Confirmed Medication Problems # = Number of Cases Pharmacists Discussed
Ther. Duplication = Therapeutic Duplication ∆ = Medication Change
Psychotropic = Inappropriate Use of Psychotropic Medication Pharm = Consultant Pharmacist
Psych-Confusion = Inappropriate Use of Psychotropic Medication with Confusion MD = Prescribing Physician
Psych-Falls = Inappropriate Use of Psychotropic Medication with Fall in Last 3 Months CM = Primary Care Manager
CV = Cardiovascular Medication Problems P/C = Participant and/or Caregiver
NSAIDs = Inappropriate Use of Non-Steroidal Anti-Inflammatory Drugs
Psychotropic Ther. Duplicaton Any Problem NSAIDs CV Psych-Falls Psych-Confusion

118

119
Medication problem and change rates for the 51 participants who lived
alone with a risk factor of falls, dizziness, or confusion are described in
Table 4.5. Those living alone with a risk factor had comparable rates of
medication change relative to the larger intervention sample at 61%.
Therapeutic duplication was still the most prevalent problem; 65% of this
group had a change. Psychotropic medication problems related to falls had
the lowest change rate at 35%.
The second and third columns report pharmacist communication with
those who lived alone with a risk factor. Thirty percent of this group
received a pharmacist contact during the intervention phase. However the
pharmacist mostly activated the assistance of the prescribing physician and
care manager for this group, as only one of the 15 pharmacist contacts was
to the participant and/or caregiver exclusively. For those with
cardiovascular medication problems, the pharmacist only addressed the
prescribing physician and care manager for resolution. Similarly the
pharmacist only contacted three participants and/or caregivers associated
with a psychotropic medication problem related to confusion, and all had
change at the 3-month follow-up.

120
Table 4.5: Medication Problems and Change for Those Who Lived Alone with Risk (N=51)
# ∆ % ∆ # ∆ % ∆ # ∆ % ∆
Medication Problem
Any 51 31 60.8% 15 9 60.0% 1 0 0.0%
Ther. Duplication 26 17 65.4% 9 5 55.6% 1 0 0.0%
Psychotropic 22 10 45.5% 8 4 50.0% 0 0 0.0%
Psych-Confusion 8 5 62.5% 3 3 100.0% 0 0 0.0%
Psych-Falls 17 6 35.3% 5 1 20.0% 0 0 0.0%
CV 8 4 50.0% 0 0 0.0% 0 0 0.0%
NSAIDs 11 5 45.5% 5 3 60.0% 0 0 0.0%
Any = Any of the 4 Confirmed Medication Problems
Ther. Duplication = Therapeutic Duplication
Psychotropic = Inappropriate Use of Psychotropic Medication
Psych-Confusion = Inappropriate Use of Psychotropic Medication with Confusion
Psych-Falls = Inappropriate Use of Psychotropic Medication with Fall in Last 3 Months
CV = Cardiovascular Medication Problems
NSAIDs = Inappropriate Use of Non-Steroidal Anti-Inflammatory Drugs
∆ = Medication Change
Pharm Talk = Pharmacist Talked with Participant &/or Caregiver
Pharm Only Talk = Pharmacist Talked Only with Participant &/or Caregiver
Risk defined as presence of falls, dizziness, or confusion
Lived Alone w/ Risk Pharm Talk Pharm Only Talk

Table 4.6 shows Pearson product-moment correlation

results for
significant participant-level characteristics related to medication change at
the p<.05 level. Socio-demographic, health status, and change variables
were entered into the correlation matrix; however the only medication
categories with significant results were psychotropic medication problems
and more specifically those related to falls. Participants living alone had a
negative relationship with changes in psychotropic medication problems
(r=-.266; p=.044); those with at least a high school education also had a
negative relationship to changes in psychotropic medications related to falls

121
(r=-.367; p=.026). No other participant- or organizational-level
characteristics showed significance in bivariate or multivariate models.

Table 4.6: Pearson Product-Moment Correlation Results for Medication Change
High School Education & Above -0.367 *
Lives Alone -0.266 *
*p<.05
Psychotropic = Inappropriate Use of Psychotropic Medication
Psych-Falls - Inappropriate Use of Psychotropic Medication Related to Falls
(N=59)
Psychotropic Psych-Falls
(N=37)

D. Discussion
This study evaluated the resolution of medication problems through a
pharmacist-centered intervention in a Medi-Cal waiver care management
program. Sixty-one percent of those who received the medication
management intervention reported a medication change at the three-month
follow up. Similar to findings by Meredith and colleagues (2002) in home
healthcare, therapeutic duplication problems had a large percent change
(62% in MSSP; 71% in home healthcare), yet the greatest change was for
psychotropic medication problems related to confusion at 68%.
Cardiovascular medications were confirmed as a medication problem for
only 4% of participants; however resolution of those confirmed with a
confirmed problem was complimentary to the home healthcare sample
(46% in MSSP; 40% in home healthcare). This finding suggests that

122
although care managers had difficulty gathering necessary data to confirm
a cardiovascular medication problem, resolution of the problems was
equally possible in this setting.
Participants with a higher number of medication problems had greater
potential to realize at least one medication change. This finding implies the
variability in what pharmacists categorized as a medication problem based
on specific criteria and the clinical realities of making a medication change.
Although it is unknown why a change did not occur in this sample, previous
research suggests that reasons for not making a change include patient
reluctance, a physician’s previous attempt to change, clinical failure of a
pharmacist recommendation, and the inability to address a medication
problem due to other, more critical health care issues (Sellors et al., 2003).
Higher level analyses to identify participant and organizational predictors
of medication change did not yield meaningful results. This likely occurred
because the true agent of change was the consultant pharmacist.
Pharmacists were the key implementer of the intervention with support from
care managers. Pharmacists developed individualized care plans and
communicated this plan to prescribing physicians as well as participant and
caregivers utilizing expert pharmacotherapy knowledge and skills. Although
officially acting in a consultant role in MSSP, pharmacists led the
intervention, soliciting assistance from care managers who had already
developed rapport with participants during the home assessment. Care

123
managers in turn supported the pharmacists by communication medication
solutions to participants and/or facilitating the pharmacist to talk with
participants directly.
The number and type of pharmacist contacts further demonstrated the
individualized and collaborative nature of the intervention. Pharmacists
communicated with any combination of prescribing physicians, care
managers, and participants and/or caregivers. Interdisciplinary
collaboration for the intervention is best defined by results from those who
lived alone with a risk factor. Pharmacists only communicated directly with
30% of these participants and/or caregivers, and instead focused attention
toward prescribing physicians and care managers to promote problem
resolution. Although delineating whether the pharmacist contacted the
participant or the caregiver was not available in the data, the fact that these
participants lived alone suggests that caregivers may not have been readily
available. As such, pharmacists communicated with these participants
when appropriate and engaged formal providers to promote continuity of
care for this higher risk group within a sample already at risk for
institutionalization.
Although the pharmacist’s role was essential, the importance of
interdisciplinary collaboration, including engagement by the prescribing
physician, to resolve medication issues can not be overstated. Pharmacists
and care managers can work together with physicians to meet the

124
medication management needs of vulnerable populations such as dual-
eligibles enrolled in Medi-Cal waiver programs (ASHP Continuity of Care
Task Force, 2005). The key to success is working collaboratively without
adding undue strain on physicians who may already be overburdened by
competing demands of the needs of chronically ill patients and health care
system regulations (Boling, 2002; Gilbert et al., 2002). One way the CBM
Intervention attempted to reduce physician burden was to provide the
physician with patient-specific recommendations limited to the four problem
types rather than a general training process on medication concerns
specific to older adults (Hanlon et al., 1996).
Limitations
This study had limitations related to solutions for NSAIDs problems, data
collection and the nature of the evaluation design. Of the four problem
types, the solution for MSSP participants with NSAIDs problems was
complicated by external factors. In the original study, most home
healthcare patients who screened in with a NSAIDs problem were switched
to the newer Cox-2 Inhibitor class of NSAIDs. Cox-2 Inhibitors include
Bextra® and Vioxx®, both of which were removed from the market in 2005
due to increased risk of cardiovascular events, and Celebrex®, which
remained on the market but was re-labeled with “serious and life-
threatening” warnings for cardiovascular events (Center for Drug Evaluation
and Research, 2005). These medications were the key solutions in the

125
home healthcare trial, yet were no longer alternatives for MSSP participants
needing pain relief and likely affected the change rate for this problem type.
In an effort to implement the intervention without dramatically increasing
physician or care manager burden, medication change was measured by
participant and/or caregiver report to the care manager during a home visit.
Prior to the visit, pharmacists provided care managers with information of
the participant’s identified medication problem and what a medication
change would specifically entail. Even with this mechanism, false positive
or false negative reporting could have biased this measure. A similar
challenge was that data did not capture any reasons that a medication
change did not occur.
Although data collection delineated pharmacist contacts made to
prescribing physicians and care managers, contacts made to participants
and/or caregivers were collected as a single measure. It is possible that
who a pharmacist contacted at the participant level might have affected the
problem resolution rate. To better understand this issue, participants who
spoke with the pharmacist were separated by living status. Caregivers not
living in the home could have been available for consultation; however it is
more likely that participants who lived alone spoke directly with the
pharmacist. Of the 58 participants who spoke directly with the pharmacist,
44% lived alone (N=26). The only significant medication change for this
subgroup was that 87% of those living with someone had a psychotropic

126
medication change related to falls than those living alone (Mann-Whitney
U=6.5; p=.019). Although this result suggests that support in the home can
affect medication improvement, further work is needed in this area to clarify
this relationship.
A final set of limitations involved the evaluation design. First, medication
change protocols were focused on reducing access to inappropriate
medications, not affecting participant adherence and/or compliance with a
medication regimen. Therefore, participant consumption patterns were not
evaluated. Also, the CBM Intervention project did not test whether it would
have been successful without direct intervention from a consultant
pharmacist. Although in the initial program design the care manager was
intended to fulfill the pharmacist recommendations by contacting the
prescribing physician and participants with medication change strategies,
translation issues surrounding care manager scope of practice dictated that
the medication management role remained with the pharmacist. This and
related issues of staff perceptions on the CBM Intervention are the primary
focus of Chapter V.
Policy Implications
There are a number of policy implications that might affect dissemination
of the CBM Intervention into Medi-Cal waiver care management or
translation of the model to other medical/social hybrid programs. A primary
issue is cost of the intervention. Although cost estimate data for the project

127
were not available from the MSSP sites, the use of a consultant pharmacist
to operate the intervention would add a considerable cost to care
management organizations. The median salary for a consultant pharmacist
in 2004 was $84,900, calculating to approximately $41.00 per hour (Bureau
of Labor Statistics, 2004). This cost would be prohibitive for most agencies
with a limited discretionary budget, and currently Medi-Cal does not
reimburse for a consultant pharmacist through waiver services. However,
this study suggests that consultant pharmacist services can be beneficial to
reducing medication problems for a targeted sample using limited problem
criteria. The Centers for Medicare and Medicaid Services (CMS) should
consider approving the use of consultant pharmacists as a waiver services
vendor under these specific conditions. Given that the purpose of Medi-
Cal-waiver care management programs, such as MSSP, is to improve the
continuity of care for vulnerable populations and reduce the need for higher
level care, CMS should support this goal by targeted reimbursement of
consultant pharmacist services.
Additionally, most social services agencies that provide care
management do not have access to a consultant pharmacist due to the
limited number of professionals with geriatric pharmacotherapy expertise
and lack of organizational capacity to support the assessment and
intervention process. Care management is a general term and does not
denote any particular staff expertise or professional capacity. Therefore

128
care management programs that do not have medically trained staff, such
as a nurse, and function from a social model of care may not be able to
effectively support consultant pharmacists with the medication screening
and intervention process.
Conclusion
Findings from the intervention phase of the CBM Intervention project
show that medication problems can be resolved for community-dwelling,
dually eligible older adults in a Medi-Cal waiver care management program.
This is an important departure from the usual medical setting where
medication management programs have demonstrated success. The key
component was the use of a consultant pharmacist in care management
that maintained the medical nature of medication problems and solutions.
To understand the dynamics of this collaborative effort, Chapter V
describes the care managers’ perspective of implementing the CBM
Intervention.

129
V: STAFF PERCEPTIONS OF IMPLEMENTING
AN EVIDENCE-BASED MEDICATION MANAGEMENT INTERVENTION
IN CARE MANAGEMENT
A. Introduction
A rapid increase of evidence-based practice implementation into applied
health and social care settings has improved service delivery effectiveness
and reduced provider variability in applying interventions (Feldman & Kane,
2003). Although the use of evidence-based practices do not necessarily
reduce program costs (Sackett et al., 1996), improved consumer outcomes
and standardization of care within and between providers are positive
results that can be realized through this method. Even with these benefits,
provider perspectives of implementing evidence-based practices have
stirred much debate in the health service professions such as medicine
(Grol, 2001), nursing (Retsas, 2000), and social work (Gira et al., 2004),
and substantive practice areas including mental health (Aarons, 2004;
Tanenbaum, 2003), HIV/AIDS (Solomon, Card, & Malow, 2006), and
rehabilitative services (Koch, Cook, Tnakersley, & Rumrill, 2006). Several
issues fundamental to this debate are: What qualifies as “evidence” when
using evidence-based practices, what is the impact of these practices on
professional autonomy and decision making, and how can protocol-driven
practices be best implemented in a variety of contexts and with various
populations?

130
One human service delivery area that has lagged behind in
implementing evidence-based practices is the network of community-based
aging service programs (Bryant et al., 2006). The aging service network
often has relied on “home grown” programs, developing interventions within
their own local context that lack evaluations to support effectiveness on
achieving desired outcomes. The trend to increase health promotion
evidence-based practices for older adults is changing due to two
implementation forces (Sabatier, 1986): “top-down” support through several
new federal and state initiatives providing political will backed by funding
(Administration on Aging, 2004, 2006); and a “bottom-up” approach of
evidence-based program development at the local level (Bryant et al.,
2006). In light of its novelty, little is known about staff perceptions of the
evidence-based implementation process for this service sector. To better
understand provider attitudes in aging service programs, this chapter
reports on staff perceptions of implementing an evidence-based practice
(CBM Intervention project) into a Medi-Cal waiver care management
(MSSP).
The previous chapter highlighted the unique and invaluable
contributions from consultant pharmacists in implementing the CBM
Intervention. However the ultimate success of the CBM Intervention rested
with the direct service providers in MSSP: nurse and social work care
managers. Care manager engagement in and satisfaction with an

131
evidence-based practice, specifically medication management, is essential
for two reasons. First, care managers provide the direct contact and
service with participants and their caregivers through standard care
management activities of assessment, treatment planning, intervention,
monitoring, evaluation (Moxley, 1989; Scharlach, Giunta, & Mills-Dick,
2001; Tahan, 2006). Assessment includes initial contact with the
participant and caregivers to ascertain the challenges they face in trying to
maintain a functionally-impaired older adult at home. Treatment planning is
the direct result of synthesizing assessment information and includes
prioritizing issues, developing a plan of action to address problems based
strengths and needs of the participant and caregivers, and identifying
formal providers to assist with service delivery. Intervention in MSSP
involves directly linking the participant and caregivers to service providers,
providing service-related counseling, and accessing available funds within
the MSSP agency for purchase of services. Next steps include monitoring
via a monthly phone call and quarterly visits to ensure service utilization
based on the treatment plan, and evaluation to ascertain value-added of
care management services on the participant’s overall functioning.
Implementation of the CBM Intervention follows the same activity pathway
building on the existing care management process, and therefore requires
care manager participation and buy-in at every step.

132
The second, and a potentially more important reason from an
operational viewpoint, is that care managers are the authorized staff to
provide and arrange for services under the Medi-Cal waiver for MSSP.
Contractual vendors outside of MSSP often provide substantial services to
participants at the intervention phase (e.g., daily personal care assistance
provided by a local home care agency). However, care managers must
clearly document the need and plan for services to address a specific
treatment goal to allow for Medi-Cal payment. The application of a new
intervention must be justified and coherently linked to the care manager’s
evaluation of a participant’s unique situation, meaning that professional
judgment is at the core of this process. Therefore care manager
perceptions of an evidence-based practice, such as the CBM Intervention,
are critical for its success in expanding the parameters of MSSP standard
practice.
Professional Differences of Evidence-Based Practice
Care manager perceptions of evidence-based practice, however, are not
likely to be uniform given that two different professions perform this role:
nurses and social workers. Historically, nurses and social workers have
different professional traditions and attitudes toward evidence-based
practice. Nurses, who operate within a medical model of care, generally
have favorable attitudes toward research and the use of research findings
in clinical practice (Fink, Thompson, & Bonnes, 2005; Meijers et al., 2006;

133
Retsas, 2000). The dynamic interplay between research and practice is
institutionalized for nurses as described in a position statement from their
national organization:
Research-based practice is essential if the nursing profession
is to meet its mandate to society for effective and efficient
patient care. The development and utilization of research in
nursing practice depends upon the effective interaction of
clinicians and scientists. Clinicians identify relevant clinical
problems for investigation; researchers design studies to
address these problems. Clinicians are then, in a very
meaningful way, the consumers of research findings
(American Nurses Association, 1994).

Conversely, the social work profession has had a tentative relationship
with research and evidence-based practice. Some have called for greater
use of clinically efficacious interventions in social work practice (Gibbs &
Gambrill, 2002; McNeece & Thyer, 2004; Sheldon, 2001). Others have
expressed concern about quality standards of what constitutes “evidence”,
and inherent power issues in defining evidence-based practice that impact
professional autonomy, clinical decision-making, and a broader political-
economic agenda driving its use (Gambrill, 2006; Murphy & McDonald,
2004; Webb, 2001). However, both nursing and social work acknowledge
similar barriers to identifying and implementing evidence-based practices
such as accessibility of research findings and lack of organizational support
needed for implementation (Gira et al., 2004).
Nurses and social workers also have differing professional perspectives
on medication management. This activity is an integral part of nursing

134
practice according to the California Nursing Practice Act, the legislative
mandate for licensed nurses in the state. The practice of nursing includes
“functions…that help people cope with difficulties in daily living that are
associated with their actual or potential health or illness problems
or…treatment” that include “direct and indirect patient care services
including…administration of medications and therapeutic agents…”
(California Board of Registered Nursing, 1997). The Act also acknowledges
that a fundamental process of nursing is collaboration with physicians in
order to improve a patient’s health status.
Unlike nurses, social workers are not required to be licensed for practice
in some capacities, and hence there are two authoritative entities that guide
social work practice. The first is the National Association of Social Workers
(NASW), a professional organization whereby members pledge adherence
to an ethical code. A key element in the code is professional competence
in which social workers should “provide services and represent themselves
as competent only within the boundaries of their education, training, license,
certification, consultation received, supervised experience, or other relevant
professional experience” (National Association of Social Workers, 1996).
As for the implementation of new interventions, NASW recommends that
social workers “provide services in substantive areas or use intervention
techniques or approaches that are new to them only after engaging in
appropriate study, training, consultation, and supervision from people who

135
are competent in those interventions or techniques” (National Association of
Social Workers, 1996). When engaging in new areas of practice or in
collaboration with providers with different expertise, NASW recommends
that “social workers should seek the advice and counsel of colleagues
whenever such consultation is in the best interests of clients” and “should
keep themselves informed about colleagues' areas of expertise and
competencies” (National Association of Social Workers, 1996).
The second authoritative body for those who are licensed in California is
the Board of Behavioral Sciences (BBS), a licensing board under the
Department of Consumer Affairs. BBS defines clinical social work practice
as:
The application of social work principles and methods
includes providing information and referral services; providing
or arranging for the provision of social services; explaining or
interpreting the psychosocial aspects in the situations of
individuals, families, or groups; helping communities to
organize, to provide, or to improve social or health services; or
doing research related to social work (California Board of
Behavioral Sciences, 2006).

Therefore the social work perspective on evidence-based practices and
medication management are open for interpretation and depend upon the
skills and training of individual practitioners. Rather than offering definitive
statements in these areas of practice, definitions of social work practice
acknowledge the importance of collaboration with knowledgeable

136
professionals who can provide clinical expertise as appropriate for a client’s
needs.
The purpose of this study was to assess nurse and social work care
managers perspectives on the implementation of an evidence-based
practice into an existing program structure in order to gain new insights and
lessons learned from those who function in this key direct service role. It
was anticipated that using a structured forum to provide feedback, care
managers could provide useful insights into the implementation process,
which could be used for continuous quality improvement and further
refinement of the CBM Intervention in MSSP and other programs. Given
the professional differences described above, it was hypothesized that
nurses would report a more positive experience with implementing the CBM
Intervention than social workers given the intervention’s origins as an
evidence–based practice from a medical orientation.
B. Methods
Sample
Upon completion of the CBM Intervention in January 2006, MSSP care
managers were solicited to complete a 33-item MSSP Staff Questionnaire
about their experience in implementing the intervention. The sample
consisted of care managers and direct supervisors of the three MSSP
program sites who participated in implementing the CBM Intervention. Care
managers who terminated employment of the MSSP program but had

137
participated in the implementation for at least 3 months were also included.
Those excluded from participation were MSSP administrative staff not
involved in participant care and care management staff employed less than
60 days at the end of the intervention period for each site.
All eligible staff employed at the three MSSP sites completed the
questionnaire (N=25), two of which were direct supervisors. Questionnaires
were sent out to seven former MSSP care managers with known
addresses. Four surveys were returned creating a total sample of twenty-
nine or 91% of possible respondents.
Procedure
The questionnaire was disseminated by the researcher at each MSSP
site during a regularly scheduled staff meeting. Participants were provided
general information about the survey’s purpose and told that completion of
the questionnaire was voluntary with responses remaining confidential.
Upon completion, staff returned the questionnaires to the researcher in a
sealed envelope. For those who were no longer working in MSSP, the
program site of previous employment mailed a questionnaire and a
stamped return envelope addressed to the researcher. A small incentive
($5 Starbucks Coffee Company card) was provided to all staff who received
a survey. Institutional Review Boards for the research and project partners
reviewed and approved the study.

138
Care Manager Questionnaire
The MSSP Staff Questionnaire was developed specifically for this
dissertation to gather information about the care managers’ experiences in
implementing the medication management intervention (see Appendix J for
sample of questionnaire). The questionnaire were developed using the
PARIHS framework and qualitative feedback gathered through staff and
advisory group meetings with the facilitation team.
The development of the questionnaire consisted of the following steps:
1. Reviewed the PARIHS concepts of evidence, context, and facilitation
and sub-criteria from literature to build item bank (Rycroft-Malone et
al., 2002);
2. Developed questions to reflect the key elements of each concept as
applicable to this implementation;
3. Received reviews by an expert in survey development and three
researchers with familiarity in translational research;
4. Incorporated feedback from reviews and scaled questionnaire back
to 33 items;
5. Pilot-tested revised questionnaire with three MSSP administrators for
face validity; and
6. Completed final modification of questionnaire based on feedback
from pilot testing.

139
Measures
Independent Variables
Several questions were included to gather employment characteristics
of the staff as reported in Table 5.1. Professional type distinguished
between nurses, social workers and other professional affiliations. Level of
education ranged from Associates to Doctoral degree. Staff were asked if
they were licensed in their given profession. Length of time in their
profession and length of time as an MSSP employee were ordinal
measures ranging from less than 1 year to over 10 years of experience.
Staff were also asked if they were a current MSSP employee. Staff self-
reported the average size of their caseload in a given month. Finally staff
were asked about their experience with implementing a medication
management intervention and implementing an evidence-based practice
prior to the CBM Intervention with response categories of “none”, “some”,
and “a lot.” Respondents affiliated with Site #1, which was the location of
the CBM Intervention facilitation team, were coded as 1 and all others as 0.

140
Table 5.1: Measures of Care Manager Characteristics
Variable Coding
Professional Type
Nurse 0
Social Worker 1
Other 2
Highest Education Level
Associates Degree 1
Bachelor's Degree 2
Master's Degree 3
Doctorate 4
Licensed Professional Yes=1; No=0
Length of Time in Profession
Less than 1 Year 1
1-3 Years 2
4-6 Years 3
7-9 Years 4
10+ Years 5
Current MSSP Employee Yes=1; No=0
Length of Time in MSSP
Less than 1 Year 1
1-3 Years 2
4-6 Years 3
7-9 Years 4
10+ Years 5
Caseload Size Self Report
Medication Management Experience
None 0
Some 1
A Lot 2
Evidence-Based Practice Experience
None 0
Some 1
A Lot 2
Employed at Site #1 (Faciliation Site) Yes=1; No=0

141
Dependent Variables
The majority of the questionnaire (see Appendix J) consisted of 21
questions to elicit the care managers’ perspective on the implementation
process. Questions were developed based on the PARIHS concepts of
evidence, context, and facilitation (Rycroft-Malone et al., 2002). Facilitation
questions (items A – E) focused on the training processes and the ability to
interact with the facilitation staff. Questions related to Evidence (items F –
M) inquired about the importance of using an evidence-based practice, the
ease of using the assessment and follow up tools, and perceived client
response to the medication management protocol. Context questions
(items N – U) focused on how well the medication management intervention
meshed with standard MSSP care management practice. These items
were scored on a 5-point Likert-scale (strongly agree – 5 points; agree – 4;
neither agree nor disagree – 3 points; disagree – 2 points; and strongly
disagree – 1 point).
The questionnaire finished with three open-ended items to gather
additional care manager feedback on the benefits, challenges, and any
additional information on the project. The 33-item questionnaire took
approximately 10 minutes to administer.
Analysis
Analytic methods included descriptive statistics, t-test for independent
means, and non-parametric statistics. Non-parametric statistics were used

142
due to the small sample size, use of ordinal measurement, and lack of a
normal distribution in the dependent variables (Pett, 1997). Fisher’s Exact
Test was specifically employed to evaluate group comparisons on nominal
measures where cell sizes were less than five. Spearman’s rank-order
correlation coefficient or Spearman’s rho, a non-parametric bivariate
statistic, was used to measure degree of association among sample
characteristics (e.g. professional type) and questionnaire response
variables that were ordinal measures. To further evaluate the degree of
association between sample characteristics and questionnaire responses,
the Mann-Whitney U test was used to assess mean and median rank of
response differences between nurses or social workers, those with and
without experience with medication management, and those with and
without experience with evidence-based practices. Qualitative responses to
the open-ended items were also reported.
C. Results
Quantitative Responses
Table 5.2 reports descriptive statistics for the sample including
differences between nursing and social work staff. Twenty-four social work
staff at the direct service (N=22) and supervisor (N=2) levels and five
nurses completed the questionnaire. Social workers had an overall higher
educational level with nearly 60% having a Master’s degree compared to
none of the nurses (p<.05; Fisher’s exact). However, all nurses were

143
licensed by the state of California to practice compared to only 17% of
social workers (p<.001; Fisher’s exact). All nurses also had at least ten
years of professional experience compared to only 17% of social workers
(p<.001; Fisher’s exact). Social workers had on average a higher case load
than nurses (t=3.65; p<.001), which was consisted with MSSP workforce
requirements at the three sites. All nurses reported at least some previous
experience with medication management whereas only 58% of the social
workers did. Similarly, 80% of nurses reported at least some previous
experience with evidence-based practices compared to 50% of social
workers. Differences between professional type and these experiences
were not significant (Fisher’s exact p-value).

144
Table 5.2: Characteristics of Care Manager Sample
N % or Mean (SD) N % or Mean (SD) N % or Mean (SD)
Site
1 7 24.1% 1 20.0% 6 25.0%
2 11 37.9% 1 20.0% 10 41.7%
3 11 37.9% 3 60.0% 8 33.3%
Highest Education Level*
Associate's 1 3.4% 1 20.0% 0
Bachelor's 14 48.3% 4 80.0% 10 41.7%
Master's 14 48.3% 0 14 58.3%
Licensed Professional*** 9 31.0% 5 100.0% 4 16.7%
Time in Profession
1-3 Years 8 27.6% 0 8 33.3%
4-6 Years 4 13.8% 0 4 16.7%
7-9 Years 6 20.7% 0 6 25.0%
10+ Years 11 37.9% 5 100.0% 4 16.7%
MSSP Employee 25 86.2% 5 100.0% 20 83.3%
Time in MSSP
Less than 1 Year 5 17.2% 0 5 20.8%
1-3 Years 11 37.9% 2 40.0% 9 37.5%
4-6 Years 8 27.6% 1 20.0% 7 29.2%
7-9 Years 3 10.3% 1 20.0% 2 8.3%
10+ Years 2 6.9% 1 20.0% 1 4.2%
Mean Case Load (28-50)*** 42.1 (6.13) 33.8 (2.06) 43.7 (5.29)
Med Mgmt. Experience
At Least Some 19 65.5% 5 100.0% 14 58.3%
EBP Experience
At Least Some 16 55.2% 4 80.0% 12 50.0%
Med Mgmt. = Medication Management
EBP = Evidence-Based Practice
P-Values from t-Test and Fisher's Exact Test as Appropriate
Total Sample (N=29) Nurses (N=5) Social Workers (N=24)
*p<.05; **p<.01; ***p<.001

Staff responses were overall positive toward the implementation of the
CBM Intervention (see Appendix K for full responses). The first five
questions (items A – E) dealt with facilitation issues, inquiring about how
well the administrative team assisted staff with implementing the

145
intervention. Sixty-five percent agreed that training on the CBM
Intervention was adequate; however the same percentage stated that
additional training would have been helpful. Staff noted the responsiveness
of the CBM Intervention administrative team as 79% reported that the team
provided assistance when needed and 69% reported that staff questions
were addressed in a timely manner.
The middle eight questions (items F – M) evaluated the staff’s
perceptions of evidence, and these scores were generally lower with more
responses in the “neither agree nor disagree” category. Sixty-five percent
agreed it was important that the medication management intervention was
tested in a research setting. Fifty-nine percent agreed that completing the
medication follow-up form was easy, but 41% acknowledged it was time
consuming to complete. Over 60% of the sample reported that their
participants were aware of the medication screening yet only 41% agreed
that participants were pleased that medication management was offered as
a service. Fifty-nine percent agreed that over time, they became more
comfortable with implementing the intervention. Only 41% of staff agreed
that medication management was part of their professional scope of
practice and 34% disagreed with this statement.
The final eight items addressed issue of context (items N – U). The
highest rated item overall was that 86% of staff agreed medication issues
and solutions should be discussed in case conference. Nearly 50%

146
reported that medication management fit easily with other care
management duties and 55% agreed that they were able to adapt the CBM
Intervention to how care management operates in MSSP. Although 52%
agreed that their professional role in the CBM Intervention was clear, 24%
of the sample disagreed with this statement. However 69% agreed that the
intervention was useful and worthwhile and 66% agreed that it should be an
ongoing service.
Complementing the descriptive data, Table 5.3 reported non-parametric
correlations using Spearman’s rho showed strong associations between
nurses and the following variables: licensure status (r
s
=-.680; p<.001),
greater length of time in profession (r
s
=-.514; p=.004), belief that medication
management was within their scope of practice (r
s
=-.558; p<.001), and
more perceived contact with physician to inform of medication problems
(r
s
=-.443; p=.030). Also, those with less time working in MSSP were
associated with agreement that additional training on medication
management intervention would have been helpful (r
s
=-.489; p=.011), the
medication follow-up form was easy to complete (r
s
=-.416; p=.034), and
medication management should be an ongoing component of MSSP (r
s
=-
.504; p=.005). Those with at least some experience with medication
management programs were associated with more years working in their
profession (r
s
=.683; p<.001) whereas those with some experience with
implementing evidence-based practices had greater agreement that

147
medication management was part of their scope of practice (r
s
=.522;
p=.004) and contact with physician to inform of medication problems
(r
s
=.510; p=.013).

Table 5.3: Spearman's Rho Results of MSSP Staff Questionnarie Responses
Highest Academic Level 0.495 **
Licensed Professional -0.680 ***
Time in Profession -0.514 ** 0.683 *** 0.599 ***
Additional training on the MM
Intervention would have been helpful
-0.489 *
Completing the medication follow-up
form was easy
-0.416 *
Medication management is typically part
of my scope of practice
-0.558 ** 0.522 **
I contacted my clients’ physicians to
inform him/her of a potential medication
problems
-0.443 * 0.444 * 0.510 *
My role in the MM Intervention was clear 0.379 *
I would like the MM Intervention to be an
ongoing service provided by MSSP
-0.375 * -0.504 **
Social Worker (1) Compared to Nurse (0)
Employee = Employed in MSSP at Questionnaire Administration
Med Mgmt. = Medication Management
Exp. = Experience
EBP = Evidence-Based Practice
*p<.05; **p<.01; ***p<.001
EBP Exp. Social Worker Employee Time in MSSP Med Mgmt. Exp.

To further understand these associations, the Mann-Whitney U test was
used to analyze differences between professional types, those with practice
experiences in medication management, implementing evidence-based
practices, and those from Site #1 housed with the facilitation team on the
questionnaire items (Table 5.4). Compared to social workers, nurses

148
agreed that medication management was within their scope of practice
(p=.003) and they reported contacting the physician to inform of a
participant’s potential medication problem (p=.034). Similarly, those with
some experience in implementing evidence-based practices agreed that
medication management was within their scope of practice (p=.009) and
reported contact with the physician to inform of potential medication
problems (p=.027).
There were several significant differences between care managers
located at Site #1 where the facilitation team was housed relative to those
at the other two sites who received intermittent contact with the facilitation
team. Care managers at Site #1 reported that the training tools and
protocols were helpful (p=.016) and that the facilitation team provided
assistance when needed (p=.025). They also reported more favorably that
medication management helped participants on their caseload reduce or
avoid medication problems (p=.046) and that the intervention was
worthwhile (p=.007). Finally, there were no significant differences related to
medication management experience on the questionnaire items.

149

Table 5.4: Mann-Whitney U Statistics of MSSP Staff Questionnarie Responses
Median 44
Mean Rank 18.10 11.03
Mann-Whitney U 20 *
Median 54
Mean Rank 20.71 13.18
Mann-Whitney U 37 *
Median 43 4 3
Mean Rank 24.90 12.94 16.54 9.17
Mann-Whitney U 11 ** 32 **
Median 43 4 3
Mean Rank 18.75 11.25 13.15 7.85
Mann-Whitney U 15 * 24 *
Median 44
Mean Rank 19.00 12.57
Mann-Whitney U 33 *
Median 44
Mean Rank 21.79 12.84
Mann-Whitney U 30 **
SW = Social Worker EBP = Evidence-Based Practice
Med Mgmt. = Medication Management Housed On-Site = Facilitation Team housed in Site #1
Exp. = Experience
None (N=12)
Professional Type EBP Exp. Housed On-Site
Yes (N=7) No (N=22) Some (N=16)
Training tools and protocols helped me learn and implement the
MM Intervention
The MM Intervention administrative team provided assistance
when needed
Nurse (N=5) SW (N=24)
*p<.05; **p<.01
Medication management is typically part of my scope of practice
I contacted my clients’ physicians to inform him/her of a potential
medication problems
In general, the MM Intervention helped my clients reduce/avoid
medication problems
The MM Intervention was useful and worthwhile

149

150
Qualitative Responses
Three open-ended items requested comments on the benefits,
challenges, and other feedback about the CBM Intervention (see Appendix
L for full responses). Staff comments on the open-ended questions were
coded for the respondent’s professional type. Themes generated about
benefits of the CBM Intervention centered on becoming more
knowledgeable about specific medications and their potential side effects.
Staff responded with comments on the following intervention benefits.
Nurse: Looking closely at the meds and evaluating the safely
and effectiveness of meds; having [medication management]
increased teaching on [medications], side effects and
therapeutic effect which is good practice in patient care

Social worker: Being able to identify risky [medications] and
duplication and informing clients or their families of potential
side effects;

Social worker: As a social worker I became aware of potential
dangers of or complications of some medications; I now look
at all medications my clients are taking

Social Worker/Supervisor: From my observation, our clients
benefited by reduction in health complications, reductions in
duplications of meds and improvement in overall functioning

Staff also acknowledged several challenges in the implementation
process. These included inter-organizational issues, such as the need for
staff collaboration to complete intervention protocols, lack of ongoing
training, and computerized systems initially designed to assist with the
implementation were not functional.

151
Social worker: If [CBM Intervention] could provide a staff
member to do quarterly [follow up] and take blood pressure
when [social worker] is not able to, [that] would be helpful

Social Worker: I received practically no orientation to [CBM
Intervention], its purpose [or] objectives when starting
employment. I was told it existed and given forms for [follow
up]. I still don’t totally get what [CBM Intervention] was about
or its purpose.

Social worker: I was asked to call clients to tell her about
potential harm of taking two medications (due to side effects)
but was not able to clearly explain to client about why that was
so. Someone from [CBM Intervention] program should make
calls to the client to better explain and inform the client

Social worker: The computer program is not user friendly. [It]
needs major improvement

Challenges stemming from external stakeholders generally focused on
patient resistance to medication change, lack of family support, and a lack
of responsiveness from physicians.
Nurse: [Physicians] do not respond to all calls

Social worker: No or slow response from the doctor. Many
clients like to keep all [medications] including those been
taking off [sic], making it very confusing. It can take a long
time to address a [medication] problem

Nurse: There was some resistance to change; follow up was
slow at times

Social worker: Some clients have taken certain medications
for so long that they were unwilling or fear change

The final question asked for any additional information to improve the
program. Responses focused on issues of additional training, scope of
practice issues for a problem perceived as a medical issue, and improved

152
collaboration among the MSSP staff within the constraints of a Medi-Cal
waiver program.
Nurse: Would like more contact with pharmacy staff

Social Worker: More training to [social work] staff on high risk
medications and side effects

Social worker: [Need for] orientation for new employees [and]
additional assistance/orientation for non-medical staff (social
workers). Also, I never knew exactly what the expectations
were of me as a participant in [medication management].
Some of my answers that are “disagree” have more to do with
the demands/constraints of my normal MSSP workload than
with [the CBM Intervention’s] goals/purpose/issues.

Nurse: We need to continue with the [medication]
management program; I think having a pharmacist reviewing
meds is very helpful

Social worker: [CBM Intervention] more applicable if overseen
by [nurses] and not [social workers]

Social worker: More training with the case managers on how
to approach client/ family/ doctor about the potential harm
would be helpful

D. Discussion
This is the first study known to ascertain care manager perceptions of
implementing an evidence-based practice into a community-based aging
service program. The uniqueness of this study was that it challenged a top-
down nature of evidence-based practice implementation, which assumes
that evidence-based practices are inherently beneficial for all stakeholders
and therefore are beyond reproach from direct service staff. The goal of
this study was to gather lessons learned from the rich experience of care

153
managers who served as the direct implementers of a new evidence-based
practice. Questions were developed using the PARIHS translational
research framework that focused on evidence, context, and facilitation as
three key components to implementing research into practice. Themes for
lessons learned are described using this framework.
Evidentiary Lessons
Lessons related to the evidence domain suggested that through the
implementation process, care managers became more knowledgeable
about medications, potential problems that medications can cause, and
warning signs of these problems. Care managers perceived a
strengthening of standard practice by adding a new dimension of problem
identification and viable solutions to the high prevalence of medication
problems. Most of the care managers expressed satisfaction that the
intervention originated from an evidence-based practice, and this
preference did not vary by professional type. This finding suggests that
although care managers might not seek out efficacious interventions, many
were receptive to using them in daily practice when introduced.
Even with these benefits, the intervention was impeded by participant
and external issues that related to the program’s evidence base.
Participant problems included difficulty gaining accurate information from
participants and caregivers to assist with assessment process, and some
resistance to changing medications even when informed of potential

154
dangers. Although physicians ultimately make the prescribing decision,
physicians may be reluctant to make change in medication regimens if a
patient resists (Little et al., 2004). Previous studies also acknowledge that
patient resistance to change can be a barrier to implementing any type of
evidence-based practice (McKenna, Ashton, & Keeney, 2004; Meijers et al.,
2006).
The main external issue that affected implementation was that a
computerized system intended to complete the initial screening process for
medication problems did not become fully operational in the intervention
period. Based on the original implementation protocols, the initial screening
process was supposed to identify participants with problems using an alert
system that incorporated data from the medication screening algorithm and
care management assessment. During training for the intervention, care
managers were informed of this system and how this technology would
assist them in bringing participants with identified problems to the attention
of the consultant pharmacists. However, interface problems between these
two databases caused the screening system to be inoperable. Several
beta-tests of the system were conducted without success. Therefore,
rather than care managers informing the pharmacists of participants with
identified problems, the pharmacists conducted both the initial screen to
identify potential medication problems and then confirmed the accuracy of
problems based on additional clinical information. This change not only

155
impacted the intended implementation of the evidence, but had a negative
result on care manager attitudes toward the intervention. Care managers
expressed frustration with the computerized system after several failures
and raised concerns about the feasibility of the intervention to function in
MSSP.
Contextual Lessons
The majority of the lessons learned from care managers related to
contextual factors. The most prominent issue focused on the interplay
between professional scope of practice and role specialization. Social
workers were more apt to respond that medication management as an
intervention was outside their scope of practice. Rather than expressing a
willingness to adopt intervention processes, they reported that nurse care
managers and consultant pharmacists should oversee the CBM
Intervention given that their medically-based expertise is more congruent
with communicating medication issues to participants and physicians. This
finding was corroborated by the finding that nurses reported participating
more fully in the intervention by contacting some physicians’ offices to
discuss medication problems.
Although communication with various providers is both a professional
norm for nurses and social workers and a routine process in MSSP,
collaboration with consultant pharmacists was novel. Care managers
reported that having access to the pharmacists for direct consultation was

156
valuable for clarifying presenting problems at assessment, creating
treatments plans and intervention strategies, and for improving staff
knowledge on medication issues. Inter-professional communication beyond
MSSP staff, however, presented some unique challenges. Care managers
reported that some physician offices were not responsive to communication
about medication problems. In addition to participant factors, they reported
that a barrier to effective medication management was physician resistance
to change. A possible reason for this interaction could be the nature of the
relationship between MSSP staff and participants’ physicians. Unlike home
healthcare that requires a physician order for services, participants may
engage in MSSP without any physician knowledge. When calling or
sending faxes regarding medications issues, MSSP staff may be contacting
a physician’s office for the first time. Physicians who receive treatment
recommendations without a prior relationship with the MSSP care manager
may perceive the communication as ill-informed, inappropriate advice
and/or may disregard it altogether.
Facilitation Lessons
Three themes emerged for related to care manager perceptions of
facilitation. First, those employed at Site #1 where the facilitation team was
housed had more favorable ratings of facilitation activities, such as training
tools and receiving assistance when requested, and the usefulness of the
intervention than their counterparts in Sites #2 and #3. Qualitative

157
responses regarding a desire for more collaboration with the consultant
pharmacists and regular communication with the project’s facilitation team
might have been site-specific and related to the proximity and access to
facilitation team. Although facilitation team members (a consultant
pharmacist and a support staff) spent approximately ten hours a month at
each site and attended care coordination meetings, care managers
responded with a greater perceived need for direct contact that impacted
their perception of the project as a whole. A parallel theme was care
managers’ requests to benefit from supplemental facilitation activities.
These included additional and ongoing training on general medication
issues and specific medication management strategies to address identified
problems. Both lessons are important integrate in efforts toward
widespread dissemination and implementation in various contexts.
Finally, the facilitation process was often impacted by evidence or
contextual factors. For example, one area of training requested was the
clarification of the care manager’s role when working with other treatment
partners due to role specialization issues. Social work care managers in
particular expressed concern about how to specifically approach the
participant, caregivers, or physician’s office with medication-related issues
given their general unfamiliarity with the subject. Based on feedback from
the CBM Intervention Project Director who worked on the original clinical
trial, these issues did not arise in home healthcare because of standard

158
practice issues for nurses to directly implement physician changes in a
treatment plan, including medication management. This issue highlights
the need for a different training approach in future implementation activities
that account for the specialized needs and potential knowledge gaps of the
direct service audience. A related example is how the failure of the
computer interfaces required the facilitation team to abandon the
computerized screening system during the project and switch to hand-
screening by the consultant pharmacists. This key change led to the
facilitation team to accept more responsibility for the implementation
process, which may have affected care manager perceptions of the
project’s sustainability for the long term. At the present time, the lead
organization (MSSP A) is working with the vendor of the computer system
to address technical malfunctions and improve the operability of medication
management screening within the agency’s database.
Limitations
Although a strength of the questionnaire was its development specific to
the CBM Intervention, the PARIHS framework it was built on has only been
partially validated for construct validity (Wallin, Estabrooks, Midodzi, &
Cummings, 2006). Also, the small sample size limited power to detect
significant associations between care manager characteristics and
questionnaire responses. Another limitation was the absence of measures
that could have offered greater explanation of the findings, such as the

159
specific amount of time facilitation team members spent at each site and its
relationship to care manager perception of the implementation process. A
second example is that 48% of care managers reported contacting the
physician’s office in the medication management intervention. However,
data from facilitation team members reported that pharmacists, with the
support of care managers, completed the physician contacts. These
conflicting results between care managers and the facilitation team suggest
both differing perceptions on implementation and how collaboration
partnerships function in real world environments.
Conclusion
In spite of the limitations, results demonstrate the intrinsic value of
ascertaining care managers perceptions, which are critically important to
understand when implementing evidence-based practices into an existing
program. Lessons learned in this study speak to the challenges and
tensions involved in translation work at the point where implementation
comes alive. These lessons mirror the larger contributions and next steps
discussed in Chapter VI.

160
VI: DISCUSSION
This dissertation used the CBM Intervention to evaluate a Type II
translation of an evidence-based practice originating from a randomized
clinical trial into the “in vivo” environment of community agencies. Chapter I
described the guiding theories, frameworks, and government initiatives
involved in shaping the future of translational research. Chapter II applied
one of these models, the PARIHS framework, to depict key evidentiary,
contextual, and facilitation elements of the CBM Intervention that served as
the programmatic foundation for the dissertation. The CBM Intervention
utilized the a medication management model developed in a home
healthcare setting (“evidence”) and translated this practice into a Medi-Cal
waiver care management program (“context”) with the active support from a
leadership and implementation team housed at one of the sites
(“facilitation”).
Chapter III substantiated the need for this intervention by reporting
prevalence data that 49% of the MSSP sample had a potential medication
problem based on four problem criteria, a rate that rivaled other studies
using Medicaid waiver participants (Golden et al., 1999; Piecoro et al.,
2000; Rigler et al., 2005). Once potential problems were confirmed through
a focused pharmacy review process, the prevalence rate dropped to 29%.
This was a reduction of 41% although the overall prevalence rate remained
high and reflected a more accurate description of this population’s

161
vulnerability to adverse drug events. Predictors of medication problems
included advanced age, Caucasian, English-speaking, new MSSP
enrollees, institutional-level health service utilization, and greater number of
medications.
Those who were confirmed to have a medication problem received a
pharmacist-driven intervention to resolve the identified problem. Chapter IV
reported that 61% of those receiving the intervention had a successful
medication change, which was comparable to improvement rates achieved
in the home healthcare randomized clinical trial. Participants with multiple
problems also had reasonable success of having a least one of their
problems addressed. Therapeutic duplication problems had the best
resolution rate at 62% and the lowest rate was for psychotropic medication
problems related to falls at 43%. A key translation issue at this stage was
the shifting of responsibility for the intervention to the pharmacist rather
than the primary care manager as originally intended. Results suggest that
even with this and other adaptations needed to implement this practice in
care management, the translation process was successful in garnering
equally effective results of reducing medication-related problems for this
population.
Chapter V described feedback from nurse and social work care
managers on implementing an evidence-based practice into an existing
service setting using a questionnaire developed specifically for this project.

162
Nurses and social workers reported differing perspectives on professional
norms related to medication management processes. Compared to social
workers, nurses agreed that medication management was within their
scope of practice and that they contacted prescribing physicians to inform
of medication problems. These differences were mirrored by those with
more experience with implementing an evidence-based practice.
Furthermore, care managers located at Site #1 where the facilitation team
was housed expressed more positive responses to the benefits of training,
access to the facilitation team, and the overall benefits of the CBM
Intervention.
Chapters III, IV, and V discussed lessons learned, limitations, and policy
implications specific to each content area covered. This chapter seeks to
capture important findings at the macro level by clarifying lessons learned
and future research issues for both the CBM Intervention specifically and
the Type II translation process. Lessons learned illustrate the
interconnections between the three PARIHS concepts of evidence, context,
and facilitation.
Medication Screening Process
Problems with the computerized medication screening system led the
facilitation team to abandon this technology in the intervention stage and
complete hand screenings to gather prevalence data on the sample. Hand
screens were performed at all three sites, and about half of those at Site #3

163
were also screened through the computer system to beta-test its reliability.
This problem contributed to several implementation challenges. First, costs
associated with the intervention likely increased due to the increased
pharmacist time to complete the hand screening. It also negatively
impacted care manager enthusiasm for the project as a whole because they
were further removed from the intervention in the initial stages and viewed
medication management as the pharmacist’s responsibility. This resulted in
a lost opportunity to bolster care manager interest in translation, as
previous research has shown that computerized systems to assist with
clinical decision-making has improved health professionals’ use of
evidence-based practice (Gira et al., 2004; McKenna et al., 2004).
Professional Differences Affecting Implementation
Role specialization, not transdisciplinarity, was the care managers’
general operating principle in the CBM Intervention. A salient example was
in the collection of assessment measures needed to identify cardiovascular
medication problems. Although assessment was a routine process in care
management, the collection of clinical indicators was highly dependent on
the care manager’s professional type and the timing of the assessment
visit. Nurses generally gathered more medically-related assessment items
such as blood pressure and pulse readings that were key clinical indicators
to determine cardiovascular medication problems. Nurses visited
participants assigned to their caseload quarterly, but visited all other

164
participants only at the initial assessment period and at a reassessment
visit, which in some cases was up to 18 months after enrollment. All other
quarterly visits for these participants were conducted by social work care
managers. Social work care manager resisted taking blood pressure and
pulse readings using standard instruments, stating these measures were
outside of their scope of practice. They attempted to gather these readings
from participant report or adult day health care providers when available.
Although electronic wrist devices could have improved the collection of this
information by non-medical staff, social work care managers did not have
regular access to these instruments and pharmacy staff also questioned the
validity of readings through this mechanism. Given that social workers
were the bulk of the MSSP workforce due to Medi-Cal waiver staffing
guidelines, few participants were able to be properly screened for
cardiovascular problems.
Role specialization was also apparent in the intervention phase of the
project as care managers, and social workers specifically, resisted
contacting the physician about medication-related issues. Care managers
were willing to participate in a consultative role to pharmacists and assist
with assessment and treatment planning. Ownership of the medication
management role, however, remained with the pharmacist and was not
successfully transferred to the care managers. This outcome corroborates
current knowledge on successful medication management interventions as

165
being primarily nurse or pharmacist driven regardless of the organizational
setting (Gilbert et al., 2002; Hanlon et al., 1996; Krska et al., 2001; Rhoads
& Thai, 2003; Sellors et al., 2003; Sorensen et al., 2004).
Although proposed in the original implementation proposal, the plan for
care managers to take a fully active role in the CBM Intervention has
drawbacks. First medication management may be perceived as too far into
the health care sphere to be adequately implemented in care management
practice without direct support from pharmacy professionals. Pharmacists
were critical partners as they were able to delineate actual medication
problems from false positives based on their particular expertise. The
demarcation of “confirmed” problems from those deemed only as “potential”
problems allowed for a 41% reduction in participants appropriate for the
intervention phase. If care managers assumed full responsibility of the
intervention with the use of a functional computerized system and
communicated all potential problems to the physician’s office, this
information could have overburdened the physician and negatively affected
their relationship with a key ally in the fight against suboptimal prescribing.
However, the availability of pharmacists with expertise in geriatric
pharmacotherapy may be difficult to access outside of a metropolitan area,
thereby creating health service disparities for those in typically underserved
areas. High cost of consultant pharmacy services may also present a
barrier to full implementation of the model. In light of these challenges,

166
some potential solutions for care management programs seeking to
implement medication management without direct access to a consultant
pharmacist are listed below:
• At assessment, gather necessary information about medication and
clinical indicators as suggested by the problem protocols and use a
computerized screening system if available;
• Train a nurse care manager to review participant assessment
information for potential problems using the hand screen or
computerized system;
• For those with a potential problem, provide participant with alert
information and refer to his/her community pharmacist where
medications are dispensed for a medication review;
• If a participant is enrolled in other community-based programs with
health professionals (adult day health care or home healthcare
services), alert these professionals of potential problems and
suggest further medication review;
• Have trained nurse educate participants about methods to decrease
medication problems such as taking all medications to all physician
visits, utilize only one pharmacy, ensure that a primary caregiver is
informed about all prescribed medications, and discard
old/discontinued medications; and

167
• Have nurse trained in medication review contact physician’s office
with information of potential medication problem with disclaimer that
alert is based on protocols and could represent a false positive
reading.
Some of these strategies might be viewed as passing the responsibility
for medication management to another service provider to do a
comprehensive review. However by involving additional partners, MSSP
care managers could promote a broader network of interagency
collaboration, which is an effective way to channel existing resource
capacities into productive partnerships and to help older adults overcome
the challenges of medication-related problems (Alkema, Shannon, & Wilber,
2003).
Facilitation Lessons
A key lesson related to embedding the CBM Intervention into MSSP
practice included balancing the need to maintain fidelity to the original home
healthcare study and adapting the program sufficiently to promote
sustainability. This tension was expressed by one facilitation team member
who questioned whether the CBM Intervention was a “replication” project
focused exclusively on fidelity or a “demonstration” project leaning more
toward adaptation to the MSSP environment. As a phase on the translation
pathway, the CBM Intervention represented both processes and it was the
job of the facilitation team to balance these tensions accordingly. One way

168
the facilitation team accomplished this goal was by supporting
communication flow and knowledge transfer across the sites to promote the
most uniform implementation possible. For example, Site #1 developed a
method on how to word a treatment plan goal on medication management
in order to meet state survey requirements. The facilitation team
incorporated this information into training materials for the other sites and
assisted with treatment plan development during case consultation
meetings. Another method used was that the two consultant pharmacists
met regularly and reviewed each other’s medication intervention
recommendations to consistently implement the Home Health Criteria
protocols. Data collection teams were also supervised by a project
associate who checked errors or inconsistencies in reporting by the
pharmacists.
Feedback from care managers identified the need for ongoing training to
further embed and sustain the CBM Intervention. It was suggested that in
addition to participation in staff and care coordination meetings, the
facilitation team provide a refresher training on medications and the
intervention process to help overcome implementation challenges and to
train new staff members. This suggestion was critical given that all three
sites averaged an 85% turnover rate during the project period. There are a
few ways to create an ongoing training program for new and current staff.
Each quarter participating sites could organize a new employee training

169
session that would address various topics related to MSSP guidelines and
interventions including medication management. For current care
managers needing a refresher, facilitation team members could spend a
brief period of time during each care coordination meeting attended
discussing necessary elements of the implementation process.
Lessons from External Factors
Although beyond the control of the CBM Intervention facilitation team,
some evidentiary and contextual factors had a profound impact on
implementation. First, Chapter IV reported the removal of Cox-2 Inhibitors
from the market in the middle of the intervention period. This change
greatly reduced solutions available for participants with a NSAIDs problem.
In the short term, the facilitation team consulted with the project’s expert
advisory panel and developed a revised NSAIDs intervention protocol.
However, this situation underscores a larger question of how to implement
interventions that are predicated on a fluid evidence base. The third phase
of translation, implementation research, attempts to address this issue
through a continuous quality improvement paradigm. This process can be
costly and to be successful, it requires a strong commitment from program
administrators to continue refining a specific intervention, such as
medication management, based on advances in basic research.
Implementation in a Medi-Cal waiver context also highlighted difficulties
of incorporating an evidence-based intervention into a current clinical

170
practice routine that was housed in an existing organizational structure and
bound by state and federal regulations. For example, prior to the
pharmacist’s intervention on behalf of participants with medication
problems, care managers needed to include medication management in
MSSP care planning documentation. Given that this intervention strategy
was a new practice in MSSP, care managers were first required to gain
approval from state auditors for their documentation of medication
management activities in the care plan. Another example was when state-
level administrators piloted a new nursing assessment during the
intervention, causing nurse care managers to accommodate a new
paperwork for a trial period while trying to incorporate medication
management processes into their practice routine. In addition, Site #2 was
acquired by MSSP A in 2003 due to poor site surveys under a previous
organization. This led to quarterly state surveys and pushed back their
implementation schedule nearly five months.
Challenges faced in MSSP differ greatly from implementation of “stand-
alone” evidence-based health promotion interventions that are not
dependent on a pre-existing clinical process, such as Lorig’s Chronic
Disease Self Management Program (Lorig et al., 1999). In this type of
program, staff is often hired and trained to administer the program as
directed specific scripts and protocols, and to monitor participant outcomes
based on existing tools. Additionally, evaluation standards devised to

171
measure implementation fidelity in the translation process were developed
using these “stand-alone” programs (Bellg et al., 2004). These and similar
standards do not account for the complexities of translation in existing
contexts, and further work is needed in this area.
A final external factor was that all three sites were implementing another
evidence-based practice (physical activity regimen) at the same time as the
CBM Intervention. The facilitation team reported that care managers
expressed concern about adopting both new practices at the same time and
often felt the need to “choose” which intervention was most important for
which participant. This issue highlights the level of change an organization
can sustain at any given time, especially given that human service
organizations have the tendency to be rigid and have fixed informal cultures
(Hasenfeld, 1983).
Lessons in Translation and Future Trajectories
This is really hard… ~ quote from facilitation team member on
translation.
The CBM Intervention exemplifies how translation is the operational
nexus of implementation theory given the iterative interaction between top-
down and bottom-up processes. Engaging a top-down process, the CBM
Intervention employed a research-tested evidence-based practice and
struggled to maintain fidelity in light of significant contextual hurdles. In
order to manage the tension between the evidence and context, facilitation

172
staff focused on bottom-up processes to sufficiently adapt it to the MSSP
environment in order to promote sustainability. The CBM Intervention also
represented a bottom up process given that medication management was
implemented through pilot funding and did not originate as a federal or state
mandate.
Findings and lessons learned from the CBM Intervention suggest that
three PARIHS factors of evidence, context, and facilitation affected the
implementation process, but not in equal doses. Instead, it is likely that
facilitation mediated the relationship between evidence and context on
implementation success (Figure 6.1). The facilitation team needed to
compensate for challenges between the evidence and context that had a
direct impact on the balance between implementation fidelity and
adaptation. This likely occurred because evidentiary and contextual
elements in the CBM Intervention were more fixed and immovable than
facilitation, such as the computerized screening system not interfacing with
current MSSP database. The facilitation team circumvented this problem
through a lower technology solution, allowing the implementation process to
continue. The facilitation team addressed many challenges of
implementing an evidence-based practice within a bureaucratic program
that used existing assessment tools and was bound by state and federal
regulations. Without this concentrated effort, it is unlikely that CBM

173
Intervention would have had nearly the success rate of both identifying and
resolving medication problems for functionally-impaired older adults.

The strategy to address medication problems in community-based aging
services has evolved down a long and winding pathway from its beginnings
in basic pharmacological and epidemiology research to its first round of
dissemination in an applied setting (Figure 6.2). Next steps for the CBM
Intervention involve expansion to eight additional Medicaid waiver care
management programs in three states The organization that operates
MSSP A received a $1.7 million grant from the John A. Hartford Foundation
to address challenges that occurred in the CBM Intervention and to
implement this translation phase. Although the medication problems will
remain the same (excluding NSAIDs), the discussion has arisen about the
need the include medication problems that are reasonable to screen in a
care management setting without excessive burden on staff or collaborative
providers. The facilitation team has discussed changing the problem set to

174
be better associated with skills and knowledge base of social workers,
considering under-treated depression and inadequate pain control as two
potential areas. A change in problem types would require an expert panel
to convene and delineate a specific protocol for identifying and addressing
these new problem types. If this occurs, it would demonstrate a natural
feedback loop in the translational research process by redefining the clinical
intervention prior to widespread implementation. At this point, however, the
translation process into the new care management programs will use the
same CBM evidence base making alterations in the implementation
strategy as needed.

175
The lead organization for MSSP B has selected a different route for
maintaining the medication management intervention. Without the support
of additional pilot funding and no clear-cut reimbursement stream for
consultant pharmacy services, Site #3 chose to adopt a more generic
medication management protocol that offers care managers a decision tree
on when to involve other health professionals in addressing potential
medication-related problems (see Appendix M). They plan to use a
different clinical approach to medication management by identifying
problematic symptoms that might be related to a medication problem (e.g.,
confusion) and then soliciting the assistance of a health professional.
Given that the organization that operates MSSP B is a hospital, Site #3
hopes to utilize healthcare resources within the larger staff network to assist
with medication screening.
Strengths and Limitations of the Dissertation
This study had several strengths that contribute to a stronger knowledge
base of Type II translational research. First, the CBM Intervention
successfully demonstrated the translation of screening and intervention
protocols developed in a randomized clinical trial to an existing care
management program inundated with federal and state regulations. As
such, this translation improved the external validity of a medication
management intervention that had previously shown good internal validity in
the research environment by demonstrating its adaptability to a “real-world”

176
setting. The impact of this strength cannot be underestimated, as the
success of a Type II translation fundamentally relies on the thoughtful
balance between implementation fidelity to the original study and
appropriate adaptation to the new program setting.
A core strength was the study’s use of mixed methods to evaluate the
process and outcomes of the medication management intervention
implementation. This dissertation incorporated rigorous quantitative
analysis with in-depth qualitative methods that collectively provided a
complete picture of successes and challenges involved in a Type II
translation. Benefits of a mixed methods approach were best exemplified
by the inclusion of care manager perspectives on the implementation
process. Several lessons learned were gained from both non-parametric
statistical analyses and direct staff feedback, including opportunities and
barriers to implementation of evidence-based practices and the steadfast
nature of role specialization in health services. These lessons helped
shape the intervention while in progress and helped guide future
dissemination strategies to new adopting programs.
Given the nature of a Type II translational research study, there were
also a few limitations to note for this dissertation. First, there were of
number of participant-, organization-, and system-level variables that were
not available for the various analyses. To promote a sustainable
intervention, the project used existing MSSP assessment tools

177
supplemented by new forms developed by the researcher that were specific
to medication management. In addition, the three MSSP sites utilized the
care management database to differing degrees for recording assessment
and treatment documentation. However, all sites were required by the state
to use the database for billing purposes, which included participant socio-
demographic information. Therefore additional variables on health status,
MSSP treatment processes, or provider information were not available.
Also, the PARIHS framework is a relatively new way of conceptualizing
the implementation of research findings into practice. Its concepts and sub-
criteria of evidence, context, and facilitation have not been thoroughly
tested for construct validity and there are no current tools to quantify their
impact on translational research. However, as an emerging field of study,
the PARIHS framework offers a conceptual mechanism for understanding
at least some of the key components that shape translation and affect
implementation outcomes.
Future Research and Conclusion
Building on this dissertation, future research should focus on three
areas. First, this dissertation evaluated an integrative treatment model
using nurses, social workers, and consultant pharmacists in community
agencies to address one of the more pressing health concerns facing
chronically ill older adults–medication problems and their potential adverse
effects. This work is yet another example of how integrative treatment

178
models that bridge medical and social care services have shown a positive
impact on health and well-being, especially when targeting those most at
risk for negative outcomes (Albert, Simone, Brassard, Stern, & Mayeux,
2005; Alkema, Wilber, Shannon, & Allen, 2007; Chatterji, Burstein, Kidder,
& White, 1998; E. A. Miller & Weissert, 2000; Shapiro & Taylor, 2002).
Future research should focus on continued development, evaluation, and
dissemination of integrative care models and interventions that support
functionally-impaired, older adults in community-based settings. Such
models include employing innovative teams to support transitions from
institutional care settings to the home environment, greater use of health
technology for those with decreased access to health care services (e.g.,
rural elders), and expansion of successful models of care to include older
adults not eligible for Medicaid waiver services where many of these
models are operational.
Second, translational research should focus on further development and
refinement of conceptual frameworks and technical assistance products
that can better guide the effective implementation of evidence-based
practices into community settings. Specifically translational researchers
should evaluate the utility of the PARIHS framework by creating a tool that
can measure the individual power and associations within and across the
three concepts. Continued work in this area will support a greater

179
understanding of the processes and constraints so that additional
connections can be established between research and practice domains.
Third, medication problems and their associated effects should be
continually evaluated as new medications arrive in the marketplace and
greater access to prescriptions become available to older adults through
Medicare Part D. This dissertation is timely given that data collection for
this study was complete when Medicare-funded prescription drug coverage
through Part D initiated in January 2006. It is likely that increased access to
prescription drugs will also increase the potential for medication problems
and adverse drug events. Future research should consider systemic
problems and solutions related to medication management, such as
improvement in health technology to maintain a current patient medical
record and address potential medication problems prior to dispensing, and
the impact of greater utilization of consultant pharmacist services in a
variety of institutional and community-level agencies. Larger health system
research should evaluate the longitudinal effects of increasing prescription
drug coverage to older adults through Medicare Part D and differential
effects of commercial Prescription Drug Plans to reduce the prevalence of
medication problems and adverse drug events through internal medication
review systems.
Implementation of the CBM Intervention characterized the benefits and
challenges for projects engaging in Type II translation. Although this work

180
is extremely challenging, Type II translation is the frontier of advancing
research findings into health and social service delivery with the ultimate
goal of improving the lives of aging individuals.

181
BIBLIOGRAPHY
Aarons, G. A. (2004). Mental health provider attitudes toward adoption of
evidence-based practice: The Evidence-Based Practice Attitude
Scale (EBPAS). Mental Health Services Research, 6(2), 61-74.

Administration on Aging. (2004). Evidence Based Disease Prevention
Grants Program. Retrieved December 5, 2005, from
http://www.aoa.gov/prof/evidence/evidence.asp

Administration on Aging. (2006). AoA Grant Programs: Empowering Older
People To Take More Control of their Health Through Evidence-
Based Prevention Programs. Retrieved September 5, 2006, from
http://www.aoa.gov/doingbus/fundopp/fundopp.asp

Agency for Healthcare Research and Quality. (1998). AHCPR Strategic
Plan. Retrieved May 4, 2005, from
http://www.ahrq.gov/about/stratpln.htm

Albert, S. M., Simone, B., Brassard, A., Stern, Y., & Mayeux, R. (2005).
Medicaid home care services and survival in New York City. The
Gerontologist, 45(5), 609-616.

Alkema, G. E., & Frey, D. (2006). Implications of translating research into
practice: A medication management intervention. Home Health Care
Services Quarterly, 25(1/2), 33-54.

Alkema, G. E., Shannon, G. R., & Wilber, K. H. (2003). Using interagency
collaboration to serve older adults with chronic care needs: The Care
Advocate Program. Family & Community Health, 26(3), 221-229.

Alkema, G. E., Wilber, K. H., Shannon, G. R., & Allen, D. (2007). Reduced
mortality: The unexpected impact of a telephone-based care
management intervention for older adults in managed care. Health
Services Research.

American Nurses Association. (1994). Position Statement: Education for
Participation in Nursing Research. Silver Spring, MD.

Aparasu, R., & Mort, JR. (2000). Inappropriate prescribing for the elderly:
Beers criteria-based review. Annals of Pharmacotherapy, 34(3), 338-
346.

182
ASHP Continuity of Care Task Force. (2005). Continuity of care in
medication management: Review of issues and considerations for
pharmacy. American Journal of Health System Pharmacy, 62, 1714-
1720.

Atkins, D., Best, D., & Briss, P. (2004). Grading quality of evidence and
strength of recommendations. British Medical Journal, 328(7454),
1490-1490.

Azoulay, L., Zargarzadeh, A., Salahshouri, Z., Oraichi, D., & Berard, A.
(2005). Inappropriate medication prescribing in community-dwelling
elderly people living in Iran. European Journal of Clinical
Pharmacology, 61(12), 913-919.

Backer, T. E. (2001). Finding the Balance: Program Fidelity and Adaptation
in Substance Abuse Prevention. Rockville, MD: Center for
Substance Abuse Prevention.

Beers, M. H. (1997). Explicit criteria for determining potentially inappropriate
medication use by the elderly. Archives of Internal Medicine, 157,
1531–1536.

Bellg, A. J., Borrelli, B., Barbara, R., Hecht, J., Minicucci, D. S., Ory, M.,
Ogedegbe, G., Orwig, D., Ernst, D., & Susan, C. (2004). Enhancing
treatment fidelity in health behavior change studies: Best practices
and recommendations from the NIH behavior change consortium.
Health Psychology, 23(5), 443-451.

Boling, P. A. (2002). Strategic use of home care pharmacy consultant may
be worthwhile. Journal of the American Geriatrics Society, 50, 1597-
1598.

Bradley, E. H., Schlesinger, M., Webster, T. R., Baker, D., & Inouye, S. K.
(2004). Translating research into clinical practice: Making change
happen. Journal of the American Geriatrics Society, 52(11), 1875-
1882.

Brown, N. J., Griffin, M. R., Ray, W. A., Meredith, S., Beers, M. H., Marren,
J., Robles, M., Stergachis, A., Wood, A. J., & Avorn, J. (1998). A
model for improving medication use in home health care patients.
Journal of the American Pharmaceutical Association, 38(6), 696-702.

183
Bryant, L. L., Altpeter, M., & Whitelaw, N. A. (2006). Evaluation of health
promotion programs for older adults: An introduction. Journal of
Applied Gerontology, 25(3), 197-213.

Bureau of Labor Statistics. (2004). Occupational Outlook Handbook:
Pharmacists. Retrieved August 23, 2006, from
http://www.bls.gov/oco/ocos079.htm

California Board of Behavioral Sciences. (2006). Statues and Regulations
Relating to the Practice of Marriage and Family Therapy, Licensed
Clinical Social Work, and Licensed Educational Psychology.
Sacramento, CA: California Department of Consumer Affairs.

California Board of Registered Nursing. (1997). Nursing Practice Act.
Sacramento, CA: California Department of Consumer Affairs -
Business and Professionals Code.

California Department of Aging. (2006). Multipurpose Senior Services
Program. Retrieved June 15, 2006, from
http://www.aging.state.ca.us/html/programs/mssp.html

Castro, F. G., Barrera, M., & Martinez, C. (2004). The cultural adaptation of
prevention interventions: Resolving tensions between fidelity and fit.
Prevention Science, 5(1), 41-45.

Center for Drug Evaluation and Research. (2005). Cox-2 Selective and
Non-Selective Non-Steroidal Anti-Inflammatory Drugs. Retrieved
September 25, 2006, from
http://www.fda.gov/cder/drug/infopage/cox2/

Centers for Medicare and Medicaid Services. (2005). Medicaid at a Glance.
Washington, DC: Department of Health and Human Services.

Chatterji, P., Burstein, N. R., Kidder, D., & White, A. (1998). Evaluation of
the Program of All-Inclusive Care for the Elderly (PACE)
Demonstration: The Impact of PACE on Participant Outcomes.
(Final report prepared for HCFA). Cambridge, MA: Abt Associates.

Chobanian, A. V., Bakris, G. L., Black, H. R., Cushman, W. C., Green, L. A.,
Izzo, J. L., Jr., Jones, D. W., Materson, B. J., Oparil, S., Wright, J. T.,
Jr., & Roccella, E. J. (2003). Seventh report of the Joint National
Committee on Prevention, Detection, Evaluation, and Treatment of
High Blood Pressure. Hypertension, 42(6), 1206-1252.

184
Clancy, C. M. (2004). AHRQ's FY 2005 budget request: New mission, new
vision. Health Services Research, 39(3), xi-xviii.

Clancy, C. M., Slutsky, J. R., & Patton, L. T. (2004). Evidence-based health
care 2004: AHRQ moves research to translation and implementation.
Health Services Research, 39(5), xv-xxiii.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences
(2nd ed.). Hillsdale, NJ: Lawrence Earlbaum Associates.

Coleman, E. A. (2003). Falling through the cracks: Challenges and
opportunities for improving transitional care for persons with
continuous complex care needs. Journal of the American Geriatrics
Society, 51(4), 549-555.

Commonwealth Fund. (2004). The Commonwealth Fund Health Care
Opinion Leaders Survey. New York.

Curtis, L. H., Ostbye, T., Sendersky, V., Hutchison, S., Dans, P. E., Wright,
A., Woosley, R. L., & Schulman, K. A. (2004). Inappropriate
prescribing for elderly Americans in a large outpatient population.
Archives of Internal Medicine, 9(23), 1621-1625.

Department of Health and Human Services. (2000). Healthy People 2010
(Objective 17.3). Washington, DC.

Dhalla, I. A., Anderson, G. M., Mamdani, M. M., Bronskill, S. E., Sykora, K.,
& Rochon, P. A. (2002). Inappropriate prescribing before and after
nursing home admission. Journal of the American Geriatrics Society,
50(6), 995-1000.

Dusenbury, L., & Hansen, W. B. (2004). Pursuing the course from research
to practice. Prevention Science, 5(1), 55-59.

Eans, T. L. (2000). New HCFA drug-prescribing criteria for nursing homes
and suggested alternate prescribing to avoid care deficiencies.
Annals of Long-Term Care, 8, 58-62.

Eccles, M., & Mittman, B. (2006). Welcome to Implementation Science.
Implementation Science, 1(1), 1.

Elliott, D. S., & Mihalic, S. (2004). Issues in disseminating and replicating
effective prevention programs. Prevention Science, 5(1), 47-53.

185
Farquhar, C. M., Stryer, D., & Slutsky, J. R. (2002). Translating research
into practice: The future ahead. International Journal for Quality in
Health Care, 14(3), 233-249.

Feldman, P. H., & Kane, R. L. (2003). Strengthening research to improve
the practice and management of long-term care. Milbank Quarterly,
81(2), 179-220.

Fialova, D., Topinkova, E., Gambassi, G., Finne-Soveri, H., Jonsson, P. V.,
Carpenter, I., Schroll, M., Onder, G., Sorbye, L. W., Wagner, C.,
Reissigova, J., Bernabei, R., & for the AdHOC Project Research
Group. (2005). Potentially inappropriate medication use among
elderly home care patients in Europe. Journal of the American
Medical Association, 293(11), 1348-1358.

Fick, D. M., Cooper, J. W., Wade, W. E., Waller, J. L., Maclean, J. R., &
Beers, M. H. (2003). Updating the Beers Criteria for potentially
inappropriate medication use in older adults: Results of a US
consensus panel of experts. Archives of Internal Medicine, 163(22),
2716-2724.

Fick, D. M., Waller, J. L., Maclean, J. R., Heuvel, R. V., Tadlock, J. G.,
Gottlieb, M., & Cangialose, C. B. (2001). Potentially inappropriate
medication use in a Medicare managed care population: Association
with higher costs and utilization. Journal of Managed Care
Pharmacy, 7, 407-413.

Field, T. S., Gilman, B. H., Subramanian, S., Fuller, J. C., Bates, D. W., &
Gurwitz, J. H. (2005). The costs associated with adverse drug events
among older adults in the ambulatory setting. Medical Care, 43(12),
1171-1176.

Fink, R., Thompson, C. J., & Bonnes, D. (2005). Overcoming barriers and
promoting the use of research in practice. Journal of Nursing
Administration, 35(3), 121-129.

Frey, D., & Rahman, A. (2003). Medication management: An evidence-
based model that decreases adverse events. Home Healthcare
Nurse, 21(6), 404-412.

Fu, A. Z., Liu, G. G., & Christensen, D. B. (2004). Inappropriate medication
use and health outcomes in the elderly. Journal of the American
Geriatrics Society, 52(11), 1934-1939.

186
Gambrill, E. (2006). Evidence-based practice and policy: Choices ahead.
Research on Social Work Practice, 16(4), 338-357.

Gary, T. L., Hill-Briggs, F., Batts-Turner, M., & Brancati, F. L. (2005).
Translational research principles of an effectiveness trial for diabetes
care in an urban African American population. The Diabetes
Educator, 31(6), 880-889.

Gerrish, K., & Clayton, J. (2004). Promoting evidence-based practice: An
organizational approach. Journal of Nursing Management, 12(2),
114-123.

Gibbs, L., & Gambrill, E. (2002). Evidence-based practice:
Counterarguments to objections. Research on Social Work Practice,
12(3), 452-476.

Gilbert, A. L., Roughead, E. E., Beilby, J., Mott, K., & Barratt, J. D. (2002).
Collaborative medication management services: Improving patient
care. Medical Journal of Australia, 177(4), 189-192.

Ginexi, E. M., & Hilton, T. F. (2006). What's next for translation research?
Evaluation and the Health Professions, 29(3), 334-347.

Gira, E. C., Kessler, M. L., & Poertner, J. (2004). Influencing social workers
to use research evidence in practice: Lessons from medicine and the
allied health professions. Research on Social Work Practice, 14(2),
68-79.

Glasgow, R. E. (2006). RE-AIMing research for application: Ways to
improve evidence for family medicine. Journal of the American Board
of Family Medicine, 19(1), 11-19.

Glasgow, R. E., Lichtenstein, E. P., & Marcus, A. C. (2003). Why don't we
see more translation of health promotion research to practice?
Rethinking the efficacy-to-effectiveness transition. American Journal
of Public Health, 93(8), 1261-1267.

Glasgow, R. E., Nutting, P. A., King, D. K., Nelson, C. C., Cutter, G., Gaglio,
B., Rahm, A. K., Whitesides, H., & Amthauer, H. (2004). A practical
randomized trial to improve diabetes care. Journal of General
Internal Medicine, 19(12), 1167-1174.

187
Glasgow, R. E., Vogt, T. M., & Boles, S. M. (1999). Evaluating the public
health impact of health promotion interventions: The RE-AIM
Framework. American Journal of Public Health, 89(9), 1322-1327.

Glasgow, R. E., Whitlock, E., Eakin, E., & Lichenstein, E. (2000). A brief
smoking cessation intervention for women in low-income Planned
Parenthood clinics. American Journal of Public Health, 90, 786-789.

Golden, A. G., Preston, R. A., Barnett, S. D., Llorente, M. M. D., Hamdan,
K. M. D., & Silverman, M. A. (1999). Inappropriate medication
prescribing in homebound older adults. Journal of the American
Geriatrics Society, 47(8), 948-953.

Gonzales, J. J., Ringeisen, H. L., & Chambers, D. A. (2002). The tangled
and thorny path of science to practice: Tensions in interpreting and
applying "evidence". Clinical Psychology: Science and Practice, 9(2),
204-209.

Goulding, M. R. (2004). Inappropriate medication prescribing for elderly
ambulatory care patients. Archives of Internal Medicine, 164(3), 305-
312.

Grimshaw, J. M., Shirran, L., Thomas, R., Mowatt, G., Fraser, C., Bero, L.,
Grilli, R., Harvey, E., Oxman, A., & O'Brien, M. A. (2001). Changing
provider behavior: An overview of systematic reviews of
interventions. Medical Care, 39(8), II-2-II-45.

Grol, R. P. (2001). Successes and failures in the implementation of
evidence-based guidelines for clinical practice. Medical Care, 39(8),
II-46-II-54.

Gurwitz, J. H. (2004). Polypharmacy: A new paradigm for quality drug
therapy in the elderly? Archives of Internal Medicine, 164(18), 1957-
1959.

Gurwitz, J. H., Field, T. S., Harrold, L. R., Rothschild, J., Debellis, K., Seger,
A. C., Cadoret, C., Fish, L. S., Garber, L., Kelleher, M., & Bates, D.
W. (2003). Incidence and preventability of adverse drug events
among older persons in the ambulatory setting. Journal of the
American Medical Association, 289(9), 1107-1116.

188
Hagedorn, H., Hogan, M., Smith, J. L., Bowman, C., Curran, G. M.,
Espadas, D., Kimmel, B., Kochevar, L., Legro, M. W., & Sales, A. E.
(2006). Lessons learned about implementing research evidence into
clinical practice. Experiences from VA QUERI. Journal of General
Internal Medicine, 21(s2), S21-S24.

Hajjar, E. R., Hanlon, J. T., Sloane, R. J., Lindblad, C. I., Pieper, C. F.,
Ruby, C. M., Branch, L. C., & Schmader, K. E. (2005). Unnecessary
drug use in frail older people at hospital discharge. Journal of the
American Geriatrics Society, 53(9), 1518-1523.

Han, B., Remsburg, R. E., Lubitz, J., & Goulding, M. (2004). Payment
source and length of use among home health agency discharges.
Medical Care, 42(11), 1081-1090.

Hanlon, J. T., Artz, M. B., Pieper, C. F., Lindblad, C. I., Sloane, R. J., Ruby,
C. M., & Schmader, K. E. (2004). Inappropriate medication use
among frail elderly inpatients. Annals of Pharmacotherapy, 38(1), 9-
14.

Hanlon, J. T., Schmader, K. E., Boult, C., Artz, M. B., Gross, C. R.,
Fillenbaum, G. G., Ruby, C. M., & Garrard, J. (2002). Use of
inappropriate prescription drugs by older people. Journal of the
American Geriatrics Society, 50(1), 26-34.

Hanlon, J. T., Schmader, K. E., Ruby, C. M., & Weinberger, M. (2001).
Suboptimal prescribing in older inpatients and outpatients. Journal of
the American Geriatrics Society, 49(2), 200-209.

Hanlon, J. T., Weinberger, M., Samsa, G. P., Schmader, K. E., Uttech, K.
M., Lewis, I. K., Cowper, P. A., Landsman, P. B., Cohen, H. J., &
Feussner, J. R. (1996). A randomized, controlled trial of a clinical
pharmacist intervention to improve inappropriate prescribing in
elderly outpatients with polypharmacy. American Journal of
Medicine, 100(4), 428-437.

Harrington, C., Carrillo, H., Wellin, V., Miller, N., & LeBlanc, A. (2000).
Predicting state Medicaid home and community based waiver
participants and expenditures, 1992-1997. The Gerontologist, 40(6),
673-686.

189
Harvey, G., Loftus-Hills, A., Rycroft-Malone, J., Titchen, A., Kitson, A.,
McCormack, B., & Seers, K. (2002). Getting evidence into practice:
the role and function of facilitation. Journal of Advanced Nursing,
37(6), 577-588.

Hasenfeld, Y. (1983). Human service organizations. Englewood Cliffs, NJ:
Prentice Hall.

Hiss, R. G., Green, L. W., & Ottoson, J. M. (2004). Fundamental issues in
translational research. Paper presented at the From Clinical Trials to
Community: The Science of Translating Diabetes and Obesity
Research, Bethesda, MD.

Hjern, B. (1982). Implementation research - the link gone missing. Journal
of Public Policy, 2(3), 301-308.

Hughes, S. L., & Pittard, M. A. (2005). The continuum of long-term care. In
C. Evashwick (Ed.), (2nd ed.). Albany, NY: Delmar.

Improved Clinical Effectiveness through Behavioural Research Group
(ICEBeRG). (2006). Designing theoretically-informed implementation
interventions. Implementation Science, 1(1), 4.

Inouye, S. K., van Dyck, C., Alessi, C., Balkin, S., Siegal, A., & Horwitz, R.
(1990). Clarifying confusion: the confusion assessment method. A
new method for detection of delirium. Annals of Internal Medicine,
113(12), 941-948.

Institute of Medicine. (2000). To err is human: Building a safer health
system. Washington, D.C.: National Academy Press.

Institute of Medicine. (2001). Crossing the quality chasm: A new health
system for the 21st century. Washington, D.C: National Academy
Press.

Institute of Medicine. (2007). Preventing medication errors. Washington,
DC: National Academies Press.

JCAHO. (2006a). National Patient Safety Goals. Retrieved April 28, 2006,
from
http://www.jointcommission.org/PatientSafety/NationalPatientSafety
Goals/

190
JCAHO. (2006b). Sentinel Event ALERT: Using Medication Reconciliation
to Prevent Errors. Oakbrook Terrace, IL: Joint Commission on
Accreditation of Healthcare Organizations.

Kasper, J. D., Elias, R., & Lyons, B. (2004). Dual Eligibles: Medicaid's Role
in Filling Medicare's Gaps. Washington, DC: Kaiser Family
Foundation - Kaiser Commission on Medicaid and the Uninsured.

Katon, W. J., Lin, E., Russo, J., & Unutzer, J. (2003). Increased medical
costs of a population-based sample of depressed elderly patients.
Archives of General Psychiatry, 60(9), 897-903.

Kaufman, D. W., Kelly, J. P., Rosenberg, L., Anderson, T. E., & Mitchell, A.
A. (2002). Recent patterns of medication use in the ambulatory adult
population of the United States - The Slone survey. Journal of the
American Medical Association, 287(3), 337-344.

Kitson, A., Harvey, G., & McCormack, B. (1998). Enabling the
implementation of evidence-based practice: A conceptual framework.
Quality in Health Care, 7, 149-158.

Koch, L., Cook, B., Tnakersley, M., & Rumrill, P. (2006). Utilizing research
in professional practice. Work, 26(3), 327-331.

Krska, J., Cromarty, J. A., Arris, F., Jamieson, D., Hansford, D., Duffus, P.
R. S., Downie, G., & Seymour, D. G. (2001). Pharmacist-led
medication review in patients over 65: A randomized, controlled trial
in primary care. Age and Ageing, 30(3), 205-211.

Lane, C. J., Bronskill, S. E., Sykora, K., Dhalla, I. A., Anderson, G. M.,
Mamdani, M. M., Gill, S. S., Gurwitz, J. H., & Rochon, P. A. (2004).
Potentially inappropriate prescribing in Ontario community-dwelling
older adults and nursing home residents. Journal of the American
Geriatrics Society, 52(6), 861-866.

Larson, P. D., & Hoot-Martin, J. (1999). Polypharmacy and elderly patients.
AORN Journal, 69(3), 619-628.

Lau, D. T., Kasper, J. D., Potter, D. E. B., & Lyles, A. (2004). Potentially
inappropriate medication prescriptions among elderly nursing home
residents: Their scope and associated resident and facility
characteristics. Health Services Research, 39(5), 1257-1276.

191
Leach, M. J. (2006). Evidence-based practice: A framework for clinical
practice and research design. International Journal of Nursing
Practice, 12(5), 248-251.

Lechevallier-Michel, N., Gautier-Bertrand, M., Alperovitch, A., Berr, C.,
Belmin, J., Legrain, S., Saint-Jean, O., Tavernier, B., Dartigues, J.-
F., & Fourrier-Reglat, A. (2005). Frequency and risk factors of
potentially inappropriate medication use in a community-dwelling
elderly population: Results from the 3C Study. European Journal of
Clinical Pharmacology, 60(11), 813-819.

Leventhal, H., & Friedman, M. A. (2004). Does establishing fidelity of
treatment help in understanding treatment efficacy? Comment on
Bellg et al. Health Psychology, 23(5), 452-456.

Lipsky, M. (1980). Street-level bureaucrats: Dilemmas of the individual in
public services. New York: Russell Sage Foundation.

Little, P., Dorward, M., Warner, G., Stephens, K., Senior, J., & Moore, M.
(2004). Importance of patient pressure and perceived pressure and
perceived medical need for investigations, referral, and prescribing in
primary care: nested observational study. British Medical Journal,
328, 444-447.

Liu, G. G., & Christensen, D. B. (2002). The continuing challenge of
inappropriate prescribing in the elderly: An update of the evidence.
Journal of the American Pharmaceutical Association, 42(6), 847-857.

Lorig, K. R., Sobel, D. S., Stewart, A. L., Brown, B. W., Jr., Bandura, A.,
Ritter, P., Gonzalez, V. M., Laurent, D. D., Holman, H. R., Lorig, K.
R., Sobel, D. S., Stewart, A. L., Brown, B. W., Jr., Bandura, A., Ritter,
P., Gonzalez, V. M., Laurent, D. D., & Holman, H. R. (1999).
Evidence suggesting that a chronic disease self-management
program can improve health status while reducing hospitalization: A
randomized trial. Medical Care, 37(1), 5-14.

Louis, M. R. (1983). Organizations as culture-bearing milieux. In L. Pondy,
P. Frost, G. Morgan & T. Dandridge (Eds.), Organizational
symbolism (pp. 39-54). Greenwich, CT: JAI Press.

192
Luber, M. P., Meyers, B. S., Williams-Russo, P. G., Hollenberg, J. P.,
DiDomenico, T. N., Charlson, M. E., & Alexopoulos, G. S. (2001).
Depression and service utilization in elderly primary care patients.
American Journal of Geriatric Psychiatry, 9(2), 169-176.

Maio, V., Hartmann, C. W., Poston, S., Liu-Chen, X., Diamond, J., &
Arenson, C. (2006). Potentially inappropriate prescribing for elderly
patients in 2 outpatient settings. American Journal of Medical
Quality, 21(3), 162-168.

Mazmanian, D. A., & Sabatier, P. A. (1989). Implementation and public
policy (2nd ed.). Landham, MD: University Press of America.

McCormack, B., Kitson, A., Harvey, G., Rycroft-Malone, J., Titchen, A., &
Seers, K. (2002). Getting evidence into practice: the meaning of
`context'. Journal of Advanced Nursing, 38(1), 94-104.

McKenna, H. P., Ashton, S., & Keeney, S. (2004). Barriers to evidence-
based practice in primary care. Journal of Advanced Nursing, 45(2),
178-189.

McNeece, C. A., & Thyer, B. A. (2004). Evidence-based practice and social
work. Journal of Evidence-Based Social Work, 1(1), 7-25.

Meijers, J. M. M., Janssen, M. A. P., Cummings, G. G., Wallin, L.,
Estabrooks, C. A., & Y.G. Halfens, R. (2006). Assessing the
relationships between contextual factors and research utilization in
nursing: systematic literature review. Journal of Advanced Nursing,
55(5), 622-635.

Meredith, S., Feldman, P. H., Frey, D., Giammarco, L., Hall, K., Arnold, K.,
Brown, N. J., & Ray, W. A. (2002). Improving medication use in
home healthcare patients: A randomized controlled trial. Journal of
the American Geriatrics Society, 50, 1484-1491.

Meredith, S., Feldman, P. H., Frey, D., Hall, K., Arnold, K., Brown, N. J., &
Ray, W. A. (2001). Possible medication errors in home healthcare
patients. Journal of the American Geriatrics Society, 49, 719-724.

Miller, E. A., & Weissert, W. G. (2000). Predicting elderly people's risk for
nursing home placement, hospitalization, functional impairment, and
mortality: a synthesis. Medical Care Research & Review, 57(3), 259-
297.

193
Miller, L. S., Clark, M. L., & Clark, W. F. (1985). The comparative evaluation
of California's Multipurpose Senior Services Project. Home Health
Care Services Quarterly, 6(3), 49-79.

Miller, N. A., Ramsland, S., & Harrington, C. (1999). Trends and issues in
the Medicaid 1915c waiver program. Health Care Financing Review,
20(4), 139-160.

Medicare Prescription Drug, Improvement, and Modernization Act. 2003,
Sec. 1013 Cong. Rec.(2003).

Moher, D., Schulz, K., & Altman, D. (2001). The CONSORT statement:
Revised recommendations for improving the quality of reports.
Journal of the American Medical Association, 285, 1987-1991.

Moon, S., & Shin, J. (2006). Health care utilization among Medicare-
Medicaid dual eligibles: A count data analysis. BMC Public Health, 6,
88-97.

Moxley, D. P. (1989). The practice of case management. Thousand Oaks,
CA: Sage Publications.

Murphy, A., & McDonald, J. (2004). Power, status and marginalisation:
Rural social workers and evidence-based practice in multidisciplinary
teams. Australian Social Work, 57(2), 127-136.

Myrtle, R. C., & Wilber, K. H. (1994). Designing service delivery systems:
Lessons from the development of community-based systems of care
for the elderly. Public Administration Review, 54(3), 245-252.

Nakamura, R. T., & Smallwood, F. (1980). The politics of policy
implementation. New York: St. Martin's Press.

National Association of Social Workers. (1996). Code of Ethics of the
National Association of Social Workers. Washington, DC: NASW
Press.

National Center for Health Statistics. (2004a). Health, United States, 2004.
Atlanta, GA: Centers for Disease Control and Prevention.

National Center for Health Statistics. (2004b). Home Health Care Patients:
Data from the 2000 National Home and Hospice Care Survey.
Hyattsville, MD.

194
National Center for HIV Prevention. (2003). Procedural Guidance for
Selected Strategies and Interventions for Community-Based
Organizations Funded Under Program Announcement. Atlanta, GA:
Centers for Disease Control and Prevention.

National Institute of Health. (2003). NIH Roadmap Initiatives. Retrieved
March 5, 2006, from http://nihroadmap.nih.gov/initiatives

National Institute of Health. (2006). PAR-06-520 Dissemination and
Implementation Research in Health (R03). Retrieved August 11,
2006, from http://grants.nih.gov

Niwata, S., Yamada, Y., & Ikegami, N. (2006). Prevalence of inappropriate
medication using Beers criteria in Japanese long-term care facilities.
BMC Geriatrics, 6(1), 1-7.

Onder, G., Landi, F., Cesari, M., Gambassi, G., Carbonin, P., & Bernabei,
R. (2003). Inappropriate medication use among hospitalized older
adults in Italy: results from the Italian Group of
Pharmacoepidemiology in the Elderly. European Journal of Clinical
Pharmacology, 59(2), 157-162.

Perri, M., III, Menon, A. M., Deshpande, A. D., Shinde, S. B., Jiang, R.,
Cooper, J. W., Cook, C. L., Griffin, S. C., & Lorys, R. A. (2005).
Adverse outcomes associated with inappropriate drug use in nursing
homes. Annals of Pharmacotherapy, 39(3), 405-411.

Pett, M. A. (1997). Nonparametric statistics in health care research.
Thousand Oaks, CA: Sage Publications.

Phillips, D. P., Christenfeld, N., & Glynn, L. M. (1998). Increase in US
medication-error deaths between 1983 and 1993. Lancet, 351(9103),
643-644.

Piecoro, L. T., Browning, S. R., Prince, T. S., Rantz, T. T., & Schutchfield,
F. D. (2000). A database analysis of potentially inappropriate drug
use in an elderly Medicaid population. Pharmacotherapy, 20(2), 221-
228.

Pillemer, K., Czaja, S., Schulz, R., & Stahl, S. M. (2003). Finding the best
way to help: Opportunities and challenges of intervention research
on aging. The Gerontologist, 43(Special Issue 1), 5-8.

195

Pitkala, K. H., Strandberg, T. E., & Tilvis, R. S. (2002). Inappropriate drug
prescribing in home-dwelling, elderly patients: A population-based
survey. Archives of Internal Medicine, 162(15), 1707-1712.

Porter, R., & Ogilvie, M. (2000). Biographical dictionary of scientists (3rd
ed.). New York: Oxford University Press.

Pressman, J. L., & Wildavsky, A. (1984). Implementation: How great
expectations in Washington are dashed in Oakland; or, why it's
amazing that Federal programs work at all, this being a saga of the
Economic Development Administration as told by two sympathetic
observers who seek to build morals on a foundation of ruined hopes.
(3rd ed.). Berkeley, CA: University of California Press.

Prohaska, T., Belansky, E., Belza, B., Buchner, D., Marshall, V., McTigue,
K., Satariano, W., & Wilcox, S. (2006). Physical activity, public
health, and aging: Critical issues and research priorities. Journals of
Gerontology: Social Sciences, 61(5), S267-273.

Pugh, M. J. V., Fincke, B. G., Bierman, A. S., Chang, B.-H., Rosen, A. K.,
Cunningham, F. E., Amuan, M. E., Burk, M. L., & Berlowitz, D. R.
(2005). Potentially inappropriate prescribing in elderly veterans: Are
we using the wrong drug, wrong dose, or wrong duration? Journal of
the American Geriatrics Society, 53(8), 1282-1289.

Quality Interagency Coordination Task Force. (2000). Doing What Counts
for Patient Safety: Federal Actions to Reduce Medical Errors and
Their Impact. Rockville, MD: Agency for Healthcare Research and
Quality.

Retsas, A. (2000). Barriers to using research evidence in nursing practice.
Journal of Advanced Nursing, 31(3), 599-606.

Reuben, D. B. (2002). Organizational interventions to improve health
outcomes of older persons. Medical Care, 40(5), 416-428.

Rhoads, M., & Thai, A. (2003). Physician acceptance rate of pharmacist
recommendations to reduce use of potentially inappropriate
medications in the assisted living setting. Consultant Pharmacist,
18(3), 241-247.

196
Rigler, S. K., Jachna, C. M., Perera, S., Shireman, T. I., & Eng, M. L.
(2005). Patterns of potentially inappropriate medication use across
three cohorts of older Medicaid recipients. Annals of
Pharmacotherapy, 39(7), 1175-1181.

Rochon, P. A., & Gurwitz, J. H. (1997). Optimizing drug treatment for elderly
people: The prescribing cascade. British Medical Journal, 315, 1096-
1099.

Rochon, P. A., & Gurwitz, J. H. (1999). Prescribing for seniors: Neither too
much nor too little. Journal of the American Medical Association,
282(2), 113-115.

Rogers, E. (1995). Diffusion of innovations (4th ed.). New York: Free Press.

Rohrbach, L. A., Grana, R., Sussman, S., & Valente, T. W. (2006). Type II
translation: Transporting prevention interventions from research to
real-world settings. Evaluation and the Health Professions, 29(3),
302-333.

Rosen, L., Manor, O., Engelhard, D., & Zucker, D. (2006). In defense of the
randomized controlled trial for health promotion research. American
Journal of Public Health, 96(7), 1181-1186.

Rossi, P. H., Lipsey, M. W., & Freeman, H. E. (2004). Evaluation: A
systematic approach (7th ed.). Thousand Oaks, CA: Sage
Publications.

Rubenstein, L. V., & Pugh, J. (2006). Strategies for promoting
organizational and practice change by advancing implementation
research. Journal of General Internal Medicine, 21(s2), S58-S64.

Ryan, J., & Super, N. (2003). Dually Eligible for Medicare and Medicaid:
Two for One or Double Jeopardy? NHPF Issue Brief # 794.
Washington, DC: National Health Policy Form.

Rycroft-Malone, J., Harvey, G., Seers, K., Kitson, A., McCormack, B., &
Titchen, A. (2004). An exploration of the factors that influence the
implementation of evidence into practice. Journal of Clinical Nursing,
13(8), 913-924.

197
Rycroft-Malone, J., Kitson, A., Harvey, G., McCormack, B., Seers, K.,
Titchen, A., & Estabrooks, C. (2002). Ingredients for change:
Revisiting a conceptual framework. Quality & Safety in Health Care,
11(2), 174-180.

Rycroft-Malone, J., Seers, K., Titchen, A., Harvey, G., Kitson, A., &
McCormack, B. (2004). What counts as evidence in evidence-based
practice? Journal of Advanced Nursing, 47(1), 81-90.

Sabatier, P. A. (1986). Top-down and bottom-up approaches to
implementation research: A critical analysis and suggested
synthesis. Journal of Public Policy, 6(1), 21-48.

Sackett, D. L., Rosenberg, W. M. C., Gray, J. A. M., Haynes, R. B., &
Richardson, W. S. (1996). Evidence based medicine: What it is and
what it isn't. British Medical Journal, 312(7023), 71-72.

Safran, D. G., Neuman, P., Schoen, C., Kitchman, M. S., Wilson, I. B.,
Cooper, B., Li, A., Chang, H., & Rogers, W. H. (2005). Prescription
drug coverage and seniors: Findings from a 2003 National Survey.
Health Affairs, 24(Supplement 1), W5-152-W155-166.

Scharlach, A., Giunta, N., & Mills-Dick, K. (2001). Case Management in
Long-Term Care Integration: An Overview of Current Programs and
Evaluations. Berkeley, CA: Center for the Advanced Study of Aging
Services - University of California.

Schmader, K. E., Hanlon, J. T., Pieper, C. F., Sloane, R., Ruby, C. M.,
Twersky, J., Francis, S. D., Branch, L. G., Lindblad, C. I., & Artz, M.
(2004). Effects of geriatric evaluation and management on adverse
drug reactions and suboptimal prescribing in the frail elderly.
American Journal of Medicine, 116(6), 394-401.

Sellors, J., Kaczorowski, J., Sellors, C., Dolovich, L., Woodward, C., Willan,
A., Goeree, R., Cosby, R., Trim, K., Sebaldt, R., Howard, M.,
Hardcastle, L., & Poston, J. (2003). A randomized controlled trial of a
pharmacist consultation program for family physicians and their
elderly patients. Canadian Medical Association Journal, 169(1), 17-
22.

198
Shapiro, A., & Taylor, M. (2002). Effects of a community-based early
intervention program on the subjective well-being, institutionalization,
and mortality of low-income elders. The Gerontologist, 42(3), 334-
341.

Sheldon, B. (2001). The validity of evidence-based practice in social work:
A reply to Stephen Webb. British Journal of Social Work, 31, 801-
809.

Shelton, P. S., Fritsch, M. A., & Scott, M. A. (2000). Assessing medication
appropriateness in the elderly - A review of available measures.
Drugs & Aging, 16(6), 437-450.

Silverman, J. B., Stapinski, C. D., Huber, C., Ghandi, T. K., & Churchill, W.
W. (2004). Computer-based system for preventing adverse drug
events. American Journal of Health System Pharmacy, 61(15), 1599-
1603.

Simon, S. R., Chan, K. A., Soumerai, S. B., Wagner, A. K., Andrade, S. E.,
Feldstein, A. C., Lafata, J. E., Davis, R. L., & Gurwitz, J. H. (2005).
Potentially inappropriate medication use by elderly persons in U.S.
Health Maintenance Organizations, 2000-2001. Journal of the
American Geriatrics Society, 53(2), 227-232.

Sloane, P. D., Zimmerman, S., Brown, L. C., Ives, T. J., & Walsh, J. F.
(2002). Inappropriate medication prescribing in residential
care/assisted living facilities. Journal of the American Geriatrics
Society, 50(6), 1001-1011.

Smith, V., Gifford, K., Kramer, S., & Elam, L. (2006). The Transition of Dual
Eligibles to Medicare Part D Prescription Drug Coverage: State
Actions during Implementation. Menlo Park, CA: Kaiser Family
Foundation.

Social Security Act, 42 U.S.C. 1396n SEC. 1915 (1981).
Solomon, J., Card, J. J., & Malow, R. M. (2006). Adapting efficacious
interventions: Advancing translational research in HIV prevention.
Evaluation and the Health Professions, s(2), 162-194.

Sorensen, L., Stokes, J. A., Purdie, D. M., Woodward, M., Elliott, R., &
Roberts, M. S. (2004). Medication reviews in the community: Results
of a randomized, controlled effectiveness trial. British Journal of
Clinical Pharmacology, 58(6), 648-664.

199
Sorensen, L., Stokes, J. A., Purdie, D. M., Woodward, M., & Roberts, M. S.
(2005). Medication management at home: Medication-related risk
factors associated with poor health outcomes. Age and Ageing,
34(6), 626-632.

Spiker, E., Emptage, R., Giannamore, M., & Pedersen, C. (2001). Potential
adverse drug events in an indigent and homeless geriatric
population. Annals of Pharmacotherapy, 35(10), 1166-1172.

Stetler, C. B., Legro, M. W., Wallace, C. M., Bowman, C., Guihan, M.,
Hagedorn, H., Kimmel, B., Sharp, N. D., & Smith, J. L. (2006). The
role of formative evaluation in implementation research and the
QUERI experience. Journal of General Internal Medicine, 21(s2), S1-
S8.

Stuck, A. E., Beers, M. H., Steiner, A., Aronow, H. U., Rubenstein, L. Z., &
Beck, J. C. (1994). Inappropriate medication use in community-
residing older persons. Archives of Internal Medicine, 154(19), 2195-
2200.

Sung, N. S., Crowley, W. F., Jr, Genel, M., Salber, P., Sandy, L., Sherwood,
L. M., Johnson, S. B., Catanese, V., Tilson, H., Getz, K., Larson, E.
L., Scheinberg, D., Reece, E. A., Slavkin, H., Dobs, A., Grebb, J.,
Martinez, R. A., Korn, A., & Rimoin, D. (2003). Central challenges
facing the national clinical research enterprise. Journal of the
American Medical Association, 289(10), 1278-1287.

Sussman, S., Valente, T. W., Rohrbach, L. A., Skara, S., & Ann Pentz, M.
(2006). Translation in the health professions: Converting science into
action. Evaluation and the Health Professions, 29(1), 7-32.

Tahan, H. (2006). Essential activities and knowledge domains of case
management: New insights from the CCMC role and functions study.
The Case Manager, 17(3), 45-48.

Tanenbaum, S. (2003). Evidence-based practice in mental health: Practical
weaknesses meet political strengths. Journal of Evaluation in Clinical
Practice, 9(2), 287-301.

Tang, K. C., Ehsani, J. P., & McQueen, D. V. (2003). Evidence based
health promotion: Recollections, reflections, and reconsiderations.
Journal of Epidemiology & Community Health, 57(11), 841-843.

200
Timmermans, S., & Mauck, A. (2005). The promises and pitfalls of
evidence-based medicine. Health Affairs, 24(1), 18-28.

Tinetti, M., & Williams, C. (1998). The effect of falls and fall injuries on
functioning in community-dwelling older persons. Journals of
Gerontology: Biological Sciences and Medical Sciences, 53(2),
M112-119.

Titler, M. G. (2004). Methods in translation science. Worldviews Evidence-
Based Nursing, 1(1), 38-48.

Toobert, D. J., Strycker, L. A., Glasgow, R. E., Barrera, M., & Bagdade, J.
D. (2002). Enhancing support for health behavior change among
women at risk for heart disease: The Mediterranean Lifestyle Trial.
Health Education Research, 17(5), 574-585.

Triller, D. M., Clause, S. L., Briceland, L. L., & Hamilton, R. A. (2003).
Resolution of drug-related problems in home care patients through a
pharmacy referral service. American Journal of Health System
Pharmacy, 60, 905-910.

VA-QUERI. (2004). Guide for Implementing Evidence-Based Clinical
Practice and Conducting Implementation Research. Retrieved June
6, 2006, from http://www.hsrd.research.va.gov/queri

van der Hooft, C. S., Jong, G. W. t., Dieleman, J. P., Verhamme, K. M. C.,
van der Cammen, T. J. M., Stricker, B. H. C., & Sturkenboom, M. C.
J. M. (2005). Inappropriate drug prescribing in older adults: The
updated 2002 Beers criteria - a population-based cohort study.
British Journal of Clinical Pharmacology, 60(2), 137-144.

Vinks, T., de Koning, F., de Lange, T., & Egberts, T. (2006). Identification of
potential drug-related problems in the elderly: The role of the
community pharmacist. Pharmacy World & Science, 28(1), 33-38.

Viswanathan, H., Bharmal, M., & Thomas III, J. (2005). Prevalence and
correlates of potentially inappropriate prescribing among ambulatory
older patients in the year 2001: Comparison of three explicit criteria.
Clinical Therapeutics, 27(1), 88-99.

Wallin, L., Estabrooks, C. A., Midodzi, W. K., & Cummings, G. G. (2006).
Development and validation of a derived measure of research
utilization by nurses. Nursing Research, 55(3), 149-160.

201
Walshe, K., & Rundall, T. (2001). Evidence-based management: From
theory to practice in health care. Milbank Quarterly, 79(3), 429-457.

Webb, S. (2001). Some considerations on the validity of evidence-based
practice in social work. British Journal of Social Work, 31(1), 57-79.

Weiss, C. H. (1997). Evaluation (2nd ed.). New York: Prentice Hall.

Weissert, W., Chernew, M., & Hirth, R. (2003). Titrating versus targeting
home care services to frail elderly clients: An application of agency
theory and cost-benefit analysis to home care policy. Journal of
Aging and Health, 15(1), 99-123.

Wensing, M., Wollersheim, H., & Grol, R. (2006). Organizational
interventions to implement improvements in patient care: A
structured review of reviews. Implementation Science, 1(1), 2.

Wholey, J. S., Hatry, H. P., & Newcomer, K. E. (Eds.). (2004). Handbook of
practical program evaluation (2nd ed.). New York: Jossey-Bass.

Will, J., Farris, R., Sanders, C., Stockmyer, C., & Finkelstein, E. (2004).
Health promotion intervention for disadvantaged women: Overview
of the WISEWOMAN projects. Journal of Women's Health, 13, 484-
502.

Willcox, S., M., Himmelstein, D. U., & Woolhandler, S. (1994). Inappropriate
drug prescribing for the community-dwelling elderly. Journal of the
American Medical Association, 272(4), 292-296.

Wu, S.-Y., & Green, A. (2000). Projection of Chronic Illness and Cost
Inflation. Santa Monica, CA: Rand Corporation.

Zecevic, A. A., Salmoni, A. W., Speechley, M., & Vandervoort, A. A. (2006).
Defining a fall and reasons for falling: Comparisons among the views
of seniors, health care providers, and the research literature. The
Gerontologist, 46(3), 367-376.

Zerhouni, E. A. (2003). The NIH Roadmap. Science, 302, 63-72.

Zerhouni, E. A. (2005a). Translational and clinical science -- Time for a new
vision. New England Journal of Medicine, 353(15), 1621-1623.

202
Zerhouni, E. A. (2005b). US biomedical research: Basic, translational, and
clinical sciences. Journal of the American Medical Association,
294(11), 1352-1358.

Zhan, C., Sangl, J., Bierman, A. S., Miller, M. R., Friedman, B., Wickizer, S.
W., & Meyer, G. S. (2001). Potentially inappropriate medication use
in the community-dwelling elderly: Findings from the 1996 Medical
Expenditure Panel Survey. Journal of the American Medical
Association, 286(22), 2823-2829.

203
APPENDICES

Appendix A: Procedure for Medication Problem Screening
Appendix B: Medications Errors Data Collection Form
Appendix C: Confirmation Procedures of Medication Problems by
Category
Appendix D: Terminated Participants Prior to Medication Problem
Confirmation and Sample
Appendix E: Reason for Termination Prior to Medication Problem
Confirmation
Appendix F: Terminated Participants Prior to Intervention and
Intervention Sample
Appendix G: Reason for Termination Prior to Intervention
Appendix H: Medication Follow-Up Form
Appendix I: Pharmacist Contacts and Medication Change
Appendix J: MSSP Staff Questionnaire
Appendix K: Responses to MSSP Staff Questionnaire
Appendix L: Qualitative Responses to MSSP Staff Questionnaire
Appendix M: Three Medication Management Protocols Created by
Site #3
Appendix N: Disclaimer

204
Appendix A: Procedure for Medication Problem Screening
For each patient and for each drug a patient is currently taking and after all
drugs have been "standardized" (via a coding system or conversion to
generic/active ingredients), do the following:

1.
Load all tables for drugs of interest (e.g. NSAIDS, benzodiazepines, etc)
by whatever method is chosen.
2. Check for hypertension
3.
Check if taking Tricyclic Antidepressant (TCA) - if yes, check for
orthostasis, confusion, recent fall
4.
Check if taking anti-psychotic of interest - if yes, check for orthostasis,
confusion, recent fall
5.
Check if taking non-ACE anti-hypertensives - if yes, check for orthostasis,
or SBP less than or equal 90, or SBP less than or equal 100 with vertigo.
6.
Check pulse - if less than 55, check if taking digitalis, thyroid, or selected
anti-hypertensives
7. Check if taking levodopa - if yes, check orthostasis
8. Check if taking benzodiazepine - if yes, check for falls or confusion
9.
Check if taking NSAIDs - if yes, check if taking more than 325 mg ASA
daily, age over 80, or taking an anti-ulcer, anti-coagulant, or steroid.
10. Check for therapeutic duplication of the following groups:
a. NSAIDs
b. Narcotics
c. H2 blockers
d. Dihydro calcium channel blockers
e. Non-dihydro calcium channel blockers
f. Benzodiazepines
g. ACE Inhibitors
h. Sulfonylureas
i. Beta-blockers
j. TCA's
k. SSRI's
l. Anti-psychotics
m. Loop diuretics
n. Thiazides
O Potassium sparing diuretics
p. Thyroid
q. Theophyllines
r. Systemic corticosteroids
s. Estrogens
11.
For dosage checking, we created a table of 94 drugs (most commonly
prescribed) with their name/code and the maximum daily dose.

205
Appendix B: Medications Errors Data Collection Form

Study Number Pharmacist

1. What type of error was detected:
a. Cardiac: 1. Uncontrolled hypertension
2. Hypotension
3. Low pulse
a. Psychotropics: 1. Diphenhydramine, etc.
2. Benzodiazapine + falls or confusion
3. Antidepressants + falls or confusion
4. Antipsychotics + falls or confusion
b. NSAID
c. Therapeutic Duplication

2. Did you speak with the care manager (by telephone, in-person, or case
conferences)?
(0) No (1) Yes

3. Was the medication error confirmed? (1) Yes (0) No
a. If no, why was error a false
positive?_________________________________________________
___________________________________________________________

4. Was the physician contacted?
(0) No (1) Yes
a. If no, please
clarify:________________________________________________

5. Did the physician agree to make the changes to the patient’s medication in
accordance with the guidelines?
(0) No (1) Yes

6. Were medication changes discussed with the patient?
(0) No (1) Yes
a. If yes, did the patient agree to the proposed changes?
(0) No (1) Yes (9) Unknown

7. So far as you know, did the patient’s medication change?
(0) No (Skip to #8) (1) Yes
a. If yes, do you think it was in response to your intervention?
(0) No (1) Yes

8. Estimate the approximate time you spent on this case: _______ Minutes

9. Comments (and if more than one problem, please describe which were dealt with):
__________________________________________________________________
__________________________________________________________________
__________________________________________________________________

206
Appendix C: Confirmation Procedures of Medication Problems by Category

The four medication problems as defined by the Home Health Criteria were
confirmed as a medication problem if the following conditions were met:

1. Therapeutic Duplication: Verified that participant was taking both
medications that were in question and that the medication combination was
not determined to be therapeutically appropriate based on clinical
diagnoses.
• Examples of therapeutically appropriate duplication: routine pain
medication coupled with PRN pain medication for osteoarthritis;
participant prescribed Dilantin and Neurontin for seizure disorders.

2. Inappropriate Use of Psychotropic medications:
• Falls: participant must have reported a fall in the last three months
and was taking a target category psychotropic medication (e.g.
benzodiazepine).

• Confusion- participant must have presence of confusion as assessed
by the care management staff and taking a target category
medication.

3. Cardiovascular:
• Orthostasis: Participant must have reported dizziness and had at
least two blood pressure readings that reported more than 20mm
change between two changing positions (e.g. sitting to standing).

• High Blood Pressure: Verified elevated blood pressure (Systolic BP
≥160 and/or diastolic BP >90 mmHg) during at least two reading.

• Low Blood Pressure: Verified that participant was taking a
medication that lowers blood pressure and had at least two low blood
pressure readings (SBP<100mmHg).

• Low Heart Rate: Verified low pulse reading (55 bpm or lower) and
taking an anti-hypertensive agent.

4. Non-Steroidal Anti-Inflammatory Drugs (NSAIDs): Verified that participant
was taking medications in question and was 1) age 80 or above, 2) taking
interacting meds such as anti-coagulants or corticosteroids, and 3) that
there was no diagnosis of rheumatoid arthritis or significant osteoarthritis.

207

Appendix D: Terminated Participants Prior to Medication Problem Confirmation & Sample (N=615)
N % Mean (SD) N % Mean (SD) t-Test Chi
2
Socio-demographics
Age (65-108) 79.00 (5.83) 80.83 (7.79) -0.739
Female 7 70.0% 483 79.8% 0.588
Race/Ethnicity 7.683
Caucasian 5 50.0% 141 23.3%
African-American 3 30.0% 238 39.3%
Latino/a 0 146 24.1%
Asian/Pacific Islander 2 20.0% 50 8.3%
Other 0 30 5.0%
English Speaking 7 70.0% 356 58.8% 0.506
Married 1 10.0% 122 20.2% 0.635
Widowed 5 50.0% 320 52.9% 0.033
High School & Above* 7 70.0% 232 38.3% 4.148
Lives Alone 4 40.0% 251 41.5% 0.011
Newly Enrolled in MSSP 3 30.0% 385 63.6% 0.179
MSSP Site
Site #1 6 60.0% 210 34.7% 2.811
Site #2 3 30.0% 270 44.6%
Site #3 1 10.0% 125 20.7%
Health Status
ER/Hospital/SNF in Last Year 4 40.0% 230 38.0% 0.016
# of Medications (0-27) 9.20 (4.98) 8.75 (4.29) 0.327
12+ Medications 3 30.0% 134 22.1% 0.350
Falls in Past 3 Months* 5 50.0% 130 21.5% 5.549
Dizziness 2 20.0% 166 27.4% 0.099
Confusion 3 30.0% 187 30.9% 0.234
Lived Alone x Risk 4 40.0% 122 20.2% 3.091
Potential Medication Problem
Any Problem** 10 100.0% 289 47.8% 10.743
Therapeutic Duplication 3 30.0% 146 24.1% 0.185
Psychotropic Medication 3 30.0% 85 14.0% 2.041
Confusion 1 10.0% 48 7.9% 0.057
Falls 2 20.0% 50 8.3% 1.750
Cardiovascular Medication*** 6 60.0% 81 13.4% 17.598
NSAIDS 1 10.0% 78 12.9% 0.074
*p<.05; **p<.01; ***p<.001
NSAIDs = Inappropriate Use of Non-Steroidal Anti-Inflammatory Drugs
Risk defined as presence of falls, dizziness, or confusion
Terminated (N=10) Remained (N=605)

208

Appendix E: Reason for Termination Prior to Medication Problem Confirmed (N=10)
Died 2 20.0%
Long-Term Institutionalization 3 30.0%
Moved Out of Area 2 20.0%
No longer MSSP/Medi-Cal Eligible 1 10.0%
No Longer Desired Services 0 0.0%
Unknown 2 20.0%

209

Appendix F: Terminated Participants Prior to Intervention and Intervention Sample (N=181)
N % Mean (SD) N % Mean (SD) t-Test Chi
2
Socio-demographics
Age (65-108) 82.95 (8.72) 80.58 (7.76) 1.240
Female 18 94.7% 131 80.9% 2.249
Race/Ethnicity 6.844
Caucasian 2 10.5% 44 27.2%
African-American 10 52.6% 62 38.3%
Latino/a 3 15.8% 40 24.7%
Asian/Pacific Islander 2 10.5% 12 7.4%
Other/Unknown 2 10.5% 4 2.5%
English Speaking 15 78.9% 99 61.1% 2.320
Married 2 10.5% 25 15.4% 0.322
Widowed 10 52.6% 93 57.4% 0.158
High School & Above 7 36.8% 69 42.6% 0.321
Lived Alone 8 42.1% 78 48.1% 0.328
MSSP Site 4.090
Site #1 3 15.8% 59 36.4%
Site #2 9 47.4% 69 42.6%
Site #3 7 36.8% 34 21.0%
Health Status
ER/Hospital/SNF in Last Year 7 36.8% 76 46.9% 0.695
# of Medications (2-26)* 8.21 (3.74) 10.73 (4.70) -2.259
12+ Medications* 2 10.5% 60 37.0% 5.307
Falls in Past 3 Months 4 21.1% 58 35.8% 1.403
Dizziness 6 31.6% 61 37.7% 0.335
Confusion 6 31.6% 58 35.8% 0.063
Lived Alone x Risk 5 26.3% 51 31.5% 0.228
Confirmed Medication Problem
Therapeutic Duplication 8 42.1% 79 48.8% 0.302
Psychotropic Medication 6 31.6% 59 36.4% 0.173
Confusion 3 15.8% 34 21.0% 0.283
Falls 3 15.8% 37 22.8% 0.491
Cardiovascular Medication 3 15.8% 24 14.8% 0.013
NSAIDs 9 47.4% 44 27.2% 3.354
*p<.05; **p<.01; ***p<.001
NSAIDs = Inappropriate Use of Non-Steroidal Anti-Inflammatory Drugs
Risk defined as presence of falls, dizziness, or confusion
Terminated (N=19) Intervention (N=162)

210

Appendix G: Reason for Termination Prior to Intervention (N=19)
N%
Died 7 36.8%
Long-Term Institutionalization 7 36.8%
Moved Out of Area 2 10.5%
No longer MSSP/Medi-Cal Eligible 1 5.3%
No Longer Desire Services 1 5.3%
Unable/Unwilling to Follow Plan 1 5.3%

211
Appendix H: Medication Follow-Up Form

[PARTICIPANT’S NAME] screened in with the following potential
medication problem:

Cardiac: Uncontrolled hypertension

Hypotension

Low pulse

Psychotropics: Diphenhydramine, etc.

Benzodiazapine + falls or confusion

Antidepressants + falls or confusion

Antipsychotics + falls or confusion
NSAID

Therapeutic Duplication

Can you please follow–up on the following issues:

[PHARMACIST ASKED PRIMARY CARE MANAGER TO ASK SPECIFIC
QUESTIONS HERE BASED ON CONFIRMED MEDICATION PROBLEMS
IDENTIFIED]
____________________________________________________________
____________________________________________________________
____________________________________________________________
____________________________________________________________
____________________________________________________________
____________________________________________________________
____________________________________________________________
____________________________________________________________
____________________________________________________________
____________________________________________________________
____________________________________________________________

Once you complete this form please place it in the appropriate follow-
up box.

212

Appendix I: Pharmacist Contacts and Medication Change
# ∆ % ∆ # ∆ % ∆ # ∆ % ∆ # ∆ % ∆ # ∆ % ∆ # ∆ % ∆ # ∆ % ∆
Pharm Contacts
MD only 7 4 57.1% 1 0 0.0% 3 2 66.7% 3 2 66.7% 2 1 50.0% 0 0 0.0% 5 2 40.0%
CM only 40 21 52.5% 18 12 66.7% 12 7 58.3% 8 6 75.0% 5 1 20.0% 11 4 36.4% 12 2 16.7%
P/C only 4 2 50.0% 2 1 50.0% 1 1 100.0% 1 1 100.0% 0 0 0.0% 1 0 0.0% 0 0 0.0%
MD & CM 57 36 63.2% 30 18 60.0% 21 9 42.9% 9 5 55.6% 17 6 35.3% 9 5 0.0% 11 8 72.7%
MD & P/C 5 2 40.0% 2 2 100.0% 2 0 0.0% 1 0 0.0% 1 0 0.0% 2 1 50.0% 0 0 0.0%
CM & P/C 27 18 66.7% 14 8 57.1% 9 5 55.6% 6 4 66.7% 5 3 60.0% 0 0 0.0% 11 7 63.6%
MD, CM, & P/C 22 16 72.7% 12 8 66.7% 11 8 72.7% 6 5 83.3% 7 5 71.4% 1 1 100.0% 5 3 60.0%
Total # and ∆ 162 99 61.1% 79 49 62.0% 59 32 54.2% 34 23 67.6% 37 16 43.2% 24 11 45.8% 44 22 50.0%
Total MD Contacts 91 58 63.7% 45 28 62.2% 37 19 51.4% 19 12 63.2% 27 12 44.4% 12 7 58.3% 21 13 61.9%
Total CM Contacts 146 91 62.3% 74 46 62.2% 53 29 54.7% 29 20 69.0% 34 15 44.1% 21 10 47.6% 39 20 51.3%
Total P/C Contacts 58 38 65.5% 30 19 63.3% 23 14 60.9% 14 10 71.4% 13 8 61.5% 4 2 50.0% 16 10 62.5%
Any = Any of the 4 Confirmed Medication Problems # = Number of Cases Pharmacists Discussed
Ther. Duplication = Therapeutic Duplication ∆ = Medication Change
Psychotropic = Inappropriate Use of Psychotropic Medication Pharm = Consultant Pharmacist
Psych-Confusion = Inappropriate Use of Psychotropic Medication with Confusion MD = Prescribing Physician
Psych-Falls = Inappropriate Use of Psychotropic Medication with Fall in Last 3 Months CM = Primary Care Manager
CV = Cardiovascular Medication Problems P/C = Participant and/or Caregiver
NSAIDs = Inappropriate Use of Non-Steroidal Anti-Inflammatory Drugs
Psych-Falls CV NSAIDs Any Problem Ther. Duplicaton Psychotropic Psych-Confusion

212

213

Appendix J: MSSP Staff Questionnaire:
Community-Based Medication Management Intervention

To better improve services for MSSP clients, we would like to solicit your feedback regarding the operation and functioning of the
Community Based Medication Management Intervention. Please take a few minutes to complete and return this survey, as your responses
will help us understand your experience with implementing medication management in the MSSP program.

The Community-Based Medication Management Intervention has been abbreviated to MM Intervention

1. Are you a (please check one):
Social worker
Nurse
Other________________________________

2. What is your highest academic level achieved?
AA
Bachelor’s Degree
Master’s Degree
Doctorate

3. Are you licensed in your profession?
Yes
No

4. How long have you been working in your profession?
Less than one year
One to three years
Four to six years
Seven to nine years
Ten or more years

5. Are you currently employed in an MSSP program at either Partners
in Care or Huntington Senior Care Network?
Yes
No

6. How long have you been working (or worked) in MSSP?
Less than one year
One to three years
Four to six years
Seven to nine years
Ten or more years

7. What is (was) your average caseload size per month, including new
clients and those currently enrolled? ________________

8. How much previous experience do you have with any type of
medication screening and management intervention?
None
Some
A lot

9. How much previous experience do you have with implementing a
research-tested intervention into clinical practice?
None
Some
A lot
213

214

10. Please indicate (circle) how much you agree or disagree with the following statements on a scale from 1 to 5 where 1 = “strongly
disagree” and 5 = “strongly agree.” If you don’t remember or cannot comment, indicate N/A (not applicable) in the last column.

Statement
Strongly
disagree
Disagree
Neither agree
nor disagree
Agree
Strongly
agree
N/A
a. MM Intervention training was adequate 1 2 3 4 5 N/A
b. Training tools and protocols helped me
learn and implement the MM Intervention
1 2 3 4 5 N/A
c. Additional training on the MM Intervention
would have been helpful
1 2 3 4 5 N/A
d. My questions about the MM Intervention
were addressed in a timely manner
1 2 3 4 5 N/A
e. The MM Intervention administrative team
provided assistance when needed
1 2 3 4 5 N/A
f. It is important to me that this intervention
was tested in a research setting
1 2 3 4 5 N/A
g. Computerized screening of medications
helped me identify potential problems
1 2 3 4 5 N/A
h. Completing the medication follow-up form
was easy
1 2 3 4 5 N/A
i. Completing the medication follow-up form
was time consuming
1 2 3 4 5 N/A
j. My clients were aware that their
medications were screened for potential
problems
1 2 3 4 5 N/A
k. My clients appeared pleased that
medication screening and management were
offered as part of MSSP
1 2 3 4 5 N/A
l. Over time, I became comfortable with
implementing the MM Intervention

1 2 3 4 5 N/A
214

215

Statement
Strongly
disagree
Disagree
Neither agree
nor disagree
Agree
Strongly
agree
N/A
m. Medication management is typically part
of my scope of practice
1 2 3 4 5 N/A
n. Medication issues and solutions should be
discussed in case conferences
1 2 3 4 5 N/A
o. I contacted my clients’ physicians to inform
him/her of a potential medication problems
1 2 3 4 5 N/A
p. I was able to help adapt the MM
Intervention to how case management
operates in MSSP
1 2 3 4 5 N/A
q. The MM Intervention fit easily with my
other case management activities
1 2 3 4 5 N/A
r. In general, the MM Intervention helped my
clients reduce/avoid medication problems
1 2 3 4 5 N/A
s. My role in the MM Intervention was clear 1 2 3 4 5 N/A
t. The MM Intervention was useful and
worthwhile
1 2 3 4 5 N/A
u. I would like the MM Intervention to be an
ongoing service provided by MSSP
1 2 3 4 5 N/A

11. Please list any benefits you or your clients experienced from the MM intervention:
____________________________________________________________________________________________________

12. Please list any challenges you or your clients experienced during the MM intervention:
____________________________________________________________________________________________________

13. Please provide any feedback or suggestions that you think would improve the MM intervention:
____________________________________________________________________________________________________

Thank you very much for your assistance!!!
215

216

Appendix K: Responses to the MSSP Staff Questionnaire
Strongly
disagree Disagree
Neither agree
nor disagree Agree Strongly agree
a. MM Intervention training was adequate
0.0% 10.3% 13.8% 51.7% 13.8%
b. Training tools and protocols helped me learn and implement the
MM Intervention
0.0% 17.2% 3.4% 55.2% 6.9%
c. Additional training on the MM Intervention would have been
helpful
0.0% 6.9% 17.2% 37.9% 27.6%
d. My questions about the MM Intervention were addressed in a
timely manner
0.0% 3.4% 20.7% 51.7% 17.2%
e. The MM Intervention administrative team provided assistance
when needed
0.0% 0.0% 20.7% 51.7% 27.6%
f. It is important to me that this intervention was tested in a
research setting
0.0% 0.0% 34.5% 41.4% 24.1%
g. Computerized screening of medications helped me identify
potential problems
0.0% 6.9% 31.0% 27.6% 20.7%
h. Completing the medication follow-up form was easy
0.0% 13.8% 17.2% 51.7% 6.9%
i. Completing the medication follow-up form was time consuming
6.9% 24.1% 20.7% 37.9% 3.4%
j. My clients were aware that their medications were screened for
potential problems
3.4% 13.8% 6.9% 44.8% 17.2%
k. My clients appeared pleased that medication screening and
management were offered as part of MSSP
0.0% 6.9% 27.6% 34.5% 6.9%

216

217

Appendix K (cont.): Responses to the MSSP Staff Questionnaire
Strongly
disagree Disagree
Neither agree
nor disagree Agree Strongly agree
l. Over time, I became comfortable with implementing the MM
Intervention
0.0% 3.4% 27.6% 44.8% 13.8%
m. Medication management is typically part of my scope of practice
17.2% 17.2% 24.1% 34.5% 6.9%
n. Medication issues and solutions should be discussed in case
conferences
0.0% 6.9% 3.4% 55.2% 31.0%
o. I contacted my clients’ physicians to inform him/her of a
potential medication problems
6.9% 10.3% 17.2% 44.8% 3.4%
p. I was able to help adapt the MM Intervention to how case
management operates in MSSP
0.0% 6.9% 34.5% 51.7% 3.4%
q. The MM Intervention fit easily with my other case management
activities
0.0% 13.8% 34.5% 44.8% 3.4%
r. In general, the MM Intervention helped my clients reduce/avoid
medication problems
0.0% 6.9% 24.1% 55.2% 6.9%
s. My role in the MM Intervention was clear
3.4% 20.7% 20.7% 41.4% 10.3%
t. The MM Intervention was useful and worthwhile
0.0% 0.0% 31.0% 55.2% 13.8%
u. I would like the MM Intervention to be an ongoing service
provided by MSSP
0.0% 3.4% 31.0% 48.3% 17.2%

217

218
Appendix L: Qualitative Responses to MSSP Staff Questionnaire

Please list any benefits you or your clients experienced from the MM
intervention

• SW: Being able to identify risky meds and duplication and informing
clients or their families of potential side effects
• SW: Avoided duplication of meds and awareness of any side effects
such as falls or dizziness
• SW: MM team able to identify RX that had adverse effects with other
medications client on
• Nurse: Looking closely at the meds and evaluating the safely and
effectiveness of meds; having MM increased teaching on meds, side
effects and therapeutic effect which is good practice in patient care
• SW: Not over medicating; avoid duplication of Rx
• SW: Double/overdose prevention; side effect prevention
• SW: I learned many different medications and the reason why clients
were taking
• SW: They were able to address some medication complications to
their doctors
• SW: More knowledgeable about the meds, easier to identify potential
problems
• SW: Less medication overdosing/errors/duplications
• Nurse: Medication dosages were sometimes decreased which may
have decreased side effects
• Nurse: None noted
• SW: Having a second professional evaluate the appropriateness of
medication
• SW: Increased awareness of Rx duplication
• SW: More aware of meds and mixing meds can do harm to a client
• SW: Made clients and me more aware of meds interactions/potential
side effects
• Supervisor: From my observation, our clients benefited by reduction
in health complications, reductions in duplications of meds and
improvement in overall functioning
• SW: As a social worker I became aware of potential dangers of or
complications of some medications; I now look at all medications my
clients are taking

219
Please list any challenges you or your clients experienced during the
MM intervention

• SW: If MM could provide a staff member to do quarterly f/u and take
blood pressure when SW is not able to would be helpful
• SW: Contacting physicians
• SW: I received practically no orientation to MM, its
purpose/objectives when starting employment. I was told it existed
and given forms for f/u. I still don’t totally get what MM was about or
its purpose.
• Nurse: none
• SW: Accurate reporting due to memory problems and lack of family
support
• SW: Hard to work with some doctors or talk to them
• SW: None mentioned
• SW: I was never trained for MM (because I was not present during
that time) and did not know exactly what I was
• SW: No or slow response from the doctor. Many clients like to keep
all meds including those been taking off, making it very confusing. It
can take a long time to address a med problem
• SW: Talking to/convincing physician to make changes based on
recommendations
• Nurse: MD’s do not respond to all calls
• Nurse: there was some resistance to change, follow up was slow at
times
• Nurse Time consuming; confusing
• SW: Some clients have taken certain medications for so long that
they were unwilling / fear to change
• SW: Physicians were not receptive to letters received regarding
medications. Client unwilling to effect any changes if MD disregarded
letters/phone calls
• SW: Client’s confusion over meds; difficulty getting in touch with
person to discuss potential problems verbally over phone
• SW: Clients often gave minimal info; contacting doctors and having
them respond was difficult
• SW: Uncomfortable addressing this issue with MDs – feel it is
beyond my scope of practice
• Supervisor: Medication library in MSSP-Care
• SW: I was asked to call clients to tell her about potential harm of
taking 2 medications (due to side effects) but was not able to clearly
explain to client about why that was so. Someone from MM program
should make calls to the client to better explain and inform the client

220

Please provide any feedback or suggestions that you think would
improve the MM intervention

• SW: More training to SW staff on high risk medications and side
effects
• SW: Orientation for new employees – additional
assistance/orientation for non-medical staff (social workers). Also, I
never knew exactly what the expectations were of me as a
participant in MM. Some of my answers that are “disagree” have
more to do with the demands/constraints of my normal MSSP
workload than with MM’s goals/purpose/issues.
• Nurse: We need to continue with the med management program; I
think having a pharmacist reviewing meds is very helpful
• SW: More interaction with the team and training staff
• SW: Need to get MD’s interested, so they would cooperate better
• SW: Maybe more training on detecting what medications are right for
clients and which ones they should discontinue
• SW: The computer program is not user friendly. Needs major
improvement
• SW: Additional training would be helpful in regards to meds
• Nurse: Would like more contact with pharmacy staff
• Nurse: More organization; better timing
• SW: MM intervention more applicable if overseen by RN’s and not
SWs
• SW: More availability to collaborate with intervention staff
• SW: More tips on eliciting more MDs’ cooperation
• SW: Should be delegated to those who are more proficient in
medication management
• Supervisor: Improve medication library to make it more user friendly.
Also think of ways to involve pharmacists clients rely on for their
meds by bringing them into the loop
• SW: More training with the case managers on how to approach
client/ family/ doctor about the potential harm would be helpful

Key:
MM = Medication Management, also known as the CBM Intervention
Nurse = Nurse Care Manager
SW = Social Worker Care Manager
Supervisor = MSSP supervisor

221
Appendix M: Three Medication Management Protocols Created by Site #3

222

Medications that Can Cause Adverse Reactions

Drug Type Symptom of Adverse Reaction

sedatives, hypnotics,
antianxietals
sedation, confusion, Parkinsonian
symptoms, lethargy, falls

antidepressants anticholinergic effects, (diff. voiding,
delirium), orthostatic hypotension, falls

antipsychotics anticholinergic effects, sedation, confusion,
falls

cardiac medications weakness, lack of appetite, confusion

antihypertensives orthostatic hypertension, falls

diuretics incontinence, dehydration, weakness

anticoagulants bleeding (increased danger of falls)

analgesics and non-
steroidal anti-
inflammatory
medications
gastric distress and bleeding, dizziness,
confusion, depression

223
Decision Tree for Adverse Medical Reaction (AMR)

Develop complete medication list
Identify how person actually takes meds
Currently have
symptoms typical of
AMR?
Symptoms include
new confusion
incontinence, falls?
Problem meds?
More than 5 prescription
meds?
Multiple prescribers?
Determine if high
risk for AMR
Promote
communica-
tion with
MD
Review of med list by nurse,
clinical pharmacist before LOC
decision is made
Arrange MD review of
symptoms and
complete med list
Encourage MD review of
complete meds list
Is client misusing
meds. or non-compliant
with med regime?
Promote
communication
with MD
Ensure client/
caregiver can
follow regimen
Simplify
regimen
Obtain
medication
organizing
devices
Instruction in med
management
Consider non med
therapy
Obtain
financial
assistance
Client’s physician
Pharm. Company
Pt. Assist.
Program
Home health nurse
Pharmacist
Home health nurse
Promote
communication
NO YES
NO
YES
YES
NO YES
NO

224
Appendix N: Disclaimer

The contents of this dissertation are solely the responsibility of the
author and do not necessarily represent the official views of the United
States Administration on Aging.

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Translating two physical activity programs for older adults into home- and community-based settings: "Active Start" and "Healthy Moves for Aging Well"

PDF

Investigating the effectiveness of a social work intervention on reducing hospital readmissions among older adults

PDF

Palliative care: breaking the knowledge barrier among community-based older adults

PDF

New directions for family caregiver interventions

PDF

Forensic markers of physical elder abuse in medical and community contexts: implications for criminal justice interventions

PDF

Examination of the long-term psychosocial and functioning effects of a healthy living, life management behavior intervention for older adults

PDF

Translating evidence-based medicine into practice in critical care: an innovation study

PDF

The role of the modern aging network: measuring innovations of Area Agencies on Aging

PDF

Acculturation team-based clinical program: pilot program to address acculturative stress and mental health in the Latino community

PDF

The Filipino family initiative: preliminary effects of an evidence-based parenting intervention offered in churches on parent and child outcomes

PDF

A biodemographic approach to understanding sociodemographic disparities in kidney functioning on three dimensions: individual, population, and cross-national

PDF

Addressing federal pain research priorities: drug policy, pain mechanisms, and integrative treatment

PDF

Investigating a physiological pathway for the effect of guided imagery on insulin resistance

Asset Metadata

Creator Alkema, Gretchen Elizabeth (author)

Core Title Translating research into practice: a community-based medication management intervention

School Leonard Davis School of Gerontology

Degree Doctor of Philosophy

Degree Program Gerontology

Publication Date 04/19/2007

Defense Date 11/13/2006

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag care management,dually-eligible,medication problems,OAI-PMH Harvest,older adults,translational research

Language English

Advisor Crimmins, Eileen M. (committee member), Myrtle, Robert C. (committee member), Wilber, Kathleen H. (committee member)

Creator Email alkema@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m230

Unique identifier UC1473881

Identifier etd-Alkema-20070125 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-166147 (legacy record id),usctheses-m230 (legacy record id)

Legacy Identifier etd-Alkema-20070125.pdf

Dmrecord 166147

Document Type Dissertation

Rights Alkema, Gretchen Elizabeth

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu