Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
The roles of social isolation and secondary caregiver support in the stress process among primary caregivers of older adults
(USC Thesis Other)
The roles of social isolation and secondary caregiver support in the stress process among primary caregivers of older adults
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
THE ROLES OF SOCIAL ISOLATION AND SECONDARY CAREGIVER SUPPORT IN
THE STRESS PROCESS AMONG PRIMARY CAREGIVERS OF OLDER ADULTS
by
Jiaming Liang
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
(SOCIAL WORK)
August 2022
Copyright 2022 Jiaming Liang
ii
ACKNOWLEDGEMENTS
At the time of writing this acknowledgement, I have already passed my defense and am
just one step away from earning my Ph.D.
Since I finished the work in April, I've been hesitant to write this acknowledgment.
Although I used to write it in my undergraduate and master's thesis to past teachers, family
members, and friends, it is tough to write it this time for the PhD dissertation. It's not that I do
not have sensations; rather, my heart is filled with conflicting emotions. I do not really know
where to begin when I think of my life in Los Angeles during the last four years.
Human fate is always surprisingly unpredictable. I could never have thought, growing up
on the other side of the ocean in a small town in southwest China, that I would earn my family's
first doctorate degree in the far western hemisphere before the age of thirty. I believe I owe a
heartfelt thank you to everyone who has supported and encouraged me since the day I was born!
A home is a safe haven. Despite the fact that my parents were not wealthy, their nurturing
helped me avoid taking a detour. Even if there are failures along the route of studying, it just
strengthens and braves me.
I am fortunate in that there are always good people willing to assist me in my studies.
Throughout my four-year PhD program, my mentor, Dr. Maria P. Aranda, provided me with the
direction and support I needed to pass test after one another, and keep moving forward
academically. I would also like to express my gratitude to Dr. Yuri Jang and Dr. Kathleen
Wilber, without whom I would not have been able to successfully complete my dissertation.
The ancient Chinese believed that a man should be self-sufficient by the age of thirty, and
it was at this age that I met the love of my life, Yuting Zhang. I was able to finish my dissertation
iii
on time and pass the defense with ease since I had my wife at my side to share all of my
pleasures and sorrows.
Finally, I would like to offer my deepest appreciation to my entire family and circle of
friends. I wish you the best of luck in the coming days, that all of your dreams come true, that
you have good health, and that you have a happy life!
Jiaming Liang
June 2, 2022
iv
TABLE OF CONTENTS
ACKNOWLEDGEMENTS ............................................................................................................ ii
LIST OF TABLES ......................................................................................................................... vi
LIST OF FIGURES ...................................................................................................................... vii
ABSTRACT ................................................................................................................................. viii
CHAPTER ONE: INTRODUCTION ............................................................................................. 1
CHAPTER TWO: CARE DEMANDS AND CAREGIVING STRESS ........................................ 4
2.1 Population Aging and Increasing Care Needs ...................................................................... 4
2.2 Role of Family Caregivers .................................................................................................... 6
2.3 Impact of Caregiving Stress on Family Caregivers .............................................................. 7
2.4 Primary vs. Secondary Caregivers ...................................................................................... 11
CHAPTER THREE: SOCIAL ISOLATION OF FAMILY CAREGIVERS ............................... 14
3.1 Defining the Construct of Social Isolation .......................................................................... 14
3.2 Measuring Social Isolation of Family Caregivers .............................................................. 17
3.3 Social Isolation May Mediate the Stress Process of Caregiving ........................................ 19
CHAPTER FOUR: SUPPORT FROM SECONDARY CAREGIVER NETWORK .................. 25
4.1 Define Secondary Caregiver Network ................................................................................ 26
4.2 Previous Research on SCN support .................................................................................... 28
4.3 The Moderation Effects of SCN Support in Family Caregiving ........................................ 29
4.4 Potential Patterns of SCN Support ...................................................................................... 31
CHAPTER FIVE: ROLES OF GENDER, RACE, AND RELATIONSHIP TYPE .................... 34
5.1 Gender, Race, and Relationship Type on Family Caregiving ............................................ 34
5.2 The Social Isolation Mediated Model May Vary by Gender, Race,
or Relationship Type ........................................................................................................... 39
5.3 The Moderation Effects of SCN Support May Vary by Gender, Race,
or Relationship Type ........................................................................................................... 42
v
5.4 The Potential Intersections of Gender, Race, and Relationship Type ................................ 44
CHAPTER SIX: CONCEPTUAL FRAMEWORK AND RESEARCH AIMS ........................... 48
CHAPTER SEVEN: METHODS ................................................................................................. 52
7.1 Data Source ......................................................................................................................... 52
7.2 Inclusion and Exclusion Criteria ......................................................................................... 52
7.3 Sample Size and Characteristics ......................................................................................... 53
7.4 Measures ............................................................................................................................. 54
7.5 Analysis ............................................................................................................................... 59
CHAPTER EIGHT: RESULTS .................................................................................................... 64
8.1 Descriptive Characteristics of the Participants ................................................................... 64
8.2 AIM 1: The Mediated Effects of Social Isolation on the Process
from Caregiving Stressors to Caregiver Well-being ........................................................... 69
8.3 AIM 2: The Moderation Effects of SCN Support
on the Social Isolation Mediated Caregiving Stress Process .............................................. 72
8.4 AIM 3: Model Comparisons by Gender, Race, and Relationship Type ............................. 80
CHAPTER NINE: DISCUSSION ................................................................................................ 97
9.1 Descriptive Characteristics of the Study Sample ................................................................ 98
9.2 The Mediated Effect of Social Isolation ........................................................................... 103
9.3 The Moderation Effects of SCN Support .......................................................................... 105
9.4 The effects of gender, race, and relationship type ............................................................ 109
9.5 Research innovations and implications ............................................................................. 115
9.6 Limitations ........................................................................................................................ 120
9.7 Conclusion ........................................................................................................................ 121
REFERENCES ........................................................................................................................... 123
vi
LIST OF TABLES
Table 1 The sociodemographics of primary caregivers in 2015 (N = 782) .................................. 53
Table 2 Sociodemographics of Older Adults in 2015 (N = 782) .................................................. 54
Table 3 Overview of variables and measurements in the proposed study .................................... 59
Table 4 Descriptive statistics of primary caregivers (N = 782) .................................................... 66
Table 5 Descriptive statistics of older adults in wave 1 (N = 782) ............................................... 67
Table 6 The characteristics of secondary caregiver network (N = 1003) ..................................... 69
Table 7 The mediated model of social isolation between caregiving stressors
and depression in primary caregivers .............................................................................. 70
Table 8 The mediated model of social isolation between caregiving stressors
and anxiety in primary caregivers .................................................................................... 72
Table 9 Latent profile solutions for SCN support from two to five classes ................................. 73
Table 10 Predicted prevalence and descriptive statistics of SCN
a
support subgroups ................. 75
Table 11 Cross-group equality test on latent measurement construct .......................................... 76
Table 12 The moderation effects of SCN
a
support on the depression model ............................... 78
Table 13 Comparing means of observed variables by gender, race, and relationship type .......... 82
Table 14 Cross-group equality on latent measurement construct by gender, race,
and relationship type ...................................................................................................... 84
Table 15 Multigroup analysis on social isolation mediated model by gender, race,
and relationship type ...................................................................................................... 87
Table 16 Multigroup analysis on moderation effects of SCN support by gender, race,
and relationship type ...................................................................................................... 90
Table 17 Multigroup analysis on moderation effects of SCN support
at the intersection of gender and race ............................................................................ 95
vii
LIST OF FIGURES
Figure 1 The Conceptual Framework of the Study ....................................................................... 48
Figure 2 The two-wave mediation model of the study ................................................................. 60
Figure 3 The mediated model of social isolation between caregiving stressors
and depression in primary caregivers ............................................................................. 71
Figure 4. The distributions of the latent subgroups on the SCN support indicators ..................... 75
Figure 5 The moderation effects of SCN support on the depression model ................................. 79
Figure 6 The impact of objective stressor on depression by race ................................................. 88
Figure 7 Moderation effects of SCN support on women and men ............................................... 92
Figure 8 Moderation effects of SCN support on White ................................................................ 93
Figure 9 Moderation effects of SCN support at the intersection of gender and race .................... 96
viii
ABSTRACT
Primary caregivers of older adults are more likely to experience social isolation and
mental health problems as a result of their responsibilities. To reduce caregiver burden and
protect their health and well-being, it is critical to investigate primary caregivers' experiences of
caregiving and identify supportive factors. Furthermore, research should take into account
sociocultural diversity and investigate how the caregiver stress process differs depending on
primary caregiver sociodemographic characteristics. The purposes of this dissertation are to
examine the mediated effect of social isolation between caregiving stress and mental health
symptoms (i.e., depression and anxiety) of primary caregivers, and the moderation effect of
support from secondary caregiver network (SCN) on the stress process, by applying the Stress
Process Model of Caregiving (SPM). A second aim is to test model variances by gender, race,
and relationship type of primary caregivers, as well as their potential intersections. Secondary
data analyses were conducted on 782 primary caregivers and 1003 secondary caregivers using
data from the 2015 and 2017 National Study of Caregiving (NOSC) and the linked 2015
National Health and Aging Trends Study (NHATS). Structural equational models were built to
examine the study aims, and the composite measurements (e.g., caregiving stress, social
isolation, and SCN support) were tested for their internal consistency and construct validity. The
findings suggest that taking on caregiving responsibilities can cause social isolation, which can
lead to depression in primary caregivers. High level of SCN support, however, may protect
primary caregivers’ social connections and mental health from the impacts of caregiving stress.
Intersectional analyses also found that high level of SCN support may be more effective in
protecting Black women caregivers’ mental health than in other gender-race groups. In
conclusion, the results of the dissertation study enable a clearer understanding of the stress
ix
process and how SCN support hold promise for protecting social networks and mental health of
primary caregivers with different sociodemographic backgrounds.
1
CHAPTER ONE: INTRODUCTION
Globally, the aging population, particularly those with chronic conditions and physical
and functional limitations, is creating unprecedented demand for long-term care. The majority of
caregiving for older adults is done by family caregivers, who are usually the older adults'
partners, relatives, or friends, and receive no compensation for their efforts. Taking care of an
older family member can be stressful, which may impact family caregivers’ emotional and
mental health, sleep quality, and family relationships (Adelman et al., 2014; Goren et al., 2016;
Liang et al., 2020; Longacre et al., 2017; Moon et al., 2017). Furthermore, intensive care work
can take up caregivers' personal time, limiting their opportunities for self-care and social
engagement, leading to burnout and unemployment, as well as financial hardship for families
(Bullock et al., 2003; Ghosh et al., 2020; Hamid & Musa, 2017). As a result, more research is
needed to understand the mechanisms that influence caregiving stress and to identify potentially
modifiable factors that can help family caregivers reduce caregiving burden while also
improving their health and quality of life.
The Stress Process Model of Caregiving suggests that one possible pathway by which
caregiving stress affects the mental health of family caregivers is by limiting caregivers' social
engagement (Pearlin et al., 1990). Due to the commitment to caregiving, caregivers of older
adults frequently experience social isolation, which describes the caregiver's objectively reduced
frequency of meeting with friends and participating in activities of interest, as well as a
subjectively perceived lack of social connection and support (Cornwell & Waite, 2009; Santini et
al., 2020; Yuen et al., 2021). Taking care of an older person with cognitive or physical
limitations is often a time-consuming and long-lasting process, which may gradually separate
family caregivers from their original social networks. Disconnection from the social network
2
may negatively impact the caregiver's own mental health, quality of life, relationship satisfaction
with the older adult, and persistence of caregiving roles (dos Santos Azevedo et al., 2021;
Greenwood et al., 2019; Perez et al., 2021; Segrin et al., 2018).
Social support is thought to be an effective way to reduce chronic stress, as well as to
protect and improve one's mental health. Significant evidence from caregiving research suggests
that social support can help to mitigate the negative effects of caregiving (Dam et al., 2016;
Halpern et al., 2017; Wang et al., 2016). However, social support is a relatively broad concept,
which includes many different sources and forms of support, and includes objectively accepted
support and subjectively perceived support (Cohen & Wills, 1985). Help from other family
members or friends with caregiving tasks and sharing caregiving responsibility may be the most
direct and effective form of social support for family caregivers. According to research, many
older adults are not helped by just one caregiver, but by a team of caregivers (Gonçalves-Pereira
et al., 2020; Huiying et al., 2019; Spillman et al., 2020). A primary caregiver is the one who
provides the most care and takes on the most responsibility, while one or more secondary
caregivers provide a smaller amount of assistance. The current understanding of the role of
secondary caregiver support in reducing caregiver social isolation is very limited. As a result,
more research is needed to see if support from secondary caregivers can help primary caregivers
preserve their social engagement and mental health.
In addition, the caregiving experience of caregivers with different background
characteristics may vary (Lin et al., 2012; Siefert et al., 2008). The perception of caregiving
stress, the impact on health outcomes, and caregivers' coping resources and strategies were found
to be varied by background factors such as race, gender, and relationship with older adults.
Women, for example, have long been thought to be more likely to take on the role of family
3
caregiver, and their workplace inequity means they have fewer social resources and are more
likely to lose their jobs after becoming caregivers (Akpınar et al., 2011; Hsiao, 2010; Penning &
Wu, 2016). Racial/Ethnic minorities, such as Blacks, are disproportionately more likely to care
for family members with disabilities due to financial and social disadvantages relative to Whites
(Chen et al., 2015; Fabius et al., 2020). Spouse caregivers may be more negatively affected by
caregiving stress due to their older age and potentially poorer physical health (Haley et al., 2010;
Jinyu, 2021). However, studies also suggest that adult children generally have their own family
and work, and taking care of their older parents is an extra burden (Penning & Wu, 2016;
Tolkacheva et al., 2011). In addition to the impact of these factors respectively, their
intersections may reflect overlaps of social disadvantages in some caregivers, thus expand
disparities in social resources and health among caregivers of different backgrounds. Therefore,
it is necessary to holistically understand the caregiving experience and the protective effects of
support from secondary caregivers among family caregivers of different genders, races, and
relationship types.
In summary, based on the Stress Process Model of caregiving, this dissertation focuses on
the caregiving experience of primary caregivers of older adults by: (1) exploring the mediated
role of social isolation on the relationship between caregiving stress and mental health, (2)
examining the moderation effects of support from secondary caregivers on this relationship, and
(3) adopting an intersectional perspective to examine the heterogeneity of the caregiving stress
process under different races, genders, and relationship types, and whether support from
secondary caregivers as a modifiable factor exhibits similar protective effects across different
race/gender/relationship type groups.
4
CHAPTER TWO: CARE DEMANDS AND CAREGIVING STRESS
2.1 Population Aging and Increasing Care Needs
Population aging is a structural challenge facing the world in the 21st century. According
to the data of the United Nations World Population Prospects 2019, the older population over the
age of 65 has reached 703 million globally in 2019, which means that there is one older person in
every 11 people (United Nations, 2019). The growth rate of the aging population is still
increasing with the overall aging of the baby boomer generation. It is estimated that by 2050, the
global aging population will double to 1.5 billion, accounting for 16% of the total population
(United Nations, 2019). American society is experiencing rapid aging too. Compared to 10 years
ago, the number of older Americans over 65 has increased by 16%, while the population under
65 has only increased by 3%. As of the end of 2019, there were more than 54 million older
people over 65 years old in the country, accounting for 16% of the country's total population,
which is much higher than the world average for the same period (Administration for
Community Living, 2021).
Increasing age is associated with chronic conditions that can affect functioning of older
adults. The advancement of modern medicine has allowed older adults to live longer while
suffering from one or more chronic diseases, extending their life expectancy. According to
surveys, the majority of older Americans suffer from at least one chronic illness, such as
hypertension, arthritis, or coronary heart disease (Administration for Community Living, 2021;
Sierra et al., 2009). In addition, limitations in physical and cognitive function are also associated
with decreased quality of life of many older adults. Hearing and vision loss, for example, can
occur at any age, but it is more common in people over 65. In the absence of appropriate
corrective aids, hearing and vision loss in older adults can affect their communication and
5
mobility (National Alliance for Caregiving & AARP, 2020). Chronic diseases that are not well
managed and controlled, such as diabetes, can also have devastating effects on physical
functioning in older adults (Mannino et al., 2008). Similarly, while cognitive decline can occur in
young and middle-aged people, the prevalence of cognitive impairment rises with age (Langa et
al., 2017). Neurodegenerative diseases, such as Alzheimer's disease and Parkinson's
disease, have significant impacts on older adults' ability to live independently. Decreased
memory function, for example, may affect older adults' ability to perform routine tasks like
laundry and cooking. Psychiatric symptoms such as anxiety and depression, as well as suicidal
ideation, may also accompany cognitive impairment (Feter et al., 2021; Kinzer & Suhr, 2016;
Steffens et al., 2009).
With the increasing number of older adults with chronic diseases and functional
limitations, the challenges faced by long-term care services also increase. Long-term care
services include a broad range of health, personal care, and supportive services that meet the
needs of frail older adults and other adults whose capacity for self-care is limited due to chronic
illness, injury, physical, cognitive, or mental disability (Harris-Kojetin et al., 2019). The
government has been financially supporting the development of long-term care services. For
example, The Patient Protection and Affordable Care Act (ACA) defines "long-term care and
supports" as including institutionally based and non-institutionally based long-term services and
supports, which attempts to expand the scope of government-supported services from traditional
health care facilities to home and community-based services (Thach & Wiener, 2018). Although
the government pays over $240 billion annually for long-term care services through programs
such as Medicaid and Medicare, the supply of home and community services is still far below
the demand (Sutcliffe et al., 2016). In fact, nearly 80% of care/help to older people is provided
6
by family and friends, which makes family caregivers an integral part of the long-term care
system (Schulz & Czaja, 2018).
2.2 Role of Family Caregivers
Family caregivers are usually relatives, friends, partners, or neighbors of older adults who
have physical, mental, or cognitive functional limitations (Gaugler et al., 2003; Gonçalves-
Pereira et al., 2020; Tausig, 1992). Support for older adults is generally free of charge, with the
possible exception of some public benefits and subsidies. The role of the family caregiver varies
widely due to caregiving-related characteristics such as the caregiver's relationship with older
adults, the type of help the caregiver provides, the caregiver's care intensity, and the caregiver's
own competence, skill, and motivation to help. Because of mobility issues, an older adult may
require assistance to get up and walk around the yard. As his family caregiver, his spouse is
responsible for not only doing the housework but also assisting him in getting up, dressing, and
showering. While the old man's neighbor, who drove him to church activities every week, also
became a caregiver who helped him get around.
A broad definition considers anyone who provides emotional, physical, or financial
support to an older adult to be a caregiver (Penrod et al., 1995; Tausig, 1992). However, such a
definition raises concerns that it will muddle caregivers' and older adults' social relationships.
Friends of an older adult at a community activity center, for example, may not consider
themselves caregivers simply because they chat with the older adult during activity time. As a
result, this study recommends that caregivers be defined as those who provide direct assistance
to older adults for the convenience of daily life, such as housekeeping, financial management,
health monitoring, and self-care activities.
7
Family caregiving is also becoming more common. According to a 2020 report, one in
every five people in the United States is a family caregiver, which includes caring for children
under the age of 18 and adults with disabilities under the age of 50 (National Alliance for
Caregiving & AARP, 2020). According to data released by the National Health and Aging
Trends Study (NHATS) and National Study of Caregiving (NSOC) in 2011, a total of 17.7
million informal caregivers are providing care for older adults over the age of 65 on instrumental
activities of daily living (IADL) and activities of daily living (ADL) (Freedman & Spillman,
2014). Such a large number of caregivers, coupled with the differences in their own backgrounds
and older adults' care needs, make it difficult for research to obtain comprehensive and accurate
data to analyze and understand the real situation and difficulties faced by family caregivers.
More research on family caregivers' caregiving experiences is needed to inform practice,
services, and policy development at all levels.
2.3 Impact of Caregiving Stress on Family Caregivers
Family caregiving is a public health issue that affects the entire society, and it has piqued
the interest of experts and scholars from various fields, including social workers, psychologists,
sociologists, and public policy researchers. From a social work and psychological perspective,
caregiving is a source of chronic stress for family caregivers. According to the Stress Process
Model of Caregiving (SPM), chronic stress causes physical and psychological strains in
caregivers, which can lead to secondary strains in other areas of their lives and a decline in their
well-being in the long run (Pearlin et al., 1990).
However, family caregivers are not always negatively affected by their caregiving
experience. In one study, 57 family caregivers were interviewed and their perceived positive
8
gains from caregiving were identified (Cheng et al., 2016). For example, family caregivers may
experience increased satisfaction, self-confidence, and self-resilience as a result of resolving a
care problem; older adults may also express gratitude to caregivers, which is an external reward.
Furthermore, receiving assistance from other family members or friends makes family caregivers
feel supported, increasing their sense of security in the environment, family cohesion, and
satisfaction with social relationships. These positive aspects are important motivators for
caregivers to persevere in the face of chronic stress, as they help them maintain self-confidence
and reduce feelings of burden.
In contrast to the findings on the negative effects of caregiving on family caregivers, the
evidence on positive aspects is still limited. Therefore, a large body of research continues to
focus on the negative effects and potential consequences of caregiving stress on family
caregivers, as well as attempts to identify risk factors in order to better understand how the
caregiving experience affects caregivers' health and well-being. To begin investigating these
issues, a literature review was conducted on the biological, psychological, and social effects of
caregiving stress on family caregivers.
A number of biological markers have been identified to reflect changes in family
caregivers' physiological status under chronic caregiving stress. For example, family caregivers
who report high levels of stress have compromised endocrine systems, including lower levels of
hormones and neurotransmitters (Brown et al., 2020; Lovell & Wetherell, 2011). The activity of
leukocytes in the blood is affected, and the level of interleukin caused by infection is lower than
that of non-caregivers with similar demographic characteristics, indicating a reduction immune
function (Kuster & Merkel, 2004; Lovell & Wetherell, 2011; Pinquart & Sorensen, 2006).
Research also found that the metabolic capacity of family caregivers fluctuated greatly, heart rate
9
and blood pressure were also affected to varying degrees, and the incidence of coronary heart
disease was significantly increased (Kim et al., 2015; Rowe & Gross-King, 2015). In terms of
overall health, family caregivers have a lower self-reported health and health-related quality of
life, a higher prevalence of chronic diseases and somatic symptoms, and a higher use of health
services and clinic visits (Litzelman et al., 2016; Polenick et al., 2018). Most of the studies on the
effects of caregiving stress on the physical health of family caregivers use a control group
design, and the results have high validity. However, related studies tend to have small sample
sizes and strict eligibility criteria for recruitment. Therefore, large representative samples were
rarely used to examine the diversity of background characteristics of caregivers, and the
generalizability of the results is limited.
Evidence has accumulated on the effects of caregiving stress on the psychological health
of family caregivers. Numerous studies have shown that caregiving stress is related to family
caregivers' perceived psychological distress, including symptoms of depression and anxiety
(García-Alberca et al., 2012; Shin et al., 2018; Wang et al., 2016). Many family caregivers,
particularly dementia caregivers, express feelings of loneliness, owing to the fact that caring for
an older person with progressive memory loss can make the caregiver feel as if they are caring
for a stranger (Vasileiou et al., 2017a). Positive interactions between caregivers and older adults
can bring caregivers a sense of accomplishment and purpose, while lack of positive feedback
may lead caregivers to feel the loss of their sense of purpose or meaning in life (Cheng et al.,
2016; Tarlow et al., 2004). Most of these studies use cross-sectional data which has a sufficient
representative sample size, but also inevitably limits the validity of the results due to the bias of
sampling and analytical methods. Longitudinal approaches are critical for research examining the
effects of caregiving stress on the psychological health of family caregivers. According to
10
research, when people take on the role of caregiver, their mental health drops significantly, and
this drop is directly related to the care demands of older adults (Rafnsson et al., 2017; Trivedi et
al., 2014). Furthermore, caregivers who have been in the role for a long time are more likely to
send older adults to nursing facilities than those who have just started (Chyr et al., 2020; Spruytte
et al., 2001).
The impact of caregiving stress on family caregivers' social lives is inextricably linked to
their psychological health. Negative emotional symptoms of caregivers may be passed down
through the family, i.e., negative emotions experienced by caregivers may be passed down to
other family members through daily communication, affecting the family atmosphere and
relationships (Falcão et al., 2016). The caregiver's relationship with the older adult is also highly
vulnerable to both parties' emotions. For example, older adults with psychiatric disorders may
frequently experience negative emotions and behavioral issues, which can have a significant
impact on the caregiver's mood and reduce satisfaction with the caregiving relationship. The
older person may also complain about the caregiver not meeting their care needs (Braun et al.,
2009; Pristavec, 2019; Segrin et al., 2018). Reduced satisfaction with the caregiving relationship
may have an impact on the quality of care, predict health declines in the older adults, and even be
linked to more hospital visits and mortality in the older adults (Braun et al., 2009; Pristavec,
2019; Thandi et al., 2017). The family relationship may also be affected when multiple family
members do not agree on care arrangements (e.g., assigned caregiving workload, financial
matters) (Xu et al., 2021). Additionally, many caregivers have to withdraw from some social
activities, reduce social interaction with friends, and reduce work participation to assure
sufficient time and energy to care for older adults at home, which may lead to social isolation, or
even unemployment (Bullock et al., 2003; Scharlach et al., 1991; Sun et al., 2019).
11
Unemployment not only completely changes the life patterns of caregivers, but also reduces the
income level of the entire family, adding more financial strain to caregivers.
2.4 Primary vs. Secondary Caregivers
Although there is ample evidence that caregiving stress has a variety of negative
consequences for caregivers, not every family caregiver is affected to the same extent. The
caregiver's responsibility for care, such as the type of care tasks and the intensity of care work, is
one of the influencing factors. An older person usually receives help from multiple family
caregivers, and these caregivers can be classified as primary or secondary caregivers based on
the amount of work they do (Andersson & Monin, 2017; Penrod et al., 1995; Spillman et al.,
2020). The primary caregiver usually provides the most assistance, while the remaining
caregivers are referred to as secondary caregivers because they provide less frequent/intensive
assistance. By comparing the intensity of care provided by multiple caregivers, primary
caregivers can be identified (Dilworth-Anderson et al., 1999; Staight & Marie Harvey, 1990;
Tennstedt et al., 1989).
A national survey shows that only 20% of family caregivers in the United States are sole
caregivers, which means they do not receive any help from other family/unpaid caregivers
(National Alliance for Caregiving & AARP, 2020). On average, each older adult with dementia
has more than 3 family caregivers to help with their daily life, and this number is increasing
(Spillman et al., 2020). Therefore, it is necessary to use a network perspective to
comprehensively consider the relationship between multiple family caregivers and the quality of
care received by older adults, as well as the interaction between family caregivers (e.g., the
support of secondary caregivers to primary caregivers), which will help understand the actual
12
family caregiving situation of older adults, as well as the experience and needs of family
caregivers, to provide evidence for the improvement of community-based services and supports.
Primary caregivers are more likely to experience higher level of caregiver burden and
mental distress (e.g., anxiety and depression),because they are more engaged in caregiving
activities and spent more time in caregiving, and are often close relatives (e.g., spouses or adult
children) of older adults (Pinquart & Sorensen, 2006). Although some research suggests that
secondary caregivers may be sympathetic to the distress experienced by older adults and primary
caregivers, and also experience burden when participating in caregiving activities (Gonçalves-
Pereira et al., 2020; Monin & Schulz, 2010), most of the available evidence indicates that
primary caregivers report higher or similar burden and distress levels relative to secondary
caregivers (Everhart et al., 2018; Marino et al., 2020; Staight & Marie Harvey, 1990). It's
difficult to generalize secondary caregiver experiences due to differences in the number of
secondary caregivers in each household and their levels of involvement in caregiving activities.
As a result, the focus of this research is on primary caregivers in order to better understand the
psychosocial mechanisms of caregiving stress on their well-being, as well as to investigate the
impact of secondary caregivers support on primary caregivers, which could be a critical
component of social support for primary caregivers.
In summary, the existing research on family caregiving for older adults can be
summarized as follows:
• The demand for family caregivers will rise as the population continues aging.
• Family caregivers bear a significant portion of the cost of long-term care and are
subjected to the biopsychosocial effects of caregiving stress.
13
• Primary caregivers have the most responsibilities and are more likely to report higher
caregiving stress than secondary caregivers, but there is little evidence about how
multiple family caregivers interact.
Understanding the experiences and needs of family caregivers can help improve social
work practices and guide public policy formulation, which is critical for improving the quality of
care for older adults while also protecting caregivers' physical and mental health. More research
on the impact of caregiving stress on primary caregivers is needed because they are the backbone
of family caregiving. The social work perspective emphasizes the interplay of mental health and
social relationships. Given that family caregivers are more vulnerable to social isolation, which
may require them to withdraw from social activities, give up social relationships, or even quit
their jobs, their social functions are undoubtedly harmed. According to the Stress Process Model
of Caregiving, family caregivers' social network constraints will eventually manifest in their
psychological health outcomes. Therefore, this dissertation intends to explore a possible pathway
by which caregiving stress affects the primary caregiver's social life and, as a result, has a
negative impact on their mental health.
Successfully identifying the pathway will provide valuable evidence for the advancement
of community-based social work practice and the formulation of public policy. For example,
more community-based services aimed at assisting family caregivers in reducing burden,
enabling self-care and social participation could be developed for the purpose of protecting their
psychological health. In the next chapter, a more detailed literature search was conducted on the
possible psychosocial mechanism of caregiving stress on family caregivers.
14
CHAPTER THREE: SOCIAL ISOLATION OF FAMILY CAREGIVERS
Social isolation and loneliness are common experiences among family caregivers (Hajek
et al., 2021; Neri et al., 2012a; Vasileiou et al., 2017b). Caring for older adults with physical and
cognitive limitations can cause chronic stress in family caregivers, and caregiver burden is
thought to limit caregiver social interactions. Many studies show that social isolation as a risk
factor has negative consequences for one's mental and physical health. Social isolation, for
example, has been linked to increased morbidity, impaired immune function, and an increased
risk of inflammation (Naito et al., 2021; Scharf et al., 2021). In terms of psychological health,
social isolation and perceived loneliness were found to be associated with faster cognitive
decline, depression, sleep disturbance, and higher mortality in older adults (Blazer, 2020; Perez
et al., 2021; Xiang et al., 2020). However, studies investigating the effects of social isolation on
family caregivers are limited, and no studies have examined whether social isolation can serve as
a mediator to explain the effects of caregiving stress on family caregivers' mental health.
Through a review of the relevant literature, this chapter aims to improve understanding of
caregivers' experiences of social isolation and its health consequences.
3.1 Defining the Construct of Social Isolation
The definition of social isolation has been controversial. Some researchers tend to define
social isolation as the degree of objective separation between individuals and the external
society, that is the, “absence of social relationship” (Coyle et al., 2021; Hawton et al., 2010). At
various stages of life, a person will meet and know a variety of people, including parents,
relatives, classmates, coworkers, and other friends, all of whom form a person's social network.
The social resources available to a person are determined by interactions with individual
15
members of a social network, including both objective contact and subjective relationship quality
(Weiai et al., 2013). Social interaction can take many forms, such as daily communication with
family members, occasional visits with friends, collaboration with colleagues on work, and
social activity participation due to personal interests, social identities, and religious preferences.
Using this definition, the measurement of objective social isolation includes two parts: (a) social
contact, which is mainly calculated by counting the social contacts between the person and the
members of the social network within a certain time and space; (b) social participation, which
calculates the frequency of individual's participation in several major social activities (Pohl et al.,
2017). The absence of objective social isolation indicates individuals' lack of social capital to
help cope with the health impacts of various stressors. Studies using objective social isolation
measures revealed its association with poor quality of life, physical disability, and cognitive
impairment (Cornwell & Waite, 2009; dos Santos Azevedo et al., 2021; Quach & Burr, 2021).
Objective social isolation conceptually separates social isolation from loneliness, which
reflects an individual's subjective perception of social relationships, social connections, and
social resources (Kovaleva et al., 2018). However, the calculation of social contacts and
participation cannot reflect the individual's integration in social relationships, especially many
people who live alone but continue to be very active in their social networks (Lubben, 1988).
Objective social isolation and loneliness, as two highly correlated constructs, must be
investigated together in research, which will inevitably make results interpretation and
mechanism discussion more difficult (Cornwell & Waite, 2009). More and more researchers
proposed that social isolation should be considered as a complex and multi-dimensional concept,
and defined it as having a limited social network, lack of social support, lower frequency of
social activities participation, as well as a self-perception of social isolation (Cornwell & Waite,
16
2009; Santini et al., 2020; Yuen et al., 2021). According to this definition, social isolation could
be divided into two dimensions. The objective dimension is social disconnectedness which
indicates the lack of social interaction with others and participation in social activities (Quach &
Burr, 2021); The subjective dimension refers to loneliness, which is defined as an unpleasant
experience that occurs when a person's social interaction is inadequate (Vasileiou et al., 2017a).
Some may argue using a scale to combine social disconnectedness and loneliness ignores
the independence of loneliness. It is true that, for some people, the size of their social network
determines the quality of their social life, but some people's perceived social isolation is not
related to their actual social network size and time spent socializing. Studies also found that
social disconnectedness and loneliness are both respectively correlated with individual mental
health, chronic conditions, frailty, and life satisfaction (Blazer, 2020; Ge et al., 2022; Sun et al.,
2019). Therefore, it seems necessary to evaluate the two constructs separately to explore the
mechanism of their impacts on health outcomes.
However, not all populations have the same relationship between social isolation and
loneliness. It has a strong relationship with sociodemographic and other contextual factors. The
most vulnerable older adults, for example, have lower levels of social engagement and are more
likely to be lonely. If healthy older people participate in community activities on a regular basis,
they can maintain a large social network, but widowhood can significantly increase their
subjective loneliness. Even if immigrants are able to find work and form a stable social network
in their new surroundings, nostalgia for their previous life and cultural environment causes them
to feel lonely (De Jong Gierveld et al., 2015; Fried et al., 2015; Wright et al., 2019). In addition,
individuals' perception of loneliness is also related to their personality traits (Peerenboom et al.,
2015). Therefore, neither social disconnectedness nor loneliness alone can reflect the actual
17
social situation of the person, which requires a composite scale to comprehensively measure
subjective and objective social isolation.
3.2 Measuring Social Isolation of Family Caregivers
According to research, family caregivers of older adults, particularly those with severe
health conditions and intensive care needs, have fewer social interactions, a more withdrawn
social life, and are more lonely. One longitudinal study, for example, looked at changes in social
isolation among family caregivers of older adults over the course of 1-2 years using a combined
measurement that assessed both objective social disconnectedness and subjective loneliness. The
researchers discovered an increasing trend in social isolation among family caregivers, which
was strongly predicted by their caregiving hours (Li et al., 2020).
While there are numerous measures of social isolation and loneliness, no scale that
combines objective social disconnectedness and subjective loneliness is available specifically for
family caregivers. The majority of the early scales were designed to assess the social interaction
of older adults and only assessed their objective social connection. Some items were obviously
sensitive to the characteristics of the measured group, such as living alone and being unmarried
(Dean et al., 2016; Hughes & Gove, 1981; Lillard et al., 2015). Although many family caregivers
live with or close to the older adults, this does not mean that they have more social connections
than those who do not. Family caregivers who do not live with older adults, on the other hand,
may have better sleep quality and more social time due to lower nighttime caregiving workload.
Similarly, spousal caregivers do not necessarily have more social connections than adult child
caregivers who are not married.
18
In recent decades, researchers have proposed using an integrated scale that measures
social isolation in two dimensions: objective and subjective. The Social Isolation Index (SII),
which includes both objective and subjective social isolation, is the most representative of these
(Cornwell & Waite, 2009). Social disconnectedness assesses an individual's social network size
(e.g., number of relatives/friends) and social participation (e.g., meetings, groups, gathering, and
volunteering). Subjective loneliness assesses perceived loneliness. The SII has been shown to
have good measurement validity and predictive power of health outcomes, including life
satisfaction, depression, self-rated health, and mental health of older adults (Wister et al., 2019).
There are studies using the conceptual framework of the SII measured social isolation for family
caregivers and found good internal consistency (Li et al., 2020). Therefore, this study adopts an
integrated approach to combine the objective social disconnectedness of family caregivers with
subjective loneliness to evaluate their social isolation.
Given that the analysis of this study is based on extant secondary data, including the
National Health and Aging Trend Study (NHATS) and National Study of Caregiving (NSOC),
the construction of the measurement should be based on available evidence and data, and the
reliability and validity of the newly constructed measurement need to be evaluated. Literature
review found studies that have constructed measures of social isolation for older adults based on
NHATS data. The main items include living alone, unmarried, having no friends/family to talk,
having no friends/family to visit, and social activity participation (Cudjoe et al., 2020; Pohl et al.,
2017). Given that the above items are also available in the NSOC survey and the impact of the
background characteristics of family caregivers on their social isolation, the measurement used
in this study will exclude the two items living alone and being unmarried. In addition, this
measurement only assessed objective social disconnectedness in older adults, this study will also
19
incorporate measures of subjective loneliness to achieve a comprehensive measure of social
isolation in family caregivers.
3.3 Social Isolation May Mediate the Stress Process of Caregiving
3.3.1 The Stress Process Model of Caregiving (SPM)
The stress process model of caregiving (SPM) was used to guide the development of the
analytic framework in this dissertation study. The SPM is a widely accepted theoretical model in
family caregiving research that incorporates factors related to caregiving stress and
comprehensively describes the stress process from care tasks to care burden, as well as other
psychological, social, and clinical health outcomes among family caregivers (Pearlin et al.,
1990). This model views caregiving stress as a primary stressor that can lead to changes in
secondary strains such as daily roles, relationships with older adults, and interpersonal
interactions. As a result of reduced social and leisure engagement due to care responsibilities,
social isolation is a major secondary strain. Secondary strain may eventually lead to changes in
individual health outcomes as a mediator (Pearlin et al., 1990).
3.3.2 Caregiving Stress May Cause Social Isolation
Social isolation among older adults with care needs has been the focus of previous
research. There have been few studies on the social isolation of family caregivers. According to a
recent rapid review, there are only 12 peer-reviewed articles on social isolation of family
caregivers in English as of 2020 (Perez et al., 2021). Research supports the SPM-based premise
that taking on the role of caregiver will reduce individual’s social participation, and the level of
social isolation of caregivers is related to the functional limitations of older adults (Gallanis,
2019; Hayes et al., 2015; Li et al., 2020). In specific aspects of social participation, studies found
20
that family caregivers would feel increasing loneliness if their participation in meaningful
activities decreased due to caregiving tasks (Sun et al., 2019).
Notably, caregiving-related contextual factors have different effects on objective social
disconnectedness and subjective loneliness. Most previous research suggests that older adult care
demands and caregiver workload (e.g., caregiving tasks, hours) are linked to family caregivers'
objective social isolation, as they must sacrifice social time, or even work time, to assist older
adults (Gallanis, 2019; Robison et al., 2009; Sun et al., 2019). For example, family caregivers
who live together with the older adults are nearly 2.5 times more likely to experience social
isolation than those who do not co-reside with the older adults (Robison et al., 2009). However,
research that used scales to distinguish between social disconnectedness and loneliness
discovered that loneliness was not always linked to care demands or caregiver workload. For
example, one study found that the prediction power of caregiver workload on social isolation is
almost three times stronger than that on loneliness (Sun et al., 2019).
A possible explanation is that the stressors affect objective and subjective aspects of
social isolation of family caregivers in different ways. Objective stressors, such as older adults'
health status, caregiving needs, and caregiving workload, indicate older adults' dependence on
caregivers, which can have a direct impact on family caregivers' social engagement and lead to
social disconnection (Sun et al., 2019). Although care activities may reduce family caregivers'
normal social participation, interaction with other caregivers and positive feedback from older
adults (e.g., gratitude, sense of accomplishment) may be substitutes for social needs (Cheng et
al., 2016; Tarlow et al., 2004; Yu et al., 2018). Subjective stressors, which are the family
caregiver's subjective perceptions of the impact of caregiving activities on their own lives, are
more likely to influence perceived loneliness (van der Lee et al., 2014). A qualitative analysis
21
suggests that the source of loneliness among family caregivers may be powerlessness and
helplessness when facing caregiving responsibilities, in addition to reduced social interaction
(Vasileiou et al., 2017a). Therefore, both objective and subjective stress need to be considered
when modeling the impact of caregiving stress on family caregivers' social isolation.
3.3.3 Impact of Social Isolation on Family Caregivers
Many studies have looked into the effects of social isolation on family caregivers' health.
The majority of them claim that social isolation has a negative impact on caregivers' sleep,
mental health, and life satisfaction (Hajek & König, 2018, 2022; Kaschowitz & Brandt, 2017;
Zwar et al., 2018). In terms of specific aspects of social isolation, studies have found that socially
disconnected family caregivers have worse self-rated health, lower immunity, and more hospital
visits (dos Santos Azevedo et al., 2021; Kovaleva et al., 2018; Pohl et al., 2017; Segrin et al.,
2018). Subjective loneliness is also related to caregivers' lower quality of life, feelings of
helplessness, and more mental health symptoms (e.g., depression, anxiety) (Perez et al., 2021;
Segrin et al., 2018).
In addition to harming the health of family caregivers, social isolation can also affect
older adults. According to studies, family caregivers with severely reduced social participation
experienced mental health declines, such as depressive symptoms, lowering their sense of
accomplishment and satisfaction with caregiving and possibly leading to a decline in the quality
of relationships with older adults (Burgener & Twigg, 2002; Luth & Pristavec, 2020; Sim et al.,
2017). A decrease in job participation or unemployment will directly affect the family's financial
situation, making it impossible to ensure the quality of care (Home, 1998; Todd et al., 2020).
Older caregivers tend to have their own health issues, which may be worsened by participating in
long-term care activities and lacking social engagement (von Känel et al., 2019). In the end,
22
caregivers will be more likely to send older people to long-term care facilities like nursing
homes. Given the potential negative effects of social isolation on family caregivers and older
adults, as well as the importance of maintaining good social connections and social participation
on the quality and sustainability of family caregiving, it is critical to investigate a potential
pathway that connects caregiver stress and health outcomes, with social isolation serving as a
mediating factor.
3.3.4 The Longitudinal Essence of the Mediated Model
The available evidence is mostly based on cross-sectional studies, which, while finding
links between caregiver stress, social isolation, and a variety of health outcomes in family
caregivers, are insufficient to support the hypothesis that social isolation mediates the impact of
caregiving stress on health. The SPM also proposed the accumulation and proliferation of
caregiving stress and its impact on health over time (Li et al., 2020; Pearlin et al., 1990), which
emphasizes the importance of longitudinal research to uncover the mechanisms by which
caregiving stress affects caregivers. For example, researchers discovered that after a year of
follow-up, family caregivers reported a significant increase in perceived caregiving burden
(Oldenkamp et al., 2016). Duration of caregiving is also associated with mental health of family
caregivers, namely those who have long-term caregiving responsibilities tend to report more
depressive symptoms than those who only provide help occasionally (Kaschowitz & Brandt,
2017; Schulz & Williamson, 1991).
Long-term caregiving may also impair family caregivers' ability to remain socially
connected and increase their perceived loneliness. A longitudinal study found that family
caregivers of older adults have an increased risk of social isolation in the long-run (Li et al.,
23
2020). Several reasons may explain the increasing trend of social isolation among family
caregivers. First, the health status of older adults, such as chronic conditions, physical disability,
or cognitive impairment, may deteriorate over time, leading to increased care demands and
caregiving hours. Family caregivers will spend less time maintaining their own social
connections and more time supporting older adults. Second, due to the increased perceived care
burden or fatigue associated with long-term care responsibilities, caregivers are less willing to
spend limited time participating in social activities and prefer to rest alone. Third, becoming a
family caregiver is accompanied by a continuous modification of the caregiver's original
lifestyle, such as withdrawing from one's original social network or resigning or retiring early. It
is difficult to rejoin a social network once left, which is why some caregivers remain socially
isolated and lonely even after completing their caregiving roles (Hajek & König, 2022).
In sum, extant research on social isolation of family caregivers can be summarized as the
following three points:
• Social isolation is a composite concept that encompasses individuals' objective social
connection and participation in social activities, as well as their subjective sense of
loneliness.
• Existing measures on family caregivers' social isolation focus primarily on their objective
social disconnectedness, and a comprehensive assessment on their subjective and
objective social isolation remains lacking.
• The impact of caregiving stress on family caregivers' health outcomes may be mediated
by social isolation, and its specific mechanisms (such as the effects of subjective and
objective caregiver stress on family caregivers' social disconnectedness and loneliness)
are still being investigated.
24
The first research question is:
Will social isolation mediate the stress process between caregiving stress and mental
health in primary caregivers?
Taken together, social isolation may be an important secondary strain that mediates the
impact of caregiving stress on the mental health symptoms of family caregivers of older adults.
Using a longitudinal analytical model will help explore the mechanism of social isolation in the
stress process of caregiving and its health consequences more thoroughly, which has the
potential to provide evidence to facilitate development and implementation of interventions to
mitigate social isolation and protect health of family caregivers.
25
CHAPTER FOUR: SUPPORT FROM SECONDARY CAREGIVER NETWORK
Taking care of an older person is demanding and can have negative health consequences
for family caregivers. Exploring the risk factors that contribute to the caregiving burden will help
researchers better understand caregiving stress and its impact on family caregivers, which is
essential for developing effective interventions aimed at supporting family caregivers' health and
well-being in order to maintain high-quality caregiving. In addition to examining modifiable risk
factors, the buffering effects of family caregivers' available external supportive resources on the
negative impact of caregiving stress should be considered. The Stress Process Model of
caregiving (SPM) also suggests that social support as a moderator provides family caregivers
with coping resources to deal with primary stressors (e.g., caregiving stress), prevent or buffer
the development of secondary stressors (e.g., social isolation), and ultimately protect health and
well-being of family caregivers (Pearlin et al., 1990).
Many studies have been conducted to explore different types of social support received
by family caregivers, to develop measurements on it, and to understand its role in the caregiving
stress process (Haley et al., 1987; Han et al., 2012; Shiba et al., 2016). However, there is still a
lack of evidence on the effects of secondary caregiver support on the stress process of primary
caregivers. According to the previous chapters, older adults typically receive assistance from
multiple caregivers who can be classified as primary caregivers or secondary caregivers based on
their caregiving responsibilities, tasks, and time (Spillman et al., 2020; Staight & Marie Harvey,
1990; Tausig, 1992). This dissertation study focuses on primary caregivers, because they tend to
have the most caregiving responsibilities and are also most likely to be affected by caregiving
stress. Although secondary caregivers may also be involved in caregiving tasks, their perceived
caregiver burden and health impact may be lower than primary caregivers (Everhart et al., 2018;
26
Gonçalves-Pereira et al., 2020; Marino et al., 2020). Secondary caregiver support should be the
most direct and caregiving-related social support received by both primary caregivers and older
adults, so its potential protective effects on the health effects of caregiving stress should be
investigated. This chapter examines the literature on secondary caregivers in order to provide
theoretical and empirical support for this study, including how secondary caregivers are defined,
the types of support secondary caregivers provide, and the mechanisms by which secondary
caregivers provide support.
4.1 Define Secondary Caregiver Network
Given that an older adult receiving help from multiple family caregivers is well
documented, many researchers tend to use a network perspective to describe the overall help an
older adult receives. The most commonly used indicators to characterize the support of an
informal caregiver network are network size (i.e., how many informal caregivers are there), type
of care tasks (i.e., direct or indirect care), and caregiving hours (i.e., how many hours are
provided by all caregivers or per capita) (Feld et al., 2012; Tausig, 1992). Indicators such as the
proportion of women and the proportion of kin in the informal caregiver network are also used to
assess the quality of caregiving in the informal caregiver network (Andersson & Monin, 2017).
Considering all family caregivers as a network can reflect the level of support provided to older
adults, but it cannot reflect the interaction and mutual support between multiple caregivers.
Primary caregivers, in particular, are critical to this network because they are responsible for the
majority of the workload and are the most vulnerable to caregiving stress. Therefore, it is
necessary to separate primary caregivers from the informal caregiver network and explore the
impact of support from secondary caregivers on primary caregivers.
27
This study uses a network perspective to describe the helpers providing care to older
adults and assisting primary caregivers, the secondary caregiver network (SCN). Secondary
caregivers are still family caregivers, meaning they are the family, friends, or neighbors of older
adults and do not directly receive financial compensations for their help from the older adults or
their families. A controversial issue is how to decide whether a person can be considered a
secondary caregiver? A study classified secondary caregivers' involvement in caregiving into
three levels. First, secondary caregivers share caregiving tasks with primary caregivers and are as
deeply involved in caregiving activities as primary caregivers. Second, by providing emotional
or financial support, secondary caregivers are only indirectly involved in caregiving. Third-level
helpers are not directly involved in caregiving activities or provide indirect support to primary
caregivers or older adults; instead, they are only involved in some key decisions affecting older
adults and their families, such as nursing home placement (Penrod et al., 1995). In some later
research, the second and third levels of caregiving could be combined to form indirect care.
(Dilworth-Anderson et al., 1999).
While the above definition of secondary caregivers intends to characterize most informal
social support for older adults and primary caregivers, the broad definition of caregiving is easily
confused with other related concepts of social support. For example, people included in level 3
may not be identified as caregivers by older adults or themselves. It is very likely that all the
adult children of an older adult, even some of them do not live in a city and rarely see each other,
would be involved in key decision-makings. The second level of indirect support is defined as
two subtypes of social support, and emotional support is difficult to measure because secondary
caregivers can provide emotional support to both older adults and primary caregivers.
Furthermore, given the purpose of this study, which is to investigate the effects of secondary
28
caregiver support on reducing the negative effects of caregiving stress on primary caregiver
outcomes, and the lack of relevant data, this study uses a relatively strict definition, in which
only family caregivers who directly participate in caregiving activities are considered secondary
caregivers. Thus, SCN's assistance with caregiving tasks can directly relieve primary caregivers
of caregiving or household responsibilities, such as driving older adults to hospitals and assisting
with housecleaning and laundry (Huiying et al., 2019; Ogletree et al., 2019).
4.2 Previous Research on SCN support
Several studies have looked into involving secondary caregivers or multiple caregivers in
caregiving activities, but the results are mixed. Early studies, for example, suggested that having
secondary caregivers did not significantly reduce primary caregivers' care intensity or perceived
care burden, nor did it affect older adults' health or formal service use (Dilworth-Anderson et al.,
1999; Penrod et al., 1995; Tennstedt et al., 1989). Many recent studies have also found that
having secondary caregivers has no effect on primary caregiver burden, and may even worsen
primary caregiver quality of life, relationship satisfaction with older adults, depression, and
nursing home placement motivation (Allen et al., 2012; Everhart et al., 2018; Keim et al., 2021).
Several reasons may explain the prior inconclusive findings. First, secondary caregivers
were not clearly defined. Some studies defined secondary caregivers as family or friends who
provide emotional, financial, and task-related support for primary caregivers, which may confuse
SCN with the general social network of primary caregivers (Everhart et al., 2018; Park et al.,
2015). Second, the measurements might not accurately reflect SCN support. Many studies only
looked at one secondary caregiver and used a dichotomized variable to indicate whether or not
there was a secondary caregiver present (Allen et al., 2012; Khalaila, 2021). Types of care tasks
29
and caregiving hours of SCN were barely measured. Additionally, limited diversity of the study
samples diminished the representativeness of findings to the general population. Most of these
studies used small and convenience samples and predominantly focused on white caregivers (80-
90%).
4.3 The Moderation Effects of SCN Support in Family Caregiving
Notably, in addition to those studies that did not find a significant association between
support from secondary caregivers and caregiving outcomes, some found negative relationships
(e.g., worse quality of life, lower satisfaction on relationship with older adults, higher depression,
and increased motivation of nursing home placement). Because most of the evidence is based on
cross-sectional data, which can only reflect associations rather than causality, it is impossible to
prove that SCN support has a negative impact on primary caregiver health outcomes. In fact, it is
reasonable to assume that older adults with worse health and more care demands receive more
help from family caregivers in cross-section (Feld et al., 2012; Wysocki et al., 2009). Primary
caregivers, as the people who devote the most time and energy to the caregiver network, may
face more negative health consequences. According to the SPM, social support has a protective
effect on family caregivers' health by reducing the negative effects of caregiving stress on their
health and well-being (Pearlin et al., 1990). Therefore, it is possible that SCN support could
mitigate the impact of caregiving stress on primary caregivers' mental health over time.
Several studies that used longitudinal data or experimental design paradigms did report
that having secondary caregivers is associated with favorable outcomes of primary caregivers.
For example, having secondary caregivers makes primary caregivers more involved in
intervention programs than those who do not have secondary caregivers, and even experienced
30
more improvements after intervention, including less caregiver burden, perceived more positive
aspects of caregiving, and better mental health (Brand et al., 2016; Paúl et al., 2019). Also,
primary caregivers of older adults with eating disorders tend to experience lower depression and
anxiety if at least one secondary caregiver was presence during the caregiving of past year
(Stefanini et al., 2018). At the family level, more involvement of secondary caregivers can
improve relationship satisfaction and family functioning perceived by primary caregivers
(Wysocki et al., 2009). Therefore, secondary caregiver support has the potential to buffer the
mental health impacts of caregiving stress.
Nonetheless, the impact of caregiving stress on primary caregivers is a complex process,
and the buffering effects of SCN support can occur along any path, including reducing social
isolation. There were no studies that looked into the protective effect of SCN support on primary
caregivers' social isolation, according to the literature review. Objectively, primary caregivers
may become socially isolated as a result of their intensive caregiving work, such as reducing
social interaction and withdrawing from social activities, and this change may continue after the
cessation of caregiver role (Gallanis, 2019; Hajek & König, 2022; Hayes et al., 2015; Li et al.,
2020). The presence of additional caregivers could help to alleviate the situation. Because
secondary caregivers are an important force in primary caregivers' immediate support network,
some studies have used the presence of at least one secondary caregiver as an indicator of social
isolation of primary caregivers (Chamberlain et al., 2022). Secondary caregivers are typically
older adults' spouses, adult children, other relatives, or friends who are also part of the primary
caregivers' social network. Taking care of older people with secondary caregivers could be a
social interaction scenario, with primary caregivers receiving both practical and emotional
support from SCN. Despite the fact that studies show that even when all secondary caregivers are
31
combined, they provide less care to older adults than primary caregivers (Dilworth-Anderson et
al., 1999; Staight & Marie Harvey, 1990), the presence of secondary caregivers provides primary
caregivers with opportunities for respite, allowing them to engage in other social activities. The
SCN's long-term and consistent involvement in caregiving may aid primary caregivers in
maintaining a stable social connection and social participation.
Primary caregivers may have fewer responsibilities and workloads as a result of SCN
support, and thus have more energy to engage in other social activities and experience less
loneliness. Additionally, when primary caregivers collaborate with SCN, their perception of
being supported by family or friends can significantly reduce their loneliness (Rodrigues et al.,
2015). A community-based intervention study aimed at reducing the caregiving strain of primary
caregivers and improving their mental health shows that if primary caregivers participated with
secondary caregivers, the level of their perceived solitude were significantly lower than those of
primary caregivers who participated alone (Paúl et al., 2019). Therefore, there is a compelling
reason to support this research into the moderating effects of SCN support on the negative effects
of caregiving stress.
4.4 Potential Patterns of SCN Support
As mentioned earlier, past research may have some issues regarding the SCN support
measurement. For example, some studies only used a binary variable (the presence of secondary
caregivers) to reflect the level of SCN support (Allen et al., 2012; Khalaila, 2021). The level of
SCN support may vary by family due to the variety of care demands of older adults and support
needs of primary caregivers. Because of their relationship with older adults or primary
32
caregivers, mastery of caregiving skills, and geographic proximity to older adults, multiple
secondary caregivers' caregiving engagement can vary even within the same family.
To assess the level of support, previous studies used the care tasks/domains and care
intensity (hours per week/month) of each secondary caregiver. Multiple indicators indicate that
secondary caregiver support can take many forms. A high secondary caregiver support, for
example, could be a large number of secondary caregivers providing a lot of care, while a low
secondary caregiver support could be a small number of secondary caregivers providing limited
help. There may also be a pattern of mixed secondary caregiver support, which means that there
are multiple secondary caregivers but they only provide limited care, or there are a small number
of secondary caregivers but they provide a lot of care. The approach of integrating multiple
indicators to assess the overall level of SCN support can provide a comprehensive analysis of the
actual SCN support received by primary caregivers. It can also be used to investigate the stress-
reduction effects of various SCN support patterns on primary caregivers. Our literature search
found no studies that tested the combined construct, which integrated multiple indicators to
identify patterns of secondary caregiver support.
In summary, the extant evidence on SCN support can be summarized in three points:
• Previous studies lacked uniform standards for the definition and measurement of
secondary caregiver support, which could explain the mixed results.
• Although some cross-sectional research linked SCN support to worse outcomes for both
older adults and primary caregivers, it still has the potential to protect primary caregivers
from the social and psychological impacts of caregiving stress in the long run.
• There are multiple indicators for measuring SCN support, and combining them could
help us understand the various patterns of SCN support more thoroughly.
33
The second research question is:
Does SCN support moderate the social isolation mediated stress process of primary
caregivers, including:
• Would SCN support buffer the impacts of caregiving stress on the development of
social isolation in primary caregivers;
• Would SCN support buffer the impacts of social isolation on the mental health
symptoms of primary caregivers?
Therefore, this dissertation study only includes those family caregivers who directly
involved in caregiving activities into the SCN, and attempts to explore potential SCN support
patterns by integrating separate indicators, such as network size, types of caregiving tasks, and
caregiving intensity (e.g., hours per month). As an SCN that can provide direct emotional and
instrumental support to primary caregivers, its protective effect may take effect through social
isolation mediated stress process. That is, SCN support can prevent primary caregivers from
experiencing increased social isolation, and thus reduced the impacts of caregiving stress on
psychological health.
34
CHAPTER FIVE: ROLES OF GENDER, RACE, AND RELATIONSHIP TYPE
The literature review in the previous chapters illustrates that family caregivers who
provide substantial assistance with daily activities for older adults with long-term care needs, are
an important force in the long-term care system in society and the assurance of the quality of life
of older adults. However, family caregiving does not come without a price, as family caregivers
are exposed to chronic stress and may suffer other health and well-being consequences. Many
previous studies point out that the demographic characteristics of family caregivers such as
gender, race, and relationship type can have an impact their caregiving experience due to
differences in cultural values, socioeconomic status, and family structure represented by these
characteristics (Friedemann & Buckwalter, 2014; Heehyul et al., 2020; Hsiao, 2010). Given the
racial and cultural diversity of American society, any study of family caregiving that does not
fully account for participant differences in background may be biased. Therefore , this chapter's
literature review takes a diversity perspective, attempting to investigate differences in caregiving
experience and health outcomes among family caregivers with various demographic
characteristics (e.g., gender, race, relationship type).
5.1 Gender, Race, and Relationship Type on Family Caregiving
5.1.1 Gender: Women vs. Men
Women are the primary providers of family care, accounting for approximately two-
thirds of all family caregivers in the United States (National Alliance for Caregiving & AARP,
2020). There is extensive discussion about the reasons why family caregivers are predominantly
women. From a sociocultural perspective, the social image of women has been fixed for a long
time as nurturing children and caring for the family (Versey, 2017). Therefore, when family
35
members require care, women usually rank first to become caregivers. For example, when an
older man suffers from health problems, his wife is most likely to be his caregiver; when an older
woman has care needs, her adult daughter is most likely to be her primary caregiver. It is also
argued that women psychologically value their social connections more and tend to respond
more proactively when someone close to them has care needs (Zwar et al., 2020). On the
contrary, in general, men are more psychologically pursuing personal independence and tend to
keep their emotions private, thus less active in taking on the caregiving responsibility. This point
of view is essentially inseparable from the historical and social bias that shape the images of men
and women. That is, it attributes gender differences in attitudes toward family caregiving to
biological differences, while ignoring the impact of sociocultural values on individual
development. It is worth noting that the number of women entering the workforce is growing, as
is the number of men caregivers. Nonetheless, women will continue to be the majority of family
caregivers for a long time to come (Robinson et al., 2014).
In terms of caregiving experience, women family caregivers generally have higher
caregiving intensity than men, as evidenced by greater time commitment and participation in
more hands-on tasks, such as assisting with dressing and bathing (Fredman et al., 2019; Steven et
al., 2019). Women caregivers report higher caregiver burden, perceived caregiving-related
distress, and overload than men caregivers (Akpınar et al., 2011; Willert & Minnotte, 2019). The
immediate explanation is that women caregivers spend more time and energy to caregiving than
men, making them more susceptible to caregiver burden. Internalized sociocultural stereotypes
that women are good at caring force them to link care quality to self-efficacy, and any negative
feedback can add to their psychological stress (Versey, 2017). In regards to health consequences,
women caregivers report higher prevalence of mental problems, such as depression and anxiety
36
than men (Akpınar et al., 2011; Penning & Wu, 2016). Impacts were also found on psychological
distress, family conflicts, guilt, quality of life, and coping ability, on which women are more
vulnerable (Ducharme et al., 2011; Losada et al., 2010; Takai et al., 2011). Additionally, women
caregivers are more vulnerable to physical impacts of caregiving, including poor sleep quality,
decreased immune function, and increased comorbidity (Mills et al., 2009a; Pöysti et al., 2012).
5.1.2 Race: Black vs. White
Numerous studies have found significant racial/ethnic disparities in caregiving
experiences. Because Black families are more culturally respectful of older adults' wisdom than
White families, they may be more willing to care for older relatives with long-term care needs
(Rozario & DeRienzis, 2008). Many African Americans live in extended families, which include
relatives other than their parents or partners. Black older adults are also more likely than White
older adults to be assisted by family caregivers due to their larger family size (Falzarano et al.,
2021). Some researchers believe that one of the reasons why Black caregivers experience less
emotional distress than White caregivers is because they have a larger caregiver network
(Dilworth-Anderson et al., 1999, 2002). In addition, long-standing racial inequities in American
society, such as lower socioeconomic status and fewer available formal resources than Whites,
may have also catalyzed the family care traditions in Blacks. For example, the percentage of
Black households with annual income below 100% of the federal poverty level is almost three
times that of White households (Fabius et al., 2020). It is worth noting that many non-social
science fields continue to use predominantly White samples, while research on Blacks has
largely relied on small convenience samples, emphasizing the importance of using nationally
representative samples to investigate racial disparities in caregiving experience.
37
According to the extant evidence, Black family caregivers provide more care than
Whites. Black caregivers, in particular, spend significantly more time and perform significantly
more tasks than White caregivers. Compared to 38 percent of White caregivers, more than half
of Black caregivers assist older adults for more than 40 hours per week (Fabius et al., 2020).
Despite the fact that many studies show that Black caregivers perceive lower emotional burden
and higher satisfaction with their relationship with older adults, they are more likely to
experience physical and financial strains than White caregivers (Rosenthal Gelman et al., 2014;
Willert & Minnotte, 2019). Regarding the health effects of caregiving stress, Black caregivers
have worse physical health and more unmet needs for supportive services, which may be related
to their higher intensity of caregiving activities and greater financial strain (Black et al., 2013;
Pinquart & Sörensen, 2005). In addition, Black caregivers are also found to have a higher risk for
depression compared to White counterparts (Pinquart & Sörensen, 2005; Sörensen et al., 2010).
Overall, while Black caregivers are more likely to appraise caregiving positively and perceive
benefits from it, their caregiving experience is less favorable than that of White caregivers, and
they are more likely to suffer health consequences as a result of their caregiving.
5.1.3 Relationship Type: Spouse vs. Adult Child
Relationship type, as a key demographic factor, has been found to have significant
impacts on caregiving experience and quality of care. Close relatives, such as spouses and adult
children, are most likely to become family caregivers for older adults. According to surveys,
about 70% of older adults in the United States have primary caregivers who are their spouses
(20%) or adult children (55 %) (National Alliance for Caregiving & AARP, 2020). Despite the
fact that spouses and adult children are both close relatives, their experiences with caregiving
may be vastly different. Research on family caregivers with various relationships with older
38
adults will aid in understanding differences in caregiving experiences, such as challenges and
needs, and will provide evidence to improve social work practices aimed at meeting the needs of
family caregivers with various contextual characteristics (Dilworth-Anderson & Cohen, 2010).
As family caregivers of older adults, spousal caregivers tend to be older than adult
children caregivers. Because older age is associated with worse health status and lower resilience
in the face of caregiving stress, all these factors above contribute to spousal caregivers having
more time to provide care to older adults and may perceive higher levels of caregiver burden,
particularly in physical aspects (Galvin et al., 2010; Hong & Kim, 2008). Spousal caregivers are
also more likely to co-reside with older adults (90% vs. 34%) because of their marital
relationship with them, allowing spousal caregivers to assist older adults at night. Marriage
creates a long-term relationship commitment that makes spouse caregiving socially and morally
acceptable (Elizabeth & Rickenbach, 2019).
However, studies comparing the caregiver burden between spousal caregiver and adult
child caregiver are inconclusive. Some studies report that spousal caregivers tend to report higher
level of burden than adult children (Galvin et al., 2010; Jinyu, 2021; Oldenkamp et al., 2016).
While some other studies show the reverse (Chappell et al., 2014; Ott et al., 2007; Rigby et al.,
2019). It may be that there are many factors that affect the perceived caregiver burden, and
previous studies cannot comprehensively consider the impact of all correlates on the caregiver
burden. For example, despite spousal caregivers provided more intensive caregiving to older
adults, such as more IADL tasks, they are more likely to perceive positive aspects of caregiving
than adult child caregivers, which are thought to reduce the emotional burden of spousal
caregivers (Polenick & DePasquale, 2018). Given that adult child caregivers' caregiving
responsibilities are not derived from marriage vows, the caregiving commitment and quality of
39
care are pertinent with the quality of previous relationships between adult child caregivers and
older adults. Both good relationship quality and secure attachment styles are thought to protect
adult child caregivers' health and well-being from the negative effects of caregiving stress
(Karantzas et al., 2010; Ward-Griffin et al., 2007).
Furthermore, spousal caregivers are often retired, and their social connections may be
less complex than those of adult child caregivers. According to role theory, adult children's
multiple social roles, such as caregiver, child, parent, and colleague, may cause them to
experience role conflicts when balancing life and work. Adult child caregivers, for example, may
feel sandwiched, meaning they must care for their parents while also nurturing their children (Do
et al., 2014; Miller, 1981). Nevertheless, there are some studies suggesting that multiple social
roles of adult children may also play a protective role to their psychological health (Mengting et
al., 2019; Rigby et al., 2019; Tolkacheva et al., 2011). One explanation is that adult children's
social connections may provide them with more social resources, which, as important coping
resources, can help reduce caregiver burden. Adult child caregivers, for example, receive more
informal support from other family members or friends than spousal caregivers, and they also
use more formal support services, such as adult day care (Karen et al., 2005).
5.2 The Social Isolation Mediated Model May Vary by Gender, Race, or Relationship Type
According to the above literature review, gender, race, and relationship type all have the
potential to affect the caregiving experience of family caregivers, and previous studies proposed
a variety of explanations for the variance of caregiving experience between family caregivers
with different background identities. However, most of these explanations take a relatively
macro perspective to describe the social cultures, systemic inequalities, and family cultures
40
represented by these demographic characteristics, and only a few analyzed the mechanisms that
cause differences in caregiving experience from a micro/individual perspective. Therefore, the
purpose of this study is to see if there are any differences in the social isolation mediated stress
process model among primary caregivers of various genders, races, and relationship types. That
is, whether the various effects of caregiving stress on mental health are the result of its various
effects on primary caregivers' social connections. Individual research evidence can support the
development and improvement of community-based practices aimed at assisting and protecting
socially isolated family caregivers' health and well-being.
When it comes to the impact of caregiving stress on social aspects, a large body of
evidence suggests that women family caregivers are more vulnerable than men. Women who
care for their families are more likely to have fewer social connections, participate in fewer
social activities, and feel more lonely (Friedemann & Buckwalter, 2014; Hsiao, 2010; Versey,
2017). Researchers believe that women caregivers have a stronger sense of responsibility than
men in the role of caregiver, so they will devote more energy to caregiving tasks to ensure the
quality of care, even if it means sacrificing some social connections (Allen et al., 1999). In
addition, women caregivers are also more likely to be impacted by caregiving stress in terms of
labor force participation, such as reduced working hours or unemployment (Schulz, 2020).
Working-place communication is an important part of an individual's social connection, and
losing a job results in a significant reduction in social connections. Unemployment causes
financial strain as well. Family caregivers may experience increased anxiety and cut back on
unnecessary family expenses, such as those related to social activities (Hussain et al., 2018; Neri
et al., 2012b). Although studies also report that men family caregivers experience significant
social isolation after taking on the caregiver roles, because men are not as socially prepared for
41
the responsibility as their women counterparts (Buchanan & Huang, 2012). Longitudinal
evidence suggests that social disconnection and loneliness are declining over time in men
caregivers, while women caregivers remain at relatively high levels (Perez et al., 2021).
There is little research on racial differences in the effects of caregiving stress on the
social isolation of family caregivers. Many European studies, such as those conducted in the
United Kingdom and Germany, have found that White family caregivers may experience high
levels of social isolation and loneliness (Hajek & König, 2016; Quirke et al., 2019; Zwar et al.,
2018). However, studies also point out that while White family caregivers experience social
isolation, they also experience an expansion of their core social network by reporting an
increased number of important people (Quirke et al., 2019). Caregiving stress may not have a
significant impact on Black family caregivers' social connections. Blacks have higher familism
values than Whites, and they tend to live in larger households, which allows them to maintain
relationships with other family members after becoming family caregivers (Cudjoe et al., 2020).
Religious praying is an important way for individuals, especially for Black people, to reduce
stress. Black family caregivers are more likely to maintain a level of social engagement by
attending weekly church services and interacting with other church members (Ong et al., 2020).
Although Black family caregivers are less likely than Whites to experience social isolation, the
psychological effects of social isolation can be severe. According to one study, Black family
caregivers who reported high levels of social isolation and perceived loneliness were more likely
than White family caregivers to be depressed (Gallagher & Wetherell, 2020).
There are no conclusive studies comparing social isolation in spousal and adult child
caregivers. Spousal caregivers reported higher levels of social isolation than adult child
caregivers in studies on family caregivers of older adults with stroke and dementia (Oldenkamp
42
et al., 2016; Rigby et al., 2019). Due to their advanced age, deteriorating health, and increased
caregiving activities, spousal caregivers are more likely to see a reduction in their social network.
Furthermore, spousal caregivers tend to receive less social support than adult children and have
less access to both informal and formal support (Karen et al., 2005). However, some studies
point out that because most adult child caregivers do not live with older adults, their social
connection is less affected. When adult children live with older parents, they may experience the
same level of isolation as spousal caregivers, or even higher. For example, spousal caregivers
tend to report fewer interruptions in social activities or social contact than adult child caregivers
(Cho et al., 2009). Another study also found that adult child caregivers who co-reside with older
adults tend to report more limitations on their personal and social life (Kang, 2008). Therefore, it
remains to be tested whether the social isolation mediated stress process differs depending on the
relationship type.
5.3 The Moderation Effects of SCN Support May Vary by Gender, Race, or Relationship Type
The extent to which primary caregivers of older adults received support from secondary
caregivers, the level of SCN engagement, and the protective effect of SCN support on primary
caregivers may all vary depending on the sociodemographic characteristics of primary
caregivers. The majority of previous research focused on determining whether SCN can reduce
primary caregiver burden and comparing primary and secondary caregivers' perceived distress
(Everhart et al., 2018; Gonçalves-Pereira et al., 2020; Marino et al., 2020; Staight & Marie
Harvey, 1990), and very few studies examined differences in the acceptance of SCN support by
gender, race, and relationship type. Examining these differences helps understand the interaction
and collaboration between primary caregivers and other family members or friends from a family
43
perspective, and provide evidence for the development of community-based interventions aimed
at promoting cooperation between family caregivers and enhancing quality of care.
Women primary caregivers tend to have a lower socioeconomic status than men, are at
higher risk of being unemployed due to caregiving, and use less formal support services
(Scharlach et al., 2006; Song et al., 2011). Generally, women may value their social connections
more and rely more on their informal social network than men to protect their mental health
(Kelley et al., 2017). Therefore, support from SCN can be an important source of instrumental
and emotional support for women primary caregivers. A social network analysis shows that the
social network composition of women family caregivers is more homogeneous than that of men
family caregivers, such as coming from the same family, and are more likely to provide
caregiving-related assistance to the women caregiver (Rodríguez-Madrid et al., 2018). Black
families have fewer accessible community support services than White families, and because of
their stronger family caregiving tradition and larger household size, Black families may form
larger caregiver networks (Dilworth-Anderson & Gibson, 2002; Wallace et al., 2002). Hence,
Black primary caregivers may receive more support from SCN than Whites, and are more likely
to benefit psychologically from family cooperation. Several studies report that spousal caregivers
receive less both formal and informal support than adult child caregivers (Kang, 2008; Mengting
et al., 2019; Tennstedt et al., 1989). So spousal caregivers may have smaller SCN size and
receive less SCN support. However, there is no direct evidence implies that the buffering effects
of SCN support differ between spousal and adult child caregivers. Despite the less SCN support
of spousal caregivers may be objectively associated with their own smaller social network and
poorer financial status, spousal caregivers may be subjectively more reluctant to seek support
from their adult children (Elizabeth & Rickenbach, 2019).
44
5.4 The Potential Intersections of Gender, Race, and Relationship Type
Although existing research suggests that gender, race, and relationship type influence
family caregivers' caregiving experiences, there is a lack of solid rationale and empirical
evidence to explain how the social isolation mediated stress process differs depending on gender,
race, and relationship type. Black family caregivers, for example, provide more support to older
adults with dementia or other medical conditions while reporting lower intrapsychic strains than
White family caregivers (Fabius et al., 2020). Women caregivers were found to have more care
intensity and mental distress (Penning & Wu, 2016; Uccheddu et al., 2019). But other studies
also found no difference between men and women in terms of care workload and perceived
burden (Akpınar et al., 2011; Willert & Minnotte, 2019), and even suggest that men caregivers
have higher risk for illness (Mills et al., 2009b). Some researchers have suggested that the gender
difference in care experience may be related to relationship types (e.g., spousal and adult-child
caregiver) (Friedemann & Buckwalter, 2014). The inconclusive findings suggest a need for
further exploration on the intersection between gender, race, and relationship type.
5.4.1 The intersectional approach
To understand how various social roles rely on each other for meaning, intersectionality
can be used to investigate multiple categories or social groups that an individual holds
simultaneously (Crenshaw, 1989). The concept of intersectionality was proposed as a way to
understand how mainstream culture discriminates against marginalized groups by combining
perspectives. Many studies have adopted an intersectional perspective to explore the interactions
of various disadvantage statuses that may or may not lead to negative consequences for the
individual, and it is becoming increasingly popular in social and health research (Chappell et al.,
45
2015; Versey, 2017). This study adopts an intersectional approach to understand the interactions
of gender, race, and relationship type, to identify variability of the combined effects on the
experience of family caregivers.
5.4.2 Research on intersectionality in family caregiving
Some research has looked at the combined effects of these factors on the caregiver
experience. Adult son caregivers, for example, reported the highest care burden and the worst
mental health, whereas husband caregivers reported the lowest care burden and the best mental
health (Conde-Sala et al., 2010). The interaction between race and gender is also found in the
caregiver burden. Black women have significantly higher care intensity than White men (Steven
et al., 2019), and being caregivers may have greater impact on depressive symptoms of Black
men than it does on White men (Mair, 2010).
Several studies have examined whether the stress process of family caregivers varies
depending on sociodemographic factors. Specifically, whether caregivers of different genders,
races, and relationship types would suffer different mental health consequences as a result of
differences in the impact of caregiving on social networks. Given Black women caregivers tend
to undertake heavier care workload than White men (Haley et al., 2010; Hinojosa et al., 2015), it
is possible that they would experience more social isolation and depression. However, given that
Blacks have a stronger tradition of family caregiving than Whites (Fabius et al., 2020; Willert &
Minnotte, 2019), the involvement of multiple relatives may challenge our conjecture that Blacks
are more likely to be socially isolated. Moreover, a longitudinal study found that men caregivers
have a higher tendency to report social isolation (Li et al., 2020). Thus, social support must be
considered to understand the intersectionality in the caregiving stress process.
46
In addition, SCN support may be useful in explaining the various effects of caregiving on
primary caregivers' social isolation and well-being. Black women caregivers, for example, may
receive more informal social support, maintain a higher level of social participation, and
experience less emotional distress than White women caregivers (Fabius et al., 2020; Willert &
Minnotte, 2019). Wife caregivers were found to receive less help and support from other family
members or friends than husband caregivers, and they are more likely to disconnect from social
network and be psychologically affected (Ornstein et al., 2019; Ott et al., 2007). As far, the
intersectional effects of different sociodemographic factors (gender, race, and relationship type)
on the caregiving experience and stress process of family caregivers are still largely unknown.
Therefore, it is necessary to take an intersectional perspective to examine whether the
combination of two disadvantaged sociodemographic identities has a greater negative impact on
primary caregivers' social connections and mental health.
In sum, extant evidence on differences in the caregiving experience of primary caregivers
of different genders, races, and relationship types is summarized as the following two points:
• Sociodemographic characteristics such as gender, race, and relationship type may
influence the caregiving experience of primary caregivers, including their social life and
mental health.
• Many previous studies that looked at the impact of only one demographic factor on
caregiving experience failed to reach consistent conclusions, suggesting using an
intersectional approach to assess the impact of multiple contextual factors altogether.
Therefore, gender, race, and relationship type are important factors that influence family
caregivers' caregiving experiences and health outcomes. In order to investigate the mechanism of
47
the effect of caregiving stress on the mental health symptoms of primary caregivers, the study
must consider the model's comparability in various sociodemographic contexts. Researchers can
use an intersectional perspective to look beyond the traditional research paradigm of
sociodemographic characteristics as single factors and investigate the combined effects of two or
more factors on caregiving experience and health outcomes.
The last research question of the study is two fold:
a) How will the social isolation mediated stress process vary by the gender, race, and
relationship type of primary caregivers?
b) How will the moderation effect of SCN support vary by the gender, race, and
relationship type of primary caregivers?
48
CHAPTER SIX: CONCEPTUAL FRAMEWORK AND RESEARCH AIMS
The dissertation research is based on the stress process model (SPM) of caregiving to
build a mediated model that describes the impact of caregiving stress on mental health symptoms
of family caregivers, and to examine the moderation effect of SCN (secondary caregiver
network) support on the stress process. An intersectional perspective is adopted to examine the
variability of the stress process of caregiving under various sociodemographic backgrounds (i.e.,
gender, race, and relationship type). The measurement of mental health symptoms in this study
focuses on the two negative affective states of depression and anxiety. Depression and anxiety
are two common psychiatric symptoms among family caregivers (Crespo et al., 2005; S. Liu et
al., 2017). While most family caregivers only have minor symptoms, allowing mental illness to
progress may increase suicidal ideation and reduce the quality of care (Matthews et al., 2004).
Therefore , it is necessary to consider depression and anxiety when studying the mental health of
family caregivers.
Figure 1 presents the conceptual framework of the study.
Figure 1 The Conceptual Framework of the Study
49
Aim 1 (in red) is to examine whether social isolation will mediate the stress process of
primary caregivers from caregiving stress to mental health symptoms. The hypotheses are:
(H1a) primary caregivers who experience higher caregiving stress are more likely to be
socially isolated;
(H1b) primary caregivers with higher level of social isolation tend to report more
depressive symptoms.
(H1c) primary caregivers with higher level of social isolation tend to report more anxiety
symptoms.
Aim 2 (in green) is to build a moderated mediation model to examine whether SCN
support has a protective effect on primary caregivers, including to prevent the increase of social
isolation and decrease of mental health. The SPM proposed that informal support from other
relatives can prevent or reduce primary caregivers from disconnecting from their social network,
as well as protecting their physical and mental health from the harmful impacts of caregiving
stress (Pearlin et al., 1990). The moderation effects of SCN support will be examined on the
three pathways:
(H2a) SCN support can buffer the path from caregiving stress to social isolation, that is,
the impact of caregiving stress on social isolation of primary caregivers will decrease
with the increase of SCN support;
(H2b) SCN support can buffer the path from social isolation to depressive symptoms,
that is, the impact of caregiving stress on depressive symptoms of primary caregivers will
decrease with the increase of SCN support;
50
(H2c) SCN support can buffer the path from social isolation to anxiety symptoms, that is,
the impact of caregiving stress on anxiety symptoms of primary caregivers will decrease
with the increase of SCN support;
(H2d) SCN support can buffer the path from caregiving stress to depressive symptoms,
that is, the impact of social isolation on the depressive symptoms of primary caregivers
would decrease with the increase of secondary caregiver support.
(H2e) SCN support can buffer the path from caregiving stress to anxiety symptoms, that
is, the impact of social isolation on the anxiety symptoms of primary caregivers would
decrease with the increase of secondary caregiver support.
Aim 3 (in blue) is to explore the variability of the social isolation mediated stress process
of caregiving and the moderation effect of SCN support under different genders (women vs.
men), races (Black vs. White), and relationship types (spouse vs. adult child). Based on the
literature review, the impact of caregiving stress on social isolation is expected to be stronger for
women, Whites, and spousal caregivers, but the impact of social isolation on mental health
symptoms may be greater among Black caregivers due to their stronger emphasis on family and
less availability and accessibility to formal support services than White caregivers (Steven et al.,
2019; Willert & Minnotte, 2019). Similarly, SCN support may be more beneficial to women and
Black caregivers. Due to a lack of robust theoretical rationale and empirical evidence,
differences in caregiving experience and SCN support between spousal caregivers and adult
child caregivers have not been hypothesized. In addition, if the mediated and moderation models
in this study differ significantly by two or more background factors, intersectionality analyses
will be performed to look at the potential combined effects of the two or three background
51
factors. Black women caregivers, for example, may feel the most isolated and have the lowest
mental health than primary caregivers of other genders and races Given the exploratory nature of
this part of the study, the analyses will be based on the two research questions:
(a) how primary caregivers of different gender, race, and relationship type experience the
social isolation mediated stress process?
(b) how SCN support moderates stress process of primary caregivers of different gender,
race, and relationship type?
52
CHAPTER SEVEN: METHODS
7.1 Data Source
The research used secondary data from the National Study of Caregiving (NSOC), a
supplementary study to the National Health and Aging Trends Study (NHATS). The NHATS
surveyed Medicare beneficiaries aged 65 and up about their health status, living conditions, and
support environment (Kasper & Freedman, 2014). Each NHATS participant identified people
who assisted them with mobility, self-care, or household activities and provided information
about the tasks and intensity of the assistance. The NSOC team then contacted the caregivers
who had been nominated to conduct interviews to obtain detailed information about caregiving
as well as personal, social, and health characteristics of caregivers (Freedman et al., 2019). The
information of caregivers in NSOC was linked to older adults in NHATS using the participant
ID. NHATS interviews have been held every year since 2011, but NSOC interviews have only
been held three times in 2011, 2015, and 2017. In 2017, the third wave of the NSOC published a
longitudinal follow-up with caregivers identified in 2015, allowing for a longitudinal analysis on
family caregivers using two waves of data.
7.2 Inclusion and Exclusion Criteria
The caregiver ID was linked to NHATS participants after merging the 2015 and 2017
NSOC data to identify older adults who met the inclusion criteria: (1) live in the community, (2)
used no proxy for the interview, and (3) no missing data on key variables like functional
limitations and cognitive status. The primary caregivers in the longitudinal sample were then
identified using the following criteria: (1) nominated by older adults as primary caregiver; (2)
self-identified as non-Hispanic White or Black; (3) no missing data on key variables such as
perceived care burden, social isolation, and mental health. Finally, the information of eligible
53
secondary caregivers was extracted if they are (1) unpaid caregivers, (2) have regular care
schedule (hours per week or month), and (3) having no missing data on key variables, such as
tasks that they helped with.
7.3 Sample Size and Characteristics
The analytical sample includes 782 primary caregivers. Correspondingly, the health
information of 782 older adults and caregiving-related information of 1003 secondary caregivers
were obtained from NHATS data.
Table 1 shows the sociodemographics of the 782 primary caregivers in 2015. Their
average age is 64.6 years old, and about half of them are younger than 65 years (48%); most are
women (67%). Two thirds of primary caregivers self-identified as non-Hispanic White (66%).
More than 85% of primary caregivers finished high school or have a higher degree. Spousal
caregiver accounted for 35% of the total, and adult child caregiver accounted for nearly 51%.
Table 1 The sociodemographics of primary caregivers in 2015 (N = 782)
f % Mean SD
Age
64.63 12.76
< 55 165 21.32%
55 - 64 208 26.87%
65 - 74 210 27.13%
75 - 191 24.68%
Gender Man 256 32.74%
Woman 526 67.26%
Race/Ethnicity Non-Hispanic White 518 66.24%
Non-Hispanic Black 264 33.76%
Education Below high school 108 13.81%
High school no college 397 50.77%
Bachelor and above 277 35.42%
Marital Status Married/with partner 501 64.07%
Unmarried 281 35.93%
Relationship Type Spouse/Partner 272 34.78%
Adult child 396 50.64%
Others 114 14.58%
Co-residence No 287 36.70%
Yes 495 63.30%
54
Table 2 presents the sociodemographics of 782 older adults who received caregiving in
2015. Nearly 80% of the older adults were over 75 years old. Women accounted for 67%. Older
adults were less educated than primary caregivers, but still 67% of them had a high school
diploma or higher.
Table 2 Sociodemographics of Older Adults in 2015 (N = 782)
f %
Age 65 - 74 157 20.08%
75 - 84 312 39.90%
85 - 313 40.03%
gender Men 257 32.86%
Women 525 67.14%
Race/Ethnicity Non-Hispanic White 510 65.22%
Non-Hispanic Black 258 32.99%
Education Below high school 255 32.61%
High school no College 378 48.34%
Bachelor and above 147 18.80%
Marital Status Married/with partner 335 42.84%
Unmarried 447 57.16%
7.4 Measures
Mental health symptoms: depression and anxiety symptoms
NSOC uses two short scales to assess depression and anxiety symptoms of caregivers
(Patient Health questionnaire-2 [PHQ-2] and Generalized Anxiety Disorder scale-2 [GAD-2]).
Both scales are widely used in public and mental health study (Monahan et al., 2009; Spitzer et
al., 2006). The items of the two scales are from two longer scales on depression and anxiety
(PHQ-9 and GAD-7), and are considered to have the advantages of shorter space, time saving,
and efficient screening (Jordan et al., 2017; Manea et al., 2012). The PHQ-2 asks how often
during the last month the respondent “had little interest or pleasure in doing things” and “felt
down, depressed, or hopeless”; while the GAD-2 asks how often the respondent “felt nervous,
55
anxious, or on edge” and “has been unable to stop or control worrying”. Each item of the two
scales was measured from 1 (not at all) to 4 (nearly every day) with a total score ranges from 2 to
8. For both PHQ-2 and GAD-2, higher scores indicate more severe depression and anxiety.
Caregiving stress: objective and subjective stressors
Two indicators were used to build a latent factor to reflect the objective stress. The first is
functional limitations which consist of difficulty with activities of daily living (ADL) or
instrumental activities of daily living (IADL) and the help of caregivers in these activities
(Bertinato et al., 2012; Depalma et al., 2013; He et al., 2015). The ADL includes dressing,
bathing, eating, toileting, get out of bed, and walk inside/outside home. The IADL includes
doing laundry, shopping, preparing meals, banking, and tracking medications. A limitation was
defined as older adults’ report of receiving help with or having difficulty conducting any of the
12 activities by themselves. The second indicator of objective stress is probable/possible
dementia status of older adults. The NHATS provides three criteria to identify the cognitive
health and probable/possible dementia of older adults (Kasper et al., 2013). The three criteria are:
(1) dementia diagnosed by the doctor; (2) scored 2 or more on AD-8 dementia screening; and (3)
scored 3 or more on the cognitive test on memory and orientation.
The subjective stress are financial, emotional, and physical strains reported by primary
caregivers. Primary caregivers were asked “how financially/emotionally/physically difficult is
helping the care recipient?”. Each strain was measured on a scale from 1 (a little difficult) to 5
(very difficult). Cronbach’s α of the three items is 0.69 in wave 1 and 0.73 in wave 2, which are
acceptable. Confirmatory factor analysis indicates the three items are loading on a same latent
factor with factor loadings range from 0.72 to 0.80. The analysis used these three items of
perceived strain to construct a latent factor to reflect subjective stress of primary caregivers. Past
56
research found that caregivers of different racial/ethnic groups may experience caregiving
burden on these three dimensions differently even after controlling for their objective care
intensity, which may be explained by differences in culture, family structure, and socioeconomic
status (Carretero et al., 2009; Heehyul et al., 2020). Therefore, constructing latent variables in a
structural equation model preserves the potential inequality of factor loadings across subgroups
on the measure of subjective stress, which may improve model efficiency than just using the
composite score.
Social isolation
The measurement of social isolation of primary caregivers in the study is based on the
composite framework which includes objective social disconnectedness and subjective loneliness
(Cornwell & Waite, 2009; Wister et al., 2019). Objective social disconnectedness was calculated
using the following five NSOC items: having family/friends to talk with about important matters;
having family/friends to visit; church participation; club participation; and volunteer work. One
point was recorded for each item that has a negative (disconnected) response with a sum score
from 0 to 5 to represent the level of objective social disconnectedness.
The objective measurement was first proposed in 2017 to assess the social isolation of
older adults (Pohl et al., 2017, 2018). The initial version also includes whether older adults being
unmarried or living alone as the sixth item. Since this is the first time using the measurement on
family caregivers, the reliability and validity of the measurement were evaluated using both two
waves data. It was found that being unmarried has a low correlation with the rest items, and its
item-rest Cronbach’s α (Wave 1: 0.75; Wave 2: 0.76) is higher than item-included Cronbach’s α
(Wave 1: 0.74; Wave 2: 0.71). The confirmatory factor analysis results also show that the factor
loading of being unmarried at the same factor with other items is very low (<0.20). Two latent
57
measurement models were fitted and the model fit comparison indicates that taking out being
unmarried can lead to a significant model improvement. Experts’ opinions also suggest that
given one-third of the sample is spousal caregivers, unmarried status may not reflect the social
disconnectedness of this group of the participants. Therefore, the study dropped being unmarried
and only use the remaining five items to measure objective social disconnectedness of primary
caregivers.
A single item loneliness scale was used to capture the subjective domain of social
isolation. The primary caregivers were asked “how often did you feel lonely during the last
month?”, and the response was from 1 (never) to 5 (every day). The composite measure that
includes both objective social disconnectedness and subjective loneliness has good internal
consistency (Cronbach’s α = 0.74 in wave 1 & 0.75 in wave 2) and showed good convergent
validity in the measurement model with only one latent factor (Wave 1: RMSEA = 0.024, SRMR
= 0.024, CFI = 0.985, TLI = 0.971; Wave 2: RMSEA = 0.031, SRMR = 0.022, CFI = 0.995, TLI
= 0.989). The total score ranges from 1 to 10, with higher scores indicate higher level of social
isolation.
Support from secondary caregiver network (SCN)
Each NHATS respondent identified people who provide assistance in the four care
domains: mobility, self-care activities, household tasks, medical care, or transportation. The
relationship between each helper and older adults, the tasks to help, and the number of hours of
help provided per week/month are available in the NHATS data. A network perspective was
adopted to treat the secondary caregivers matched with each primary caregiver as a whole, and
three indicators were used to measure the help provided by the secondary caregiver network to
the primary caregiver, including the size of secondary caregiver network (number of secondary
58
caregivers), number of care domains that secondary caregivers helped with (1-4), and care
intensity of all secondary caregivers (hours/month).
Intersectional variables: race, gender, relationship type
Gender of primary caregivers is binary (women/men). Race and ethnicity were measured
first by race (White/Black/Native American/Asian/Hawaiian or Pacific Islander), and then by
ethnic Hispanic/Latino identity (yes/no). The study only included non-Hispanic white and black
primary caregivers, thus race is a binary variable too (black/white). Relationship type between
primary caregivers and older adults has three categories that are spouse, adult child, and others.
Other covariates
Across our analyses of social isolation, SCN support, and intersectionality, study also
adjusted for sociodemographic characteristics, health status, and caregiving-related contextual
factor of primary caregivers. The selection of covariates was based on the SPM (Pearlin et al.,
1990). Sociodemographics include age (years) and education level (< high school, high
school, >high school). Health status include primary caregivers’ self-rated health (1-5) and
number of chronic conditions (1-9: heart attack, other heart disease, hypertension, arthritis,
osteoporosis, diabetes, lung disease, stroke, and cancer). Caregiving-related contextual factor is
whether primary caregiver co-reside with the older adult (yes/no). An overview of the variables
and measurements in the proposed study is provided in Table 3.
59
Table 3 Overview of variables and measurements in the proposed study
Respondents
Variables Measures
Primary
caregiver
Older
adult
Secondary
caregiver
Measured
wave
Mental health
symptoms
PHQ-2 X
2015, 2017
GAD-2 X
2015, 2017
Caregiving stress Functional limitations
X
2015
Probable/possible
dementia
X
2015
Caregiving strain X
2015
Social isolation Social disconnectedness X
2015, 2017
Subjective loneliness X
2015, 2017
Secondary caregiver
support
No. of Secondary
caregivers
X 2015
No. of care domains
X 2015
Care intensity
X 2015
Other covariates Sociodemographics X
2015
Health status X
2015
Caregiving contextual
factor X 2015
7.5 Analysis
The STATA/SE version 16.1 was used for statistical analysis of the study and the test
level was treated at a bilateral α = 0.05. Descriptive statistics were calculated on key variables
and covariates of primary caregivers, older adults, and secondary caregiver networks: categorical
variables were presented in frequency and percentage, and continuous variables in mean and
standard deviation. For caregiving stressors that were introduced in the SEM as latent constructs,
the distributions of each item were presented. McNemar’s χ² and paired t-test were conducted to
compare the changes of the categorical and continuous key variables between the two waves
respectively.
For Aim 1, a two-wave mediation model was adopted to examine longitudinally if the
impact of caregiving stress on mental health symptoms would be mediated by social isolation of
primary caregivers (Preacher & Hayes, 2008; Sussman et al., 2006). A structural equation
60
modeling (SEM) procedure was used to build the mediated model. Two latent variables were
created to predict objective and subjective caregiving stressors, and two separated models were
fitted to check the changes in depression and anxiety symptoms of primary caregivers
respectively (Figure 2).
Figure 2 The two-wave mediation model of the study
In the models, the two independent variables (i.e., subjective and objective caregiving
stress) are constructed as latent variables. The three subjective caregiving stress items assessed
primary caregivers' physical, emotional, and financial strains while caring for older adults.
Previous research tends to look at the three items separately rather than combining them (R. Liu
et al., 2021; Rodrigues et al., 2015). The analysis of this study uses these three items to construct
a latent variable to release the constraints with equal loadings for each factor in order to have
better predictive power for subjective caregiving stress. The number of functional limitations and
the probable/possible dementia status of older adults are two items that contribute to objective
caregiving stress. Because the measurement scales of the two items are different, the results of
the two items cannot be added up, so latent construct is more appropriate.
61
The changes in primary caregivers' social isolation between waves 1 and 2, with a
positive value indicating that the degree of social isolation has increased. The reliability and
validity of the composite social isolation index were tested during the measurement development
phase, allowing us to use the total score without creating a latent variable. In structural analysis
models, the Stata program does not support arithmetic operations on latent variables.
Furthermore, using the score differences between the two waves eliminates the problem of using
only one wave of data as the mediator in a two-wave mediation model, thereby preserving the
predictive power of longitudinal data analysis.
Estimated SEM models examines only one outcome variable at a time. The models
directly use the total scores of depression and anxiety because the scales have been widely used
in previous studies, have good reliability and validity, and both have only two items (PHQ-2 and
GAD-2) (Monahan et al., 2009; Spitzer et al., 2006).
The direct effects were the impact of the two caregiving stressors of wave 1 on the
depression or anxiety of wave 2 (c1 & c2). The indirect effect refers to the impact of caregiving
stress of wave 1 to the depression or anxiety of wave 2 through the changes of social isolation
between wave 1 and 2 (a1*b & a2*b).
Aim 2 intends to examine whether support from SCN can moderate the social isolation
mediated stress process of primary caregivers. A latent profile analysis (LPA) was conducted to
examine the heterogeneity of the support from SCN based on the profile of means of the three
indicators (size, number of care domains, and care intensity per capita). The analysis chooses
care intensity per capita rather than total care intensity per SCN because (1) total care hours is
highly correlated with SCN size, (2) total care hours has large values and variances, which may
dominate the model estimating process of LPA and weaken contributions of the other two
62
variables, and (3) it was expected that mixed SCN support patterns exist (e.g., large SCN but
limited hours per secondary caregiver), so using care intensity per capita in LPA can better serve
this goal. LPA and LCA (latent class analysis) have the same theoretical basis, but the latter is
for discrete variables (Hickendorff et al., 2018). The three indicators related to secondary
caregiver support in this study are all continuous variables, so LPA is an appropriate approach.
Several subgroups which represent that different pattern of secondary caregiver support shall be
identified.
Multigroup SEM models were fitted to examine the moderation effects of SCN support
on the mediation model by estimating SEM models in Aim 1 of each subgroup and comparing
estimated pathway coefficients across groups. Given the models contained latent variables, the
cross-group consistency of latent measurement constructs was tested before performing
multigroup analysis. Three latent measurement models (factor loadings inequality, error variance
inequality, factor variance/covariance inequality) were compared with the baseline model (the
model without cross-group constraint on any parameters) to determine the constraints on the
latent variable in the multigroup SEM, and to ensure that differences in structural models derived
solely from differences in path coefficients.
Given that LPA is a person-centered approach which has the potential to identify typical
subgroups that reflects real-life situations that primary caregivers receive secondary caregiver
support. Compared to using each individual indicator of SCN support (i.e., SCN size, care
domains of SCN, and care intensity of SCN), LPA has the potential to identify various patterns
of the combination of the three indicators. Namely, some primary caregivers may receive a lot of
support from a single secondary caregiver, while others only receive limited assistance from a
large SCN. In addition, even if the SCN support received by all primary caregivers follows a
63
similar distribution trend on the three indicators, the subgroups identified by LPA still retain the
continuity of the three indicators, that is, the continuous variables are ordinally classified.
Therefore, LPA is more suitable for detecting moderation effects in analytical models than
directly using these three continuous or dichotomized indicators.
For Aim 3, analyses were conducted to examine if the social isolation mediated model
(Aim 1) and the moderation effect of SCN support (Aim 2) would vary by gender (women vs.
men), race (Black vs. White), and relationship type (spouse vs. adult child). Intersectionality
would be tested based on an exploratory procedure to first identify factors that are affecting the
models fitted in Aim 1 and 2, then combine two identified intersectional factors to divide the
sample into four separate groups, and use the multigroup approach of SEM to compare the
models of different subgroups. For example, if a model in Aim 1 is found to be different between
two gender and race groups, then four gender-race subgroups (Black women, Black men, White
women, White men) will be generated for the next analyses. Because the multigroup approach
tends to split the sample into subgroups with limited sizes and introduce too many estimation
parameters, which will lead to the risk of low model fit (Morin et al., 2015). The bootstrap
sampling approach would be used to assess the significance of the intersectional effect of race,
gender, and relationship type, if all three intersectional factors were found affecting the estimated
parameters at their respective levels.
All the SEM models will be constantly refined to verify model fit and the best suited
model will be selected. Indicators to evaluate model fit in the proposed study include the relative
chi-square (χ2/df) statistic, the root mean-square error of approximation (RMSEA), The
Comparative fit index (CFI), and the Tucker-Lewis index (TLI). Recommended criteria of these
indicators were χ2/df < 2, RMSEA < 0.05, CFI > 0.90, and TLI > 0.90 (Her et al., 2019).
64
CHAPTER EIGHT: RESULTS
8.1 Descriptive Characteristics of the Participants
The study included 782 primary caregivers, 782 older adults, and 1003 secondary
caregivers. Basic sociodemographics of participants (e.g., age, gender, race, education) have
been presented in the previous chapter (CHAPTER SEVEN: METHODS), this section will only
present descriptive statistics for key variables.
8.1.1 Primary Caregivers
Table 4 presents descriptive statistics of primary caregivers. For 782 primary caregivers,
their reported mean self-rated health was 3.41 (1.08) on a 1 to 5 scale in wave 1, which
decreased significantly in wave 2 reports (Mean = 3.33, sd = 1.07; t = 2.78, p < 0.01). On
average, each primary caregiver reported 1.86 (1.41) chronic conditions, of which the three
conditions with the highest prevalence were hypertension (54%), arthritis (47%), and diabetes
(22%). The prevalence of rest conditions was all below 15%.
In terms of subjectively perceived caregiving strain, primary caregivers reported an
average of 4.73 (2.56) on a 3 to 15 scale in wave 1, and it increased significantly to 5.02 (2.84) in
wave 2 (t = 3.22, p < 0.01). For each item, both emotional and physical difficulty increased
significantly, but financial difficulty did not change. The overall score for social isolation
increased from 4.02 (1.66) in wave 1 to 4.55 (1.64) in wave 2 (t = 7.63, p < .01). In structural
equation modeling, we used the score difference of social isolation between the two waves as the
mediator, which is 0.53 (1.79).
On a scale from 2 to 8, primary caregivers reported a mean depression score of 3.05
(1.33), which increased significantly to 3.45 (1.57) in wave 2 (t = 7.43, p < 0.001). A similar
65
situation was observed for the anxiety, which increased from 3.05 (1.35) in wave 1 to 3.16 (1.45)
in wave 2 (t = 2.22, p < 0.05).
66
Wave 1 (2015) Wave 2 (2017)
McNemar / paired t
f % Mean SD f % Mean SD
Self-rated health
3.41 1.08
3.33 1.07 2.78**
Chronic Conditions
1.86 1.41
Heart attack 77 9.85%
Other heart disease 84 10.74%
Hypertension 425 54.35%
Arthritis 369 47.19%
Osteoporosis 118 15.09%
Diabetes 169 21.61%
Lung disease 107 13.68%
Cancer 109 13.94%
Perceived difficulty
4.73 2.56
5.02 2.84 3.22**
Financial difficulty
1.41 0.99
1.40 1.02 0.23
Emotional difficulty
1.82 1.25
1.99 1.41 3.69***
Physical difficulty
1.50 1.11
1.62 1.18 2.75**
Social isolation
4.02 1.66
4.55 1.92 8.31***
Social disconnectedness
2.18 1.27
2.63 1.64 7.63***
No friend/family to talk 128 16.37%
244 31.20%
No visit to friend/family 170 21.74%
262 33.50%
No church participation 306 39.13%
388 49.62%
No club participation 516 65.98%
538 68.80%
No volunteering 587 75.06%
622 79.54%
Loneliness
1.83 0.94
1.92 1.04 2.51*
Depression
3.05 1.33
3.45 1.57 7.43***
Anxiety 3.05 1.35 3.16 1.45 2.22*
Note: * p < 0.05, ** p < 0.01, *** p < 0.001.
Table 4 Descriptive statistics of primary caregivers (N = 782)
67
8.1.2 Older Adults
Table 5 presents descriptive statistics of older adults. For 782 older adults, their average
self-rated health score was 2.64 (1.02) and 2.96 chronic conditions were reported. Except for
heart attack (10%) and cancer (14%), the prevalence of all other chronic conditions exceeded
20%, including hypertension (76%), arthritis (69%), diabetes (33%), other heart diseases (26%),
osteoporosis (26%), lung disease (21%), and stroke (21%).
Functional limitations and probable/possible dementia status are indicators for the latent
construct of the objective caregiving stressor. On 12 functional items, older adults reported an
average of 6.35 (3.07) limitations, and 39% of them had possible/probable dementia.
Table 5 Descriptive statistics of older adults in wave 1 (N = 782)
f % Mean SD
Self-rated health
2.64 1.02
Chronic Conditions
2.96 1.48
Heart attack 78 9.97%
Other heart disease 206 26.34%
Hypertension 593 75.83%
Arthritis 539 68.93%
Osteoporosis 206 26.34%
Diabetes 259 33.12%
Lung disease 164 20.97%
Stroke 161 20.59%
Cancer 112 14.32%
Functional Limitations
6.35 3.07
Probable/Possible Dementia (yes) 304 38.87%
8.1.3 Secondary Caregiver Networks
Table 6 shows the characteristics of secondary caregivers. Size indicates the number of
secondary caregivers matched by each primary caregiver. There are 228 primary caregivers
served as the sole caregiver for older adults, and 292 primary caregivers have one secondary
caregiver to form a double-caregiver team. There are also 262 primary caregivers received help
68
in care activities from two or more secondary caregivers. All care tasks were classified into four
types, including mobility and self-care, household activities, transportation, and medical care.
173 (31%) primary caregivers reported that their SCN was involved in only one care domain,
160 (29%) reported that SCN provided assistance on two domains, and 127 (23%) and 94 (17%)
SCN provided support for primary caregivers on three and four care domains, respectively. In
terms of care intensity (hours/month), each SCN contributed a total of 72.79 (64.33) hours of
care per month. Of these, 39% of SCN assisted less than 10 hours per month, 20% assisted 10-29
hours per month, 22% assisted 30-99 hours per month, and 19% provided more than 100 hours
of support per month to primary caregivers.
Given that the analysis will use latent profile analysis to treat these three SCN support
indicators as continuous variables, these three indicators should satisfy both the assumptions of
consistency and difference. Correlation analysis shows that these three indicators are correlated
with each other, and the correlation coefficients are less than 0.40 except the correlation between
SCN size and care intensity (r = 0.74, p < 0.001). The high correlation may hinder LPA from
identifying latent patterns of common contribution of the three indicators. So we used the SCN
care intensity per capita as the indicator in LPA. On average, the care intensity per secondary
caregiver was 37.71 (38.79) hours per month.
69
Table 6 The characteristics of secondary caregiver network (N = 1003)
f % Mean SD
Size
1.81 1.13
0 228 29.16%
1 292 37.34%
2 146 18.67%
3+ 116 14.83%
No. of Domains
2.26 1.08
1 173 31.23%
2 160 28.88%
3 127 22.92%
4 94 16.97%
Care Intensity Overall
72.79 64.33
(hours/month) Per secondary caregiver
37.71 38.79
1 - 9 214 38.63%
10 - 29 113 20.40%
30 - 99 120 21.66%
100 - 107 19.31%
8.2 AIM 1: The Mediated Effects of Social Isolation on the Process from Caregiving Stressors to
Caregiver Well-being
Structural equation modeling established two two-wave mediation models for depression
and anxiety, respectively. For both models, we used two latent variables (subjective stressor and
objective stressor) as independent variables, the score difference of social isolation as mediator,
and depression or anxiety as the dependent variable. Covariates were introduced into models to
adjust model explanatory efficacy on the mediator and dependent variables, including primary
caregivers' age, gender, race, relationship type, self-rated health, number of chronic conditions,
and coresidential status with older adults.
8.2.1 Depression Model
Table 7 presents the output of SEM on depression. The results of the two-wave mediation
model indicate that there is a mediated path linking objective caregiving stressor and depression
70
through social isolation (β = 0.18, se = 0.03, p < 0.001), suggesting that heavier objective
caregiving stress increases social isolation of primary caregivers, thereby aggravating their
depression. However, social isolation did not mediate the impact of subjective caregiving
stressor on depression (β = 0.04, se = 0.03, p = 0.18), because the increase of subjective
caregiving stress did not lead to worsening of social isolation (β = 0.16, se = 0.12, p = 0.17).
The mediated effect of social isolation does not fully explain the impacts of the two
caregiving stressors on the depression of primary caregivers. Subjective stressor in wave 1 was
directly related to depression in wave 2 (β = 0.32, se = 0.09, p < 0.001), as was objective stressor
(β = 0.21, se = 0.07, p < 0.01). Model fit indices shows that the structural equation model is good
(RMSEA = 0.028, SRMR = 0.020, CFI = 0.990, TLI = 0.980).
Table 7 The mediated model of social isolation between caregiving stressors and depression in primary
caregivers
β (se) p
Direct paths
Change of social isolation (Mediator)
Subjective stressor 0.16 (0.12) 0.17
Objective stressor 0.32 (0.09) < 0.001
Depression (Dependent variable)
Change of social isolation 0.26 (0.03) < 0.001
Subjective stressor 0.32 (0.09) < 0.001
Objective stressor 0.21 (0.07) 0.003
Indirect paths
SS
a
-- > CSI
b
-- > Depression 0.04 (0.03) 0.18
OS
c
-- > CSI -- > Depression 0.18 (0.03) < 0.001
Note: Model was adjusted for primary caregivers’ age, gender, race, relationship type, self-rated health,
number of chronic conditions, and coresidential status.
a
SS: Subjective Stressor
b
CSI: Change of Social Isolation
c
OS: Objective Stressor
71
Figure 3 The mediated model of social isolation between caregiving stressors and depression in primary
caregivers
8.2.2 Anxiety Model
Another two-wave mediation model was estimated to examine if social isolation would
mediate the impact of caregiving stressors on anxiety of primary caregivers (Table 8). Results
suggest that social isolation did not mediate the impact of neither subjective stressor (β = 0.01, se
= 0.08, p = 0.68) nor objective stressor (β = 0.27, se = 0.29, p = 0.39) on the anxiety level of
primary caregivers.
For each individual path, only objective stressor to social isolation is significant (β =
0.32, se = 0.09, p < 0.001), which is consistent with the results of the depression model. None of
the remaining paths are significant which suggests that neither subjective nor objective stressor
have impact on anxiety of primary caregivers.
72
Table 8 The mediated model of social isolation between caregiving stressors and anxiety in primary
caregivers
β (se) p
Direct paths
Change of social isolation (mediator)
Subjective stressor 0.31 (0.23) 0.18
Objective stressor 0.32 (0.09) < 0.001
Anxiety
Isolation change 0.07 (4.03) 0.98
Subjective stressor 0.53 (0.82) 0.06
Objective stressor -0.09 (0.72) 0.98
Indirect paths
SS
a
-- > CSI
b
-- > Anxiety 0.01 (0.08) 0.68
OS
c
-- > CSI -- > Anxiety 0.27 (0.29) 0.39
Note: Model was adjusted for primary caregivers’ age, gender, race, relationship type, self-rated health,
number of chronic conditions, and coresidential status.
a
SS: Subjective Stressor
b
CSI: Change of Social Isolation
c
OS: Objective Stressor
8.3 AIM 2: The Moderation Effects of SCN Support on the Social Isolation Mediated Caregiving
Stress Process
8.3.1 Latent Profile Analysis on SCN Support
Table 9 shows model fit indices of latent profile solutions for SCN support from two to
five classes. To help determine the best solution, six goodness-of-fit indices are presented. The
prediction error of each model is indicated by the AIC, BIC, and aBIC (sample size adjusted
BIC), with a smaller value indicating better model prediction quality. Entropy measures how
distinct identified subgroups are in each model, and a value greater than 0.8 indicates that
subgroup identification accuracy is greater than 90%. LMR-LRT and BLRT indicate the relative
adequacy of a (k-1)-class model compared to a k-class model. A significant p value indicates that
the k-class model is superior.
73
According to multiple indices, the changes in AIC and BIC show that the 2-class and 3-
class models are significantly better than the 4-class and 5-class models. The entropy of the four
models is greater than 0.9, indicating that they all have high subgroup identification accuracy.
The 3-class model has a significant improvement over the 2-class model in LMR-LRT and
BLRT, implying that the 3-class model has the best overall model fit. By comparing the
predicted probability and estimated variable means of each subgroup in different models, the 4-
class and 5-class models both were just splitting the last class on the basis of the 3-class model,
so their contribution to the improvement of model fit is limited. In sum, it was decided to choose
the 3-class model due to its best model fit, relatively parsimonious structure, and having a
sufficient and similar number of participants per class.
Table 9 Latent profile solutions for SCN support from two to five classes
In contrast to latent class analysis, which identifies latent subclasses represented by
several categorical variables, the classification result of latent profile analysis may be several
latent classes with qualitatively continuous attributes. By combining SCN size, domains, and
care intensity per capita, the three identified latent subgroups indicate three distinct SCN support
patterns. The results review confirmed that these three latent subgroups represent the conceptual
meaning of SCN support levels ranging from low to high, hence each group was given a
AIC BIC aBIC Entropy LMR-LRT
BLRT
(2*ΔLR)
2-class 7854 7897 7866 0.914 423.76 440.53***
3-class 7769 7829 7785 0.916 268.55** 279.17***
4-class 8110 8188 8131 0.926 190.38 197.91
5-class 8096 8191 8121 0.921 127.25* 132.29***
AIC: Akaike Information Criteria;
BIC: Bayesian Information Criteria;
aBIC: sample size adjusted BIC;
LMR-LRT: Lo-Mendell-Rubin adjusted likelihood ratio test;
BLRT: bootstrap likelihood ratio test;
*p < 0.05, **p < 0.01, ***p < 0.001.
74
shorthand label: low SCN support, medium SCN support, and high SCN support. Table 10
shows the predicted prevalence of the four SCN support subgroups (including one without SCN)
in the study sample and the descriptive statistics of each subgroup on the three SCN support
indicators.
SCN support subgroup 1 (no SCN): About 29% of older adults in the study have only one
caregiver to help with their daily activities. SCN support subgroup 2 (low SCN support): About
22% of primary caregivers received a low level of SCN support, which means low estimates on
all three indicators. On average, each primary caregiver reported 1.16 secondary caregivers who
assisted with 1.69 care domains and each secondary caregiver provided 12 hours of assistance
per month. SCN support subgroup 3 (medium SCN support): About 20% of primary caregivers
received moderate SCN support. On average, there were 1.69 secondary caregivers per primary
caregiver. They provided help on 2.75 care domains and 29 hours of assistance per month per
secondary caregiver. SCN support subgroup 4 (high SCN support): The last subgroup contains
about 28% of primary caregivers who received the highest level of assistance from SCN. For the
high SCN support subgroup, there are on average 2.41 secondary caregivers per primary
caregiver who participated in 3.42 care domains. On average, each of them provided nearly 65
hours of help per month. The distribution of the three identified latent groups across the three
indicators is depicted in Figure 4 for a more intuitive comparison.
75
Table 10 Predicted prevalence and descriptive statistics of SCN
a
support subgroups
n % SCN size
b
No. of domains
c
Care intensity
d
No SCN 228 29.16% / / /
Low SCN support 173 22.12% 1.16 (0.42) 1.69 (0.56) 10.92 (10.81)
Medium SCN support 160 20.46% 1.69 (0.97) 2.75 (0.71) 29.02 (21.89)
High SCN support 221 28.26% 2.41 (1.29) 3.42 (0.49) 64.96 (49.29)
a
SCN: Secondary Caregiver Network
b
SCN size: Number of secondary caregivers of each primary caregiver
c
No. of domains: Number of care domains that each SCN participated in
d
Care intensity: Mean number of hours that each secondary caregiver assisted per month
Figure 4. The distributions of the latent subgroups on the SCN support indicators
8.3.2 Depression Model
Before conducting multigroup analysis on the two-wave mediation model to test the
moderation effect of SCN support, the analysis needs to first confirm that the factor loadings,
error variance, factor variance, and covariance of latent factors are consistent across groups, so
as to determine the significant between-model difference comes from the difference in path
coefficients. Table 11 shows the results of the model comparison. Model 1 (baseline model) does
not constrain any parameters across groups; Model 2 constrained factor loadings to be equal
across the four subgroups; Model 3 added constraint of error variance to Model 2; and Model 4
76
added constraint of factor variance/covariance on the basis of Model 3. The χ
2
of the four models
are not significant at their respective degrees of freedom levels. The difference in χ
2
between
model 1 and model 2, as well as model 1 and model 3, was not significant. The χ
2
value of
Model 4 was significantly increased compared to Model 1 (Δ χ
2
= 30.79, p < 0.05), indicating
that the factor variance or covariance of latent variables differed among the four subgroups. In
sum, it can be determined that the latent measurement construct has cross-group consistency, and
the constraints on factor variance and covariance should be released when performing
multigroup comparison on the SEM model.
Table 11 Cross-group equality test on latent measurement construct
χ
2
Δ χ
2
RMSEA SRMR CFI TLI
Model 1 32.45 / 0.001 0.038 0.999 0.999
Model 2 34.25 1.8 0.001 0.043 0.999 0.997
Model 3 59.47 27.02 0.027 0.047 0.982 0.986
Model 4 63.24 30.79* 0.028 0.044 0.98 0.985
* p < 0.05
Model 1: baseline model (no constraint);
Model 2: factor loadings constrained;
Model 3: factor loadings and error variance constrained;
Model 4: factor loadings, error variance, factor variance/covariance constrained.
Multigroup analysis was conducted to examine whether the pathways of two-wave
mediation model were consistent across groups. Table 12 shows the path coefficients for each
subgroup and the results of cross-group coefficient comparisons. The results showed that there
were three direct path coefficients with cross-group differences, namely the moderation effects
of SCN support, including subjective stressor to social isolation (χ
2
= 8.35, p < 0.05), subjective
stressor to depression (χ
2
= 7.35, p < 0.05), objective stressor to depression (χ
2
= 7.55, p < 0.05).
The results of pairwise comparison showed that the path coefficient of subjective stressor to
social isolation was significantly higher in the low SCN support group than in the high SCN
support group (χ
2
= 4.30, p < 0.05), indicating that SCN support may buffer the impact of
77
subjective stressor on social isolation. The path coefficient of subjective stressor to depression
was significantly higher in the medium SCN support group than in the high SCN support group
(χ
2
= 3.98, p < 0.05), indicating that SCN support may buffer the impact of subjective stressor on
depression. The path coefficient of objective stressor to depression was significantly higher in
the low SCN support group than in the medium SCN support group (χ
2
= 4.70, p < 0.05),
indicating that SCN may buffer the impact of objective stressor on depression. Figure 5a, 5b, and
4c were generated to help demonstrate the moderation effects of SCN support on the three paths.
Test results for the two indirect pathways showed no cross-group differences. Overall, the four
cross-group models have good model fit with SRMRs all smaller than 0.05.
78
Table 12 The moderation effects of SCN
a
support on the depression model
β (se) p Wald Test (χ
2
) Pairwise comparison (χ
2
)
Change of social isolation (Mediator)
No SCN
Low SCN
support
Medium SCN
support
Subjective stressor No SCN 0.06 (0.19) 0.74
8.35*
/
Low SCN support 0.44 (0.18) 0.02 0.57 /
Medium SCN support 0.16 (0.22) 0.46 0.11 1.00 /
High SCN support -0.18 (0.26) 0.49 2.11 4.30* 0.99
Objective stressor No SCN 0.36 (0.14) 0.01
6.68
/
Low SCN support 0.39 (0.16) 0.02 0.02 /
Medium SCN support 0.28 (0.16) 0.08 0.15 0.26 /
High SCN support 0.15 (0.09) 0.10 1.61 1.69 0.53
Depression (Dependent variable)
Change of social
isolation
No SCN 0.23 (0.06) <.001
1.32
/
Low SCN support 0.24 (0.06) <.001 0.04 /
Medium SCN support 0.32 (0.07) <.001 1.12 0.90 /
High SCN support 0.25 (0.05) <.001 0.11 0.02 0.72
Subjective stressor No SCN 0.31 (0.16) 0.06
7.35*
/
Low SCN support 0.43 (0.18) 0.02 0.26 /
Medium SCN support 0.47 (0.14) 0.001 1.35 2.47 /
High SCN support 0.03 (0.17) 0.855 0.63 0.04 3.98*
Objective stressor No SCN 0.27 (0.13) 0.03
7.55*
/
Low SCN support 0.36 (0.14) 0.01 3.08 /
Medium SCN support -0.13 (0.11) 0.91 0.26 4.70* /
High SCN support 0.14 (0.07) 0.06 0.90 1.32 2.26
a
SCN: Secondary Caregiver Network;
Sample size: No SCN (n = 228), Low SCN support (n = 173), Medium SCN support (n = 160), High SCN support (n = 221);
* p < 0.05
79
a.
b.
c.
Figure 5 The moderation effects of SCN support on the depression model
80
8.3.3 Anxiety Model
The same approach was also applied to the two-wave mediation model with anxiety as
the outcome variable. Only one direct path was found having cross-group differences, which is
subjective stressor to social isolation, that is, SCN support moderates this path effect (χ
2
= 8.29,
p = 0.02). Since this path used the same data in the depression model, no pairwise comparison of
this path is needed. Test results for the two indirect pathways showed no cross-group differences.
Overall, due to the low data correlation, the model fit indices of each group of models are only
marginally acceptable, and the SRMRs are 0.051 (no SCN), 0.059 (low SCN support), 0.062
(medium SCN support), 0.057 (high SCN support).
8.4 AIM 3: Model Comparisons by Gender, Race, and Relationship Type
Multigroup analysis has examined the moderation effects of SCN support on the two-
wave mediation model, and identified three pathways that SCN support may buffer the negative
impact of caregiving stressors on primary caregivers' social isolation and depression. The
analysis in 8.4 was built on the previous findings, using multigroup analysis to examine whether
the two-wave mediation model (depression) differs between different gender, race, and
relationship type groups of primary caregivers. In addition, the analysis also examined whether
there are differences by gender, race, and relationship type in the moderation effects of SCN
support on the above three paths.
8.4.1 Comparing Means and Variances/covariance of All Variables
The analysis examined the mean and variance of all variables used in the two-wave
mediation model (Table 13). It was found that changes in social isolation scores differed between
the two relationship types, with adult child caregivers experiencing a greater increase in social
81
isolation than spousal caregiver (t = 2.16, p < 0.05). In terms of SCN support level, a difference
was found between spousal caregiver and adult child caregiver (χ
2
= 20.26, p < 0.001). Nearly
half (47%) of spousal caregivers reported no assistance from secondary caregivers, the
proportion in the adult child group was only 20%. Only 8.5% of spousal caregivers reported
receiving high level of SCN support, compared with 39% in the adult child group.
82
Table 13 Comparing means of observed variables by gender, race, and relationship type
Gender
t / (χ
2
)
Race
t / (χ
2
)
Relationship type
t / (χ
2
)
Women
(n = 526)
Men
(n = 256)
Black
(n = 264)
White
(n = 518)
Spouse
(n = 272)
Adult child
(n = 396)
Change of social
isolation
0.50 (1.89) 0.55 (1.74) 0.32 0.44 (1.72) 0.58 (1.83) 1.04 0.32 (1.67) 0.62 (1.83) 2.16*
Depression 3.50 (1.62) 3.34 (1.48) 1.37 3.52 (1.61) 3.41 (1.56) 0.97 3.44 (1.57) 3.52 (1.61) 0.61
Anxiety 3.25 (1.49) 2.97 (1.35) 2.57* 3.02 (1.43) 3.23 (1.46) 1.89 3.20 (1.46) 3.19 (1.48) 0.09
SCN
a
support
No SCN 30.61% 26.17%
-3.51
21.97% 32.82%
-10.38
47.43% 20.20%
(116.34***)
Low SCN 20.34% 25.78% 23.11% 21.62% 30.88% 17.42%
Medium SCN 20.53% 20.31% 23.11% 19.11% 13.24% 23.23%
High SCN 28.52% 27.73% 31.82% 26.45% 8.46% 39.14%
a
SCN: Secondary Caregiver Network
* p < 0.05, *** p < 0.001
83
For latent measurement construct, caregiving stressors, three sets of multigroup models
by primary caregivers' gender, race, and relationship type were conducted to examine cross-
group equality of factor loadings, error variance, and factor variance/covariance. Table 14 shows
that the mediated model can assume cross-group equality between women and men, and between
spouse and adult children. However, between Black and White, models 2, 3, and 4 all showed
significant χ
2
increases relative to the baseline model, suggesting that for Black and White
caregivers, the caregiving stress experienced by them are significantly different in factor
loadings (Δχ
2
= 16.69, p < 0.05), error variance (Δχ
2
= 28.33, p < 0.05), and factor
variance/covariance (Δχ
2
= 29.02, p < 0.05). Equality constraints on the latent measurement will
be released when comparing racial differences in the mediated model and moderation effects of
SCN support.
84
Table 14 Cross-group equality on latent measurement construct by gender, race, and relationship type
χ
2
Δχ
2
RMSEA SRMR CFI TLI
Gender: women vs. men
Model 1 25.51 / 0.001 0.039 999 0.999
Model 2 27.09 1.58 0.045 0.035 0.969 0.959
Model 3 33.87 8.36 0.017 0.041 0.983 0.983
Model 4 34.71 9.20 0.015 0.014 0.981 0.982
Race: Black vs. White
Model 1 29.75 / 0.001 0.025 0.999 0.999
Model 2 46.44 16.69* 0.037 0.04 0.972 0.93
Model 3 58.08 28.33* 0.028 0.018 0.96 0.96
Model 4 58.77 29.02* 0.038 0.035 0.953 0.96
Relationship type: spouse vs. adult child
Model 1 27.81 / 0.001 0.031 0.999 0.999
Model 2 30.27 2.46 0.026 0.044 0.96 0.964
Model 3 36.66 8.85 0.036 0.016 0.992 0.992
Model 4 36.73 8.92 0.037 0.026 0.952 0.953
* p < 0.05
Model 1: baseline model (no constraint)
Model 2: factor loadings constrained
Model 3: factor loadings and error variance constrained
Model 4: factor loadings, error variance, factor variance/covariance constrained
85
8.4.2 Comparing the Social Isolation Mediated Models
Multigroup analysis was conducted to examine the two-wave mediation model by gender,
race, and relationship type, respectively. Panel A of Table 15 shows that the structural equation
model by gender. The five direct paths of the mediated model were not significantly different
between women and men. There is no gender difference in the two indirect pathways either, that
is, for primary caregivers of different genders, there is a similar social isolation mediated path
between objective stressor and depression. The SRMRs of the two models are 0.037 (women)
and 0.020 (men), respectively.
Panel B of Table 15 shows the structural equation model by race, and the results indicate
that there is a racial difference in the path of objective stressor to depression (χ
2
= 5.64, p <
0.05). There is no significant difference in the cross-group comparison of other direct and
indirect paths, that is, for primary caregivers of different races, there is a similar social isolation
mediated path between objective stressor and depression. The SRMRs of the two models are
0.024 (white) and 0.051 (black), respectively. A figure showing the relationship between
objective stress and depression in different racial groups was generated to help clarify the
findings (Figure 6). The impact of objective stressor on primary caregiver’s depression tends to
be stronger in black than in white.
Panel C of Table 15 presents the structural equation model by relationship type, which
shows that the five direct paths and two indirect paths of the mediated model were not
significantly different between spousal and adult child caregivers. The SRMRs of the two models
are 0.041 (spouse) and 0.022 (adult child), respectively. Combining the results of the three
multigroup analyses above, only a racial difference was found, which highlights that black
caregivers are more vulnerable to objective stressor and experience higher depression than their
86
white counterparts. Given cross-group differences in the mediated model were found only in
race, it was decided not to examine differences in the models at the intersections of multiple
factors.
87
Table 15 Multigroup analysis on social isolation mediated model by gender, race, and relationship type
* p < 0.05
β (se) p Wald Test (χ
2
)
Panel A: women (n = 526) vs. men (n = 256)
Change of social isolation (Mediator)
Subjective stressor Men 0.28 (0.22) 0.19
1.51
Women -0.06 (0.73) 0.73
Objective stressor Men 0.26 (0.10) 0.01
0.82
Women 0.42 (0.14) 0.004
Depression (Dependent variable)
Change of social isolation Men 0.19 (0.05) < 0.001
0.72
Women 0.25 (0.05) < 0.001
Subjective stressor Men 0.58 (0.17) 0.001
1.52
Women 0.32 (0.14) 0.02
Objective stressor Men 0.15 (0.08) 0.06
0.76
Women 0.28 (0.12) 0.02
Panel B: Black (n = 264) vs. White (n = 518)
Change of social isolation (Mediator)
Subjective stressor White 0.19 (0.16) 0.24
0.60
Black 0.02 (0.15) 0.91
Objective stressor White 0.24 (0.09) 0.01
1.71
Black 0.47 (0.15) 0.002
Depression (Dependent variable)
Change of social isolation White 0.27 (0.04) < 0.001
2.29
Black 0.15 (0.07) 0.04
Subjective stressor White 0.46 (0.13) < 0.001
0.01
Black 0.46 (0.13) < 0.001
Objective stressor White 0.15 (0.07) 0.06
5.64*
Black 0.29 (0.14) 0.03
Panel C: Spouse (n = 272) vs. Adult child (n = 396)
Change of social isolation (Mediator)
Subjective stressor Spouse -0.06 (0.18) 0.73
1.68
Adult child 0.23 (0.14) 0.09
Objective stressor Men 0.31 (0.13) 0.02
0.08
Women 0.27 (0.09) 0.002
Depression (Dependent variable)
Change of social isolation Men 0.24 (0.06) < 0.001
0.01
Women 0.23 (0.04) < 0.001
Subjective stressor Men 0.33 (0.16) 0.04
0.45
Women 0.45 (0.11) < 0.001
Objective stressor Men 0.21 (0.12) 0.08
0.08
Women 0.17 (0.07) 0.02
88
Figure 6 The impact of objective stressor on depression by race
8.4.3 Comparing the Moderation Effects of SCN Support
Multigroup analysis was conducted to compare whether the moderation effects of SCN
support differ between different gender, race, and relationship type groups. In the analysis of
Aim 2, moderation effects of SCN have been identified on only three direct pathways, including
subjective stressor to social isolation, subjective stressor to depression, and objective stressor to
depression. Therefore, to test whether there is a difference in the effect of SCN support in
women and men, multigroup structural equation models were established for the two groups of
primary caregivers, respectively (Panel A of Table 16). In the women model, the relationship
between subjective stressor and social isolation was not significant at all four SCN support
levels, which is consistent with the results of the full-sample model. The regression coefficients
of subjective stressor to depression showed a trend of decrease with the increase of SCN support
level, but the pairwise comparison showed that there was no significant difference between the
coefficients of the four groups. A significant cross-group difference was found in the path of
objective stressor to depression, that is, SCN support could buffer the impact of objective
89
stressor on the depression of women primary caregivers (low SCN support vs. medium SCN
support, χ
2
= 7.56, p < 0.05). Figure 7a shows the effect of objective stressor on the depression of
women caregivers at different SCN support levels. In men, with the improvement of SCN
support level, the impacts of both subjective and objective stressors on the social isolation and
depression of men primary caregivers are decreasing. But pairwise comparison only found a
significant moderation effect on the path of subjective stressor to social isolation (low SCN
support vs. high SCN support, χ
2
= 4.65, p < 0.05). Figure 7b shows the effect of subjective
stressor on social isolation of men primary caregivers with different SCN support levels.
The same approach was used to examine whether the moderation effects of SCN support
differ between black and white (Panel B of Table 16). In the black model, the regression
coefficients of the three paths were consistent with the results of the full-sample model, but no
significant between-group differences were found. In the white model, the regression coefficients
on the three paths all show a downward trend with the increase of the SCN support level.
Pairwise comparison indicated significant moderation effects of SCN support on the paths of
subjective stressor to social isolation (low SCN support vs. high SCN support, χ
2
= 5.64, p <
0.05) and objective stressor to depression (low SCN support vs. medium SCN support, χ
2
= 5.85,
p < 0.05; low SCN support vs. high SCN support, χ
2
= 4.66, p < 0.05). Figure 8a and 8b shows
the moderation effects of SCN support on these two paths.
Likewise, the difference in the moderation effects of SCN support between spouse and
adult children were examined. Panel C of Table 16 shows the results of the test. However, no
moderation effects of SCN support were found in neither the spouse nor the adult child groups.
90
Table 16 Multigroup analysis on moderation effects of SCN support by gender, race, and relationship type
β (se) p
Wald Test (χ
2
)
β (se) p
Wald Test
(χ
2
)
Panel A: women vs. men
Women (n = 526)
Men (n = 256)
Change of social isolation (Mediator)
Subjective stressor No SCN 0.37 (0.64) 0.56
/
1.02 (0.54) 0.06
4.65*
Low SCN -0.09 (0.72) 0.9
1.45 (0.45) 0.002
Medium SCN 0.19 (0.62) 0.75
0.95 (0.44) 0.03
High SCN 0.45 (0.71) 0.62
-0.11 (0.52) 0.84
Depression (Dependent variable)
Subjective stressor No SCN 0.25 (0.21) 0.23
1.86
0.78 (0.38) 0.04
0.39
Low SCN 0.61 (0.23) 0.008
0.53 (0.34) 0.12
Medium SCN -0.05 (0.24) 0.83
0.70 (0.32) 0.03
High SCN 0.39 (0.20) 0.06
0.32 (0.34) 0.35
Objective stressor No SCN 0.29 (0.16) 0.07
7.56*
0.23 (0.16) 0.16
0.83
Low SCN 0.50 (0.17) 0.004
0.36 (0.18) 0.04
Medium SCN -0.13 (0.16) 0.41
0.13 (0.15) 0.39
High SCN 0.04 (0.14) 0.09
0.15 (0.13) 0.24
Panel B: Black vs. White
Black (n = 264)
White (n = 518)
Change of social isolation (Mediator)
Subjective stressor No SCN -0.09 (0.32) 0.78
/
0.14 (0.24) 0.55
5.64*
Low SCN 0.01 (0.28) 0.99
0.63 (0.27) 0.02
Medium SCN -0.07 (0.31) 0.82
0.22 (0.29) 0.45
High SCN -0.15 (0.31) 0.62
-0.25 (0.33) 0.44
Depression (Dependent variable)
Subjective stressor No SCN 0.49 (0.31) 0.1
/
0.21 (0.21) 0.31
1.32
Low SCN 0.40 (0.22) 0.06
0.35 (0.27) 0.19
Medium SCN 0.09 (0.29) 0.75
0.47 (0.22) 0.04
High SCN 0.45 (0.23) 0.05
-0.14 (0.28) 0.62
Objective stressor No SCN 0.32 (0.34) 0.35
1.31
0.30 (0.12) 0.01
5.26*
Low SCN 0.67 (0.34) 0.04
0.49 (0.16) 0.003
Medium SCN 0.48 (0.32) 0.13
0.01 (0.13) 0.96
High SCN 0.13 (0.24) 0.59
0.08 (0.11) 0.46
Panel C: spouse vs. adult child
Spouse (n = 272)
Adult Child (n = 396)
Change of social isolation (Mediator)
Subjective stressor No SCN -0.01 (0.69) 0.98
/
-0.06 (0.41) 0.88
/
Low SCN -0.17 (0.73) 0.82
-0.36 (0.43) 0.41
Medium SCN -0.01 (0.95) 0.99
-0.07 (0.32) 0.85
High SCN 11.45 (17.08) 0.91
-0.07 (0.38) 0.86
Depression (Dependent variable)
91
(continued)
β (se) p
Wald Test (χ
2
)
β (se) p
Wald Test
(χ
2
)
Objective stressor No SCN 0.32 (0.34) 0.35
1.31
0.30 (0.12) 0.01
5.26*
Low SCN 0.67 (0.34) 0.04
0.49 (0.16) 0.003
Medium SCN 0.48 (0.32) 0.13
0.01 (0.13) 0.96
High SCN 0.13 (0.24) 0.59
0.08 (0.11) 0.46
Panel C: spouse vs. adult child
Spouse (n = 272)
Adult Child (n = 396)
Change of social isolation (Mediator)
Subjective stressor No SCN -0.01 (0.69) 0.98
/
-0.06 (0.41) 0.88
/
Low SCN -0.17 (0.73) 0.82
-0.36 (0.43) 0.41
Medium SCN -0.01 (0.95) 0.99
-0.07 (0.32) 0.85
High SCN 11.45 (17.08) 0.91
-0.07 (0.38) 0.86
Depression (Dependent variable)
Subjective stressor No SCN 0.45 (0.29) 0.12
/
-0.14 (0.42) 0.73
/
Low SCN 0.25 0.31 0.41
0.44 (0.33) 0.18
Medium SCN -0.31 (0.68 0.64
0.51 (0.27) 0.06
High SCN 0.95 (3.49) 0.91
-0.24 (0.34) 0.47
Objective stressor No SCN 0.41 (0.37) 0.37
/
0.77 (0.47) 0.14
1.96
Low SCN -0.13 (0.21) 0.54
0.86 (0.38) 0.03
Medium SCN 0.66 (0.44) 0.13
0.17 (0.36) 0.63
High SCN 0.44 (0.35) 0.74 0.06 (0.33) 0.84
* p < 0.05
92
a. Women
b. Men
Figure 7 Moderation effects of SCN support on women and men
93
a.
b.
Figure 8 Moderation effects of SCN support on White
In sum, the analytic results show that there are gender and racial differences in the
moderation effects of SCN support. Therefore, it is necessary to continue examining the
moderation effects of SCN support at the intersection of gender and race. In the previous section,
gender and racial differences were only found in paths of subjective stressor to social isolation
and objective stressor to depression, the following test only examined these two paths. Four
subgroups were generated based on primary caregivers' gender and racial identities, including
White men, Black men, White women, and Black women. The results of multigroup analysis
94
were presented in Table 17. For White men, there is a trend that the regression coefficient
decreases as the SCN support level increases in the two paths, but the pairwise comparison
shows that the significant moderation effect is only on the path of subjective stressor to social
isolation (low SCN support vs. high SCN support, χ
2
= 11.05, p < 0.001). Figure 9a shows the
moderation effect of SCN support on this path in White men.
For Black men and White women, pairwise comparison did not find moderation effect of
SCN support on either path. For Black women, the results show that SCN support does not play
a significant moderation effect on the path of subjective stressor to social isolation, but on the
path of objective stressor to depression. Compared to Black women who received low (χ
2
= 6.93,
p < 0.01) and medium (χ
2
= 5.65, p < 0.05) level of SCN support, objective caregiving stress had
a significantly lower impact on depression in those who received a high level of SCN support.
Figure 9b shows the moderation effect of SCN support on this path in Black women.
95
Table 17 Multigroup analysis on moderation effects of SCN support at the intersection of gender and race
White men
(n = 176)
Black men
(n = 80)
White women
(n = 342)
Black women
(n = 184)
β (se) p β (se) p β (se) p β (se) p
Change of social isolation
(Mediator)
Subjective
stressor No SCN 1.04 (0.54) 0.06
-0.04 (0.85) 0.96
.21 (0.84) 0.81
0.29 (0.48) 0.54
Low SCN 1.48 (0.48) 0.002
1.41 (0.75) 0.06
-1.09 (0.94) 0.25
0.50 (0.53) 0.35
Medium SCN 1.02 (0.49) 0.04
2.05 (1.92) 0.28
-0.36 (0.65) 0.57
0.69 (0.64) 0.27
High SCN -0.07 (0.73) 0.92
0.68 (0.76) 0.37
-0.05 (0.76) 0.94
-0.31 (0.51) 0.55
Wald Test (χ
2
)
11.05**
/
/
/
Depression (Dependent variable)
Objective
stressor No SCN 0.23 (0.17) 0.19
0.17 (0.08) 0.04
0.14 (0.26) 0.59
0.83 (0.42) 0.05
Low SCN 0.88 (0.39) 0.02
13.42 (76.35) 0.86
-0.15 (0.25) 0.55
1.23 (0.44) 0.005
Medium SCN 0.11 (0.18) 0.61
-0.28 (0.21) 0.16
0.04 (0.26) 0.86
1.11 (0.41) 0.007
High SCN 0.19 (0.16) 0.25
57.04 (80.87) 0.88
0.04 (0.22) 0.84
0.34 (0.37) 0.37
Wald Test (χ
2
) 3.27 1.26 / 6.29**
** p < 0.01
96
a. White men
b. Black women
Figure 9 Moderation effects of SCN support at the intersection of gender and race
97
CHAPTER NINE: DISCUSSION
The dissertation study explored the social isolation mediated stress process in primary
caregivers of older adults using two paired nationally representative datasets. The study used an
integrated measurement of family caregivers' social isolation, including objective social
disconnectedness and subjective loneliness. Longitudinal data enables research to construct a
two-wave mediation model that examines the effects of caregiving stress on family caregivers'
mental health symptoms through the mediated effect of social isolation. A network perspective
was also adopted to look at the possible support of an older adult from multiple family
caregivers. By refining the definition and measurement of SCN, this dissertation study identified
the type of support primary caregivers received from secondary caregivers, and examined the
moderation effects of different levels of SCN support on their social isolation mediated stress
process. In addition, the study considered the complexity and diversity of the background
characteristics of family caregivers, as well as the differences in the impact of caregiving stress
on caregiving experiences and health consequences among different primary caregivers.
Intersectionality analysis was used to examine whether the social isolation mediated stress
process and SCN support vary by gender (women vs. men), race (Black vs. White), and
relationship type (spouse vs. adult child) of primary caregivers.
Overall, the findings support the hypotheses in Aims 1 and 2, namely, that caregiving
stress can increase primary caregivers' social isolation and, as a result, their depression. As
hypothesized, SCN support also protected primary caregivers' social connections and mental
health from caregiving stress. However, mediated effects were found only in the depression
model, and the moderation effects of SCN support was primarily to reduce the impact of
caregiving stress on social isolation and the direct effects of caregiving stress on depression. Due
98
to its exploratory nature, Aim 3 did not set research hypotheses. The results found that the social
isolation mediated model of primary caregivers only differed between Black and White, that is,
Black primary caregivers tend to experience a greater increase in depression than Whites. The
moderation effect of SCN support is different between women and men, as well as between
Blacks and Whites. Intersectionality analysis revealed that receiving SCN support can
significantly protect White men caregivers from becoming socially isolated, and Black women
caregivers from depression. A detailed discussion of the major findings in the study follows.
9.1 Descriptive Characteristics of the Study Sample
9.1.1 Sample representativeness
The sample for this dissertation study included a total of 782 primary caregivers, 782
older adults, and 1003 secondary caregivers. A direct comparison with the original samples in
the NHATS and NSOC datasets shows that some key characteristics of the study sample are
consistent with the original data (Wolff et al., 2016). For example, according to whether
caregivers assist older adults with daily activities, the proportion of women caregivers in NSOC
ranges from 56.6 to 69.3%, which is 67.3% in this study. Because the inclusion criteria of this
study excluded paid caregivers and participants with missing caregiving-related information, the
proportion of close relative caregivers in the study is 85.4%, higher than in the full NSOC
sample (80.4%). The sample for this study only included non-Hispanic Black and non-Hispanic
White caregivers, so the racial percentage is different from NSOC, but the ratio of the two
groups is close to 1:2. Another key variable is probable/possible dementia prevalence among
older adults, it was found that the proportion of caregivers in NSOC who provided help to older
adults with probable/possible dementia ranged from 20.6% (no help with health care activities)
99
to 45.5% (substantial help with healthcare activities), the proportion of 38.9% in this study is in
that range. The proportion in this study is close to the high value of the NSOC data also because
the inclusion criteria excluded participants who were not directly involved in caregiving
activities and had missing values, so the final sample is more similar to those who provided
substantial help in the NSOC. Therefore, the sample of this study has a low risk of decreased
representativeness due to the data screening.
9.1.2 The increasing trend of social isolation
Because this study's mediated model is based on data from two waves in 2015 and 2017,
changes in primary caregivers' social isolation between the two waves were computed and used
in the model analysis. The results of the paired t-test revealed that primary caregivers had a
significant increase in overall social isolation, as well as both dimensions of social
disconnectedness and loneliness. This supports previous research that assuming the role of
family caregiver impacts one's social network, reduces social connection and participation, and
increases loneliness (Hallikainen et al., 2018; Li et al., 2020; Powers et al., 2002).
The most noticeable changes on each specific item are less interaction with friends and
less participation in church activities. Fewer interactions with friends may be the most common
change in family caregivers' social interactions, implying that family caregivers' access to
informal social support is dwindling, and their resilience to caregiving stress is dwindling as
well, raising the risk of mental health issues (Kovaleva et al., 2018). Women caregivers may be
the hardest hit, as a lack of social connections limits their access to emotional support from
friends and family. Because they have fewer external resources than men, women are more likely
to be affected by caregiving stress on health outcomes (Rozario & Simpson, 2018).
100
Reduced involvement in church activities is also reasonable. Because of their religious
characteristics and the uniqueness of group rituals, church activities often require a set period of
time, such as the morning, in comparison to other social activities. Other social activities, such as
clubs or volunteer activities, are more flexible in terms of time and can be adjusted to fit
participants' schedules. In the United States, Blacks are more likely than Whites to engage in
religious pursuits, which may protect their psychological health, provide meaning and purpose in
life, expand their social circle, and improve social support (Henderson et al., 2021). As a result,
the decline in church attendance may have a greater impact on Black caregivers' mental health
than White caregivers. In general, longitudinal data shows that the level of social isolation
among family caregivers is increasing, and its potential risk to family caregivers' mental health
cannot be overlooked.
9.1.3 The change of mental health outcomes
The results of the paired t-test show that both the depression and anxiety levels reported
by primary caregivers were elevated. Depression levels increased significantly, which is
consistent with previous research findings (Bradley et al., 2004; Crespo et al., 2005; S. Liu et al.,
2018; Wang et al., 2016). Taking care of an older adult with disabilities is a long-term chronic
stress with few effective solutions, because many older people with long-term care needs are
unable to regain their ability to live independently as before, and their physical and cognitive
functions are continuously deteriorating. This process undoubtedly increases family caregivers'
caregiving stress, which leads to an increase in depressive symptoms (Crespo et al., 2005).
Rising caregiver depression may affect primary caregivers' perceptions of caregiving
responsibilities and may even increase their resentful feelings toward older adults, thereby
reducing the quality of care (Beeson, 2003).
101
Anxiety levels were also higher, as family caregivers face greater challenges as a result of
increased caregiving stress over time. Caregivers may need to devote more time and energy to
caregiving if older adults' ability to perform activities of daily living independently deteriorates.
When faced with new caregiving tasks, caregivers may feel helpless and stressed, which can lead
to anxiety (S. Liu et al., 2017; Shin et al., 2018). In comparison to depression, anxiety shows a
smaller change, which could be due to a difference in how caregiving stress affects the two
emotions. It is widely assumed that caregivers' activities do not cause them to become depressed.
Instead, caregivers' depressive symptoms stem from their sacrifice of time and energy, as well as
their frustration with caregiving tasks. Anxiety is primarily caused by caregiving tasks that
exceed caregivers' coping capacity, causing them to feel overwhelmed (García-Alberca et al.,
2012). In the long run, the change of caregiving tasks is relatively slow, so family caregivers can
improve their caregiving skills after a period of learning, so as to adapt to the caregiving tasks
and reduce anxiety. Primary caregivers' depressive emotions will build up gradually once they
take on caregiving responsibilities, and research shows that depressive symptoms can last for at
least a year even after the caregiving roles are completed (Magan et al., 2021).
9.1.4 The recipient of SCN support
The size of SCN in this study is also slightly smaller than in previous studies due to the
strict definition of secondary caregivers used. Each primary caregiver had an average of 1.81
secondary caregivers assisting older adults. Thus, in the sample of this study, there are
approximately 2.81 family caregivers per older adult, which is still comparable to the size of
caregiver networks (around 3) reported in previous studies (Andersson & Monin, 2017; Spillman
et al., 2020). In addition to older adults with only one family caregiver, 56% had one or two
secondary caregivers, and 15% had three or more secondary caregivers.
102
Despite the fact that the size of the SCN did not change significantly between 2015 and
2017, the researchers believe that future American society will face a shortage of family
caregivers (Schulz & Czaja, 2018), because nuclear families or single parent families, in which
only one or two parents live with underage children, are becoming more common. Adult children
choose not to live with older parents for better work opportunities and life independence. Rising
women labor force participation and declining birth rates could also herald a future shortage of
family caregivers. In fact, data shows that 26% of family caregivers are not actually involved in
the day-to-day activities of older adults, but only help manage finances or handle bills (Wolff et
al., 2016). Therefore, more and more older people may face the issue of aging alone. The limited
accessibility and high cost of assisted living facilities and nursing homes cannot meet everyone's
needs, especially for older adults with poor health and long-term care needs. As a result, based
on the assumption that government and social services cannot completely replace family
caregiving, public policies are needed to encourage multiple family members to take on the
responsibility, thereby sharing the burden with primary caregivers and improving the quality of
life for older adults.
On average, each SCN participates in 2.26 caregiving domains, but nearly 60% of SCNs
participate in less than 2 caregiving domains. The average monthly care intensity of SCN is 73
hours, and the average secondary caregiver provides less than 10 hours of care to older adults per
week, while the average caregiving intensity of primary caregivers is close to 40 hours per week,
which is equivalent to a full-time job. The data indicates that secondary caregivers still serve as
assistant to primary caregivers, and primary caregivers remain the main force in family
caregiving, providing nearly 70% of help to older adults. Primary caregivers are at the highest
risk of health consequences from caregiving stress, because of the unbalanced division of care
103
responsibilities between primary and secondary caregivers. Promoting SCNs to take on more
caregiving responsibilities and participate in more caregiving tasks may be an effective way to
help primary caregivers reduce their caregiving burden and improve health outcomes.
9.2 The Mediated Effect of Social Isolation
The analysis of Aim 1 in this study supports the hypothesis that social isolation can
mediate the stress process of primary caregivers. That is, caregiving stress can affect the mental
health of primary caregivers by increasing their level of social isolation, though this effect was
only found in the depression model. It is reasonable that one sacrifices personal time to take on
the role of family caregiver, thus reduces participation in social activities and interaction with
friends, and results in compromised emotional stability (Crespo et al., 2005; García-Alberca et
al., 2012). The more intensive care activities, the more likely to increase social isolation of
caregivers, and lead to higher levels of depression (Wang et al., 2016). The structural pathway
model found that when distinguishing between subjective stressor and objective stressor, social
isolation only mediated the effect of objective stressor on depression. This may be because
objective stressor is the direct measure of the caregiving workload, such as functional limitations
and dementia status of older adults, which reflect the time and effort that primary caregivers
devote to caregiving. While subjective stressor is a measure of perceived caregiver burden,
which is an appraisal process that takes into account the caregiver's resilience and available
social support resources (Haley et al., 1987; Seiffge-Krenke, 2011). Primary caregivers who
experience social isolation or have limited access to social support may perceive higher levels of
caregiving stress than those with relatively adequate informal and formal support resources (dos
Santos Azevedo et al., 2021; Sun et al., 2019). Therefore, subjective caregiving stress may not
104
lead to an increase in social isolation, but social isolation may lead to an increase in the
individual's perceived caregiver burden.
In addition to the indirect paths, the two direct paths significantly indicate that both
subjective stressor and objective stressor have an impact on the levels of depression of primary
caregivers. This finding is consistent with previous studies (George & Gwyther, 1986; S. Liu et
al., 2018; Shin et al., 2018; Wang et al., 2016). In the Stress Process Model of caregiving, the
effect of caregiving stress on the health outcomes of caregivers is very complex (Pearlin et al.,
1990). Although the social isolation mediated path explored in this study is just one of the
possible mechanisms, accounting for about 28.4% of the total effect, it still shows that the
limitation of social life has a significant impact on the mental health of family caregivers. Future
research is necessary to continue exploring the mechanism by which caregiving stress affects
family caregivers' social life and thus their health outcomes, so as to provide empirical evidence
for the development of interventions aimed at protecting the health and well-being of caregivers.
Although the analysis found a significant mediated effect in the model of depression,
nothing was found in the model of anxiety. Caregiving stress at wave 1 did not directly lead to
the rise of primary caregivers' anxiety at wave 2, nor did it indirectly affect their anxiety by
affecting their social isolation. This may be related to the previously analyzed differences in the
mechanisms of depression and anxiety in family caregivers. One of the primary causes of anxiety
is that the difficulty of caregiving tasks exceeds family caregivers' coping capacity, thus unable
to meet the care needs of older adults even with full commitment (Cantor, 1983; George &
Gwyther, 1986). Studies found that anxiety of family caregivers is also related to their own self-
efficacy or self-competence (Lloyd et al., 2018; Savla et al., 2019; Shin et al., 2018). So
subjective stressor may have a direct effect on the anxiety level of primary caregivers, and this
105
trend was found in this study though not significant (p = 0.06). Despite the fact that long-term
caregiving stress may cause primary caregivers to live under chronic anxious emotion, but as the
mastery level of caregiving skills increases, their appraisal to caregiver burden may not change
significantly and anxiety levels may not continue rising either.
9.3 The Moderation Effects of SCN Support
9.3.1 The latent profiles of SCN support
Aim 2 of this dissertation study assumes that there are several subtypes of SCN support
that can be used to describe the support that primary caregivers received from other secondary
caregivers. The results of the latent profile analysis (LPA) identified three subtypes for the level
of SCN support, including low SCN support, medium SCN support, and high SCN support. This
means that in this study, the level of SCN support is related to the three indicators (SCN size,
number of care domains, caregiving hours per month). No mixed SCN support type was found
(e.g., large SCN size but low participation in care tasks), indicating that primary caregivers with
larger SCNs are more likely to obtain more support for caregiving.
The identified three subtypes describing the SCN support level from low to high show
good reliability, because we used the average monthly care hours of each secondary caregivers in
the analysis instead of the total care hours of the SCN to avoid collinearity between SCN size
and SCN monthly care hours. However, the interpretation of the results should be cautious, that
is, it cannot be concluded that there is no situation in which older people have large SCN but the
caregiving engagement of secondary caregivers is low. One of the major reasons may be that this
study adopted a relatively strict definition on secondary caregivers in order to obtain
accurate/non-missing information of each secondary caregiver on the three indicators. For
106
example, people who are not regular/monthly caregivers and who are not directly involved in
caregiving activities were excluded. Doing so can ensure the quality of the data, but may affect
the comprehensiveness and complexity of the description on older adults' support network.
As a data-driven analysis approach, the participants assigned to each subgroup are the
most similar in terms of each indicator, which helps qualitatively describe the results of complex
multivariate measurements. Although, the three indicators used in this study to measure the SCN
support level (SCN size, number of care domains, and caregiving intensity per capita) can all
reflect the level of SCN participation in caregiving, because structural equational modeling
requires categorical variables for cross-group comparison, LPA can distinguish different SCN
support levels with the best goodness of fit, which is more scientific than using an arbitrary cut-
off point. It also eliminates the need to perform a cross-group SEM analysis separately using
three indicators.
9.3.2 SCN support buffers the impact of caregiving stress on the depression of primary
caregivers
The moderation effect of SCN support was found to be significant on three pathways,
including subjective stress to social isolation, subjective stress to depression, and objective stress
to depression. The analysis on social isolation mediated indirect effect did not find significant
moderation effects, that is, SCN support may not protect primary caregivers' depression level
from the influence of social isolation.
According to the results of cross-group SEM analysis, subjective caregiving stress was
only found to aggravate the social isolation of primary caregivers when SCN support is very
limited. We did not find a correlation between subjective stress and social isolation in the Aim 1
analysis, and speculated that social isolation might be the cause of aggravated caregiver burden
107
perceived by primary caregivers (dos Santos Azevedo et al., 2021; Sun et al., 2019). But the
significant correlation between the two in the Aim 2 analysis suggests that their relationship may
be bidirectional rather than unidirectional. Primary caregivers who perceive greater caregiving
stress can still maintain a certain level of social engagement and social interaction if they receive
adequate social support. Some other primary caregivers with lower levels of SCN support might
compare their caregiving responsibilities and caregiving intensity with those of secondary
caregivers. This comparison may significantly reduce their perceived social support and increase
their loneliness (Evans et al., 2012).
Subjective stress has no effect on social isolation for primary caregivers with high SCN
support. This finding suggests that SCN has the potential to replace primary caregivers' general
social networks. Secondary caregivers in this study are all family/unpaid caregivers who are
usually relatives or friends of older adults and have a high likelihood of knowing primary
caregivers. Due to their high level of involvement in caregiving activities, secondary caregivers
may communicate with primary caregivers frequently. Communicating with others who have had
similar experiences as primary caregivers can provide emotional support and reduce the effects
of caregiving stress on loneliness (Kelley et al., 2017).
No significant moderation effect was found on the impact of objective stress on social
isolation of primary caregivers. This suggests that SCNs, even with high involvement in
caregiving activities, are unable to make up for the sacrificed social life of primary caregivers
due to caregiving responsibilities. While SCN support moderated the relationship between
subjective stress and perceived loneliness, interaction with secondary caregivers did not appear
to improve social activity participation for primary caregivers with high objective stress (i.e.,
older adults with intensive care needs). There may be three explanations. First, SCN support may
108
not directly reduce primary caregivers' care intensity. In cross-sectional analyses, primary
caregivers' care intensity is more correlated with the care needs of older adults, which also
explains its positive correlation with SCN size (Feld et al., 2012; Penrod et al., 1995; Tausig,
1992). As a result, SCN's high involvement in caregiving activities is more likely to make
primary caregivers feel supported on a subjective level (this has been discussed in the previous
paragraph). Second, because of their strong sense of responsibility, primary caregivers may not
leave older adults even when secondary caregivers are present. Third, individuals will gradually
withdraw from their social network and reduce their participation in social activities once they
take on the role of caregiver, making it difficult to reintegrate (Williamson et al., 1998;
Zegwaard et al., 2013). Because the measures used in this study did not assess the frequency of
communication between primary caregivers and the SCN, it is unlikely that there will be a
significant change in primary caregivers' social disconnectedness.
Our findings also revealed that SCN support has a moderating effect on the effects of
both subjective and objective stress on primary caregivers' depression levels. These findings
support the SPM's assertion that social support can buffer the effects of caregiving stress on the
majority of mental, social, and health outcomes (Halpern et al., 2017; Pearlin et al., 1990; Wang
et al., 2016). In addition, the impact of social isolation on depression is not moderated by SCN
support, which supports our view that despite SCN cannot completely replace the social network
of primary caregivers, the social support obtained during cooperation with SCN can contribute to
the mental health of primary caregivers. Combined the three pathways on which SCN support
has moderation effects, having secondary caregivers participating in caregiving activities is more
likely to provide emotional support for primary caregivers, so as to reduce perceived loneliness
109
and protect mental health. And SCN support does not appear to protect the social connection of
primary caregivers.
It is worth noting that the high SCN support group that identified in this study contains an
average of 2.4 secondary caregivers, participates in 3.4 care domains, and provides more than 60
hours of care per month per capita. SCN's level of caregiving is already comparable to that of
primary caregivers in this case, and multiple caregivers can be considered to provide assistance
to an older adult at a similar level of engagement. Therefore, this cooperative mode, in which
multiple caregivers participate in caregiving activities at a similar level, will be beneficial to each
caregiver's social and mental health. Future research should focus on how secondary caregivers
can become more involved in caregiving and take on more responsibilities, as well as the effect
of this cooperative mode on care quality.
9.4 The effects of gender, race, and relationship type
9.4.1 Women vs. men
Women and men caregivers had similar levels of social isolation, depression, and SCN
support, according to the results of a between-group comparison. The social isolation mediated
model showed no gender differences, implying that both women and men caregivers may
experience increased depression as a result of social isolation brought on by high objective
caregiving stress. These findings contradict previous research that suggests that because women
are more likely to play the role of family caregiver and provide more intensive care than men,
their social networks are more likely to be affected, and they are more likely to experience
depression and receive less social support (Dill & Duffy, 2022; Rozario & Simpson, 2018;
Versey, 2017). One possible explanation is that the current study only compared caregiving
110
experience of primary caregivers, which did not appear to differ by gender since primary
caregivers are typically deeply involved in caregiving tasks. The absence of gender differences in
SCN support may also be explained by the adoption of the strict definition of secondary
caregivers that only includes those who provide support with caregiving tasks. Men caregivers
are thought to have larger social support networks, more financial support, and higher formal
service use than women caregivers (Scharlach et al., 2006; Song et al., 2011), all of which were
not measured in the analysis.
Significant gender differences were found in the moderation effect of SCN support. For
men caregivers, higher SCN support could significantly reduce the impact of objective
caregiving stress on their social isolation, while this effect was not significant in women group.
This may be explained by differences in women and men caregivers' social networks and need
for social support. Men caregivers may have more complex social networks and more sources of
social support. When SCN has a higher level of caregiving participation, men caregivers may be
more likely to withdraw to participate in other social activities for social support (Mulud &
McCarthy, 2017). Women caregivers' social networks are relatively simple, and they rely on
people with similar backgrounds for emotional support in order to reduce caregiver burden
(Kelley et al., 2017). Coupled with the socially gendered role of caregiver, women caregivers
have a higher sense of responsibility than their men counterparts (Versey, 2017). For the direct
path of objective stress to depression, the moderation effect of SCN support was significant only
for women, which also supports the hypothesis that women are more inclined to obtain
emotional support from those with the same caregiving experience. Therefore, when it comes to
the preferred cooperative mode, women caregivers may benefit more from sharing caregiving
111
intensity with SCN, whereas men caregivers may prefer a shift mode to allow them to participate
in other social activities.
9.4.2 Black vs. White
Similarly, racial differences were not found at the social isolation, depression, and SCN
support levels. The mediated effects of social isolation also not varied between Black and White
caregivers. However, racial differences were found on the path of objective caregiving stress to
depression, with higher objective stress leading to increased depression in Black people.
Although Black caregivers tended to report less caregiving burden (i.e., subjective caregiving
stress) when assuming similar caregiving responsibilities as White caregivers, the prevalence of
depression in Black caregivers is significantly higher than their White counterparts (Cynthia et
al., 2020; Fabius et al., 2020). Black older adults tend to have more care needs and greater
reliance on caregivers, but the average income of Black families is lower than that of White
families, and they have less access to external support resources due to the disadvantage of
socioeconomic status. So Black caregivers' coping capacity and resilience may be lower than
Whites, which significantly predicts their higher prevalence of depression and higher possibility
of institutionalizing older adults (Jenkins Morales & Robert, 2020; Sörensen et al., 2010).
Placing older adults in care facilities is contrary to the strong familism value in Black families,
and may place a greater burden on them emotionally (Cox & Monk, 1990). Furthermore, when
depressive symptoms are present, Black caregivers are less concerned about their own mental
health, as evidenced by fewer doctor visits and a lower likelihood of taking antidepressant
medication (Sleath et al., 2005).
The Moderation effects of SCN support were found to differ between Black and White
caregivers. When higher levels of SCN support were obtained, the impact of subjective stress on
112
White caregivers' social isolation was significantly reduced. While Black caregivers may
experience more caregiving stress, especially objective stress, SCN support did not appear to
protect their social engagement from caregiving stress. It may be because the familism value of
Black families makes primary caregivers take the help from other family members for granted,
thus SCN support may not have a significant impact on their perceived loneliness. Also, the
measurement of social isolation in this study did not include the frequency of interaction with
secondary caregivers, so SCN support was not well reflected in the social disconnectedness score
of primary caregivers. Notably, religious activity participation was identified as an important
coping strategy for Black caregivers (Henderson et al., 2021), but the high levels of SCN support
in this study also did not appear to help maintain church participation in Black primary
caregivers.
The comparison between Blacks and Whites found that SCN support may have a
protective effect on White primary caregivers. For Black primary caregivers, high SCN
involvement may be an important reason for their perceived lower caregiver burden, but appears
to have limited protective effects on their social functioning and mental health. The
DISCUSSION section analyzes two possible reasons, namely the emphasis on the caregiving
culture of Black families prevents primary caregivers from taking part in personal social
activities even with assistance from SCN; and the limited external support resources and less
formal service use of Black caregivers compared to Whites may predict their social isolation and
depression. Future work could compare the protective effects of informal and formal support on
family caregivers of different races, as well as promote external support resources in favor of
Blacks to improve their availability and accessibility to formal services.
9.4.3 Spouse vs. adult child
113
Neither the social isolation mediated stress process model nor the moderation effects of
SCN support were found to differ between spouse and adult child caregivers. However, the
bivariate comparison found that between wave 1 and wave 2, the increase in social isolation
experienced by adult child caregivers was significantly higher than that of spousal caregivers.
Since previous studies have not reached a unified conclusion on social isolation of spouse and
adult child caregivers, the results of this study seem to favor the view that adult children are
more vulnerable in family caregiving. Most adult child caregivers may experience the
sandwiched dilemma, that is, having to sacrifice work and social time for caring for their
underage children as well as their older parents (Huiying et al., 2019; Oldenkamp et al., 2016).
Adult child caregivers may have larger social networks than spousal caregivers.
Therefore, despite adult children experienced a greater impact on their social interactions when
assuming care responsibilities, they still have a greater potential to obtain more social support
from multiple sources (Ornstein et al., 2019). This study found that nearly 40% of the adult child
caregivers received a high level of SCN support, while nearly half of the spousal caregivers were
sole caregiver, that is, without the help of any secondary caregiver. More resources should be
navigated to support spousal caregivers, especially those who are older, have limited social
connections, and are less able to access social support.
9.4.5 The intersection between gender and race
Because there was no difference in the mediated effects of social isolation and the
moderation effects of SCN support between spouse and adult child caregivers, the intersectional
analysis only looked at differences in primary caregiver's experience across gender-race groups.
The intersectionality perspective is to explore the experience of discrimination experienced by
individuals due to two or more sociodemographic identities, so a large number of studies are
114
basically exploring the marginalized experience of Black women in society due to their
experience of sexism and racism (Crenshaw, 1989, 1991). Compared to studies focusing on
Black women's job opportunities, income, civil rights, and health, discriminatory experiences or
cultures of implicit discrimination may not have a direct impact on caregiving activities, but on
caregivers' attitudes and their perceived experience of caregiving (Dill & Duffy, 2022; O’Malley
et al., 2000). Therefore, the analysis of this dissertation study adopted an exploratory strategy to
first identify the demographic factors by which the mediated effects of social isolation and
moderation effects of SCN support are different, and then examine the intersectional effects of
the two identified significant factors.
Only in the Black women group did SCN support have a moderating effect on the
relationship between objective stress and depression. This suggests that SCN support may
protect Black women's mental health from the stress of caregiving. This finding supports the
hypothesis that primary caregivers who are Black women are more likely to benefit from SCN
support. One possible explanation is that the gendered role of caregiver, coupled with Black
families' emphasis on family caregiving and the lack of external support resources relative to
Whites, makes SCN support the main source of social support for Black women caregivers
(Rozario & DeRienzis, 2008; Watson-Singleton, 2017). Black women are considered to provide
the most intensive care to older adults, and instrumental support from the SCN can greatly share
caregiving intensity and protect the physical and mental health of Black women primary
caregivers (Lawrence et al., 2008).
The analysis also found that SCN support was able to moderate the relationship between
subjective stress and social isolation in White men group. It was found in the previous analysis
that men and White primary caregivers are more likely to benefit from a high level of SCN
115
support on this pathway. It may be because men and White caregivers are generally considered
to have lower caregiving engagement and caregiving intensity than women and Black caregivers,
respectively (Akpınar et al., 2011; Martin, 2000). Therefore, White men caregivers might be
more likely to engage in social activities when obtaining instrumental support from SCN.
Another explanation is that White men caregivers may have smaller informal social networks
and are less able to obtain emotional support under caregiver stress than Black women caregivers
(Lopez–Anuarbe & Kohli, 2019; Robinson et al., 2014). Therefore, the presence of SCN in
caregiving can significantly help reduce loneliness.
Combined with the above findings, the buffering effects of SCN support on caregiving
stress are reflected differently in primary caregivers across gender-race groups. Given the role of
primary caregivers, the commitment to caregiving responsibility, the participation in caregiving
activities, and the difference in their own socioeconomic status, the impact mechanism of SCN
support is still understudied. More quantitative research is needed to establish robust analytic
models, as well as qualitative research to explore caregivers' subjective perceptions toward
different types of social support.
9.5 Research innovations and implications
This study contains numerous research design and methodology innovations that cleverly
solve some difficulties in the field of family caregiving research and avoid some issues with
using secondary data in analysis and modeling. The majority of the findings are consistent with
theoretical hypotheses, providing important empirical evidence for the study of family
caregiving, particularly the mechanism of cooperation and social support among family
116
caregivers. Furthermore, the findings may have implications for policy development as well as
community-based social work practice.
9.5.1 Research innovations
The dissertation study has three major innovations in design and methodology. First, this
study adopted a longitudinal perspective to explore the social isolation mediated stress process in
family caregivers. The two-wave mediation model, the simplest longitudinal mediation model,
served as an appropriate approach to examine causal relationship between variables within the
stress process (Valente et al., 2021). It also serves as good approach to explore the moderation
effect of different levels of SCN support on primary caregiver’s stress process. SCN support, as
an important component of social support for family caregivers, has the dynamics of change
longitudinally. Using data from two closely adjacent waves maximizes the stability of the SCN
support obtained by primary caregivers, and thus enables longitudinal evaluation on its
moderation effects. The interval between the two waves used in this study is 2 years, which is
shorter than previous studies reported 4 years of average caregiving duration, and it can be
considered that caregiver networks remained relatively stable during this period (Family
Caregiver Alliance, 2019).
Second, to our knowledge, no study has used latent profile analysis identified subgroups
to describe the different patterns of SCN support. LPA allows researchers to characterize
qualitative differences between support received by primary caregivers, which was missed in
previous studies that used several indicators to reflect secondary caregiver support respectively.
The results of this study found that the three indicators of SCN support (i.e., SCN size, number
of care domains, care intensity per secondary caregiver) were highly correlated, and identified
three latent groups that can be described simply as low, medium, and high SCN support. This
117
finding confirms that most secondary caregivers in this study were highly engaged in caregiving
tasks, that is, there may be few cases where SCN size is large but engagement is low. The
identification of the latent group solves the difficulty of categorizing continuous variables in the
analysis, and avoids the time-consuming process of analyzing the moderation effects of the three
indicators separately.
Third, adopting an intersectional perspective adds theoretical attention to diversity in the
study, which is aligned with the calling for diversity and equality in the study of aging and
family caregiving (Dilworth-Anderson et al., 2020). The accumulation of empirical evidence can
support the development of research, intervention, practice, and public policy. Focusing solely
on basic conceptual models, such as the stress and coping models, while revealing the general
caregiving experience of family caregivers, fails to take into account the variance of
demographic and sociocultural diversity and their associations with caregiving experience of
family caregivers. This study considered gender, race, and relationship type, the three
demographic characteristics most likely to influence caregiver experience, and further examined
differences in the stress process of primary caregivers at the intersection of gender and race.
Significant findings indicate that the stress process and stress-reduction mechanisms of primary
caregivers with varied sociodemographic backgrounds might be different, which are critical to
achieving the social work practice aimed at mobilizing support resources effectively to give
family caregivers the most appropriate support. Future research on the stress-reduction
mechanisms of family caregiver should incorporate sociocultural theory to fully consider the
diversity of caregivers and the inclusivity of study design.
9.5.2 Practice and policy implications
118
The findings of this dissertation study have potent contributing to social work research
and practice, as well as policymaking. First, the findings of the study extend our understanding
of the fit between social environment of family caregivers and their physical and mental health.
Social isolation as a mediator can explain part of the effect of caregiving stress on the depression
of primary caregivers, which highlights the importance of interventions aimed at promoting
caregivers' social connection and reducing loneliness. At present, most intervention programs to
help family caregivers reduce caregiver burden are primarily from the perspective of improving
caregiving skills and reducing caregiving intensity. For example, some intervention programs
provide psychoeducation to family caregivers to improve their health literacy, sense of mastery,
and skills of maintaining their own mental health (Aranda MP et al., 2007; Hepburn et al., 2007;
Kwok et al., 2013). There are also programs that provide solutions to caregiver’s difficulties,
such as connecting with formal service/respite care, referring to professional assistance, and
providing assistive technology (Gilmour, 2002; Holm & Ziguras, 2003; Lariviere et al., 2021).
However, only a few programs pay attention and try to improve the social isolation of family
caregivers. Some studies attempted to provide family caregivers with digital tools to help
connect with peers for informal social support, which showed the potential to help caregivers
regain a sense of social inclusion and belongings (Newman et al., 2019). When family caregivers
are burdened with onerous caregiving responsibilities and cannot participate in in-person social
activities, web-based social participation may address some of the social needs. Future research
could try to examine the compensation of online socialization for the social needs of family
caregivers, especially those formal and meaningful activities, such as church and volunteer
activities.
119
Second, the findings of this study advance the understanding of the cooperative mode
between primary caregivers and secondary caregivers. SCN is not equivalent to a general social
network, which is confirmed in this study. Social work practices or intervention programs that
target at promoting primary caregivers' access and use of support from social network should
help primary caregivers differentiate secondary caregivers from general social network members,
which would make it easier to form a caregiver network and get help from the appropriate
persons. Secondary caregivers and primary caregivers have similar backgrounds, such as similar
living environment, provide assistance to the same older adult, and understand each other's
caregiving styles. In addition to providing instrumental support to primary caregivers, SCN can
also provide emotional and informational support by sharing experience and information.
Although this study found that SCN support can buffer the negative impacts of caregiving stress
on the social isolation and depression of primary caregivers, its specific mechanism is not clear
enough, which requires more quantitative and qualitative exploration. The role of SCN should be
emphasized in social work practice, taking them into account when developing services and
interventions. For example, psychoeducation programs could include modules on family
resource mapping and help-seeking to teach family caregivers how to work together more
effectively, which may contribute to the caregivers' health and quality of care.
Third, the intersectional approach identified the differences in the demand for and the
effect of SCN support by primary caregivers of different gender-race groups. When Black
women caregivers received high levels of SCN support, the effects of caregiving stress on their
mental health were attenuated. This finding underscores the importance of considering diversity
of caregivers' sociodemographic and cultural backgrounds when designing research and practice.
For example, gendered role of caregiving and Black families' emphasis on familism prompts
120
Black women to internalize caregiving for the older adults at home as an inescapable
responsibility, which to a certain extent reduces their motivation to seek instrumental support and
brings additional burdens on mental health (Dilworth-Anderson et al., 2005). Modules that aim
to change perceptions of caregiving responsibility and encourage seeking informal support from
SCN should be included in the intervention program for Black women caregivers.
9.6 Limitations
The dissertation study also has some limitations. First, although this is a longitudinal
design, only the data of the two waves are available for analysis, so the interpretation of causal
relationship generated by the study results might be limited. Second, some of the measurements
used in this study, such as social isolation, are constructed based on theories, and no previous
studies have tested the reliability and validity of the composite social isolation framework among
the population of caregivers. This study uses the structural equation modeling to construct a
latent variable to predict the items of social isolation, and good model fit provides evidence that
partially supports the validity of the measurement. More research is needed to validate this
measurement in order to provide scientific and reliable tools for the research of caregiving stress
process. Third, while LPA can quantitatively reflect primary caregivers' SCN support using the
three indicators (i.e., SCN size, SCN domains, and SCN care hours), it cannot qualitatively
reflect the quality of caregiver collaboration, such as conflicts among caregivers and satisfaction
with the collaborative relationship. A possible SCN support pattern is large SCN but more
conflicts/dissatisfaction. Including quality-related variables perceived by primary caregivers in
LPA may aid in identifying heterogeneity of SCN support patterns. Fourth, this study only
examined the impacts on the two core mental health components, i.e., depression and anxiety.
121
Other metrics related to psychological well-being of family caregivers, such as subjective
happiness, satisfaction, and resilience, should be included in the model to examine the primary
caregiver's experience more thoroughly. Last, although the sample characteristics of this study
are not much different from the original NSOC samples, the process of data cleaning excluded
racial/ethnic minority family caregivers other than non-Hispanic Black, which may limit the
generalizability of research findings to caregivers of other minority groups. It is necessary for
future research to use more inclusive samples to examine the social isolation mediated stress
process of caregivers of other racial/ethnic minority groups, as well as cross-cultural differences
in the effect and mechanism of SCN support on caregiver experience.
9.7 Conclusion
This dissertation study examines the caregiving experience and SCN support received by
primary family caregivers of older adults, using a longitudinal research approach and an
intersectional perspective. The findings made important empirical and theoretical contributions
to studies on family caregiving. More specifically, the dissertation 1) added to our understanding
of primary caregiver stress by examining the mediated effect of social isolation on the impact of
caregiving stress on primary caregivers' mental health symptoms; 2) demonstrated how SCN
support can protect primary caregivers' social functioning and mental health from caregiving
stress; and 3) examined the differences of both social isolation mediated stress process and SCN
support in primary caregivers of varied genders, races, and relationship types, as well as the
intersection of gender and race. By promoting recognition of the importance of secondary
caregivers among families, service providers, and policy makers, and the potential benefits of
mutual support and cooperation between multiple family caregivers, the findings of the
122
dissertation study will have implications for future research design, social work
practice/intervention development, and public policy making.
123
REFERENCES
Adelman, R. D., Tmanova, L. L., Delgado, D., Dion, S., & Lachs, M. S. (2014). Caregiver
Burden. JAMA, 311(10), 1052. https://doi.org/10.1001/jama.2014.304
Administration for Community Living. (2021). 2020 Profile of Older Americans.
Akpınar, B., Küçükgüçlü, Ö., & Yener, G. (2011). Effects of Gender on Burden Among
Caregivers of Alzheimer’s Patients. Journal of Nursing Scholarship, 43(3), no-no.
https://doi.org/10.1111/j.1547-5069.2011.01402.x
Allen, S. M., Goldscheider, F., & Ciambrone, D. A. (1999). Gender Roles, Marital Intimacy, and
Nomination of Spouse as Primary Caregiver. The Gerontologist, 39(2), 150–158.
https://doi.org/10.1093/geront/39.2.150
Allen, S. M., Lima, J. C., Goldscheider, F. K., & Roy, J. (2012). Primary Caregiver
Characteristics and Transitions in Community-Based Care. The Journals of Gerontology
Series B: Psychological Sciences and Social Sciences, 67B(3), 362–371.
https://doi.org/10.1093/geronb/gbs032
Andersson, M. A., & Monin, J. K. (2017). Informal Care Networks in the Context of
Multimorbidity. Journal of Aging and Health, 898264316687623.
https://doi.org/10.1177/0898264316687623
Aranda MP, M. C., Aranda, M., & Morano, C. (2007). Psychoeducational strategies for Latino
caregivers. In C. Cox (Ed.), Dementia and Social Work Practice: Research and
Interventions (pp. 189–204). Springer Publishing.
Beeson, R. A. (2003). Loneliness and depression in spousal caregivers of those with Alzheimer’s
disease versus non-caregiving spouses. Archives of Psychiatric Nursing, 17(3), 135–143.
124
Bertinato, J., Sherrard, L., & Plouffe, L. J. (2012). EDTA disodium zinc has superior
bioavailability compared to common inorganic or chelated zinc compounds in rats fed a
high phytic acid diet. J Trace Elem Med Biol, 26(4), 227–233.
https://doi.org/10.1016/j.jtemb.2012.04.008
Black, B. S., Johnston, D., Rabins, P. V., Morrison, A., Lyketsos, C., & Samus, Q. M. (2013).
Unmet Needs of Community-Residing Persons with Dementia and Their Informal
Caregivers: Findings from the Maximizing Independence at Home Study. Journal of the
American Geriatrics Society, 61(12), 2087–2095. https://doi.org/10.1111/JGS.12549
Blazer, D. (2020). Social Isolation and Loneliness in Older Adults—A Mental Health/Public
Health Challenge. JAMA Psychiatry. https://doi.org/10.1001/jamapsychiatry.2020.1054
Bradley, E. H., Prigerson, H., Carlson, M. D., Cherlin, E., Johnson-Hurzeler, R., & Kasl, S. V.
(2004). Depression Among Surviving Caregivers: Does Length of Hospice Enrollment
Matter? The American Journal of Psychiatry, 161(12), 2257–2262.
Brand, C., Barry, L., & Gallagher, S. (2016). Social support mediates the association between
benefit finding and quality of life in caregivers. Journal of Health Psychology, 21(6),
1126–1136. https://doi.org/10.1177/1359105314547244
Braun, M., Scholz, U., Bailey, B., Perren, S., Hornung, R., & Martin, M. (2009). Dementia
caregiving in spousal relationships: A dyadic perspective. Aging & Mental Health, 13(3),
426–436. https://doi.org/10.1080/13607860902879441
Brown, L. L., Mitchell, U. A., & Ailshire, J. A. (2020). Disentangling the Stress Process:
Race/Ethnic Differences in the Exposure and Appraisal of Chronic Stressors Among
Older Adults. The Journals of Gerontology: Series B, 75(3), 650–660.
https://doi.org/10.1093/geronb/gby072
125
Buchanan, R. J., & Huang, C. (2012). Caregiver perceptions of accomplishment from assisting
people with multiple sclerosis. Disability and Rehabilitation, 34(1), 53–61.
https://doi.org/10.3109/09638288.2011.587091
Bullock, K., Crawford, S. L., & Tennstedt, S. L. (2003). Employment and Caregiving:
Exploration of African American Caregivers. Social Work, 48(2), 150–162.
Burgener, S., & Twigg, P. (2002). Relationships Among Caregiver Factors and Quality of Life in
Care Recipients with Irreversible Dementia. Alzheimer Disease & Associated Disorders,
16(2), 88–102.
Cantor, M. H. (1983). Strain Among Caregivers: A Study of Experience in the United States. The
Gerontologist, 23(6), 597–604. https://doi.org/10.1093/geront/23.6.597
Carretero, S., Garcés, J., Ródenas, F., & Sanjosé, V. (2009). The informal caregiver’s burden of
dependent people: Theory and empirical review. Archives of Gerontology and Geriatrics,
49(1), 74–79.
Chamberlain, S. A., Bronskill, S. E., Hsu, Z., Youngson, E., & Gruneir, A. (2022). Resident
loneliness, social isolation and unplanned emergency department visits from supportive
living facilities: A population-based study in Alberta, Canada. BMC Geriatrics, 22(1), 1–
11. https://doi.org/10.1186/S12877-021-02718-5/TABLES/3
Chappell, N. L., Dujela, C., & Smith, A. (2014). Spouse and Adult Child Differences in
Caregiving Burden. Canadian Journal on Aging / La Revue Canadienne Du
Vieillissement, 33(4), 35. https://doi.org/10.1017/S0714980814000336
Chappell, N. L., Dujela, C., & Smith, A. (2015). Caregiver Well-Being: Intersections of
Relationship and Gender. Research on Aging, 37(6), 623–645.
https://doi.org/10.1177/0164027514549258
126
Chen, F., Mair, C. A., Bao, L., & Yang, Y. C. (2015). Race/Ethnic Differentials in the Health
Consequences of Caring for Grandchildren for Grandparents. The Journals of
Gerontology Series B: Psychological Sciences and Social Sciences, 70(5), 793–803.
https://doi.org/10.1093/geronb/gbu160
Cheng, S.-T., Mak, E. P. M., Lau, R. W. L., Ng, N. S. S., & Lam, L. C. W. (2016). Voices of
Alzheimer Caregivers on Positive Aspects of Caregiving. The Gerontologist, 56(3), 451–
460. https://doi.org/10.1093/geront/gnu118
Cho, S., Zarit, S. H., & Chiriboga, D. A. (2009). Wives and Daughters: The Differential Role of
Day Care Use in the Nursing Home Placement of Cognitively Impaired Family Members.
The Gerontologist, 49(1), 57–67. https://doi.org/10.1093/GERONT/GNP010
Chyr, L. C., Drabo, E. F., & Fabius, C. D. (2020). Patterns and Predictors of Transitions across
Residential Care Settings and Nursing Homes Among Community-Dwelling Older
Adults in the United States. The Gerontologist. https://doi.org/10.1093/geront/gnaa070
Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis.
Psychological Bulletin, 98(2), 310–357. https://doi.org/10.1037/0033-2909.98.2.310
Conde-Sala, J. L., Garre-Olmo, J., Turró-Garriga, O., Vilalta-Franch, J., & López-Pousa, S.
(2010). Differential features of burden between spouse and adult-child caregivers of
patients with Alzheimer’s disease: An exploratory comparative design. International
Journal of Nursing Studies, 47(10), 1262–1273.
https://doi.org/10.1016/J.IJNURSTU.2010.03.001
Cornwell, E. Y., & Waite, L. J. (2009). Social disconnectedness, perceived isolation, and health
among older adults. Journal of Health and Social Behavior, 50(1), 31–48.
https://doi.org/10.1177/002214650905000103
127
Cox, C., & Monk, A. (1990). Minority caregivers of dementia victims: A comparison of Black
and Hispanic families. Journal of Applied Gerontology, 9(3), 340–354.
Coyle, C. E., Gleason, S. R., & Mutchler, J. E. (2021). Spillover benefits and achieving
sustainability of age-friendly communities. The Gerontologist.
https://doi.org/10.1093/geront/gnab060
Crenshaw, K. (1989). Demarginalizing the Intersection of Race and Sex: A Black Feminist
Critique of Antidiscrimination Doctrine, Feminist Theory and Antiracist Politics.
University of Chicago Legal Forum, 1989.
Crenshaw, K. (1991). Mapping the margins: Identity politics, intersectionality, and violence
against women. Stanford Law Review, 43(6), 1241–1299.
Crespo, M., Lopez, J., & Zarit, S. H. (2005). Depression and anxiety in primary caregivers: A
comparative study of caregivers of demented and nondemented older persons.
International Journal of Geriatric Psychiatry, 20, 591–592.
Cudjoe, T. K. M., Roth, D. L., Szanton, S. L., Wolff, J. L., Boyd, C. M., & Thorpe, R. J. (2020).
The Epidemiology of Social Isolation: National Health and Aging Trends Study. Journals
of Gerontology - Series B Psychological Sciences and Social Sciences, 75(1), 107–113.
https://doi.org/10.1093/geronb/gby037
Cynthia, C., Thunell, J., & Zissimopoulos, J. (2020). Changes in physical and mental health of
Black, Hispanic, and White caregivers and non‐caregivers associated with onset of
spousal dementia. Alzheimer’s & Dementia: Translational Research & Clinical
Interventions, 6(1), e12082.
Dam, A. E. H., de Vugt, M. E., Klinkenberg, I. P. M., Verhey, F. R. J., & van Boxtel, M. P. J.
(2016). A systematic review of social support interventions for caregivers of people with
128
dementia: Are they doing what they promise? Maturitas, 85, 117–130.
https://doi.org/10.1016/J.MATURITAS.2015.12.008
De Jong Gierveld, J., Van der Pas, S., & Keating, N. (2015). Loneliness of Older Immigrant
Groups in Canada: Effects of Ethnic-Cultural Background. Journal of Cross-Cultural
Gerontology, 30(3), 251–268. https://doi.org/10.1007/S10823-015-9265-X/TABLES/2
Dean, A., Kolody, B., Wood, P., & Matt, G. E. (2016). The Influence of Living Alone on
Depression in Elderly Persons:
Http://Dx.Doi.Org.Libproxy2.Usc.Edu/10.1177/089826439200400101, 4(1), 3–18.
https://doi.org/10.1177/089826439200400101
Depalma, G., Xu, H., Covinsky, K. E., Craig, B. A., Stallard, E., Thomas, J., & Sands, L. P.
(2013). Hospital readmission among older adults who return home with unmet need for
ADL disability. Gerontologist, 53(3), 454–461. https://doi.org/10.1093/geront/gns103
Dill, J., & Duffy, M. (2022). Structural Racism And Black Women’s Employment In The US
Health Care Sector. Health Affairs, 41(2), 265–272.
https://doi.org/10.1377/hlthaff.2021.01400
Dilworth-Anderson, P., Brummett, B. H., Goodwin, P., Williams, S. W., Williams, R. B., &
Siegler, I. C. (2005). Effect of Race on Cultural Justifications for Caregiving. The
Journals of Gerontology: Series B, 60(5), S257–S262.
https://doi.org/10.1093/geronb/60.5.S257
Dilworth-Anderson, P., & Cohen, M. D. (2010). Beyond diversity to inclusion: Recruitment and
retention of diverse groups in Alzheimer research. Alzheimer Disease & Associated
Disorders, 24, S14–S18.
129
Dilworth-Anderson, P., & Gibson, B. E. (2002). The Cultural Influence of Values, Norms,
Meanings, and Perceptions in Understanding Dementia in Ethnic Minorities. Alzheimer
Disease & Associated Disorders, Volume 16(2), S56–S63.
Dilworth-Anderson, P., Moon, H., & Aranda, M. P. (2020). Dementia caregiving research:
Expanding and reframing the lens of diversity, inclusivity, and intersectionality. The
Gerontologist. https://doi.org/10.1093/geront/gnaa050. In press
Dilworth-Anderson, P., Wallace, W. S., Cooper, T., Williams, S. W., & Cooper, T. (1999).
Family Caregiving to Elderly African Americans: Caregiver Types and Structures. The
Journals of Gerentology, 54B(4), S237–S241. https://doi.org/10.1093/geronb/54B.4.S237
Dilworth-Anderson, P., Williams, I. C., & Gibson, B. E. (2002). Issues of race, ethnicity, and
culture in caregiving research: A 20-year review (1980-2000). The Gerontologist, 42(2),
237–272. https://doi.org/10.1093/geront/42.2.237
Do, E. K., Cohen, S. A., & Brown, M. J. (2014). Socioeconomic and demographic factors
modify the association between informal caregiving and health in the Sandwich
Generation. BMC Public Health, 14(1), 1–8. https://doi.org/10.1186/1471-2458-14-
362/TABLES/3
dos Santos Azevedo, L. V., Calandri, I. L., Slachevsky, A., Graviotto, H. G., Santos Vieira, M.
C., Andrade, C. B. de, Rossetti, A. P., Generoso, A. B., Carmona, K. C., & Pinto, L. A.
C. (2021). Impact of Social Isolation on People with Dementia and Their Family
Caregivers. Journal of Alzheimer’s Disease, Preprint, 1–11.
Ducharme, F., Lévesque Louise, L., Lachance, L., Kergoat, M. J., & Coulombe, R. (2011).
Challenges associated with transition to caregiver role following diagnostic disclosure of
130
Alzheimer disease: A descriptive study. International Journal of Nursing Studies, 48(9),
1109–1119. https://doi.org/10.1016/J.IJNURSTU.2011.02.011
Elizabeth, G., & Rickenbach, E. H. (2019). Perceptions of couplehood among community-
dwelling spousal caregivers. Aging & Mental Health, 1–8.
https://doi.org/10.1080/13607863.2019.1594168
Evans, R., Catapano, M. A., Brooks, D., Goldstein, R. S., & Avendano, M. (2012). Family
caregiver perspectives on caring for ventilator-assisted individuals at home. Canadian
Respiratory Journal : Journal of the Canadian Thoracic Society, 19(6), 373–379.
https://doi.org/10.1155/2012/452898
Everhart, R. S., Greenlee, J. L., Winter, M. A., & Fiese, B. H. (2018). Primary and Secondary
Caregiver Reports of Quality of Life in Pediatric Asthma: Are they Comparable? Applied
Research in Quality of Life, 13(2), 371–383. https://doi.org/10.1007/S11482-017-9528-
5/TABLES/4
Fabius, C. D., Wolff, J. L., & Kasper, J. D. (2020). Race Differences in Characteristics and
Experiences of Black and White Caregivers of Older Americans. The Gerontologist,
60(7), 1244–1253. https://doi.org/10.1093/geront/gnaa042
Falcão, D. V. D. S., Teodoro, M. L. M., & Bucher-Maluschke, J. S. N. F. (2016). Family
cohesion: A study on caregiving daughters of parents with alzheimer’s disease.
Interpersona, 10(supp1), 61–74. https://doi.org/10.5964/ijpr.v10isupp1.244
Falzarano, F., Moxley, J., Pillemer, K., & Czaja, S. J. (2021). Family Matters: Cross-Cultural
Differences in Familism and Caregiving Outcomes. The Journals of Gerontology: Series
B. https://doi.org/10.1093/geronb/gbab160
131
Family Caregiver Alliance. (2019). Caregiver Statistics: Work and Caregiving. Family
Caregiver Alliance. https://www.caregiver.org/resource/caregiver-statistics-work-and-
caregiving/
Feld, S., Dunkle, R. E., & Schroepfer, T. (2012). When do couples expand their ADL caregiver
network beyond the Marital Dyad? In Challenges of Aging on U.S. Families: Policy and
Practice Implications (pp. 27–44). Taylor and Francis.
https://doi.org/10.1300/J002v37n01_04
Feter, N., Caputo, E. L., Smith, E. C., Doring, I. R., Cassuriaga, J., Leite, J. S., Reichert, F. F., da
Silva, M. C., Coombes, J. S., & Rombaldi, A. J. (2021). Association between physical
activity and subjective memory decline triggered by the COVID-19 pandemic: Findings
from the PAMPA cohort. Preventive Medicine, 145, 106415.
Fredman, L., Ranker, L. R., Strunin, L., Smith, M. L., & Applebaum, K. M. (2019). Caregiving
Intensity and Mortality in Older Women, Accounting for Time-Varying and Lagged
Caregiver Status: The Caregiver-Study of Osteoporotic Fractures Study. The
Gerontologist. https://doi.org/10.1093/geront/gny135
Freedman, V. A., Skehan, M. E., Wolff, J., & Kasper, J. D. (2019). National Study of Caregiving
(NSOC) I-III User Guide. Johns Hopkins Bloomberg School of Public Health.
Freedman, V. A., & Spillman, B. C. (2014). The residential continuum from home to nursing
home: Size, characteristics and unmet needs of older adults. The Journals of
Gerontology. Series B, Psychological Sciences and Social Sciences, 69(Suppl 1), S42–
S50. https://doi.org/10.1093/geronb/gbu120
Fried, E. I., Bockting, C., Arjadi, R., Borsboom, D., Amshoff, M., Cramer, A. O. J., Epskamp,
S., Tuerlinckx, F., Carr, D., & Stroebe, M. (2015). From loss to loneliness: The
132
relationship between bereavement and depressive symptoms. Journal of Abnormal
Psychology, 124(2), 256–265. https://doi.org/10.1037/ABN0000028
Friedemann, M.-L., & Buckwalter, K. C. (2014). Family Caregiver Role and Burden Related to
Gender and Family Relationships. Journal of Family Nursing, 20(3), 313–336.
https://doi.org/10.1177/1074840714532715
Gallagher, S., & Wetherell, M. A. (2020). Risk of depression in family caregivers: Unintended
consequence of COVID-19. BJPsych Open, 6(6). https://doi.org/10.1192/BJO.2020.99
Gallanis, T. (2019). The Effect of Activities of Daily Living on Family Caregiver Social
Isolation. Innovation in Aging, 3(Suppl 1), S899.
https://doi.org/10.1093/GERONI/IGZ038.3286
Galvin, J. E., Duda, J. E., Kaufer, D. I., Lippa, C. F., Taylor, A., & Zarit, S. H. (2010). Lewy
body dementia: Caregiver burden and unmet needs. Alzheimer Disease and Associated
Disorders, 24(2), 177–181. https://doi.org/10.1097/WAD.0B013E3181C72B5D
García-Alberca, J. M., Cruz, B., Lara, J. P., Garrido, V., Lara, A., & Gris, E. (2012). Anxiety and
depression are associated with coping strategies in caregivers of Alzheimer’s disease
patients: Results from the MÁLAGA-AD study. International Psychogeriatrics, 24(08),
1325–1334. https://doi.org/10.1017/S1041610211002948
Gaugler, J. E., Mendiondo, M., Smith, C. D., & Schmitt, F. A. (2003). Secondary dementia
caregiving and its consequences. American Journal of Alzheimer’s Disease and Other
Dementias. https://doi.org/10.1177/153331750301800505
Ge, L., Yap, C. W., & Heng, B. H. (2022). Associations of social isolation, social participation,
and loneliness with frailty in older adults in Singapore: A panel data analysis. BMC
Geriatrics 2022 22:1, 22(1), 1–10. https://doi.org/10.1186/S12877-021-02745-2
133
George, L. K., & Gwyther, L. P. (1986). Caregiver Weil-Being: A Multidimensional
Examination of Family Caregivers of Demented Adults. The Gerontologist, 26(3), 253–
259. https://doi.org/10.1093/geront/26.3.253
Ghosh, S., Choi, W. Y., Williams, A., Duggleby, W., Ploeg, J., Markle-Reid, M., & Peacock, S.
(2020). Burden on Caregivers of Adults with Multiple Chronic Conditions:
Intersectionality of Age, Gender, Education level, Employment Status, and Impact on
Social Life. Canadian Journal on Aging, 39(3), 456–467.
https://doi.org/10.1017/S071498081900045X
Gilmour, J. A. (2002). Dis/integrated care: Family caregivers and in-hospital respite care.
Journal of Advanced Nursing, 39(6), 546–553. https://doi.org/10.1046/j.1365-
2648.2002.02323.x
Gonçalves-Pereira, M., Zarit, S. H., Cardoso, A. M., Alves da Silva, J., Papoila, A. L., & Mateos,
R. (2020). A comparison of primary and secondary caregivers of persons with dementia.
Psychology and Aging, 35(1), 20–27. https://doi.org/10.1037/pag0000380
Goren, A., Montgomery, W., Kahle-Wrobleski, K., Nakamura, T., & Ueda, K. (2016). Impact of
caring for persons with Alzheimer’s disease or dementia on caregivers’ health outcomes:
Findings from a community based survey in Japan. BMC Geriatrics, 16(1), 1–12.
Greenwood, N., Pound, C., Brearley, S., & Smith, R. (2019). A qualitative study of older
informal carers’ experiences and perceptions of their caring role. Maturitas, 124, 1–7.
https://doi.org/10.1016/J.MATURITAS.2019.03.006
Hajek, A., & König, H. H. (2016). Informal Caregiving and Subjective Well-Being: Evidence of
a Population-Based Longitudinal Study of Older Adults in Germany. Journal of the
134
American Medical Directors Association, 17(4), 300–305.
https://doi.org/10.1016/J.JAMDA.2015.10.015
Hajek, A., & König, H. H. (2018). The relation between personality, informal caregiving, life
satisfaction and health-related quality of life: Evidence of a longitudinal study. Quality of
Life Research, 27(5), 1249–1256. https://doi.org/10.1007/S11136-018-1787-
6/TABLES/3
Hajek, A., & König, H. H. (2022). How does beginning or ceasing of informal caregiving of
individuals in poor health influence sleep quality? Findings from a nationally
representative longitudinal study. Aging & Mental Health, 26(2), 314–319.
https://doi.org/10.1080/13607863.2020.1857700
Hajek, A., Kretzler, B., & König, H. H. (2021). Informal Caregiving, Loneliness and Social
Isolation: A Systematic Review. International Journal of Environmental Research and
Public Health 2021, Vol. 18, Page 12101, 18(22), 12101.
https://doi.org/10.3390/IJERPH182212101
Haley, W. E., Levine, E. G., Brown, S. L., & Bartolucci, A. A. (1987). Stress, appraisal, coping,
and social support as predictors of adaptational outcome among dementia caregivers.
Psychology and Aging, 2(4), 323–330. https://doi.org/10.1037//0882-7974.2.4.323
Haley, W. E., Roth, D. L., Howard, G., & Safford, M. M. (2010). Caregiving Strain and
Estimated Risk for Stroke and Coronary Heart Disease Among Spouse Caregivers.
Stroke, 41(2), 331–336. https://doi.org/10.1161/STROKEAHA.109.568279
Hallikainen, I., Koivisto, A. M., & Välimäki, T. (2018). The influence of the individual
neuropsychiatric symptoms of people with Alzheimer disease on family caregiver
135
distress-A longitudinal ALSOVA study. International Journal of Geriatric Psychiatry,
33(9), 1207–1212. https://doi.org/10.1002/gps.4911
Halpern, M. T., Fiero, M. H., & Bell, M. L. (2017). Impact of caregiver activities and social
supports on multidimensional caregiver burden: Analyses from nationally-representative
surveys of cancer patients and their caregivers. Quality of Life Research, 26(6), 1587–
1595. https://doi.org/10.1007/s11136-017-1505-9
Hamid, A. A. R. M., & Musa, S. A. (2017). The mediating effects of coping strategies on the
relationship between secondary traumatic stress and burnout in professional caregivers in
the UAE. Journal of Mental Health, 26(1), 28–35.
https://doi.org/10.1080/09638237.2016.1244714
Han, J., Kim, K. W., Kim, T. H., Jeong, H., Park, J. Y., Youn, J., Park, J. H., Jhoo, J. H., Lee, J.
J., Kim, J.-L., Kim, B.-J., Kim, S. G., Ryu, S.-H., Choo, I. H., Lee, D. Y., Lee, D. W.,
Lee, S. B., Kwak, K. P., & Kang, S. Y. (2012). The impact of social support on caregiver
burden in dementia. Alzheimer’s & Dementia, 8(4), P249.
https://doi.org/10.1016/j.jalz.2012.05.661
Harris-Kojetin, L. D., Sengupta, M., Lendon, J. P., Rome, V., Valverde, R., & Caffrey, C.
(2019). Long-term care providers and services users in the United States, 2015-2016.
Vital and Health Statistics, 3(43), 2019–1427.
Hawton, A., Green, C., Dickens, A. P., Richards, S. H., Taylor, R. S., Edwards, R., Greaves, C.
J., & Campbell, J. L. (2010). The impact of social isolation on the health status and
health-related quality of life of older people. Quality of Life Research 2010 20:1, 20(1),
57–67. https://doi.org/10.1007/S11136-010-9717-2
136
Hayes, L., Hawthorne, G., Farhall, J., O’Hanlon, B., & Harvey, C. (2015). Quality of Life and
Social Isolation Among Caregivers of Adults with Schizophrenia: Policy and Outcomes.
Community Mental Health Journal 2015 51:5, 51(5), 591–597.
https://doi.org/10.1007/S10597-015-9848-6
He, S., Craig, B. A., Xu, H., Covinsky, K. E., Stallard, E., Thomas, J., Hass, Z., & Sands, L. P.
(2015). Unmet need for ADL assistance is associated with mortality among older adults
with mild disability. Journals of Gerontology - Series A Biological Sciences and Medical
Sciences, 70(9), 1128–1132. https://doi.org/10.1093/gerona/glv028
Heehyul, M. E., Haley, W. E., Rote, S. M., & Sears, J. S. (2020). Caregiver Well-Being and
Burden: Variations by Race/Ethnicity and Care Recipient Nativity Status. Innovation in
Aging, 4(6), 1–13. https://doi.org/10.1093/GERONI/IGAA045
Henderson, A. K., Walsemann, K. M., & Ailshire, J. A. (2021). Religious Involvement and
Cognitive Functioning at the Intersection of Race/Ethnicity and Gender in Mid-Life and
Older Adults. The Journals of Gerontology: Series B.
https://doi.org/10.1093/geronb/gbab034
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.
Her, Y. W., Shin, H., & Pae, S. (2019). A multigroup SEM analysis of moderating role of task
uncertainty on budgetary participation-performance relationship: Evidence from Korea.
Asia Pacific Management Review, 24(2), 140–153.
https://doi.org/10.1016/j.apmrv.2018.02.001
137
Hickendorff, M., Edelsbrunner, P. A., McMullen, J., Schneider, M., & Trezise, K. (2018).
Informative tools for characterizing individual differences in learning: Latent class, latent
profile, and latent transition analysis. Learning and Individual Differences, 66, 4–15.
https://doi.org/10.1016/j.lindif.2017.11.001
Hinojosa, R., Haun, J., Hinojosa, M. S., & Rittman, M. (2015). Social Isolation Poststroke:
Relationship Between Race/Ethnicity, Depression, and Functional Independence.
Https://Doi-Org.Libproxy2.Usc.Edu/10.1310/Tsr1801-79, 18(1), 79–86.
https://doi.org/10.1310/TSR1801-79
Holm, S., & Ziguras, S. (2003). The host-homes program: An innovative model of respite for
carers of people with dementia. Australasian Journal on Ageing, 22(3), 141–145.
https://doi.org/10.1111/j.1741-6612.2003.tb00485.x
Home, A. M. (1998). Predicting Role Conflict, Overload and Contagion in Adult Women
University Students with Families and Jobs. Adult Education Quarterly, 48(2), 85–97.
https://doi.org/10.1177/074171369804800204
Hong, G. R. S., & Kim, H. (2008). Family Caregiver Burden by Relationship to Care Recipient
with Dementia in Korea. Geriatric Nursing, 29(4), 267–274.
https://doi.org/10.1016/J.GERINURSE.2007.09.004
Hsiao, C.-Y. (2010). Family demands, social support and caregiver burden in Taiwanese family
caregivers living with mental illness: The role of family caregiver gender. Journal of
Clinical Nursing, 19(23–24), 3494–3503. https://doi.org/10.1111/j.1365-
2702.2010.03315.x
Hughes, M., & Gove, W. R. (1981). Living alone, social integration, and mental health. AJS;
American Journal of Sociology, 87(1), 48–74. https://doi.org/10.1086/227419
138
Huiying, L., Fang, B., Chan, J., & Chen, G. (2019). The relationship between comorbidities in
dementia patients and burden on adult–child primary caregivers: Does having a
secondary caregiver matter? International Journal of Mental Health Nursing, 28(6),
1306–1317. https://doi.org/10.1111/inm.12640
Hussain, R., Wark, S., & Ryan, P. (2018). Caregiving, Employment and Social Isolation:
Challenges for Rural Carers in Australia. International Journal of Environmental
Research and Public Health 2018, Vol. 15, Page 2267, 15(10), 2267.
https://doi.org/10.3390/IJERPH15102267
Jenkins Morales, M., & Robert, S. A. (2020). Black–White Disparities in Moves to Assisted
Living and Nursing Homes Among Older Medicare Beneficiaries. The Journals of
Gerontology: Series B, 75(9), 1972–1982. https://doi.org/10.1093/geronb/gbz141
Jinyu, L. (2021). Spouse and Adult-Child Dementia Caregivers in Chinese American Families:
Who Are More Stressed Out? Journal of the American Medical Directors Association.
https://doi.org/10.1016/j.jamda.2020.12.012
Jordan, P., Shedden-Mora, M. C., & Löwe, B. (2017). Psychometric analysis of the Generalized
Anxiety Disorder scale (GAD-7) in primary care using modern item response theory.
PloS One, 12(8), e0182162.
Kang, S. Y. (2008). Predictors of Emotional Strain Among Spouse and Adult Child Caregivers.
Http://Dx.Doi.Org/10.1300/J083v47n01_08, 47(1–2), 107–131.
https://doi.org/10.1300/J083V47N01_08
Karantzas, G. C., Evans, L., & Foddy, M. (2010). The role of attachment in current and future
parent caregiving. Journals of Gerontology - Series B Psychological Sciences and Social
Sciences, 65 B(5), 573–580. https://doi.org/10.1093/GERONB/GBQ047
139
Karen, R. M., Buckwalter, K. C., & Reed, D. (2005). Predictors of use of services among
dementia caregivers. Western Journal of Nursing Research, 27(2), 126–140.
https://doi.org/10.1177/0193945904272453
Kaschowitz, J., & Brandt, M. (2017). Health effects of informal caregiving across Europe: A
longitudinal approach. Social Science & Medicine, 173, 72–80.
https://doi.org/10.1016/J.SOCSCIMED.2016.11.036
Kasper, J. D., & Freedman, V. A. (2014). National Health and Aging Trends Study: User guide
(Round 1 and 2, Final Release). MD: Johns Hopkins University School of Public Health.
Kasper, J. D., Freedman, V. A., & Spillman, B. C. (2013). Classification of Persons by Dementia
Status in the National Health and Aging Trends Study.
Keim, M. C., Fladeboe, K., Galtieri, L. R., Kawamura, J., King, K., Friedman, D., Compas, B.,
Breiger, D., Lengua, L., & Katz, L. F. (2021). Primary and secondary caregiver
depressive symptoms and family functioning following a pediatric cancer diagnosis: An
exploration of the buffering hypothesis. Psycho-Oncology, 30(6), 928–935.
https://doi.org/10.1002/PON.5676
Kelley, D. E., Lewis, M. A., & Southwell, B. G. (2017). Perceived support from a caregiver’s
social ties predicts subsequent care-recipient health. Preventive Medicine Reports, 8,
108–111. https://doi.org/10.1016/J.PMEDR.2017.08.001
Khalaila, R. (2021). Caregiver Burden and Compassion Fatigue Among Arab Family Caregivers
of Older Relatives. Journal of Applied Gerontology, 40(7), 722–730.
https://doi.org/10.1177/0733464820920100
140
Kim, Y., Carver, C. S., Shaffer, K. M., Gansler, T., & Cannady, R. S. (2015). Cancer caregiving
predicts physical impairments: Roles of earlier caregiving stress and being a spousal
caregiver. Cancer, 121(2), 302–310. https://doi.org/10.1002/CNCR.29040
Kinzer, A., & Suhr, J. A. (2016). Dementia worry and its relationship to dementia exposure,
psychological factors, and subjective memory concerns.
https://doi.org/10.1080/23279095.2015.1030669
Kovaleva, M., Spangler, S., Clevenger, C., & Hepburn, K. (2018). Chronic stress, social
isolation, and perceived loneliness in dementia caregivers. Journal of Psychosocial
Nursing and Mental Health Services, 56(10), 36–43.
Kuster, P. A., & Merkel, C. J. (2004). Caregiving Stress, Immune Function, And Health:
Implications for Research with Parents of Medically Fragile Children. Issues in
Comprehensive Pediatric Nursing, 27(4), 257–276.
https://doi.org/10.1080/01460860490884165
Kwok, T., Wong, B., Ip, I., Chui, K., Young, D., & Ho, F. (2013). Telephone-delivered
psychoeducational intervention for Hong Kong Chinese dementia caregivers: A single-
blinded randomized controlled trial. Clinical Interventions in Aging, 8, 1191–1197.
https://doi.org/10.2147/CIA.S48264
Langa, K. M., Larson, E. B., Crimmins, E. M., Faul, J. D., Levine, D. A., Kabeto, M. U., &
Weir, D. R. (2017). A Comparison of the Prevalence of Dementia in the United States in
2000 and 2012. JAMA Internal Medicine, 177(1), 51–58.
https://doi.org/10.1001/jamainternmed.2016.6807
Lariviere, M., Poland, F., Woolham, J., Newman, S., & Fox, C. (2021). Placing assistive
technology and telecare in everyday practices of people with dementia and their
141
caregivers: Findings from an embedded ethnography of a national dementia trial. BMC
Geriatrics, 21(1), 121. https://doi.org/10.1186/s12877-020-01896-y
Lawrence, V., Murray, J., Samsi, K., & Banerjee, S. (2008). Attitudes and support needs of
Black Caribbean, South Asian and White British carers of people with dementia in the
UK. The British Journal of Psychiatry, 193, 240–246.
Li, L., Wister, A. V, & Mitchell, B. (2020). Social Isolation Among Spousal and Adult-Child
Caregivers: Findings From the Canadian Longitudinal Study on Aging. The Journals of
Gerontology: Series B, 76(7), 1415–1429. https://doi.org/10.1093/geronb/gbaa197
Liang, J., Aranda, M. P., & Lloyd, D. A. (2020). Association between Role Overload and Sleep
Disturbance among Dementia Caregivers: The Impact of Social Support and Social
Engagement. Journal of Aging and Health, 32(10), 1345–1354.
https://doi.org/10.1177/0898264320926062
Lillard, L. A., Linda, R., Waite, J., & Waite, L. J. (2015). ’Til Death Do Us Part: Marital
Disruption and Mortality. American Journal of Sociology, 100(5), 1131–1156.
https://doi.org/10.1086/230634
Lin, I.-F., Fee, H. R., & Wu, H.-S. (2012). Negative and Positive Caregiving Experiences: A
Closer Look at the Intersection of Gender and Relationship. Family Relations, 61(2),
343–358. https://doi.org/10.1111/j.1741-3729.2011.00692.x
Litzelman, K., Kent, E. E., & Rowland, J. H. (2016). Social Factors in Informal Cancer
Caregivers: The Interrelationships Among Social Stressors, Relationship Quality, and
Family Functioning in the CanCORS Data Set. Cancer, 122, 278–286.
https://doi.org/10.1002/cncr.29741
142
Liu, R., Chi, I., & Wu, S. (2021). Caregiving Burden Among Caregivers of People With
Dementia Through the Lens of Intersectionality. The Gerontologist, gnab146.
https://doi.org/10.1093/geront/gnab146
Liu, S., Li, C., Shi, Z., Wang, X., Zhou, Y., Liu, S., Liu, J., Yu, T., & Ji, Y. (2017). Caregiver
burden and prevalence of depression, anxiety and sleep disturbances in Alzheimer’s
disease caregivers in China. Journal of Clinical Nursing, 26(9–10), 1291–1300.
https://doi.org/10.1111/jocn.13601
Liu, S., Liu, J., Wang, X.-D., Shi, Z., Zhou, Y., Li, J., Yu, T., & Ji, Y. (2018). Caregiver burden,
sleep quality, depression, and anxiety in dementia caregivers: A comparison of
frontotemporal lobar degeneration, dementia with Lewy bodies, and Alzheimer’s disease.
International Psychogeriatrics, 30(8), 1131–1138.
https://doi.org/10.1017/S1041610217002630
Lloyd, J., Muers, J., Patterson, T. G., & Marczak, M. (2018). Self-Compassion, Coping
Strategies, and Caregiver Burden in Caregivers of People with Dementia. Clinical
Gerontologist. https://doi.org/10.1080/07317115.2018.1461162
Longacre, M. L., Valdmanis, V. G., Handorf, E. A., & Fang, C. Y. (2017). Work Impact and
Emotional Stress Among Informal Caregivers for Older Adults. The Journals of
Gerontology: Series B, 72(3), 522–531. https://doi.org/10.1093/GERONB/GBW027
Lopez–Anuarbe, M., & Kohli, P. (2019). Understanding Male Caregivers’ Emotional, Financial,
and Physical Burden in the United States. Healthcare, 7(2), 72.
https://doi.org/10.3390/healthcare7020072
143
Losada, A., Márquez-González, M., Peñacoba, C., & Romero-Moreno, R. (2010). Development
and validation of the Caregiver Guilt Questionnaire. International Psychogeriatrics,
22(4), 650–660. https://doi.org/10.1017/S1041610210000074
Lovell, B., & Wetherell, M. A. (2011). The cost of caregiving: Endocrine and immune
implications in elderly and non elderly caregivers. Neuroscience & Biobehavioral
Reviews, 35(6), 1342–1352. https://doi.org/10.1016/J.NEUBIOREV.2011.02.007
Lubben, J. E. (1988). Assessing social networks among elderly populations. Family &
Community Health, 11(3), 42–52.
Luth, E. A., & Pristavec, T. (2020). Do caregiver experiences shape end-of life care perceptions?
Burden,benefits, and care quality assessment. Journal of Pain and Symptom
Management, 59(1), 77. https://doi.org/10.1016/J.JPAINSYMMAN.2019.08.012
Magan, K. C., Hirschman, K. B., Starr, L. T., & Meghani, S. H. (2021). Longitudinal
Trajectories and Correlates of Post-Caregiving Depressive Symptoms in Former
Caregivers of Persons With Alzheimer’s Disease or a Related Dementia. American
Journal of Hospice and Palliative Medicine®, 10499091211042836.
https://doi.org/10.1177/10499091211042836
Mair, C. A. (2010). Social Ties and Depression: An Intersectional Examination of Black and
White Community-Dwelling Older Adults. Journal of Applied Gerontology, 29(6), 667–
696. https://doi.org/10.1177/0733464809350167
Manea, L., Gilbody, S., & McMillan, D. (2012). Optimal cut-off score for diagnosing depression
with the Patient Health Questionnaire (PHQ-9): A meta-analysis. CMAJ, 184(3).
144
Mannino, D. M., Thorn, D., Swensen, A., & Holguin, F. (2008). Prevalence and outcomes of
diabetes, hypertension and cardiovascular disease in COPD. European Respiratory
Journal, 32(4), 962–969. https://doi.org/10.1183/09031936.00012408
Marino, V. R., Badana, A. N. S., & Haley, W. E. (2020). Care Demands and Well-Being of
Primary and Secondary Non-Spousal Caregivers of Aging Adults.
Https://Doi.Org/10.1080/07317115.2020.1759748, 43(5), 558–571.
https://doi.org/10.1080/07317115.2020.1759748
Martin, C. D. (2000). More Than the Work: Race and Gender Differences in Caregiving Burden.
Journal of Family Issues, 21(8), 986–1005.
https://doi.org/10.1177/019251300021008003
Matthews, B. A., Baker, F., & Spillers, R. L. (2004). Family caregivers’ quality of life: Influence
of health protective stance and emotional strain. Psychology & Health, 19(5), 625–641.
https://doi.org/10.1080/0887044042000205594
Mengting, L., Mao, W., Chi, I., & Lou, V. W. Q. (2019). Geographical proximity and depressive
symptoms among adult child caregivers: Social support as a moderator. Aging and
Mental Health, 23(2), 205–213. https://doi.org/10.1080/13607863.2017.1399349
Miller, D. A. (1981). The ‘sandwich’generation: Adult children of the aging. Social Work, 26(5),
419–423.
Mills, P. J., Ancoli-Israel, S., Känel, R. von, Mausbach, B. T., Aschbacher, K., Patterson, T. L.,
Ziegler, M. G., Dimsdale, J. E., & Grant, I. (2009a). Effects of gender and dementia
severity on Alzheimer’s disease caregivers’ sleep and biomarkers of coagulation and
inflammation. Brain, Behavior, and Immunity, 23(5), 605–610.
https://doi.org/10.1016/J.BBI.2008.09.014
145
Mills, P. J., Ancoli-Israel, S., Känel, R. von, Mausbach, B. T., Aschbacher, K., Patterson, T. L.,
Ziegler, M. G., Dimsdale, J. E., & Grant, I. (2009b). Effects of gender and dementia
severity on Alzheimer’s disease caregivers’ sleep and biomarkers of coagulation and
inflammation. Brain, Behavior, and Immunity, 23(5), 605–610.
https://doi.org/10.1016/J.BBI.2008.09.014
Monahan, P. O., Shacham, E., Reece, M., Kroenke, K., Ong’Or, W. O., Omollo, O., Yebei, V.
N., & Ojwang, C. (2009). Validity/reliability of PHQ-9 and PHQ-2 depression scales
among adults living with HIV/AIDS in Western Kenya. Journal of General Internal
Medicine, 24(2), 189–197. https://doi.org/10.1007/s11606-008-0846-z
Monin, J. K., & Schulz, R. (2010). The effects of suffering in chronically ill older adults on the
health and well-being of family members involved in their care. GeroPsych: The Journal
of Gerontopsychology and Geriatric Psychiatry, 23(4), 207–213.
https://doi.org/10.1024/1662-9647/a000024
Moon, H., Rote, S., & Beaty, J. A. (2017). Caregiving setting and Baby Boomer caregiver stress
processes: Findings from the National Study of Caregiving (NSOC). Geriatric Nursing,
38(1), 57–62. https://doi.org/10.1016/J.GERINURSE.2016.07.006
Mulud, Z. A., & McCarthy, G. (2017). Caregiver Burden Among Caregivers of Individuals With
Severe Mental Illness: Testing the Moderation and Mediation Models of Resilience.
Archives of Psychiatric Nursing, 31(1), 24–30.
https://doi.org/10.1016/j.apnu.2016.07.019
Naito, R., Leong, D. P., Bangdiwala, S. I., McKee, M., Subramanian, S. V., Rangarajan, S.,
Islam, S., Avezum, A., Yeates, K. E., Lear, S. A., Gupta, R., Yusufali, A., Dans, A. L.,
Szuba, A., Alhabib, K. F., Kaur, M., Rahman, O., Seron, P., Diaz, R., … Yusuf, S.
146
(2021). Impact of social isolation on mortality and morbidity in 20 high-income, middle-
income and low-income countries in five continents. BMJ Global Health, 6(3), e004124.
https://doi.org/10.1136/BMJGH-2020-004124
National Alliance for Caregiving, & AARP. (2020). Caregiving in the U.S. 2020.
Neri, A. L., Yassuda, M. S., Fortes-Burgos, A. C. G., Mantovani, E. P., Arbex, F. S., De Souza
Torres, S. V., Perracini, M. R., & Guariento, M. E. (2012a). Relationships between
gender, age, family conditions, physical and mental health, and social isolation of elderly
caregivers. International Psychogeriatrics, 24(3), 472–483.
https://doi.org/10.1017/S1041610211001700
Neri, A. L., Yassuda, M. S., Fortes-Burgos, A. C. G., Mantovani, E. P., Arbex, F. S., De Souza
Torres, S. V., Perracini, M. R., & Guariento, M. E. (2012b). Relationships between
gender, age, family conditions, physical and mental health, and social isolation of elderly
caregivers. International Psychogeriatrics, 24(3), 472–483.
https://doi.org/10.1017/S1041610211001700
Newman, K., Wang, A. H., Wang, A. Z. Y., & Hanna, D. (2019). The role of internet-based
digital tools in reducing social isolation and addressing support needs among informal
caregivers: A scoping review. BMC Public Health, 19(1), 1–12.
https://doi.org/10.1186/S12889-019-7837-3/TABLES/1
Ogletree, A. M., Brennan-Ing, M., Blieszner, R., Karpiak, S. E., Sands, L. P., & Castle, N. G.
(2019). Health Burden, Support Adequacy, and Depressive Symptoms in Older Men
With HIV. The Gerontologist, 59(6), 1131–1140.
https://doi.org/10.1093/GERONT/GNY169
147
Oldenkamp, M., Hagedoorn, M., Slaets, J., Stolk, R., Wittek, R., & Smidt, N. (2016). Subjective
burden among spousal and adult-child informal caregivers of older adults: Results from a
longitudinal cohort study. BMC Geriatrics, 16(1), 1–11. https://doi.org/10.1186/S12877-
016-0387-Y/TABLES/5
O’Malley, A. S., Forrest, C. B., & O’Malley, P. G. (2000). Low-income women’s priorities for
primary care: A qualitative study. Journal of Family Practice, 49(2), 141–146.
Ong, G., Pibarot, P., Redfors, B., Weissman, N. J., Jaber, W. A., Makkar, R. R., Lerakis, S.,
Gopal, D., Khalique, O., Kodali, S. K., Thourani, V. H., Anwaruddin, S., McAndrew, T.,
Zhang, Y., Alu, M. C., Douglas, P. S., & Hahn, R. T. (2020). Diastolic Function and
Clinical Outcomes After Transcatheter Aortic Valve Replacement: PARTNER 2
SAPIEN 3 Registry. Journal of the American College of Cardiology, 76(25), 2940–2951.
https://doi.org/10.1016/j.jacc.2020.10.032
Ornstein, K. A., Wolff, J. L., Bollens-Lund, E., Rahman, O.-K., & Kelley, A. S. (2019). Spousal
Caregivers Are Caregiving Alone In The Last Years Of Life. Aging & Health, 38(6),
964–972. https://doi.org/10.1377/hlthaff.2019.00087
Ott, C. H., Sanders, S., & Kelber, S. T. (2007). Grief and Personal Growth Experience of
Spouses and Adult-Child Caregivers of Individuals With Alzheimer’s Disease and
Related Dementias. The Gerontologist, 47(6), 798–809.
https://doi.org/10.1093/GERONT/47.6.798
Park, M., Sung, M., Kim, S. K., Kim, S., & Lee, D. Y. (2015). Multidimensional determinants of
family caregiver burden in Alzheimer’s disease. International Psychogeriatrics, 27(8).
https://doi.org/10.1017/S1041610215000460
148
Paúl, C., Teixeira, L., Duarte, N., Pires, C. L., & Ribeiro, O. (2019). Effects of a community
intervention program for dementia on mental health: The importance of secondary
caregivers in promoting positive aspects and reducing strain. Community Mental Health
Journal. https://doi.org/10.1007/s10597-018-0345-6
Pearlin, L. I., Mullan, J. T., Semple, S. J., & Skaff, M. M. (1990). Caregiving and the Stress
Process: An Overview of Concepts and Their Measures. The Gerontologist, 30(5), 583–
594. https://doi.org/10.1093/geront/30.5.583
Peerenboom, L., Collard, R. M., Naarding, P., & Comijs, H. C. (2015). The association between
depression and emotional and social loneliness in older persons and the influence of
social support, cognitive functioning and personality: A cross-sectional study. Journal of
Affective Disorders, 182, 26–31. https://doi.org/10.1016/J.JAD.2015.04.033
Penning, M. J., & Wu, Z. (2016). Caregiver Stress and Mental Health: Impact of Caregiving
Relationship and Gender. The Gerontologist, 56(6), 1102–1113.
https://doi.org/10.1093/geront/gnv038
Penrod, J. D., Kane, R. A., Kane, R. L., & Finch, M. D. (1995). Who Cares? The Size, Scope,
and Composition of the Caregiver Support System. The Gerontologist, 35(4), 489–497.
https://doi.org/10.1093/geront/35.4.489
Perez, S. G., Nuccio, A. G., & Stripling, A. M. (2021). A Rapid Review of the Detrimental
Impact of Loneliness and Social Isolation in Caregivers of Older Adults. The American
Journal of Geriatric Psychiatry, 29(4), S122–S123.
https://doi.org/10.1016/J.JAGP.2021.01.117
Pinquart, M., & Sörensen, S. (2005). Ethnic differences in stressors, resources, and psychological
outcomes of family caregiving: A meta-analysis. The Gerontologist, 45(1), 90–106.
149
Pinquart, M., & Sorensen, S. (2006). Gender Differences in Caregiver Stressors, Social
Resources, and Health: An Updated Meta-Analysis. The Journals of Gerontology Series
B: Psychological Sciences and Social Sciences, 61(1), P33–P45.
https://doi.org/10.1093/geronb/61.1.P33
Pohl, J. S., Cochrane, B. B., Schepp, K. G., & Woods, N. F. (2017). Measuring Social Isolation
in the National Health and Aging Trends Study. Research in Gerontological Nursing •,
10(6), 277–287. https://doi.org/10.3928/19404921-20171002-01
Pohl, J. S., Cochrane, B. B., Schepp, K. G., & Woods, N. F. (2018). Falls and Social Isolation of
Older Adults in the National Health and Aging Trends Study. Research in Gerontological
Nursing, 11(2), 61–70. https://doi.org/10.3928/19404921-20180216-02
Polenick, C. A., & DePasquale, N. (2018). Predictors of Secondary Role Strains Among Spousal
Caregivers of Older Adults With Functional Disability. The Gerontologist.
https://doi.org/10.1093/geront/gnx204
Polenick, C. A., Leggett, A. N., Maust, D. T., & Kales, H. C. (2018). Medical Care Tasks among
Spousal Dementia Caregivers: Links to Care-Related Sleep Disturbances. The American
Journal of Geriatric Psychiatry, 26(5), 589–597.
https://doi.org/10.1016/J.JAGP.2018.01.206
Powers, D. V, Gallagher-Thompson, D., & Kraemer, H. C. (2002). Coping and Depression in
Alzheimer’s Caregivers: Longitudinal Evidence of Stability. Journal of Gerontology:
Psychological Sciences, 57B(3), P205–P211.
Pöysti, M. M., Laakkonen, M. L., Strandberg, T., Savikko, N., Tilvis, R. S., Eloniemi-Sulkava,
U., & Pitkälä, K. H. (2012). Gender differences in dementia spousal caregiving.
International Journal of Alzheimer’s Disease, 2012. https://doi.org/10.1155/2012/162960
150
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and
comparing indirect effects in multiple mediator models. Behavior Research Methods
2008 40:3, 40(3), 879–891. https://doi.org/10.3758/BRM.40.3.879
Pristavec, T. (2019). The caregiving dyad: Do caregivers’ appraisals of caregiving matter for
care recipients’ health? Archives of Gerontology and Geriatrics, 82, 50–60.
https://doi.org/10.1016/j.archger.2019.01.020
Quach, L. T., & Burr, J. A. (2021). Perceived social isolation, social disconnectedness and falls:
The mediating role of depression. Aging and Mental Health, 25(6), 1029–1034.
https://doi.org/10.1080/13607863.2020.1732294/SUPPL_FILE/CAMH_A_1732294_SM
3689.DOCX
Quirke, E., König, H. H., & Hajek, A. (2019). Association between caring for grandchildren and
feelings of loneliness, social isolation and social network size: A cross-sectional study of
community dwelling adults in Germany. BMJ Open, 9(12), e029605.
https://doi.org/10.1136/BMJOPEN-2019-029605
Rafnsson, S. B., Shankar, A., & Steptoe, A. (2017). Informal caregiving transitions, subjective
well-being and depressed mood: Findings from the English Longitudinal Study of
Ageing. Aging and Mental Health, 21(1), 104–112.
https://doi.org/10.1080/13607863.2015.1088510
Rigby, T., Ashwill, R. T., Johnson, D. K., & Galvin, J. E. (2019). Differences in the Experience
of Caregiving Between Spouse and Adult Child Caregivers in Dementia With Lewy
Bodies. Innovation in Aging, 3(3), 1–15. https://doi.org/10.1093/GERONI/IGZ027
151
Robinson, C., Bottorff, J., Pesut, B., Oliffe, J., & Tomlinson, J. (2014). The Male Face of
Caregiving: A Scoping Review of Men Caring for a Person With Dementia. American
Journal of Men’s Health, 8(5), 409–426. https://doi.org/10.1177/1557988313519671
Robison, J., Fortinsky, R., Kleppinger, A., Shugrue, N., & Porter, M. (2009). A Broader View of
Family Caregiving: Effects of Caregiving and Caregiver Conditions on Depressive
Symptoms, Health, Work, and Social Isolation. The Journals of Gerontology: Series B,
64B(6), 788–798. https://doi.org/10.1093/GERONB/GBP015
Rodrigues, A. M., Ferré-Grau, C., & Ferreira, P. L. (2015). Being an Informal Caregiver of a
Person with a Pressure Ulcer in the Azores Islands. Advances in Skin and Wound Care,
28(10), 452–459. https://doi.org/10.1097/01.ASW.0000471191.11548.DD
Rodríguez-Madrid, M., del Río-Lozano, M., Fernandez-Peña, R., Jiménez-Pernett, J., García-
Mochón, L., Lupiañez-Castillo, A., & García-Calvente, M. (2018). Gender Differences in
Social Support Received by Informal Caregivers: A Personal Network Analysis
Approach. International Journal of Environmental Research and Public Health, 16(1),
91. https://doi.org/10.3390/ijerph16010091
Rosenthal Gelman, C., Sokoloff, T., Graziani, N., Arias, E., & Peralta, A. (2014). Individually-
Tailored Support for Ethnically-Diverse Caregivers: Enhancing Our Understanding of
What Is Needed and What Works. Https://Doi-
Org.Libproxy2.Usc.Edu/10.1080/01634372.2014.881451, 57(6–7), 662–680.
https://doi.org/10.1080/01634372.2014.881451
Rowe, M., & Gross-King, M. (2015). Improving Dementia Caregiver Sleep & the Effect on
Heart Disease Biomarkers. Nursing Faculty Publications.
152
Rozario, P. A., & DeRienzis, D. (2008). Familism Beliefs and Psychological Distress Among
African American Women Caregivers. The Gerontologist, 48(6), 772–780.
Rozario, P. A., & Simpson, G. M. (2018). Social Support and Self-Rated Health of African
American Women Informal Caregivers: Urban and Rural Differences. Journal of
Gerontological Social Work, 61(1), 16–30.
https://doi.org/10.1080/01634372.2017.1391918
Santini, Z. I., Jose, P. E., York Cornwell, E., Koyanagi, A., Nielsen, L., Hinrichsen, C.,
Meilstrup, C., Madsen, K. R., & Koushede, V. (2020). Social disconnectedness,
perceived isolation, and symptoms of depression and anxiety among older Americans
(NSHAP): A longitudinal mediation analysis. The Lancet Public Health, 5(1), e62–e70.
https://doi.org/10.1016/S2468-2667(19)30230-0
Savla, J., Wang, Z., Zhu, J., Brossoie, N., Roberto, K. A., & Blieszner, R. (2019). Mastery and
Longevity in Spousal Caregivers of Persons with Dementia. The Journals of
Gerontology: Series B. https://doi.org/10.1093/geronb/gbz028
Scharf, I., Stoldt, M., Libbrecht, R., Höpfner, A. L., Jongepier, E., Kever, M., & Foitzik, S.
(2021). Social isolation causes downregulation of immune and stress response genes and
behavioural changes in a social insect. Molecular Ecology, 30(10), 2378–2389.
https://doi.org/10.1111/MEC.15902
Scharlach, A. E., Kellam, R., Ong, N., Baskin, A., Goldstein, C., & Fox, P. J. (2006). Cultural
Attitudes and Caregiver Service Use. Journal of Gerontological Social Work, 47(1–2),
133–156. https://doi.org/10.1300/J083v47n01_09
Scharlach, A. E., Sobel, E. L., & Roberts, R. E. (1991). Employment and caregiver strain: An
integrative model. Gerontologist, 31(6), 778–787.
153
Schulz, R. (2020). The Intersection of Family Caregiving and Work: Labor Force Participation,
Productivity, and Caregiver Well-Being. In Current and Emerging Trends in Aging and
Work (pp. 399–413). Springer International Publishing. https://doi.org/10.1007/978-3-
030-24135-3_20
Schulz, R., & Czaja, S. J. (2018). Family Caregiving: A Vision for the Future. The American
Journal of Geriatric Psychiatry, 26(3), 358–363.
https://doi.org/10.1016/J.JAGP.2017.06.023
Schulz, R., & Williamson, G. M. (1991). A 2-year longitudinal study of depression among
Alzheimer’s caregivers. Psychology and Aging, 6(4), 569–578.
https://doi.org/10.1037/0882-7974.6.4.569
Segrin, C., Badger, T. A., Sikorskii, A., Crane, T. E., & Pace, T. W. W. (2018). A dyadic
analysis of stress processes in Latinas with breast cancer and their family caregivers.
Psycho-Oncology, 27(3), 838–846. https://doi.org/10.1002/PON.4580
Seiffge-Krenke, I. (2011). Coping with relationship stressors: A decade review. Journal of
Research on Adolescence, 21(1), 196–210. https://doi.org/10.1111/j.1532-
7795.2010.00723.x
Shiba, K., Kondo, N., & Kondo, K. (2016). Informal and Formal Social Support and Caregiver
Burden: The AGES Caregiver Survey. https://doi.org/10.2188/jea.JE20150263
Shin, J. Y., Lim, J.-W., Shin, D. W., Kim, S. Y., Yang, H.-K., Cho, J., Jeong, A., Jo, D., Yim,
C.-Y., Park, K., & Park, J.-H. (2018). Underestimated caregiver burden by cancer
patients and its association with quality of life, depression and anxiety among caregivers.
European Journal of Cancer Care, 27(2), e12814. https://doi.org/10.1111/ecc.12814
154
Siefert, M. Lou, Williams, A., Dowd, M. F., Chappel-Aiken, L., & McCorkle, R. (2008). The
Caregiving Experience in a Racially Diverse Sample of Cancer Family Caregivers.
Cancer Nursing, 31(5), 399. https://doi.org/10.1097/01.NCC.0000305760.04357.96
Sierra, F., Hadley, E., Suzman, R., & Hodes, R. (2009). Prospects for Life Span Extension.
Https://Doi-Org.Libproxy2.Usc.Edu/10.1146/Annurev.Med.60.061607.220533, 60, 457–
469. https://doi.org/10.1146/ANNUREV.MED.60.061607.220533
Sim, A., Cordier, R., Vaz, S., Parsons, R., & Falkmer, T. (2017). Relationship Satisfaction and
Dyadic Coping in Couples with a Child with Autism Spectrum Disorder. Journal of
Autism and Developmental Disorders, 47(11), 3562–3573.
https://doi.org/10.1007/s10803-017-3275-1
Sleath, B., Thorpe, J., Landerman, L. R., Doyle, M., & Clipp, E. (2005). African-American and
White Caregivers of Older Adults with Dementia: Differences in Depressive
Symptomatology and Psychotropic Drug Use. Journal of the American Geriatrics
Society, 53(3), 397–404. https://doi.org/10.1111/J.1532-5415.2005.53155.X
Song, H. I., Hasche, L., & Lee, M. J. (2011). Service use barriers differentiating care-givers’
service use patterns. Ageing and Society, 31(8), 1307–1329.
https://doi.org/10.1017/S0144686X10001418
Sörensen, S., Pinquart, M., So¨rensen, S., & So¨rensen, S. (2010). Racial and ethnic differences
in the relationship of caregiving stressors, resources, and sociodemographic variables to
caregiver depression and perceived physical health.
Http://Dx.Doi.Org.Libproxy2.Usc.Edu/10.1080/13607860500142796, 9(5), 482–495.
https://doi.org/10.1080/13607860500142796
155
Spillman, B. C., Freedman, V. A., Kasper, J. D., & Wolff, J. L. (2020). Change Over Time in
Caregiving Networks for Older Adults With and Without Dementia. The Journals of
Gerontology: Series B, 75(7), 1563–1572. https://doi.org/10.1093/geronb/gbz065
Spitzer, R. L., Kroenke, K., Williams, J. B. W., & Löwe, B. (2006). A brief measure for
assessing generalized anxiety disorder: The GAD-7. Archives of Internal Medicine,
166(10), 1092–1097.
Spruytte, N., Van Audenhove, C., & Lammertyn, F. (2001). Predictors of institutionalization of
cognitiely-impared elderly cared for by their relatives. International Journal of Geriatric
Psychiatry, 21, 1119–1128.
Staight, P. R., & Marie Harvey, S. (1990). Caregiver burden: A comparison between elderly
women as primary and secondary caregivers for their spouses. Journal of Gerontological
Social Work, 15(1–2), 89–104. https://doi.org/10.1300/J083v15n01_10
Stefanini, M. C., Troiani, M. R., Caselli, M., Dirindelli, P., Lucarelli, S., Caini, S., & Martinetti,
M. G. (2018). Living with someone with an eating disorder: Factors affecting the
caregivers’ burden. Eating and Weight Disorders - Studies on Anorexia, Bulimia and
Obesity 2018 24:6, 24(6), 1209–1214. https://doi.org/10.1007/S40519-018-0480-7
Steffens, D. C., Fisher, G. G., K.M., Langa., & al, et. (2009). Prevalence of depression among
older Americans: The Aging, Demographics and Memory Study. Int Psychogeriatr, 21,
879–888.
Steven, C. A., Sabik, N. J., Cook, S. K., Azzoli, A. B., & Mendez-Luck, C. A. (2019).
Differences within Differences: Gender Inequalities in Caregiving Intensity Vary by Race
and Ethnicity in Informal Caregivers. Journal of Cross-Cultural Gerontology, 34(3),
245–263. https://doi.org/10.1007/s10823-019-09381-9
156
Sun, A. Y., Finkelstein, E., & Ouchida, K. (2019). Social Isolation, Loneliness, and Caregiver
Burden among Paid and Unpaid Caregivers of Homebound Older Adults. Innovation in
Aging, 3(S1), 967.
Sussman, S., Valente, T. W., Rohrbach, L. A., Skara, S., & Ann Pentz, M. (2006). Translation in
the Health Professions:Converting Science into Action. Evaluation & the Health
Professions, 29(1), 7–32. https://doi.org/10.1177/0163278705284441
Sutcliffe, C. L., Giebel, C. M., Jolley, D., & Challis, D. J. (2016). Experience of burden in carers
of people with dementia on the margins of long-term care. International Journal of
Geriatric Psychiatry, 31(2), 101–108. https://doi.org/10.1002/gps.4295
Takai, M., Takahashi, M., Iwamitsu, Y., Oishi, S., & Miyaoka, H. (2011). Subjective experiences
of family caregivers of patients with dementia as predictive factors of quality of life.
Psychogeriatrics : The Official Journal of the Japanese Psychogeriatric Society, 11(2),
98–104. https://doi.org/10.1111/J.1479-8301.2011.00354.X
Tarlow, B. J., Wisniewski, S. R., Belle, S. H., Rubert, M., Ory, M. G., & Gallagher-Thompson,
D. (2004). Positive Aspects of Caregiving. Research on Aging, 26(4), 429–453.
https://doi.org/10.1177/0164027504264493
Tausig, M. (1992). Caregiver Network Structure, Support and Caregiver Distress. American
Journal of Community Psychology, 20(1).
Tennstedt, S. L., McKinlay, J. B., & Sullivan, L. M. (1989). Informal Care for Frail Elders: The
Role of Secondary Caregivers. The Gerontologist, 29(5), 677–683.
https://doi.org/10.1093/geront/29.5.677
Thach, N., & Wiener, J. (2018). An Overview of Long-term Services and Supports and Medicaid:
Final Report.
157
Thandi, G., Oram, S., Verey, A., Greenberg, N., & Fear, N. T. (2017). Informal caregiving and
intimate relationships: The experiences of spouses of UK military personnel. Journal of
the Royal Army Medical Corps, 163(4), 266–272. https://doi.org/10.1136/jramc-2016-
000679
Todd, C. S., Jones, H. E., Langwenya, N., Hoover, D. R., Chen, P.-L., Petro, G., & Myer, L.
(2020). Safety and continued use of the levonorgestrel intrauterine system as compared
with the copper intrauterine device among women living with HIV in South Africa: A
randomized controlled trial. PLoS Medicine, 17(5), e1003110.
https://doi.org/10.1371/journal.pmed.1003110
Tolkacheva, N., Van Groenou, M. B., De Boer, A., & Van Tilburg, T. (2011). The impact of
informal care-giving networks on adult children’s care-giver burden. Ageing and Society,
31(1), 34–51. https://doi.org/10.1017/S0144686X10000711
Trivedi, R., Beaver, K., Bouldin, E. D., Eugenio, E., Zeliadt, S. B., Nelson, K., Rosland, A. M.,
Szarka, J. G., & Piette, J. D. (2014). Characteristics and well-being of informal
caregivers: Results from a nationally-representative US survey. Chronic Illness, 10(3),
167–179. https://doi.org/10.1177/1742395313506947
Uccheddu, D., Gauthier, A. H., Steverink, N., & Emery, T. (2019). The pains and reliefs of the
transitions into and out of spousal caregiving. A cross-national comparison of the health
consequences of caregiving by gender. Social Science and Medicine, 240, 112517.
https://doi.org/10.1016/j.socscimed.2019.112517
United Nations. (2019). World Population Ageing 2019: Highlights.
Valente, M. J., Georgeson, A. R., & Gonzalez, O. (2021). Clarifying the Implicit Assumptions of
Two-Wave Mediation Models via the Latent Change Score Specification: An Evaluation
158
of Model Fit Indices. Frontiers in Psychology, 12, 3873.
https://doi.org/10.3389/FPSYG.2021.709198/BIBTEX
van der Lee, J., Bakker, T. J. E. M., Duivenvoorden, H. J., & Dröes, R.-M. (2014). Multivariate
models of subjective caregiver burden in dementia: A systematic review. Ageing
Research Reviews, 15, 76–93.
Vasileiou, K., Barnett, J., Barreto, M., Vines, J., Atkinson, M., Lawson, S., & Wilson, M.
(2017a). Experiences of loneliness associated with being an informal caregiver: A
qualitative investigation. Frontiers in Psychology, 8(APR), 585.
https://doi.org/10.3389/FPSYG.2017.00585/BIBTEX
Vasileiou, K., Barnett, J., Barreto, M., Vines, J., Atkinson, M., Lawson, S., & Wilson, M.
(2017b). Experiences of loneliness associated with being an informal caregiver: A
qualitative investigation. Frontiers in Psychology, 8(APR), 585.
https://doi.org/10.3389/FPSYG.2017.00585/BIBTEX
Versey, H. S. (2017). Caregiving and Women’s Health: Toward an Intersectional Approach. In
Women’s Health Issues (Vol. 27, Issue 2, pp. 117–120). Elsevier USA.
https://doi.org/10.1016/j.whi.2017.01.009
von Känel, R., Mausbach, B. T., Dimsdale, J. E., Ziegler, M. G., Mills, P. J., Allison, M. A.,
Patterson, T. L., Ancoli-Israel, S., & Grant, I. (2019). Refining caregiver vulnerability for
clinical practice: Determinants of self-rated health in spousal dementia caregivers. BMC
Geriatrics, 19(1), 18. https://doi.org/10.1186/s12877-019-1033-2
Wallace, W. S. P., Dilworth-Anderson, P. P., Williams, S. W., & Dilworth-Anderson, P. P.
(2002). Systems of social support in families who care for dependent African American
elders. The Gerontologist, 42(2), 224–236.
159
Wang, L.-J., Zhong, W.-X., Ji, X.-D., & Chen, J. (2016). Depression, caregiver burden and social
support among caregivers of retinoblastoma patients in China. International Journal of
Nursing Practice, 22(5), 478–485. https://doi.org/10.1111/ijn.12458
Ward-Griffin, C., Oudshoorn, A., Clark, K., & Bol, N. (2007). Mother-adult daughter
relationships within Dementia care: A critical analysis. Journal of Family Nursing, 13(1),
13–32. https://doi.org/10.1177/1074840706297424
Watson-Singleton, N. N. (2017). Strong Black Woman Schema and Psychological Distress: The
Mediating Role of Perceived Emotional Support. Journal of Black Psychology, 43(8),
778–788. https://doi.org/10.1177/0095798417732414
Weiai, X. W., Stefanone, M. A., & Rui, J. R. (2013). The benefits and burdens of network
diversity: Political engagement on social networking sites. First Monday, 18(9).
https://doi.org/10.5210/fm.v18i9.4822
Willert, B., & Minnotte, K. L. (2019). Informal Caregiving and Strains: Exploring the Impacts of
Gender, Race, and Income. Applied Research in Quality of Life .
https://doi.org/10.1007/s11482-019-09786-1
Williamson, G., Shaffer, D. R., & Schulz, R. (1998). Activity restriction and prior relationship
history as contributors to mental health outcomes among middle-aged and older spousal
caregivers. Health Psychology, 17, 152–162.
Wister, A., Cosco, T., Mitchell, B., Menec, V., & Fyffe, I. (2019). Development and Concurrent
Validity of a Composite Social Isolation Index for Older Adults Using the CLSA.
Canadian Journal on Aging / La Revue Canadienne Du Vieillissement, 38(2), 180–192.
https://doi.org/10.1017/S0714980818000612
160
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. https://doi.org/10.1001/jamainternmed.2015.7664
Wright, M. R., Hammersmith, A. M., Brown, S. L., & Lin, I.-F. (2019). The Roles of Marital
Dissolution and Subsequent Repartnering on Loneliness in Later Life. The Journals of
Gerontology: Series B. https://doi.org/10.1093/geronb/gbz121
Wysocki, T., Nansel, T. R., Holmbeck, G. N., Chen, R., Laffel, L., Anderson, B. J., Weissberg-
Benchell, J., & Study, for the S. C. of the F. M. of C. D. (2009). Collaborative
Involvement of Primary and Secondary Caregivers: Associations with Youths’ Diabetes
Outcomes. Journal of Pediatric Psychology, 34(8), 869–881.
https://doi.org/10.1093/JPEPSY/JSN136
Xiang, X., Lai, P. H. L., Bao, L., Sun, Y., Chen, J., Dunkle, R. E., & Maust, D. (2020). Dual
Trajectories of Social Isolation and Dementia in Older Adults: A Population-Based
Longitudinal Study. Journal of Aging and Health, 33(1–2), 63–74.
https://doi.org/10.1177/0898264320953693
Xu, J., Liu, P.-J., & Beach, S. (2021). Multiple Caregivers, Many Minds: Family Discord and
Caregiver Outcomes. The Gerontologist, 61(5), 661–669.
https://doi.org/10.1093/geront/gnaa067
Yu, D. S. F., Cheng, S.-T., & Wang, J. (2018). Unravelling positive aspects of caregiving in
dementia: An integrative review of research literature. International Journal of Nursing
Studies, 79, 1–26. https://doi.org/10.1016/J.IJNURSTU.2017.10.008
Yuen, E. Y. N. ;, Wilson, C. J., Yuen, E. Y. N., & Wilson, C. J. (2021). The Relationship
between Cancer Caregiver Burden and Psychological Outcomes: The Moderating Role of
161
Social Connectedness. Current Oncology 2022, Vol. 29, Pages 14-26, 29(1), 14–26.
https://doi.org/10.3390/CURRONCOL29010002
Zegwaard, M. I., Aartsen, M. J., Grypdonck, M. H. F., & Cuijpers, P. (2013). Differences in
impact of long term caregiving for mentally ill older adults on the daily life of informal
caregivers: A qualitative study. BMC Psychiatry, 13(1), 1–9.
https://doi.org/10.1186/1471-244X-13-103/TABLES/1
Zwar, L., König, H. H., & Hajek, A. (2018). The impact of different types of informal caregiving
on cognitive functioning of older caregivers: Evidence from a longitudinal, population-
based study in Germany. Social Science and Medicine, 214, 12–19.
https://doi.org/10.1016/j.socscimed.2018.07.048
Zwar, L., König, H. H., & Hajek, A. (2020). Psychosocial consequences of transitioning into
informal caregiving in male and female caregivers: Findings from a population-based
panel study. Social Science and Medicine, 264, 113281.
https://doi.org/10.1016/j.socscimed.2020.113281
Abstract (if available)
Abstract
Primary caregivers of older adults are more likely to experience social isolation and mental health problems as a result of their responsibilities. To reduce caregiver burden and protect their health and well-being, it is critical to investigate primary caregivers' experiences of caregiving and identify supportive factors. Furthermore, research should take into account sociocultural diversity and investigate how the caregiver stress process differs depending on primary caregiver sociodemographic characteristics. The purposes of this dissertation are to examine the mediated effect of social isolation between caregiving stress and mental health symptoms (i.e., depression and anxiety) of primary caregivers, and the moderation effect of support from secondary caregiver network (SCN) on the stress process, by applying the Stress Process Model of Caregiving (SPM). A second aim is to test model variances by gender, race, and relationship type of primary caregivers, as well as their potential intersections. Secondary data analyses were conducted on 782 primary caregivers and 1003 secondary caregivers using data from the 2015 and 2017 National Study of Caregiving (NOSC) and the linked 2015 National Health and Aging Trends Study (NHATS). Structural equational models were built to examine the study aims, and the composite measurements (e.g., caregiving stress, social isolation, and SCN support) were tested for their internal consistency and construct validity. The findings suggest that taking on caregiving responsibilities can cause social isolation, which can lead to depression in primary caregivers. High level of SCN support, however, may protect primary caregivers’ social connections and mental health from the impacts of caregiving stress. Intersectional analyses also found that high level of SCN support may be more effective in protecting Black women caregivers’ mental health than in other gender-race groups. In conclusion, the results of the dissertation study enable a clearer understanding of the stress process and how SCN support hold promise for protecting social networks and mental health of primary caregivers with different sociodemographic backgrounds.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Impacts of caregiving on wellbeing among older adults and their spousal caregivers in the United States
PDF
A targeted culturally-informed approach for caregiver stress among Vietnamese caregivers of family members [executive summary]
PDF
A targeted culturally-informed approach for caregiver stress among Vietnamese caregivers of family members [oral defense]
PDF
A targeted culturally-informed approach for caregiver stress among Vietnamese caregivers of family members [capstone paper]
PDF
The art connection: disrupting social isolation and loneliness among transition age youth in eastern Kentucky
PDF
An exploratory study of the role of religion and religious involvement among Latinos with schizophrenia and family caregivers
PDF
Collective impact addressing social isolation for older adults living independently in the community; a capstone innovation project
PDF
Examining the longitudinal influence of the physical and social environments on social isolation and cognitive health: contextualizing the role of technology
PDF
COACH: Connecting Older Adults to Community for Health - an evidence-based program to address older adult social isolation and loneliness
PDF
Making connections saves lives: social prescribing linking older adults to their communities
PDF
Project social: a systems approach to reduce social isolation in aging adults
PDF
Full spectrum transition assistance: preventing loneliness and social isolation in military members during moves
PDF
La cocina de abuela: eradicating social isolation with Latino older adults
PDF
My Life, My Wishes, a death education program for community-dwelling Chinese older adults: development and pilot testing
PDF
Belonging as Black women: Black grandmothers pilot program
PDF
Tobacco use change among formerly homeless supportive housing residents: socioecological barriers and facilitators to cessation
PDF
Close the health gap: improving patient access to psychiatric treatment through primary care and telepsychiatry integration
PDF
Catalysts redefining resilient communities and leadership excellence: an executive leadership program for social service managers
PDF
Blue Star Families Connected Communities Pilot
PDF
Acculturation team-based clinical program: pilot program to address acculturative stress and mental health in the Latino community
Asset Metadata
Creator
Liang, Jiaming
(author)
Core Title
The roles of social isolation and secondary caregiver support in the stress process among primary caregivers of older adults
School
Suzanne Dworak-Peck School of Social Work
Degree
Doctor of Philosophy
Degree Program
Social Work
Degree Conferral Date
2022-08
Publication Date
06/23/2022
Defense Date
05/24/2022
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
aging,family caregiving,OAI-PMH Harvest,primary caregiver,secondary caregiver network,social isolation
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Aranda, Maria P. (
committee chair
), Jang, Yuri (
committee member
), Wilber, Kathleen H. (
committee member
)
Creator Email
jimmy.liangljm@gmail.com,jliang@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-oUC111345523
Unique identifier
UC111345523
Legacy Identifier
etd-LiangJiami-10785
Document Type
Dissertation
Rights
Liang, Jiaming
Type
texts
Source
20220623-usctheses-batch949
(batch),
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 author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.
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
Repository Email
cisadmin@lib.usc.edu
Tags
family caregiving
primary caregiver
secondary caregiver network
social isolation