University of Southern California Dissertations and Theses

New directions for family caregiver interventions

(USC Thesis Other)

New directions for family caregiver interventions

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content 1

NEW DIRECTIONS FOR FAMILY CAREGIVER INTERVENTIONS

by
Kylie N Meyer

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(GERONTOLOGY)
December, 2018

Copyright 2018 Kylie Nicole Meyer
2
Acknowledgements
I would like to thank the committee of scholars whose guidance helped to shape this
dissertation: Donna Benton, Susan Enguidanos, Zachary Gassoumis, and Kathleen Wilber. Your
support and feedback have been invaluable. I would like to especially acknowledge Dr. Benton,
for your perspective and critical insight on caregiving. I am incredibly grateful for the support of
Dr. Wilber. Thank you for your encouragement, the ample opportunities you have provided, and
for sharing not only your academic insights on caregiving, but also your personal ones.
While at USC, I have benefited from generous peer mentors, including Jeanine
Yonashiro-Cho, Jaclyn Portenova, Lucila Torres, and each of my Secure Old Age Lab team
mates. Thank you for your support and encouragement.
I also want to thank members of the California Task Force on Family Caregiving for
sharing their vast knowledge of caregiving. Three members deserve special recognition:
Kathleen Kelly for her forward-looking perspective, Sandi Fitzpatrick for her policy savoir-faire,
and Robert Lesh, a dedicated caregiver who also promised to read this dissertation.
Dr. Gassoumis’s name bears mentioning again: your keen intellect and statistical
savviness is only surpassed by your booming laugh. Thank you for generously sharing both.
Finally, considerable gratitude is owed to Kelly, Matt, and Jaclyn for your patient ears
and kind words throughout this process. Most importantly, I am so incredibly fortunate to have
the overwhelming support of my parents. There is no greater gift.

3
Funding
This dissertation was made possible through the financial support of the USC Provost
Doctoral Fellowship, USC Final Year Fellowship, the California Department of Health and
Human Services, the National Institutes of Justice, AARP California, and Archstone Foundation.
The opinions, findings, and conclusions or recommendations expressed in this document are
those of the author and do not necessarily reflect those of the aforementioned organizations.

4
Table of Contents
NEW DIRECTIONS FOR FAMILY CAREGIVER INTERVENTIONS ..................................... 1
Acknowledgements ......................................................................................................................... 2
Table of Contents ............................................................................................................................ 4
Chapter 1: Introduction ................................................................................................................... 5
Chapter 2: What are the characteristics of caregivers using online support services? ................. 22
Chapter 3: What can elder mistreatment researchers learn about primary prevention from family
violence intervention models? ...................................................................................................... 41
Chapter 4: The effect of number of years caregiving on physician visits and hospitalizations.... 61
Chapter 5: Building the policy case to scale caregiver interventions ........................................... 73
Chapter 6: Discussion and Conclusion ......................................................................................... 89
References ................................................................................................................................... 104
Tables .......................................................................................................................................... 131
Figures......................................................................................................................................... 142

5
Chapter 1: Introduction
There are approximately 18 million family caregivers in the U.S. (Schulz et al., 2016).
They are the family members and friends who assist older adults with an illness or disability, and
often make it possible for the people they support to remain living in the community. When they
begin this role, caregivers are often underprepared its many demands, and are faced with a range
of challenges, including: difficulty accessing services and information (Moon, 2017; Samia,
Hepburn & Nichols, 2012; Reinhard, Levine & Samis, 2012), trouble juggling employment and
caregiving (AARP Public Policy Institute [AARP] & National Alliance for Caregiving [NAC],
2015), and declines in mental and physical health (e.g., Capistrant, 2016; Monin, Schulz,
Martire, Jennings, Lingler & Greenberg, 2010). There are now hundreds of evidence-based
interventions to support family caregivers in this role and prevent negative consequences that can
result from caring for an aging family member (e.g., Gitlin & Hodgson, 2015). However, despite
decades of research to develop and evaluate evidence-based interventions, existing programs
remain inadequately tailored to the needs of caregivers and few caregivers ever access programs.
This dissertation explores promising intervention approaches—including mode of service
delivery, targeted outcomes, and timing of program administration— that can increase the value
of and access to interventions that can alleviate some of the challenges experienced by family
members taking on a caregiving role.
Who are Family Caregivers?
America’s 17.7 million family caregivers are the relatives, spouses, partners, friends, and
neighbors who have a significant relationship with, and who provide a broad range of assistance
to, an older person or an adult with a chronic or disabling condition (Reinhard, Given, Petlick &
Bemis, 2008; Schulz et al., 2016). The ranks of caregivers are expected to grow with the aging
Baby Boomer cohort (Ortman, Velkoff & Hogan, 2014), the increasing amount of time older
6
adults spend living with long-term chronic conditions (Crimmins & Beltrán-Sánchez, 2011), and
the shift to providing long-term services and supports in the community rather than in facilities
(Fields, Anderson & Dabelko-Schoeny, 2011). Not only is the number of caregivers growing, but
the typical caregiving role is changing for many of the same reasons. Nearly half of caregivers to
individuals with multiple conditions report doing at least one medical or complex care tasks
(Reinhard et al., 2012), a consequence of increasing levels of need combined with restricted
healthcare services (Levine, Halper, Peist & Gould, 2010).
It is challenging to describe the exact number and characteristics of family caregivers. To
provide a “whole picture” description of this population, I primarily draw upon data from two
nationally-representative sources: 1) NAC’s & AARP’s 2015 survey of caregivers to individuals
aged 50 and older and 2) analyses of the National Study on Caregiving presented in the 2016
report for the National Academies of Sciences, Engineering, and Medicine (Schulz et al., 2016).
Findings indicate that caregivers are, on average, 50.3 years-old (NAC & AARP, 2015), and
over half are between ages 45 to 64 (Schulz et al., 2016). Notably, there is an increasingly-
recognized population of younger caregivers. One-in-four caregivers are Millennials, those born
from 1980 to 1996 (Flinn, 2018). The number of women who are caregivers is disproportionate
to their percentage of the population. Approximately sixty percent of caregivers are women
(NAC & AARP, 2015). Among those providing assistance to someone with high care needs (i.e.,
someone with a cognitive impairment and/or requiring assistance with multiple self-care tasks),
this figure rises to 70%. At the same time, in recent cohorts, men comprise nearly half of all
caregivers (47%) (Flinn, 2018). Sixty-two percent of caregivers are White/Caucasian, 16% are
Hispanic, 13% are African American/Black, and 7% are Asian (NAC & AARP, 2015). These
proportions, however, are drastically different among younger caregivers; just 17% of Millennial
caregivers are White/Caucasian. Spouses comprise between one-tenth and one-fifth of
7
caregivers, and children make up little over half of caregivers to older adults (NAC & AARP,
2015; Schulz et al., 2018). The average length of time in the caregiving role is 3.7 years (NAC &
AARP, 2015), but this varies considerably, and many caregivers perform this role multiple times
during their life (Schulz et al., 2018). Approximately one-third of caregivers assist a person with
dementia (NAC & AARP, 2015). Nearly half of all caregivers indicate they felt they had no
choice in taking on this role (49%) (NAC & AARP, 2015).
While averages like those described above are informative, when discussing caregiver
interventions, it is important to note that caregivers are a diverse group in terms of culture,
background, capacity, and need for support.
Consequences of Caregiving
While many family members describe their caregiving experience positively (Cheng,
Mak, Lau, Ng & Lam, 2015), at least some caregivers experience negative consequences to their
health and/or financial wellbeing attributable to caregiving.
Health consequences of caregiving
Descriptions of poor health among caregivers have been difficult to ascertain in the past
given a lack of longitudinal data on this population. Frequently, studies have found caregivers
are healthier than non-caregivers, reflecting the fact that those who are unhealthy select out of
this role and those who experience health decline often relinquish caregiving responsibilities
(Fredman, Lyons, Cauley, Hochberg, Applebaum, 2015; Schulz et al., 2016). However, there is a
growing body of research on the health outcomes of caregiving which indicate this role
contributes to: an increased risk of developing depressive symptoms (Capistrant, 2016; Pinquart
& Sorenson, 2003), an increased risk of developing cardiovascular disease (Monin et al., 2010),
compromised immune functioning (Damjanovic et al., 2007), and a higher risk of experiencing
musculoskeletal injury from engaging in tasks such as lifting and transferring the care recipient
8
(Darragh, Sommerich, Lavender, Tanner, Vogel & Campo, 2015). There is some evidence that
caregiving increases risk of mortality (Schulz & Beach, 1999), however this finding has not been
consistent across studies (Fredman et al., 2015).
Causes of poor health among caregivers vary. Compared to non-caregivers, caregivers
report a higher odds of engaging in negative health behaviors including smoking, eating fast
food, and drinking sugary beverages than non-caregivers (Hoffman, Lee & Mendez-Luck, 2012).
A variety of stressors have been implicated as causes of health decline among caregivers.
Caregiver burden—the perceived stress and negative appraisal of social, financial, emotional,
physical, and spiritual outcomes attributed to caregiving (Zarit, Reever, & Bach-Peterson,
1980)—affects 40% of caregivers and has also been linked to a range of clinically-relevant
outcomes (Adelman, Tmanova, Delgado, Dion & Lachs, 2014; NAC & AARP, 2015). For some
caregivers, this role is associated with interpersonal challenges related to wellbeing, including
reduced relationship satisfaction with the care recipient (Monin & Schulz, 2009), increased
family conflict (Zarit & Edwards, 2008), and social isolation (Gwyther, 1998).
Financial consequences of caregiving
In addition to effects on health, caregiving often contributes to serious financial risks and
consequences. When caregivers leave the workforce to provide care, a step taken by
approximately 20% of caregivers to older adults either permanently or temporarily (NAC &
AARP, 2015), their financial wellbeing is seriously compromised. Women who leave the
workforce to provide care are 4.3 times as likely to experience poverty as non-caregivers who do
so (Wakabayashi & Donato, 2006). The estimated cumulative cost of leaving the workforce to
provide care ranges in the hundreds of thousands of dollars-worth of lost retirement savings and
Social Security benefits (MetLife Mature Market Institute, 2011). Caregivers may be particularly
at risk of economic harm during periods of financial downturn; compared to non-caregivers,
9
caregivers were 51% more likely to have lost a job during the Great Recession (Meyer,
Gassoumis & Wilber, in review). Further, the out-of-pockets costs of caregiving are substantial,
averaging nearly $7,000 per year (Rainville, Skufca & Mehegan, 2016). For caregivers to a
person with dementia, the average out-of-pocket cost of caregiving is closer to $11,000 per year.
Adding to caregivers’ financial vulnerability, in the U.S. there are few policies available
to protect family caregivers from financial harm. Just 60% of caregivers are eligible for job
protections under the Family Medical Leave Act, and just five states offer a form of paid leave to
employees who need to take time off to provide assistance to a family member (Findlay, 2016).
Even in these states, awareness of paid leave benefits is low. Just one-third of California voters
are aware of the state’s paid leave program (DiCamillo & Field, 2015). Human resource
professionals also describe insufficient understanding of these programs; 78% of HR
professionals indicate interest in training on paid leave (Andrew Chang & Company, 2015).
Unprepared Caregivers
Also concerning the lack of preparedness with which caregivers enter this role. A veteran
aging and long-term care researcher described her experience when taking on a caregiving role
herself: “As [my husband’s] sole caregiver, I found myself plunged into a confusing world of
poorly coordinated care, confusing systems, and an expectation that the caregiver could take on
full-time responsibilities” (Moon, 2017). Or, as another caregiver more colloquially described
her experience, “We are flying by the seat of our pants” (Samia et al., 2012).
Information and training are among the most pressing needs of caregivers. Over 80% of
caregivers indicate they need more information on caregiving-related topics (NAC & AARP,
2015). Topics on which caregivers most often need additional information and training include:
the care recipient’s condition, the care recipient’s prognosis, services and supports for caregivers
and care recipients, and financial assistance (Brossoie, Roberto, Willis-Walton & Reynolds,
10
2011; Li, 2015; Washington, Meadows, Elliott & Koopman, 2011).

There is, however,
considerable variation in the type of training caregivers require. For example, caregivers who
spend a greater amount of time caregiving per week are more likely to report an interest in
learning about stress management (52% compared to 42% among caregivers in general) (NAC &
AARP, 2015). Family members assisting someone with complex care tasks require guidance
with practical tasks related to care. Nearly half of family caregivers perform complex
care/nursing tasks (e.g., medication management, wound care, preparing special diets), yet many
face challenges when learning how to do so (Reinhard et al., 2012). For example, while 78% of
family caregivers who provide complex care perform medication management, 66% found this
task to be challenging. Approximately 29% of those managing medications reported this task to
be challenging due to fear of causing harm, and 24% reported an additional need for training.
Most family caregivers (61%) report learning how to do this task on their own.
Perhaps one of the most difficult aspects of keeping caregivers informed and adequately
trained is that caregiver needs evolve over time, such as when the needs of the care recipient
escalate and when the caregiving network changes (e.g., change in availability of formal and
informal assistance) (e.g., Schulz, 2013). Consequently, static, low-intensity information
resources such as fact sheets are unlikely to adequately equip caregivers who are in this role for
multiple years.
Evidence-Based Caregiver Interventions
The challenges experienced by family caregivers have not gone entirely unrecognized. As
required levels of care for older family members exceeded what would previously have been
considered an expected part of familial duties for an increasing number of Americans,
interventions were developed to alleviate challenges associated with the caregiving role. The
earliest interventions were developed in the 1960s and have continued to evolve (Gitlin &
11
Hodgson, 2015). As described by the National Academies of Sciences Engineering and Medicine
report, Families Caring for an Aging America, interventions for caregivers have been developed
at multiple levels of society (Schulz et al., 2016). At the highest level, societal interventions
include job protections under the Family Medical Leave Act (1993) and the National
Alzheimer’s Project Act to support research on caregivers to people with Alzheimer’s Disease
(National Alzheimer’s Project Act, 2011). Organization-level interventions are less broad, but
still affect multiple potential caregivers, such as employee assistance programs to provide care
management. This dissertation is primarily focused on individual-level, evidence-based
interventions. Evidence-based interventions, specifically, are interventions that have been
systematically evaluated and consistently render positive outcomes (e.g. improved health) that
can be attributed primarily if not entirely to the intervention (National Association for Social
Workers, n.d.). Evidence-based interventions ensure that resources are well-used by raising the
likelihood of positive outcomes.
There is growing consensus on key components to include in efficacious evidence-based
caregiver interventions. Evidence-based caregiver interventions most often include
psychoeducational approaches (e.g. coping with stress, learning about the recipient’s condition),
behavior management and skills training (e.g. lifting the care recipient, developing problem
solving skills), and relaxation training (e.g. yoga and meditation). Provision of information,
education and training are central to most evidence-based caregiver interventions (Schulz et al.,
2016). Psychoeducational approaches have consistently shown promise in pooled studies, and
are especially efficacious at reducing burden and depression (e.g., Gitlin et al., 2003; Pinquart &
Sörensen, 2006).

Multicomponent interventions, where education and training modules are
offered in consortium with other supports (e.g. counseling), are increasingly common and show
considerable promise (Gitlin & Hodgson, 2015; Landry & Keller-Allen, 2017; Schulz et al.,
12
2016). To illustrate some of the different types of multicomponent, psychoeducational
interventions that exist, below is a brief overview of two interventions, Savvy Caregiver and
Powerful Tools for Caregivers. Both programs are notable for having been successfully
replicated and scaled.
Savvy Caregiver. Savvy Caregiver is a psychoeducational intervention developed for
caregivers to people with dementia. The creators of the program posit that knowledge about
dementia, developing the right outlook, and appropriate caregiving skills will prevent deleterious
effects of caregiving (Hepburn, Lewis, Tornatore, Sherman & Bremer, 2007; Kally et al., 2014).
Improving the appraisal of the caregiving situation is primary to the intervention, followed by
acquisition of problem-solving skills. Savvy Caregiver is delivered through 6 two-hour
classroom-style sessions. Topics covered include: acknowledging the presence of dementia,
understanding more about dementia, developing emotional tolerance while caregiving, taking
responsibility for another adult, establishing realistic care goals for caregiving, engaging and
communicating with the recipient, and problem solving.
Powerful Tools for Caregivers. Powerful Tools for Caregivers (Powerful Tools) is another
psychoeducational intervention program, but it is not tailored for caregivers to people with
dementia. Like the Savvy Caregiver program, Powerful Tools is delivered over a 6-week period
and sessions last approximately 2 hours each. Powerful Tools is based on social learning theory
and cognitive behavioral approaches, where improved self-efficacy from program participation
and learning is posited to yield long-term behavior change (Boise, Congleton & Shannon, 2005).
In this case, the desired behavior change is improved self-care and caregiving skills. During
classes, topics include locating resources, managing stress, communication skills, emotional
coping and relaxation techniques, and self-care.
13
Evidence-based caregiver interventions positively impact a range of outcomes. To date,
these outcomes include:
• Reduced depression (Griffiths, Whitney, Kovaleva & Hepburn, 2016; Gitlin et al.,
2003; Kally et al., 2014; Smith & Bell, 2005);
• Reduced anxiety (Griffiths et al., 2016);
• Improved self-care behaviors (Boise et al., 2005; Kuhn, Hollinger-Smith, Presser,
Civian & Batsch, 2008);
• Reduced caregiver burden (Griffiths et al., 2016; Gitlin et al., 2003;
Savundranayagam, Montgomery, Kosloski & Little, 2011);
• Reduced frequency of behavioral symptoms of dementia (e.g. wandering) (Griffiths et
al., 2016);
• Improved reaction to disruptive behaviors (Griffiths, Whitney, Kovaleva, & Hepburn,
2016; Kally et al., 2014);
• Improved caregiver competence and self-efficacy (Boise et al., 2005; Kuhn et al.,
2008);
• Increased knowledge of community resources (Boise et al., 2005);
• Reduced job stress (Kuhn et al., 2008).
Results from meta-analyses indicate that caregiver interventions have an overall small to
moderate effect on frequently-measured outcomes including depression and burden (Gitlin,
Marx, Stanley & Hodgson, 2015; Gitlin & Hodgson, 2015). Although modest, effect sizes are
comparable and, in some cases, exceed those of Food and Drug Administration-approved
pharmacological therapies to treat depression (Shedler, 2010).
14
Caregiver interventions have also demonstrated cost-effectiveness at the state-level. An
evaluation of Mittlman’s New York University Caregiver Intervention, where caregivers receive
individual and family counseling, found that those receiving care from someone who participated
in the program were less likely to be admitted to a nursing home (Foldes, Moriarty, Farseth,
Mittelman & Long, 2017). The authors estimate that by keeping 5-6% more Medicaid-eligible
care recipients in the community per year through the provision of the intervention, Minnesota
could render a $40.4 million Medicaid cost savings over 15 years.
Limitations in Existing Caregiver Interventions

Despite demonstration of their considerable benefits to individual caregivers and state
programs, just a fraction of caregivers ever access evidence-based interventions (Gitlin &
Hodgson, 2005; Schulz et al., 2016). The exact proportion of caregivers who have participated in
intervention programs is difficult to ascertain given: 1) previously described discrepancies in the
size of the caregiving population; 2) the tendency among users of caregiver services to be repeat
clients (Mensie & Steffen, 2011), and 3) and the patchwork network of organizations serving this
population (see Schulz et al., 2016). However, a preliminary idea of the very low proportion of
caregivers accessing evidence-based interventions is illustrated through data from California’s
Department on Aging, a state unit on aging which serves the highest population of caregivers in
the country in absolute numbers under the National Family Caregiver Support Program.
1
In
2014, the most recent year on which data are posted, 13,646 caregivers in the state received
support services including counseling, support groups, and training, including access to
multicomponent evidence-based interventions such as Savvy Caregiver (California Department
on Aging, n.d.). Given an estimated population of 4.45 million caregivers in the state (Reinhard,

1
See Chapter 5 for further discussion on the structure and availability of government-funded caregiver
support programs.
15
Feinberg, Choula & Houser, 2015), this means that just 0.31% California caregivers received
access to caregiver programming that year through the Department on Aging. Even if some
caregivers had already received services in previous years and/or did not wish to use formal
services, this figure is extremely low.
There are many reasons caregivers do not access evidence-based intervention programs
despite demonstrated need within this population. However, a major cause for limited access is
that very few evaluated interventions have ever been translated and scaled in community
settings. Gitlin & Hodgson (2015) identified more than 200 interventions for caregivers to people
with dementia found in systematic reviews and meta-analyses published from 1966 and 2013,
and found only a handful of these interventions have been scaled to any degree. This is resonant
with other areas of health research; it is estimated to take 17 years on average to translate health
research to practice and policies (Morris, Wooding, & Grant, 2011). At the same time, evidence-
based interventions have been under development for decades, suggesting there that caregiver
intervention research is uniquely hindered at the translational stage.
Part of the reason translation and scaling efforts have been marred is due to the value—
both perceived and actual— of existing programs. Programs frequently require caregivers to
invest hours of their time over weeks and months into intervention activities, and service
organizations must front a high cost to provide many evidence-based intervention programs. For
example, the cost of providing an 8-session intervention (6 in-home and 2 telephone contacts)
delivered by occupational therapists to people with dementia and their families was $941.63 per
dyad (Gitlin, Hodgson, Jutkowitz & Pizzi, 2010). A cost evaluation of the Resources for
Alzheimer's Caregivers Health II intervention revealed an average cost of $1,214 per dyad to
provide 9 home-visits, 3 telephone consults, and 5 telephone support groups delivered by
supervised interventionists (Nichols et al., 2008). While both evaluations found that individual
16
caregivers saved time through spending less time caregiving after completing interventions, the
upfront cost placed upon organizations is beyond the reach for many service organizations which
administer caregiver interventions in the community.
Organizations also have to weigh the potential value of intervention programs for the
clients they serve, and consider their appeal to caregivers whose time is already limited by care
duties. Several weaknesses in existing programs undermine their worth. Demographic
descriptions of the caregiving population reveal considerable diversity in the characteristics and
experiences of caregivers, and yet interventions are rarely designed to integrate these differences.
Specifically, few interventions have been adapted and designed to meet the needs of a
racially/ethnically and culturally diverse population of caregivers (Gitlin & Hodgson, 2015;
Gitlin et al., 2015; Landry & Keller-Allen, 2017). Phase within the caregiving trajectory is
another characteristic that has been neglected in caregiver intervention research, despite its
obvious relevance to the caregiving experience. Interventionists are increasingly calling for
programs to be designed around phases or critical periods within the caregiving journey, such as
following a health event, care transition, new diagnosis, change in the level of care, and
caregivers’ readiness (Cameron & Gignac, 2008; Schulz, 2013; Wald, Fahy, Walker &
Livingston, 2003). And yet there is little practical guidance on how to tailor interventions based
on variation in the caregiving trajectory. Further, the intended outcomes of existing interventions
merit critique. Interventions, to date, have focused on a narrow range of psychosocial outcomes
that are not necessarily the most relevant to caregivers’ experiences (Gitlin & Hodgson, 2015;
Gitlin, et al., 2015; Landry & Keller-Allen, 2017). Savings in time and costs related to
caregiving, successful access to available benefits and job protections, and preserved caregiver
health are likely more germane to families than oft-measured outcomes such as caregiver burden,
competence, and reaction to problem behaviors (Gitlin et al., 2010; Johnson, Hofacker, Boyken
17
& Eisenstein, 2016). Interventions with outcomes which benefit the care recipient are also likely
to appeal to caregivers, who typically report a desire to provide quality care (Gitlin & Hodgson,
2015; Scharlach, Kellam, Ong, Baskin, Goldstein & Fox, 2006).
New Directions for Caregiver Intervention
This dissertation considers new directions for caregiver interventions to overcome
limitations in existing research, and, in doing so, illustrates the many directions interventionists
and researchers can take to develop evidence-based programs that are accessible and relevant to
caregivers’ experiences as this population grows and evolves. There are four main chapters to
this dissertation, each of which employs a different methodological approach to explore
promising ways to increase the value of caregiver interventions as well as access to these
programs. The forthcoming chapters describe new ways of delivering interventions and services,
promising approaches to address understudied outcomes, findings to support the case for early
administration of interventions, and a discussion on strategic approaches at the policy-level to
better scale interventions. The chapters of this dissertation are intended to guide the translation of
caregiver interventions so that they benefit those in this role, rather than accumulate in the
archives of peer-reviewed scholarly journals without benefiting caregivers. Below, I briefly
describe each the chapters in this dissertation in greater detail.
In Chapter 2, I describe the characteristics of caregivers logging in for online support
services through the Family Caregiver Alliance CareJourney (FCA CJ) program using
quantitative analyses of client data. Learning more about caregivers who are willing to access
services and interventions online is important because: 1) online service delivery modes may
allow service providers to serve a greater number of caregivers through cost savings compared to
traditional modes of service delivery (Blom, Zarit, Zwaaftink, Cuijpers, & Pot, 2015; Czaja et
al., 2006), and 2) online-delivered services may reach populations of caregivers who typically do
18
not approach social service programs (Navaie, 2011). The potential of online-delivered services
to accomplish this is finally a reality as secure web 2.0 technologies become more available and
as caregivers are increasingly comfortable using these technologies. However, little is known
about the characteristics of caregivers who are most likely to use online-delivered supports over
usual services when both options are available. Leaders of organizations contemplating
investment in online support services and interventions need this information to understand how
shifting to online services will 1) affect access to services among different segments of their
caregiving clientele and 2) to make decisions about what types of content to provide using online
versus usual service delivery modes. For example, if caregivers who use online services are more
likely to be employed, it may be all the more pertinent to provide information about the Family
Medical Leave Act and Paid Family Leave options online. The findings from this chapter can be
applied to understand which caregivers would be most likely to use online-delivered
interventions provided by community-based organizations.
Chapter 3 describes the structure and components of an intervention to prevent elder
mistreatment of those receiving care from a family member or friend by drawing on literature of
evidence-based practices in other fields of family violence. This chapter makes an important
addition to the field of caregiver interventions because elder mistreatment has seldom been
evaluated in the hundreds of home-based interventions for family caregivers despite heightened
rates of mistreatment in caregiving dyads compared to population samples generally. In a sample
of caregivers assisting an older adult with dementia, nearly half of caregivers reported engaging
in at least one form of mistreatment (Wiglesworth, Mosqueda, Mulnard, Liao & Gibbs, 2010).
By comparison, estimated annual rates of mistreatment among older adults generally is 16%
(Yon, Mikton, Gassoumis, & Wilber, 2017). If implemented, the intervention described in this
chapter could prevent mistreatment of care recipients by providing caregivers with tools to cope
19
with challenges that may be associated with heightened risk of elder mistreatment (e.g., access to
respite and training on stress management techniques to address high levels of caregiver burden).
The focus of this intervention on supporting caregivers to provide safe and high-quality care may
be particularly valued by families. As previously discussed, family members and friends
frequently begin caregiving with little preparation on how to handle the practical, psychological,
and social challenges that can accompany this role, sometimes to dangerous ends.
Next, in Chapter 4, I make the case for administering health promotion intervention
programs early in the caregiving career using secondary data on spousal caregivers from the
nationally-representative Health and Retirement Study. To do this, I longitudinally examine
variation in rates of health service use based on the number of years spent in the caregiving role,
and compare these rates to those of spouses who were not providing care. I examine two types of
health services broadly representing preventive and acute health services: rates of physician
visits and whether participants experienced a hospitalization. Chapter 4 is unique from previous
chapters in that findings serve a dual purpose: in addition to supporting the need for early
intervention to protect caregivers’ health, this chapter helps to build the policy case for greater
access to publicly-funded caregiver interventions by estimating the additional costs associated
with heightened rates of health service use among spousal caregivers, much of which is paid for
using the federal Medicare and Medicaid programs. As discussed in Chapter 5, identifying a
potential federal cost-savings rendered through intervention delivery is a promising path for
increased federal funding that is critical to scaling caregiver interventions.
Finally, Chapter 5 looks beyond intervention delivery, design, and administration to
consider policy barriers to scaling interventions that could otherwise support caregivers faced
with the challenges and consequences of caregiving for an aging family member or friend. This
chapter considers two theories of policymaking in order to identify actionable steps to expand
20
access to caregiver interventions. I apply policy-making tools identified by Sandra Levitsky
(2014) from her focus group study with family caregivers and service providers to John
Kingdon’s (1984) revised garbage can model of how issues become a part of the policy agenda.
The recommendations derived from this exercise can be used to create and expand funding
mechanisms to provide access to evidence-based caregiver intervention programs.
In no way do the chapters of this dissertation discount the important work that has
already been done to develop and improve existing caregiver interventions. However, by
addressing widely-recognized weaknesses within existing interventions that prevent their
translation into community settings (e.g., Gitlin et al., 2015), researchers can work towards better
meeting the needs of today’s caregivers using one or more of the approaches described in the
following chapters. Nor is this dissertation is not intended to be a “roadmap” to improve
caregiver interventions, although Chapter 5 could be used this way for the purpose of improving
efforts to scale interventions at a policy-level. Instead, the chapters in this dissertation can be
best used by researchers and funders to identify untried or under-considered approaches to
improve upon existing evidence-based interventions by making them more accessible to family
caregivers. The so-called new directions explored in each chapter are again highlighted below, as
well as a short description of each chapter.
• Chapter 2. Description: Evaluation of an online mode of service delivery that compares
the characteristics of caregivers accessing services online or using traditional modes (e.g.,
by telephone). New direction or approach: Intervention delivery mode.
• Chapter 3. Description: Proposal for the structure and components of a first-of-its-kind
elder mistreatment prevention program using promising practices from other fields of
family violence. New direction or approach: Intended intervention outcome.
21
• Chapter 4. Description: Examination of variation in health service use by the number of
years spent as a caregiver. New direction or approach: Timing of intervention
administration.
• Chapter 5. Description: Critical analysis of policy barriers to scaling interventions, and
application of two policy theories to identify actionable steps to expand access to
caregiver interventions. New direction or approach: Policy strategies to improve access
to interventions.
By the end of this dissertation, readers should have 1) an increased awareness of barriers
to translating behavioral interventions so they are accessible to family caregivers to older adults
and 2) ideas about how interventions can change to become both more accessible and valuable to
family caregivers. Further, it is my hope that readers will be left with a sense of “skeptical
optimism” wrought from awareness of the many opportunities to more effectively put to use
decades of research on caregiver interventions that can address the needs of this growing
population.
Some parts of this chapter were previously reported in: Meyer, K., Kaiser, N., Benton,
D., Fitzpatrick, S. Gassoumis, Z., Wilber, K., & the California Task Force on Family
Caregiving (2018, July). Picking Up the Pace of Change for California’s Family Caregivers: A
Report from the California Task Force on Family Caregiving. Los Angeles, CA: USC Leonard
Davis School of Gerontology.

22
Chapter 2: What are the characteristics of caregivers using online support services?
Caregivers to older adults often encounter new information and education needs when
adopting a caregiving role. Those providing assistance to an older family member or friend with
an illness or disability need information tailored to their recipients’ condition, degree of disease
progression, and their own readiness to access information (Wald, Fahy, Walker, & Livingston,
2003; Washington, Meadows, Elliott, & Koopman, 2011). Interventions with education
components to meet caregivers’ information needs improve psychological wellbeing (Sörensen,
Pinquart, & Duberstein, 2002). Increasingly, these information and education needs are met
using online technologies (Hopwood et al., 2018).
As of 2016, 70% of caregivers to people aged 50 and older were interested in using
technology to receive personalized information to assist with their caregiving tasks (AARP,
2016), and the vast majority of caregivers in the U.S. have access to the internet (86% vs. 78% of
non-caregivers) (Fox, Duggan, & Purcell, 2013). Caregivers report using technology in a self-
directed fashion to seek information related to caregiving, including practical aspects of how to
provide care (AARP, 2016; Fox et al., 2013; Kernisan, Sudore, & Knight, 2010), to learn about
recipient health conditions (Kernisan et al., 2010; Peterson, Hahn, Lee, Madison, & Atri, 2016),
and to find ways to cope with stress from caregiving (AARP, 2016; Fox et al., 2013). Web-based
platforms are preferable to smartphone-based apps; 97% of caregivers are comfortable with
computers while 80% report being comfortable with tablets and smartphones (AARP, 2016).
From a provider perspective, technology can expand services available to caregivers
beyond what is possible using traditional modes of delivery (e.g., in-person, phone lines staffed
during business hours). Technology may lower the cost of delivering services so organizations
can afford to reach more caregivers with their budgets (Blom et al., 2015; Czaja et al., 2006). In
addition, services provided to caregivers online (i.e., online-delivered services) can encourage
23
new populations of caregivers to use formal support services to access high-quality information.
For caregivers who cannot easily access in-person services—like those who live in rural areas or
who are employed and cannot call during regular service hours—technology may further enable
access to information (Navaie, 2011). Further, technology-based interventions for caregivers to
older adults such as supportive peer and professional messaging, web video education, and
specialized caregiver websites, have demonstrated favorable outcomes (Boots, Vugt,
Knippenberg, Kempen, & Verhey, 2014; Hopwood et al., 2018; McKechnie, Barker, & Stott,
2014).
Of particular interest in social service settings are technologies that tailor educational
resources and information to meet the needs of family caregivers to older adults. Caregivers to
older adults often have trouble finding the information they need through self-directed searches
(Funk, Dansereau, & Novek, 2017; Hopwood et al., 2018). Among caregivers searching an
online site developed for caregivers, just 57% indicated they found what they were looking for
(Kernisan et al., 2010). Social service organizations can play a role in streamlining and
simplifying access to appropriate information by offering web-based platforms with information
and resources that are curated for caregivers and which utilize algorithms that match caregivers’
needs and characteristics with recommended information (AARP, 2016; Gaugler, Reese, &
Tanler, 2015).
Given new opportunities for organizations to include technology-based services in their
repertoires, it is important to understand which caregivers are likely to utilize these services. If
new technologies do not reach caregivers with high levels of need or traditionally underserved
populations, organizations may need to weigh how much to invest in these services or at least
how to better tailor and market these tools for the community they serve (Fogel, Ribisl, Morgan,
Humphreys, & Lyons, 2008; Kovaleva, Blevins, Griffiths, & Hepburn, 2017). Moreover, the
24
caregiving population changes over time, as new caregivers begin this role and others leave
(Wolff & Kasper, 2006); successive cohorts of caregivers bring with them distinct attitudes and
experiences, and thus caregivers’ interest in technological supports is likely to increase.
Background and Objectives
Survey data provide some information on the characteristics of caregivers most likely to
use online-delivered services when these are available. Survey research consistently shows
younger caregivers are more likely to access caregiving information online (Fox et al., 2013;
Kim, 2015; Li, 2015). Some suggest there may be differences in online service access by race
and/or ethnicity (Fox et al., 2013), however, the literature is not consistent in this finding (Kim,
2015; Li, 2015). Socioeconomic status, measured by both income and educational attainment, is
consistently found to be a predictor of online information-seeking among caregivers (Fox et al.,
2013; Kim, 2015; Li, 2015). Although it is conceivable that employed caregivers would be
disproportionately interested in using online services given the ability to access information at
any time, previous research does not always demonstrate this (Li, 2015). Finally, although few
studies report on which aspects of the caregiving situation predict the use of online information
services, Li (2015) observed higher rates of use among non-primary caregivers. Similarly, Kim
(2015) found that spending a greater number of hours caregiving per week was associated with
reduced likelihood of using the internet to access health information related to caregiving. Kim
(2015) also found caregivers experiencing higher levels of emotional stress were more likely to
use the internet to access health information.
There are several limitations to existing research on caregivers’ use of online services
that undermine the application of previous findings in social service settings. First, it is unclear
which caregivers are more likely to use online services to access tailored information when both
online and usual service delivery modes are available in social service settings. Secondly, many
25
available online support programs are targeted at caregivers of people with dementia, rather than
the general caregiver population (Beauchamp et al., 2005; Gaugler et al., 2015; van der Roest et
al., 2010). Thus, conclusions from previous research are not generalizable to those assisting older
adults living with other types of conditions. Finally, both available technologies and the
population of caregivers are constantly evolving. For service providers seeking to expand the
reach of information services using technology, it is necessary to have current information on
which caregivers are likely to use web-based services and which caregivers opt for usual modes
of service delivery.
To understand which caregivers are most likely to use online services in social service
settings, this article explores the characteristics of caregivers using a new online caregiver
service delivery program called Family Caregiver Alliance Care Journey (FCA CJ), a program of
the Family Caregiver Alliance. FCA CJ delivers individualized information and resources to
caregivers based on caregivers’ responses to questions in a digital assessment tool. FCA CJ is
being used by local organizations serving caregivers in two large California metropolises as an
alternative mode of service delivery—one in northern California (Site 1) and one in southern
California (Site 2). Clients at both sites have the option of using FCA CJ or usual modes of
services delivery (e.g., by phone) to access information at no charge and while using the same
service deliver model, described below. Over the course of the evaluation period examined here,
advertisement for FCA CJ included a press release, information placed on each service site’s
home webpage, public educational events for caregivers, a postcard mailed to existing clients,
newsletter notifications, and notification by program staff to incoming clients about the option to
complete assessment information online.
Using service assessment data on caregivers receiving services through FCA CJ as well
as those accessing services through the usual mode, we explore the question: What are the
26
characteristics of caregivers choosing to use online-delivered social services compared to those
of caregivers accessing services delivered by usual modes?
The FCA CJ program
FCA CJ is a web-based support tool for family caregivers. Caregivers log in to the
program with an email address and complete an intake form that requests basic information
about themselves (e.g., demographics) and their caregiving situation (e.g., relationship to the
person they assist). To be eligible for both usual- and online-delivered services, caregivers must
provide assistance to someone with multiple chronic conditions or a cognitive impairment who is
living within the service region.
For decades, service specialists, called family consultants, at each service site have
administered caregiver intakes and assessments to eligible clients by phone or in-person. After
receiving an assessment administered by a family consultant, caregivers are mailed a packet of
information pertaining to their caregiving situation. Depending on the caregiver’s assessed
needs, mailed packets include items such as support group schedules, fact sheets, and
information on educational events. After receiving materials, caregivers may call their
designated family consultant at any time if their needs change.
Caregivers have had the option of enrolling in FCA CJ since September 2016 at Site 1
and since April 2017 at Site 2 as an alternative way of accessing services. Caregivers could learn
about and access FCA CJ through the website for each service organization and/or prompt from
a family consultant to consider completing intake forms online. Clients who access services
online through FCA CJ self-administer the intake and assessment, and are shown suggested
resources on a digital dashboard based on their responses. Caregivers using FCA CJ can contact
family consultants using a secure messaging program found within FCA CJ for additional
assistance. After assessment, caregivers at both sites receive monthly check-in calls and are
27
reassessed four months later. The service model used at both sites for online and usual services is
illustrated in Figure 2.1.
Conceptual model
Our primary research question was: how did the characteristics of caregivers using FCA
CJ differ from those accessing services delivered the usual way? To answer this question, we
applied Andersen’s model of healthcare utilization (Andersen, 1995). According to this model,
health service use is predicted by predisposing, enabling, and need-based factors. Although all
caregivers included in this study accessed services, these categories can assist with
understanding why a caregiver used one service delivery approach over the other.
We considered predisposing characteristics to using online-delivered services to be
primarily demographic, following previous applications of the Andersen model (Wu, Luo, Flint,
& Qin, 2015). Existing literature (AARP, 2016; Fox et al., 2013), suggested that younger people
would be more likely to use online-delivered services. We also explore whether the use of online
services might vary by race/ethnicity given the so-called digital divide posited to discourage
online service use among some racial/ethnic minority caregivers (Fogel, 2008).
Enabling characteristics include those factors that support the use of online services. As
markers of socioeconomic status, educational attainment and employment status imply access to
technologies needed to use FCA CJ (e.g., computer and internet), as well as comfort and skill
with using online technologies. We thought that employed caregivers might be more likely to use
online services because of restricted opportunities to access service by phone during traditional
business/service hours. Aspects specific to each study site were also believed to affect access,
such as persistence among service specialists in promoting FCA CJ and frequency of educational
events to promote online service delivery at each site. We also wanted to explore how the way
that caregivers learned about the services effect whether caregivers used online-delivered
28
services versus services delivered by phone or in-person. For example, we suspected that
caregivers who learn about services online will be more likely to use online-delivered services.
Finally, need-based factors pertain to those characteristics that necessitate a caregiver use
online- or usual-delivered services. We had conflicting ideas regarding how caregiving intensity
might affect service delivery preference. On the one hand, those with more intensive caregiver
duties, such as caring for someone with a cognitive impairment or greater functional disabilities,
might prefer online-delivered services given the opportunity to stop and start assessments when
responding to care demands. On the other hand, those with more demanding caregiving duties
may desire the therapeutic aspect of talking to a service specialist by phone. The model we
propose is illustrated in Figure 2.2.
Research Design and Methods
Data
Data were collected in two ways. Either data was entered by users who logged into the
system and complete a service assessment themselves, or service providers entered information
provided by clients over the phone. For this study, de-identified client data was accessed using a
password-protected data extraction form. The initial sample included client data from both sites
collected from September 2016 through June 2018 (N=797). We included among our study
sample only individuals who were caring for someone with multiple chronic conditions or
someone with a cognitive impairment (n=690). We excluded: (1) data entered during the first
four months of program implementation at Site 1 and the first two months of implementation at
Site 2 to minimize the impact of factors pertaining to implementation (e.g., programming errors,
training among service specialists) (n=62); (2) caregivers to recipients under age 60 (n=31); and
(3) caregivers whose primary language is not English, as FCA CJ is not currently translated into
29
other languages (n=57). After the application of inclusion and exclusion criteria, the analytic
sample was 540.
Measures
Dependent variable. The dependent variable in this study was whether the service
assessment was completed online by the caregiver (i.e., self-administered) or by a family
consultant (i.e., non-self-administered). This variable was used to indicate whether the caregiver
accessed services using usual means (i.e., by phone or in-person), or online using FCA CJ. This
information is collected automatically by the FCA CJ system.
Independent variables. Independent variables were those items collected on either the
intake or assessment forms, and were selected based on the Andersen (1995) model. (See Figure
2.2.)
Predisposing variables. Predisposing variables included age, gender, race/ethnicity, and
marital status. Caregiver age was calculated by subtracting each caregiver’s reported year of
birth from the calendar year when the caregiver approached services. Gender was coded as male
or female. Race and ethnicity were collected as a combined category, which allowed
Caucasian/White, African-American/Black, Asian, Latino/Hispanic, Native American, Pacific
Islander, and Other. Pacific Islander (n=5) responses were collapsed with the Other option to
avoid low expected cell frequencies in bivariate analyses (i.e., n<5). Marital status was coded as
“married” (married, domestic partners) and “unmarried” (single, divorced, separated, widowed).
Enabling variables. Enabling variables were divided into two sub-types, which we term
“individual-level” and “system-level” factors. Individual-level enabling characteristics included
employment status and educational attainment. In the assessment, employment status options
included full-time, part-time, leave, retired, and unemployed. We recoded those indicating they
were on leave as missing (n=8), again to avoid low expected cell frequencies. Caregivers were
30
asked to select their highest level of educational attainment, and chose among the options: some
high school, high school degree, some college, college degree, and post-graduate degree.
System-level enabling factors included both the service site (either Site 1 or Site 2) and how the
caregiver learned about services (through a healthcare provider, social service provider, online,
or another source [e.g., family and friends, public event]).
Need variables. Need-based factors included whether the recipient had a cognitive
impairment, the recipient’s functional ability, completion of medical tasks by the caregiver,
caregiver burden, and hours of caregiving per week. Recipients were coded as having a cognitive
impairment if caregivers indicated the recipient experienced memory problems or had any of the
following conditions: Alzheimer’s Disease, vascular dementia, Lewy Body dementia,
frontotemporal dementia, or dementia (non-specified). Caregivers were asked about whether they
provided assistance with any activities of daily living (ADLs), including eating, bathing,
dressing, grooming, toileting, and transferring (Katz, 1983). These were added such that scores
for needing any assistance with ADLs ranged from 0 to 6. Caregivers were asked whether or not
they completed medical tasks when assisting the recipient, and were given examples such as
wound care, administering medications, and preparing special diets. Burden was measured using
the 4-item version of the Zarit Burden screener instrument (Bédard, Molloy, Squire, Dubois,
Lever, & O'Donnell, 2001). Caregivers with a score of 8 or more were considered burdened. The
estimated number of hours during which caregivers provided assistance in a given week was
asked in 10-hour increments (e.g., 1 to 10). We recoded these as providing fewer than 20, 20 to
40, or greater than 40 hours per week.
Analysis
Analyses began with descriptive statistics, stratified by users of FCA CJ and the usual
service delivery mode. To explore differences in characteristics between caregivers using FCA
31
CJ over usual services, we applied bivariate statistics including Pearson chi-square tests and t-
tests. Before proceeding to regression models, we assessed contingency tables to ensure there
were no cells with expected frequencies of fewer than 5 to prevent biased estimates. Next, we
applied step-wise logistic regression models with use of online service delivery as the outcome.
Model 1 regressed predisposing factors on FCA CJ use. Models 2 and 3 added enabling factors
to Model 1, where Model 2 included only individual-level enabling factors (i.e., employment,
educational attainment) and Model 3 added system-level enabling factors (i.e., how caregiver
learned about services, service site). Model 4 included predisposing, enabling, and need-based
factors. While building regression models, we balanced statistical demands keep models
parsimonious given a relatively small analytic sample size while keeping models conceptually-
driven. Variables with a bivariate outcome where the p-value was greater than 0.25 were
removed from the model one at a time. We compared the coefficients for other variables between
nested models (i.e., models with and without the non-associated variable). If the coefficients of
remaining variables differed more than 20% between models, the variable was retained for the
final model (Hosmer, Lemeshow, & Sturdivant, 2013). After following this process, we did not
remove any covariates. We also examined continuous variables including age and ADL
functional ability for linearity on a logit distribution. Due to a notable violation of linearity
assumptions, we recoded age into a categorical variable that included less than 50, 50 to 65, and
over 65). We also assessed predictor variables for missing data. The highest proportion of
missing data were found for: how clients learned about services (25.6%), educational attainment
(22.2%), and burden (14.6%). Bivariate associations indicated that data was not missing
completely at random. To prevent biased estimates in the regression, missing data were handled
using multiple imputation by chained equations. The imputation model included auxiliary
variables associated with missing information, including caregiver’s self-assessed health,
32
poverty, social isolation, and month when the assessment was administered. Based on the highest
fraction of missing information found in our models, we used 40 imputed datasets. Using a
conservative guide for analytic sample size, we confirmed our largest model was adequately
powered for the number of predictors included (VanVoorhis & Morgan, 2007). To assess model
fit, we applied a Hosmer-Lemeshow statistic to the last 10 imputed data sets for each model. The
statistic was non-significant for each model, suggesting adequate fit across models.
This study was approved as exempt by the University of Southern California’s University
Park Campus Institutional Review Board in November 2016 (UPC-16-00544). All analyses were
completed in Stata 15.1.
Results
Sample characteristics
The analytic sample encompassed 540 caregivers who completed an assessment from
January 2017 (June 2017 at Site 2) through June 2018. On average, caregivers were 61.3 years
old (SD=13.2) ranging in age from 23 to 94 years. Care recipients were an average of 80.7 years-
old (SD 9.3). Just over half of clients assisted a parent (54.4%), while 36.6% of clients assisted a
spouse or partner. Clients approached services from all walks of caregiving; 41.5% had been a
caregiver for less than 2 years, 30.6% had been caring for 2 to 5 years, and 27.9% had been
providing care for more than 5 years. Seventy-five percent of caregivers assisted someone with a
cognitive impairment. On average, recipients had 3.7 ADL impairments (SD=2.2). Just 23.0% of
clients indicated an interest in learning about technologies to help assist with caregiving.
Bivariate associations
During the study period, 74 (13.7%) of caregivers accessed services using the online
FCA CJ system. Several bivariate relationships emerged when comparing caregivers who
33
accessed online-delivered versus usual services. We describe these differences according to the
proposed application of the Andersen model.
Predisposing characteristics. We did not find any statistically significant relationships
between predisposing demographic variables and use of online-delivered services, but did
observe two results near the significance threshold (i.e., p<0.05). On average, online service
users were younger than those using regular services ( χ
2
=5.54; p=0.07). FCA CJ users were more
likely to be under age 50 (27.9% vs. 17.0% among usual services) and less likely to be older than
65 (31.1% vs. 43.6% among usual services). The proportion of women caregivers was
approximately 10% higher among those using FCA CJ ( χ
2
=3.43; p=0.06). Bivariate associations
were not found for race/ethnicity or marital status.
Enabling characteristics. We discovered bivariate associations among each enabling
characteristic we examined. Regarding individual-level factors, over twice as many part-time
employees used the online service delivery option as usual-delivered services (26.4% vs. 12.5%;
χ
2
=10.17; p=0.02). Caregivers using FCA CJ had higher levels of educational attainment.
Whereas 10.3% of those with a post-graduate degree used usual service delivery, 32.3% of CJ
users had a post-graduate degree ( χ
2
=27.17; p<0.001). System-level enabling factors also showed
significant associations. One-fifth of those referred by social service providers used online-
delivered services (20.0%), while about 31.7% of those referred by healthcare providers did so
( χ
2
=42.80; p<0.001). Those who received service at Site 2 were far more likely to use online-
delivered services than Site 1 clients (28.2% compared to 8.2%; χ
2
=36.51, p<0.001).
Need characteristics. No need-based characteristics were associated with online versus
usual service use in bivariate analyses. We did, however, observe results near the significance
threshold for caregivers who completed of medical tasks. Nearly 10% fewer caregivers using
online services reported completing medical tasks as those using the usual-service delivery mode
34
( χ
2
=2.99; p=0.08). Similarly, whereas nearly one-third (31.3%) of FCA CJ users provided care
for less than 20 hours per week, this proportion was just one-fifth (20.1%) among those using
usual service delivery ( χ
2
=4.21; p=0.12). See Table 2.1 for additional descriptive and bivariate
information.
Logistic regression results
The first model we ran (Model 1) included only predisposing variables. The only
statistically significant predictor was the age of the caregiver: those aged 65 and older had 0.42
times the odds of using online-delivered services compared to those under age 50 (CI: 0.19-
0.91). When individual-level enabling factors were added, age was no longer a significant
predictor of using online-delivered services. In this model, caregivers employed part-time had
2.69 times the odds of using online-delivered services compared to caregivers employed full-
time (CI: 1.21-5.99). Compared to caregivers who completed some high school, those who had a
graduate degree had 11.57 times the odds of using FCA CJ, though the confidence interval was
notably wide (CI: 2.67-50.07). System-level enabling factors were also strong predictors of
online service delivery in Model 3. Caregivers who learned about services through health care
providers had 2.91 times the odds of accessing services online than those who learned about
services through social service providers (CI: 1.30-6.55). Those who learned about services
online had 5.17 times the odds of using FCA CJ versus those who learned about services from
social service providers (CI: 2.10-12.74). Those who received services at Site 1 had 0.28 times
the odds of using FCA CJ (CI: 0.15-0.53). None of the need-based factors added in Model 4
were significant predictors of using online-delivered services. See Table 2.2.
Discussion and Implications
This study compared the characteristics of caregivers using an online mode of service
delivery with those receiving usual service delivery modes at two social service sites. The
35
services received were the same using each mode of service delivery, and both modes were
available to eligible caregivers without charge. We found enabling factors were most strongly
related to using online-delivered over usual services, including educational attainment,
employment status, how clients learned about services, and the service site itself. Rates of
caregivers using the online service delivery option were surprisingly low.
We examined two types of enabling factors: individual- and system-level. Both types of
enabling factors predicted whether caregivers used the online service delivery option, but not
necessarily in the way we anticipated. We expected primary and full-time employed caregivers
to prefer online over usual service delivery because of the added convenience of being able to
access services at any time. We found that part-time employees were more likely than full-time
employees to use FCA CJ. Part-time employees may have more time available to complete
internet searches and find out about online resources compared to those employed full-time. This
is consistent with previous research wherein non-primary caregivers were more likely to use
online resources than primary caregivers (Kim, 2015; Li, 2015). It takes time to filter through
online search results before identifying a trustworthy resource (Funk et al., 2017), and primary
caregivers may not have sufficient time to do this filtering. On the other hand, the number of
hours of care provided per week did not predict online service use in regression models. We
believe this is because reported time estimates reflect many different caregiving experiences that
differentially affect access (e.g., companionship versus constant monitoring), and because of
relatively small cell sizes among FCA CJ users which may have prevented detection of small
effects. Findings that higher education predicted use of online-delivered services is consistent
with previous surveys of caregiver internet use (Kim, 2015; Li, 2015). Still, wide confidence
intervals for clients with a post-graduate degree indicate uncertainty in the extent to which more
educated caregivers select online service options, again a possible consequence of low cell
36
frequencies. It is also worth noting that Site 2 was located at a university setting; perceptions of
services housed in this setting may have contributed to different rates of service use by education
level.
Predisposing and need-based characteristics appeared to have little impact on online
service use. Our regression findings differ from previous survey studies where younger
caregivers were more likely to access online services than older caregivers (AARP, 2016; Li,
2015; Kim, 2015). We suspect this difference is in-part a result of the older age of our study
sample compared to survey research samples. Consistent with previous evaluations of caregivers
using social services, the caregivers in our sample were older than caregivers from national
studies (Herrera, Angel, Markides & Torres-Gil, 2013). In addition, we did not find variation in
online-delivered service use varied by racial/ethnic minority status (Fox et al., 2013). Lack of
difference by race/ethnicity has been previously observed (Kim, 2015; Li, 2015).
It is more challenging to draw comparisons between our results and previous studies
when considering need-based factors, since these are less consistently reported. One item on
which our findings deviated from previous research was emotional strain (Li 2015; Kim, 2015).
We did not find the emotionally distressed (i.e., burdened) caregivers were any more likely to
access services online. However, caregivers have demonstrated less enthusiasm for using online
resources to access emotional support. Whereas 70% of caregivers in the AARP survey of
caregivers to people aged 50 and older indicated they would likely use online sources to receive
personalized information, just 52% expressed interest in learning how to access emotional
support from an online source (AARP, 2016). Li (2015) reported similar findings. Null findings
regarding recipient cognitive impairment and functional ability were consistent with survey
results (Kim, 2015; Li, 2015). Bivariate results suggested an association between selection of
usual service delivery and the completion of medical tasks. We are not aware of any other
37
studies in which the completion of medical tasks was examined in relation to online information-
seeking, however Fox and colleagues (2013) reported relatively low use of technology supports
among caregivers who managed medications for recipients (18%).
Overall, we were also surprised by low uptake of online-delivered services in a social
service setting. Just 23.0% of clients expressed an interest in using technology to access services,
and only 13.7% actually used the online-delivered service. This is far below rates indicated in
national survey data (AARP, 2016; Kim et al., 2015; Li, 2015). Even among those who learned
about services online, 62.2% still opted to use services delivered by phone or in-person. While
low rates of online service utilization might be attributed to implementation factors, this is
unlikely. We removed the first months’ data from both sites, and did not observe a discernable
pattern in online service use over time. Adding a control variable for month of data collection did
not appreciably alter findings. The focus on health information in some previous studies (Fox et
al., 2013; Kim, 2015) likely underlies some of the discrepancy; still, even in these surveys, health
information was broadly defined and largely consistent with the kinds of information provided
by the service organizations we examined. We believe relatively low uptake of online service
delivery is primarily attributable to the differences between caregivers using social services and
those from nationally-representative survey samples. Similar to previous research on service-
using caregivers (Herrara et al., 2013), caregivers in this sample provided care to recipients with
a greater number of functional impairments and were older than caregivers nationally. Although
we observed no difference in ADL functioning among caregivers using online service delivery,
difference in levels of functional disability in a high-need population may not have been
substantial enough to affect results.
We do not take these results to mean that caregivers using social services generally reject
online service delivery. Two of the most significant predictors of online service use were how
38
caregivers learned about services and the service site. The prominence of these enabling
characteristics under Andersen’s (1995) model suggests that uptake of online service delivery is
modifiable. Indeed, nearly one-third of caregivers at Site 2 used the online service delivery
mode. While we do not know what about each service site generated such different rates FCA CJ
utilization, variation in how service specialists promoted the online service delivery option could
be one cause of this difference. Variation in the types of partner organizations and referring
agencies at both sites also likely played a role. Caregivers who learned about services through
social service providers were less likely to use the online delivery option compared to those who
learned about services through a healthcare provider or online. At Site 1, nearly twice as many
clients learned about services through a social service provider compared to Site 2 (54.9% vs.
26.4%). It is plausible that rates of online service delivery will increase as community
organizations learn about new service options to share with the caregivers they refer, and through
implementation of novel marketing approaches for online service delivery options (e.g.,
advertising in healthcare settings).
Limitations
This study provides one of the first looks at the characteristics of caregivers who use
online-delivered social services to access informational resources when both online and usual
services are available within the same service system. By drawing on a sample of clients from
two social service settings, we are able to provide findings that are applicable to actual service
users as opposed to caregivers generally. There are, however, several limitations with this study.
First, clients received services from service agencies based in large urban regions, and thus
findings may not be generalizable to rural populations. Secondly, low rates of online service
delivery uptake resulted in low cell sizes, and may have hidden weaker associations in regression
models. To account for this, we ran analyses several times after collecting additional months of
39
data, but cell sizes for FCA CJ users remained relatively small. Third, despite advertising at both
sites, we do not know the extent to which caregivers made an explicit choice between FCA CJ or
usual services; thus, we cannot know whether usual services were pursued by caregivers because
of an actual preference against online services or simply due to a lack of awareness of the FCA
CJ option. At the same time, the procedures in place at the two service delivery systems mirror
how clients would typically approach services; our findings likely reflect rates of online service
use that providers would find if they implemented similar online delivery mechanisms. Fourth,
we do not know exactly which factors led to variation in rates of caregivers using the online
service delivery option at each service site. Qualitative research to discern these factors is
underway, although findings regarding how caregivers learned about services provides some
insight into potential causes. Finally, it is not yet possible to tell whether those who logged into
FCA CJ returned to the site and reviewed resources after taking the online assessment. Still, even
if caregivers do not log back in, FCA CJ may be a beneficial tool in that caregivers know where
they can go for information should they need it (van der Roest et al., 2010).
Implications
Answers to the question of which caregivers use online-delivered services have
considerable implications for the way caregiver services are delivered. Results showed that there
is an interest in using online-delivered services to access information by at least some family
caregivers. Lower rates of online service delivery among less-educated caregivers support the
need to maintain social services delivered by usual means for some caregivers, as does the
selection of usual-delivered services among a considerable proportion of clients who found out
about services online. Variation in system-level enabling factors suggest that utilization of online
services is likely modifiable. Service organizations likely have an opportunity to increase use of
online service delivery options among clients, through means such as marketing to caregivers at
40
healthcare organizations and making online options easier to find for time-stretched caregivers
(e.g., increasing precision of tags used to locate programs in internet searches). Novel marketing
of online service delivery options, such as by partnering with employers, could also alter rates of
online use by attracting caregivers who traditionally have been less likely to approach social
services (e.g., younger caregivers).
Conclusion
Using actual client data, we compared the characteristics of caregivers using online and
usual service delivery modes when the service model was the same for both delivery approaches.
Overall, we observed few predisposing and need-based differences between caregivers using the
online FCA CJ system versus usual service delivery at two service sites. Those differences we
did observe suggest the relevance of enabling factors—education, employment status, how
caregivers learn about services, and aspects related to the service site—in whether caregivers use
online-delivered services. For service organizations and caregivers, online service delivery
options can be an appealing supplement to usual services. Future research should focus on
learning why caregivers opt to use online versus usual service delivery options when both are
available, the difference in service delivery mode on frequency of accessing resources, and how
to modify rates of online service delivery uptake through different marketing approaches.
41
Chapter 3: What can elder mistreatment researchers learn about primary prevention from
family violence intervention models?
Having looked at the cutting edge of service delivery in the previous chapter, I now turn
to intervention approaches responding to a long under-addressed aspect of family caregiving:
elder mistreatment. Although descriptions of family caregivers and potential perpetrators of
mistreatment has not always been welcomed (e.g., Brandl & Raymond, 2012), recognizing and
understanding risk of violence within caregiving dyads paves the way for structuring
interventions to prevent its occurrence.
2
A focus on elder mistreatment committed by family
caregivers is critical given the disproportionately high rates of elder mistreatment experienced by
older adults receiving care from a family member in comparison to samples of older adults
generally. Building on shared risk and theoretical mechanisms, I describe evidence-based
intervention programs that have demonstrated efficacy in preventing child maltreatment and
intimate partner violence. Drawing on promising approaches found in these fields, I describe the
structure and components of a home visiting program to intervene to prevent elder mistreatment
in caregiving dyads, as well as challenges related to evaluation and implementation of such a
program.
Compared to survey population samples generally (see Yon et al., 2017), family
caregivers perpetuate elder mistreatment at relatively high rates. A nationally representative
study found that 43% of those who verbally mistreated an older adult were immediate family

2
“Violence” as it is used here refers to actions which, following the definition by Hamby (2017),
are intentional, unwanted, non-essential, and harmful. While the term might evoke images of
“granny battering,” its application to more subtle acts of such as verbal abuse and neglect
acknowledges the equally harmful consequences of these behaviors (e.g., Dong& Simon, 2013).
Further, it distinguishes the intended target of an elder mistreatment intervention from normal
forms of aggression that can occur even in healthy relationships (e.g., shouting, teasing).

42
members (Laumann et al., 2008); another report found that 55% of financial abuse cases are
committed by the family (Metlife Mature Market Institute, 2009). Covering all five mistreatment
subtypes, the National Elder Mistreatment Study found the rate of family members as the
primary perpetrators in: 43% of financial abuse cases, 52% of sexual abuse cases, 57% of
emotional abuse cases, 74% of neglect cases, and 76% of physical abuse cases (Acierno,
Hernandez-Tejada, Muzzy, & Steve, 2009). Studies relying on self-report of care recipients only
also suggest very high rates of mistreatment in a family context; 26% of care recipients report
experiencing either physical mistreatment or exposure to potentially harmful behaviors (e.g.,
withholding food) at the hands of a family caregiver (Beach, Schulz, Williamson, Miller,
Weiner, & Lance, 2005; VandeWeerd, Paveza, Walsh, & Corvin, 2013). Increased levels of
dependency among care recipients may drive high rates of mistreatment in this population
(Dong, Simon, & Evans, 2012). In one study, 47% of those caring for family members with
Alzheimer’s Disease and related dementias acknowledged committing at least one type of
mistreatment (Wiglesworth et al., 2010). Interestingly, the willingness of caregivers to admit to
mistreatment in this study suggests family members are able to identify their problematic
behaviors towards the recipient, which may indicate receptiveness to receiving help that may
prevent these behaviors in the future. Further, caregivers generally express an interest in
providing quality care to the person they assist (Cheng et al., 2015). In this way, intervening to
prevent elder mistreatment may serve benefit not only at-risk older adults, but also those charged
with their care.
Given the 17.7 million family members adopting a caregiving role—a figure that is
expected to grow (Schulz et al., 2016)—and the potential for mistreatment to arise in the context
of caregiving, preventive tools targeting family caregivers are critical. However, intervention
research on preventing elder mistreatment is sparse, with few studies showing positive outcomes
43
(Pillemer, Burnes, Riffin, & Lachs, 2016). Preventive interventions that have been developed are
secondary and tertiary in nature, aiming to stop ongoing abuse and/or mitigate its detrimental
effects. There are no evidence-based primary prevention interventions specifically developed to
target elder mistreatment, let alone programs specific aimed at preventing elder mistreatment
among family caregivers.
Foundations in Family Violence: Share Risks and Underlying Theoretical Mechanims
Prevention efforts developed to address other types of family violence, such as child
maltreatment and intimate partner violence (IPV), offer possible applications for elder
mistreatment to guide the development of an intervention (Teresi, Burnes, Skowron, Dutton,
Mosqueda, Lachs & Pillemer, 2016). Hamby & Grych (2013) describe “webs of violence”
flowing through different forms of family conflict, wherein exposure at any point in the
lifecourse increases risk of eventually experiencing and/or perpetuating violence. They urge
researchers to follow these complex webs, rather than limiting their research to fields focused on
a specific part of the lifecourse. And there are many threads to follow. Shared risks and
theoretical underpinnings running throughout different types of family violence form a basis for
comparisons between child maltreatment, IPV, and elder mistreatment, as does the context of
trusting relationships such as that between a caregiver and recipient. Examining the causal
mechanisms that drive shared risk factors may help elder mistreatment researchers identify
adaptable prevention strategies in order to develop interventions (Burnes, 2016). The theories
used to organize these causal mechanisms are summarized in Table 3.1, and the causal
mechanisms and their specific risk factors are described below in detail.
Social Learning Theory
Previous exposure to family violence is associated with both future victimization and
perpetration of abuse. For example, adults who were abused as children are at a greater risk of
44
experiencing revictimization (Widom, Czaja, & Dutton, 2008) and are more likely to abuse their
own children (Merrill, Thomsen, Crouch, & May, 2005). Further, the Adverse Childhood
Experiences study strongly suggests that exposure to negative experiences early in life—
including abuse, domestic violence, and other forms of household dysfunction—increases the
likelihood of participating in IPV; those with high ACE scores have five times greater odds of
perpetrating IPV than those with low scores (Anda et al., 2006).
As the web of violence concept suggests, patterns of violence developed over the
lifecourse can yield devastating results in old age (Hamby & Grych, 2013). Shedding light on
such patterns though qualitative interviews, Pickering and colleagues observed that adult
daughters who were abused as children often engaged in what the authors called reciprocal
aggression towards their elderly mothers years later (Pickering, Moon, Pieters, Menten &
Phillips, 2015). The abuse was largely verbal—name-calling, yelling, and using isolating
language—and in some cases it escalated into hitting and possible neglect. Experiences of
childhood trauma also have been associated with increased risk of perpetration and victimization
specifically among caregivers and care recipients (Fulmer, Paveza, VandeWeerd, Fairchild,
Guadagno, Bolton-Blatt, & Norman, 2005). Social learning theory explains this patterned
violence, positing that abusive behaviors are learned through observation and are replicated after
having been established as normative behaviors (Bandura, 1977; McDonald & Thomas, 2013).
Caregiver stress
High stress levels among perpetrators have been identified as a risk factor for family
violence (Tucker & Rodriguez, 2014), including elder mistreatment. Although caregiver stress
theory suggests that overwhelmed caregivers are more likely to mistreat care recipients, the
evidence is mixed. Pillemer’s & Finklehor’s (1989) seminal study showed higher care needs
among recipients were not associated with elder mistreatment. Rather, those who mistreated
45
older adults were more likely to have experienced other life stresses (e.g., death of a relative)
within the year that mistreatment occurred. A more recent study found that among caregivers for
older adults, more hours spent providing care and higher levels of subjective burden were
predictive of mistreatment (Cooper, Selwood, Blanchard, Walker, Blizard, & Livingston, 2010).
Similarly, parents who commit child maltreatment find caregiving more stressful than other
parental caregivers (Teresi et al., 2016), suggesting that subjective assessments of caregiving are
central to understanding the role of stress in risk of mistreatment.
Reactive abuse
Reactive abuse conceives that certain victim behaviors are associated with risk, rather
than care needs of the care recipient. In child maltreatment (Thornberry, Matsuda, Greenman,
Augustyn, Henry, Smith, & Ireland, 2014), IPV (Kuijpers, van der Knaap, & Winkel, 2012), and
elder mistreatment (Cooper et al., 2010), aggressive behaviors from the victim are associated
with abuse, potentially eliciting negative reactions from perpetrators. Older adults who exhibit
aggressive behaviors related to behavioral symptoms of dementia are more likely to experience
mistreatment than those without dementia (Wiglesworth et al., 2010). Indeed, odds of physical
violence towards care recipients are four times higher when the older adult is physically violent
towards the caregiver (VandeWeerd et al., 2013).
Attachment Theory
Poor relationship quality between potential victims and perpetrators also is related to risk
of child maltreatment, IPV, and elder mistreatment. In cases of child maltreatment, this includes
low levels of maternal warmth (Brown, Cohen, Johnson, & Salzinger, 1998), minimal empathy,
negative attributions of children by parents (Tucker & Rodriguez, 2014), and low levels of
attachment (Thornberry et al., 2014). Attachment theory suggests that positive caregiver
interactions with a child promote emotional and behavioral regulation while increasing the
46
parent’s investment in the child’s wellbeing (Daro & McCurdy, 2008; Dishion, Mun, Drake,
Tein, Shaw, & Wilson, 2015; Toth, Gravener-Davis, Guild & Cicchetti, 2013). Poor attachment
in grown children has been implicated as a factor contributing to neglect of their older parents
(Cicirelli, 1986; Perkins, Spira, & Key, 2018).
Cumulative Risk
Risk factors for mistreatment can be additive and even multiplicative. Cumulative risk
theory suggests that compound risks severely heighten the likelihood of perpetuating violence. In
the child maltreatment literature, cumulative risk has been measured at both the individual and
family level. Families with high cumulative risk scores show greater likelihood of maltreatment
(Thornberry et al., 2014), while individual risk factors such as antisocial behaviors in
adolescence have been associated with greater likelihood of later committing child maltreatment
(Smith, Knoble, Zerr, Dishion, & Stormshak, 2014). Those with a higher ACE score during
childhood also had greater perceived stress and difficulty controlling anger in adulthood (Anda et
al., 2006). If these adults become caregivers, they may be more susceptible to perpetuating abuse
or reacting in a violent manner to challenging behaviors from the care recipient. However,
evidence on predictors of elder mistreatment behaviors is almost non-existent given the lack of
longitudinal data on this phenomenon.
Factors unique to elder mistreatment
While applying risk factors and theories from different fields of family violence to elder
mistreatment, it is important to recognize dynamics that make elder mistreatment different.
Aging is associated with illnesses and disabilities that may change an adult’s status from
independent to dependent, requiring different behaviors from family members. Because of the
older age of potential victims, an elder mistreatment prevention program must recognize: 1) the
possibility that the care recipient has a cognitive impairment or will experience cognitive decline
47
and 2) the likelihood that the caregiver is engaged in complex caregiving tasks with little
preparation or choice (NAC & AARP, 2015; Reinhard et al., 2012). For example, family norms
of gift-giving may turn exploitative if the older adult loses financial decision-making capacity
(Wilber & Reynolds, 1996). Family members may need to refrain from accepting gifts if the
older adult loses financial decision-making capacity. Such awareness of others’ cognitive
capacity and behavior monitoring is typically not called for in intimate partner relationships. And
while parental caregivers must take care to respond to children in developmentally appropriate
ways, an important distinction when assisting older adults is the assumption of autonomy. Unless
it is explicitly determined that the older adult lacks capacity, they retain their decision-making
autonomy, which includes the autonomy to make seemingly poor decisions, place themselves in
harmful or detrimental situations, and disagree with caregivers. This level of autonomy is not
assumed among children. Secondly, a need for demanding medical and personal care tasks to be
performed by caregivers also opens the door to mistreatment in a manner that is not often seen in
child maltreatment and IPV. Family caregivers may be more likely to neglect older adults when
high care demands and lack of support exceed their capacity. Some neglect is unintentional,
resulting from this imbalance of care recipient need and caregiver capacity, yet still produces
egregious outcomes (DeLiema, Homeier, Anglin, Li, & Wilber, 2016). Other times neglect is
willful and combined with other types of abuse. These added complexities necessitate a
gerontological-informed interventionists and curricula.
Community-based Intervention Models in Family Violence
Like theories and risk factors, community-based prevention models for other types of
family violence offer useful advances for addressing elder mistreatment. Because of the broad
scope of community-based intervention models existing in other areas of family violence, this
section largely focuses on home visiting models seen primarily in the child maltreatment field.
48
There are several reasons to focus on home visiting models. As with parents raising children,
family caregiving for older adults takes place primarily within the home. A home setting is
valuable for an intervention because it gives providers insight into contextual risks (e.g.,
substance abuse, home safety/fall risks, availability of nutritious food and medications,
accessibility of assistive devices, whether others are living in the home) that may not be apparent
in other settings (Mosqueda, Burnight, Gironda, Moore, Robinson & Olsen, 2016). Further,
home visiting interventions support the application of individualized intervention features to
meet the diverse needs of families, as opposed to more general prevention tools provided in
group-based interventions (e.g., group counseling) taking place outside of the home. Home
visiting programs occasionally have been adopted to address intimate partner violence (IPV), but
these programs tend to focus on secondary and tertiary prevention (Whitaker, Murphy, Eckhardt,
Hodges, & Cowart, 2013). As such, IPV home visiting models are less instructive than those
developed for child maltreatment. Home visiting models are particularly attractive in an elder
mistreatment prevention context, as a strong tradition of home-based intervention already exists
for family caregivers (Gitlin, Hodgson, & Choi, 2016). In reviewing approaches to community-
based intervention, components that contribute most to success are identified in this section.
Table 3.2 summarizes the intervention models described.
Home visiting programs to prevent child maltreatment
Home visiting programs in child maltreatment date back to the 1970s (Toth et al., 2013).
Some programs provide generalized supportive services, such as the Healthy Start program,
which connects new parents to community resources (Bair-Merritt, Jennings, Chen, Burrell,
McFarlane, Fuddy, & Duggan, 2010). Other interventions target specific mechanisms thought to
underlie child maltreatment. This strategy is used in the Promoting First Relationships
intervention, which aims to improve parents’ sensitivity to child needs and promote attachment
49
(Spieker, Oxford, Kelly, Nelson, & Fleming, 2012). Still other programs can be described as
multicomponent, offering a range of services and support to families, such as counselling,
training, education, and referrals to additional services (MacLeod & Nelson, 2000).
Despite variation, some designs and strategies are consistently more successful than others. For
example, longitudinal programs where home visits are conducted for two years or more and
provide 12+ visits are among the most successful at preventing child maltreatment (Daro &
McCurdy, 2008). Resonant with these findings, programs occurring in less than six months and
delivered in 12 or fewer visits appear to have lower impact (MacLeod & Nelson, 2000).
Home visiting interventions that include a training component also outperform others.
Parent-Child Interaction Therapy, for example, provides 12 to 14 sessions wherein parents are
taught parenting skills—typically while both the parent and child are present—including
alternatives to physical discipline (Toth et al., 2013). The program, which is based on social
learning and attachment theories, aims to break cycles of violence and improve relationship
quality. Other training approaches focus on improving parents’ perceptions of children. In an
adapted version of the Healthy Start program, at-risk families were trained to alter negative
appraisals of children; the control group had a 24% rate of physical abuse at follow up, while
those receiving the enhanced version of the Healthy Start intervention showed a rate of 4%
(Bugental, Ellerson, Lin, Rainey, Kokotovic, & O’Hara, 2010).
Although not specifically designed to prevent child maltreatment, the Family Check-Up
model offers several promising approaches. A home visiting program addressing maladaptive
behavior patterns in children, it is tailored to the unique needs of families over the course of 3
visits (Dishion, Forgatch, Chamberlain, & Pelham, 2016). Parents are interviewed about their
situation during the first visit, assessed and observed interacting with the child during the second
visit, and receive feedback during a final visit to discuss parenting strengths and weaknesses as
50
well as available services and supports to meet the family’s needs (Dishion et al., 2015, 2016;
Smith et al., 2014). The Family Check-Up model is based on the transtheoretical model of
change, a dynamic and person-centered approach that encourages behavior change in stages
based on readiness (Procheska, Redding, & Evers, 2008). Moving forward through stages is
believed to improve self-efficacy, an important factor to keep participants engaged in behavior
change (Bandura, 1977). In the case of the Family Check-Up model, motivational interviewing
techniques prepare parents to change reactions that contribute to difficult behaviors such as
defiance and poor self-regulation (Dishion et al., 2016). Feedback helps parents recognize areas
of strength and areas where they can improve (Dishion et al., 2015, 2016; Smith et al., 2014).
While its impact on preventing child maltreatment specifically remains largely unstudied, recent
findings by Dishion and colleagues (2015) show lower levels of neglect among families
participating in the Family Check-Up program 2 years after the intervention.
Community-based programs to prevent intimate partner violence
While addressing unique dynamics, there is considerable overlap between child
maltreatment and IPV prevention models. As with child maltreatment programs, more intensive
IPV interventions are most effective for primary prevention (Whitaker et al., 2013). Moreover,
some child maltreatment prevention programs double as IPV prevention. Bair-Merritt and
colleagues (2010) found that aspects of the Healthy Start program, including providing
emotional support and promoting access to resources, reduced physical IPV among participating
parents compared to a control group. Although home visiting approaches are far more common
in child maltreatment interventions, IPV interventions offer valuable lessons especially given
overlapping dynamics with elder mistreatment that are not seen in child maltreatment (e.g.
spousal abuse that continues into old age).
51
Drawing on child maltreatment, the Family Check-Up model described above also has
been effectively revised to prevent IPV in the form of the Men’s Domestic Abuse Check-Up, a
secondary prevention approach. The program is like the Family Check-Up but provides the
intervention via telephone to men who self-identify as engaging in violent behaviors toward their
partners. As with the Family Check-Up, motivational interviewing and feedback are central to
the model (Mbilinyi, Neighbors, Walker, Roffman, Zegree, Edleson, & O’Rourke, 2010). In
addition, because perpetrators of IPV tend to overestimate the normality of violent behaviors
towards partners (Neighbors, Walker, Mbilinyi, O’Rourke, Edleson, Zegree, & Roffman, 2010),
the program provides participants with a brochure detailing the actual prevalence of interpersonal
violence within the general male population. Revealing the discrepancy between participants’
perceptions of IPV occurrence and what is normal promotes self-awareness and draws upon
social learning theory, encouraging men to accept a more accurate perspective of IPV.
In contrast to many child maltreatment interventions, some IPV home visiting programs also
involve law enforcement. For example, one program used a one-time home visit following a
reported dispute, where police provided victims with information about community resources
and the reporting process (Casey, Berkman, Stover, Gill, Durso, & Marans, 2007; Stover,
Berkman, Desai, & Marans, 2010). An initial evaluation showed significant reduction in
subsequent abuse (Casey et al., 2007). A later study showed those receiving the intervention
were more likely to call the police (Stover et al., 2010), but this was possibly because of victims’
increased trust in community resources rather than an increase in violence.
In addition to or as part of home visiting, interventions to improve communication skills
between partners have led to reductions in revictimization (Whitaker et al., 2013). In a seminal
study, Markman and colleagues tested the Prevention Relationship Enhancement Program, where
couples learned techniques such as active listening and expressive communication during
52
therapy sessions (Markman, Renick, Floyd, Stanley, & Clements, 1993). Researchers found that
during 4- and 5-year follow ups, participants in the intervention reported better communication
and lower levels of violence in their relationships than controls. Communication The success of
these programs may be attributed to their accounting for varying stages of risk, an important
factor to consider since potentially abusive behaviors often increase over time (Whitaker, Hall, &
Coker, 2009).
Lessons from community-based interventions that have not worked in other fields of family
violence
Given the long history of prevention programs in other areas of family violence, it is
instructive to consider intervention approaches that have not been effective. Low-intensity
interventions primarily using media, such as a parenting newsletter, had the smallest effect size
in a meta-analysis of child maltreatment interventions (MacLeod & Nelson, 2000). On the other
hand, there appears to be an upper limit on intensity. The meta-analysis found lower effect sizes
for interventions with more than 33 visits (MacLeod & Nelson, 2000), a result possibly due to
additional logistical burden placed on at-risk families (Daro & McCurdy, 2008). Finally, training
interventionists is critical for success. The disappointing results of a randomized control trial for
Hawaii’s Healthy Start program were attributed to poor preparation of staff who were not
equipped to recognize risk factors, produce action plans tailored to families, and connect families
to community resources (Duggan, McFarlane, Fuddy, Burrell, Higman, Windham & Sia, 2004).
Applying Caregiving Interventions to Elder Mistreatment
An intervention targeting family caregivers at risk of engaging in mistreatment—
particularly one that focuses on the role of precipitating risk factors—should also consider
components of caregiver interventions that have shown success. Caregiver interventions tend to
focus on outcomes including strain, burden, depression, and caregiver health (Schulz et al.,
53
2016), as opposed to abuse and mistreatment. Home visiting approaches have also been widely
used for caregiver interventions; a recent review found 49 different home intervention studies for
caregivers of people with dementia (Gitlin et al., 2016). Like interventions addressing family
violence, the most effective caregiver interventions are multicomponent, providing a range of
services including dementia education, counseling, skills training, referral, and case management
(Gitlin et al., 2016; Schulz et al., 2016). However, it is common for interventions to have a
particular focus, such as improving communication skills or coping. Livingston and colleagues
(2013) used the Coping with Caregiving program to provide an in-home intervention to
caregivers of people with dementia with a focus on education, and observed reductions in
anxiety and depression in the intervention group. Interestingly, these authors also measured
family conflict as a secondary outcome, but results were not significantly different between the
intervention and control groups.
Building on Family Violence Interventions to Identify Key Components for a Home
Visiting Model to Prevent Elder Mistreatment
Drawing on lessons learned from child maltreatment and IPV, we argue for the adoption
of a home visiting model delivered as a primary prevention intervention for elder mistreatment.
Based on successful child maltreatment and IPV models, we propose that the structure of a
preventive EM intervention should target caregivers facing new or changing conditions, take
place over several months, and be tailored to individual and family needs, preferences, and
culture. Components should include regular assessments, training, and strengths-based feedback.
Figure 3.1 illustrates how these components might be structured within an intervention. Although
not exhaustive, these approaches reflect the most promising practices from family violence.
Readers seeking additional information about specific program structure and components may
wish to review MacLeod and Nelson (2000) or Whitaker and colleagues (2013).
54
Although we propose a standardized structure and several key components for an
intervention, the exact intervention tools provided to each family will depend on specific risk
factors identified during assessment. Previously, we suggested that risks fall into three general
theoretical categories: briefly, those related to relationship quality, stress, and knowledge. The
intervention tools selected for each dyad should be tailored based on their risk profile that is
identified from the assessment and accompanying theoretical mechanism (e.g., assessment of
high burden and high perceived stress would indicate a need to address a caregiver stress
theoretical mechanism). For example, whereas training for someone who is experiencing high
levels of stress might focus on behavior management and assistance accessing respite, caregivers
who are struggling with a difficult relationship history will likely receive greater benefits from
counseling. Figure 3.2 illustrates an intervention based on a hypothetical case in which stress-
based risk factors are central; Figures 3.3 and 3.4 illustrate what an intervention might include
for the other theoretical mechanisms discussed.
The rest of this section expands on the lessons learned from successful models in child
maltreatment and IPV about how to structure an intervention.
Target caregivers at the start of caregiving and those facing changing conditions
As Teresi and colleagues (2016) point out, interventions in child maltreatment are most
effective when delivered before maltreatment occurs. Similarly, EM prevention programs
implemented near the beginning of a caregiving journey or at critical junctures after a change in
care level occurs may be most effective. For example, an intervention may target caregivers
following a hospital transition. However, unlike new parents who readily identify with their role,
caregivers who are just beginning this role may require additional guidance during recruitment
given that many family members do not perceive themselves as caregivers (Levitskey, 2014;
Schulz et al., 2016).
55
Take place over an extended period
Further, in reviews of both child maltreatment and IPV interventions, the most successful
home visiting programs were found to be relatively intensive and moderately long in duration.
Based on studies from other forms of family violence, regular and frequent visits—perhaps 12 or
more (MacLeod & Nelson, 2000)—over at least 6 months are most likely to improve the success
of a home visiting intervention. Even longer intervention periods may be appropriate, possibly
lasting 2 years or more, depending on other design components (e.g., qualifications of the
interventionist) (Daro & McCurdy, 2008).
Tailored to individual and family needs, preferences, and culture
Interventionists should utilize the flexibility of a multicomponent approach to deliver
services and supports that meet the unique needs of families wherever they are on the spectrum
between a healthy and abusive relationship. For example, interventionists in the Healthy Start
program individualize the intervention to families by guiding them to relevant services based on
specifically identified needs, and decrease the number of visits as parent competency grows
(Bair-Merritt et al., 2010). Multicomponent options may also be preferable compared to “one-
size-fits-all” approaches (e.g., Alkema, Reyes, & Wilber, 2006), as some families will be more
amenable to and will engage with some intervention tools more than others. In addition, an EM
intervention should meet the needs of culturally and ethnically diverse caregiving dyads and
families. The Family Check-Up model’s success in racially and ethnically diverse samples has
been attributed to its ability to address multiple types of stressors linked to culturally specific
norms and attitudes (Smith et al., 2014).
Guided by regular assessment
To effectively meet specific and evolving needs and identify whether progress has been
made toward reducing risk, family assessment at baseline and regularly thereafter is necessary.
56
Assessment, as described here, is both a means of evaluating a program, and a component of the
intervention itself. In the Family and Men’s Check-Up models, both inspired by the
transtheoretical model, assessment is the cornerstone of increased self-awareness underlying
behavior change (Prochaska et al., 2008). Although assessment should avoid being burdensome,
it is important to include contextual factors (e.g., family dynamics, characteristics of the
caregiving situation) that are likely to evolve, such as cognitive impairment of the care recipient.
An assessment should inform a care plan (i.e., intervention components and tools) that is specific
to the needs of the family.
Inclusion of a training component
A home visiting model to prevent elder mistreatment among caregiving dyads should also
include a caregiver training component for most cases. The success of training is observed in the
positive results of Parent-Child Interaction Therapy (Toth et al., 2013). Success is attributed not
only to skill-building in lower-performing areas but also to increases in self-efficacy that in turn
supports behavior change. Like training options offered in caregiver and child maltreatment
interventions, content for an elder mistreatment intervention might include practical
demonstrations on providing care, managing money, and managing difficult care recipient
behaviors. Training could also address relationship quality, especially in situations involving
recipients with dementia or those who may have difficulty appropriately expressing their needs.
In-line with Promoting First Relationships (Spieker et al., 2012), relationship quality can be
improved by assisting caregivers in understanding recipient needs. Although training a dyad on
communication skills (e.g., active listening) could be useful in some cases (Markman et al.,
1993), there are likely limitations to implementing communication-based training in families
with long-standing relationship conflicts or where the recipient is experiencing severe cognitive
impairment. Such training, if used, should include realistic goals as to what can be accomplished
57
(e.g. de-escalation, reduction of potentially harmful interactions). This approach is supported by
the transtheoretical model, in which progress is tailored to the individual (Prochaska et al., 2008).
Provision of strengths-based feedback
Strengths-based feedback builds self-efficacy, encourages participant retention, and
motivates progress toward acquiring the skills necessary to provide quality care. Such feedback
should be provided following assessment, reassessment, and while providing training modules.
For example, while a caregiver might respond negatively to recipient behaviors such as yelling
(e.g., yelling back, insulting the recipient), they may show strengths in other areas (e.g., finding
suitable respite care when they need a break, being attentive to recipient needs). According to
transtheoretical model of behavior change, promoting self-efficacy through recognition of
strengths can encourage engagement in interventions (Prochaska et al., 2008), which has been
illustrated by Family Check-Up programs (Dishion et al., 2015). Long considered an outcome of
caregiver interventions (Schulz et al., 2016), self-efficacy should also be considered as a
moderator by elder mistreatment interventions.
Creating an Evidence Base
One of the most complex aspects of the home visiting model is determining what
standard ensures that the model is evidence-based. As with any primary prevention intervention,
it is difficult to prove that the intervention prevented an incident that otherwise would have likely
occurred. Following in the footsteps of child maltreatment and IPV studies, a control group is
required. More specifically, a randomized control trial would provide the most reliable evidence.
More complex is the matter of measuring the success of a program given variation in
caregiving circumstances; inevitably, some dyads will have greater susceptibility to program
benefits than others. For example, Toth et al. (2013) describe how genetic, epigenetic, and
neurocognitive differences in children may impact the effect of maltreatment interventions.
58
Heterogeneity within characteristics, predisposing some dyads to benefit from the intervention
versus others, are likely to increase with age, which lends itself to different functional and
cognitive statuses among care recipients and caregivers. A randomized control trial design,
however, should serve to balance this heterogeneity across the intervention and control groups.
Once efficacy is established, large sample sizes should be collected in order to evaluate relevant
moderators of the intervention’s impact. By doing so, researchers can learn which
subpopulations are most likely to benefit and achieve preliminary guidance on how to adjust the
intervention to benefit groups of caregivers who receive little benefit. The greatest challenge to
illustrating the effectiveness of home visiting models is the possibility of home visits increasing
detection of maltreatment, thus making it appear that maltreatment occurs at a similar or greater
rate in an intervention group compared to a control group (MacLeod & Nelson, 2000). Such was
the case in the police visiting program studied by Stover and colleagues (2010); even when an
intervention is likely working, design challenges might suggest otherwise. Adding a qualitative
component when evaluating an intervention would help to explain any upticks in reported abuse
rates that may occur in the intervention group. For example, pre- and post-intervention
interviews with participants, delving into knowledge of community resources, attitudes and
understandings of risk factors, and the nature of any experiences with mistreatment, could
explain factors underlying reporting besides the dichotomous occurrence of mistreatment.
Resources to Support a Home Visiting Program
Another challenge to implementation OF what? is the cost of a home visiting program.
Given the time it would take to deliver the intervention over the course of at least 12 visits, the
program would be resource intensive. On the other hand, when Dalzeil & Segal (2012) evaluated
the costs of implementing 32 different home visiting programs ($1,800 to $30,000 per family)
and compared these with estimated lifetime costs of child maltreatment, they found a moderate
59
savings. Still, the lifetime cost for elder mistreatment would likely be lower than that of child
maltreatment, given shorter life expectancy among older adults. Nevertheless, the association
between elder mistreatment and both hospitalization (Dong & Simon, 2013) and nursing home
placement (Lachs, Williams, O’Brien, & Pillemer, 2002) may still render a primary prevention
program cost effective. Indeed, interventions that engage caregivers during transition from
hospital to home have been shown to effectively reduce hospitalization, lowering mean hospital
costs by approximately $500 (Coleman, Parry, Chalmers, & Min, 2006). Similar cost avoidances
from EM may be found from reduced reliance on emergency medical, legal, and social services
responses and long-term punitive actions if prevention of mistreatment is successful. Moreover,
preventing elder maltreatment extends beyond financial value; return on investment is also
achieved through the promotion of social justice and the preservation of dignity in old age.
There may be less expensive delivery models suitable for low-risk caregiving dyads, such as
interventions delivered via telephone and smart phone apps. Technology-based delivery methods
are particularly apropos given that 71% of caregivers show interest in using technology to
support caregiving activities (AARP, 2016). However, despite the promise shown in the
telephone-based Men’s Domestic Abuse Check-Up (Mbilinyi et al., 2010), other telephone-based
IPV programs have not replicated this success (Stevens et al., 2015). At the same time, an online
version of the PREP program (ePREP) has demonstrated efficacy in reducing physical and
psychological aggression among university students (Braithwaite & Fincham, 2009). Telephone-
and internet-based interventions might be suitable for low-risk dyads. Thus, low-cost delivery
options provided via technology have potential, but cannot yet be counted on as an effective
cost-saving option.

60
Conclusion
In the search for evidence-based primary prevention strategies to address elder
mistreatment, interventions from other fields of family violence provide evidence-based ideas,
inspiration and guidance. Of particular interest are home visiting models aimed at caregivers,
which offer opportunities to target underlying risk factors shared among different forms of
family violence. We describe starting points for such an intervention but acknowledge that any
program will be shaped by the available resources and specific agency goals. Regardless, given
what we know about prevention programs from child maltreatment and IPV, we believe there is
adequate information to pilot a promising prevention intervention for elder mistreatment.

This chapter was previously published in The Gerontologist: Meyer, K., Yonashiro Cho,
J., Gassoumis, Z.D., Mosqueda, L., Han, S.D., Wilber, K. (2017). What can elder mistreatment
researchers learn about primary prevention from family violence intervention models? The
Gerontologist.

61
Chapter 4: The effect of number of years caregiving on physician visits and hospitalizations
Having previously described intervention approaches to prevent elder mistreatment in
caregiving dyads, in this chapter I turn to the health and wellbeing of caregivers themselves.
Caregiving has been associated with decline in both physical and mental aspects of health
(Capistrant, 2016). Despite the risk of negative health consequences, caregiving is rarely
described as a public health problem. In the U.S., caregiving is typically considered a private
family matter (Levitsky, 2014). Consequently, the U.S. has few policies and programs available
to protect and improve caregiver health and wellbeing. Nor is there much guidance as to which
caregivers could most benefit from health-focused programming. Given this, the purpose of this
chapter is twofold: 1) to develop a policy case for increased access to health-focused caregiver
interventions and 2) to establish when during the caregiving trajectory health interventions
should be administered.
Health consequences of caregiving include both acute and chronic conditions, and span
mental and physical aspects of health. Caregivers are prone to musculoskeletal injury from
physical caregiving tasks (Darragh, et al., 2015). They are more likely than non-caregivers to
accumulate cardiovascular risk factors (Monin et al., 2010). Caregivers are particularly at risk of
mental health morbidities, including depression (Pinquart & Sorenson, 2003). Stress has long
been recognized as a primary culprit for health decline among caregivers (Son, Erno, Shea,
Femia, Zarit, & Parris Stephens, 2007; Shaw, Patterson, Semple, Ho, Irwin, Hauger, & Grant,
1997). According to Pearlin’s Stress Process Model, stressors related to caregiving accumulate
and interact to affect health outcomes over time (Pearlin, Mullan, Seple & Skaff, 1990).
Compared to adult children providing care, spousal caregivers report worse physical
health and higher rates of depression (Pinquart & Sörensen, 2011). Certain stressors experienced
primarily by spousal caregivers place them at increased risk of health decline. Spouses provide
62
care for a longer period of time than non-spousal caregivers (Pinquart & Sorenson, 2011); longer
time spent in a caregiving role is associated with poorer physical health (Pinquart & Sorenson,
2007). Some interpersonal stressors also weigh more heavily on spousal caregivers, such as
witnessing a partner suffer and the complex loss of a partner who has a cognitive impairment
(Blieszner, Roberto, Wilcox, Barham, & Winston, 2007; Monin & Schulz, 2009) Cohabitation
with the care recipient, more frequent among spousal caregivers (Pinquart & Sörensen, 2011),
has also been identified as a contributing factor to poorer physical health and depression
(Pinquart & Sörensen, 2007).

Policies and programs to attenuate risk to caregiver health are few. The National Family
Caregiver Support Program (NFCSP) is the primary federal program responsible for providing
caregiver services (Schulz et al., 2016). And although health benefits have been associated with
many NFCSP-funded services including counseling, support groups, respite care, and
psychoeducational programs, very few caregivers ever access these services

(Chien, Chu, Guo,
Liao, Chang, Chen, & Chou, 2011; Gallagher-Thompson & Coon, 2007; Kally et al., 2014;
Klein, Kim, Almeida, Femia, Rovine, & Zarit, 2016). Of an estimated 17.7 million caregivers to
older adults in the U.S., just 106,795 (0.60%) received counseling services and 57,30 (0.03%)
received respite in 2016 (Schulz et al., 2016; Administration for Community Living, n.d.).
3

Further, Medicaid and Medicare services distributed to care recipients typically do not
incorporate the needs of family caregivers, even when caregivers are essential to patient care

3
Data on clients served was found using the Administration for Community Living’s AGing
Integrated Database (https://agid.acl.gov/CustomTables/). Data was selected from 2016, and
included caregivers from all age groups and all states. Counseling was described as services “to
assist [caregivers] in making decisions and solving problems relating to their caregiver roles.
This includes counseling individuals, support groups, and caregiver training (of individual
caregivers and families).” Respite was defined as, “Services which offer temporary, substitute
supports or living arrangements for care recipients in order to provide a brief period of relief or
rest for caregivers.”
63
plans developed under these programs (Kelly, Wolfe, Gibson, & Feinberg, 2013). And while
many spousal caregivers have access to Medicare services, health service providers typically are
not trained to ask about caregiving duties or how to respond when a caregiver expresses concerns
(Adelman et al., 2014).
One barrier to advancing programs promoting caregiver health is the inconsistent
findings on the extent to which caregiving places health at risk (Capistrant, 2016; Schulz et al.,
2016). Caregivers may appear healthier than non-caregiving individuals because those who are
unhealthy select out of this role. Moreover, health declines take time to manifest and may occur
after individuals relinquish their caregiving role (Fredman et al., 2015; Schulz et al., 2016).
In addition to questions about how caregiving impacts health, there is little consensus to
date on whether caregiving for an older family member increases health service use. Several
studies have found caregivers have more emergency department visits than non-caregivers
(Kolanowski, Fick, Waller & Shea, 2004; Musich, Wang, Kraemer, Hawkins, & Wicker, 2017;
Shaw et al., 1997; Suehs et al, 2014). Other studies have not supported the finding that caregivers
use more health services than non-caregivers when considering hospitalizations (Buyck et al.,
2011), outpatient visits (Kolanowsk et al., 2004), office visits (Kolanowsk et al., 2004; Wu, Luo,
Flint & Qin, 2015), and screenings (Kinnear, Connolly, Rosato, Hall, Mairs, & O'Reilly, 2010).

Two studies focusing on caregivers to individuals with Alzheimer’s Disease, who
comprise one-third of all caregivers to individuals age 50 and older (NAC & AARP, 2015) have
limited and inconclusive findings. One study linked objective and subjective stressors from
caregiving to increased healthcare visits and expenditures (Son et al., 2007). Yet, because
caregiving lasts an average of 4 years and often longer (NAC & AARP, 2015), the study’s cross-
sectional design could not address how length of exposure effects rates of healthcare service
utilization.

Another study by Shaw and colleagues (1997) used a longitudinal design to examine
64
hospitalizations between spousal caregivers to matched controls, and did not observe differences
between the two groups.

However, for much of the caregiving research, lack of longitudinal data
on caregivers makes it difficult to understand the extent to which changes in health and health
service use are caused by caregiving.
In the current study, we address discrepancies found in previous studies by examining the
effect of time spent as a caregiver on caregiver health service use by using Health and
Retirement Study (HRS) data from 2002 to 2014. The HRS is sponsored by the National Institute
on Aging (grant number NIA U01AG009740) and is conducted by the University of Michigan.
We hypothesize that caregiving for a spouse, with its many associated stressors, will result in
greater use of healthcare services when compared to other adults in the HRS sample. We expect
rates of health service use will be positively associated with the length of time a spouse serves as
caregiver, as health risks accumulate over time.
Methods
Data
To determine whether the number of years spent as a caregiver impacts rates of
healthcare utilization, we use data from the HRS. Participants in the HRS sample include adults
ages 51 and older and their spouses. HRS uses a longitudinal panel design, where participants are
surveyed every 2 years. In this study, we use data collected from 2002 to 2014 and include
caregivers to individuals with a variety of conditions (i.e., not only dementia). To determine
caregiver status, HRS participants are asked if they have any functional disabilities and, if so,
who most often helps them with these. Caregivers were interviewed based on recipient
responses.

65
Variables
Dependent variables. We ran two sets of models to capture use of both preventive health
services and services to address acute health conditions. The dependent variable in the first set of
models was the number of physician visits experienced in the last two years. In the second set of
models, we considered whether or not the participant experienced a hospitalization in the
previous two years.
Independent variables. We created a variable for the number of years a study participant
spent as a caregiver by adding together the number of waves where the respondent indicated they
provided care to a spouse. If a participant indicated they were a caregiver when they entered the
study, they were removed from the sample to prevent a biased understanding of the impact of
length of caregiving on outcomes (n=1742), since there is no way of knowing how long these
individuals had already been a caregiver. Although there were spouses who had been caregivers
for up to 12 years in this sample, data on caregivers providing care for 10 or more years was
sparse (n=89 caregivers were in this role for 10 years in either 2012 or 2014, and n=19
caregivers were in this role for 12 years by the 2014 wave). Given this, if a caregiver had been
providing assistance for 10 or 12 years, the number of years spent as a caregiver was recoded to
indicate having spent 6 or more years in this role.
Covariates. Model covariates included demographic characteristics age, gender, race,
ethnicity, and educational attainment. We anticipated that having health insurance would
increase rates of health service use, and thus added this as a covariate (McWilliams, Zaslavsky,
Meara, & Ayanian, 2003). Notably, we did not include covariates such as health and health
behaviors in regression models, these are likely mediators of health services use and caregiving
experience (Schulz et al., 2016).
66
Analysis. To examine the relationship between the number of years spent as a caregiver
and rates of healthcare utilization, we elected to use the “hybrid” fixed effects (FE) modeling
approach described by Allison (2009). We decided the hybrid approach was most appropriate
after discovering over-dispersion when analyzing the number of physician visits participants
reported in the previous two years. Over-dispersion was detected using a Cameron-Trivedi test
(Cameron & Trivedi, 2010). Although poisson models are often used to model count data, an
underlying assumption of these models is that the variance is equal to the mean in the dependent
variable. Failure to account for over-dispersed count data could result in downwardly-biased
estimates. A frequently used solution when handling over-dispersed count data is to employ a
negative binomial distribution. However, the parameter added to the negative binomial equation
when using Stata’s xtnbreg command results in a model that is not a true FE model (Allison &
Waterman, 2002). The hybrid model makes it possible to use a negative binomial distribution to
achieve fixed effects estimates by regressing the mean and mean deviation of time-varying
predictors on outcome variables using a negative binomial link. By comparing the equivalence of
the mean and deviation variable estimates after running this model, we were also able to confirm
our choice of an FE model given likely violation of the assumption of non-correlation for
between-person heterogeneity and independent variables made with applying random effects
models (e.g., health, personality) (Allison, 2009). We ran the hybrid models using a generalized
estimating equation model (GEE) to estimate population average fixed effect coefficients using
an exchangeable correlation. Others have addressed the limitation of negative binomial models
by bootstrapping standard errors in FE models using a poisson distribution (Allison, 2009;
Cameron & Trivedi, 2010). We present results for both types of models for the sake of
comparison, and because post-estimation commands cannot be used directly after using the
hybrid model, where regression predictors are the mean and deviation of independent and
67
covariate variables. Even though over-dispersion was not a concern when examining whether or
not participants experienced a hospitalization, we also applied the hybrid model for logistic
models since a benefit of these models is that they can also be used to estimate time-invariant
variables (e.g., race). In both sets of models for physician visits and hospitalizations, we include
a quadratic term for number of years caregiving. However, we also present results from a model
where the number of years caregiving is treated as a factor variable (i.e., excluding the quadratic
term) to facilitate easier interpretation of odds ratios; this was appropriate given the very small
coefficient we found for the quadratic term when assessing hospitalizations. Maximum
likelihood estimation was used to address missing data in FE models.
Limitations. There are limitations to this study. First, we recognize that potential
confounders could explain both the likelihood of being a caregiver and using healthcare services,
such as personality characteristics (e.g., conscientiousness). In FE models, this is handled by
controlling for within-persons heterogeneity using a fixed constant (Johnson, 1995). However,
the estimates for time-invariant variables provided under the hybrid approach are still vulnerable
to omitted variable bias. Another consequence of using FE models is that participants with one
observation and/or little or no change in the outcome variable are dropped from models, reducing
statistical power in these models. Finally, when observing caregivers who had been in this role
for many years, sample sizes are considerably reduced; it is possible that data sparsity for
caregivers who had been in this role for a long period of time (e.g., 8 years) could bias results for
those with many years of caregiving. We addressed this by aggregating data for those with more
than 6 years of caregiving experience.

68
Results
Sample characteristics
The study sample included 28,017 unique participants over seven waves, with a total of
128,714 observations. After excluding participants who were caregivers in who were already
providing care when they were first surveyed, there were 530 new caregivers in 2004, about
2.78% of the HRS sample. By the 2014 wave, there were a total of 1073 caregivers (5.98% of the
sample). Across all waves, the average length of time caregivers spent in the role was 3.24 years
(SD 1.95). Across waves, participants reported 9.77 (SD 15.51) physician visits over the
previous two years and 51.64% of participants experienced a hospitalization.
Overall, there were few differences between the characteristics of caregivers and non-
caregivers. Caregivers were slightly older than non-caregivers (71.85[SD 11.07] vs. 67.66 [SD
11.35]). Caregivers were, on average, similar to non-caregivers in terms of race and ethnicity;
80.12% of caregivers were Caucasian/white compared to 77.00% of non-caregivers, and 13.07%
of caregivers indicated Hispanic ethnicity compared to 11.01% of non-caregivers. Caregivers
were slightly more educated than non-caregivers; compared to 29.92% of non-caregivers,
20.45% of caregivers had a high school education. Caregivers were also more likely to be
insured compared to non-caregivers (88.67% vs. 82.01%). Interestingly, just 43.69% of
caregivers were female compared to 59.76% of non-caregivers. Given conventional knowledge
that women provide the majority of care, we explored the unexpectedly high proportion of male
caregivers. We suspected the high proportion of male caregivers was caused by the higher age of
the sample compared to samples found in other national studies. This was confirmed when the
study sample was stratified by age, where the proportion of women caregivers was higher at
younger ages. Removal of those providing care prior to 2002 exacerbated this difference. Table
69
4.1. provides a comparison of caregivers and non-caregivers, pooled across waves. See Table 4.2
for characteristics of study participants at each wave.
Physician visits
After controlling for covariates, we found a significant increase in the number of
physician visits in the previous two years by the number of years spent in a caregiving role
(B=0.20; p<0.001). However, rates of physician visits declined as the number of years spent
caregiving increased, indicated by a significant and negative quadratic term (B=-0.03; p<0.001).
Predicted values derived from the bootstrap model indicate the greatest change in rates of
physician visits occurs in the first two years of caregiving. (See Figure 4.1.) Estimates for both
the hybrid model and bootstrap model were similar, although confidence intervals for the hybrid
model were larger than for the poisson model. Regression results are available in Table 4.3.
Hospitalizations
Similarly, we found higher-odds of experiencing a hospitalization in the past two years
with more years spent caregiving for a spouse (OR 1.38; p<0.001), an effect that lessened with
time, shown by the slight but significantly lower odds in the quadratic term (OR 0.96; p<0.001).
Here, too, estimates for the hybrid and poisson model were largely similar. Complete regression
results available in Table 4.4.
We also examined the odds of experiencing a hospitalization when years of caregiving is
treated as a factor variable. These estimates are more intuitive to interpret, but do not account for
the slight but significant quadratic relationship. We found that compared to those with 0 years of
caregiving experience, participants with 2 or fewer years of caregiving experience had 1.90 times
the odds of experiencing a hospitalization (CI 1.71 to 2.10; p<0.001), and those with 4 or fewer
years of caregiving had 1.53 times the odds of experiencing a hospitalization (CI 1.31 to 1.78;
p<0.001). Participants with 6 or more years of caregiving, however, had just 1.08 times the odds
70
of experiencing a hospitalization, a difference that was no longer significant (CI 0.81 to 1.45;
p=0.58). Logistic regression results are displayed in Table 4.5.
Discussion
Our results describe the dynamic trajectory of health service use among spousal
caregivers. We hypothesized that more years spent as a caregiver would result in higher rates of
physician visits and odds of hospitalizations. Our hypotheses were partially supported.
Consistent with previous studies, caregivers did use healthcare services more often than non-
caregivers at the onset of this role and rates initially increased the longer one remained a
caregiver (Kolanowski et al., 2004; Musich et al., 2017; Suehs et al, 2014). However, rates of
health service use did not increase linearly, such that the number of physician visits among
caregivers began to fall after about four years in this role. Increases in the rate of hospitalizations
slowed as well. This pattern could account for apparent lack of differences in rates of health
service use observed in some studies, particularly those with a cross-sectional design (Buyck et
al., 2011; Kinnear et al., 2010; Kolanowsk et al., 2004; Wu et al., 2015; Son et al., 2007).
According to the Stress Process Model, stressors accumulate and interact over time,
eventually resulting in negative health outcomes, and presumably greater health service use
(Pearlin et al., 1990). If interpreted through using this model, it is possible that decline in service
use after many years of caregiving is attributable to increased capacity to cope with caregiving
stressors that are known to increase use of healthcare services (Son et al., 2007).

Similarly, it
may be that caregivers who remain in this role for a long time might have more resilient health
profiles than those who relinquish caregiving. A third alternative for declines in utilization after
many years of caregiving is that spousal caregivers experience reduced access to healthcare as
the recipient’s condition progresses and demand for care increases. However, this explanation is
71
less likely when examining hospitalizations, where caregivers would likely have less discretion
over whether to access services than for physician visits.
We also tested models where number of years of caregiving was treated as a cumulative
variable, such that if one ended caregiving after 6 years, they were treated as having 6 years of
caregiving experience in later waves. While results from these models are more challenging to
interpret and apply to policymaking, this approach may more closely reflect caregivers’ lived
experiences and tenants of the Stress Process Model. A cumulative variable for the number of
years caregiving treats caregiving as a stressor that could affect rates health services utilization
even after ending this role. However, we did not observe notable differences in results from the
models already presented in this paper and models with the cumulative variable for years of
caregiving. (See Table 4.6.) Further research on a sample where more long-term caregivers have
shifted out of this role might better capture the cumulative effects of caregiving on health service
use.
Overall, our results show caregiving has significant effect on rates of health service use,
and varies by the number of years in a caregiving role. To understand the cost of higher rates of
health services use by spousal caregivers, we use the predicted values from regression models of
physician visits. Using the Medicare Healthcare Common Procedure Coding System code for an
outpatient physician visit (code 99213), we estimate that, among spousal caregivers, this role
contributes to an additional $98.8 million in health spending per year, ranging between $63.2
million and $134.4 million.
4
Given that 62.67% of the sample had some form of government

4
We reached this figure by differencing the predicted value of physician visits for non-
caregivers with predicated values by years spent caregiving. We then multiplied each of these
values by the proportion of caregivers in the 2014 HRS wave according to years spent in this role
(e.g., 0.42 for the 42% of caregivers who had been in this role for 2 years or less). We multiplied
each of these products by $48.40, the average payment amount for an outpatient physician visits
by Medicare in 2016 (Healthcare Common Procedure Coding System, code 99213) (Center for
72
insurance, this equates to roughly $61.9 million in government health spending. (We did not
estimate costs associated with hospitalizations, since these are far too variable to estimate and no
one code can be readily applied.) With the number of caregivers expected to increase in coming
years, we expect this figure to continue to grow. Notably, this estimate accounts for just a
fraction of health services, and does not consider physical therapy, counseling, and other health
services spousal caregivers might use, and only includes spousal caregivers.
As far as we know, this is the first study to examine the effects of spousal caregiving for
any type of condition on healthcare utilization over multiple years to understand how time spent
in a caregiving role impacts health service use. This study establishes that caregivers experience
higher rates of physicians visits and odds of hospitalizations, particularly early in the caregiving
role. Future research should consider mediators that explain why rates of health service use in
this population are higher (e.g., physical and mental health), factors that contribute to reduced
rates of healthcare service use the longer one is in this role (e.g., access to services), and spiking
rates of health service use in the initial years of caregiving. Our results suggest policymakers and
service providers should consider targeting programs and services that protect caregiver health
and wellbeing early in the care journey to protect caregivers’ health as they adjust to new
stressors.

Medicare & Medicaid Services, n.d.,[a]). This was multiple by the estimated proportion of
spousal caregivers to older adults nationally, estimated 21.5% the U.S.’s 17.7 million caregivers,
and divided this by two to get an annual rate (Schulz et al., 2016). To get the range of predicated
costs, we conducted the same procedure using the upper and lower confidence intervals from
bootstrap models.
73
Chapter 5: Building the policy case to scale caregiver interventions
The chapters in this dissertation have so far explored ways to improve upon existing
caregiver interventions in order to increase access to and the value of intervention programs. At
the same time, it would be naive to ignore the role of policy when trying to understand why
caregiver interventions have been accessed by so few in this role. Over the past two decades,
there has been a remarkable growth in the number of government initiatives to recognize and
support family caregivers. While the rate of development is notable during this period, even the
current pace of change will not help many of today’s caregivers, particularly those who will exit
this role long before existing programs expand to reach them. This chapter explores avenues at
the policy-level to scale individual-level interventions so that they are accessible to more
caregivers who need them.
Federal Programs Supporting Caregiver Interventions and Program Limitations
Since 2016, the National Institutes of Health (NIH) have contributed $253 million to
caregiving research, including the development and evaluation of caregiver intervention
programs and research that can inform these programs (NIH, n.d.). Following development and
evaluation of interventions, translation and scaling primarily occurs at two federal-level
agencies: the Administration for Community Living (ACL) and the Department of Veterans
Affairs (VA). (Notably, intervention research and development also occurs at these agencies.
Interventions do not necessarily flow down a single “pipeline” and often begin at state and local-
level organizations overseen by these agencies.) Specifically, the ACL oversees the Older
Americans Act (OAA), the Johnson-era legislation to which the National Family Caregiver
Support Program (NFCSP) was added in 2000 (Link, 2016). The NFCSP has since funded
services including information and referral, educational programs, counseling, support groups,
and some respite for caregivers through local Area Agencies on Aging and Title IIIE contractors
74
(Older Americans Act Reauthorization, 2000). In 2010, the VA added supports for caregivers
under the Veterans’ Caregiver and Omnibus Health Benefits Act, including respite, education
and training, and counseling (Ramchand et al., 2014).
However, lack of resources at these agencies undermines their ability to scale evidence-
based interventions for caregivers. In 2016, the total budget for the NFCSP was $125,720,936,
including all IIIE caregiver services, not just intervention programs (e.g., respite, counseling,
information and assistance) (ACL, 2017). This divides out to about $14 per caregiver based on
recent national estimates of older adult caregivers in the U.S. (Schulz et al., 2016). And while
some types of programming can be provided at this rate (e.g., referral), this amount is only
enough to provide a small fraction of caregivers with high-intensity interventions, such as home
visiting programs and multi-session counseling. Indeed, less than one percent of caregivers
received such services through the NFCSP in 2016 (ACL, n.d.). Moreover, historically, funding
for the NFCSP has not consistently increased. The NFCSP budget for 2016 was approximately
$3.5 million less than the amount budgeted for the program ten years prior (ACL, n.d.). Although
not all caregivers need or want to use formal services, caregivers seeking such supports still
report limited access (NAC & AARP, 2015). And, while the VA provides services to one
segment of the caregiving population, the majority of caregivers served are caregivers to post-
9/11 combat veterans rather than caregivers to older adults (Ramchand et al., 2014).
State and local governments, too, have overseen delivery of caregiver supports and
interventions. In California, for example, 11 Caregiver Resource Centers (CRCs) have served
caregivers in the state for over three decades (California Department of Mental Health &
California’s Caregiver Resource Centers, 2009). However, the scope of state and local programs
varies considerably. Further, state-level funding is far more precarious than federal support, in
part due to the requirement that states produce balanced budgets each year (National Conference
75
of State Legislature, 2010). When California encountered a deficit in 2010, funding cuts to the
state’s CRC programs severely curtailed their ability to serve the state’s caregivers (Senate
Select Committee on Aging and Long Term Care, 2015). Thus, while an important resource,
state and local funding sources and agencies alone cannot reliably deliver caregiver
interventions.
Surprisingly, in most states, Medicaid programs typically do not deliver caregiver
interventions and services. Following the 1999 Olmstead decision (Olmstead v. L.C., 1999),
Medicaid has funded a range of Home and Community Based services (HCBS) to Medicaid-
eligible adults who would previously have received services in a skilled nursing facility. Even
when—as is true of many cases— a caregiver is essential to an older adult’s being able to remain
in the community, caregiver needs remain overlooked by HCBS. The presence of a caregiver, in
fact, may reduce the number of services hours that are allocated to care recipients (Justice in
Aging, 2016).
Potential Cost Savings Associated with Caregiver Intervention Programs
Failure to address these limitations in federal programs is still more striking given the
value of the care provided by caregivers, and potential cost savings accrued from providing
interventions. Describing caregivers as the foundation to the U.S.’s long-term services and
supports system is more than rhetoric. If a dollar amount were applied to the contributions of
caregivers, the value of their labor would be over $470 billion per year (Reinhard et al., 2015). In
California, the contribution of caregivers exceeds the cost of the state’s Medicaid program.
Further, multiple evaluations have demonstrated cost savings to state Medicaid programs when
caregivers are supported. For example, among caregiving dyads where the caregiver participated
in the New York University Caregiver Intervention, 5-6% more Medicaid-eligible care recipients
remained in the community each year compared to controls, rendering a potential savings of
76
$40.4 million over 15 years (Foldes et al., 2017). In the state of Washington, a $3.45 million
increase in the state’s caregiver program in 2012 enabled organizations to provide more intensive
services to caregivers who would otherwise be assessed at a lower level of need. This one-year
budget bump resulted in a 2% drop in the number of care recipients using Medicaid LTSS
services, from 11% to 9% (Lavelle, Mancuso, Huber & Felver, 2014). Other potential savings
can be realized when supporting caregivers during hospital transitions.
Even beyond the cost savings, researchers are identifying a connection between caregiver
and care recipient outcomes. In addition to delaying recipient placement in a nursing home and
reducing hospital readmissions (Coleman et al., 2006; Foldes et al., 2017), recent research
suggests that supporting caregiver wellbeing supports recipient health. For example, one study
found that older adults with a neurodegenerative disorder who received assistance from a
caregiver with poor mental health experienced earlier mortality (Lwi, Ford, Casey, Miller &
Levenson, 2017). Specifically, anxiety, depression hostility, obsessive-compulsivity, and phobia
in caregivers were significantly associated with higher rates of mortality among recipients.
Caregiver depression and burden also have been associated with an increased risk of committing
elder mistreatment (Arai, Noguchi & Zarit, 2017; Beach et al., 2005; Wiglesworth et al., 2010).
Existing evidence-based interventions have been found to be effective at reducing symptoms
including depression, burden, and anxiety (e.g., Gitlin et al., 2003; Griffiths et al., 2016), thereby
raising the possibility that timely administration of caregiver interventions could benefit care
recipients as well as caregivers.
Worth noting when considering positives outcomes rendered by caregiver interventions is
the fact that benefits and potential cost savings appear to be largely realized through Medicaid
and Medicare savings and often conferred upon the patients (care recipients) receiving services
through these programs. And yet, social service organizations charged with supporting
77
caregivers typically front the high costs of interventions when they are delivered outside of a
research context. These misaligned incentives are likely an additional roadblock to translation
and scaling of caregiver interventions. Indeed, with an annual budget of $126 million, the
NFCSP simply cannot afford to provide interventions which cost hundreds and thousands of
dollars per participant to administer (e.g., Gitlin et al., 2010). And while not all caregivers
require intensive intervention programs, turnover within the caregiver population and changing
needs over time generates constant demand. This set of circumstances supports the need to not
only develop and evaluate cost-efficient interventions, but also more robust investment in the
NFCSP, and/or extension of Medicaid HCBS to include family caregiver interventions.
Policy Barriers Expanding Access to Interventions
The set of circumstances described above raises the question: why has the policy
response to provide caregivers with opportunities to participate in evidence-based interventions–
many of which were developed and evaluated using taxpayer dollars—not been more successful?
Adding to this conundrum is the apparent bipartisan support for legislation supporting
family caregivers, including several programs with major upfront costs that might otherwise be
unappealing to fiscally conservative policymakers. In 2017, major legislative efforts on behalf of
caregivers were introduced by both Republicans and Democrats in Congress, including tax
credits towards the out-of-pocket costs of caregiving (S. 1151 Ernest/H.R. 2505-Reed),
reauthorization of respite care funding (S. 1188 Collins/ H.R. 2535-Langevin), and the
Recognize, Assist, Include, Support, and Engage Family Caregivers Act of 2017 (RAISE Family
Caregivers’ Act) to set a national caregiving strategy (S. 1028-Collins/ H.R. 3759-Harper). And,
while major differences exist in the plans, both parties introduced legislation to create a national
paid family leave program (S. 337- Gillibrand/ H.R. 947- DeLauro and S. 344-Fischer/ H.R.
78
3595-Kelly). Despite demonstrated interest from both parties, by the end of 2017, only the
RAISE Act had passed in both houses of Congress.
Given potential cost savings for federal programs to support caregivers, dual benefit to
caregivers and recipients, and bipartisan support to improve and add to policies supporting
family caregivers, other factors must be explored to explain the delayed federal-level policy
response to generate robust access to caregiver interventions. To understand the cause of this
discrepancy and why polices to support caregiver interventions have been slow to develop, two
models of policymaking are considered: Kingdon’s (1984) revised “garbage can” model of
policy choice and Levitsky’s (2014) application of cultural hegemony and policy drift.
Applying complementary models of policy-making
Cultural hegemony and policy drift. Drawing on interviews and focus groups she
conducted with family caregivers, social service professionals, and leaders of caregiving
organizations, Levitsky (2014) explores why social policies to support caregivers have not
gained more traction in the U.S. She finds that many caregivers believe families bear primary
responsibility for providing care and do not look to public programs for support. To explain why
caregivers expect so little of policy, she draws on Gramsci’s theory of cultural hegemony (1971),
which suggests that present laws and policies shape citizens’ beliefs about what is owed to them.
Following this theory, because so few supports currently exist to support caregivers, it is difficult
for caregivers to envision—let alone demand—policy change that could expand access to
services and interventions. Although beliefs can change even in a policy context where no
supporting legislation exists, it takes time for these beliefs take hold. Slow changing beliefs
about the role of the state combined with fast-moving social changes (e.g., population aging)
leads to policy drift and resulting unmet need among caregivers. (See Rocco, 2017 for further
discussion on causes of policy drift as it related to policies to support family caregivers.) Even in
79
cases where caregivers seek greater state support, their efforts to expand supportive policies are
dampened by the demands of their role. For many, the caregiver role affords little time for
political advocacy.
Potential on the policy stream. Alternatively, Kingdon’s (1984) adaptation of the
“garbage can” model of organizational choice (Cohen, Mark and Olden, 1972, cited in Kingdon,
1984) can be repurposed to identify the reason policies to adequately fund intervention programs
have not been more successful. Kingdon’s revised garbage can model begins with the coupling
of societal problems with policy alternatives (i.e., solutions) and politics. Policy alternatives are
developed in a “primeval soup,” a slow-moving arena wherein ideas produced by policy
communities stew and evolve. Policy alternatives join the policy stream only when they reach a
threshold of technical feasibility and alignment with public values. When a problem is
recognized, policymakers draw on these alternatives and pull them into the policy stream.
Problem recognition can occur as a result of feedback from constituents or government agencies,
new information such as the release of major reports, or a focusing event. Policy entrepreneurs
facilitate “coupling” by connecting ideas and people, including reframing problems so they
capture the attention of policymakers. In the policy stream, proposals are moved forwards
through political bargaining and compromise. They are slowed or stopped due to high costs
associated with the proposal, changes in administration, a change in national mood, pressure
from interest groups, or even new issues that take attention away from a given proposal. Policy
alternatives must pass through the policy stream during the short time when policymakers are
receptive—a policy window. This window closely quickly, making it essential that policy
alternatives are ready to advance to the policy stream when opportunity strikes.
Although Kingdon (1984) and Levitsky (2014) approach policymaking from different
angles, the theories are complementary and point to similar barriers to advancing federal-level
80
support for polices that would increase access to evidence-based caregiver interventions. Both
models describe the agenda-setting process, and identify barriers and facilitators to this process.
Figure 5.1 illustrates how an issue moves along the policy stream per Kingdon’s (1984) model,
and incorporates Levitsky’s (2014) concept of policy drift as a barrier at the problem generation
stage. As these theories relate to policies supporting family caregivers, the two most relevant
barriers are: 1) failure to understand caregiving as an issue meriting a public policy response and
2) finding policy alternatives that reach a level of acceptability in accordance with social values
and beliefs.
5
An essential requirement for a policy alternative to make it out of the “primeval
soup” and into the policy stream, according to Kingdon (1984), is the coupling of a problem with
a policy solution. But if an issue is never defined as a problem, it cannot be coupled with a
solution. As Levitsky (2014) points out, the challenges faced by caregivers do not immediately
present as a problem meriting a policy response. If caregivers are not demanding caregiver
interventions to alleviate consequences of caregiving such as declines in mental and physical
health on their own (e.g., Capistrant, 2016), what else would compel the state to provide them?
Further, according to Kingdon (1984), policy alternatives must be adequately aligned with social
values before they are drawn into the policy stream. In the focus groups Levitsky (2014)
conducted with caregivers, she found that the way many caregivers understood family
responsibility as a social value was incongruent with advocating for additional social supports.
“Family responsibility,” to these caregivers, made caregiving an issue to be dealt with privately
rather than with public support.

5
Although Kingdon (1984) describes other obstacles to policymaking that occur along the policy
stream, often these are unavoidable. Changes in the administration, shifts in national mood, and
the emergence of new issues and funding priorities will inevitably occur, and cause other issues
to pull attention away from policy proposals to support family caregivers. These events should
not be ignored, yet foundational barriers exist even before caregiving alternatives reach the
policy stream, and contribute to policy drift.
81
Overcoming Policy Barriers to Support Access to Caregiver Interventions
To address the barriers to policymaking described in her model, Levitsky (2014) floats
two solutions. The first is to modify existing beliefs among caregivers regarding family
responsibility to make state support more acceptable. “Discursive integration,” as it is used by
Levitsky, entails incorporating novel ideas with familiar ones to alter beliefs that dampen
political demand. The author applies discursive integration to demonstrate a way to overcome
caregivers’ rejection of Medicaid HCBS services as a supportive resource for caregivers, caused
by the program’s association with so-called polluted social groups. Countering this, some
caregivers reframed Medicaid as middle-class entitlement following the spend down of assets.
By altering how they understood and talked about Medicaid, these caregivers summoned an
additional supportive resource without violating existing values. In the model illustrated by
Figure 5.1, discursive integration can help to propel policy alternatives into the policy stream by
aligning alternatives with existing beliefs. A second tool proposed by Levitsky (2014) is
“political consciousness-raising,” or the creation of a political constituency through the
generation of a group identity bound by shared political will. In focus groups, those who
provided care and did not identify as a caregiver saw their grievances as personal, not political.
Alternatively, those who accessed social services where the term “caregiver” was used were
more likely to construct a political identity. Even without overt advocacy, the language used by
social service organizations lent itself to the construction of a political constituency that had the
potential to form a caregivers’ rights movement. As shown in Figure 5.1, political consciousness-
raising supports problem generation so that problems can more easily join the policy stream
through feedback to policymakers. By developing an identity as a caregiver in the first place,
caregivers are better positioned to demand access to interventions.
82
Discursive integration and political consciousness-raising, as described by Levitsky
(2014), provide the beginnings of a toolbox to advance supportive policies including access to
interventions for family caregivers. Still, in the current political context, these tools are less
effective than they might have been when interviews were collected in 2004. At that time,
government-funded healthcare was experiencing an expansion with the passage of Medicare Part
D in 2003 (Starr, 2011) and the expansion of HCBS following the Olmstead decision five years
prior (Olmstead v. L.C., 1999). The NFCSP was still a relatively new program, signaling the
growth of caregiver supports. The current political climate is less receptive to expansive social
policies. Recent changes in the administration raised the possibility that Medicaid expansion
passed under the Patient Protection and Affordable Care Act would be retracted (Kaiser Family
Foundation, 2017). (See also Miller, Nadahs, Gusmano, Simpson & Ronneberg, 2018.) The
increased uncertainty surrounding the size of Medicaid makes the idea of reframing Medicaid a
middle-class program less tenable. The application of political conscious-raising, too, is a
compelling idea to problematize caregiving. But there is a catch-22 to this approach. If the
language of social service organizations are the drivers of political consciousness-raising,
specifically their use of the term “caregiver,” the reach of consciousness-raising will remain
limited. With a weak network of social support services available through the modest Title IIIE
budget, opportunities for political consciousness-raising are constrained. A constituency is
needed to demand more intervention services, but more services are needed to generate a
constituency—a circular dynamic foiling forward progress.
The advancement of policies to support access to caregiver interventions in the current
political climate requires new tools and approaches. These tools must be applicable in a political
climate that is less receptive to expansive social policies. At the same time, they must be capable
of rendering changes that are more than incremental in nature, since gradual change could
83
perpetuate policy drift wherein a shrinking proportion of caregivers ever access programs and
interventions. The tools described below provide specific and up-to-date pathways to overcome
policy drift, and expand access to evidence-based caregiver interventions.
Political Consciousness-Raising through Encounters with Health Services and Employers
Social services to family caregivers reach very few in this role, such that political
consciousness-raising through these services would be inefficient. Further, there is often a delay
between the start of the caregiving role and when caregivers begin to identify with the term
“caregiver” (Peterson, 2016). Consequently, newer caregivers are even less likely to access
services compared to caregivers with more experience in this role, despite the fact, as
demonstrated in Chapter 4, they may experience the greatest need for some types of intervention.
In contrast, caregivers frequent other institutional settings, including healthcare
organizations and workplaces. In a survey, 55% of caregivers reported the person they provide
care to visited the hospital within the last year (AARP & NAC, 2015). Moreover, health events
preceding visits to a healthcare facility (e.g., fractures, stroke) and diagnoses that occur in these
settings can initiate the caregiving role. Survey data also reveals that 59% of caregivers work in
paid employment (AARP & NAC, 2015). The caregiving role may be particularly salient for
employers to discuss with employees in order to reduce the risk of “presenteeism,” wherein
employed caregivers are present but less productive given caregiving demands (Wolff, Spillman,
Freedman & Kasper, 2016). Integrating policies that encourage the identification of caregivers in
healthcare and workplace settings could expand the ranks of family members who identify with
the term “caregiver.” By identifying as a caregiver, those assisting a family member would be
84
better equipped to join a constituency of caregivers demanding supportive policies including
access to interventions (Levitsky, 2014).
6

In addition to serving as venues where families can begin identifying as caregivers,
healthcare organizations and workplaces are also places where interventions can be feasibly
delivered. Several interventions have been developed to address caregiver needs during hospital
transitions (Gibson, Kelly & Kaplan, 2012).
7
Workplaces, too, are suitable for intervention
delivery. As employers seek new ways to recruit employees, employee assistance programs that
meet the needs of employed caregivers are a promising new benefit (Schulz et al., 2016). Digital
intervention tools hold particular appeal for working caregivers. The Powerful Tools
psychoeducational program has been successfully translated as a digital intervention provided to
caregivers employed at Exxon Mobil Corporation, IBM and Texas Instruments (Kuhn et al.,
2008). A digital service delivery program such as FCA CJ, described in Chapter 1, could have
particular appeal as an employee assistance program.
Problematizing caregiving using data
Another way to transform caregiving into a policy problem and support widespread
provision of evidence-based intervention is by collecting and reporting on data regarding the
needs of caregivers. In order to create a political demand for increased access to caregiver
interventions, high-quality data is needed to demonstrate the pervasive occurrence of the
problems that interventions have been developed to alleviate. While there is extensive research

6
Explicit identification of caregivers in healthcare settings is already occurring in 36 states
through various renditions of the Caregiver Advise Recognize Enable (CARE) Act legislation,
which requires caregivers to be added to patient health records in hospitals when reported by the
patient (AARP, 2018).
7
Medicare recently added a billing code for caregiver consultations following diagnosis of
cognitive impairment in 2017 (Department of Health and Human Services, 2016). This could be
expanded to cover early interventions to prepare to caregivers as conditions progress.
85
reporting on the effects of caregiving, much of this literature has suffered from methodological
weaknesses that can be attributed to current data sources, including minimal longitudinal
datasets, small and/or unrepresentative samples, and inadequate measures (Schulz et al., 2016).
Collecting and reporting on data about the types and extent of challenges faced by family
caregivers to policymakers could redefine the experiences of caregivers as a public health and
policy problem, even if caregivers themselves are not pressing for a policy response. Indeed,
informative reports are described as one way in which problems can join the policy stream by
Kingdon (1984). Regular collection of panel data under the National Study of Caregiving
(NSOC) should remain an advocacy priority, as well as regular inclusion of detailed caregiving
modules in other longitudinal data sets such as the Health and Retirement Study.
8

Coupling caregiving as a solution to problems within healthcare
Kingdon’s (1984) model indicates another way to move caregiving policies along the
policy stream: by fitting alternatives generated by policy specialists in one field to a problem
from a different field when the opportunity arises. Specifically, supportive caregiver policies can
be framed as a solution to a problem that frequently captures the interests of policymakers: the
high costs of healthcare. Findings described in Chapter 4 demonstrated higher rates of healthcare
utilization among spousal caregivers compared to non-caregivers, contributing to a predicted
$98.8 million in additional physician visits per year. Curbing health service utilization could
lower the amount the federal government pays into healthcare. Caregiver interventions can also

8
In addition to helping to frame caregiving as a policy problem, large secondary data can be
used to improve existing intervention programs. Large samples of caregivers would enable
researchers to stratify analyses, enabling better understanding of the unique needs of sub-
populations of caregivers, such as Asian and LGBT caregivers on whom there is little data given
small sub-samples of these populations in existing surveys. This information could be used to
increase the value of interventions to family caregivers and the organizations that currently
deliver them.
86
reduce the cost of providing health services to care recipients, as already demonstrated in the
savings realized through Medicaid. An evaluation of a multi-month health education intervention
delivered to spousal caregivers at an HMO revealed a $309,461.14 savings in healthcare
expenditures among caregivers and care recipients compared to controls (Toseland & Smith,
2006).
9

Discursive integration of caregiver supports as a Medicaid and Medicare benefit
Reframing caregiver interventions as a way to support care recipients also provides a path
to sustainable funding. Medicaid HCBS have been described as “person-centered” while
excluding family caregivers who make it possible for recipients to remain in the community. Yet,
the benefits realized by care recipients of caregivers who access interventions provides a strong
justification for expanding Medicaid HCBS and Medicare post-hospital benefits to include
caregiver interventions.
10
Discursive integration can be used to frame caregiver interventions as a
way to promote health and wellbeing among care recipients to gain traction for Medicaid and
Medicare reimbursement of caregiver interventions.
11
The intervention described in Chapter 3
for example, certainly would benefit care recipients, if effective at preventing mistreatment, at
the same time as supporting caregivers. Describing caregiver interventions as a way to support

9
It is worth considering the possibility that participation in interventions could actually increase
in service utilization due to an increased awareness of recipient needs and opportunities to access
services. Indeed, not all evaluations of social service interventions have revealed healthcare
savings (Weinberge et al., 1993). Even if interventions lead to higher use in some cases, it would
not necessarily be a negative outcome, and could prevent catastrophic care needs from
developing in cases where the caregiver is dangerously unable to assist (e.g., DeLiema et al.,
2016).
10
Notably, the definition of person-centered care established in recent research does include
families. Using consensus methods among providers, Coulourides, Wilber & Mosqueda (2016)
conclude that “Person-centered care is achieved through a dynamic relationship among
individuals, others who are important to them, and all relevant providers.”
11
There is a notable critique to this approach. A policy case to support caregivers that is built on
the potential benefits to recipients perpetuates the idea that caregivers are not deserving of
support because of their own needs.
87
recipient of care could also appeal to and generate demand from caregivers themselves, many of
whom often report an interest in learning ways to better support care recipients (Scharlach et al.,
2006).
Discussion
Despite considerable investment towards developing and evaluating caregiver
interventions over the past several decades, only a small portion of caregivers ever access
intervention programs. Even with a considerable evidence-base, potential cost savings, and
bipartisan support, only a small proportion of caregivers access interventions each year outside
of a research context. Existing pathways to implementation are crippled by inadequate and
unreliable funding. Models of policymaking developed by Levitsky (2014) and Kingdon (1984)
can be fruitfully applied to identify barriers to creating policies that increase access to caregiver
interventions and to develop tools to overcome existing barriers. Promising strategies include:
political consciousness-raising through encounters with health services and employers,
problematization of caregiving using quality data, coupling caregiving as a solution to the high
costs of healthcare, and discursive integration of caregiver supports as Medicaid and Medicare
benefits.
Potential weaknesses in the identified policymaking tools
The tools proposed to advance policy-level support for caregiver interventions do have
limitations. First, they do not address barriers to policymaking that occur along the policy stream
(e.g., change in administration, change in policy priorities by policymakers), and instead focus
on barriers to placing the need for caregiver interventions on the policy agenda in the first place.
I prioritize the strategies that can be implemented at the agenda-setting stage since barriers at this
stage must resolved in order to advance major legislation along the policy stream. Secondly, it is
88
not known whether the tools described will be effective. The suggested tools are best understood
as promising approaches.
Conclusion
By drawing on models of policymaking and the discussion of caregiving experts, I
identify tools to increase access to caregiver interventions at the policy-level. The slow rate of
change in caregiving policies, even as the size of and challenges encountered by this population
have rapidly changed, has led to serious health and financial consequences for family members
providing valuable support to older adults. Incremental increases in funding for caregiver
interventions from the National Family Caregiver Support Program will not keep pace with
changing social contexts, and will leave many caregivers vulnerable. To stem the growth of gaps
between caregiver needs and policy responses, strategic approaches like those suggested here are
necessary.

Parts of this chapter were previously published in the Journal of Aging and Social Policy:
Meyer, K., Rath, L., Kaiser, N., Gassoumis, Z.D., & Wilber, K. (2018). What are strategies to
advance policies supporting family caregivers?: Promising approaches from a statewide task
force. Aging and Social Policy. doi: 10.1080/08959420.2018.1485395

89
Chapter 6: Discussion and Conclusion
This dissertation described new ways to add value to and increase access to interventions
to attenuate negative aspects of caregiving experienced by some caregivers. Despite decades of
research, too few caregivers ever benefit from interventions demonstrated to alleviate declines in
caregivers’ health and wellbeing, and overall lack of preparation to complete the duties required
within the caregiving role. Even as a growing number of Americans are likely to become
caregivers, demand for care is expected to surpass the availability of family caregivers given
ongoing and anticipated demographic changes (e.g., population aging, geographic dispersion of
families) (Schulz et al., 2016). Already, nearly half of caregivers feel they have no choice in
whether to provide care (NAC & AARP, 2015), a consequence of demand outstripping the
number of those willing and able to provide assistance to an older relative with an illness or
disability. It is thus critical to protect the health and wellbeing of those in this role, and equip
them with the resources they need to provide care to the best of their ability. While a
considerable amount of resources have already been invested to develop and evaluate efficacious
behavioral interventions, too few caregivers ever access these programs, and consequently
remain less-equipped than they could be to provide care to another without risking their own
wellbeing. The intent of this dissertation was to explore promising ways to remedy the
accumulation of research on behavioral interventions for family caregivers that is never or only
modestly ever translated and scaled in the community.
To encourage successful translation and scaling of evidence-based caregiver
interventions, each chapter presented in this dissertation explored a “new direction” or way that
caregiver interventions can become more accessible and/or valuable to caregivers, service
providers, and/or funders. The new directions considered in Chapters 2 to 4 include the
intervention delivery mode, targeted outcomes, and timing of intervention administration. Each
90
chapter applies a different type of methodological approach to examine these new directions,
including an evaluation of social service client data, a literature review drawing on lessons
learned from different fields of family violence, and an analysis of longitudinal data from a large
secondary dataset. Chapter 5 does not consider specific intervention features, but applies theories
of policymaking to derive strategies 1) to hasten translation of research on caregiver
interventions to make evidence-based programs available to caregivers in service settings, and 2)
to create sustainable mechanisms for intervention delivery in communities. For readers’
convenience, I provide a summary of each chapter and key findings below.
Summary of chapter findings
Chapter 2: What are the characteristics of caregivers using online support services?
Chapter 2 explored the characteristics of caregivers approaching information and
education resources using an online mode of service delivery over the usual service delivery
mode (e.g., telephone or in-person assessment followed by mailed materials) at two social
service sites in California. Online service delivery is appealing as a way 1) to reach caregivers
who typically do not access social services, and 2) to lower the cost of service delivery so
provider organizations can reach more caregivers than is possible while exclusively using usual
service delivery modes. Knowing the characteristics of those caregivers using online modes of
service delivery is important for provider organizations to know when deciding 1) whether to
invest in online programs and interventions and 2) what types of content to include in online
versus usual service delivery options.
Results from our evaluation showed that relatively few caregivers approached services
online (13.7%) over usual modes of service delivery, a result that did not vary even after the
online option was available to caregivers for many months. Logistic regression model results
indicated there were few predisposing demographic or need-based (i.e., caregiving intensity)
91
differences between caregivers who accessed services online over usual service delivery modes.
In some respects, these findings are encouraging for community organizations seeking to invest
resources in providing services online. Results suggest that expanding online intervention
options would likely not negatively impact underserved populations of caregivers so long as
online modes of delivery do not replace usual service delivery options. Moreover, we found that
factors like the service delivery site and how caregivers learned about services were among the
strongest factors associated with online service use. These findings suggest that rates of online
service delivery are likely modifiable. Targeted marketing strategies of online services (e.g.,
advertisement in health service settings) and training of service providers could increase uptake
of online delivered services. Chapter 2 added to previous survey data on the characteristics of
caregivers using online information services (e.g., Li, 2015) by providing an opportunity to
observe the actual behaviors of caregivers approaching services. This is an important addition,
since caregiving research volunteers typically do not reflect the actual caregiving population
(Pruchno, Brill, Shands, Gordon, Genderson, Rose & Cartwright, 2008). Additional evaluation is
needed to understand the reasons that some caregivers elect to use online-delivered services (e.g.,
greater flexibility as to when to access services) and the extent to which caregivers continue to
engage with online programs like FCA CJ over time (e.g., how often do caregivers log into their
digital resource dashboard). Further, it will be important to learn whether caregivers who access
programs like FCA CJ experience similar outcomes as caregivers who use the usual service
delivery modes (e.g., changes in caregiver burden, increased caregiver competency).
Chapter 3: What can elder mistreatment researchers learn about primary prevention from
family violence intervention models?
Chapter 3 drew upon research from child maltreatment and intimate partner violence to
identify promising components of a first-of-its-kind home visiting intervention to prevent elder
92
mistreatment by family caregivers. Elder mistreatment and quality of care are infrequent targets
of caregiver interventions, which typically focus on the wellbeing of the caregiver. However, as
important as it is to preserve caregiver wellbeing, there are compelling reasons to expand
targeted outcomes of caregiver interventions to include elder mistreatment prevention and the
related issues of family conflict, quality of care, and potentially harmful behaviors by a
caregiver. Approximately one-quarter to one-half of caregivers to older adults engage in at least
one type of elder mistreatment or potentially harmful behaviors (e.g., Beach et al., 2005;
Wigglesworth et al., 2010). At the same time, caregivers express a desire to provide high quality
care (Cheng et al., 2015), and are surprisingly willing to disclose engagement in mistreatment
behaviors (Wigglesworth et al., 2010). Given disproportionately high prevalence of elder
mistreatment of care recipients compared to prevalence rates generally, and indications that
caregivers are willing to address issues related to quality of care, developing an elder
mistreatment prevention intervention targeted at family caregivers holds great promise. At the
same time, there are no existing elder mistreatment prevention programs to provide guidance on
how to go about developing such a program.
In Chapter 3, I addressed the lack of previous research on elder mistreatment prevention
interventions with family caregivers by examining promising intervention approaches found in
other fields of family violence. The choice to focus on home visiting interventions was driven by
primarily by the frequent and successful use of home visiting programs in child maltreatment
prevention and other family caregiver interventions programs. A home setting also enables
interventionists to account for ecological risk and protective factors that can be most easily be
observed in a home setting (e.g., safety of the home). Based on our review of the literature, we
proposed the structure and components of an intervention program to prevent elder mistreatment
by a family caregiver. First, we recommended that a prevention program target those caregivers
93
who are new to this role, when negative events are less likely to have already occurred, and those
who are experiencing a major change in the caregiving role (e.g., acute health event). Secondly,
the most effective prevention interventions from other fields of family violence occur over an
extended period of time. Given this, we proposed a program with at least 12 contacts delivered
over 6 or more months. Third, an intervention should be tailored to the unique risk profiles found
in dyads or families, rather than a one-size-fits-all curriculum. Fourth, the intervention should be
assessment-driven in order to identify the specific needs of each family, and should provide
feedback to participants so they are aware of both their strengths and aspects of caregiving they
can improve upon (e.g., communication with the care recipient). Moreover, given the extended
delivery period of the proposed intervention, we recommended regularly reassessing families as
needs and risk profiles evolve. Fifth, the intervention should include a training component rather
than passive learning alone, a feature that has regularly demonstrates improved outcomes in
meta-analyses of caregiver interventions. Finally, the program we proposed includes strengths-
based feedback in order to encourage continued engagement in the intervention and investment
in behavior change through gains in self-efficacy.
While we are hopeful about the potential of the intervention program described to prevent
elder mistreatment and improve quality of care, we recognize that proposing an intervention is
very different from developing, evaluating, and implementing such a program. A next step for
this research is to pilot the proposed intervention to determine feasibility and acceptability
among caregivers, as well as potential safety risks, and identify opportunities to mitigate the
challenges described in Chapter 3(e.g., recruitment and retention) before administering a
randomized control trial to determine program efficacy.

94
Chapter 4: The effect of number of years caregiving on physician visits and hospitalizations
In Chapter 4, I used longitudinal data from the Health and Retirement Study to examine
variation in the use of healthcare services by spousal caregivers compared to non-caregivers
according to the number of years spent in a caregiving role. In addition to determining whether
spousal caregivers use more health services than non-caregivers, this study adds to previous
literature by shedding light on when health promotion interventions should be administered to
caregivers. To date, few intervention programs have considered the ideal timing of intervention
administration. The results of fixed effects models indicated that rates of physician visits and
odds of experiencing a hospitalization peaked relatively early in the caregiving role, and then
declined after approximately four years spent as a caregiver. These findings suggest not only that
caregivers experience greater health service use than non-caregivers, but that there is variation in
the need for health services among caregivers at different points along the caregiving trajectory.
Interventions to address caregiver health that are delivered early in the caregiving career may be
most useful to caregivers, as they begin to cope with stressful new caregiving challenges. If
effective, early intervention to protect caregivers’ health could attenuate the $98.8 million in
additional yearly costs associated with physician visits by spousal caregivers. Future research in
this area should consider whether declines in rates of health service use at later points in the
caregiving career are the result of impaired access to services, including lack of available respite
and high costs associated with some healthcare services. In addition, researchers should examine
variation in rates of health service use within the caregiving population (e.g., assisting someone
with dementia, adult child caregivers) to further inform appropriate timing of intervention
administration.

95
Chapter 5: Building the policy case to scale caregiver interventions
Finally, Chapter 5 drew on theories of policymaking to describe barriers to translating
and scaling existing intervention programs, and to identify opportunities to overcome these
barriers. Applying complimentary theories of policymaking described by Sandra Levitsky (2014)
and John Kingdon (1984), I described two primary barriers to expanding access to behavioral
interventions to support family caregivers at the policy level. First, caregiving is rarely
recognized as a policy problem even among caregivers, who perceive caregiving and its effects
as private family matters. Consequently, the need for caregiver interventions has largely failed to
make it onto the policy agenda due to lack of perceived need for intervention programs and
constituent demand. Secondly, intervening to provide supports and services to family caregivers
runs contrary to U.S. ethos of “personal responsibility.” Given this, it is challenging to find
policy alternatives (i.e., solutions) that capture widespread support among policymakers and
constituents.
To overcome these barriers to generating caregiver-supportive policies and to promote
translation and scaling of interventions, I proposed four novel strategies to increase the flow of
federal funding to enable sustainable implementation evidence-based caregiver interventions.
First, I recommended using healthcare and employment venues to encourage family members
and friends to self-identify as caregivers. Self-identification as a “caregiver” in settings
frequented by caregivers can promote political consciousness-raising, or the creation of a
political constituency through the generation of a group identity bound by shared political will.
By forming a caregiver political constituency, caregivers would be better able to exert pressure
on policymakers to treat their concerns as a policy problem. Secondly, I recommended collecting
and reporting on data to strengthen evidence on the challenges caregivers encounter throughout
the caregiver trajectory. This is important so that, even if caregivers do not perceive caregiving
96
as a policy problem, data reports can be used to show the need for supportive interventions.
Third, I recommended coupling caregiver interventions as a solution to other pressing policy
problems, namely the high costs of healthcare. Even if protecting the health and wellbeing of
family caregivers in itself is not widely perceived as reason enough to fund greater access to
interventions, associating caregiving with a policy topic that has high traction, like the high costs
of healthcare, may generate support. Evidence such as that presented in Chapter 4 can be applied
to begin making this case. Finally, I recommended framing caregiver interventions as a way to
support the health of care recipients. Caregiver health and wellbeing is often neglected even
when caregivers are essential to recipient care plans; framing caregiver interventions as a way to
improve the health of care recipients can pave the way for a policy case for Medicare and
Medicaid reimbursement for caregiver behavioral interventions.
Further evaluation is needed to determine which of the described strategies has the
greatest potential to impact change in caregiving policies. As a next step, policy researchers
might compare political candidate’s policy platforms over time to observe whether the
candidates include more caregiving issues given changes such as: the increasing availability of
quality longitudinal data on caregivers, widespread dissemination of research on the added health
service costs associated with caregiving (e.g., popular news segments, special issue journals),
and the incorporation of caregiver supports in employment settings (e.g., caregiver-focused
employee assistance programs). It is likely that a combination of the strategies described will
yield the greatest potential to generate the political will needed to create supportive caregiver
polices, including access to evidence-based interventions.
Next steps: Developing and evaluating sustainable intervention programs
In addition to the takeaways described in each chapter of this dissertation, I make one
further recommendation for fellow researchers when approaching the next generation of
97
caregiver interventions. I urge researchers to add a new criterion when developing and refining
intervention programs for caregivers: can the intervention be funded using existing and reliable
funding mechanisms? Developing intensive interventions that adhere exclusively to what is
considered the most optimal intervention approach according to meta-analyses is a futile
exercise, and will continue to miss the mark of supporting caregivers so long as what is
considered optimal is not recognized as fundable. Despite comparisons of the effects of
behavioral intervention programs to those of pharmaceutical interventions (e.g., Gitlin et al.,
2016), caregiving researchers must also acknowledge an important difference: we do not yet
have the luxury of ready-made reimbursement options (e.g., Medicare) enjoyed by other aspects
of healthcare. While it is important to continue advocating for the extension of funding to
support access to caregiver interventions, as described in Chapter 5—including those programs
that are costly but also highly valuable for families with high levels of need— balancing efficacy
with effectiveness may help shorten the oft-quoted 17-year delay between research and
translation (Morris et al., 2011). The “new directions” described in this dissertation, including
online delivery mechanisms and administration of interventions early in the caregiving
trajectory, can help balance the need to produce high-quality intervention programs with the need
for programs to be feasible to translate and scale. Digital interventions, for example, once
produced, may cost less than traditional in-person intervention options (e.g. Blom et al., 2015),
thereby supporting their scalability using existing funding sources. Similarly, prevention or early
intervention can lower level of need among families so that fewer caregivers require high-
intensity and high-cost interventions.
The recommendation that researchers focus on intervention sustainability rather than just
efficacy may elicit concerns about the preserving quality of intervention programs. Granted,
results from interventions that fail to produce positive outcomes could undermine the policy case
98
to reimburse interventions through Medicare and Medicaid or to increase funding for
interventions through the National Family Caregiver Support Program. However, the
accumulation of decades-worth of efficacious but untranslated behavioral intervention research
suggests that willful naivety with regards to intervention implementation and sustainability is a
sure path to failure if our purpose is to improve the lives of family caregivers.
The choice between efficacious and effective intervention design is not “all or nothing”;
there are ways researchers can develop fundable and sustainable intervention programs while
preserving program efficacy. One way to do this is to design flexible programs that can be
tailored by community organizations given unique community and client needs and available
resources. While the intervention described in Chapter 3 is an intensive, multicomponent
intervention—features which are associated with improved caregiver outcomes but also high
administration costs (e.g., Nichols et al., 2008)— the focus on intervention components and
structure rather than a more detailed description of an intervention model was intentional. An
expectation of variation in the intervention across sites would allow service organizations to use
available resources to administer the intervention, rather than forgoing a promising program
altogether because it is perceived by organization leaders to be infeasible to implement. For
example, how organizations provide training to caregivers in the intervention described in
Chapter 3 could vary depending on the staff expertise. If the intervention is administered in a
healthcare setting, families might receive training on providing complex care tasks directly from
service providers; if administered in a social service setting, this aspect of the program might be
best delivered using the growing number of online resources on how to provide complex care
(e.g., AARP’s Home Alone Alliance web-video series:
https://www.aarp.org/ppi/initiatives/home-alone-alliance/).
99
Undoubtedly, variation in intervention administration will lead to variation in program
outcomes, but compromising outcomes—to an extent—to support increased access is worth
considering. Prioritization of strict adherence to a complex intervention programs can undermine
their replication (Gitlin & Hodgson, 2015). Relaxing adherence standards may improve access
by reducing the resource-intensity of program administration. As others have pointed out,
behavioral interventions are not like pharmaceutical interventions (Glasgow et al., 2003); precise
replication of behavioral interventions is already an unrealistic goal. Why insist on a high degree
of fidelity when, for example, interventionists inevitably vary according to their own
interpersonal skills, training, and experiences? The many opportunities for variation to occur
makes administration of behavioral interventions an inherently imprecise endeavor. Nor is exact
replication necessarily a desirable goal. Not only do the capacity of organizations to administer
interventions vary, but families, too, differ in their needs and preferences. A degree of flexibility
supports provision of relevant programming to participants. Variation in caregiver interventions
has already been accepted by experts in the field, although it is seldom observed in practice. In
1995, the National Institute on Health funded evaluation of many different caregiver
interventions under the Resources for Enhancing Alzheimer’s Caregiver Health (REACH). In a
summary of the results from the REACH evaluation, the leaders of this research argued against
one-size-fit-all interventions: “. . . we subscribe to a structured—but at the same time, tailored—
approach to delivering interventions that are responsive to individual risk profiles” (Schultz et
al., 2016, pp. 519).
At the same time, I recognize the need for “quality control,” and thus encourage
evaluations of intervention programs which examine the essential components and structural
aspects which are necessary to achieve an acceptable level of efficacy. Rather than examining
administration of a single, static intervention program, specific structures and key components of
100
an intervention should be examined for how much they contribute to program success for a given
outcome. Features like the timing of program administration (e.g., immediately after diagnosis of
the recipient condition), as discussed in Chapter 4, might be a highly relevant feature to effecting
more positive outcomes for caregivers. Likewise, the intervention delivery mode may also
moderate success, where caregivers providing high intensity care (e.g., assisting someone with
dementia) may be best served through in-person and telephone-delivery rather than online, a
possibility raised in Chapter 2. Evaluating a dynamic intervention program with intentional
variation across sites could yield valuable information about critical aspects to include in an
intervention, and be used to identify essential components to assess when examining fidelity
during replication.
Another way for researchers to contribute to sustainable caregiver interventions is to hone
existing evidence-based programs and make these interventions more feasible to fund,
administer, and participate in. For example, while the Savvy Caregiver program described in
Chapter 1 is typically delivered over six modules, a three-module version is being evaluated in
Los Angeles County. Reduced staff time to administer the program may make it more viable to
provide to a larger number of caregivers. Preliminary findings suggest this approach holds
promise for reducing depression (Aranda, 2018). Consolidation of the Savvy Caregiver program
modules also meets a specific community need; high traffic levels in Los Angeles may deter
individuals from completing the intervention given the additional time required to participate in
the program due to transport time. Consolidation of the intervention modules reflects
responsiveness to a community need, and makes the program more viable to provide.
Translation of existing programming using digital modes of delivery and development of
new programs using digital technologies can also increase access to intervention programs. Both
the Savvy Caregiver and the Powerful Tools for Caregivers interventions, also described in
101
Chapter 1, have been successfully delivered using digital tools including a web-platform and app
(Griffiths et al., 2016; Kuhn et al., 2008). And while the elder mistreatment prevention program
described in Chapter 3 includes a home-visiting program, an online intervention program may
have additional merit for some families. The added degree of anonymity rendered through online
delivery may hold greater benefit and appeal for caregivers who are at risk of mistreating the
person they assist or who have already done so. Anonymity may enable caregivers to be more
willing to build self-awareness of their potentially harmful caregiving behaviors, rather than
becoming defensive while working with someone in-person. Online intervention delivery may
also be a way to sustainably fund access to intervention programs by encouraging employers
cover the costs. Demonstrated outcomes such as reduced job stress—found in the online
Powerful Tools for Caregivers program when it was delivered at several business (Kuhn et al.,
2008)—may be appealing to employers seeking ways to retain employees who are caregivers.
Digital delivery options may make it easier for employers to provide employees with caregiver
responsibilities with evidence-based interventions given: 1) the option to have expert
organization administer the program from anywhere in the world; 2) convenience when
replicating the intervention when more employees become caregivers; 3) the option for
caregivers to complete the program at convenient times (i.e. non-work hours). At the same time,
low-touch digital interventions may not be appropriate for caregivers with high-needs levels.
However, as indicated in Chapter 4 when examining health service utilization by caregivers,
negative outcomes appear to increase in the first years of caregivers. Digital interventions may
be most appropriate for caregivers who are early in this role when risk profiles are relatively low,
and may provide a way to prevent risks from increasing. Still, more research is needed to
determine when in the caregiving trajectory the administration of digital interventions is
preferable to in-person delivery.
102
Conclusion
Previous research has consistently demonstrated the capacity for caregiver interventions
to reduce some of the negative consequences associated with caregiving. Researchers in this field
have at least a foundational idea about which approaches work best (e.g., multicomponent
programs, psychoeducational programs, inclusion of training components). Given this stage of
development, researchers and interventionists ought to place higher priority on the fundability
and sustainability of intervention programs for caregivers, provider organizations, and funders,
including policymakers who oversee budgets for federal programs. For interventions to be
valuable to caregivers, delivery options should be convenient to potential users (Chapter 2),
intended outcomes must address problems that pervade their caregivers’ daily experiences
(Chapter 3), and interventions should be tailored to meet relevant needs and preferences based on
both intervention timing and assessed need (Chapter 3 & Chapter 4). Even while building the
policy case to sustain caregiver interventions using strategic approaches described in Chapter 5,
researchers can encourage the translation and scaling of caregiver interventions by developing
flexible programs, as well as honing existing interventions so they can be more easily accessed.
The chapters in this dissertation demonstrate promising ways to increase the value of
current caregiver intervention programs, and strategies to scale programs so they can reach a
greater proportion of caregivers. Half a century ago, awareness of the evolving demands of
families resulted in the development of early caregiver interventions. Since then, demands on
caregivers have changed, a new and diverse cohort of caregivers has emerged, technologies have
generated new opportunities to support families, data is available to describe previously
unrealized consequences of caregiving, and there is an array of programs and agencies available
to oversee implementation of caregiver interventions. There is immense opportunity to build
upon what is already known from robust body of existing research to develop an improved
103
generation of caregiver intervention programs that are accessible and responsive to the unique
and evolving needs of family caregivers.

104
References

Aarons, G. A., & Palinkas, L. A. (2007). Implementation of evidence-based practice in child
welfare: Service provider perspectives. Administration and Policy in Mental Health and
Mental Health Services Research, 34(4), 411-419.
AARP. (2018, August). New State Law to Help Family Caregivers. Retrieved online at:
https://www.aarp.org/politics-society/advocacy/caregiving-advocacy/info-2014/aarp-
creates-model-state-bill.html
AARP. (2016, April). Caregivers & Technology: What They Want and Need. Retrieved online
at: http://www.aarp.org/content/dam/aarp/home-and-family/personal-
technology/2016/04/Caregivers-and-Technology-AARP.pdf
Acierno, R., Hernandez-Tejada, M., Muzzy, W., & Steve, K. (2009). Final report: The National
Elder Mistreatment Study. Washington, DC: U.S. Department of Justice.
Adelman, R. D., Tmanova, L. L., Delgado, D., Dion, S., & Lachs, M. S. (2014). Caregiver
burden: A clinical review. JAMA 311(10), 1052-1060.
Administration on Community Living. (2017). Older Americans Act (OAA). Retrieved online at:
https://www.acl.gov/about-acl/older-americans-act-oaa
Administration on Community Living. (n.d.). AGing Integrated Database. Retrieved online at:
https://agid.acl.gov/CustomTables/
Alkema, G. E., Reyes, J. Y., & Wilber, K. H. (2006). Characteristics Associated With Home-
and Community-Based Service Utilization for Medicare Managed Care Consumers. The
Gerontologist, 46(2), 173-182.
Allison, P. (2009). Fixed Effects Regression Models. Los Angeles, CA: SAGE
Allison, P. D., & Waterman, R. P. (2002). Fixed-effects negative binomial regression models.
Sociological Methodology, 32(1), 247-265.
105
Anda, R. F., Felitti, V. J., Bremner, J. D., Walker, J. D., Whitfield, C. H., Perry, B. D., ... &
Giles, W. H. (2006). The enduring effects of abuse and related adverse experiences in
childhood. European Archives of Psychiatry and Clinical Neuroscience, 256(3), 174-186.
Andersen, R. M. (1995). Revisiting the behavioral model and access to medical care: Does it
matter? Journal of Health and Social Behavior, 36(1), 1–10.
Andrew Chang & Company. (2015, July) Paid Family Leave Market Research (Report for the
Employment Development Department for the State of California). Retrieved online at:
http://www.edd.ca.gov/disability/pdf/Paid_Family_Leave_Market_Research_Report_201
5.pdf
Aranda, M. (2018, April). Savvy Caregiver Express: Preliminary Pilot Results and a Bit on
Challenges to Scaling Up. What’s Hot in Aging at the Leonard Davis School of
Gerontology. Los Angeles, CA.
Arai, Y., Noguchi, C. & Zarit, S. H. (2017). Potentially harmful behavior by caregivers may be
predicted by a caregiver burden scale. International Journal of Geriatric Psychiatry, 32
(5): 582-583.
Bair-Merritt, M. H., Jennings, J. M., Chen, R., Burrell, L., McFarlane, E., Fuddy, L., & Duggan,
A. K. (2010). Reducing maternal intimate partner violence after the birth of a child: a
randomized controlled trial of the Hawaii Healthy Start Home Visitation Program.
Archives of Pediatrics & Adolescent Medicine, 164(1), 16-23.
Bandura, A. (1977). Self-efficacy: toward a unifying theory of behavioral change. Psychological
review, 84(2), 191-215.
Beach, S. R., Schulz, R., Williamson, G. M., Miller, L. S., Weiner, M. F., & Lance, C. E. (2005).
Risk factors for potentially harmful informal caregiver behavior. Journal of the American
Geriatrics Society, 53(2), 255-261
106
Beauchamp, N., Irvine, A. B., Seeley, J., & Johnson, B. (2005). Worksite-based internet
multimedia program for family caregivers of persons with dementia. The Gerontologist,
45(6), 793-801.
Bédard, M., Molloy, D. W., Squire, L., Dubois, S., Lever, J. A., & O'Donnell, M. (2001). The
Zarit Burden interview a new short version and screening version. The Gerontologist,
41(5), 652-657.
Blieszner, Rosemary, Roberto, Karen A., Wilcox, Karen L., Barham, Elizabeth J., & Winston,
Brianne L. (2007). Dimensions of Ambiguous Loss in Couples Coping with Mild
Cognitive Impairment. Family Relations, 56(2), 196-209.
Blom, M. M., Zarit, S. H., Zwaaftink, R. B. G., Cuijpers, P., & Pot, A. M. (2015). Effectiveness
of an Internet intervention for family caregivers of people with dementia: results of a
randomized controlled trial. PloS One, 10(2), e0116622.
Boise, L., Congleton, L., & Shannon, K. (2005). Empowering family caregivers: The powerful
tools for caregiving program. Educational Gerontology, 31(7), 573-586.
Boots, L. M. M., Vugt, M. E., Knippenberg, R. J. M., Kempen, G. I. J. M., & Verhey, F. R. J.
(2014). A systematic review of Internet ‐based supportive interventions for caregivers of
patients with dementia. International Journal of Geriatric Psychiatry, 29(4), 331-344.
Braithwaite, S. R., & Fincham, F. D. (2009). A randomized clinical trial of a computer based
preventive intervention: Replication and extension of ePREP. Journal of Family
Psychology, 23(1), 32.
Brandl, B., & Raymond, J. (2012). Policy implications of recognizing that caregiver stress is not
the primary cause of elder abuse. Generations, 36(3), 32-39.
107
Brossoie, N., Roberto, K. A., Willis-Walton, S. & Reynolds, S. (2011). Report on baby boomers
and older adults: Information and service needs. Retrieved online at:
http://www.gerontology.vt.edu/docs/n4a%20Report.pdf
Brown, J., Cohen, P., Johnson, J. G., & Salzinger, S. (1998). A longitudinal analysis of risk
factors for child maltreatment: Findings of a 17-year prospective study of officially
recorded and self-reported child abuse and neglect. Child Abuse & Neglect, 22(11), 1065-
1078.
Bugental, D. B., Ellerson, P. C., Lin, E. K., Rainey, B., Kokotovic, A., & O’Hara, N. (2010). A
cognitive approach to child abuse prevention. Psychology of Violence,16(3):243-58.
Burnes, D. (2016). Community Elder Mistreatment Intervention with Capable Older Adults:
Toward a Conceptual Practice Model. The Gerontologist.
Buyck, J. F., Bonnaud, S., Boumendil, A., Andrieu, S., Bonenfant, S., Goldberg, M., ... & Ankri,
J. (2011). Informal caregiving and self-reported mental and physical health: results from
the Gazel Cohort Study. American Journal of Public Health, 101(10), 1971-1979.
California Department of Mental Health & California’s Caregiver Resource Centers (2009,
May). California’s Caregiver Resource Center System Annual Report Fiscal Year 2007-
08. Retrieved online at:
http://www.dhcs.ca.gov/services/MH/Documents/CRC_Annual_Report_2007-2008.pdf
California Department on Aging. (n.d.). National Aging Program Information Systems (NAPIS)
State Program Report (SPR) for California. Retrieved online at:
https://www.aging.ca.gov/Data_and_Statistics/#NAPIS
Cameron, J. I., & Gignac, M. A. (2008). “Timing It Right”: A conceptual framework for
addressing the support needs of family caregivers to stroke survivors from the hospital to
the home. Patient Education and Counseling, 70(3), 305-314.
108
Cameron, A. C., & Trivedi, P. K. (2010). Microeconometrics Using Stata (Vol. 2). College
Station, TX: Stata press.
Capaldi, D. M., Knoble, N. B., Shortt, J. W., & Kim, H. K. (2012). A systematic review of risk
factors for intimate partner violence. Partner Abuse, 3(2), 231.
Capistrant, B. D. (2016). Caregiving for older adults and the caregivers’ health: An
epidemiologic review. Current Epidemiology Reports 3(1), 72-80.
Casey, R. L., Berkman, M., Stover, C. S., Gill, K., Durso, S., & Marans, S. (2007). Preliminary
results of a police-advocate home-visit intervention project for victims of domestic
violence. Journal of Psychological Trauma, 6(1), 39-49.
Center for Medicare & Medicaid Services, (n.d.[a]) Medicare National HCPCS Aggregate
Summary Table. Retrieved online at: https://data.cms.gov/Medicare-Physician-
Supplier/Medicare-National-HCPCS-Aggregate-Summary-Table-CY/jtra-d83c/data
Center for Medicare & Medicaid Services, (n.d.[b]) What Part A covers. Retrieved online at:
https://www.medicare.gov/what-medicare-covers/part-a/what-part-a-covers.html
Cheng, S. T., Mak, E. P., Lau, R. W., Ng, N. S., & Lam, L. C. (2015). Voices of Alzheimer
caregivers on positive aspects of caregiving. The Gerontologist, 56(3), 451-460.
Chien, L. Y., Chu, H., Guo, J. L., Liao, Y. M., Chang, L. I., Chen, C. H., & Chou, K. R. (2011).
Caregiver support groups in patients with dementia: a meta ‐analysis. International
Journal of Geriatric Psychiatry, 26(10), 1089-1098.
Chiu, T., Marziali, E., Colantonio, A., Carswell, A., Gruneir, M., Tang, M., & Eysenbach, G.
(2009). Internet-based caregiver support for Chinese Canadians taking care of a family
member with Alzheimer disease and related dementia. Canadian Journal on Aging/La
Revue canadienne du vieillissement, 28(4), 323-336.
109
Cicirelli, V. (1986). The Helping Relationship and Family Neglect in Later Life. In Pillemer, K.
A., & Wolf, R. S. (Eds). Elder abuse: Conflict in the Family. Dover, Mass: Auburn
House Pub. Co.
Coleman, E. A., Parry, C., Chalmers, S., & Min, S. J. (2006). The care transitions intervention:
results of a randomized controlled trial. Archives of Internal medicine, 166(17), 1822-
1828.
Cooper, C., Selwood, A., Blanchard, M., Walker, Z., Blizard, R., & Livingston, G. (2010). The
determinants of family carers' abusive behaviour to people with dementia: Results of the
CARD study. Journal of Affective Disorders, 121(1), 136-142.
Coulourides Kogan, A., Wilber, K., & Mosqueda, L. (2016). Moving Toward Implementation of
Person ‐Centered Care for Older Adults in Community ‐Based Medical and Social Service
Settings: “You Only Get Things Done When Working in Concert with Clients”. Journal
of the American Geriatrics Society, 64(1), e8-e14.
Credit for Caring Act of 2017. HR 2505. 115th Cong. (2017).
Credit for Caring Act of 2017. S 1151. 115th Cong. (2017).
Crimmins, E. M., & Beltrán-Sánchez, H. (2011). Mortality and morbidity trends: is there
compression of morbidity? The Journals of Gerontology Series B: Psychological
Sciences and Social Sciences, 66(1), 75-86.
Czaja, S. J., Schulz, R., Belle, S. H., Burgio, L. D., Armstrong, N., Gitlin, L. N., ... & Stahl, S.
M. (2006). Data and safety monitoring in social behavioral intervention trials: the
REACH II experience. Clinical Trials, 3(2), 107-118.
Dalziel, K., & Segal, L. (2012). Home visiting programmes for the prevention of child
maltreatment: cost-effectiveness of 33 programmes. Archives of Disease in Childhood,
97(9), 787-798.
110
Damjanovic, A. K., Yang, Y., Glaser, R., Kiecolt-Glaser, J. K., Nguyen, H., Laskowski, B., ... &
Weng, N. P. (2007). Accelerated telomere erosion is associated with a declining immune
function of caregivers of Alzheimer’s disease patients. The Journal of Immunology,
179(6), 4249-4254.
Daro, D. A., & McCurdy, K. P. (2008). Interventions to prevent child maltreatment. In L.S.,
Doll, S.E. Bonzo, J.A. Mercy, D.A. Sleet, & E.N. Haas (Eds). Handbook of Injury and
Violence Prevention (pp. 137-155). New York, NY: Springer.
Darragh, A. R., Sommerich, C. M., Lavender, S. A., Tanner, K. J., Vogel, K., & Campo, M.
(2015). Musculoskeletal discomfort, physical demand, and caregiving activities in
informal caregivers. Journal of Applied Gerontology, 34(6), 734-760.
DeLiema, M., Homeier, D. C., Anglin, D., Li, D., & Wilber, K. H. (2016). The forensic lens:
Bringing elder neglect into focus in the emergency department. Annals of Emergency
Medicine, 68(3), 371-377.
Department of Health and Human Services. (2016, November 15). Assessment and Care
Planning for Patients with Cognitive Impairment (GPPP6). Federal Register, 81(220):
80252-80254.
DiCamillo, M., & Field, M. (2015, January). Just 36% of Voters Aware of State’s Paid Family
Leave Program. (Report for the California Center for Research on Women and Families
by Field Research Corporation.) Retrieved online at:
http://www.field.com/fieldpollonline/subscribers/Rls2494.pdf
Dishion, T. J., Mun, C. J., Drake, E. C., Tein, J. Y., Shaw, D. S., & Wilson, M. (2015). A
transactional approach to preventing early childhood neglect: The Family Check-Up as a
public health strategy. Development and Psychopathology, 27(4pt2), 1647-1660.
111
Dishion, T., Forgatch, M., Chamberlain, P., & Pelham, W. (n.d.). The Oregon Model of Behavior
Family Therapy: From Intervention Design to Promoting Large-Scale System Change.
Behavior Therapy, 47(6), 812–837.
Dong, X., & Simon, M. A. (2013). Elder abuse as a risk factor for hospitalization in older
persons. JAMA Internal Medicine, 173(10), 911-917.
Dong, X., Simon, M., & Evans, D. (2012). Decline in Physical Function and Risk of Elder Abuse
Reported to Social Services in a Community ‐Dwelling Population of Older Adults.
Journal of the American Geriatrics Society, 60(10), 1922-1928.
Duggan, A., McFarlane, E., Fuddy, L., Burrell, L., Higman, S. M., Windham, A., & Sia, C.
(2004). Randomized trial of a statewide home visiting program: Impact in preventing
child abuse and neglect. Child Abuse and Neglect, 28(6), 597-622. doi:
10.1016/j.chiabu.2003.08.007
Enguidanos, S. M., DeLiema, M., Aguilar, I., Lambrinos, J., & Wilber, K. H. (2014).
Multicultural voices: attitudes of older adults in the United States of America about elder
mistreatment. Ageing and society, 34(05), 877-903.
Family and Medical Insurance Leave Act f 2017. S 337. 115th Cong. (2017).
Family and Medical Insurance Leave Act of 2017. HR 947. 115th Cong. (2017).
Family Medical Leave Act of 1993, 29 U.S.C. Sec. 2601.
Findlay, S. (2016). Paid Family and Medical Leave. Health Affairs. Retrieved online at:
http://www.healthaffairs.org/healthpolicybriefs/brief.php?brief_id=165
Flinn, B. (2018, May). Millennials: The Emerging Generation of Family Caregivers.
Washington, DC: AARP Public Policy Institute.
112
Fogel, J., Ribisl, K., Morgan, P., Humphreys, K. & Lyons, E. J. (2008). The underrepresentation
of African Americans in online cancer support groups. Journal of the National Medical
Association, 100(6), 705.
Foldes, S. S., Moriarty, J. P., Farseth, P. H., Mittelman, M. S., & Long, K. H. (2017). Medicaid
Savings From The New York University Caregiver Intervention for Families with
Dementia. The Gerontologist.
Fox, S., Duggan, M., & Purcell, K. (2013). Family caregivers are wired for health. Pew
Research Center's Internet & American Life Project. Retrieved online at:
http://www.pewinternet.org/2013/06/20/family-caregivers-are-wired-for-health/
Fredman L., Lyons J.G., Cauley J.A., Hochberg M., Applebaum K.M. (2015). The relationship
between caregiving and mortality after accounting for time-varying caregiver status and
addressing the healthy caregiver hypothesis. The Journals of Gerontology Series A:
Biological Sciences and Medical Sciences, 70(9), 1163–1168.
Fulmer, T., Paveza, G., VandeWeerd, C., Fairchild, S., Guadagno, L., Bolton-Blatt, M., &
Norman, R. (2005). Dyadic vulnerability and risk profiling for elder neglect. The
Gerontologist, 45(4), 525-534.
Funk, L.M., Dansereau, L. & Novek, S. (2017). Carers as System Navigators: Exploring
Sources, Processes and Outcomes of Structural Burden. The Gerontologist.
Gallagher-Thompson, D., & Coon, D. W. (2007). Evidence-based psychological treatments for
distress in family caregivers of older adults. Psychology and Aging, 22(1), 37.
Gaugler, J. E., Reese, M., & Tanler, R. (2015). Care to Plan: An online tool that offers tailored
support to dementia caregivers. The Gerontologist, 56(6), 1161-1174.
Gibson, M. J., Kelly, K. A., & Kaplan, A. K. (2012). Family caregiving and transitional care: A
critical review. Family Caregiver Alliance, San Francisco, CA. Retrieved online at:
113
https://www.caregiver.org/sites/caregiver.org/files/pdfs/FamCGing_TransCare_CritRvw_
FINAL10.31.2012.pdf
Gitlin, L., Belle, S., Burgio, L., Czaja, S., Mahoney, D., Gallagher-Thompson, D., . . . REACH.
(2003). Effect of multicomponent interventions on caregiver burden and depression: The
REACH multisite initiative at 6-month follow-up. Psychology and Aging, 18(3), 361-374.
Gitlin, L. N., Hodgson, N., Jutkowitz, E., & Pizzi, L. (2010). The cost-effectiveness of a
nonpharmacologic intervention for individuals with dementia and family caregivers: the
tailored activity program. The American Journal of Geriatric Psychiatry, 18(6), 510-519.
Gitlin, L. N., & Hodgson, N. (2015). Caregivers as therapeutic agents in dementia care: The
context of caregiving and the evidence base for interventions. In R. Kane & J. Gaugler
(Eds). Family Caregiving in the New Normal, (pp. 305-353). Academic Press.
Gitlin, L.N., Marx, K., Stanley, I.H., & Hodgson, N. (2015). Translating evidence-based
dementia caregiving interventions into practice: State-of-the-science and next steps. The
Gerontologist 55(2): 210-226.
Gitlin, L. N., Hodgson, N. A., & Choi, S. S. W. (2016). Home-based interventions targeting
persons with dementia: what is the evidence and where do we go from here?. In M. Boltz
& J.E. Galvin (Eds) Dementia Care (pp. 167-188). Springer International Publishing.
Glasgow, R. E., Lichtenstein, E., & 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.
Glueckauf, R. L., Ketterson, T. U., Loomis, J. S., & Dages, P. (2004). Online support and
education for dementia caregivers: overview, utilization, and initial program evaluation.
Telemedicine Journal & E-Health, 10(2), 223-232. DOI: 10.1089/tmj.2004.10.223
114
Griffiths, P. C., Whitney, M. K., Kovaleva, M., & Hepburn, K. (2016). Development and
implementation of tele-savvy for dementia caregivers: A department of veterans affairs
clinical demonstration project. The Gerontologist, 56(1), 145-154.
Gwyther, L. P. (1998). Social issues of the Alzheimer’s patient and family. The American
journal of medicine, 104(4), 17S-21S.
Hamby, S. (2017). On defining violence, and why it matters. Psychology of Violence, 7(2), 167-
180.
Hamby, S., & Grych, J. (2013). The web of violence: Exploring connections among different
forms of interpersonal violence and abuse. Netherlands: Springer.
Health and Retirement Study, (2002 to 2014 HRS Core) [dataset on the internet]. Ann Arbor
(MI): Produced and distributed by the University of Michigan with funding from the
National Institute on Aging (grant number NIA U01AG009740). 2017 December [cited
2018 Aug 26].
Hepburn, K., Lewis, M., Tornatore, J., Sherman, C. W., & Bremer, K. L. (2007). The Savvy
Caregiver program: the demonstrated effectiveness of a transportable dementia caregiver
psychoeducation program. Journal of Gerontological Nursing, 33(3), 30-36.
Herrera, A. P. George R., Angel J. L., Markides K., & Torres-Gil, F. (2013). Variation in Older
Americans Act caregiver service use, unmet hours of care, and independence among
Hispanics, African Americans, and Whites. Home Health Care Services Quarterly, 32,
35–56.
Hoffman, G. J., Lee, J., & Mendez-Luck, C. A. (2012). Health behaviors among baby boomer
informal caregivers. The Gerontologist, 52(2), 219-230.
115
Hopwood, J., Walker, N., McDonagh, L., Rait, G., Walters, K., Iliffe, S., ... & Davies, N. (2018).
Internet-Based Interventions Aimed at Supporting Family Caregivers of People With
Dementia: Systematic Review. Journal of Medical Internet Research, 20(6), e216.
Hosmer, D., Lemeshow, Stanley, & Sturdivant, Rodney X. (2013). Applied Logistic Regression
(3rd ed., Wiley Series in Probability and Statistics). New York: Wiley.
Johnson, D. R. (1995). Alternative methods for the quantitative analysis of panel data in family
research: Pooled time-series models. Journal of Marriage and the Family, 57(4), 1065–
1077.
Johnson, R., Hofacker, J., Boyken, L., & Eisenstein, A. (2016). Sustaining Chicago’s informal
caregivers: an age- friendly approach. Journal of Urban Health, 93(4), 639-651.
Justice in Aging. (May, 2016). Voluntary Means Voluntary: Coordinating Medicaid HCBS with
Family Assistance. Retrieved online at: http://www.justiceinaging.org/wp-
content/uploads/2016/05/Voluntary-Means-Voluntary-Coordinating-Medicaid-HCBS-
with-Family-Assistance.pdf
Katz, S. (1983). Assessing self ‐maintenance: activities of daily living, mobility, and instrumental
activities of daily living. Journal of the American Geriatrics Society, 31(12), 721-727.
Kaiser Family Foundation (2017). Compare Proposals to Replace The Affordable Care Act.
Retrieved online at: https://www.kff.org/interactive/proposals-to-replace-the-affordable-
care-act/
Kally, Z., Cote, S. D., Gonzalez, J., Villarruel, M., Cherry, D. L., Howland, S., ... & Hepburn, K.
(2014). The Savvy Caregiver Program: Impact of an evidence-based intervention on the
well-being of ethnically diverse caregivers. Journal of Gerontological Social Work, 57(6-
7), 681-693.
116
Kelly, K. (2003). Link2Care: Internet-based information and support for caregivers."
Generations 27(4):
Kelly, K., Wolfe, N., Gibson, M. J., & Feinberg, L. (2013). Listening to family caregivers: The
need to include family caregiver assessment in Medicaid home-and community-based
service waiver programs. Washington, DC: AARP Public Policy Institute.
Kernisan, L. P., Sudore, R. L., & Knight, S. J. (2010). Information-seeking at a caregiving
website: a qualitative analysis. Journal of Medical Internet Research, 12(3).
Kingdon, J. (1984). Agendas, alternatives, and public policies. Boston: Little, Brown.
Kinnear, H., Connolly, S., Rosato, M., Hall, C., Mairs, A., & O'Reilly, D. (2010). Are caregiving
responsibilities associated with non-attendance at breast screening? BMC public health,
10(1), 749.
Kim, H. (2015). Understanding Internet use among dementia caregivers: results of secondary
data analysis using the US caregiver survey data. Interactive Journal of Medical
Research, 4(1).
Klein, L. C., K. Kim, D. M. Almeida, E. E. Femia, M. J. Rovine & Zarit, S.H. (2016).
Anticipating an easier day: Effects of adult day services on daily cortisol and stress. The
Gerontologist 56(2), 303-312.
Kolanowski, A. M., Fick, D., Waller, J. L., & Shea, D. (2004). Spouses of persons with
dementia: their healthcare problems, utilization, and costs. Research in Nursing &
Health, 27(5), 296-306.
Kovaleva, M., Blevins, L., Griffiths, P. C., & Hepburn, K. (2017). An Online Program for
Caregivers of Persons Living With Dementia: Lessons Learned. Journal of Applied
Gerontology.
117
Kuhn, D., Hollinger-Smith, L., Presser, J., Civian, J., & Batsch, N. (2008). Powerful tools for
caregivers online: An innovative approach to support employees. Journal of Workplace
Behavioral Health, 23(1, part 2), 51-69.
Kuijpers, K. F., van der Knaap, L. M., & Winkel, F. W. (2012). Risk of revictimization of
intimate partner violence: The role of attachment, anger and violent behavior of the
victim. Journal of Family Violence, 27(1), 33-44.
Lachs, M. S., Williams, C. S., O’Brien, S., & Pillemer, K. A. (2002). Adult protective service
use and nursing home placement. The Gerontologist, 42(6), 734-739.
Landry, S., & Keller-Allen, C. (2017, October). Innovating the Next Generation of Dementia and
Alzheimer’s Disease Care Interventions: Addressing the Needs of Persons Living with
Dementia, Caregivers, and Care Providers. Retrieved online at:
https://aspe.hhs.gov/system/files/pdf/257861/Innovating.pdf
Laumann, E., Leitsch, S., & Waite, L. J. (2008). Elder Mistreatment in the United States:
Prevalence Estimates from a Nationally Representative Study. Journal of Gerontology B:
Psychology and Social Science, 63(4), S248-S254.
Lavelle, B., Mancuso, D., Huber, A., & Felver, B. E. (2014). Expanding Eligibility for the
Family Caregiver Support Program in SFY 2012: Updated Findings. Department of
Social and Health Services, Research and Data Analysis Division. Retrieved online at:
http://leg.wa.gov/JointCommittees/ADJLEC/Documents/2014-05-
19/Expanding_Eligibility_FCS_Report.pdf
Levine, C., Halper, D., Peist, A., & Gould, D. A. (2010). Bridging troubled waters: family
caregivers, transitions, and long-term care. Health affairs, 29(1), 116-124.
Levitsky, S. R. (2014). Caring for our own: Why there is no political demand for new American
social welfare rights. Oxford; New York: Oxford University Press.
118
Lewis, M., Hepburn, K., Narayan, S., & Kirk, L. (2005). Relationship matters in dementia
caregiving. American Journal of Alzheimer's Disease and Other Dementias, 20(6), 341-
347.
Li, H. (2015). Informal caregivers’ use of the internet for caregiving information. Social Work in
Health Care 54(6): 532–546.
Lifespan Respite Care Reauthorization Act of 2017. H.R.2535. 115th Cong. (2017).
Lifespan Respite Care Reauthorization Act of 2017. S 1188. 115th Cong. (2017).
Link, G. (2016, January 21). The Administration for Community Living: Programs and
Initiatives Providing Family Caregiver Support. Retrieved online at:
http://www.asaging.org/blog/administration-community-living-programs-and-initiatives-
providing-family-caregiver-support
Liu, Y., Kong, J., Bangerter, L. R., Zarit, S. H., & Almeida, D. M. (2018). Early Parental Abuse
and Daily Assistance to Aging Parents With Disability: Associations With the Middle-
Aged Adults’ Daily Well-being. The Journals of Gerontology: Series B. 73(5), e59-e68
Livingston, G., Barber, J., Rapaport, P., Knapp, M., Griffin, M., King, D., . . . Cooper, C. (2013).
Clinical effectiveness of a manual based coping strategy programme (START,
STrAtegies for RelaTives) in promoting the mental health of carers of family members
with dementia: Pragmatic randomised controlled trial. BMJ, 347(oct25 2), f6276-f6276.
Lwi, S. J., Ford, B. Q., Casey, J. J., Miller, B. L., & Levenson, R. W. (2017). Poor caregiver
mental health predicts mortality of patients with neurodegenerative disease. Proceedings
of the National Academy of Sciences, 114(28), 7319–7324
MacLeod, J., & Nelson, G. (2000). Programs for the promotion of family wellness and the
prevention of child maltreatment: A meta-analytic review. Child Abuse & Neglect, 24(9),
1127-1149.
119
Markman, H. J., Renick, M. J., Floyd, F. J., Stanley, S. M., & Clements, M. (1993). Preventing
marital distress through communication and conflict management training: a 4-and 5-year
follow-up. Journal of Consulting and Clinical Psychology, 61(1), 70.
Mbilinyi, L. F., Neighbors, C., Walker, D., Roffman, R., Zegree, J., Edleson, J., & O'Rourke, A.
(2010). A telephone intervention for substance-using adult male perpetrators of intimate
partner violence. Research on Social Work Practice, 21(1), 43-56.
McDonald, L., & Thomas, C. (2013). Elder abuse through a life course lens. International
Psychogeriatrics, 25(08), 1235-1243.
McKechnie, V., Barker, C., & Stott, J. (2014). Effectiveness of computer-mediated interventions
for informal carers of people with dementia—a systematic review. International
Psychogeriatrics, 26(10), 1619-1637.
McWilliams, J. M., Zaslavsky, A. M., Meara, E., & Ayanian, J. Z. (2003). Impact of Medicare
coverage on basic clinical services for previously uninsured adults. JAMA, 290(6), 757-
764.
Mensie, L. C., & Steffen, A. M. (2011). Predicting in-home respite utilization by family
caregivers of older adults: Results of a community study. Home Health Care
Management & Practice, 23(2), 109-117.
Merrill, L. L., Thomse, C. J., Crouch, J. L., & May, P. (2005). Predicting adult risk of child
physical abuse from childhood exposure to violence: Can interpersonal schemata explain
the association? Journal of Social and Clinical Psychology, 24(7), 981-1002.
Metlife Mature Market Institute. (2009). Broken Trust: Elders, Families, and Finances. Retrieved
online at: https://www.metlife.com/assets/cao/mmi/publications/studies/mmi-study-
broken-trust-elders-family-finances.pdf
120
MetLife Mature Market Institute. (2011, June). The MetLife Study of Caregiving Costs to
Working Caregivers: Double Jeopardy for Baby Boomers Caring for Their Parents.
Retrieved online at: https://www.metlife.com/mmi/research/caregiving-cost-working-
caregivers.html#key%20findings
Meyer, K., Rath, L., Kaiser, N., Gassoumis, Z.D., & Wilber, K. (2018). Prioritizing family
caregivers across policy systems: Findings from the California Task Force on Family
Caregiving. Journal of Aging and Social Policy.
Meyer, K., Wilber, K., & Gassoumis, Z.D. (In review). Are caregivers more vulnerable to
economic downturn than non-caregivers?: The impact of the Great Recession. Journal of
Aging & Social Policy.
Miller, E. A., Nadash, P., Gusmano, M. K., Simpson, E., & Ronneberg, C. R. (2018). The state
of aging policy and politics in the trump era. Journal of Aging & Social Policy.
Monin, J. K., Poulin, M. J., Brown, S. L., & Langa, K. M. (2017). Spouses’ daily feelings of
appreciation and self-reported well-being. Health Psychology, 36(12), 1135.
Monin, J. K., & Schulz, R. (2009). Interpersonal effects of suffering in older adult caregiving
relationships. Psychology and Aging, 24(3), 681-695.
Monin, J. K., Schulz, R., Martire, L. M., Jennings, J. R., Lingler, J. H., & Greenberg, M. S.
(2010). Spouses’ cardiovascular reactivity to their partners’ suffering. Journals of
Gerontology Series B: Psychological Sciences and Social Sciences, 65(2), 195-201.
Moon, M. (2017). The unprepared caregiver. The Gerontologist, 57(1), 26-31.
Morris, Z. S., Wooding, S., & Grant, J. (2011). The answer is 17 years, what is the question:
understanding time lags in translational research. Journal of the Royal Society of
Medicine, 104(12), 510-520.
121
Mosqueda, L., Burnight, K., Gironda, M. W., Moore, A. A., Robinson, J., & Olsen, B. (2016).
The abuse intervention model: a pragmatic approach to intervention for elder
mistreatment. Journal of the American Geriatrics Society, 64(9), 1879-1883.
Musich, S., Wang, S. S., Kraemer, S., Hawkins, K., & Wicker, E. (2017). Caregivers for older
adults: Prevalence, characteristics, and health care utilization and expenditures. Geriatric
Nursing, 38(1), 9-16.
National Alliance for Caregiving and AARP Public Policy Institute. (2015, June). Caregivers of
Older Adults: A Focused Look at Those Caring for Someone Age 50+ Retrieved online
at: https://www.aarp.org/content/dam/aarp/ppi/2015/caregivers-of-older-adults-focused-
look.pdf
National Alzheimer’s Project Act, 42 U.S. Code § 11225 (2011)
National Association of Social Workers. (n.d.) Evidence-Based Practice. Retrieved online at:
https://www.socialworkers.org/News/Research-Data/Social-Work-Policy-
Research/Evidence-Based-Practice
National Conference of State Legislatures. (2010). NCSL fiscal brief: State balanced budget
provisions. Washington, DC: NCSL.
National Institutes of Health. (n.d.) Funding, Retrieved online at:
https://grants.nih.gov/funding/index.htm
Navaie, M. (2011). Accessibility of Caregiver Education and Support Programs: Reaching Hard-
to-Reach Caregivers. In Education and Support Programs for Caregivers (pp. 13-28).
Springer New York.
Neighbors, C., Walker, D. D., Mbilinyi, L. F., O'Rourke, A., Edleson, J. L., Zegree, J., &
Roffman, R. A. (2010). Normative misperceptions of abuse among perpetrators of
intimate partner violence. Violence Against Women, 16(4), 370-386.
122
Nichols, L. O., Chang, C., Lummus, A., Burns, R., Martindale ‐Adams, J., Graney, M. J., ... &
Resources for Enhancing Alzheimer's Caregivers Health II Investigators. (2008). The
cost ‐effectiveness of a behavior intervention with caregivers of patients with Alzheimer's
disease. Journal of the American Geriatrics Society, 56(3): 413-420.
Older Americans Act Reauthorization of 2000. Public Law 106–501.
Olmstead v. L.C. 527 US 581 (1999).
Ortman, J. M., Velkoff, V. A., & Hogan, H. (2014, May). An aging nation: the older population
in the United States. Washington, DC: US Census Bureau.
Pearlin, Leonard I., Mullan, Joseph T., Semple, Shirley J., & Skaff, Marilyn M. (1990).
Caregiving and the stress process: An overview of concepts and their measures. The
Gerontologist, 30(5), 583-594.
Perkins, N. H., Spira, M., & Key, J. E. (2018). Intergenerational Transmission of Physical and
Emotional Sibling Violence: A Potential Connection to Elder Abuse. Families in Society,
99(3), 256-268.
Peterson, K., Hahn, H., Lee, A. J., Madison, C. A., & Atri, A. (2016). In the Information Age, do
dementia caregivers get the information they need? Semi-structured interviews to
determine informal caregivers’ education needs, barriers, and preferences. BMC
Geriatrics, 16(1), 164.
Pickering, C., Moon, A., Pieters, H., Mentes, J., & Phillips, L. (2015). Relationship management
strategies for daughters in conflicted relationships with their ageing mothers. Journal of
Advanced Nursing, 71(3), 609-619.
Pillemer, K., & Finkelhor, D. (1989). Causes of elder abuse: Caregiver stress versus problem
relatives. American Journal of Orthopsychiatry, 59(2), 179.
123
Pillemer, K., Burnes, D., Riffin, C., & Lachs, M. S. (2016). Elder Abuse: Global Situation, Risk
Factors, and Prevention Strategies. The Gerontologist, 56(Suppl 2), S194-S205.
Pinquart, M., & Sörensen, S. (2003). Differences between caregivers and noncaregivers in
psychological health and physical health: a meta-analysis. Psychology and Aging, 18(2),
250.
Pinquart, M., & Sörensen, S. (2006). Helping caregivers of persons with dementia: Which
interventions work and how large are their effects? International Psychogeriatrics, 18(4),
577-595.
Pinquart, M., & Sö rensen, S. (2007). Correlates of physical health of informal caregivers: A
meta-analysis. Journal of Gerontology: Psychological and Social Sciences, 62(2), 126–
137
Pinquart, M., & Sörensen, S. (2011). Spouses, adult children, and children-in-law as caregivers
of older adults: a meta-analytic comparison. Psychology and Aging, 26(1), 1-14.
Prochaska, J., Redding, C. and Evers, K. (2008). The Transtheoretical Model and Stages of
Change. In K. Glanz, B. K. Rimer, & K. Viswanath (Eds.), Health Behaviour and Health
Education: Theory, Research and Practice (pp. 97-121). San Francisco: Jossey-Bass.
Pruchno, R. A., Brill, J. E., Shands, Y., Gordon, J. R., Genderson, M. W., Rose, M., &
Cartwright, F. (2008). Convenience samples and caregiving research: How generalizable
are the findings?. The Gerontologist, 48(6), 820-827.
Rainville, C., Skufca, L., & Mehegan, L. (2016, November). Family Caregiving and Out-of-
Pocket Costs: 2016 Report. (Report for the AARP) Retrieved online at:
http://www.aarp.org/content/dam/aarp/research/surveys_statistics/ltc/2016/family-
caregiving-cost-survey-res-ltc.pdf
124
Ramchand, R., Tanielian, T., Fisher, M. P., Vaughan, C. A., Trail, T. E., Epley, C., ... & Ghosh-
Dastidar, B. (2014). Hidden heroes: America's military caregivers. Rand Corporation.
Santa Monica, CA.
RAND HRS (2002 to 2014 Fat Files) [dataset on the internet]. Santa Monica (CA): U.S:
Produced by the RAND Center for the Study of Aging, with funding from the National
Institute on Aging and the Social Security Administration; 2018 February [cited 2018
Aug 26].
Recognize, Assist, Include, Support, and Engage Family Caregivers Act of 2017. HR 3759. 115
th

Cong. (2017).
Recognize, Assist, Include, Support, and Engage Family Caregivers Act of 2017. S 1028. 115
th

Cong. (2017).
Reinhard, S.C., Given, B & Petlick, N.H. (2008). Supporting Family Caregivers in Providing
Care. In Hughes R.G. (Ed). Patient Safety and Quality: An Evidence-Based Handbook for
Nurses. Rockville, MD: Agency for Healthcare Research and Quality (US).
Reinhard, S. C., Levine, C., & Samis, S. (2012). Home alone: Family caregivers providing
complex chronic care. Washington, DC: AARP Public Policy Institute.
Reinhard, S. C., Feinberg, L. F., Choula, R., & Houser, A. (2015, July). Valuing the invaluable:
2015 update. Insight on the Issues, 104. Retrieved online at:
http://www.aarp.org/content/dam/aarp/ppi/2015/valuing-the-invaluable-2015-update-
new.pdf
Rocco, P. (2017). Informal caregiving and the Politics of Policy Drift in the United States.
Journal of Aging & Social Policy, 29(5), 413-432.
125
Samia, L. W., Hepburn, K., & Nichols, L. (2012). “Flying by the seat of our pants”: What
dementia family caregivers want in an advanced caregiver training program. Research in
Nursing & Health, 35(6), 598-609.
Savundranayagam, M. Y., Montgomery, R. J., Kosloski, K., & Little, T. D. (2011). Impact of a
psychoeducational program on three types of caregiver burden among
spouses. International Journal of Geriatric psychiatry, 26(4), 388-396.
Scharlach, A. E., Kellam, R., Ong, N., Baskin, A., Goldstein, C., & Fox, P. J. (2006). Cultural
attitudes and caregiver service use: Lessons from focus groups with racially and
ethnically diverse family caregivers. Journal of Gerontological Social Work, 47(1-2),
133-156.
Senate Select Committee on Aging and Long Term Care. (2015, January). A Shattered System:
Reforming Long-Term Care in California. Retrieved online at:
http://www.cicaihss.org/sites/default/files/aginglong_termcarereport.pdf
Shedler, J. (2010). The efficacy of psychodynamic psychotherapy. American Psychologist, 65(2),
98.
Schulz, R., & Beach, S. R. (1999). Caregiving as a risk factor for mortality: the Caregiver Health
Effects Study. JAMA, 282(23), 2215-2219.
Schulz, R. (2013). Research priorities in geriatric palliative care: Informal caregiving. Journal of
Palliative Medicine, 16(9), 1008-1012.
Schulz, R., Aranda, M., Beane, S., Czaja, S., Duke, B., Feder, J., ... & Hinton, L. (2016).
Families caring for an aging America. Washington, DC: Committee on Family
Caregiving for Older Adults, National Academies of Sciences, Engineering, and
Medicine.
126
Schulz, R., Burgio, L., Burns, R., Eisdorfer, C., Gallagher-Thompson, D., Gitlin, L.N.,
Mahoney, D.F. (2003). Resources for Enhancing Alzheimer’s Caregiver Health
(REACH): Overview, Site-Specific Outcomes, and Future Direction. The Gerontologist,
43(4), 514-520.
Shaw, W. S., Patterson, T. L., Semple, S. J., Ho, S., Irwin, M. R., Hauger, R. L., & Grant, I.
(1997). Longitudinal analysis of multiple indicators of health decline among spousal
caregivers. Annals of Behavioral Medicine, 19(2), 101-9.
Shugrue, N., Kellett, K., Gruman, C., Tomisek, A., Straker, J., Kunkel, S., & Robison, J. (2017).
Progress and Policy Opportunities in Family Caregiver Assessment: Results From a
National Survey. Journal of Applied Gerontology.
Smith, J. D., Knoble, N. B., Zerr, A. A., Dishion, T. J., & Stormshak, E. A. (2014). Family
check-up effects across diverse ethnic groups: Reducing early-adolescence antisocial
behavior by reducing family conflict. Journal of Clinical Child & Adolescent
Psychology, 43(3), 400-414.
Smith, S. A., & Bell, P. A. (2005). Examining the effectiveness of the Savvy Caregiver Program
among rural Colorado residents. Rural and Remote Health, 5 (3), 466
Son, J., Erno, A., Shea, D. G., Femia, E. E., Zarit, S. H., & Parris Stephens, M. A. (2007). The
caregiver stress process and health outcomes. Journal of Aging and Health, 19(6), 871-
887.
Spieker, S. J., Oxford, M. L., Kelly, J. F., Nelson, E. M., & Fleming, C. B. (2012). Promoting
first relationships: Randomized trial of a relationship-based intervention for toddlers in
child welfare. Child Maltreatment, 17(4), 271-286.
Starr, P. (2011). Remedy and reaction: The peculiar American struggle over health care reform.
New Haven: Yale University Press.
127
Stevens, J., Scribano, P. V., Marshall, J., Nadkarni, R., Hayes, J., & Kelleher, K. J. (2015). A
trial of telephone support services to prevent further intimate partner violence. Violence
against women, 1(12), 1528-1547.
Stover, C. S., Berkman, M., Desai, R., & Marans, S. (2010). The efficacy of a police-advocacy
intervention for victims of domestic violence: 12 month follow-up data. Violence Against
Women, 16(4), 410-425.
Strong Families Act. HR 3595. 115
th
Cong. (2017).
Strong Families Act. S 344. 115
th
Cong. (2017).
Suehs, B. T., Shah, S. N., Davis, C. D., Alvir, J., Faison, W. E., Patel, N. C., ... & Bobula, J.
(2014). Household members of persons with Alzheimer's disease: health conditions,
healthcare resource use, and healthcare costs. Journal of the American Geriatrics Society,
62(3), 435-441.
Teresi, J. A., Burnes, D., Skowron, E. A., Dutton, M. A., Mosqueda, L., Lachs, M. S., &
Pillemer, K. (2016). State of the science on prevention of elder abuse and lessons learned
from child abuse and domestic violence prevention: Toward a conceptual framework for
research. Journal of Elder Abuse & Neglect, 28(4-5), 263-300.
Thornberry, T. P., Matsuda, M., Greenman, S. J., Augustyn, M. B., Henry, K. L., Smith, C. A., &
Ireland, T. O. (2014). Adolescent risk factors for child maltreatment. Child Abuse &
Neglect, 38(4), 706-722.
Toseland, R. W., & Smith, T. L. (2006). The impact of a caregiver health education program on
health care costs. Research on Social Work Practice, 16(1), 9-19.
Toth, S. L., Gravener-Davis, J. A., Guild, D. J., & Cicchetti, D. (2013). Relational interventions
for child maltreatment: Past, present, and future perspectives. Development and
Psychopathology, 25(4, part 2), 1601-1617.
128
Tucker, M. C., & Rodriguez, C. M. (2014). Family dysfunction and social isolation as
moderators between stress and child physical abuse risk. Journal of Family Violence,
29(2), 175-186.
van der Roest, H. G., Meiland, F. J., Jonker, C., & Dröes, R. M. (2010). User evaluation of the
DEMentia-specific Digital Interactive Social Chart (DEM-DISC). A pilot study among
informal carers on its impact, user friendliness and, usefulness. Aging & Mental Health,
14(4), 461-470.
Vandeweerd, C., Paveza, G. J., Walsh, M., & Corvin, J. (2013). Physical mistreatment in persons
with Alzheimer's Disease. Journal of Aging Research.
VanVoorhis, C. W., & Morgan, B. L. (2007). Understanding power and rules of thumb for
determining sample sizes. Tutorials in Quantitative Methods for Psychology, 3(2), 43-50.
Wakabayashi, C., & Donato, K. M. (2006). Does caregiving increase poverty among women in
later life? Evidence from the Health and Retirement Survey. Journal of Health and Social
Behavior, 47(3), 258-274.
Wald, C., Fahy, M., Walker, Z., & Livingston, G. (2003). What to tell dementia caregivers—the
rule of threes. International Journal of Geriatric Psychiatry, 18(4), 313-317.
Washington, K. T., Meadows, S. E., Elliott, S. G., & Koopman, R. J. (2011). Information needs
of informal caregivers of older adults with chronic health conditions. Patient Education
and Counseling, 83(1), 37-44.
Weinberger, M., Gold, D. T., Divine, G. W., Cowper, P. A., Hodgson, L. G., Schreiner, P. J., &
George, L. K. (1993). Social service interventions for caregivers of patients with
dementia: impact on health care utilization and expenditures. Journal of the American
Geriatrics Society, 41(2), 153-156.
129
Whitaker, D. J., Hall, D. M., & Coker, A. L. (2009). Primary prevention of intimate partner
violence: Toward a developmental, social-ecological model. Intimate Partner Violence:
A Health-Based Perspective. In C. Mitchell & A. Deirdre (Eds.), (pp. 289-306). New
York: Oxford University Press.
Whitaker, D. J., Murphy, C. M., Eckhardt, C. I., Hodges, A. E., & Cowart, M. (2013).
Effectiveness of primary prevention efforts for intimate partner violence. Partner Abuse,
4(2), 175-195.
Widom, C. S., Czaja, S. J., & Dutton, M. A. (2008). Childhood victimization and lifetime
revictimization. Child Abuse & Neglect, 32(8), 785-796.
Wiglesworth, A., Mosqueda, L., Mulnard, R., Liao, S. Gibbs, L. (2010). Screening for Abuse and
Neglect of People with Dementia. Journal of the American Geriatrics Society, 58(3),
493-500.
Wilber, K. H., & Reynolds, S. L. (1997). Introducing a framework for defining financial abuse of
the elderly. Journal of Elder Abuse & Neglect, 8(2), 61-80.
Wolff, J. L., & Kasper, J. D. (2006). Caregivers of frail elders: Updating a national profile. The
Gerontologist, 46(3), 344-356.
Wolff, J. L., Spillman, B. C., Freedman, V. A., & Kasper, J. D. (2016). A National Profile of
Family and Unpaid Caregivers Who Assist Older Adults With Health Care Activities.
JAMA Internal Medicine, 176(3), 372-379.
Wu, B., Luo, H., Flint, E., & Qin, L. (2015). Dental care utilization among caregivers who care
for older adults. Research on Aging, 37(4), 388-412.
Yon, Y., Mikton, C., Gassoumis, Z., Wilber, K. (2017). Elder abuse in community settings: a
systemic review and meta-analysis. Lancet Global Health, 5 e147-e156.
130
Yonashiro Cho, J., Meyer, K., Wilber, K. (2017). The Aging Network. In XinQi Dong (Ed),
Elder Abuse: Research, Practice, and Policy. United States: Springer.
Zarit, S. & Edwards, A. (2008). Family caregiving: research and clinical intervention. In Bob
Wood & Linda Clare (Eds.), Handbook of the Clinical Psychology of Ageing. (2nd ed.).
Hoboken, NJ: John Wiley & Sons.
Zarit, S. H., Reever, K. E., & Bach-Peterson, J. (1980). Relatives of the impaired elderly:
correlates of feelings of burden. The Gerontologist, 20(6), 649-655.

131
Tables

Table 2.1: Comparing client characteristics at each service site (N=540)

Variable
Usual services
N(%)
(n=466)
CareJourney
N(%)
(n=74)
Missing
N(%)
X
2
/t-test p-value
Predisposing
Age

44(8.2) 5.46 0.07
Less than 50 72(17.0) 17(27.9)
50 to 65
167(39.4) 25(41.0)
Older than 65
185(43.6) 19(31.2)
Female
340(73.43) 61(83.6) 4(0.7) 3.43 0.06
Race/ethnicity

0(0.0) 2.39 0.49
Caucasian 194(41.6) 32(43.2)
African American 92(19.7) 19(25.7)

Asian
90(19.3) 10(13.5)

Latino
90(19.3) 13(17.6)

Married
302(65.4) 40(56.3) 7(1.3) 2.18 0.14
Enabling

Employment 27(5.0) 10.17 0.02
Full time 132(29.9) 19(26.4)
Part time 55(12.5) 19(26.4)
Retired 179(40.6) 22(30.6)
Unemployed 75(17.0) 12(16.67)
Education

120(22.2) 27.17 <0.001
Some high school
56(15.6) 2(3.2)
Completed high school
42(11.7) 3(4.8)
Some college
77(21.5) 14(22.6)
College 146(40.8) 23(37.1)

Post graduate
37(10.3) 20(32.3)

How client learned about
service
138(25.6) 42.80 <0.001
Healthcare provider 67(19.6) 19(31.7)
Social service provider 213(62.3) 12(20.0)
Online 28(8.2) 17(28.3)
Other 34(9.9) 12(20.0)
Service site 1 359(77.0) 32(43.2) 0(0.0) 36.51 <0.001
Need

Care recipient has a
cognitive impairment
340(74.2) 59(79.7) 8(1.5) 1.03 0.31
Number of ADL tasks
a
3.7(2.2) 3.6(2.3) 14(2.6) 0.28 0.78
Completes medical tasks 149(34.3) 13(22.8) 48(8.9) 2.99 0.08
Burden 262(65.2) 38(64.4) 79(14.6) 0.01 0.91
Hours caregiving/week 44(8.2) 4.21 0.12
Less than 20 87(20.1) 20(31.3)
20 to 40 92(21.3) 13(20.3)
Greater than 40 253(58.6) 31(48.4)
a
Descriptive statistics presented for ADLs are M(SD).
132
Table 2.2: Characteristics of caregivers using online- and usual-delivered service modes (N=540)
Model 1 Model 2 Model 3 Model 4

Variable OR CI OR CI OR p-value OR CI
Predisposing
Age
Less than 50 (omit)
50 to 64 0.61 0.31-1.21 0.67 0.31-1.42 0.59 0.26-1.34 0.61
0.26-1.43
Older than 65 0.40 0.19-0.84 0.41 0.15-1.14 0.34 0.11-1.01 0.34 0.11-1.08
Female 1.78 0.92-3.44 1.51 0.76-3.01 1.50 0.72-3.13 1.47 0.70-3.09
Race/ethnicity

Caucasian (omit)
African
American
1.13 0.59-2.15 1.39 0.69-2.78 0.99 0.45-2.17 0.98 0.44-2.19
Asian 0.58 0.27-1.27 0.66 0.29-1.47 0.56 0.23-1.35 0.64 0.26-1.57
Latino 0.69 0.33-1.42 0.89 0.41-1.90 0.63 0.27-1.47 0.66 0.28-1.57
Married 0.87 0.50-1.51 0.84 0.47-1.53 0.95 0.50-1.81 0.95 0.49-1.82
Enabling
Employment
Full time (omit)
Part time 3.13 1.44-6.81 3.51 1.48-8.29 3.82 1.58-9.22
Retired 1.60 0.64-4.05 1.67 0.61-4.60 1.88 0.65-5.39
Unemployed 1.34 0.59-3.05 1.13 0.46-2.79 1.19 0.46-3.09
Education

Some high
school
(omit)
Completed
high school

1.64 0.27-9.79 1.21 0.17-8.45 1.26 0.17-9.18
Some college 3.74 0.85-16.52 2.57 0.55-12.15 2.70 0.55-13.30
College 3.69 0.88-15.50 2.39 0.53-10.82 2.50 0.51-12.29
Post graduate

11.79 2.71-51.24 6.88 1.45-32.58 7.14 1.42-36.04
How did client learn about service
Social service
provider
(omit)
Healthcare
provider
2.91 1.30-6.55 2.91 1.28-6.62
Online 5.17 2.10-12.74 5.42 2.15-13.66
Other 3.65 1.45-9.20 3.53 1.38-9.04
Service site 1 0.31 0.17-0.57 0.31 0.17-0.58
Need

Cognitive
impairment

1.49 0.72-3.08
Number of ADL
tasks
1.03 0.89-1.19
Completes
medical tasks
0.78 0.35-1.75
Burden 0.94 0.49-1.81
Hours caregiving/week
Less than 20 (omit)
20 to 40 0.65 0.26-1.63
Greater than
40
0.83 0.37-1.88
Constant 0.21 0.09-0.51 0.04 0.01-0.19 0.06 0.01-0.43 0.05 0.01-0.42
*
p<0.05; **p<0.01; ***p<0.001
133

Table 3.1: Theoretical Basis for Shared Risk Factors
Theory Premise
References from
Child
Maltreatment
Literature
References
from IPV
Literature
References from
Elder
Mistreatment
Literature
Social Learning
Theory
Abusive behaviors are learned by
observation. Normalization of
these behaviors leads to their
replication.
No child
maltreatment
research directly
addressing this
theory.
Capaldi et al.,
2012; Whitaker
et al., 2013
McDonald &
Thomas, 2013
Caregiver Stress
High levels of stress that
overwhelm a family caregiver
lead the caregiver to treat
recipients abusively.
Tucker &
Rodriguez, 2014
Not applicable
in the absence
of a caregiving
relationship.
Cooper et al.,
2010;
Reactive Abuse
Caregivers who experience
violent behaviors from care
recipients react in harmful or
abusive ways.
Thornberry et al.,
2014
Kuijpers et al.,
2012
Cooper et al.,
2010;
Wiglesworth et
al., 2010
Attachment
Theory
Positive caregiver interactions
with a child promote emotional
and behavioral regulation while
increasing the parent’s
investment in the child’s
wellbeing. Poor attachment
increases risk of abuse in the
dyad and undermines child
development.
Daro & McCurdy,
2008; Dishion et
al., 2015;
Thornberry et al.,
2014; Toth et al.,
2013
Kuijpers et al.,
2012
Cicirelli, 1986;
Perkins, et al.,
2018
Cumulative Risk
Predisposing risk factors for
abuse accumulate over time,
rendering severely heightened
risk for perpetuating violence.
Smith et al., 2014;
Thornberry et al.,
2014
Anda et al.,
2006
No elder
mistreatment
research directly
addressing this
theory.

134

Table 3.2: Prevention Interventions from Child Maltreatment and IPV
Intervention Name Intervention Program Relevant References
Healthy Start
Within 1 week after birth, a trained paraprofessional
visits the home of an at-risk family to provide direct
services (e.g., education on child development,
positive appraisal), as well as links to relevant
supports and services in the community. Visits occur
until the child is aged 3 to 5 years old. Frequency of
visits varies by risk for family violence.
Bair-Merritt et al., 2010;
Bugental et al., 2010
Promoting First
Relationships
A weekly 10-session intervention to improve parent-
child relationships by helping caregivers understand
the child’s needs and behaviors. Video feedback of
parent-child interactions is used to help parents
better interpret child behaviors. Activities are guided
by trained interventionists and include feedback on
strengths and weaknesses.
Described in Toth et al.,
2013; Spieker et al., 2012
Parent-Child
Interaction Therapy
Originally aimed at parent(s) with children aged 2 to
7 with behavioral problems, this intervention is
provided over the course of 12 to 14 weekly sessions
and aims at improving relationship quality and
disrupting coercive tendencies between parents and
children. During the first 7 sessions parents follow
the child’s lead during play in order to strengthen
the relationship. In the final 7 sessions, the parents
are taught skills to direct the child’s behavior,
thereby better managing it.
Described in Toth et al.,
2013
Family Check-Up
Comprised of 3 sessions, the Family Check-Up begins
with an interview with the parent(s) to learn about
the family situation and engage in motivational
interviewing techniques. This is followed by an
assessment, which includes a questionnaire and
observation of the parent and child interacting. At
the third session, feedback is provided on the
family’s strengths and weaknesses, as well as a
discussion of accessing additional supportive
services.
Dishion et al., 2015,
Smith et al., 2014
135
Men’s Domestic Abuse
Check-Up
Men respond to program recruitment invitations
and are asked to complete a survey about their
interpersonal relationships and engagement in
aggressive or violent behaviors towards their
partners. They are provided with a personalized
survey summary which presents their responses
alongside answers from the general population,
which often serves to highlight participants’
deviation from societal norms. Participants then
engage in individual telephone counseling with
study personnel trained in motivational interviewing
techniques to promote change.
Mbilinyi et al., 2010
Prevention
Relationship
Enhancement Program
Couples are taught skills (e.g., active listening) to
better manage negative affect and improve
communication. Skills are taught during 5 sessions,
scheduled 1 week apart, with 3 to 5 other couples,
lasting for 3 hours each. Sessions are devoted to 1 or
2 topics and led by trained psychology
undergraduate and graduate students. Outside of
sessions, couples are assigned reading and asked to
complete exercises.
Markman et al., 1993

136

Table 4.1: Sample characteristics (percent, pooled across waves)

Non-caregivers Caregivers
Average number of waves
a
4.02(1.99) 4.86(1.65)
Female
b
59.76 43.69
Age
a
67.66(11.35) 71.85(11.07)
Race
Caucasian/White 77.00 80.12
African American/Black 16.57 13.06
Other race 6.43 6.82
Hispanic ethnicity 11.01 13.07
Education
Less than high school 29.92 20.45
High school 35.27 34.71
Some college 19.86 23.12
College or more 14.94 21.72
Health insurance 82.01 88.67
Physician visits
a
9.46(14.65) 18.04(29.01)
Hospitalizations 25.36 49.70
a
Mean/standard of deviation
b
Given conventional wisdom that women provide the majority of care,
we explored the unexpectedly high proportion of male caregivers. We
suspected was caused by the higher age of the sample compared to
other national studies. This was confirmed when the sample was
stratified by age, where the proportion of women caregivers was higher
at younger ages. Removal of those providing care prior to the first time
data was collected exacerbated this difference

137

Table 4.2: Sample Characteristics by wave (frequency/percent)
Year/wave 2002 2004 2006 2008 2010 2012 2014
Number of
participants
17032 19047 17577 16480 20992 19631 17955
Caregivers 0(0.00) 530(2.78) 775(4.41) 767(4.65) 885(4.22) 1036(5.28) 1073(5.98)
Number of years caregiving
0 years 17032(100.00) 18517(97.22) 16354(94.32) 14395(92.36) 18383(93.18) 16582(91.20) 14491(89.54)
2 years or less N/A 530(2.78) 772(4.45) 772(4.95) 828(4.20) 962(5.29) 950(5.87)
4 years or less N/A N/A 212(1.22) 321(2.06) 306(1.55) 342(1.88) 366(2.26)
6 years or less N/A N/A N/A 98(0.63) 159(0.81) 160(0.88) 196(1.21)
More than 6 years N/A N/A N/A N/A 52(0.26) 136(0.75) 181(1.12)
Female 10192(59.84) 11247(59.05) 10438(59.38) 9816(59.56) 12275(58.47) 11522(58.69) 10607(59.10)
Age 68.68(10.49) 66.96(11.52) 68.43(11.11) 69.71(10.75) 65.88(11.98) 67.30(11.63) 68.49(11.30)
Race
Caucasian/White 14040(82.45) 15348(80.60) 14241(81.03) 13295(80.68) 15199(72.60) 14167(72.38) 12844( 71.74)
African
American/Black
2345(13.77) 2710(14.23) 2456(13.97) 2335(14.17) 4020(19.20) 3753(19.17) 3503(19.57)
Other race 643(3.78) 984(5.17) 878(5.00) 849(5.15) 1717(8.20) 1654(8.45) 1556(8.69)
Hispanic ethnicity 1384(8.13) 1811(9.51) 1,616(9.19) 1573(9.55) 2764(13.18) 2639(13.46) 2473(13.79)
Education
Less than high
school
4283(25.15) 4306(22.61) 3833(21.81) 3474(21.09) 4033(19.22) 3641(18.55) 3230(17.99)
High school 6167(36.22) 6700(35.18) 6221(35.40) 5826(35.36 ) 7095(33.81) 6652(33.89) 6032(33.60)
Some college 3389(19.90) 4134(21.71) 3828(21.78) 3637(22.07) 5178(24.67) 4881(24.87) 4535(25.26)
College or more 3188(18.72) 3903(20.50) 3691(21.00) 3539(21.48) 4682(22.31) 4453(22.69) 4153(23.14)
Average number of
physician visits
a

9.98(14.86) 9.68(14.57) 10.11(15.53) 10.28(15.66) 10.18(17.06) 9.15(15.20) 9.10(15.37)
Hospitalizations
(15470)55.22 (13791 )49.22 (15141)54.04 ( 16121 )57.54 (12955)46.24 (13319)47.54 (14486)51.70
a
Mean/Standard of deviation

138

Table 4.3: Number of years caregiving and rates of physician visits
Hybrid model
Number of observation=106,096
Number of groups (participants)=23,665
Average observations per group (participant)=4.5

LR Chi-squared= 1310.28 p-value < 0.0001

Estimate Standard Error Z p-value CI
Years of caregiving 0.20 0.02 8.67 0.00 0.16 0.25
Years of caregiving
2
-0.03 0.00 -6.61 0.00 -0.03 -0.02
Age 0.02 0.00 9.53 0.00 0.01 0.02
Female 0.22 0.02 12.60 0.00 0.19 0.26
Race
Caucasian (omitted)
African American 0.08 0.03 2.86 0.00 0.02 0.13
Other -0.04 0.04 -0.94 0.35 -0.12 0.04
Hispanic ethnicity -0.13 0.03 -4.56 0.00 -0.19 -0.08
Education
Less than high school (omitted)
High school/GED 0.00 0.02 -0.01 0.99 -0.05 0.05
Some college 0.07 0.03 2.51 0.01 0.01 0.12
College and above 0.02 0.03 0.86 0.39 -0.03 0.08
Insurance 0.10 0.03 3.70 0.00 0.05 0.15
Constant 0.89 0.06 13.83 0.00 0.76 1.01

Bootstrap model
Number of observation= 116,528
Number of groups (participants)=24,644
Average observations per group (participant)=4.7

LR Chi-squared= 429.91

p-value < 0.0001
Loglikelihood= -420052.85

Estimate Standard Error Z p-value CI
Years of caregiving 0.15 0.02 6.60 0.00 0.10 0.19
Years of caregiving
2
-0.02 0.00 -6.33 0.00 -0.03 -0.01
Age 0.02 0.00 16.85 0.00 0.02 0.03
Insurance 0.09 0.02 4.96 0.00 0.05 0.13

139
Table 4.4: Number of years caregiving and rates of hospitalizations
Hybrid model
Number of observation=111,239

Number of groups (participants)= 23,780

Average observations per group (participant)=4.6

LR Chi-squared=2517.62 p-value<0.0001

Estimate Standard Error Z p-value CI
Years of caregiving 1.38 0.05 9.71 0.00 1.29 1.47
Years of caregiving
2
0.96 0.01 -7.68 0.00 0.95 0.97
Age 1.04 0.00 18.46 0.00 1.04 1.05
Female 1.03 0.02 1.43 0.15 0.99 1.08
Race

Caucasian (omitted)
African American 1.12 0.04 3.38 0.00 1.05 1.20
Other 1.00 0.06 -0.08 0.94 0.89 1.12
Hispanic ethnicity 0.81 0.04 -4.44 0.00 0.74 0.89
Education

Less than high school (omitted)
High school/GED 0.90 0.03 -3.41 0.00 0.84 0.95
Some college 0.86 0.03 -4.33 0.00 0.80 0.92
College and above 0.62 0.02
-
12.58
0.00 0.58 0.67
Insurance 1.08 0.04 1.95 0.05 1.00 1.17
Constant 0.02 0.00
-
40.35
0.00 0.02 0.03

Bootstrap model
Number of observation=75,283
Number of groups (participants)=14,037
Average observations per group
(participant)=5.4

LR Chi-squared= 1434.71 p<0.0001
Loglikelihood=-28774.992
Estimate Standard Error Z p-value CI
Years of caregiving 1.39 0.04 11.15 0.00 1.31 1.47
Years of caregiving
2
0.96 0.00 -8.48 0.00 0.95 0.97
Age 1.07 0.00 27.68 0.00 1.07 1.08
Insurance 1.04 0.05 1.01 0.31 0.96 1.14

140
Table 4.5: Logistic regression where number of years caregiving is a factor variable
Number of observation=75,283

Number of groups (participants)=14,037

Average observations per group (participant)=5.4

LR Chi-squared=1253.79 p-value < 0.0001

Loglikelihood=-28762.816

Odds ratio Standard Error Z p-value CI
Years of caregiving

2 years or less
1.90 0.10 12.17 0.00 1.71 2.10
4 years or less
1.53 0.12 5.44 0.00 1.31 1.78
6 years or less
1.28 0.15 2.20 0.03 1.03 1.61
More than 6 years
1.08 0.16 0.55 0.58 0.81 1.45
Age
1.07 0.00 29.96 0.00 1.07 1.08
Insurance
1.05 0.04 1.21 0.23 0.97 1.12

141
Table 4.6: Hybrid models using cumulative years of caregiving variable
Physician visits
Number of observation=103,064

Number of groups (participants)=23,620

Average observations per group (participant)=4.4

LR Chi-squared=1255.09 p-value < 0.0001

Estimate Standard Error Z p-value CI
Years of caregiving 0.20 0.02 8.60 0.00 0.15 0.25
Years of caregiving
2
-0.02 0.00 -6.28 0.00 -0.03 -0.02
Age 0.01 0.00 6.84 0.00 0.01 0.01
Female 0.22 0.02 12.25 0.00 0.19 0.25
Race
Caucasian Omitted
African American 0.08 0.03 2.84 0.01 0.02 0.13
Other -0.05 0.04 -1.27 0.21 0.02 0.13
Hispanic ethnicity -0.13 0.03 -4.61 0.00 -0.13 0.03
Education
Less than high school
High school/GED 0.00 0.02 -0.16 0.87 -0.05 0.04
Some college 0.06 0.03 2.26 0.02 0.01 0.11
College and above 0.03 0.03 0.93 0.36 -0.03 0.08
Insurance 0.10 0.03 3.59 0.00 0.04 0.15
Constant 0.89 0.07 13.56 0.00 0.76 1.02
Hospitalizations
Number of observation=107,818

Number of groups (participants)=23,764

Average observations per group (participant)=4.5

LR Chi-squared=2439.06 p-value < 0.0001

Estimate Standard Error Z p-value CI
Years of caregiving 1.41 0.05 10.83 0.00 1.33 1.51
Years of caregiving
2
0.96 0.00 -8.86 0.00 0.95 0.97
Age 1.04 0.00 15.23 0.00 1.03 1.04
Female 1.03 0.02 1.39 0.13 0.98 1.16
Race
Caucasian Omitted
African American 1.12 0.04 3.19 0.00 1.04 1.19
Other 0.98 0.06 -0.41 0.68 0.87 1.10
Hispanic ethnicity 0.81 0.04 -4.35 0.00 0.73 0.89
Education
Less than high school Omitted
High school/GED 0.88 0.03 -3.77 0.00 0.83 0.94
Some college 0.84 0.03 -4.76 0.00 0.78 0.90
College and above 0.62 0.02 -12.66 0.00 0.57 0.66
Insurance 1.07 0.04 1.52 0.13 0.90 1.16
Constant 0.02 0.00 -39.61 0.00 0.02 0.03
142
Figures

Month 2
Check in call
Month 1
Intake
Month 5
Reassessment
Month 3
Check in call
Month 4
Check in call
Month 1
Assessment
All clients at the both sites receive this
model of service delivery, but only
some use the online service delivery
option through the FCA CJ system.
Clients using traditional services
complete the intake, assessment, and
reassessment by phone rather than
online, and receive resources by mail
rather than on their online dashboard.
Caregivers who are not eligible for
local services can complete an
intake online or by phone, and then
receive some information either by
mail or their online dashboard.
Caregivers who are eligible for
services are provided with tailored
information based on both an
intake and assessment. Resources
are shared either by mail or digitally
through the FCA CJ dashboard.
Figure 2.1: Service delivery model for local services
143

Predisposing factors
Age
Gender
• Race/ethnicity
• Marital status
•
Enabling factors
• Employment status
• Educational attainment
• How caregiver learned about services
• Service site

Need factors
• Recipient cognitive impairment
Recipient activities of daily living
Completion of medical tasks
Caregiver burden
• Hours per week caregiving

Usual service delivery
approach (i.e., telephone,
in-person)
Online-delivered services
through FCA CJ
Figure 2.2: Application of Anderson Model to Healthcare Utilization to online versus usual
service use by caregivers
144

Figure 3.1: Model of intervention structure and key components

Recruitment

Target and
recruit
caregivers
starting in this
role or facing a
sudden change
in the care
situation
Assessment

Assess
caregivers for
risks as well as
protective
factors
Tailored
intervention plan

Provide feedback to
caregivers, including
recognition of strengths,
based on assessment and
reassessment

Develop (refine) a
tailored plan of which
tools to use in meeting
unique caregiver needs
Home visits

Provide intervention
over at least 12 visits

Offer relevant training,
education, and
support

Include strengths-
based feedback in
training
Reassessment

Regularly reassess caregivers,
seeking reduction of risk and
increase in protective factors
145
Figure 3.2: Example of Intervention Approach Addressing Stress Theories

Possible risk factors Theory
Tools/
Components
Primary outcomes
Secondary
outcomes
1

Recent stressful life
event
Counseling Improved coping
Increased self-
efficacy
Caregiver burden
Caregiver stress
and/or reactive
abuse
Respite
Knowledge of
conditions

Behavioral
symptoms of
dementia
Training on coping
Knowledge of
community
resources
Reduced stress
and burden

Training on behavior
management

Improved help-
seeking

Training on help-
seeking

Education on
dementia and other
health conditions

146

Figure 3.3: Example of Intervention Approach Addressing Relationship Quality

Possible risk
factors
Theory
Tools/
Components
Primary outcomes Secondary outcome
Poor relationship
quality
Counseling Improved coping
Increased self-
efficacy
1

History of violence
in relationship

Social learning/
Attachment
theory
Support group
Better
communication
skills
Improved relationship
quality

Depression
Training on coping
Reduced
depression
Identification of
possible alternatives
to provide care, if
caregiving is not
feasible

Training on
communication
Increased self-
awareness

Norms-based
feedback

1
Secondary outcomes are related to reduced risk of elder mistreatment. Self-efficacy is recognized as both
promoting engagement in the intervention and supporting behavior change.
147

Figure 3.4: Example of Intervention Approach Addressing Knowledge and Experience
Possible risk
factors
Theory
Tools/
Components
Primary outcomes Secondary outcome

Low educational
attainment

Education on
dementia
and other
conditions
Increased knowledge
of dementia and
other medical
conditions
Increased self-
efficacy
1

Recipient with
dementia

Knowledge and
experience
Support group
2

Improved problem
solving skills
Better
understanding of
dementia behaviors

Complex care tasks

Training on honing
problem-solving
skills
Increased capacity to
complete complex
care tasks
Increased ability to
provide quality care

Training on complex
care tasks

1
Secondary outcomes are related to reduced risk of elder mistreatment. Self-efficacy is recognized as both
promoting engagement in the intervention and supporting behavior change.

2
Subjective knowledge learned from other caregivers, such as during a support group, is considered valuable by
many caregivers.
148

0
1
2
3
4
5
6
7
8
9
0 years 2 years or less 4 years or less 6 years or less More than 6 years
Figure 4.1: Number of years caregiving and predicted rates of physician
visits in the last two years
Predicted value Lower CI Upper CI
Figure 4.1: Number of years spent as a caregiver and number of physician visits over the
previous year. Predicted values presented here are smaller than average visits in the
previous two years because we intentionally excluded key predictors of health service use in
our models. We did this for two reasons: 1) to keep models parsimonious and 2) to avoid
including potential mediators of health service use by caregivers in our models (e.g., self-
assessed health, depression). Predicted values presented here describe the effect of
caregiving on physician visits controlling for key demographic variables and insurance status.
149

Figure 1: A model of the agenda setting and policymaking process as described by Kingdon ’s (1984) garbage can model,
supplemented by tools to policy progress proposed by Levitsky (2014).
Figure 5.1: A model of the agenda setting and policymaking process as described by Kingdon's garbage can
model, supplemented by tad to policy progress proposed by Levitsky (2014).

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

What is success? an exploration of person-centered intervention of complex elder mistreatment

PDF

Life course implications of adverse childhood experiences: impacts on elder mistreatment, subjective cognitive decline, and caregivers' health

PDF

Developing a better understanding of elder abuse multidisciplinary teams: addressing gaps for research, policy, and practice

PDF

The dollars and sense of caregiving: exploring the financial financial realities of providing informal care to an older adult

PDF

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

PDF

Family caregiver well-being: generational differences and coping strategies

PDF

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

PDF

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

PDF

Using mixed methods to identify the characteristics of older fraud victims

PDF

The economic security of an aging minority population: a profile of Latino baby boomers to inform future retirees

PDF

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

PDF

Racial disparities in advance care planning and directives completion and end-of-life outcomes in the United States

PDF

Exploring end-of-life care experiences and disparities among Hispanic populations in the United States

PDF

Thinking generatively versus acting generatively: exploring the associations of generative self-concept and generative activity with cognitive function among older adults

PDF

Hidden perspectives: narratives of female family caregiver experiences

PDF

The effects of wisdom-related personality traits on caregivers’ health: an application of the resilience model

PDF

The role of public policy in the decisions of parents and caregivers: an examination of work, fertility, and informal caregiving

PDF

Nursing home transitions: a new framework for understanding preferences, barriers and outcomes

PDF

Disrupting career exploration: a curriculum for traditionally marginalized students to become agents of their own future

PDF

A targeted culturally-informed approach for caregiver stress among Vietnamese caregivers of family members [executive summary]

Asset Metadata

Creator Meyer, Kylie Nicole (author)

Core Title New directions for family caregiver interventions

School Leonard Davis School of Gerontology

Degree Doctor of Philosophy

Degree Program Gerontology

Publication Date 11/09/2018

Defense Date 10/16/2018

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

Tag elder abuse,elder mistreatment,family caregiving,healthcare utilization,intervention,OAI-PMH Harvest,online services,policymaking,translation

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Wilber, Kathleen (committee chair), Benton, Donna (committee member), Enguidanos, Susan (committee member), Gassoumis, Zachary (committee member)

Creator Email kyliemey@usc.edu,kylienmeyer1@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-105958

Unique identifier UC11675439

Identifier etd-MeyerKylie-6944.pdf (filename),usctheses-c89-105958 (legacy record id)

Legacy Identifier etd-MeyerKylie-6944.pdf

Dmrecord 105958

Document Type Dissertation

Format application/pdf (imt)

Rights Meyer, Kylie Nicole

Type texts

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

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA