University of Southern California Dissertations and Theses

The effects of familism and cultural justification on the mental and physical health of family caregivers

(USC Thesis Other)

The effects of familism and cultural justification on the mental and physical health of family caregivers

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

THE EFFECTS OF FAMILISM AND CULTURAL JUSTIFICATION ON THE
MENTAL AND PHYSICAL HEALTH OF FAMILY CAREGIVERS

by

Philip S. Sayegh

A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF ARTS
(PSYCHOLOGY)

December 2009

Copyright 2009 Philip S. Sayegh

ii

Acknowledgments
I gratefully acknowledge Bob G. Knight for his careful guidance and valuable
contributions to this paper. In addition, I am grateful to Margaret Gatz, Mara Mather, and
Brian Lickel for their helpful input.

iii

Table of Contents
Acknowledgements ii
List of Tables iv
List of Figures vi
Abstract viii
Chapter 1: Introduction 1
Chapter 2: Methods 15
Chapter 3: Results 29

Chapter 4: Discussion 65
Bibliography 80

iv

List of Tables
Table 1: Mean Demographic, Caregiving, and Key Variable Information 18
of Caregivers (N = 160)

Table 2: Familial Obligations: Items from the Sabogal et al. (1987) 22
Familism Scale

Table 3: Perceived Support from the Family: Items from the Sabogal et al. 22
(1987) Familism Scale

Table 4: Cultural Justification: Items from the CJCS (Dilworth-Anderson, 23
1995) and TCI Subscale of the Lawton et al. (1989) Caregiving
Appraisal Measure

Table 5: Correlation Coefficients among Key Variables 27

Table 6: Goodness of Model Fit Statistics of the First Models 30

Table 7: Goodness of Model Fit Statistics of the Reduced Models 33

Table 8: Changes in Chi-Square Fit Indices from the First Models to the 35
Reduced Models

Table 9: Unstandardized Regression Coefficients Estimated from a Fully 37
Recursive Path Model Predicting Caregivers’ Familial
Obligations, Caregiver Burden, Coping Style, and Mental Health
(N = 160)

Table 10: Unstandardized Regression Coefficients Estimated from a Fully 38
Recursive Path Model Predicting Caregivers’ Familial
Obligations, Caregiver Burden, Coping Style, and Self-Reported
General Physical Health (N = 160)

Table 11: Unstandardized Regression Coefficients Estimated from a Fully 39
Recursive Path Model Predicting Caregivers’ Familial
Obligations, Caregiver Burden, Coping Style, and Cardiovascular
Health (N = 160)

v

Table 12: Unstandardized Regression Coefficients Estimated from a Fully 40
Recursive Path Model Predicting Caregivers’ Perceived Support
from the Family, Caregiver Burden, Coping Style, and Mental
Health (N = 160)

Table 13: Unstandardized Regression Coefficients Estimated from a Fully 41
Recursive Path Model Predicting Caregivers’ Perceived Support
from the Family, Caregiver Burden, Coping Style, and Self-
Reported General Physical Health (N = 160)

Table 14: Unstandardized Regression Coefficients Estimated from a Fully 42
Recursive Path Model Predicting Caregivers’ Perceived Support
from the Family, Caregiver Burden, Coping Style, and
Cardiovascular Health (N = 160)

Table 15: Unstandardized Regression Coefficients Estimated from a Fully 43
Recursive Path Model Predicting Caregivers’ Cultural
Justification, Caregiver Burden, Coping Style, and Mental Health
(N = 160)

Table 16: Unstandardized Regression Coefficients Estimated from a Fully 44
Recursive Path Model Predicting Caregivers’ Cultural
Justification, Caregiver Burden, Coping Style, and Self-Reported
General Physical Health (N = 160)

Table 17: Unstandardized Regression Coefficients Estimated from a Fully 45
Recursive Path Model Predicting Caregivers’ Cultural
Justification, Caregiver Burden, Coping Style, and Cardiovascular
Health (N = 160)

vi

List of Figures

Figure 1: Fitted Path Model of Familial Obligations, Burden, Coping 47
Style, and Mental Health of Caregivers (Standardized Path
Coefficients)

Figure 2: Fitted Path Model of Familial Obligations, Burden, Coping 48
Style, Education, and Mental Health of Caregivers
(Standardized Path Coefficients)

Figure 3: Fitted Path Model of Familial Obligations, Burden, Coping 49
Style, and Self-Reported General Physical Health of
Caregivers (Standardized Path Coefficients)

Figure 4: Fitted Path Model of Familial Obligations, Burden, Coping 50
Style, Education, and Self-Reported General Physical Health
of Caregivers (Standardized Path Coefficients)

Figure 5: Fitted Path Model of Familial Obligations, Burden, Coping 51
Style, and Cardiovascular Health of Caregivers (Standardized
Path Coefficients)

Figure 6: Fitted Path Model of Familial Obligations, Burden, Coping 52
Style, Education, and Cardiovascular Health of Caregivers
(Standardized Path Coefficients)

Figure 7: Fitted Path Model of Perceived Support from the Family, 53
Burden, Coping Style, and Mental Health of Caregivers
(Standardized Path Coefficients)

Figure 8: Fitted Path Model of Perceived Support from the Family, 54
Burden, Coping Style, Education, and Mental Health of
Caregivers (Standardized Path Coefficients)

Figure 9: Fitted Path Model of Perceived Support from the Family, 55
Burden, Coping Style, and Self-Reported General Physical
Health of Caregivers (Standardized Path Coefficients)

Figure 10: Fitted Path Model of Perceived Support from the Family, 56
Burden, Coping Style, Education, and Self-Reported General
Physical Health of Caregivers (Standardized Path Coefficients)

vii

Figure 11: Fitted Path Model of Perceived Support from the Family, 57
Burden, Coping Style, and Cardiovascular Health of Caregivers
(Standardized Path Coefficients)

Figure 12: Fitted Path Model of Perceived Support from the Family, 58
Burden, Coping Style, Education, and Cardiovascular Health of
Caregivers (Standardized Path Coefficients)

Figure 13: Fitted Path Model of Cultural Justification, Burden, Coping 59
Style, and Mental Health of Caregivers (Standardized Path
Coefficients)

Figure 14: Fitted Path Model of Cultural Justification, Burden, Coping 60
Style, Education, and Mental Health of Caregivers
(Standardized Path Coefficients)

Figure 15: Fitted Path Model of Cultural Justification, Burden, Coping 61
Style, and Self-Reported General Physical Health of Caregivers
(Standardized Path Coefficients)

Figure 16: Fitted Path Model of Cultural Justification, Burden, Coping 62
Style, Education, and Self-Reported General Physical Health
of Caregivers (Standardized Path Coefficients)

Figure 17: Fitted Path Model of Cultural Justification, Burden, Coping 63
Style, and Cardiovascular Health of Caregivers (Standardized
Path Coefficients)

Figure 18: Fitted Path Model of Cultural Justification, Burden, Coping 64
Style, Education, and Cardiovascular Health of Caregivers
(Standardized Path Coefficients)

viii

Abstract

This study examined how burden and coping styles mediate the relation between
three specific cultural factors and the mental and physical health of African American
and White family caregivers of dementia patients. Two factors of familism (Familial
Obligations and Perceived Support from the Family) and the Cultural Justification factor
were examined. Consistent with the hypothesis, Familial Obligations had an indirect
effect on poor mental and self-reported physical health outcomes via avoidant coping.
Contrary to the hypothesis, Perceived Support from the Family had no effect on
caregivers’ mental and physical health outcomes. Cultural Justification was associated
with active coping yet did not have the hypothesized protective effect on health
outcomes. This factor was also associated with avoidant coping, which resulted in poor
mental and self-reported physical health outcomes. These findings suggest that feelings
of obligation may account for a significant proportion of the negative effects of familism
on caregivers’ health.

1

Chapter 1: Introduction
Providing care for a family member with dementia has been shown to be
associated with many detrimental mental and physical health consequences. Numerous
studies have documented higher rates of feelings of burden, symptoms of depression and
anxiety, clinical depression, anxiety disorders, and psychotropic medication use among
caregivers compared to noncaregivers (e.g., Cannuscio et al., 2002; Coon, Ory, & Schulz,
2003; Ory, Hoffman, Yee, Tennstedt, & Schulz, 1999; Schulz, O’Brien, Bookwalla, &
Fleissner, 1995; Zarit, Johansson, & Jarrot, 1998). In addition to psychiatric morbidity,
dementia caregivers have an increased risk for various health problems, including
clinically significant reduced immunity, infectious illness, higher levels of stress
hormones, and increased risk for cardiovascular (CV) diseases, hyperlipidemia, and
hyperglycemia (Davies, 1996; Kiecolt-Glaser, Dura, Speicher, Trask, & Glaser, 1991;
Lee, Colditz, Berkman, & Kawachi, 2003; Mausbach et. al, 2007; Vitaliano, Zhang, &
Scanlan, 2003; Vitaliano, Russo, & Niaura, 1995). Moreover, family caregivers tend to
perceive and rate their health as worse compared to noncaregivers even when these
groups do not differ in the more objective measures of the number of diseases or
prescription medications they report (Abramson, Ritter, Gofin, & Kark, 1992;
Baumgarten et al., 1992; Grafstrom, Fratiglioni, Sandman, & Winblad, 1992; Soskolne,
Halevy-Levin, & Ben-Yehuda, 2007).
An increased focus on cultural factors that affect caregiver burden and outcomes
is crucial, as the number of Americans with Alzheimer’s disease (AD) is expected to
surge. This predicted rapid increased in the prevalence of AD results from an increased

2

number of Americans living into their 80s and 90s due to advances in medicine, medical
technology, and environmental and social conditions (Alzheimer’s Association, 2008).
The number of people aged 65 and over with AD is estimated to jump to 7.7 million in
2030, representing a greater than 50 percent increase over the number affected in 2007
(Hebert, Beckett, Scherr, & Evans, 2001). By 2050, this number is predicted to soar to
13.2 million Americans (Hebert, Scherr, Bienias, Bennett, & Evans, 2003). Moreover, the
U.S. Bureau of the Census (2000) predicts that between 2000 and 2050, the population of
older White adults will double while the population of older African American adults will
quadruple. In addition, the prevalence of debilitating functional limitations among racial
and ethnic minority older adults is higher compared to White older adults (U.S.
Department of Health & Human Services, 1991). As a result, it is likely that there will be
a significant increase in the number of minority family caregivers of older adults, calling
to attention the need for a richer understanding of the effects of cultural values on
caregiving both for methodological and clinical purposes.
Familism and Caregiving
One cultural factor that has been examined across diverse ethnic and cultural
groups in the context of caregiving is familism. Familism refers to the strong normative
feelings of loyalty, dedication, reciprocity, and attachment of family members to their
family and familial relationships, both nuclear and extended, as well as strong
identification and solidarity with their family members (Heller, 1970; Sabogal, Marin,
Otero-Sabogal, Marin, & Perez-Stable, 1987). The first scale created to measure familism
was developed by Bardis (1959) and was validated on both a sample of Greek students

3

from a familistic community and students from Michigan. Since its inception, this scale
has been used in a number of different cultures (e.g., Aldrich, Lipman, & Goldman, 1973;
Bardis, 1959; Blair, 1972; Jean-Van Hell, 2001; Kassees, 1972; Knight et al., 2002; Rao
& Rao, 1979; Youn, Knight, Jeong, & Benton, 1999). Recently, Schwartz (2007)
conducted several analyses to assess the applicability of familism to other ethnic groups
aside from Hispanics, with whom familism is most strongly associated and often
researched, and demonstrated few significant differences between his samples of
Hispanic, non-Hispanic White, and non-Hispanic African American young adults. His
results suggest that the cultural value of familism may be appropriately applied to diverse
ethnic groups and that it may operate similarly across ethnic groups.
Several studies have examined the effects of familism on burden and the mental
and physical health outcomes of family caregivers of dementia patients. It may seem
intuitive to predict that higher levels of familism would result in better outcomes for
family caregivers of people with dementia, as they may have more of an underlying,
deep-seated desire to care for their loved ones. However, the findings of such studies
have been mixed, suggesting that familism may be a somewhat complex value whose
impact on the caregiving process may be influenced by numerous other factors. For
example, Coon et al. (2004) compared the appraisals of stress associated with caregiving
in both Latina and White family caregivers of dementia patients and found that Latina
caregivers described the behavioral problems of their care recipients as less bothersome
and ascribed more positive aspects to caregiving than did Caucasian caregivers. The
authors suggested that familism may have served as a protective factor against deleterious

4

caregiving outcomes for the Latina caregivers in their study. However, a study by Losada
et al. (2006) comparing the association of familism with caregiver burden and depression
among Spanish and Hispanic-American caregivers produced conflicting results. While
familism was significantly correlated with lower levels of burden among Hispanic-
American caregivers, consistent with Coon et al.’s suggestion, it was also associated with
higher levels of depressive symptomatology among Spanish caregivers. In addition, the
relationship between familism and burden in their Spanish sample was nonsignificant yet
positive.
The discrepancy between the findings of these studies suggests that familism may
in fact be affected by other variables that need to be more closely examined. One study
that specifically addressed how other contextual factors influenced the effects of familism
on caregiver outcomes was performed by Kim, Knight, and Flynn Longmire (2007). The
authors used path modeling to examine how familism, burden, and coping styles mediate
the association of ethnicity with the mental and physical health of African American and
White caregivers of family members with dementia. With regard to familism specifically,
they hypothesized that African Americans would have higher levels than Whites, which
would be indirectly associated with lower burden levels. They based their hypothesis in
part on the findings from a study by McCallum, Flynn Longmire, and Knight (2007) that
showed that African American caregivers more firmly viewed caregiving as a familial
duty and reported higher levels of familism than did White caregivers, which may
provide at least a partial explanation for the lower levels of burden among their African
American sample. Contrary to their hypothesis, Kim and colleagues found that familism

5

was significantly influenced not by ethnicity but rather by education, with higher levels
of familism being associated with lower levels of education. In order to account for these
unexpected findings, the authors suggested that higher levels of education likely result in
increased amounts of individualism, which was likely more influential than ethnicity.
They argued that differences between African Americans and Whites attributed to
ethnicity could be more appropriately explained by socioeconomic status instead of
culture. This finding emphasizes the need to examine the effects of familism on burden
and mental and physical health outcomes in the context of other variables.
The Sociocultural Stress and Coping Model for Caregivers
Caregiver burden is often conceptualized from a stress and coping model,
originally developed by Lazarus and Folkman (1984). This model posits that the effects
of negative stress on individuals are mediated by variables including appraisals, coping
styles, and social support. In an effort to integrate ethnicity into the Lazarus and Folkman
stress and coping model, Aranda and Knight (1997) developed the Socio-cultural Stress
and Coping Model to better understand and conceptualize caregiving across culturally
diverse groups. This model emphasizes the role of cultural and ethnic dimensions within
the stress and coping process, rather than simply viewing ethnicity as a proxy for
presumed differences in cultural values that may affect the caregiving experience. More
specifically, they argued that ethnicity and culture would affect the stress and coping
process of caregivers due to varying risks for specific health problems and disabilities,
differences in the appraisals of possible stressors, and their impact on stress-mediating
variables, such as coping and social support. This model was subsequently modified by

6

Knight, Silverstein, McCallum, and Fox (2000) through the reinterpretation of the effects
of ethnicity on the stress and coping process as influencing other status variables,
caregiving demands that may differ by group, caregivers’ appraisals based on cultural
values pertaining to the appraisal of caregiving as stressful, coping styles that are
influenced by cultural differences, and the mental health outcomes of caregivers resulting
from the stress and coping process. They argued that these ethnic and cultural differences
can concomitantly affect any and all stages of the stress and coping model in varying
directions. This conceptualization allows for a more informed understanding of
caregiving across diverse groups.
Different coping styles can affect the stress process associated with caregiving
through both efforts to alter the stressful circumstances and regulate the emotional
distress associated with the situation (e.g., Lazarus, 1991; Lazarus & Folkman, 1984;
Penley, Tomaka & and Weibe, 2002). According to Monat and Lazarus (1991), active
coping reduces the potential unfavorable effects of stressful situations, thereby resulting
in better mental and physical health outcomes. Some studies have demonstrated that
active coping results in fewer symptoms of depression for dementia caregivers (e.g.,
Haley, Levine, Brown, & Bartolucci, 1987; Vitaliano, Russo, Carr, Maiuro, & Becker,
1985). Conversely, avoidant coping styles may result in poorer outcomes because they
consist of thoughts and actions such as denial and disengagement that aim to reduce the
emotional effects of stress. In a meta-analysis of 10 studies and 897 participants, Penley
et al. found that avoidant coping was related to poorer mental health. Several studies have
demonstrated that avoidant coping styles specifically among dementia caregivers are

7

associated with negative outcomes, including depressive symptoms (Powers, Gallagher-
Thompson, & Kraemer, 2002), higher levels of measured burden (Papastavrou,
Kalokerinou, Papacostas, Tsangari, & Sourtzi, 2007), and reduced life satisfaction and
psychological well-being (Garrity, 1997; Wright, Lund, Caserta, & Pratt, 1991).
It has been hypothesized that coping styles are influenced by cultural beliefs
(Aranda & Knight, 1997; Dilworth-Anderson, Williams, & Gibson, 2002; Janevic &
Connell, 2001; Knight et al., 2002). Studies by Knight et al. (2000) and Haley et al.
(1996) examining ethnic and cultural differences in coping styles among caregivers both
found that African American caregivers were more likely to use emotion-focused, or
avoidant, coping styles than White caregivers, who were more likely to engage in
problem-focused, or active, coping styles. However, these studies reported different
results in terms of how ethnicity affected coping styles and indirectly affected emotional
distress. Haley and colleagues found that African American caregivers used a decreased
amount of all coping styles compared to White caregivers, and that lower levels of
avoidant coping and stress appraisal may have resulted in lower levels of depression
among this group. Knight and colleagues found that African American caregivers in their
sample used higher levels of avoidant coping, which led to higher levels of emotional
distress; however, lower burden scores among this group counterbalanced overall distress
scores such that they were similar to the White caregivers in their sample. Despite their
differing results, both of these studies demonstrate how coping styles mediate the process
by which culture and ethnicity impact emotional distress among caregivers. Based on
such findings, it is possible that specific cultural values and norms affect both how family

8

caregivers appraise their caregiving situations and the strategies they use in coping with
the demands of caregiving (Wallace, Witucki, Boland, & Tuck, 1998).
Familism and the Sociocultural Stress and Coping Model for Caregivers
Familism is a specific cultural value that may affect coping styles, yet studies
examining this association have reported conflicting findings. In a study comparing
African American and White female caregivers of family members with dementia,
McCallum et al. (2007) reported that African Americans scored higher on familism than
their White counterparts, yet no significant differences were found between the two
groups in terms of the amount of use of active and avoidant coping styles as well as
mental health outcomes. However, Kim et al. (2007) found that familism actually
resulted in increased avoidant coping styles, which in turn led to worse mental and
physical health outcomes for the caregivers in their sample. The authors argued that the
association of familism with avoidant coping suggests that familism may in fact more
strongly represent feelings of obligation toward the family rather than positive feelings
regarding supporting the family. The unexpected findings from both of these studies
suggest that different factors of familism such as family obligation should be more
closely examined in order to gain more insight into what specific dimensions of familism
may be affecting coping styles and resulting in negative outcomes for caregivers.
One possible explanation for these unanticipated findings may be attributable to
the use of the Bardis (1959) familism scale in both of these studies, which may not be the
most adequate measure of familism. Rogers and Sebald (1962) criticized this scale for its
failure to appropriately operationalize the construct of familism, as it has been applied

9

indiscriminately to describe both nuclear and extended familial relationships. They held
that familism values focused on extended family systems do not necessarily share
common characteristics with the values focused on nuclear family systems. Rao and Rao
(1979) factor analyzed this scale and reported that it consists of two distinct subscales
assessing nuclear and extended familism, which intercorrelate with one another. These
criticisms suggest that the Bardis familism scale may not be best suited as a measure of
this cultural value.
The Sabogal et al. (1987) familism scale differs from the Bardis (1959) scale in
that it taps different and more specific components of familism involving both attitudinal
and behavioral aspects of familism. This scale, which combines three slightly revised
items from the Bardis familism scale and eleven from the Triandis, Marin, Betancourt,
Lisansky, and Chang (1982) familism scale, was created to study the effects of
acculturation on the familism values of Hispanics compared to European Americans.
Sabogal and colleagues conducted an exploratory factor analysis of this scale that
revealed three different factors: Familial Obligations, Perceived Support from the Family,
and Family as Referents. The Familial Obligations factor, which is composed of six
items, was described as the perception of obligation to provide financial and emotional
support to extended family members. The Perceived Support from the Family factor is
made up of three items and assesses the extent to which individuals believe that they
receive adequate material and emotional support from their family. The third factor,
Family as Referents, is composed of five items and measures the view of relatives as
behavioral and attitudinal referents. These findings provide a more specific understanding

10

of familism by providing three distinct factors addressing different dimensions of this
cultural value. These more focused dimensions that the Sabogal et al. familism scale
assesses can be more useful and illuminating in family dementia caregiving research than
the more broad dimensions of familism that the Bardis scale measures.
While Sabogal et al. (1987) succeeded in creating a familism scale that can be
analyzed in terms of these three specific cultural dimensions, they do not provide data
regarding the equivalence of the instrument across the different cultural groups to which
the measures were administered. A confirmatory factor analysis was performed by
McClendon Baumann (2007) to examine whether this scale would demonstrate factorial
invariance across African American and White dementia caregivers and noncaregivers.
The results of this analysis demonstrated that factor loadings did not differ significantly
across groups, indicating that this scale is best represented by a three factor model for
both African American and White individuals. Because factorial invariance was found,
comparisons can be meaningfully made with these different ethnic groups using this
scale. The results also suggest that this scale can be reliably and validly used to measure
specific components of familism with regard to caregiving above and beyond the more
broad definitions of this cultural value.
Cultural Justification for Caregiving
Various cultural values and norms can strongly influence outlooks, choices, and
actions regarding family caregiving. Such norms and values provide explicit rules and
affect the decisions made about both who is to give care for a family member and

11

choosing to provide care. They also have an effect on relations that occur between
caregivers and other individuals in the larger network of obtainable resources (Dilworth-
Anderson et al., 2002). In addition to familism, several other cultural factors have been
proposed to explain why individuals decide to provide care for a family member. For
example, Silverstone (1978) noted the cultural value of filial obligation as one of the
potential motivating factors in providing care for a family member. Reciprocity in the
context of an adult son or daughter providing care for an ailing or disabled parent who
had cared for the child earlier in life is also an important cultural factor in family
caregiving (Wallace et al., 1998). Individuals from various ethnic minority cultures
describe caregiving as involving behavior that is reciprocal in nature (John, Hennessey,
Dyeson, & Garret, 2001; Wallace et al., 1998). In addition, role modeling is often cited as
an important value associated with caregiving that is experienced by different ethnic
groups. Nkongho and Archbold (1995) reported that nearly half of the participants in
their caregiver study acknowledged the importance of the value of providing a model for
younger family members regarding caregiving behaviors. Religious and spiritual values
are also often discussed in the context of providing care for elderly family members. In
an investigation of the role of religion in the caregiving experience, Levkoff, Levy, and
Weitzman (1999) found an ethic to care for family members that was strongly rooted in
individual religious values.
The Cultural Justifications for Caregiving Scale (CJCS; Dilworth-Anderson,
1995) assesses these cultural values with regard to the caregiving experience. The ten-
item CJCS measures cultural socialization values by measuring attitudes about cultural

12

reasons and expectations for providing care. Dilworth-Anderson et al. (2005) evaluated
the factor structure and equivalence of this scale across 48 African American and 121
White dementia caregivers using a two-group confirmatory factor analysis. A one-factor
model of cultural justifications for caregiving was analyzed for goodness of fit, and their
results demonstrated that the CJCS assesses cultural justifications equally well in both
groups. The Cronbach’s alpha for the authors’ sample of African American and White
dementia caregivers was .86.
The Traditional Caregiving Ideology (TCI) subscale of the Lawton Caregiving
Appraisal measure (Lawton, Kleban, Moss, Rovine, & Glicksman, 1989) assesses
traditional attitudes toward the caregiving role assumed to incorporate cultural values
concerning caregiving perceptions. The four items that compose the TCI subscale
combine to form a single factor composed of both family and religious tradition,
modeling for one’s children, and reciprocity as motivation for caregiving. Lawton,
Rajagopal, Brody, and Kleban (1992) evaluated this four-item factor used to measure
caregiving attitudes and reported a Cronbach’s alpha of .71 for their African American
sample and .67 for their White sample.
McClendon Baumann (2007) performed an exploratory factor analysis to identify
possible cultural values underlying different measures of these values that may impact an
individual’s behaviors and attitudes toward providing care for a family member. Both of
these scales were included in the analysis, which revealed a twelve-item factor, Cultural
Justification, composed of all ten items from the CJCS as well as two items from the TCI
subscale. McClendon Baumann (p. 96) defined this factor as “caregiving based on a

13

broad collection of values learned within the family… exclusively comprised of items
describing explicitly stated cultural reasons for caregiving (e.g., caregiving based on
religious beliefs, based on role modeling, based on duty).” The items included in this
factor constitute a more appropriate cultural values scale with more defined cultural
components that can be useful for examining cultural differences in dementia caregiving.
Having strong cultural reasons for providing care would appear to be protective against
poor outcomes for caregivers based on a stress and coping framework that would view
such justifications for caregiving as a caregiving resource in the face of caregiving
demands (Dilworth-Anderson, Goodwin, & Williams, 2004).
Hypotheses
The overall objective of this research investigation is to provide a more detailed
examination of how specific factors of familism and cultural justifications for caregiving,
as mediated by burden and coping, affect the caregiving process and outcomes for family
caregivers of people with dementia. Using the sociocultural stress and coping model as a
foundation, which posits that cultural values operate through both appraisals of
caregiving as burdensome and coping styles to affect mental and physical health
outcomes, I hypothesized the following:
1) Higher scores on the Familial Obligation factor will result in higher levels of both
burden and use of an avoidant coping style, which will result in poorer mental and
physical health outcomes for family caregivers of people with dementia.
2) Higher scores on the Perceived Support from the Family factor will result in
lower levels of both burden and use of an active coping style, which will result in

14

better mental and physical health outcomes for family caregivers of people with
dementia.
3) Higher scores on the Cultural Justification factor will result in lower levels of
both burden and use of an active coping style, which will result in better mental
and physical health outcomes for family caregivers of people with dementia.
No hypotheses were made using the Family as Referents factor of familism in this
study, as its potential role in the context of the stress and coping process for caregivers
was not conceptually clear. In addition, ethnic differences were not hypothesized,
because previous studies using the same sample (e.g., Kim et al, 2007; Knight, Flynn
Longmire, Dave, Kim, and David. 2007) have found ethnic differences in terms of mean
levels of familism yet no direct effects of ethnicity on health outcomes. One exception is
blood pressure, which is directly associated with ethnicity, but this relationship is neither
mediated nor moderated by cultural values such as familism. Effects of ethnicity on self-
reported general physical health have also been found, specifically that African
Americans self-reported worse general physical health, but these effects disappeared
when education was controlled. This finding suggests that this effect may be more due to
socioeconomic (SES) than to ethnic differences with regard to health. This relationship
was also neither mediated nor moderated by familism.

15

Chapter 2: Methods
Participants
This study’s sample included 95 African American and 65 White family
caregivers of individuals with dementia from the Stress and Ethnicity Caregiving Study
(SECS) who were recruited by multistage sample collection. Participants were recruited
through phone contacts in census tracts known to contain a high percentage of residents
aged 65 and older (10% or greater) and known to be largely African American or non-
Hispanic White (60% or greater). Neighborhoods that had average household incomes
below the median for Los Angeles County, California were selected in order to minimize
potential confounds resulting from SES. This technique served to prevent the comparison
of primarily lower SES African Americans to Whites from predominantly higher SES
brackets. Age for the sample was restricted to individuals aged 18 years or older. This
sample of caregivers is the same that was used in the Kim et al. (2007) study, which was
derived from the subsample of caregivers in a study by Knight et al. (2007). Both of these
studies examined mental and physical health outcomes for family caregivers of
individuals with dementia. While the study by Knight and colleagues focused on ethnic
and caregiver status differences in health outcomes, the Kim et al. study examined how
familism impacted these outcomes among caregivers by applying the sociocultural stress
and coping model (Aranda & Knight, 1997; Knight et al., 2000). The current study differs
from both of these studies in that it aimed to examine how more specific cultural values
(the Familial Obligations and Perceived Support from the Family factors of familism and

16

the Cultural Justification factor) affect caregivers’ mental and physical health outcomes
in the context of the sociocultural stress and coping model.
Project telephone interviewers then contacted these randomly selected households
and questioned an adult residing in the household to determine if anyone in the household
was aged 50 years or older and had memory problems or was a caregiver of an older
relative with memory problems who resided in another household. Individuals were
considered primary caregivers if they had a family member with dementia and either
resided with the person with dementia and/or provided at least 8 hours of care per week.
If more than one potential caregiver lived in the household, the one who managed
personal care or supervision was selected. A brochure describing the purpose of the study
was mailed to those households in which a caregiver was identified. After receipt of the
brochure, the project telephone interviewers contacted the potential participants again to
determine whether they would be willing to participate in the study and to obtain the
diagnosis of dementia and its source. Only those who had a professional diagnosis of
dementia were recruited. Approximately one third of the obtained sample was providing
care for an individual diagnosed with AD, while 30.5% were diagnosed with dementia
without further specification. The remainder were nearly equally divided between various
vascular dementias and Parkinson’s related dementias.
In-person interviews and CV testing sessions with the participants were then
arranged by a graduate student research assistant. The interviews and CV testing sessions,
which lasted for about 70-100 minutes, were conducted by ethnically diverse research
assistants who were assigned arbitrarily to interview the participants. The participants

17

received $20 as compensation. The majority (greater than 90%) of the interviews took
place in the participants’ homes, while the rest occurred on the University of Southern
California campus. After being contacted by telephone by a graduate student research
assistant in an attempt to schedule in-person interviews and CV testing sessions,
approximately 69% of the African American and 37% of the White eligible caregivers
were interviewed for the study, whereas the remaining eligible caregivers that were
reached declined to participate. This difference across these two groups is statistically
significant and may reflect the fact that lower SES Whites, who are often not targeted for
recruitment in caregiver studies, are difficult to recruit (Knight et al., 2007).
Table 1 provides demographic and descriptive information for the caregivers in
this study. The ages of the participants who completed the study ranged from 21 to 87
years (M = 56.7 years, SD = 15.9). The majority of caregivers in this study were female
(68.8%) and the mean level of education for the caregivers was “some college.” Eighty
percent of caregivers from both ethnic groups resided in the same household as their care
recipient. The number of years as primary caregivers reported by the participants ranged
from fewer than 1 to 42 (M = 5.1 years, SD = 5.6); however, after dropping out 4 outliers
(greater than or equal to 27 years) from the analysis, the highest number of years as a
primary caregiver was reduced to 14 (M = 4.4 years, SD = 3.1). The mean number of
hours spent on caregiving per week was approximately 18 (SD = 11.6).

18

Table 1: Mean Demographic, Caregiving, and Key Variable Information of Caregivers
(N = 160)
Note. ADL = Activities of Daily Living; IADL = Instrumental Activities of Daily
Living; CES-D = Center for Epidemiological Studies- Depression Scale; GSI = General
Symptom Index.
a
Income was measured using the following scale: 1 = Under $10,000, 2 = $10,000 –
$19,999, 3 = $20,000 – $29,999, 4 = $30,000 – $39,999, 5 = $40,000 - $49,999, 6 =
$50,000 or More.
b
Highest level of education was measured using the following scale: 1 = Elementary
School, 2 = Middle School/Junior High, 3 = High School, 4 = Some College, 5 =
Undergraduate Degree, 6 = Master’s Degree or Beyond.
c
The mean and standard deviation of the length of time in caregiving are presented after
dropping four outliers.
d
ADL and IADL were measured using the following scale: 1 = No help; 2 = Some help;
3 = A lot of help. Total possible scores ranged from 5 – 15 and 8 – 24, respectively.

19

Table 1 (Continued)
M SD %
Demographic characteristic
Age (years) 56.70 15.85
Gender
Male (coded 0) 31.2
Female (coded 1) 68.8
Ethnicity
White (coded 0) 40.6
African American (coded 1) 59.4
Income
a
3.31 1.67
Education
b
4.06 0.88
Caregiving-related characteristic
Living with care recipient
Yes (coded 0) 80.0
No (coded 1) 20.0
Length of time in caregiving (years)
c
4.35 3.08
Caregiving hours per week 17.95 11.61
Care recipients’ problems with ADL
d
9.48 3.32
Care recipients’ problems with IADL
d
20.45 3.37
Key variable
Care recipients’ behavioral problems 8.33 6.76
Familial Obligations 16.97 3.70
Perceived Support from the Family 7.37 2.68
Cultural Justification 36.67 8.03
Burden 30.92 15.70
Active coping 58.50 10.85
Avoidant coping 31.04 7.55
Depression (CES-D) 10.89 6.71
Psychological symptoms (GSI) 0.50 0.53
Self-reported general health 2.16 0.79
Systolic blood pressure 133.91 23.58
Diastolic blood pressure 74.41 12.98
Heart rate 73.76 11.63

20

Measures
Dependent Variables: Caregiver Mental Health
Depression.
Caregivers’ depression was measured in this data set using the Center for
Epidemiological Studies—Depression Scale (CES-D; Radloff, 1977), which is a 20-item
self-report scale that was created to screen for depressive symptoms in the general
population. Responses to each of the items are scored from 0 to 3 and total scores range
from 0 to 60, with higher scores representing a greater amount of depressive symptoms.
Kim et al. (2007) reported Cronbach’s alphas of .80 in African Americans and .77 in
Whites in this sample of caregivers.
Psychological symptoms: Psychological symptoms were measured using the 53-item
Brief Symptom Inventory (BSI; Derogatis & Spencer, 1985). The BSI is a short measure
used to calculate a global measure of psychological symptoms referred to as the General
Symptom Index (GSI). Convergent validity has been demonstrated between the subscales
of the BSI and the Minnesota Multiphasic Personality Inventory (MMPI) scales
(Derogatis & Spencer, 1985). Kim et al. (2007) reported Cronbach’s alphas for the GSI
of .97 for both African Americans and Whites in this sample of caregivers.
Dependent Variables: Caregiver Physical Health
Self-reported general physical health.
Caregivers’ self-reported general physical health was measured by a one-item
measure of current physical health. The item was rated on a 4-point Likert scale (1 =
excellent; 2 = good; 3 = fair; 4 = poor). Several previous studies have reported that self-

21

reported health is highly correlated with physician ratings of health and mortality
(Benyamini, Blumstein, Lusky, & Modan, 2003; George, 2001). Therefore, self-reported
general physical health measures are regarded as valuable proxy measures of physical
health status.
Cardiovascular measures.
Physical health was also measured by CV indicators, as they are important
antecedents of stress-related CV diseases, such as hypertension (Fredrikson & Matthews,
1990) and coronary heart disease (Blascovich & Katkin, 1993). CV indicators are also
psychophysiological variables that may serve as intervening variables between objective
stress and its appraisal and health outcomes. Heart rate, systolic blood pressure, and
diastolic blood pressure readings were taken with a Dinamap Pro 400 (Critikon, Tampa,
FL) automatic blood pressure monitor with the cuff attached to the nondominant arm.
Independent Variables
Familism factors.
The scores on the two factors of familism (Familial Obligations and Perceived
Support from the Family) derived from the 14-item Sabogal et al. (1987) familism scale
were measured by calculating participants’ total scores from the six items that create the
Familial Obligations factor and the three items that compose the Perceived Support from
the Family factor (see Tables 2 and 3, respectively). This scale presents hypothetical
situations involving family crises or events (e.g., “I would help within my means if a
relative told me that he/she is in financial difficulty,” “When one has problems, one can
count on the help of relatives”), which respondents score on a scale of 0 to 4, with higher

22

scores indicating greater levels of the familism factors. The Cronbach’s alpha values for
these two factors are: .69 (African American), .80 (Whites), and .75 (both groups) for
Familial Obligations; and .82 (African American), .88 (Whites), and .85 (both groups) for
Perceived Support from the Family.
Table 2: Familial Obligations: Items from the Sabogal et al. (1987) Familism Scale
1. One should make great sacrifices in order to guarantee a good education for his/her
children.
2. I would help within my means if a relative told me that he/she is in financial
difficulty.
3. One should have the hope of living long enough to see his/her grandchildren grow
up.
4. Aging parents should live with their relatives.
5. A person should always share his/her home with his uncles, aunts, or first cousins if
he/she is in need.
6. A person should always help his/her parents with the support of his/her younger
brothers and sisters if necessary.

Table 3: Perceived Support from the Family: Items from the Sabogal et al. (1987)
Familism Scale
1. When someone has problems, he/she can count on help from his/her relatives.
2. When one has problems, one can count on the help of relatives.
3. One can count on the help from his/her relatives to solve most problems.

Cultural Justification.
The Cultural Justification factor was measured by calculating participants’ total
scores on the 10 items from the CJCS (Dilworth-Anderson, 1995) as well as the two
items from the TCI (Lawton et al., 1989) that combine to create this factor (see Table 4).
The CJCS requires respondents to indicate the degree to which they agree with reasons to
provide care for older family members (e.g., “By giving care to the elderly dependent
family members, I am giving back what has been given to me,” “A strong reason to care
for a family member is to provide a good model for your own children to follow”) on a

23

scale from 1 to 4, with higher scores indicative of greater amounts of cultural
justifications for caregiving. The two items from the TCI pertain to cultural reasons for
caregiving (caregiving based on religious beliefs and role modeling for one’s children).
Respondents indicate the degree to which they agree or disagree with each item on a
scale from 0 to 4, with higher scores representing greater amounts of cultural reasons for
caregiving. The Cronbach’s alpha values are .88 for African Americans, .87 for Whites,
and .88 for both groups.
Table 4: Cultural Justification: Items from the CJCS (Dilworth-Anderson, 1995) and TCI
Subscale of the Lawton et al. (1989) Caregiving Appraisal Measure
CJCS Items:
I give care because:
1. It is my duty to provide care to elderly dependent family members.
2. It is important to set an example for the children in the family.
3. I was taught by my parents to take care of elderly dependent family
members.
4. Of my religious and spiritual beliefs.
5. By giving care to the elderly dependent family members, I am giving back
what has been given to me.
6. It strengthens the bonds between me them.
7. I was raised to believe care should be provided in the family.
8. It is what my people have always done.
9. I feel as though I am being useful and making a family contribution.
10. My family expects me to provide care.
TCI Subscale Items:
1. Taking care of a family member is a way for you to follow religious principles.
2. A strong reason to care for a family member is to provide a good model for your
own children to follow.

Caregiver burden.
Subjective burden was measured with the 22-item Zarit Burden Interview (Zarit,
Reeves, & Bach-Peterson, 1980). Items are scored 0 to 4, with higher scores indicating

24

greater caregiver burden. Kim et al. (2007) reported Cronbach’s alphas of .95 for African
Americans and .92 for Whites in this sample of caregivers.
Coping style.
Sixty items of the COPE scale (Carver, 1997) were used as a measure of coping
styles. Caregivers were asked to respond to each item while focusing specifically on
caregiving. The response choices were 1 (“I haven’t been doing this at all”), 2 (“I have
been doing this a little bit”), 3 (“I have been doing this a medium amount”), and 4 (“I
have been doing this a lot”). McCallum et al. (2007) identified three factors (Active,
Avoidant, and Faith-Oriented Coping) through factor analysis, yet a confirmatory factor
analysis found that these three coping subscales did not fit the SECS data well (χ
2
/df =
3.562, normed fit index (NFI) = 0.654, comparative fit index (CFI) = 0.713, and root-
mean-square error of approximation (RMSEA) = 0.127). Kim et al. (2007) explored
alternative two- and three-factor structures and deleted subscales with low loadings on
the factors, which resulted in a two-factor model including the same Avoidant Coping
factor and a modified Active Coping fitting the data well (χ
2
/df = 2.72, CFI = 0.90, and
RMSEA = 0.10). The modified Active Coping factor was composed of active coping,
planning, and positive reappraisal from McCallum et al.’s (2007) Active Coping factor in
addition to the restraint coping and suppression of competing activities subscales from
the Faith-Oriented Coping factor. Cronbach’s alphas were .82 (African Americans) and
.81 (Whites) for Active Coping and .73 (African Americans and Whites) for Avoidant
Coping.

25

Care recipient behavioral problems.
The behavioral problems of the care recipients, which represent a caregiver
stressor, were measured with nine items of the Revised Memory and Behavior Problems
Checklist (RMBPC; Teri et al., 1992). Responses were scored from 0 (not at all) to 4
(always), with higher scores representing increased caregiving demands. Kim et al.
(2007) reported Cronbach’s alphas of .88 in African Americans and .78 in Whites in this
sample of caregivers.
Control Variables
Age, ethnicity, and education level were included as background variables, as
they are known to be highly correlated with physical health. Education level was assessed
using six categories: elementary school, middle school/junior high, high school, some
college, undergraduate degree, and master’s degree or beyond (scored from 1 to 6).
Although income level has been demonstrated to be an important factor affecting
physical health, it was not included as a control variable, as it was found to be highly
correlated with education level in a previous study that employed the same sample (Kim
et al., 2007). In addition, the familial relationship of the caregiver to the care recipient
and coresidence were not included because analyses by Kim et al. revealed that
correlations between both of these factors and caregiver burden and physical health were
low (r < .20).
Analysis
In order to test the study hypotheses, path models were assessed using the
statistical software AMOS 17.0 by examination of the statistical significance of estimated

26

path coefficients and various statistics indicating goodness-of-fit for the models as a
whole, including NFI, CFI, χ
2
/df, and RMSEA. Models with a better fit generally have
higher NFI and CFI and lower χ
2
/df and RMSEA values. The rules for establishing
goodness-of-fit for models vary. However, Bentler (1992) stated that values greater than
.90 for both the NFI and CFI represent well-fitting models, as do χ
2
/df values of less than
3.0 (e.g., Hu & Bentler, 1999). Browne and Cudeck (1993) reported that RMSEA values
of up to 0.1 signify reasonable approximation errors in the population.
The adequacy of models were tested in which all unidirectional paths excluding
paths from the two familism factors and the Cultural Justification factor were freely
estimated. The direct paths from these three cultural factors to the outcome variables
were omitted from the models for theoretical reasons, as the effects of these factors are
viewed as being mediated by coping and appraisal as opposed to being directly associated
with the outcomes of interest. The covariances between the disturbance terms of the
outcome variables (depression, psychological symptoms, heart rate, and systolic and
diastolic blood pressure) were estimated in the model. Correlations among the exogenous
variables were examined before restricting the covariances between them to zero. Zero-
order correlations among model variables are reported in Table 5.

27

Table 5.
Correlation Coefficients among Key Variables
Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1. 1.00
2. .153 1.00
3. -.052 -.046 1.00
4. -.323
**
-.097 -.122 1.00
5. -.180
*
- .020 -.197
*
.671
***
1.00
6. .065 -.046 -.037 .390
***
.310
***
1.00
7. -006 .077 .459
***
-.191
*
-.218
**
-.100 1.00
8. .061 .058 .096 .072 .046 .281
**
.030 1.00
9. -.158 -.095 .147 .127 .064 .150 .433
***
.026 1.00
10. -.026 -.084 .140 .062 -.032 .013 .448
***
-.123 .431
***
1.00
11. -.182
*
-.031 .096 -.006 -.023 .009 .391
***
-.061 .423
***
.805
***
1.00

12. .153 -.097 -.117 -.049 -.171
*
.071 .129 -.029 .256
***
.291
***
.148 1.00
13. .305
***
-.094 -.026 -.106 -.042 -.123 .036 -.031 .043 -.066 -.147 .182
*
1.00
14. -.024 -.206
*
.117 .034 .084 -.179
*
.056 -.111 .109 -.033 -.069 .089 .719
***
1.00
15. -.190
*
-.118 .000 .111 .063 -.131 .033 -.086 .132 .066 .002 -.001 .215
***
.429
***
1.00

Note. 1 = Caregivers’ age; 2 = Caregiver’s education; 3 = Care recipients’ behavioral problems; 4 = Familial Obligations; 5 = Perceived Support
from the Family; 6 = Cultural Justification; 7 = Caregiver burden; 8 = Active Coping; 9 = Avoidant coping; 10 = Depression (Center for
Epidemiological Studies – Depression Scale); 11 = Psychological symptoms (General Symptom Index); 12 = Self-reported general physical
health; 13 = Systolic blood pressure; 14 = Diastolic blood pressure; 15 = Heart rate.

*p < .05. ** p < .01. *** p < .001

28

Less than 10% of the cases in this data set had item-level missing values. Prorated
scores were calculated when there were fewer than three missing items on scales with
multiple items. When a reading was missing on CV indicators, the other two readings
were used to calculate the average reading for that participant. With the inclusion of these
interpolated values, scale scores were missing in less than 5% of cases for all scales and
single-item variables. The percentages of missing values did not significantly differ
between African American and White participants.
Sufficient power was expected on the basis of the Kim et al. (2007) study, which
employed the same sample and similar path modeling analyses and reported significant
findings. Path coefficients with absolute values as small as .14 were found to be
significant in this sample.

29

Chapter 3: Results
Analyses of the hypothesized path models revealed that in nearly all cases, the fit
statistics indicated acceptable levels of model fit with the exception of the model for
Perceived Support from the Family and self-reported general physical health without
controlling for education (Table 6). Aside from this model, chi-square values suggested
that the data adequately fit the hypothesized models. Similarly, the other model fit
statistics that take into account sample sizes or degrees of freedom also demonstrated
acceptable levels of model fit.

30

Table 6: Goodness of Model Fit Statistics of the First Models
Model χ
2
df χ
2
/df p NFI CFI RMSEA
(90%)
ECVI
Without controlling for education
Familial
Obligations &
Mental Health
16.17 7 2.310 .024 .954 .970 .091
(.031,
.150)
.693
Familial
Obligations &
Physical Health
9.04 6 1.507 .171 .941 .974 .056
(.000,
.127)
.535
Familial
Obligations &
Cardiovascular
Health
9.60 8 1.200 .294 .971 .994 .035
(.000,
.104)
.777
Perceived
Support &
Mental Health
13.63 7 1.947 .058 .960 .977 .077
(.000,
.138)
.677
Perceived
Support &
Physical Health
19.88 6 3.313 .003 .866 .876 .121
(.064,
.181)
.603
Perceived
Support &
Cardiovascular
Health
16.30 8 2.038 .038 .949 .969 .081
(.018,
.137)
.820
Cultural
Justification &
Mental Health
5.17 7 .738 .639 .985 1.000 .000
(.000,
.080)
.624
Cultural
Justification &
Physical Health
4.90 6 .816 .557 .965 1.000 .000
(.000,
.092)
.509
Cultural
Justification &
Cardiovascular
Health
10.36 8 1.294 .241 .968 .991 .043
(.000,
.108)
.782
Controlling for education
Familial
Obligations &
Mental Health
18.90 9 2.100 .026 .950 .969 .083
(.023,
.136)
.823
Familial
Obligations &
Physical Health
11.49 8 1.437 .175 .934 .973 .052
(.000,
.115)
.651

31

Table 6 (Continued)
Familial
Obligations &
Cardiovascular
Health
11.95 10 1.195 .288 .966 .993 .035
(.000,
.097)
.918
Perceived Support
& Mental Health
16.34 10 1.634 .090 .955 .979 .063
(.000,
.116)
.795
Perceived Support
& Physical Health
22.24 9 2.471 .008 .869 .894 .096
(.046,
.147)
.706
Perceived Support
& Cardiovascular
Health
18.95 11 1.723 .062 .945 .971 .067
(.000,
.117)
.949
Cultural
Justification &
Mental Health
7.42 9 .825 .593 .979 1.000 .000
(.000,
.078)
.751
Cultural
Justification &
Physical Health
7.06 8 .883 .530 .956 1.000 .000
(.000,
.086)
.623
Cultural
Justification &
Cardiovascular
Health
12.79 10 1.279 .235 .963 .990 .042
(.000,
.101)
.923
Note. NFI = normed fit index; CFI = comparative fit index; RMSEA = root-mean-square
error of approximation; ECVI = expected cross-validation index.

32

Final models were then estimated after nonsignificant paths were deleted from the
originally proposed models in the interest of scientific parsimony. The reduced model for
Perceived Support from the Family and self-reported general physical health without
controlling for education did not fit the data well based on the model’s fit statistics.
However, this reduced model did fit reasonably well when controlling for education. In
addition, the fit statistics of the remaining reduced models suggested that the data fit
these models well (Table 7).

33

Table 7: Goodness of Model Fit Statistics of the Reduced Models
Model χ
2
df χ
2
/df p NFI CFI RMSEA
(90%)
ECVI
Without controlling for education
Familial
Obligations &
Mental Health
39.88 24 1.662 .022 .888 .949 .065
(.025,
.099)
.628
Familial
Obligations &
Physical Health
26.25 18 1.458 .094 .829 .930 .054
(.000,
.095)
.492
Familial
Obligations &
Cardiovascular
Health
47.37 33 1.435 .050 .857 .948 .052
(.000,
.084)
.700
Perceived
Support & Mental
Health
46.03 26 1.771 .009 .863 .931 .070
(.035,
.102)
.642
Perceived
Support &
Physical Health
45.54 20 2.277 .001 .693 .772 .090
(.055,
.124)
.588
Perceived
Support &
Cardiovascular
Health
62.92 35 1.798 .003 .804 .895 .071
(.041,
.099)
.773
Cultural
Justification &
Mental Health
23.29 23 1.013 .444 .931 .999 .009
(.000,
.066)
.536
Cultural
Justification &
Physical Health
15.81 17 .930 .538 .887 1.000 .000
(.000,
.067)
.439
Cultural
Justification &
Cardiovascular
Health
42.40 32 1.325 .103 .869 .961 .045
(.000,
.079)
.682
Controlling for education
Familial
Obligations &
Mental Health
49.90 32 1.559 .023 .868 .945 .059
(.023,
.090)
.729
Familial
Obligations &
Physical Health
34.07 25 1.363 .106 .805 .930 .048
(.000,
.085)
.579

34

Table 7 (Continued)
Familial
Obligations &
Cardiovascular
Health
56.71 42 1.350 .064 .840 .949 .047
(.000,
.076)
.797
Perceived
Support & Mental
Health
56.23 34 1.654 .010 .844 .927 .064
(.032,
.093)
.744
Perceived
Support &
Physical Health
53.47 27 1.980 .002 .684 .787 .079
(.047,
.109)
.676
Perceived
Support &
Cardiovascular
Health
72.77 44 1.654 .004 .789 .897 .064
(.036,
.090)
.873
Cultural
Justification &
Mental Health
37.31 32 1.166 .238 .896 .983 .032
(.000,
.070)
.650
Cultural
Justification &
Physical Health
27.87 25 1.115 .314 .827 .975 .027
(.000,
.071)
.540
Cultural
Justification &
Cardiovascular
Health
56.31 42 1.341 .069 .837 .949 .046
(.000,
.075)
.794
Note. NFI = normed fit index; CFI = comparative fit index; RMSEA = root-mean-square
error of approximation; ECVI = expected cross-validation index.

35

While some NFIs did not suggest that the reduced models fit best, CFIs for nearly
all of the reduced models fell within the acceptable range. Given that CFIs have been
suggested as a better index of model fit than NFIs, which have shown a tendency to
underestimate fit in small samples (Bentler, 1992), and that RMSEA values fell within
the acceptable range as well, the remaining reduced models were accepted as the final
models. The reduced models were not significantly different in comparison to the
originally proposed models, as indicated by the changes in the chi-square values (Table
8).
Table 8: Changes in Chi-Square Fit Indices from the First Models to the Reduced Models
Model Δχ
2
Δdf Δχ
2
/Δdf
Without controlling for education
Familial Obligations & Mental Health 23.714 17 1.395
Familial Obligations & Physical Health 17.204 12 1.434
Familial Obligations & Cardiovascular Health 37.768 25 1.511
Perceived Support & Mental Health 32.405 19 1.706
Perceived Support & Physical Health 25.659 14 1.833
Perceived Support & Cardiovascular Health 46.618 27 1.727
Cultural Justification & Mental Health 18.122 16 1.133
Cultural Justification & Physical Health 10.912 11 0.992
Cultural Justification & Cardiovascular Health 32.045 24 1.335
Controlling for education
Familial Obligations & Mental Health 30.998 23 1.348
Familial Obligations & Physical Health 22.574 17 1.328
Familial Obligations & Cardiovascular Health 44.754 32 1.399
Perceived Support & Mental Health 39.887 24 1.662
Perceived Support & Physical Health 31.228 18 1.735
Perceived Support & Cardiovascular Health 54.014 33 1.637
Cultural Justification & Mental Health 29.884 23 1.299
Cultural Justification & Physical Health 20.805 17 1.224
Cultural Justification & Cardiovascular Health 43.521 32 1.360

36

The final path models with standardized path coefficients for all three cultural
values and mental health outcomes, self-reported general physical health outcomes, and
CV measures are presented in Figures 1 – 18. The unstandardized coefficients and their
significance levels, both with and without controlling for education, for all of the
originally proposed models are reported in Tables 9 – 17.

37

Table 9: Unstandardized Regression Coefficients Estimated from a Fully Recursive Path
Model Predicting Caregivers’ Familial Obligations, Caregiver Burden, Coping Style,
and Mental Health (N = 160)
Endogenous variable

Independent variable
Familial
Obligations

Burden
Active
coping
Avoidant
coping

CES-D

GSI
Without controlling for education
Exogenous
Age -0.07
***
-0.05 0.09 -0.04 0.00 -0.01
*

Ethnicity 1.19
*
-4.22 4.05
*
0.75 -0.66 -0.13
Behavioral
problem
— 1.03
***
0.15 -0.07 -0.03 -0.01
Mediating
Familial
Obligations
— -0.49 0.27 0.34
*
— —
Burden — — 0.02 0.24
***
0.14
***
0.01
***

Active coping — — — — -0.09 0.00
Avoidant coping — — — — 0.26
***
0.02
***

Controlling for education
Exogenous
Age -0.07
***
-0.06 0.08 -0.04 0.01 -0.01
*

Education -0.05 1.27 1.37 -0.89 -0.78 -0.02
Ethnicity 1.16
*
-3.65 4.68
*
0.37 -1.05 -0.14
Behavioral
problem
— 1.04
***
0.16 -0.08 -0.04 -0.01
Mediating
Familial
Obligations
— -0.48 0.27 0.34
*
— —
Burden — — 0.01 0.24
***
0.15
***
0.01
***

Active coping — — — — -0.08 0.00
Avoidant coping — — — — 0.25
***
0.02
***

Note. Dashes indicate the regression coefficients were not estimated because the paths
were not hypothesized in the model. CES-D = Center for Epidemiological Studies-
Depression Scale; GSI = General Symptom Index.
* p < .05. ** p < .01. *** p < .001.

38

Table 10: Unstandardized Regression Coefficients Estimated from a Fully Recursive Path
Model Predicting Caregivers’ Familial Obligations, Caregiver Burden, Coping Style,
and Self-reported General Physical Health (N = 160)
Endogenous variable

Independent variable

Familial
Obligations

Burden

Active
coping

Avoidant
coping
Self-reported
physical health
Without controlling for education
Exogenous
Age -0.07
***
-0.04 0.09 -0.05 0.01
**

Ethnicity 1.19
*
-4.21 4.08
*
0.69 0.35
**

Behavioral problem — 1.03
***
0.16 -0.07 -0.03
**

Mediating
Familial Obligations — -0.51 0.31 0.35
*
—
Burden — — 0.01 0.24
***
0.01
Active coping — — — — -0.01
Avoidant coping — — — — 0.03
**

Controlling for education
Exogenous
Age -0.07
***
-0.05 0.08 -0.04 0.01
**

Education -0.05 1.29 1.42 -0.88 -0.06
Ethnicity 1.16
*
-3.64 4.72
*
0.32 0.32
*

Behavioral problem — 1.04
***
0.17 -0.08 -0.03
**

Mediating
Familial Obligations — -0.50 0.31 0.35
*
—
Burden — — 0.00 0.24
***
0.01
Active coping — — — — 0.00
Avoidant coping — — — — 0.02
*

Note. Dashes indicate the regression coefficients were not estimated because
the paths were not hypothesized in the model.
* p < .05. ** p < .01. *** p < .001.

39

Table 11: Unstandardized Regression Coefficients Estimated from a Fully Recursive Path
Model Predicting Caregivers’ Familial Obligations, Caregiver Burden, Coping Style,
and Cardiovascular Health (N = 160)
Endogenous variable

Independent
variable

Familial
Obliga-
tions

Burden

Act-
ive
coping
Avoid
-ant
coping SBP DBP
Heart
rate
Without controlling for education
Exogenous
Age -0.07
***
-0.04 0.09 -0.05 0.55
***
0.05 -0.11
Ethnicity 1.19
*
-4.26 4.22
*
0.73 14.57
***
8.60
***
2.79
Behavioral
problem
— 1.03
***
0.16 -0.08 -0.25 0.20 -0.01
Mediating
Familial
Obligations
— -0.51 0.28 0.36
*
— — —
Burden — — 0.01 0.25
***
0.12 0.00 0.01
Active
coping
— — — — -0.22 -0.24
*
-0.10
Avoidant
coping
— — — — 0.19 0.17 0.14
Controlling for education
Exogenous
Age -0.07
***
-0.05 0.08 -0.04 0.56
***
0.06 -0.10
Education -0.04 1.37 1.54 -0.91 -2.37 -1.77 0.71
Ethnicity 1.16
*
-3.65 4.90
**
0.35 13.41
***
7.72
***
2.43
Behavioral
problem
— 1.04
***
0.18 -0.09 -0.27 0.18 -0.02
Mediating
Familial
Obligations
— -0.50 0.28 0.36
*
— — —
Burden — — 0.00 0.25
***
0.15 0.02 0.01
Active
coping
— — — — -0.20 -0.22
*
-0.10
Avoidant
coping
— — — — 0.15 0.15 0.13
Note. Dashes indicate the regression coefficients were not estimated because the paths
were not hypothesized in the model. SBP = systolic blood pressure. DBP = diastolic
blood pressure. * p < .05. ** p < .01. *** p < .001.

40

Table 12: Unstandardized Regression Coefficients Estimated from a Fully Recursive Path
Model Predicting Caregivers’ Perceived Support from the Family, Caregiver Burden,
Coping Style, and Mental Health (N = 160)
Endogenous variable

Independent variable
Perceived
Support
from the
Family

Burden

Active
coping

Avoidant
coping

CES-D

GSI
Without controlling for education
Exogenous
Age -0.03
*
-0.03 0.07 -0.06 0.00 -0.01
*

Ethnicity 0.81 -4.22 4.14
*
0.87 -0.65 -0.13
Behavioral
problem
— 1.02
***
0.15 -0.06 -0.03 -0.01
Mediating
Perceived Support
from the Family
— -0.67 0.25 0.34 — —
Burden — — 0.02 0.23
***
0.15
***
0.01
***

Active coping — — — — -0.09 0.00
Avoidant coping — — — — 0.26
***
0.02
***

Controlling for education
Exogenous
Age -0.03
*
-0.04 0.06 -0.05 0.01 -0.01
*

Education — 1.36 1.32 -0.93 -0.78 -0.02
Ethnicity 0.81 -3.60 4.77
*
0.46 -1.04 -0.14
Behavioral
problem
— 1.03
***
0.16 -0.08 -0.04 -0.01
Mediating
Perceived Support
from the Family
— -0.67 0.23 0.35 — —
Burden — — 0.01 0.24
***
0.15
***
0.01
***

Active coping — — — — -0.08 0.00
Avoidant coping — — — — 0.24
***
0.02
***

Note. Dashes indicate the regression coefficients were not estimated because the paths
were not hypothesized in the model. CES-D = Center for Epidemiological Studies-
Depression Scale; GSI = General Symptom Index.
* p < .05. ** p < .01. *** p < .001.

41

Table 13: Unstandardized Regression Coefficients Estimated from a Fully Recursive Path
Model Predicting Caregivers’ Perceived Support from the Family, Caregiver Burden,
Coping Style, and Self-reported General Physical Health (N = 160)
Endogenous variable

Independent variable

Perceived
Support
from the
Family

Burden

Active
coping

Avoidant
coping

Self-
reported
physical
health
Without controlling for education
Exogenous
Age -0.03
*
-0.03 0.07 -0.06 0.01
**

Ethnicity 0.82 -4.26 4.19
*
0.81 0.35
**

Behavioral problem — 1.01
***
0.16 -0.07 -0.03
**

Mediating
Perceived Support from
the Family
— -0.66 0.27 0.35 —
Burden — — 0.01 0.24
***
0.01
Active coping — — — — -0.01
Avoidant coping — — — — 0.03
**

Controlling for education
Exogenous
Age -0.03
*
-0.04 0.06 -0.05 0.01
**

Education — 1.39 1.38 -0.93 -0.06
Ethnicity 0.82 -3.62 4.83
*
0.41 0.32
*

Behavioral problem — 1.02
***
0.17 -0.07 -0.03
**

Mediating
Perceived Support from
the Family
— -0.66 0.24 0.36 —
Burden — — 0.00 0.24
***
0.01
Active coping — — — — 0.00
Avoidant coping — — — — 0.02
**

Note. Dashes indicate the regression coefficients were not estimated because
the paths were not hypothesized in the model.
* p < .05. ** p < .01. *** p < .001.

42

Table 14: Unstandardized Regression Coefficients Estimated from a Fully Recursive Path
Model Predicting Caregivers’ Perceived Support from the Family, Caregiver Burden,
Coping Style, and Cardiovascular Health (N = 160)
Endogenous variable

Independent
variable PSFF

Burden

Active
coping
Avoid-
ant
coping SBP DBP
Heart
rate
Without controlling for education
Exogenous
Age -0.03
*
-0.02 0.08 -0.06 0.55
***
0.05 -0.11
Ethnicity 0.82 -4.30 4.32
*
0.86 14.57
***
8.60
***
2.79
Behavior
problem
— 1.01
***
0.16 -0.08 -0.25 0.20 -0.01
Mediating
PSFF — -0.65 0.24 0.36 — — —
Burden — — 0.00 0.24
***
0.12 0.00 0.01
Active
coping
— — — — -0.22 -0.24
*
-0.10
Avoidant
coping
— — — — 0.19 0.17 0.14
Controlling for education
Exogenous
Age -0.03
*
-0.03 0.07 -0.05 0.56
***
0.06 -0.10
Education — 1.47 1.50 -0.96 -2.37 -1.76 -0.71
Ethnicity 0.82 -3.63 5.01
**
0.45 13.42
***
7.73
***
2.43
Behavioral
problem
— 1.02
***
0.17 -0.09 -0.27 0.18 -0.02
Mediating
PSFF — -0.66 0.22 0.37 — — —
Burden — — -0.01 0.25
***
0.15 0.02 0.01
Active
coping
— — — — -0.20 -0.23
*
-0.10
Avoidant
coping
— — — — 0.15 0.14 0.13
Note. Dashes indicate the regression coefficients were not estimated because the paths
were not hypothesized in the model. PSFF = Perceived Support from the Family. SBP =
systolic blood pressure. DBP = diastolic blood pressure. * p < .05. ** p < .01. *** p <
.001.

43

Table 15: Unstandardized Regression Coefficients Estimated from a Fully Recursive Path
Model Predicting Caregivers’ Cultural Justification, Caregiver Burden, Coping Style,
and Mental Health (N = 160)
Endogenous variable
Independent
variable
Cultural
Justification

Burden
Active
coping
Avoidant
coping

CES-D

GSI
Without controlling for education
Exogenous
Age 0.04 -0.01 0.05 -0.07
*
0.00 -0.01
*

Ethnicity 1.66 -4.59
*
3.82
*
0.87 -0.66 -0.13
Behavioral
problem
— 1.07
***
0.13 -0.09 -0.03 -0.01
Mediating
Cultural
Justification
— -0.12 0.35
**
0.18
*
— —
Burden — — 0.03 0.23
***
0.15
***
0.01
***

Active coping — — — — -0.09 0.00
Avoidant
coping
— — — — 0.26
***
0.02
***

Controlling for education
Exogenous
Age 0.04 -0.01 0.05 -0.07
*
0.01 -0.01
*

Education -0.20 1.28 1.32 -0.87 -0.79 -0.02
Ethnicity 1.56 -4.02 4.42
*
0.50 -1.05 -0.14
Behavioral
problem
— 1.08
***
0.14 -0.10 -0.04 -0.01
Mediating
Cultural
Justification
— -0.12 0.35
**
0.17
*
— —
Burden — — 0.02 0.24
***
0.15
***
0.01
***

Active coping — — — — -0.08 0.00
Avoidant
coping
— — — — 0.25
***
0.02
***

Note. Dashes indicate the regression coefficients were not estimated because the paths
were not hypothesized in the model. CES-D = Center for Epidemiological Studies-
Depression Scale; GSI = General Symptom Index.
* p < .05. ** p < .01. *** p < .001.

44

Table 16: Unstandardized Regression Coefficients Estimated from a Fully Recursive Path
Model Predicting Caregivers’ Cultural Justification, Caregiver Burden, Coping Style,
and Self-reported General Physical Health (N = 160)
Endogenous variable

Independent variable

Cultural
Justification

Burden

Active
coping

Avoidant
coping

Self-reported
physical
health
Without controlling for education
Exogenous
Age 0.04 0.00 0.06 0.08
*
0.01
**

Ethnicity 1.66 -4.61
*
3.88
*
0.82 0.35
**

Behavioral problem — 1.06
***
0.13 -0.09 -0.03
*

Mediating
Cultural Justification — -0.13 0.36
***
0.18
*
—
Burden — — 0.02 0.24
***
0.01
Active coping — — — — -0.01
Avoidant coping — — — — 0.03
**

Controlling for education
Exogenous
Age 0.04 -0.01 0.05 -0.07
*
0.01
**

Education -0.19 1.30 1.36 -0.86 -0.06
Ethnicity 1.57 -4.02 4.49
*
0.45 0.32
*

Behavioral problem — 1.07
***
0.15 -0.10 -0.03
**

Mediating
Cultural Justification — -0.12 0.35
***
0.17
*
—
Burden — — 0.01 0.24
***
0.01
Active coping — — — — -0.01
Avoidant coping — — — — 0.02
**

Note. Dashes indicate the regression coefficients were not estimated because
the paths were not hypothesized in the model.
* p < .05. ** p < .01. *** p < .001.

45

Table 17: Unstandardized Regression Coefficients Estimated from a Fully Recursive Path
Model Predicting Caregivers’ Cultural Justification, Caregiver Burden, Coping Style,
and Cardiovascular Health (N = 160)
Endogenous variable

Independent
variable

Cultural
Justifica
-tion

Burden

Active
coping
Avoid-
ant
coping SBP DBP
Heart
rate
Without controlling for education
Exogenous
Age 0.04 0.00 0.06 -0.08
*
0.55
***
0.05 -0.11
Ethnicity 1.67 -4.66
*
3.98
*
0.87 14.58
***
8.63
***
2.80
Behavioral
problem
— 1.06
***
0.14 -0.10 -0.25 0.20 -0.02
Mediating
Cultural
Justification
— -0.12 0.36
***
0.18
**
— — —
Burden — — 0.02 0.24
***
0.12 0.00 0.01
Active
coping
— — — — -0.23 -0.25
*
-0.11
Avoidant
coping
— — — — 0.19 0.17 0.14
Controlling for education
Exogenous
Age 0.04 -0.01 0.05 -0.07
*
0.56
***
0.06 -0.10
Education -0.10 1.39 1.45 -0.90 -2.35 -1.76 -0.71
Ethnicity 1.62 -4.03 4.63
*
0.49 13.45
***
7.76
***
2.44
Behavioral
problem
— 1.07
***
0.15 -0.11 -0.28 0.18 0.01
Mediating
Cultural
Justification
— -0.12 0.36
***
0.18
*
— — —
Burden — — 0.01 0.25
***
0.14 0.02 0.01
Active
coping
— — — — -0.21 -0.23
*
-0.10
Avoidant
coping
— — — — 0.15 0.14 0.12
Note. Dashes indicate the regression coefficients were not estimated because the paths
were not hypothesized in the model. SBP = systolic blood pressure. DBP = diastolic
blood pressure. * p < .05. ** p < .01. *** p < .001.

46

Hypothesis 1 predicted that Familial Obligations would be associated with higher
levels of both burden and use of avoidant coping, which would result in negative mental
and physical health outcomes among family caregivers. This hypothesis received partial
support (Figures 1 – 6) in that while this cultural value was not associated with burden in
any of the models, it was in fact associated with avoidant coping in all of the models.
Moreover, avoidant coping in all of these models was associated with both poor mental
and self-reported general physical health outcomes yet not with poor CV health
indicators. In addition, higher levels of burden were related to avoidant coping in all of
these models, which in turn led to poor mental and self-reported general physical health
outcomes in the models but not to poor CV health outcomes. Burden also exhibited a
significant direct effect on poor mental health outcomes but not on poor self-reported
general physical health or CV health indicators.

47

Figure 1: Fitted Path Model of Familial Obligations, Burden, Coping Style, and Mental
Health of Caregivers (Standardized Path Coefficients)
Age
Ethnicity
Behavioral
Problems
Familial
Obligations
Burden
Avoidant
Coping
Active
Coping
Depression
Psychological
Symptoms
-.30
.19
.46
e1
e2
.46
e3
e4
e5
e6
.75
.16
.28
.28
.28
-.14
.21
.34

48

Figure 2: Fitted Path Model of Familial Obligations, Burden, Coping Style, Education,
and Mental Health of Caregivers (Standardized Path Coefficients)
Age
Ethnicity
Education
Behavioral
Problems
-.29
Familial
Obligations
Burden
Avoidant
Coping
Active
Coping
Depression
Psychological
Symptoms
.19
-.14
.46
e1
e2
.46
.28
.28
e3
e4
e5
e6
.75
.16
.28
.34
.21
-.30

49

Figure 3: Fitted Path Model of Familial Obligations, Burden, Coping Style, and Self-
reported General Physical Health of Caregivers (Standardized Path Coefficients)
Age
Ethnicity
Behavioral
Problems
Familial
Obligations
Burden
Avoidant
Coping
Active
Coping
Poor Physical
Health
-.30
.19
.22
.45
e1
e2
.46
.21
.30
e3
e4
e5
.16
-.16
.18

50

Figure 4: Fitted Path Model of Familial Obligations, Burden, Coping Style, Education,
and Self-reported General Physical Health of Caregivers (Standardized Path
Coefficients)
Age
Ethnicity
Education
Behavioral
Problems
-.29
Familial
Obligations
Burden
Avoidant
Coping
Active
Coping
Poor Physical
Health
-.30
.19
.45
e1
e2
.46
.21
.30
e3
e4
e5
.18
.16
-.16
.22

51

Figure 5: Fitted Path Model of Familial Obligations, Burden, Coping Style, and
Cardiovascular Health of Caregivers (Standardized Path Coefficients)
Age
Ethnicity
Behavioral
Problems
Familial
Obligations
Burden
Avoidant
Coping
Active
Coping
Systolic BP
Heart Rate
-.30
.45
e1
e2
.47
.21
e3
e4
e5
e7
.28 Diastolic BP e6
.74
.41
.23
.19
.16
.24
.32

52

Figure 6: Fitted Path Model of Familial Obligations, Burden, Coping Style, Education,
and Cardiovascular Health of Caregivers (Standardized Path Coefficients)
Age
Ethnicity
Education
Behavioral
Problems
-.29
Familial
Obligations
Burden
Avoidant
Coping
Active
Coping
Systolic BP
Heart Rate
-.30
.45
e1
e2
.47
.21
e3
e4
e5
e7
.28 Diastolic BP e6
.74
.41
.16
.24
.32
.23
.19

Hypothesis 2, that Perceived Support from the Family would be related to lower
levels of both burden and use of active coping that would in turn result in better mental
and physical health outcomes among caregivers, was not supported at all. Unexpectedly,
this cultural value did not have any direct or indirect effects on any of the variables
examined in the models in which it was included (Figures 7 – 12). Rather, the mental and
physical health outcomes in these models were only affected by other variables in the
models. In partial support of this hypothesis, higher levels of burden had a significant
direct association with poor mental health outcomes as predicted but not with any of the
other outcomes. Burden also had a significant indirect effect on poor mental and self-
reported general physical health outcomes via avoidant coping. Active coping, however,
was not associated with any of the health outcomes in these models.

53

Figure 7: Fitted Path Model of Perceived Support from the Family, Burden, Coping Style,
and Mental Health of Caregivers (Standardized Path Coefficients)
Age
Ethnicity
Behavioral
Problems
Perceived
Support
Burden
Avoidant
Coping
Active
Coping
Depression
Psychological
Symptoms
-.18
.19
-.14
.46
e1
e2
.43
.27
.28
.29
.34
e3
e4
e5
e6
.74

54

Figure 8: Fitted Path Model of Perceived Support from the Family, Burden, Coping Style,
Education, and Mental Health of Caregivers (Standardized Path Coefficients)
Age
Ethnicity
Education
Behavioral
Problems
-.29
Perceived
Support
Burden
Avoidant
Coping
Active
Coping
Depression
Psychological
Symptoms
-.18
.19
-.14
.46
e1
e2
.43
.27
.28
.29
.34
e3
e4
e5
e6
.74

55

Figure 9: Fitted Path Model of Perceived Support from the Family, Burden, Coping Style,
and Self-reported General Physical Health of Caregivers (Standardized Path
Coefficients)
Age
Ethnicity
Behavioral
Problems
Perceived
Support
Burden
Avoidant
Coping
Active
Coping
Poor Physical
Health
.19
.22
.45
e1
e2
.43
.30
e3
e4
e5
.18
-.16
-.18

56

Figure 10: Fitted Path Model of Perceived Support from the Family, Burden, Coping
Style, Education, and Self-reported General Physical Health of Caregivers (Standardized
Path Coefficients)
Age
Ethnicity
Education
Behavioral
Problems
-.29
Perceived
Support
Burden
Avoidant
Coping
Active
Coping
Poor Physical
Health
-.18
.45
e1
e2
.43
.30
e3
e4
e5
-.16
.22
.18
.19

57

Figure 11: Fitted Path Model of Perceived Support from the Family, Burden, Coping
Style, and Cardiovascular Health of Caregivers (Standardized Path Coefficients)
Age
Ethnicity
Behavioral
Problems
Perceived
Support
Burden
Avoidant
Coping
Active
Coping
Systolic BP
Heart Rate
-.18
.46
e1
e2
.44
e3
e4
e5
e7
.28 Diastolic BP e6
.74
.41
.32
.19
.24
.23

58

Figure 12: Fitted Path Model of Perceived Support from the Family, Burden, Coping
Style, Education, and Cardiovascular Health of Caregivers (Standardized Path
Coefficients)
Age
Ethnicity
Education
Behavioral
Problems
-.29
Perceived
Support
Burden
Avoidant
Coping
Active
Coping
Systolic BP
Heart Rate
-.18
.46
e1
e2
.44
e3
e4
e5
e7
.28 Diastolic BP e6
.74
.41
.32
.24
.23
.19

Hypothesis 3 predicted that Cultural Justification would be associated with lower
levels of both burden and use of active coping, which would lead to better mental and
physical health outcomes among caregivers. As was the case with Hypothesis 1
pertaining to Familial Obligations, this hypothesis also received partial support (Figures
13 – 18). Similar to the previously discussed findings from the models that examined
Familial Obligations and Perceived Support from the Family, this cultural value did not
have a significant effect on burden in any of the models in which it was included,
contrary to expectation. However, Cultural Justification was in fact significantly
associated with active coping, yet active coping did not have a significant effect on any of
the mental and physical health outcomes in these models or in any of the models in this

59

study. Interestingly, Cultural Justification was also significantly related to avoidant
coping, which was in turn related to poor mental and self-reported general physical health
outcomes but not poor CV health indicators. Nonetheless, the association of Cultural
Justification with active coping was stronger than its association with avoidant coping.
As was the case with all of the models tested in this study, burden had a direct effect on
poor mental health outcomes as well as an indirect effect on poor self-reported general
physical health outcomes via avoidant coping.
Figure 13: Fitted Path Model of Cultural Justification, Burden, Coping Style, and Mental
Health of Caregivers (Standardized Path Coefficients)
Age
Ethnicity
Behavioral
Problems
Cultural
Justification
Burden
Avoidant
Coping
Active
Coping
Depression
Psychological
Symptoms
.16
.46
e1
e2
.44
.26
.19
.28
e3
e4
e5
e6
.75
-.15
-.16
.28
-.14
.28
.34

60

Figure 14: Fitted Path Model of Cultural Justification, Burden, Coping Style, Education,
and Mental Health of Caregivers (Standardized Path Coefficients)
Age
Ethnicity
Education
Behavioral
Problems
-.29
Cultural
Justification
Burden
Avoidant
Coping
Active
Coping
Depression
Psychological
Symptoms
.46
e1
e2
.44
.21
.19
.28
.28
e3
e4
e5
e6
.75
-.14
.21
.16
.27
.34
-.16

61

Figure 15: Fitted Path Model of Cultural Justification, Burden, Coping Style, and Self-
reported General Physical Health of Caregivers (Standardized Path Coefficients)
Age
Ethnicity
Behavioral
Problems
Cultural
Justification
Burden
Avoidant
Coping
Active
Coping
Poor Physical
Health
.16
.22
.46
e1
e2
.44
.26
.19
.30
e3
e4
e5
-.16
.18
-.17
-.15

62

Figure 16: Fitted Path Model of Cultural Justification, Burden, Coping Style, Education,
and Self-reported General Physical Health of Caregivers (Standardized Path
Coefficients)
Age
Ethnicity
Education
Behavioral
Problems
-.29
Cultural
Justification
Burden
Avoidant
Coping
Active
Coping
Poor Physical
Health
.16
.22
.45
e1
e2
.44
.26
.19
.30
e3
e4
e5
-.16
.18
-.17

63

Figure 17: Fitted Path Model of Cultural Justification, Burden, Coping Style, and
Cardiovascular Health of Caregivers (Standardized Path Coefficients)
Age
Ethnicity
Behavioral
Problems
Cultural
Justification
Burden
Avoidant
Coping
Active
Coping
Systolic BP
Heart Rate
.46
e1
e2
.45
.19
e3
e4
e5
e7
.28 Diastolic BP e6
.74
.41
-.15
.26
.32
-.17
.24
.16
.23

64

Figure 18: Fitted Path Model of Cultural Justification, Burden, Coping Style, Education
and Cardiovascular Health of Caregivers (Standardized Path Coefficients)
Age
Ethnicity
Education
Behavioral
Problems
-.29
Cultural
Justification
Burden
Avoidant
Coping
Active
Coping
Systolic BP
Heart Rate
.45
e1
e2
.45
.19
e3
e4
e5
e7
.28 Diastolic BP e6
.74
.41
-.17
.32
.23
.16
.24
.26

65

Chapter 4: Discussion

The findings of this study suggest that only one of the two factors of familism—
Familial Obligations—affects the mental and physical health of family caregivers of
individuals with dementia. This factor detrimentally affected the mental and self-reported
general physical health of the caregivers in this study through avoidant coping but did not
affect the CV measures. This finding supports the conjecture by Kim et al. (2007) that
“familism…may represent obligation more than positive feelings about family support”
(p. 573). This result is also somewhat consistent with a recent study by Losada et al. (in
press) that found that Familial Obligations was associated with dysfunctional thoughts,
which in turn led to depressive symptomatology among their sample of 334 Spanish
caregivers. Indeed, it appears as though the Familial Obligations factor accounts for a
large proportion of the harmful effects of familism on caregivers’ mental and physical
health that has been reported in previous studies.
Interestingly, the Perceived Support from the Family factor of familism had
neither any direct nor indirect effect on caregivers’ health outcomes. While it is
conceptually appealing to view positive perceptions of familial social support as
protective against poor health outcomes for caregivers, this hypothesis did not receive
any empirical support in this study. Several other studies have reported that various kinds
of social support protect against negative health outcomes among caregivers (e.g.,
Barusch & Spaid, 1989; Losada et al. (in press); Pinquart & Sörensen, 2007). The finding
regarding the null effect of Perceived Support from the Family in this study is somewhat
consistent with the results of a study by Chun, Knight, and Youn (2007) that included a

66

sample of Korean, Korean American, and White American caregivers for elderly patients
with dementia. The authors reported that neither instrumental nor emotional support had
a significant effect on depression or anxiety outcomes among the White American
caregivers in their study, whereas emotional support had a protective effect on anxiety
among the Korean American caregivers in their study. Chun and colleagues suggested
that Korean American caregivers may benefit more from emotional support, as they may
be coping with the stress associated with acculturation and new social roles such as
children and spouses taking over the caregiving role that is traditionally held by
daughters-in-law in Korea. The caregivers in the current study were generally
nonimmigrants and thus may not have benefited from emotional support, similar to the
White American caregivers in the Chun et al. study. It should be noted, however, that
Chun and colleagues’ study did not specifically ask about familial support but rather
support from anyone in the caregivers’ lives.
Other possible explanations for the finding that Perceived Support from the
Family did not affect health outcomes among the caregivers in this study can be posited.
For example, it is possible that this factor is in fact tapping something other than
perceptions of social support from the family (e.g., expectations of or beliefs about social
support, actual amount of social support), which may not have an impact on caregivers’
mental and physical health. The wording of the items that combine to create this factor
may lend itself to such various interpretations; specifically, these three items all use the
words, “One can count on…” when referring to familial support, which may be construed
as what one expects or believes should be the case or what one actually experiences. In

67

contrast, the items that belong to the Familial Obligations factor include terminology
suggesting rigid rules or beliefs (e.g., “One should…” or “Children should…”). It is
possible that the belief or expectation that family members should provide support to
caregivers may in fact counteract any potential protective effects on outcomes if the
actual support is not obtained or does not meet expectations.
In addition, the principal components factor analysis with varimax rotation used
in Sabogal et al.’s (1987) exploratory factor analysis of this familism scale revealed that
the Perceived Support from the Family factor accounted for 10.9% of the variance of the
total variance of 48.4% accounted for by familism’s three factors. On the contrary,
Familial Obligations accounted for a much larger proportion of the total variance
(27.7%). The third factor (Family as Referents), which was not examined in this study,
accounted for 9.8% of the total variance. Based on the logic behind principal components
analysis (see Jolliffe, 2002), the Familial Obligations factor explains the greatest amount
of variance of familism as compared to the other factors. The comparatively small
amount of variance accounted for by the Perceived Support from the Family factor,
coupled with this study’s relatively small sample size, may partially explain why no
significant associations were obtained with this factor and caregivers’ health outcomes in
this study.
The findings regarding the effects of Cultural Justification were somewhat mixed
and only partially supported the hypothesis. For example, this cultural value was
significantly associated with both active and avoidant coping, whereas it was
hypothesized that it would only be related to active coping. Nonetheless, the association

68

of this cultural value was stronger with active coping than avoidant coping, consistent
with the hypothesis. The finding that caregivers sometimes utilized high levels of both
active and avoidant coping is in line with prior research that suggests that people use
higher levels of varying coping styles when they appraise a stressor as being significantly
hazardous to their well-being (e.g., Lazarus & Folkman, 1984). Similar findings were
reported by Haley et al. (1987) in their study of African American and White family
caregivers of elderly dementia patients; specifically, White caregivers used higher levels
of both approach (similar to active) and avoidant coping than African American
caregivers in their study sample. Therefore, the finding that Cultural Justification was
significantly related to both of the coping styles examined in this study does not
necessarily seem unreasonable.
Another unexpected finding with regard to Cultural Justification is that its
significant association with active coping did not in turn have a protective effect on
health outcomes for these caregivers. In fact, active coping had no significant effect on
the mental and physical health outcomes for caregivers in all of the models that were
tested in this study. The finding that Cultural Justification was associated with avoidant
coping, which in turn led to poor mental and self-reported general physical health
outcomes among these caregivers, was unanticipated. However, this finding is at least
somewhat consistent with the findings of a study by Dilworth-Anderson et al. (2004)
involving African American caregivers that reported that cultural justifications for
caregiving (measured by the CJCS) had a curvilinear effect on caregivers’ psychosocial
health as measured by the five-item self-reported general health subscale of the MOS-36

69

(McHorney, Ware, Lu, & Sherbourne, 1994). Specifically, very weak and very strong
cultural justifications for caregiving predicted poor psychosocial health for the caregivers
in this study. The authors suggested that caregivers who provide care due to a strong
identification with cultural values and beliefs may be driven to do so out of a sense of
obligation. In contrast, caregivers with low levels of such cultural values and beliefs may
be providing care out of necessity, because there may be no one else available to serve as
a caregiver.
While the current study employed a somewhat different measure of cultural justifications
for caregiving (the Cultural Justification factor) than the Dilworth-Anderson et al. (2004)
study, this factor is composed largely of the same items that make up the CJCS, which
may provide at least a partial explanation for the fact that higher levels of Cultural
Justification were associated with poorer mental and self-reported general physical health
outcomes via avoidant coping. A post hoc analysis was performed to examine CJCS
scores in this study’s sample, which revealed that the mean CJCS score for the caregivers
was 31.6 (SD = 6.4) with scores ranging from 12 to 40. These values are quite similar to
the mean and range of CJCS scores from the Dilworth-Anderson et al. study sample,
which reported a mean CJCS score of 36.4 (SD = 4.4) with scores ranging from 10 to 40.
In addition, post-hoc analyses in this study revealed that the lowest CJCS scores (scores
less than 25, as used by Dilworth-Anderson et al.) were associated with a significantly
better mean value on self-reported general physical health ratings (M = 1.9, SD = 0.8)
compared to the mean value (M = 2.3, SD = 0.7) of caregivers with moderate to strong
CJCS scores (scores between 25 and 31, as used by Dilworth-Anderson et al.), t(61) = -

70

2.01, p < .05. The highest CJCS scores in this study (scores greater than or equal to 31, as
used by Dilworth-Anderson et al.) did not differ significantly from the other two levels of
CJCS scores in terms of mean values on self-reported general physical health ratings,
with the mean rating value falling in between the other lower CJCS score groups (M =
2.1, SD = 0.8). This pattern represents a somewhat of an opposite effect compared to the
effect found in the Dilworth-Anderson et al. study. In this study, it is clear that moderate
to strong CJCS scores were generally associated with lower self-reported general
physical health ratings as compared to the lowest CJCS scores. Given that higher scores
reflect having stronger cultural reasons for providing care, it is plausible that higher
levels of cultural justification for caregiving may be accounting for a large amount of the
poor self-reported general physical health outcomes as mediated by avoidant coping in
these models among this sample of caregivers, as these caregivers may be driven to
provide care out of a sense of obligation.
With regard to some of the broad findings related to the hypotheses in this study,
three findings in particular stand out and warrant further attention. First, there was no
pathway from any of the cultural values to burden in the models tested in this study, as
hypothesized. This finding runs counter to the originally proposed sociocultural stress
and coping model by Aranda and Knight (1997). However, a recently completed review
by Knight and Sayegh (2009) that reevaluated the sociocultural stress and coping model
in light of more recent research that had applied this model since its inception
recommended that the idea that cultural values operate primarily through caregivers’
appraisals of burden should be abandoned. Rather, it appears as though in most instances,

71

cultural values such as familism operate through coping style and social support when
they impact mental or physical health outcomes at all. In support of the sociocultural
stress and coping model and consistent with the hypotheses, the effects of the cultural
values that did impact health outcomes were mediated by coping style; specifically,
avoidant coping resulted in negative mental and self-reported general physical health
outcomes among caregivers. In addition, burden had both direct and indirect effects on
health, which is consistent with the original model.
Secondly, active coping had no effect whatsoever on mental or physical health
outcomes for caregivers in the final path models in this study. This finding is in
disagreement with the sociocultural stress and coping model (Aranda & Knight, 1997;
Knight et al., 2000), which posits that active coping is protective against poor mental and
physical health outcomes among caregivers. This finding may simply be an issue of
statistical power due to this study’s relatively small sample size. Interestingly, the study
by Kim et al. (2007) that used the same caregiver sample reported that active coping had
a protective effect against depression only when not controlling for education and against
diastolic blood pressure both with and without controlling for education. It is important to
note that the Kim et al. study examined familism as a unitary construct rather than its
more specific dimensions such as Familial Obligations and Perceived Support from the
Family. However, it is also possible that active coping may not necessarily result in
positive health outcomes for caregivers yet may simply buffer them from poorer
outcomes.

72

Moreover, Folkman and Lazarus (1988) argued that coping effectiveness depends
largely on whether the coping style that is chosen is in line with the possibilities for
coping in a given situation, specifically with regard to the degree to which a desired
outcome is something the individual can control. Problem-focused coping styles aimed at
achieving the desired outcome are well-suited if the outcome is in an individual’s control,
whereas emotion-focused coping styles are considered unsuitable as they may hamper
problem-focused coping. In contrast, for situations in which the desired outcome is
beyond an individual’s control, they argued that problem-focused coping styles are
unsuitable, whereas emotion-focused coping styles that have distress-reducing effects are
well-suited. The likelihood that many family caregivers for dementia patients may view
both the cognitive and functional decline of their family members as well as the
caregiving process as beyond their control provides support for the finding that active
coping did not have the anticipated protective effect on their health outcomes.
Interestingly, these postulations conflict with the findings of numerous studies (e.g.,
Haley et al., 1987; Kim et al., 2007; Vitaliano et al., 1985) that have found active coping
styles to be beneficial for caregivers’ outcomes in contrast to avoidant coping styles.
Clearly, further research is required to examine the role of active coping in affecting
mental and physical health outcomes among family caregivers in the stress and coping
process that takes into account perceptions of control of the situation, for example.
A third unanticipated finding was the lack of effect of cultural values on CV
health indicators via burden and coping style. This finding is in disagreement with the
hypotheses as well as Kim et al.’s (2007) findings of a modest but statistically significant

73

protective effect of active coping on diastolic blood pressure in their path models based
on the sociocultural stress and coping model. One reason for this particular discrepancy is
that the cultural values examined in each study differed, with the Kim et al. study
examining familism as a unitary construct (as measured by the Bardis (1959) scale) and
the current study examining two specific factors of familism (as measured by the Sabogal
et al. (1989) scale) and the Cultural Justification factor. With regard to the null effects of
cultural values on caregivers’ CV outcomes in this study, it is possible that the statistical
power may have been too low or that the powerful effects of age and ethnicity on CV
measures overwhelmed any potential effects attributable to these other variables.
However, it is also plausible that the caregivers who participated in this study had better
objective physical health (including CV health) than those who chose not to participate,
as they may be more physically able to participate. Moreover, Knight et al. (2007)
suggested that it is possible that caregivers are selected based on physical (e.g., CV)
health, as they are more able to fulfill caregiving duties if they are in better physical
health. The finding that caregivers’ self-reported general physical health was generally
poorer when examining certain cultural values, despite the lack of significant findings
regarding CV health indicators, suggests that their health may in fact be detrimentally
affected by caregiving in ways that are not captured by these CV measures.
Other findings from this study warrant comparison to the findings of other studies
that have examined similar cultural values in the context of caregiving. For example, Kim
et al. (2007) found that higher levels of education were inversely related to familism in
their path models. In the current study, education was not associated with either of the

74

factors of familism or the Cultural Justification factor. It is possible that this difference is
attributable to the use of different familism measures. Kim and colleagues utilized the
Bardis (1959) familism scale, which, as discussed earlier, may not be the most suitable
measure of familism, as it has been implemented haphazardly to characterize both
nuclear and extended familial relationships. Furthermore, caregivers with higher levels of
education may have had lower scores on the Bardis familism scale, because higher levels
of education may increase individualism and thereby weaken ties with nuclear and/or
extended family members. The current study, which used the Sabogal et al. (1987)
familism scale to derive the familism factors, may be a more appropriate and useful
measure of familism, as it taps more specific factors of familism involving both
attitudinal and behavioral elements of this cultural value.
Although specific ethnic differences were not hypothesized in this study, certain
incidental findings with regard to ethnic differences are worthy of closer attention and
can be compared to the findings of similar studies. For example, African American
ethnicity was associated with Familial Obligations in this study’s models, a finding that
somewhat conflicts with Kim et al.’s (2007) finding that ethnicity was not associated with
familism in their models. African American caregivers tend to have higher levels of
familism, so it is not completely unanticipated that they would score higher on this
specific factor of familism as well, of which familism is largely composed. What is
somewhat perplexing is why African American caregivers in this study would be more
likely to perceive caregiving more as an obligation rather than associate caregiving with
feelings of familial support and solidarity. One possible explanation for this finding is

75

that African American caregivers in this study had significantly lower mean income
levels than the White caregivers, χ
2
(5, N = 160) = 13.43, p < .05, which may have
prevented them from obtaining formal care for their family members due to financial
constraints. Thus, providing care may have been perceived as an obligation rather than a
reflection of family solidarity if these caregivers in fact would prefer to obtain formal
care services for their care recipients but are unable to do so. It is important to note that
despite this ethnic difference with regard to Familial Obligations in the path models in
this study, Familial Obligations had similar influences on the rest of these models in that
it operated through avoidant coping to detrimentally affect mental and self-reported
general physical health outcomes.
In addition, this study found no association of African American ethnicity with
Cultural Justification. In contrast, Dilworth-Anderson et al. (2005) reported that African
Americans in their caregiving sample had significantly higher CJCS scores than their
White counterparts, which suggested that African American caregivers more firmly held
beliefs regarding cultural reason for providing care to their family members than did
White caregivers. While the Dilworth-Anderson and colleagues’ CJCS measure differs to
a certain degree from the Cultural Justification factor used in the current study, it is
interesting to note that the African American caregivers in the current study did not score
significantly higher than the White caregivers on the Cultural Justification factor, which
is composed largely of the CJCS items. One possible explanation is that in this study,
African American caregivers had significantly lower mean levels of education, χ
2
(5, N =
160) = 14.11, p < .01, compared to White caregivers. In contrast, the mean levels of

76

education in Dilworth-Anderson et al.’s study did not significantly differ across the
African American and White caregivers, with the majority of caregivers from both ethnic
groups having more than 12 years of education. Dilworth-Anderson and colleagues
suggested that higher levels of education may result in lower levels on the CJCS scale,
which they reported in their study, for a number of reasons. For example, education could
potentially modify the manner in which caregivers perceive what their roles and duties
are in the caregiving process. In addition, African American caregivers with higher levels
of education may hold more hegemonic beliefs that weaken their identification with their
cultural group. In the current study, although the White caregivers had higher mean levels
of education than the African American caregivers, they did not score lower on the
Cultural Justification factor. It is possible that the effect of higher levels of education on
reducing levels on the CJCS may more significantly affect African American caregivers
rather than White caregivers, as suggested by Dilworth-Anderson et al. However, higher
levels of education were not significantly inversely related to Cultural Justification in this
study, as suggested by Dilworth-Anderson et al. Remaining possible explanations for
these differences regarding CJCS scores between these two studies include both sampling
procedure and regional variations across the two samples (probability sampling in the Los
Angeles, California area in the current study that excluded White caregivers from higher
SES groups; convenience sampling in the Chapel Hill, North Carolina in the Dilworth-
Anderson et al. study).
This study has certain limitations that suggest that caution should be used in the
interpretation of these results. First, this study employed a cross-sectional design and

77

involved a relatively small sample size, which limits both the power of the models to
assess smaller effects and discern changes in small effects. Second, these models are not
causal; instead, they should be regarded as consistent with the observed multivariate
covariance structure of this caregiver sample’s data. For both of these reasons, replication
with a larger and more diverse sample is needed to clarify the most truly likely paths in
these models. In addition, a longitudinal analysis of the influence of these cultural values
on caregivers’ health outcomes would allow for a better understanding of the relation
between the variables examined in this study. However, this study does have the benefit
of having used probability sampling methods that resulted in a population-based sample
recruited from tracts with roughly similar SES backgrounds. This sampling method not
only has the advantage over convenience sampling of being able to screen for SES
differences, but it also allows for the ability to report rejection rates and thus determine
levels of potential sampling bias from the population of interest.
In conclusion, the results of this study make several important contributions to the
cultural values and caregiving literature. First, none of the three cultural values examined
in this study was associated with caregivers’ burden, contrary to hypothesis, suggesting
that cultural values should not be viewed as operating through burden to affect health
outcomes in the sociocultural stress and coping model. This finding bears theoretical and
methodological importance with regard to the examination of cultural values using the
sociocultural stress and coping model. Secondly, this study examined some of the
specific underlying dimensions that compose the cultural value of familism, rather than
simply examining familism as a unitary construct, which resulted in a richer

78

understanding of how familism operates to affect caregivers’ mental and physical health
outcomes in the sociocultural stress and coping model. As a result, it is evident that
Familial Obligations accounts for a substantial proportion of the negative mental and self-
reported general health outcomes that caregivers experience as an indirect result of
familism. Similar findings were found with the Cultural Justification factor, which may
be attributable to providing care out of a sense of obligation or duty based on firm
cultural values and beliefs. In contrast, Perceived Support from the Family had no
significant effect on caregivers’ health outcomes in these models. It appears as though
cultural values that emphasize feelings of obligation rather than familial solidarity or
support may be harmful for caregivers. Therefore, a broader finding that can be derived
from these results is that it should not simply be assumed that cultural values invariably
have positive effects on the caregiving stress and coping process or health outcomes.
Certain cultural values may have negative or null effects for certain caregivers. Future
research should focus on examining the effects of these more specific cultural values on
caregivers’ health outcomes across diverse ethnic groups.
Finally, these findings also bear significance in terms of clinical implications, in
that cultural values that negatively impact caregivers’ health should be more closely
studied in search for therapeutic tools that could assist caregivers in highlighting the
positive effects of cultural values and decreasing the impact of their negative effects on
caregivers’ health. Caregivers should be evaluated to assess whether they hold more rigid
cultural values that are associated with detrimental mental and physical health outcomes.
Those who do should be targeted for interventions aimed at modifying these rigid cultural

79

values in the context of caregiving, perhaps using cognitive restructuring. If caregivers’
cultural values prove to be difficult to modify, given their deep-seated nature, then efforts
to modify avoidant coping style may prove useful in ameliorating the negative health
effects associated with caregiving.

80

Bibliography
Abramson, J. H., Ritter, M., Gofin, J., & Kark, J. D. (1992). A simplified index of
physical health for use in epidemiological studies. Journal of Clinical Epidemiology, 45,
651-658.

Aldrich, B. W., Lipman, B. W., & Goldman, F. P. (1973). Kinship obligation in a
developing nation: An inquiry into the socioeconomic and communal sources of familism
in Portugal. International Journal of Sociology of the Family, 3, 145-161.

Alzheimer’s Association (2008). 2008 Alzheimer’s disease facts and figures. Alzheimer’s
and Dementia, 4, 110-133.

Aranda, M. P., & Knight, B. G. (1997). The influences of ethnicity and culture on the
caregiver stress and coping process: A sociocultural review and analysis. The
Gerontologist, 37, 342-354.

Bardis, P. D. (1959). A familism scale. Marriage and Family Living, 21, 340-341.
Barusch, A. S., & Spaid, W. M. (1989). Gender differences in caregiving: Why do wives
report greater burden? The Gerontologist, 29, 667-676.

Baumgarten, M., Battista, R. N., Infante-Rivard, C., Hanley, J. A., Becker, R., &
Gauthier, S. (1992). The psychological and physical health of family members caring for
an elderly person with dementia. Journal of Clinical Epidemiology, 45, 61-70.

Bentler, P. M. (1992). On the fit of models of covariance and methodology to the
Bulletin. Psychological Bulletin, 112, 400-404.

Benyamini, Y., Blumstein, T., Lusky, A., & Modan, B. (2003). Gender differences in the
self-rated health-mortality association: Is it poor self-rated health that predicts mortality
or excellent self-rated health that predicts survival? The Gerontologist, 43, 396-405.

Blair, M. J. (1972). An evaluation of the Bardis familism scale. Journal of Marriage and
the Family, 34, 265-268.

Blascovich, J., & Katkin, E. (1993). Psychological stress testing for coronary heart
disease. In J. Blascovich & E. S. Katkin (Eds.), Cardiovascular reactivity to
psychological stress and disease (pp. 27-48). Washington, DC: American Psychological
Association.

Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit in the
analysis of covariance structure. Psychological Bulletin, 88, 588-606.

81

Cannuscio, C. C., Jones, C., Kawachi, I., Colditz, G. A., Berkman, L., & Rimm, E.
(2002). Reverberations of family illness: A longitudinal assessment of informal
caregiving and mental health status in the Nurses’ Health Study. American Journal of
Public Health, 92, 1305-1311.

Carver, C. S. (1997). You want to measure coping but your protocol’s too long: Consider
the Brief COPE. International Journal of Behavioral Medicine, 4, 92-100.

Chun, M., Knight, B. G., & Youn, G. (2007). Differences in stress and coping models of
emotional distress among Korean, Korean-American and White-American caregivers.
Aging & Mental Health, 11, 20-29.

Coon, D. W., Ory, M. G., & Schulz, R. (2003). Family caregivers: Enduring and
emergent themes. In D. W. Coon, D. Gallagher-Thompson, & L. W. Thompson (Eds.),
Innovative interventions to reduce dementia caregiver distress (pp. 3-27). New York:
Springer.

Coon, D. W., Rubert, M., Solano, N., Mausbach, B., Kraemer, H., Argüelles, T., et al.
(2004). Well-being, appraisal, and coping in Latina and Caucasian female dementia
caregivers: Findings from the REACH study. Aging & Mental Health, 8, 330-345.

Davies, M. (1996). The contribution of thrombosis to the clinical expression of coronary
atherosclerosis. Thrombosis Research, 83, 1-32.

Derogatis, L. R., & Spencer, P. M. (1985). The Brief Symptoms Inventory. Baltimore,
MD: Johns Hopkins University.

Dilworth-Anderson, P. (1995). Cultural Justifications for Caregiving Scale. Unpublished
manuscript.

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,
237-272.

Dilworth-Anderson, P., Goodwin, P. Y., & Williams, S. W. (2004). Can culture help
explain the physical health effects of caregiving over time among African American
caregivers? Journal of Gerontology: Social Sciences, 59, S138-S145.

Dilworth-Anderson, P., Brummett, B. H., Goodwin, P., Williams, S. W., Williams, R. B.,
& Siegler, I. C. (2005). Effects of race on cultural justification for caregiving. Journal of
Gerontology: Social Sciences, 60B, S257-S262.

Folkman, S., & Lazarus, R. S. (1988). The relationship between coping and emotion:
Implications for theory and research. Social Science Medicine, 26, 309-317.

82

Fredrikson, M., & Matthews, K. A. (1990). Cardiovascular responses to behavioral stress
and hypertension: A meta-analytic review. Annals of Behavioral Medicine, 12, 30-39.

Garrity, J. (1997). Stress, learning style, resilience factors, and ways of coping in
Alzheimer family caregivers. American Journal of Alzheimer’s Disease, 12, 171-178.

George, L. K. (2001). The social psychology of health. In R. H. Binstock & L. K. George
(Eds.), Handbook of aging and the social sciences (pp. 217-237). San Diego, CA:
Academic Press.

Grafstrom, M., Fratiglioni, L., Sandman, P. O., & Winblad, B. (1992). Health and social
consequences for relatives of demented and nondemented elderly: A population based
study. Journal of Clinical Epidemiology, 45, 861-870.

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, 323-330.

Haley, W. E., Roth, D. L., Coleton, M. I., Ford, G. R., West, C. A., Collins, R. P., et al.
(1996). Appraisal, coping, and social support as mediators of well-being in Black and
White family caregivers of patients with Alzheimer’s disease. Journal of Consulting and
Clinical Psychology, 64, 121-129.

Hebert, L. E., Beckett, L. A., Scherr, P. A., & Evans, D. A. (2001). Annual incidence of
Alzheimer disease in the United States projected to the years 2000 through 2050.
Alzheimer Disease and Associated Disorders, 15, 169-173.

Hebert, L. E, Scherr, P. A., Bienias, J. L., Bennett, D. A., & Evans, D. A. (2003).
Alzheimer Disease in the U.S. population: Prevalence estimates using the 2000 Census.
Archives of Neurology, 60, 1119-1122.

Heller, P. (1970). Familism scale: A measure of family solidarity. Journal of Marriage
and the Family, 32, 73-80.

Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure
analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A
Multidisciplinary Journal, 6, 1-55.

Janevic, M. R., & Connell, C. M. (2001). Racial, ethnic, and cultural differences in the
dementia caregiving experience: Recent findings. The Gerontologist, 41, 334-347.

Jean-Van Hell, R. (2001). Hispanic college students’ adjustment: The influence of
familism, acculturation and social support. Dissertation Abstracts International, 61,
6164. (UMI No. 9995926)

83

John, R., Hennessy, C. H., Dyeson, T. B., & Garret, M. D. (2001). Toward the
conceptualization and measurement of caregiver burden among Pueblo Indian family
caregivers. The Gerontologist, 41, 210-219.

Jolliffe, I. T. (2002). Principal component analysis (2
nd
ed.). New York: Springer.
Kassees, A. S. (1972). Cross-cultural comparative familism of a Christian Arab people.
Journal of Marriage and the Family, 38, 423-429.

Kiecolt-Glaser, J. K., Dura, J. R., Speicher, C. E., Trask, O. J., & Glaser, R. (1991).
Spousal caregivers of dementia victims: Longitudinal changes in immunity and health.
Psychosomatic Medicine, 53, 345-362.

Kim, J., Knight, B. G., & Flynn Longmire, C. V. (2007). The role of familism values in
stress and coping processes among African American and White dementia caregivers:
Effects on mental and physical health. Health Psychology, 26, 564-576.

Knight, B., Silverstein, M., McCallum, T. J., & Fox, L. (2000). A sociocultural stress and
coping model for mental health outcomes among African American caregivers in
Southern California. Journal of Gerontology: Psychological Sciences, 55B, P142-P150.

Knight, B., Robinson, G. S., Flynn Longmire, C. V., Chun, M., Nakao, K., & Kim, J.
(2002). Cross cultural issues in caregiving for dementia: Do familism values reduce
burden and distress? Ageing International, 27, 70-93.

Knight, B. G., Flynn Longmire, C. V., Dave, J., Kim, J. H., & David, S. (2007). Mental
health and physical health of family caregivers for persons with dementia: A comparison
of African American and white caregivers. Aging & Mental Health, 11, 538-546.

Knight, B. G., & Sayegh, P. (2009). Cultural values and caregiving: The sociocultural
stress and coping model revisited. Manuscript submitted for publication (copy on file
with author).

Lawton, M. P., Kleban, M. H., Moss, M., Rovine, M., & Glicksman, A. (1989).
Measuring caregiving appraisal. Journal of Gerontology: Social Sciences, 44, P61-P71.

Lawton, M. P., Rajagopal, D., Brody, E., & Kleban, M. H. (1992). The dynamics of
caregiving for a demented elder among Black and White families. Journal of
Gerontology: Social Sciences, 47, S156-S164.

Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York:
Springer.

Lazarus, R. (1991). Emotion and adaptation. New York: Oxford University Press.

84

Lee, S., Colditz, G. A., Berkman, L. F., & Kawachi, I. (2003). Caregiving and risk of
coronary heart disease in U.S. women: A prospective study. American Journal of
Preventive Medicine, 24, 113-119.

Levkoff, S. E., Levy, B. R., & Weitzman, P. E. (1999). The role of religion and ethnicity
in the help seeking of family caregivers of elders with Alzheimer's disease and related
disorders. Journal of Cross-Cultural Gerontology, 14, 335-356.

Losada, A., Shurgot, G. R., Knight, B. G., Marquez, M., Montorio, I., Izal, M., et al.
(2006). Cross-cultural study comparing the association of familism with burden and
depressive symptoms in two samples of Hispanic dementia caregivers. Aging & Mental
Health, 10, 69-76.

Losada, A., Márquez-González, M., Knight, B. G., Yanguas, J., Sayegh, P., & Romero-
Moreno, R. (in press). Psychosocial factors and caregivers’ distress: Effects of familism
and dysfunctional thoughts. Aging & Mental Health.

Mausbach, B. T., von Känel, R., Aschbacher, K., Roepke, S., Dimsdale, J. E., Ziegler, M.
G., et al. (2007). Spousal caregivers of patients with Alzheimer's disease show
longitudinal increases in plasma level of tissue-type plasminogen activator antigen.
Psychosomatic Medicine, 69, 816-822.

McCallum, T. J., Flynn Longmire, C. V., & Knight, B. G. (2007). African American and
White female caregivers and the Sociocultural Stress and Coping Model. Clinical
Gerontologist, 30, 25-37.

McClendon Baumann, V. A. (2007). Measurement of cultural values regarding dementia
caregiving. Dissertation Abstracts International, 69, 336. (UMI No. 3291828)

McHorney, C., Ware, J. E., Lu, R., & Sherbourne, C. D. (1994). The MOS 36-Item
Short-Form Health Survey (SF-36): III. Tests of data quality, scaling assumptions, and
reliability across diverse patient groups. Medical Care, 32, 40-66.

Monat, A., & Lazarus, R. (1991). Stress and coping: An anthology (3rd ed.). New York:
Columbia University Press.

Nkongho, N. O. & Archbold, P. G. (1995). Reasons for caregiving in African American
families. Journal of Cultural Diversity, 2, 116-123.

Ory, M. G., Hoffman, R. R. III, Yee, J. L., Tennstedt, S., & Schulz, R. (1999). Prevalence
and impact of caregiving: A detailed comparison between dementia and nondementia
caregivers. The Gerontologist, 39, 177-185.

85

Papastavrou, E., Kalokerinou, A., Papacostas, S. S., Tsangari, H., & Sourtzi, P. (2007)
Caring for a relative with dementia: Family caregiver burden. Journal of Advanced
Nursing, 58, 446-457.

Penley, J., Tomaka, J., & Wiebe, J. (2002). The association of coping to physical and
psychological health outcomes: A meta-analytic review. Journal of Behavioral Medicine,
25, 551-603.

Pinquart, M., & Sörensen, S. (2007). Correlates of physical health of informal caregivers:
A meta-analysis. Journal of Gerontology: Psychological Sciences, 62B, P126-P137.

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, P205-P211.

Radloff, L. S. (1977). The CES-D Scale: A self-report depression scale for research in the
general population. Applied Psychological Measurement, 1, 385-401.

Rao, V. V. P., & Rao, V. N. (1979). An evaluation of the Bardis familism scale in India.
Journal of Marriage and the Family, 41, 417-421.

Rogers, E. M., & Sebald, H. (1962). A distinction between familism, family integration,
and kinship orientation. Marriage and Family Living, 24, 25-30.

Sabogal, F., Marin, G., Otero-Sabogal, R., Marin, B. V., & Perez-Stable, E. J. (1987).
Hispanic familism and acculturation: What changes and what doesn’t? Hispanic Journal
of Behavioral Sciences, 9, 397-412.

Schulz, R., O’Brien, A. T., Bookwala, J., & Fleissner, K. (1995). Psychiatric and physical
morbidity effects of dementia caregiving: Prevalence, correlates, and causes. The
Gerontologist, 35, 771-791.

Schwartz, S. J. (2007). The applicability of familism to diverse ethnic groups: A
preliminary study. Journal of Social Psychology, 147, 101-118.

Silverstone, B. (1978). An overview of research on informal social supports: Implications
for policy and practice. Aged Care and Services Review, 1, 23-35.

Soskolne, V., Halevy-Levin, S., & Ben-Yehuda, A. (2007). The context of caregiving,
kinship tie and health: A comparative study of caregivers and non-caregivers. Women &
Health, 45, 75-94.

86

Teri, L., Truax, P., Logsdon, R., Uomoto, J., Zarit, S., & Vitaliano, P. P. (1992).
Assessment of behavioral problems in dementia: The revised memory and behavior
problems checklist. Psychology and Aging, 7, 622-629.

Triandis, H., Marin, G., Betancourt, H., Lisansky, J., & Chang, B. (1982). Dimensions of
familism among Hispanic and mainstream Navy recruits (Tech. Rep. No. 14). University
of Illinois, Champaign. Department of Psychology.

U.S. Bureau of the Census. (2000). Projections of the total resident population by 5-year
age groups, race, and Hispanic origin with special age categories: Middle series 1999-
2000; middle series 2050-2070. Retrieved March 13, 2008, from
http://www.census.gov/population/projections/nation/summary/np-t4-a.txt.

U.S. Department of Health and Human Services. (1991). Health status of minorities and
low-income groups: Third edition (DHHS Pub. No. 271-848/40085). Washington, DC:
U.S. Government Printing Office.

Vitaliano, P. P., Russo, J., Carr, J. E., Maiuro, R. D., & Becker, J. (1985). The ways of
coping checklist: A test of the hypothesis of fit. Journal of Personality and Social
Psychology, 59, 582-592.

Vitaliano, P. P., Russo, J., & Niaura, R. (1995). Plasma lipids and their relationships with
psychosocial factors in older adults. Journal of Gerontology, 50, 18-24.

Vitaliano, P. P., Zhang, J., & Scanlan, J. M. (2003). Is caregiving hazardous to one’s
physical health? A meta-analysis. Psychological Bulletin, 129, 946-972.

Wallace, D. C., Witucki, J. M., Boland, C. S., & Tuck, I. (1998). Cultural context of
caregiving with elders. Journal of Multicultural Nursing and Health, 4, 42-48.

Wright, S. D., Lund, D. A, Caserta, M. S., & Pratt, C. (1991). Coping and caregiver well-
being: The impact of maladaptive strategies. Journal of Gerontological Social Work, 17,
75-91.

Youn, G., Knight, B. G., Jeong, H., & Benton, D. (1999). Differences in familism values
and caregiving outcomes among Korean, Korean American, and White American
caregivers. Psychology and Aging, 14, 355-364.

Zarit, S. H., Reeves, K. E., & Bach-Peterson, J. (1980). Relatives of the impaired elderly:
Correlates of feelings of burden. The Gerontologist, 20, 649-655.

Zarit, S. H., Johansson, L., & Jarrot, S. E. (1998). In I. H. Nordhus, G. R.

87

VandenBos, S. Berg, & P. Fromholt (Eds.), Clinical gerontology (pp. 345-360).
Washington, DC: American Psychological Association.

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Measurement of cultural values regarding dementia caregiving

PDF

Evaluating a cultural process model of depression and suicidality in Latino adolescents

PDF

Mental health service use by children and youth in the child welfare system: a focus on need and predisposing factors and caregiver type

PDF

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

PDF

The effects of health-related spousal influence on couples coping with chronic heart failure: an application of the actor-partner interdependence model

PDF

Communication and home-based interventions in Alzheimer’s disease -- a review

PDF

Mechanisms of the effect of involuntary retirement on older adults' health and mental health

PDF

Family caregiver well-being: generational differences and coping strategies

PDF

Age differences in the effects of anxiety on selective attention: a test of the vigilance-avoidance hypothesis

PDF

Effectiveness of a heart plus brain health-focused nutrition intervention among Latinas: a randomized controlled trial of Buenos Hábitos Alimenticios Para Una Buena Salud (Good Eating Habits for ...

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 [capstone paper]

PDF

Influence of age and anxiety on recognition of facial expressions of emotion: exploring the role of attentional processes

PDF

Examining the effect of fertility on lifespan using data from the Swedish Twin Registry

PDF

Crosscultural differences in dementia diagnosis and care-seeking in Hispanic and non-Hispanic white outpatients

PDF

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

PDF

Health-related quality of life correlates of the treatment decision making process of newly diagnosed prostate cancer patients

PDF

From “soul calling” to calling a therapist: meeting the mental health needs of Hmong youth through the integration of spiritual healing, culturally responsive practice and technology

PDF

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

PDF

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

Asset Metadata

Creator Sayegh, Philip S. (author)

Core Title The effects of familism and cultural justification on the mental and physical health of family caregivers

School College of Letters, Arts and Sciences

Degree Master of Arts

Degree Program Psychology

Degree Conferral Date 2009-12

Publication Date 10/13/2009

Defense Date 05/11/2009

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

Tag caregiver burden,caregiving,coping,cultural values,culture,Ethnicity,familism,OAI-PMH Harvest,Stress

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Knight, Bob G. (committee chair), Gatz, Margaret (committee member), Mather, Mara (committee member)

Creator Email psayegh@gmail.com,psayegh@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2663

Unique identifier UC1501988

Identifier etd-Sayegh-2938 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-268521 (legacy record id),usctheses-m2663 (legacy record id)

Legacy Identifier etd-Sayegh-2938.pdf

Dmrecord 268521

Document Type Thesis

Rights Sayegh, Philip S.

Type texts

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

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu