Differences between caregivers and noncaregivers in psychological health and physical health: A meta-analysis

Abstract

Providing care for a frail older adult has been described as a stressful experience that may erode psychological well-being and physical health of caregivers. In this meta-analysis, the authors integrated findings from 84 articles on differences between caregivers and noncaregivers in perceived stress, depression, general subjective well-being, physical health, and self-efficacy. The largest differences were found with regard to depression (g = .58), stress (g = .55), self-efficacy (g = .54), and general subjective well-being (g = -.40). Differences in the levels of physical health in favor of noncaregivers were statistically significant, but small (g = .18). However, larger differences were found between dementia caregivers and noncaregivers than between heterogeneous samples of caregivers and noncaregivers. Differences were also influenced by the quality of the study, relationship of caregiver to the care recipient, gender, and mean age of caregivers.

Full text

Differences Between Caregivers and Noncaregivers in Psychological
Health and Physical Health: A Meta-Analysis
Martin Pinquart
Friedrich Schiller University
Silvia So¨rensen
University of Rochester
Providing care for a frail older adult has been described as a stressful experience that may erode
psychological well-being and physical health of caregivers. In this meta-analysis, the authors integrated
findings from 84 articles on differences between caregivers and noncaregivers in perceived stress,
depression, general subjective well-being, physical health, and self-efficacy. The largest differences were
found with regard to depression (g ⫽ .58), stress (g ⫽ .55), self-efficacy (g ⫽ .54), and general subjective
well-being (g ⫽⫺.40). Differences in the levels of physical health in favor of noncaregivers were
statistically significant, but small (g ⫽ .18). However, larger differences were found between dementia
caregivers and noncaregivers than between heterogeneous samples of caregivers and noncaregivers.
Differences were also influenced by the quality of the study, relationship of caregiver to the care
recipient, gender, and mean age of caregivers.
In the United States, care services to the elderly are provided
primarily by informal helpers, such as spouses and adult children
(Stone, Cafferata, & Sangl, 1987). Providing care to an elderly
relative often restricts the personal life, social life, and employ-
ment of the caregiver. For example, caregivers may have less time
to spend with friends, to fulfill other family obligations, or to
pursue leisure pursuits (e.g., Gilleard, Gilleard, Gledhill, & Whit-
tick, 1984; Kosberg & Cairl, 1986; Zarit, Reever, & Bach-
Petersen, 1980). Furthermore, caregivers are often faced with
difficult caregiving tasks (e.g., Steinmetz, 1988) and behavior
problems of demented care recipients, such as verbal and physical
aggression and confusion (Teri et al., 1992). Because the progres-
sion of the care receivers’ illnesses and care needs are difficult to
foresee, care receivers face increased uncertainty (e.g., Poulshok &
Deimling, 1984).
A substantial literature shows that providing care to an older
family member is associated with increased psychological distress
(e.g., Donaldson, Tarrier, & Burns, 1998; Schulz, O’Brien, Book-
wala, & Fleissner, 1995). For example, up to 48% of dementia
caregivers have been identified as being at risk for psychiatric
symptomatology (Brodarty & Hadzi-Pavlovic, 1990; Draper, Pou-
los, Cole, Poulos, & Ehrlich, 1992). However, many caregivers are
able to cope quite well with their role. They experience few
symptoms of distress and often report positive gain from the
experience (e.g., Kramer, 1997; Schulz et al., 1997).
The inconsistencies in studies comparing caregivers and non-
caregivers make it difficult to draw clear conclusions about the
prevalence of poor psychological and physical health of caregiv-
ers. For example, whereas Vitaliano, Russo, Scanlan, and Greeno
(1996) and Bodnar and Kiecolt-Glaser (1994) reported higher
levels of depression in caregivers than noncaregivers, Haley et al.
(1995) and Loomis and Booth (1995) did not find significant
differences. Similar contradictions are found for physical health
(e.g., Rose-Rego, Strauss, & Smyth, 1998, vs. Barnes & Patrick,
2000; Haley et al., 1995).
There are both methodological and conceptual reasons for these
inconsistencies. First, there are sampling issues with caregiver
research. Many studies on caregiving are based on nonrepresenta-
tive samples that overrepresent distressed caregivers (Schulz et al.,
1995, 1997). Thus, caregivers might not experience significantly
higher levels of distress than the general population, if they were
sampled differently. On the other hand, Schulz et al. (1997) have
also suggested that more representative studies may underestimate
caregiver stress because of how they define caregivers: In some
studies a caregiver is defined as a person sharing the household
with an impaired family member without clarifying whether the
respondent provides care or not.
Second, many studies have large random sampling errors due to
small sample sizes. If the magnitude of population effect sizes is
low to medium, then the effects may not be detectable in these
studies (e.g., Rosenthal, 1991).
Third, caregivers often derive benefits from the caregiving ex-
perience, such as increased closeness to the family member being
cared for, and satisfaction at fulfilling one’s duty (for overview,
Kramer, 1997). Thus, caregivers may not be disproportionately
stressed compared to noncaregivers, unless the caregiving situation
is unusually stressful and there are no positive aspects to their
caregiving experience.
Fourth, caregivers to older adults with dementia may be con-
siderably more distressed than caregivers to other older adults
(e.g., Clipp & George, 1993). However, these groups are rarely
differentiated in studies of caregiving (e.g., Cattanach & Tebes,
Martin Pinquart, Department of Developmental Psychology, Friedrich
Schiller University, Jena, Germany; Silvia So¨rensen, University of Roch-
ester Medical Center.
Correspondence concerning this article should be addressed to Martin
Pinquart, Department of Developmental Psychology, Friedrich Schiller
University, Am Steiger 3 Haus 1, D-07743 Jena, Germany. E-mail:
Martin.Pinquart@rz.uni-jena.de
Psychology and Aging Copyright 2003 by the American Psychological Association, Inc.
2003, Vol. 18, No. 2, 250–267 0882-7974/03/$12.00 DOI: 10.1037/0882-7974.18.2.250
250

1991; Teel & Press, 1999; Wallsten, 2000), thus influencing ob-
served differences between caregivers and noncaregivers.
Finally, research on the protection of psychological well-being
and of a positive self-concept in adulthood has shown a consider-
able resilience of the self (Brandtsta¨dter & Greve, 1994). Thus,
many caregivers may cope well with their role without showing
negative effects on psychological or physical health (e.g., Garity,
1997).
Given the inconsistent findings in the literature and the potential
reasons why many caregivers may, in fact, not experience more
psychological distress than noncaregivers, a systematic integration
of these findings would be helpful at this time. Meta-analysis is an
ideal tool to do this.
In this article, we focus on three research questions. In our first
research question, we ask whether caregivers and noncaregivers
differ in psychological and physical health, specifically in their
levels of perceived stress, depression, general subjective well-
being, physical health, and self-efficacy. In our second research
question, we ask which aspects of psychological health and phys-
ical health evidence the largest differences between caregivers and
noncaregivers. In the third research question, we analyze whether
observed differences between caregivers and noncaregivers are
influenced by moderator variables, such as caring for demented
older adults versus caring for physically impaired older adults.
Hypotheses
Differences Between Caregivers and Noncaregivers
To compare caregivers to noncaregivers, we focus on variables
that can be meaningfully assessed in both groups, namely, on
perceived general symptoms of stress not specific to caregiving,
depression, general subjective well-being, physical health, and
self-efficacy, but not stress measures unique to caregivers, such as
caregiver burden, perceived economic costs, or social costs.
Difficult caregiving tasks (e.g., Steinmetz, 1988), restrictions in
personal life (e.g., Kosberg & Cairl, 1986), and illness-specific
problems of the care receiver (e.g., dementia care recipients’
behavior disturbances; Teri et al., 1992) may impair caregivers’
psychological and physical health. However, this negative effect
may be reduced by positive aspects of caregiving (Kramer, 1997),
psychological resilience, and effective coping (Garity, 1997). Nev-
ertheless, not all caregivers may have the necessary skills to cope
well with the caregiving demands (Lutzky & Knight, 1994), and
very severe stressors may exceed individual coping resources and
make it difficult to find positive aspects of providing support (e.g.,
Faßmann & Grillenberger, 1996; Ory, Yee, Tennstedt, & Schulz,
2000). On the basis of these considerations, we expected with
regard to our first research question that caregivers would have, on
average, significantly lower levels of psychological and physical
health than matched controls, but that the effect sizes would be
small to medium.
1
In the second research question, we investigated which variables
show the largest differences between caregivers and noncaregiv-
ers. We expected larger differences between caregivers and non-
caregivers for stress and self-efficacy than for depression, general
subjective well-being, and physical health for two reasons. First,
caregiver stress, difficult-to-manage situations, and loss of control
over one’s life have been described as direct negative effects of
caregiving (e.g., Donaldson et al., 1998; Kinney, Stephens, Franks,
& Norris, 1995; Wallhagen, 1992–1993). In contrast, influences of
caregiving on depression, general well-being, and health may be
more indirect and mediated through caregiver stress (e.g., Yates,
Tennstedt, & Chang, 2000). Second, depression, general subjective
well-being, and physical health are less situation-specific than
stress; they are influenced by many experiences beyond the care-
giver role, such as other social roles, socioeconomic status, the
quality of social relations, health promoting habits, personality,
and even genetic factors (e.g., Blazer, 1993; Hooker, Monahan,
Bowman, Frazier, & Shifren, 1998; Hooker, Monahan, Shifren, &
Hutchinson, 1992; Huck & Armer, 1996; Pinquart, 1998).
Moderators of Differences Between Caregivers and
Noncaregivers
In the third research question, we asked whether the size of
observed differences between caregivers and noncaregivers is
moderated by caregiver characteristics (the nature of the care
receiver’s illness, the relationship to the care receiver, caregiver’s
age and gender), and sample characteristics (e.g., representative-
ness of the sample). Other aspects of the caregiving situation, such
as the amount and duration of care provision, could not be included
in the meta-analysis because of insufficient data.
Caregiver characteristics. Caring for demented older adults
may be more stressful than caring for physically frail older adults,
because of five problems specific to dementia caregiving: (a)
dementia-related behavioral problems, disorientation, and shifts in
personality; (b) the increased need for supervision and the associ-
ated lack of spare time; (c) the isolation of the caregiver due to the
care receiver’s behavior problems; (d) the limited ability of care
receivers to express gratitude and the associated reduction in
uplifts of caregiving; and (e) the progressive deterioration of the
care receiver, which reduces or eliminates visible positive long-
term effects of caregivers’ engagement (e.g., Clipp & George,
1993; Ory et al., 2000). On the basis of these considerations, we
expected larger differences in psychological and physical health
between caregivers and noncaregivers for studies with dementia
caregivers than for studies of caregivers for nondemented older
adults and studies that include both caregivers for demented el-
derly and other caregivers.
Also, we expected larger differences between caregivers and
noncaregivers in psychological and physical health for spouses
than for other relatives (e.g., adult children), for women than for
men, and for older samples as compared with younger samples.
Spousal caregivers were expected to show higher levels of distress
than other caregivers because (a) they are more likely to have
age-associated illnesses and disabilities (Schneider, Murray, Ban-
erjee, & Mann, 1999); (b) they provide up to four times the amount
of care provided by nonspousal family caregivers (Tennstedt,
McKinlay, & Sullivan, 1988); and (c) they are more likely than
other caregivers to report a lack of alternative roles and social
activities outside the home, which might function as buffers
against caregiver stress (Barber & Pasley, 1994).
We expected women to evidence greater psychological distress
than men because women (a) provide more caregiving assistance
1
According to J. Cohen (1992), effect sizes of g ⫽ .20 are interpreted as
small, g ⫽ .50 as medium, and g ⫽ .80 and above as large.
251
DIFFERENCES BETWEEN CAREGIVERS

in general and personal care tasks in particular; (b) are more likely
to assume the primary caregiver role, whereas men often become
secondary caregivers; (c) are less likely to obtain informal and
formal assistance with caregiving; (d) are more likely to experi-
ence social pressure to become caregivers, whereas male caregiv-
ers are more likely to feel that they have chosen to assume the
caregiving role; (e) are more likely to stay in the caregiver role
even if it becomes very stressful; and (f) tend to have fewer coping
resources, such as internal locus of control (e.g., Miller & Cafasso,
1992; Yee & Schulz, 2000).
Psychological and physical health is more likely to be compro-
mised for caregivers than noncaregivers in older samples: Com-
pared with younger adults, older adults (a) have lower levels of
psychological, physical, and financial resources (e.g., Baltes &
Mayer, 1999); (b) have fewer stress-buffering roles and activities
due to age-associated losses of social roles (e.g., Barber & Pasley,
1994); and (c) may be less likely to use formal support (Hooker et
al., 1998), for example, because of a lack of knowledge about
available sources.
Quality of the study. We expected that greater differences in
psychological health and physical health between caregivers and
noncaregivers would be found in nonrepresentative samples than
in representative samples: Negative effects of caregiving may be
overestimated for samples of highly distressed caregivers, such as
people who seek help (Schulz et al., 1995, 1997). In addition,
population-based studies with representative samples may under-
estimate effects of caregiving because they do not directly assess
whether the respondent provides care, but rather they define care-
giving as the provision of a minimum level of support to an older
adult, or even as merely living with an impaired family member
(Schulz et al., 1997).
In addition, we expected the effect of caregiving to be larger in
studies that do not ensure demographic equivalence of caregivers
and controls: For example, women and individuals with lower
socioeconomic status (SES) are more likely to become caregivers
(Gage & Kinney, 1995; Haley, Levine, Brown, Berry, & Hughes,
1987; Yee & Schulz, 2000) and to report poor psychological and
physical health, irrespective of caregiving (Pinquart, 1998). Fur-
thermore, we expected larger differences between caregivers and
noncaregivers in published, peer-reviewed studies than in unpub-
lished studies or chapters because nonsignificant results are less
likely to be published (the file-drawer problem; Rosenthal, 1991).
In sum, we expected
1. that caregivers would report higher levels of stress and
depression, and lower levels of subjective well-being,
physical health, and self-efficacy than noncaregivers;
2. that differences between caregivers and noncaregivers
would be larger with regard to stress and self-efficacy
than in the other measures;
3. that larger differences between caregivers and noncare-
givers would emerge for studies with dementia caregivers
than with other caregivers, for spouses than for other
relatives, for women than for men, and for older samples
than for younger samples, for nonrepresentative samples
than for representative samples, for studies that do not
ensure demographic equivalence of caregivers and con-
trols, and for studies that were published in peer-
reviewed journals.
Method
Sample
Studies were identified from the developmental and gerontological lit-
erature through electronic databases {e.g., PsycINFO, Medline, Psyndex;
search terms: [(caregiving, caregiver, carer, or support provider) and non-
caregivers and (elderly or old age)]}, browsing through library shelves, and
looking up references found in other articles. Criteria for inclusion of
studies in the meta-analysis were as follows:
1. A sample of informal caregivers of older adults is compared to a
sample of noncaregivers with regard to perceived stress, depression, gen-
eral subjective well-being, physical health, or self-efficacy.
2
2. Differences between caregivers and noncaregivers can be converted to
standard deviation units.
3. The studies are written in English, French, German, or Russian.
Ninety-seven studies met these criteria, but about 15% of the total
number of publications surveyed had to be eliminated, for the most part
because they reported insufficient information about the magnitude of the
relationship (zero-order effect sizes) between variables. After exclusion of
such studies, we were able to include 84 studies in the meta-analysis. Most
of these were from English-language journals; only one German study and
one French-language study could be used. The majority of articles were
from the Gerontologist (11); others were from Psychology and Aging (7),
the Journal of Aging and Health (4), the Journal of Gerontology (4), the
Journal of Abnormal Psychology (4), Psychosomatic Medicine (4), Family
Relations (2), the Journal of the American Geriatrics Society (2), and other
journals and books (35). An additional 11 studies were taken from presen-
tations at conventions or dissertations. The studies were published or
presented between 1987 and the summer of 2002. Additional information
about the studies is provided in Figures 1 through 5 in the Appendix.
We entered the year of publication; the numbers of caregivers and
noncaregivers; the representativeness of the samples (representative ⫽ 1,
nonrepresentative ⫽ 0); the sociodemographic equivalence of the caregiv-
ers and noncaregivers (1 ⫽ equivalent/statistical control for nonequiva-
lence, 0 ⫽ equivalence not tested/no statistical control for differences
between caregivers and noncaregivers); the quality of the source (1 ⫽
peer-reviewed journal, 0 ⫽ others); the impairment of the care receiver
(3 ⫽ dementia, 2 ⫽ combined sample of demented and nondemented care
receivers, 1 ⫽ nondemented care receivers, e.g., cancer); the measurement
of stress, depression, subjective well-being, physical health, and self-
2
An alternative method might be to compute comparisons of health
measures from caregiver studies with health measures from age-matched
noncaregiving community studies. Although this procedure would enable
us to include more studies in the present meta-analysis, it would have
severe limitations: First, noncaregiving studies differ considerably in psy-
chological outcomes (e.g., average level of depression; Yee & Schulz,
2000). Thus, it would be difficult to decide which study might be the best
empirical base for a comparison, even after controlling for differences in
the distribution of age and gender of caregivers and noncaregivers. Second,
community samples may differ from caregiver samples in other character-
istics that are related to psychological outcomes, such as SES. We would
not be able to statistically control for such differences in a meta-analysis
without having access to the complete data files. Third, some respondents
from community samples may be caregivers thus biasing a comparison
with caregiver studies. Therefore, we decided to focus our analysis on
empirical studies that compared caregivers and noncaregivers and, in most
cases, controlled for differences in sociodemographic variables between
caregivers and noncaregivers.
252
PINQUART AND SO
¨
RENSEN

efficacy; and the size of differences between caregivers and noncaregivers
with regard to these variables.
Measures
Perceived stress was most often assessed with the Perceived Stress Scale
(S. Cohen, Kamarck, & Mermelstein, 1983; 6 studies), single-item indica-
tors (2 studies), and other scales (14 studies). Depression was most often
assessed with the Hamilton Depression Rating Scale (Hamilton, 1967; 15
studies), the Center for Epidemiological Studies—Depression Scale
(CES–D; Radloff, 1977; 13 studies), the Beck Depression Inventory (Beck
& Steer, 1987; 11 studies), clinical interviews (4 studies), and other scales
(13 studies). Caregivers’ subjective well-being was most often assessed
with life-satisfaction scales (e.g., Life Satisfaction Index; Neugarten, Havi-
ghurst, & Tobin, 1961; 11 studies), positive affect scales (e.g., Positive and
Negative Affect Schedule; Watson, Clark, & Tellegen, 1988; 9 studies),
single-item indicators on happiness (5 studies), and other scales (9 studies).
Health was assessed with single-item indicators on subjective health (18
studies), symptom checklists (17 studies), and items on the frequency of
medication use and the number of hospitalizations (5 studies). Self-efficacy
was assessed with self-efficacy scales (e.g., Pearlin & Schooler, 1978; 4
studies), single-item indicators (2 studies), and sum-scales on perceived
control over one’s life (3 studies).
Statistical Integration of the Findings
Two common groups of statistical procedures are used in meta-analysis:
fixed- and random-effect models (e.g., Hedges & Vevea, 1998). If a
common effect size is hypothesized for all studies, then fixed-effects
models are appropriate. If effect sizes vary systematically between studies
according to moderator variables (e.g., whether caregivers for demented
and nondemented older adults are assessed), then random-effect models are
more appropriate because fixed-effect models can lead to inflated Type I
errors. Thus, the random-effect model was used, primarily based on pro-
cedures outlined by Hedges and Vevea (1998) and Raudenbush (1994).
We computed effect sizes for each study as the difference in psycho-
logical/physical health between caregivers and noncaregivers divided by
the pooled standard deviation. Effect sizes were also derived from t values,
F values, and exact p values. In cases where the direction of differences
between caregivers and noncaregivers but no effect size was reported, we
used vote counts to estimate the effect size, as suggested by Bushman and
Wang (1996). This procedure enabled us to include two additional studies
in the present meta-analysis. If in one study effect sizes were reported for
more than one subsample (e.g., caregivers for cognitively impaired older
adults and for physically impaired adults), then separate effect sizes were
computed for these subsamples rather than computing an average effect
size for the whole sample. The effect-size estimates were adjusted for
biases due to overestimation of the population effect size (common for
small samples), based on Hedges (1981). Confidence intervals that in-
cluded 95% of the effects were computed for each effect size.
The homogeneity of effect sizes was tested by using the homogeneity
statistic Q, which is distributed approximately as chi-square with k – 1 dfs,
where k is the number of effect sizes. In the case of heterogeneous effect
sizes, we estimated the between-studies variance of the effects, based on
Hedges and Vevea (1998).
Studies were weighted by the reciprocal of the sum of the between-
studies variance component and the random sampling error of the study. If
more than one outcome was provided for a sample with regard to one group
of outcome measures (e.g., clinician-rated depression and self-rated de-
pression were assessed), then we weighted studies with more than one
outcome measure, as suggested by Rosenthal (1991).
The significance of the mean effect size was tested by dividing the
weighted mean effect size by the estimation of the standard deviation.
Differences between two conditions were interpreted as significant when
the 95% confidence intervals did not overlap.
To test the influence of several moderators simultaneously, we used
weighted ordinary least squares regression analyses, following the ap-
proach outlined by Raudenbush (1994).
Results
Most of the studies included in this analysis were exclusively
focused on caregivers of demented elderly (63.3%); other studies
investigated other forms of illness (e.g., cancer, stroke; 4.7%) and
a mix of caregivers of older adults with physical and cognitive
impairments (32%). Eighty-six percent of the studies tested for
demographic differences between caregivers and control-group
members: In 68% of all studies, there were no significant differ-
ences or the authors controlled for these differences. Seven studies
did not control for emerging between-groups differences in SES
(six studies reported lower SES in caregivers, and one study
reported higher SES in caregivers) and age (age of caregivers was
higher in six studies and lower in one study). Because of the
limited number of available studies for the present meta-analysis,
we decided not to exclude these studies from our analysis but
rather to statistically control for the impact of nonequivalence of
caregivers and controls.
The respondents (caregivers and controls) had a mean age
of 62.5 years (SD ⫽ 8.6 years). Almost 72% were women. Most of
the respondents were spouses (58.5%) and adult children (35.8%).
Seventy-eight percent had completed high school. About 14% of
the respondents were members of ethnic minorities.
Seventy-one percent of the caregivers shared the household with
the care receiver. About 47% of the caregivers were employed full
time or part time. The caregivers have been providing care on
average for about 55 months (SD ⫽ 22) and spent on average 42.9
hr per week (SD ⫽ 24.3) doing caregiving tasks. The care recip-
ients’ age was, on average, 75.6 years (SD ⫽ 4.4 years). About half
(49.5%) of the care receivers were women.
Effects of Caregiver Status
First we tested the effect of caregiver status on psychological
health and physical health. As hypothesized, caregivers had higher
levels of stress and depression as well as lower levels of subjective
well-being, physical health, and self-efficacy than noncaregivers
(Table 1). Caregiver status explained 0.8% (physical health)
to 7.8% (depression) of the variance of the dependent variables.
The size of differences between caregivers and noncaregivers did
not vary for self-rated depression versus clinician-rated depression,
or for subjective health versus objective health, as measured by
symptom checklists and medication use.
Which Outcomes Show the Largest Effects?
With regard to our second research question, we had expected
larger differences between caregivers and noncaregivers for stress
and self-efficacy than for depression, subjective well-being, and
physical health. The results confirm our expectations only with
regard to physical health. Differences between caregivers and
noncaregivers were significantly smaller for physical health than
for the other outcome variables (see Table 1). The strongest
negative effects of caregiving were observed for clinician-rated
depression.
253
DIFFERENCES BETWEEN CAREGIVERS

Moderators of Differences Between Caregivers and
Noncaregivers
In the third research question, we explored whether differences
between caregivers and noncaregivers would vary by caregiver
characteristics and study characteristics. We first compared studies
with dementia caregivers, a mix of caregivers of demented and
nondemented older adults, and studies that focused on caregivers
of physically frail elderly adults (e.g., cancer, Parkinson’s Dis-
ease). Unfortunately, only six studies were found for the latter
group. Differences in favor of noncaregivers were significantly
larger in studies with only dementia caregivers compared with
studies in which caregivers of demented and nondemented older
adults were combined for all outcomes except depression (Table
2). However, although they were smaller, the differences between
caregivers and noncaregivers were still significant in the combined
group. Caregivers of nondemented older adults reported higher
levels of stress and lower levels of subjective well-being and
physical health than noncaregivers, but there were no significant
differences in depression and self-efficacy.
We further analyzed whether the observed differences between
caregivers and noncaregivers would be larger in spouses than in
adult children. Because only a small number of studies were
available for homogeneous samples of spouses or adult children,
our results have to be interpreted with caution. Differences in
perceived stress and depression between caregivers and noncare-
givers were significantly larger in samples consisting of spouses
than in samples of adult children (Table 3). We also found larger
differences between caregivers and noncaregivers in nonrepresen-
tative samples than in representative samples with regard to all
variables except stress. However, outcome variables did not differ
for studies that used demographically equivalent samples of care-
givers and noncaregivers versus those that did not (see Table 3).
The moderators may not be statistically independent. For exam-
ple, 67% of caregivers for demented older adults were spouses,
compared with 43% of caregivers from other studies; and 93% of
studies on dementia caregiving used nonrepresentative samples,
whereas only 46% of comparisons of combined samples of de-
mentia caregivers and other caregivers with noncaregivers used
nonrepresentative samples. Thus, we also tested whether the mod-
erator effects described above would remain significant after sta-
tistically controlling for other moderators. Weighted ordinary least
squares regression analyses were computed that included the ill-
ness of the care recipient (1 ⫽ dementia, 0 ⫽ nondementia/mix of
demented and nondemented care receivers), the percentage of
spousal caregivers, the percentage of female caregivers, the mean
age of the caregivers, the representativeness of the sample (1 ⫽
yes, 0 ⫽ no), the demographic equivalence of caregivers and
controls (1 ⫽ yes, 0 ⫽ no), and the quality of the source of
publication (1 ⫽ peer reviewed journal, 0 ⫽ others) as independent
variables. The dependent variables were stress, depression, sub-
jective well-being, physical health, and self-efficacy. As shown in
Table 4, the four univariate effects for dementia caregiving were
replicated in the multivariate analysis. In addition, we found a
marginal effect indicating that the tendency of caregivers to be
more depressed than noncaregivers was more pronounced for
dementia caregivers than for mixed caregivers.
Consistent with the univariate analysis, we found that differ-
ences between caregivers and noncaregivers were more pro-
nounced the more spousal caregivers were in the sample. In
addition, we found significant effects of caregivers’ gender and
age: In samples with a higher percentage of female respondents,
caregivers were more impaired than noncaregivers with regard to
all outcome variables except for stress. Furthermore, caregivers
were more likely to report higher levels of depression and lower
levels of self-efficacy in older than in younger samples. However,
younger caregivers were more likely to report increased stress than
older caregivers.
The quality of the study influenced the size of observed differ-
ences between caregivers and noncaregivers: Smaller differences
in four out of five outcome variables were revealed in represen-
Table 1
Differences in Psychological Health and Physical Health Between Caregivers and Noncaregivers
Variable kN
cg
N
ncg
g
CI
Q
W
Significance
of the mean
difference
Lower
limit
Upper
limit
Stress 30 2,696 4,578 0.55 0.41 0.69 182.68*** 7.58***
Depression 81 6,237 14,383 0.58 0.46 0.70 482.59*** 9.16***
Self-rated 57 5,463 11,856 0.51 0.40 0.62 387.96*** 9.03***
Clinician rated 23 1,176 2,507 0.66 0.49 0.84 84.81*** 7.55***
Low SWB 48 4,263 17,063 0.40 0.32 0.48 177.67*** 9.14***
Poor health 66 6,716 24,597 0.18 0.13 0.23 133.78*** 6.93***
Subjective 25 2,981 12,157 0.20 0.12 0.28 65.99*** 4.43***
Objective 39 3,655 12,405 0.15 0.09 0.22 66.48** 4.79***
Low self-efficacy 14 898 1,092 0.54 0.31 0.77 58.88*** 4.68***
Note. To make the table easier to read, we coded all dependent variables in a way that positive coefficients
indicate larger impairments of caregivers compared with noncaregivers. k ⫽ number of effect sizes; N
cg
⫽
number of caregivers; N
ncg
⫽ number of noncaregivers; g ⫽ mean difference (measured in standard deviation
units): Positive values indicate higher levels of psychological and physical distress of caregivers than non-
caregivers; CI ⫽ confidence interval of the mean effect size; Q
W
⫽ Significant values indicate heterogeneity of
the effect size; SWB ⫽ subjective well-being.
** p ⬍ .01. *** p ⬍ .001.
254
PINQUART AND SO
¨
RENSEN

tative samples than in nonrepresentative samples. In addition,
studies that controlled for the demographic equivalence of care-
givers and noncaregivers revealed smaller differences with regard
to subjective well-being and physical health. Furthermore, we
found larger differences between caregivers and noncaregivers in
studies that were published in peer-reviewed journals than in other
studies with regard to stress, depression, subjective well-being and
self-efficacy. However, the differences with regard to physical
health were smaller (see Table 4).
Discussion
The present meta-analysis suggests that caregivers fare worse
than noncaregivers with respect to five indicators of psychological
and physical health. Caregivers are more stressed, depressed, and
have lower levels of subjective well-being, physical health, and
self-efficacy than noncaregivers. However, caregiver status by
itself explains less than 8% of the variance of the dependent
variables. Although caring for an impaired relative has been de-
scribed as a very stressful experience, the size of the differences
between caregivers and noncaregivers in stress, depression, sub-
jective well-being, physical health, and self-efficacy is medium to
small according to J. Cohen’s (1992) guidelines. There are three
possible explanations for this finding. First, a substantial propor-
tion of caregivers report very low levels of burden and other
negative effects of caregiving or even no aversive effects of
caregiving (Schulz et al., 1997), probably because of below-
average involvement in caregiving (e.g., Beery et al., 1997), care
providers’ receipt of stress-buffering support (e.g., Barusch &
Spaid, 1989), positive aspects of caregiving (such as feeling use-
ful; e.g., Kramer, 1997), psychological resiliency (Garity, 1997),
or because caregivers—similar to other adults—may not be willing
to admit negative feelings to the researcher (Diener, Sandvik, &
Larsen, 1985). Second, negative effects of caregiving may also
have been underestimated in comparative studies of caregivers and
noncaregivers because caregiving-specific stressors were not as-
sessed. Third, the size of the observed effects may have been
reduced because of large random sampling error due to small
sample sizes and imperfect reliability of the measures. Because the
reliability of measures was not reported in many studies, we were
not able to correct for this source of bias in our meta-analysis.
Differences between caregivers and noncaregivers were greater
for stress, self-efficacy, depression, and subjective well-being than
for physical health. This may be the case because stress indicates
immediate negative effects of caregiving (e.g., Donaldson et al.,
1998; Kinney et al., 1995), and because it is difficult to maintain
one’s sense of competence and confidence when caregiving tasks
are difficult to manage and when one has little control over
symptoms of the care receiver. The above-average effect of care-
giving on depression may reflect symptoms of fatigue due to
chronic stress (Teel & Press, 1999), psychological responses to the
Table 2
Moderating Effects of the Type of Illness of the Care Recipient on Differences in Psychological
and Physical Health of Caregivers and Noncaregivers
Variable kN
cg
N
ncg
g
CI
Q
w
Significance
of the mean
difference
Lower
limit
Upper
limit
Stress
Dementia CG 14 728 831 0.85*** 0.58 1.12 74.94*** Dementia CG ⬎ mixed CG
Mixed CG 13 1,802 3,630 0.32*** 0.16 0.48 69.51***
Nondementia CG 3 166 117 0.47*** 0.23 0.71 1.95
Depression
Dementia CG 60 2,933 3,874 0.65*** 0.56 0.74 159.66***
Mixed CG 18 3,160 10,383 0.23*** 0.19 0.27 688.36***
Nondementia CG 3 144 126 0.27 ⫺0.08 0.62 3.61
Low SWB
Dementia CG 25 1,645 2,469 0.55*** 0.43 0.67 70.77*** Dementia CG, nondementia
CG ⬎ mixed CGMixed CG 20 2,422 14,456 0.20*** 0.11 0.29 49.35***
Nondementia CG 3 196 138 0.56*** 0.33 0.78 0.86
Poor health
Dementia CG 34 2,249 2,238 0.26*** 0.19 0.34 46.21* Dementia CG ⬎ mixed CG
Mixed CG 28 4,212 22,174 0.08** 0.02 0.14 57.47***
Nondementia CG 4 255 185 0.31** 0.12 0.50 4.29
Low self-efficacy
Dementia CG 7 371 351 0.80*** 0.53 1.07 18.29* Dementia CG ⬎ mixed CG
Mixed CG 6 497 708 0.19* 0.03 0.35 6.25
Nondementia CG 1 30 33 0.45 ⫺0.05 0.95 —
Note. To make the table easier to read, we coded all dependent variables in a way that positive coefficients
indicate larger impairments of caregivers compared with noncaregivers. The dash indicates that the heterogeneity
of effect sizes could not be computed, because only one study was available. k ⫽ number of effect sizes; N
cg
⫽
number of caregivers; N
ncg
⫽ number of noncaregivers; g ⫽ mean difference (measured in standard deviation
units): Positive values indicate higher levels of psychological and physical distress of caregivers than non-
caregivers; CI ⫽ confidence interval of the mean effect size; Q
W
⫽ Significant values indicate heterogeneity of
the effect size; CG ⫽ caregiver; SWB ⫽ subjective well-being.
* p ⬍ .05. ** p ⬍ .01. *** p ⬍ .001.
255
DIFFERENCES BETWEEN CAREGIVERS

Table 3
Moderating Effects of Type of Relationship to the Care Recipient, Representativeness of the Sample on Differences Between Caregivers’ and Noncaregivers’ Psychological
and Physical Health, and Sociodemographic Equivalence of Caregivers and Noncaregivers
Variable kN
cg
N
ncg
g
CI
Q
W
kN
cg
N
ncg
g
CI
Q
W
Contrast
effect
Lower
limit
Upper
limit
Lower
limit
Upper
limit
Spouses Adult children
Stress 8 654 823 1.05*** 0.60 1.50 78.31*** 8 830 1,449 0.23*** 0.13 0.33 7.53 Sign.
Depression 43 2,398 2,645 0.69*** 0.54 0.84 238.66*** 8 814 634 0.29*** 0.10 0.48 16.33* Sign.
Low SWB 22 1,375 2,127 0.43*** 0.29 0.57 61.45*** 4 499 940 0.16 ⫺0.03 0.35 9.28* —
Poor health 25 1,905 2,034 0.24*** 0.16 0.32 31.51 6 595 581 0.10 ⫺0.13 0.33 14.54* —
Low self-efficacy 7 519 398 0.69*** 0.40 0.98 25.66*** 4 239 140 0.24* 0.03 0.45 2.67 —
Representative samples Nonrepresentative samples
Stress 8 1,429 3,564 0.37*** 0.20 0.54 43.56*** 22 1,267 1,014 0.68*** 0.45 0.90 128.63*** —
Depression 12 2,563 9,888 0.26*** 0.03 0.49 209.86*** 69 3,674 4,495 0.64*** 0.56 0.72 182.06*** Sign.
Low SWB 11 1,789 13,787 0.14** 0.04 0.24 30.21*** 37 2,474 3,276 0.51*** 0.41 0.61 87.89*** Sign.
Poor health 20 3,852 21,759 0.09** 0.02 0.16 49.98*** 46 2,864 2,839 0.23*** 0.16 0.30 65.61* Sign.
Low self-efficacy 3 349 649 0.18 ⫺0.06 0.40 4.47 11 549 443 0.66*** 0.42 0.90 30.72*** Sign.
Equivalent samples Equivalence not tested/nonequivalent samples
Stress 16 1,405 2,426 0.64*** 0.43 0.85 108.08*** 14 1,291 2,152 0.47*** 0.27 0.67 71.32*** —
Depression 65 5,155 12,618 0.56*** 0.45 0.67 426.90*** 16 1,082 1,765 0.63*** 0.46 0.80 43.64*** —
Low SWB 32 2,913 14,489 0.36*** 0.28 0.44 118.22*** 16 1,350 2,574 0.43*** 0.27 0.59 55.13*** —
Poor health 47 5,368 22,937 0.14*** 0.09 0.20 97.15*** 19 1,348 1,660 0.24*** 0.13 0.34 34.74** —
Low self-efficacy 14 898 1,092 0.54*** 0.31 0.77 58.88*** 0
Note. Dashes indicate that there were no significant contrast effects. k ⫽ number of effect sizes; N
cg
⫽ number of caregivers; N
ncg
⫽ number of noncaregivers; g ⫽ mean difference (measured in
standard deviation units): Positive values indicate higher levels of psychological and physical distress of caregivers than noncaregivers; CI ⫽ confidence interval of the effect size; Q
W
⫽ significant
values indicate heterogeneity of the effect size; Sign. ⫽ significant; SWB ⫽ subjective well-being.
* p ⬍ .05. ** p ⬍ .01. *** p ⬍ .001.
256
PINQUART AND SO
¨
RENSEN

lack of control over the caregiving demands (Miller, Campbell,
Farran, & Kaufman, 1995), and because in some studies depres-
sion was measured by clinician ratings, which are less susceptible
to dissimulation of symptoms than self-ratings (Eaton, Neufeld,
Chen, & Cai, 2000).
Effects of Sample Characteristics
Although caregivers were in worse psychological and physical
health than noncaregivers, these effects were moderated by sample
characteristics. Differences between caregivers and noncaregivers
were considerably larger in dementia caregivers than in studies
that included a combination of caregivers for demented and non-
demented older adults. Probably because of the small number of
studies that compared nondementia caregivers to noncaregivers,
we were not able to detect significant differences to studies on
dementia caregiving. According to J. Cohen’s (1992) guidelines,
differences between dementia caregivers and noncaregivers for
stress and self-efficacy can be described as being large and differ-
ences in depression and subjective well-being as being medium,
thus supporting the notion that caregiving for demented older
adults is especially stressful (e.g., Ory et al., 2000).
The difference in stress and depression between caregivers and
noncaregivers was greater for spouses than for adult children.
Spousal caregivers have higher levels of objective burden than
adult child caregivers and fewer psychological and physical re-
sources to cope with stressors, due to age-associated losses and
declines. In fact, after statistically controlling for care receiver’s
illness and caregiver age in a multivariate analysis, the negative
effect of caregiving was larger in samples with more spouses only
for stress, physical health, and self-efficacy. Note that greater
impairments of spousal caregivers’ health compared with adult
children may not exclusively reflect above-average negative ef-
fects of caregiving. Because married couples are likely to have
similar risks for poor health outcomes as a result of selection,
lifestyle factors, and access to health care, having a disabled
spouse increases the risk of poor health, even if no care is provided
(e.g., Bookwala & Schulz, 1996). Future studies that compare
caregiving spouses and adult children should control for these
factors.
Differences between caregivers and noncaregivers were larger
in samples with a high percentage of female participants: Women
tend to provide more personal and instrumental care than men
(Miller & Cafasso, 1992; Yee & Schulz, 2000). In addition, they
are more likely to report impaired well-being than men because of
their greater sensitivity to negative feelings, greater willingness to
report negative feelings, and less effective coping styles (Lutzky &
Knight, 1994).
We also identified moderating effects of age on differences
between caregivers and noncaregivers, but these were somewhat
inconsistent. Differences between caregivers and noncaregivers
were larger for depression and self-efficacy in the older than in the
younger samples, thus indicating that older adults have fewer
coping resources (e.g., Baltes & Mayer, 1999). However, differ-
ences in stress were larger for younger than for older samples,
probably because younger caregivers may have more competing
roles to fulfill (e.g., work and family responsibilities), thus increas-
ing the risk that caregiving demands exceed the individual re-
sources (e.g., Moen, Robison, & Dempster-McClean, 1995).
Effects of the Quality of the Studies
Although it has been suggested that differences between care-
givers and noncaregivers may be larger in nonrepresentative sam-
ples than in representative samples, previous qualitative reviews of
caregiving research were not able to test for this difference (e.g.,
Schulz et al., 1995). In the present meta-analysis, we were able to
show that the tendency of caregivers to report lower levels of
psychological and physical health than noncaregivers is signifi-
cantly larger in nonrepresentative samples than in representative
samples. Many caregiver studies recruit convenience samples
through self-help organizations or service providers and are, there-
Table 4
Predictors of Differences in Caregivers’ and Noncaregivers’ Psychological and Physical Health (Weighted Ordinary Least Squares
Regression Analysis)
Variable
Stress Depression Low SWB Poor physical health Low self-efficacy
B
␤
B
␤
B
␤
B
␤
B
␤
Dementia CGs
a
0.32*** 0.30 0.08 0.09† 0.11*** 0.17 0.15*** 0.33 0.18* 0.23
Spousal CGs (%) 0.01*** 0.55 0.00 0.11 ⫺0.00 ⫺0.02 0.002*** 0.27 0.01* 0.53
Female CGs (%) ⫺0.00 ⫺0.09 0.002* 0.11 0.002*** 0.14 0.002*** 0.19 0.003* 0.18
M age of CGs ⫺0.01* 0.30 0.01*** 0.26 0.00 0.09 0.00 0.02 0.07*** 1.05
Representative sample
b
⫺0.18* ⫺0.16 ⫺0.23*** ⫺0.23 ⫺0.27*** ⫺0.38 ⫺0.02 ⫺0.04 ⫺0.33*** ⫺0.37
Demographic equivalence of CGs
and controls ⫺0.03 ⫺0.03 ⫺0.05 ⫺0.05 ⫺0.09*** ⫺0.13 ⫺0.10*** ⫺0.20 ——
Quality of source
c
0.22* 0.09 0.08* 0.09 0.24*** 0.28 ⫺0.06*** ⫺0.11 0.54*** 0.49
Constant 1.13* ⫺0.40 ⫺0.06 ⫺0.01 ⫺4.02**
R
2
0.35 0.26 0.43 0.24 0.72
Note. Dependent variable is the difference between caregivers’ (CGs) and noncaregivers’ psychological and physical distress. Higher positive values
indicate higher distress of CGs than non-CGs. B ⫽ unstandardized regression coefficient;
␤
⫽ standardized regression coefficient; R
2
⫽ explained variance
by the moderators; SWB ⫽ subjective well-being. Dashes indicate that no effect size could be computed for demographic equivalence, because this variable
did not vary between studies.
a
1 ⫽ yes, 0 ⫽ no/mix of CG.
b
1 ⫽ yes, 0 ⫽ no.
c
1 ⫽ peer-reviewed journal, 0 ⫽ other.
† p ⬍ .10. * p ⬍ .05. ** p ⬍ .01. *** p ⬍ .001.
257
DIFFERENCES BETWEEN CAREGIVERS

fore, likely to overrepresent distressed caregivers (see also Schulz
et al., 1995). In fact, in the subsample of studies that reported the
average number of caregiving hours per week, we found higher
caregiving demands in nonrepresentative than in representative
samples (36.7 hr/week vs. 26.8 hr/week), t(13720) ⫽ 27.96, p ⬍
.001.
Schulz et al. (1997) have also suggested that representative
studies may underestimate caregiver stress because they define
caregivers as individuals sharing a household with an impaired
family member without assessing whether the respondent provides
care. We ensured that this was not the case in our analysis by
excluding studies on family members who shared the household
with ill older adults without providing support.
After controlling for other moderators, we found smaller differ-
ences between caregivers and noncaregivers in subjective well-
being and physical health in studies that controlled for the demo-
graphic equivalence of caregivers and noncaregivers. This
indicates that uncontrolled demographic differences between care-
givers and noncaregivers may inflate observed differences in psy-
chological and physical health. For example, in some studies
caregivers were older or had lower SES than noncaregivers. How-
ever, uncontrolled demographic differences did not exert a consis-
tent effect: In some studies there were stronger demographic risk
factors for low psychological and physical health in control-group
members than in caregivers. These inconsistencies may have con-
tributed to a reduction in the difference between caregivers and
control-group members for depression and stress.
For four out of five outcome variables, differences between
caregivers and noncaregivers were larger in studies that were
published in peer-reviewed journals. This can be interpreted in
terms of the file drawer problem: Significant results are more
likely to be published than nonsignificant results (Rosenthal,
1991). The opposite was found with regard to physical health,
probably based on some published large studies that found low
caregiving effects on physical health (e.g., Schulz et al., 1997).
Limitations and Conclusions
A few caveats need to be considered regarding the present study.
First, we focused on five outcome variables related to stress,
well-being, and physical health. Other outcome variables, such as
the level of anxiety, the level of social integration, or leisure time
activities, should also be considered. Unfortunately, we did not
have access to a sufficient number of studies that compared care-
givers and noncaregivers with regard to these variables. Second,
only a small number of studies that compared caregivers of phys-
ically frail, nondemented older adults with noncaregivers were
available. Thus, the results regarding this group have to be inter-
preted with caution. Third, in the present meta-analysis we focused
on only seven moderator variables. Other variables may influence
observed psychological differences between caregivers and non-
caregivers as well, for example, SES and the duration of caregiv-
ing. Because most of the studies reviewed here did not provide
information on these variables, we were not able to include them.
Fourth, random-effect models may slightly overestimate the actual
between-studies variability of effect sizes and make the identifi-
cation of moderators more difficult (e.g., Hedges & Vevea, 1998).
Nevertheless, this more conservative approach is appropriate for
low information situations (Overton, 1998). Such is the case in our
study because other potential moderators cannot be controlled for
and because only very few studies on nondementia caregivers were
available for making statistical inferences.
Despite these caveats, several conclusions can be drawn from
the present study. First, although caregivers have higher levels of
stress and depression and lower levels of subjective well-being,
physical health, and self-efficacy than noncaregivers, most differ-
ences are small to medium, and the size of difference varies by
study characteristics. However, caring for demented relatives is
consistently associated with considerably higher stress levels and a
high risk for poor psychological and physical health. Therefore,
caregivers for demented persons should be considered for psycho-
social interventions.
Second, because the largest differences between caregivers and
noncaregivers were found with regard to depression, stress, and
self-efficacy, the main points of an intervention should be to
reduce the objective amount of care they need to provide and give
them more freedom in their time use (e.g., by providing flexible
respite; Zarit, Stephens, Townsend, & Greene, 1998), to increase
caregivers’ ability to control the caregiving experience (e.g.,
increasing skills to manage behavior disturbances by means
of problem-solving and self-management therapy; Gallagher-
Thompson et al., 2000), and to reduce negative thoughts that might
contribute to stress and depression (e.g., Chang, 1999). A recent
meta-analysis by So¨rensen, Pinquart, and Duberstein (2002)
showed that psychotherapeutic interventions improved caregivers’
stress, depression, and self-efficacy by about one third of a stan-
dard deviation, and that respite or adult day care had similar effects
on the stress reduction.
Third, with regard to research methodology, we have concluded
that results from studies that use nonrepresentative samples of
caregivers have to be interpreted with caution because most of the
studies exclusively focus on highly distressed caregivers and are
therefore likely to overestimate the negative effects of caregiving.
Future epidemiological studies on physical and mental impairment
in old age may be used to get representative caregiver samples that
would enable less biased estimates of impact of caregiving. Sim-
ilarly, studies that compare caregivers with noncaregivers must
assess the equivalence of the samples with regard to important
demographic variables, such as age and SES.
Fourth, because most studies report neither the levels of care
receivers’ impairments in activities of daily living and instrumen-
tal activities of daily living, nor the amount of care provision, we
were not able to include these variables in our meta-analysis. To
understand moderators on the impact of caregiving, we recom-
mend that future studies provide more descriptive information with
regard to these and other potential moderator variables of caregiv-
ing outcomes.
Fifth, because we were only able to locate a very small number
of studies that compared homogeneous samples of caregivers for
nondemented older adults to noncaregivers, we recommend more
research in this field. Finally, we encourage research that compares
caregivers and noncaregivers with regard to other outcome vari-
ables that could not be included in the present meta-analysis, such
as future time perspective or social integration. These variables
would add to the understanding of caregiver stress by further
probing the initial vulnerability of particular groups or types of
caregivers and enabling more tailored interventions.
258
PINQUART AND SO
¨
RENSEN

References
References marked with an asterisk indicate studies included in the
meta-analysis.
*Acton, G. J. (2002). Health-promoting self-care in family caregivers.
Western Journal of Nursing Research, 24(1), 73–88.
*Allen-Holmes, L. M., Markides, K., & Chiriboga, D. (2000). Physical
health and service use utilization among elderly Hispanic caregivers
and non-caregivers. Paper presented at the 53rd annual scientific meet-
ing of the Gerontological Society of America, Washington, DC.
*Almberg, B., Jansson, W., Grafstro¨m, M., & Winblad, B. (1998). Differ-
ences between and within genders in caregiving strain: A comparison
between caregivers of demented and non-caregivers of non-demented
elderly people. Journal of Advanced Nursing, 28, 849–858.
Baltes, P. B., & Mayer, K. U. (Eds.). (1999). The Berlin Aging Study:
Aging from 70 to 100. New York: Cambridge University Press.
Barber, C. E., & Pasley, B. K. (1994). Family care of Alzheimer’s patients:
The role of gender and generational relationship on caregiver outcomes.
Journal of Applied Gerontology, 14, 172–192.
*Barnes, K. A., Hicks, P., & Kirk, M. (2001, November). Caregiving
among older husbands. Paper presented at the 54th annual meeting of
the Gerontological Society of America, Chicago.
*Barnes, K. A., & Patrick, J. H. (2000). Effects of caregiving on older
husbands. Paper presented at the 53rd annual scientific meeting of the
Gerontological Society of America, Washington, DC.
Barusch, A. S., & Spaid, W. M. (1989). Gender differences in caregiving:
Why do wives report greater burden. Gerontologist, 29, 667–676.
*Bauer, M. E., Vedhara, K., Perks, P., Wilcock, G. K., Lightman, S. L., &
Shanks, N. (2000). Chronic stress in caregivers of dementia patients is
associated with reduced lymphocyte sensitivity to glucocorticoids. Jour-
nal of Neuroimmunology, 103, 84–92.
*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.
Beck, A. T., & Steer, R. A. (1987). Beck Depression Inventory—Manual.
San Antonio, TX: Psychological Corporation.
Beery, L. C., Prigerson, H. G., Bierhals, A. J., et al. (1997). Traumatic
grief, depression and caregiving in elderly spouses of the terminal ill.
OMEGA, 35, 261–279.
*Bertram, H. (1995). Das Individuum und seine Familie: Lebensformen,
Familienbeziehungen und Lebensereignisse im Erwachsenenalter [The
individual and his/her family: Living arrangements, family relations, and
life-events in adulthood]. Opladen, Germany: Leske & Budrich.
Blazer, D. G. (1993). Depression in late life (2nd ed.). St. Louis, MO:
Mosby.
*Bodnar, J. C., & Kiecolt-Glaser, J. K. (1994). Caregiver depression after
bereavement: Chronic stress isn’t over when it’s over. Psychology and
Aging, 9, 372–380.
Bookwala, J., & Schulz, R. (1996). Spousal similarity in subjective well-
being: The cardiovascular health study. Psychology and Aging, 11,
582–590.
*Bougard, L. L. (1995). Existential growth resulting from giving care to a
spouse with Alzheimer’s disease. Unpublished doctoral dissertation,
Arizona State University.
Brandtsta¨dter, J., & Greve, W. (1994). The aging self: Stabilizing and
protective processes. Developmental Review, 14, 52–80.
Brodarty, H., & Hadzi-Pavlovic, D. (1990). Psychosocial effects on carers
of living with persons with dementia. Australian and New Zealand
Journal of Psychiatry, 24, 351–361.
*Brody, E. M., Hoffman, C., Kleban, M. H., & Schoonover, C. B. (1989).
Caregiving daughters and their local siblings: Perceptions, strains, and
interactions. Gerontologist, 29, 529–538.
Bushman, B. J., & Wang, M. C. (1996). A procedure for combining sample
standardized mean differences and vote counts to estimate the popula-
tion standardized mean difference in fixed effect models. Psychological
Methods, 1, 66–80.
*Cacioppo, J. T., Burleson, M. H., Poehlmann, K. M., Malarkey, W. B., et
al. (2000). Autonomic and neuroendocrinic responses to mild psycho-
logical stressors: Effects of chronic stress on older women. Annals of
Behavioral Medicine, 22, 140–148.
*Carnwath, T. C., & Johnson, D. A. (1987). Psychiatric morbidity among
spouses of patients with stroke. British Medical Journal, 294, 409–411.
*Cattanach, L., & Tebes, J. K. (1991). The nature of elder impairment and
its impact on family caregivers’ health and psychosocial functioning.
Gerontologist, 31, 246–255.
Chang, B. L. (1999). Cognitive-behavioral intervention for homebound
caregivers of persons with dementia. Nursing Research, 48, 173–182.
Clipp, E. C., & George, L. K. (1993). Dementia and cancer: A comparison
of spouse caregivers. Gerontologist, 33, 534–541.
Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155–159.
Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of
perceived stress. Journal of Health and Social Behavior, 24, 385–396.
*Dautzenberg, M. G., Diedriks, J. P., Philipsen, H., & Tan, F. E. (1999).
Multigenerational caregiving and well-being: Distress in middle-aged
daughters providing assistance to elderly parents. Women and Health,
29(4), 57–74.
Diener, E., Sandvik, E., & Larsen, R. J. (1985). Age and sex differences for
emotional intensity. Developmental Psychology, 21, 542–546.
Donaldson, C., Tarrier, N., & Burns, A. (1998). Determinants of carer
stress in Alzheimer’s disease. International Journal of Geriatric Psy-
chiatry, 13, 248–256.
Draper, B. M., Poulos, C. J., Cole, A. D., Poulos, R. G., & Ehrlich, F.
(1992). A comparison of caregivers for elderly stroke and dementia
victims. Journal of the American Geriatrics Society, 40, 896–901.
*Dura, J. R., Haywood-Niler, E., & Kiecolt-Glaser, J. K. (1990). Spousal
caregivers of persons with Alzheimer’s disease and Parkinson’s disease
dementia. Gerontologist, 30, 332–336.
*Dura, J. R., Stukenberg, K. W., & Kiecolt-Glaser, J. K. (1990). Chronic
stress and depressive disorders in older adults. Journal of Abnormal
Psychology, 99, 284–290.
*Dura, J. R., Stukenberg, K. W., & Kiecolt-Glaser, J. K. (1991). Anxiety
and depressive disorders in adult children caring for demented parents.
Psychology and Aging, 6, 467–473.
Eaton, W., Neufeld, K., Chen, L. S., & Cai, G. (2000). A comparison of
self-report and clinical diagnostic interviews for depression: Diagnostic
Interview Schedule and Schedules for Clinical Assessment in Neuro-
psychiatry in the Baltimore Epidemiologic Catchment Area follow-up.
Archives of General Psychiatry, 57, 217–222.
Faßmann, H., & Grillenberger, R. (1996). Burnout bei Pflegepersonen von
Schwerpflegebedu¨rftigen [Burnout in caregivers of the severely ill].
Report Psychologie, 21, 788–798.
*France, A. C., & Alpher, V. (1995). Structural analysis of social behavior
and perceptions of caregiving. Journal of Psychology, 129, 375–388.
*Fuller-Jonap, F., & Haley, W. W. (1995). Mental and physical health of
male caregivers of a spouse with Alzheimer’s disease. Journal of Aging
and Health, 7, 99–118.
*Gage, M. J., & Kinney, J. M. (1995). They aren’t for everyone: The
impact of support group participation on caregivers’ well-being. Clinical
Gerontologist, 16(2), 21–34.
*Gallagher-Thompson, D., DalCanto, P. G., Jacob, T., & Thompson, L. W.
(2001). A comparison of marital interaction patterns between couples in
which the husband does or does not have Alzheimer’s disease. Journals
of Gerontology. Series B, Psychological Sciences and Social Sci-
ences, 56, S140-S150.
Gallagher-Thompson, D., Lovett, S., Rose, J., McKibbin, C., Coon, D.,
Futterman, A., & Thompson, L. W. (2000). Impact of psychoeducational
259
DIFFERENCES BETWEEN CAREGIVERS

interventions on distressed family caregivers. Journal of Clinical Gero-
psychology, 6, 91–110.
Garity, J. (1997). Stress, learning style, resilience factors, and ways of
coping in Alzheimer family caregivers. American Journal of Alzheimer’s
Disease, 12, 171–178.
*Gaynor, S. E. (1988). The effects of home care on elderly female care-
givers. Unpublished doctoral dissertation, Rush University.
Gilleard, C. J., Gilleard, K., Gledhill, K., & Whittick, J. (1984). Caring for
the mentally infirm at home: A survey of the supporters. Journal of
Epidemiology and Community Health, 38, 319–325.
*Glaser, R., Sheridan, J., Malarkey, W. B., MacCallum, R. C., & Kiecolt-
Glaser, J. K. (2000). Chronic stress modulates the immune response to
a pneumococcal pneumonia vaccine. Psychosomatic Medicine, 62, 804–
807.
*Golden-Kreutz, D. M. (1993). Social networks, support perception, and
depressive symptomatology in dementia family caregivers and noncar-
egivers. Unpublished doctoral dissertation, Ohio State University.
*Grafstro¨m, M., Fratiglioni, L., Sandman, P. -O., & Winblad, B. (1992).
Health and social consequences for relatives of demented and non-
demented elderly: A population-based study. Journal of Clinical Epide-
miology, 45, 861–870.
*Grant, I., Adler, K. A., Patterson, T. L., Dimsdale, J. E., Ziegler, M. G.,
& Irwin, M. R. (2002). Health consequences of Alzheimer’s caregiving
transitions: Effects of placement and bereavement. Psychosomatic Med-
icine, 64, 477–486.
*Haley, W. E., Levine, E. G., Brown, S. L., Berry, J. W., & Hughes, G. H.
(1987). Psychological, social, and health consequences of caring for a
relative with senile dementia. Journal of the American Geriatrics Soci-
ety, 35, 405–411.
*Haley, W. E., West, C. A., Wadley, V. G., Ford, G. R., White, F. A.,
Barrett, J. J., et al. (1995). Psychosocial, social, and health impact of
caregiving: A comparison of Black and White dementia family caregiv-
ers and noncaregivers. Psychology and Aging, 10, 540–552.
Hamilton, M. (1967). Development of a rating scale for primary depressive
illness. British Journal of Social and Clinical Psychology, 6, 278–296.
Hedges, L. V. (1981). Distribution theory for Glass’s estimator of effect
size and related estimators. Journal of Educational Statistics, 6, 107–
128.
Hedges, L. V., & Vevea, J. L. (1998). Fixed- and random-effects models in
meta-analysis. Psychological Methods, 3, 486–504.
Hooker, K., Monahan, D. J., Bowman, S. R., Frazier, L. D., & Shifren, K.
(1998). Personality counts a lot: Predictors of mental and physical health
of spouse caregivers in two disease groups. Journals of Gerontology.
Series B, Psychological Sciences and Social Sciences, 53, P73-P85.
Hooker, K., Monahan, D., Shifren, K., & Hutchinson, C. (1992). Mental
and physical health of spouse caregivers: The role of personality. Psy-
chology and Aging, 7, 367–375.
*Hoyert, D. L., & Seltzer, M. M. (1992). Factors related to the well-being
and life activities of family caregivers. Family Relations, 41, 74–81.
*Hubley, A. M., Hemmingway, D., Michalos, A. C., & Zumbo, B. D.
(2001). Quality of life and health of caregivers and non-caregivers who
are 65 years or older. Paper presented at the 17th world congress of the
International Association of Gerontology, Vancouver, British Columbia,
Canada.
Huck, D. M., & Armer, J. M. (1996). Health perceptions and health-
promoting behaviors among elderly Catholic nuns. Family and Commu-
nity Health, 18, 81–91.
*Irwin, M., Brown, M., Patterson, T., Hauger, R., Mascovich, A., & Grant,
I. (1991). Neuropeptide Y and natural killer cell activity: Findings in
depression and Alzheimer caregiver stress. FASEB Journal, 5, 3100–
3107.
*Irwin, M., Hauger, R., Patterson, T. L., Semple, S., Ziegler, M., & Grant,
I. (1997). Alzheimer caregiver stress: Basal natural killer cell activity,
pituitary-adrenal cortical function, and sympathetic tone. Annals of
Behavioral Medicine, 19, 83–90.
*Jutras, S., & Lavoie, J. -P. (1995). Living with an impaired elderly person:
The informal caregiver’s physical and mental health. Journal of Aging
and Health, 7, 46–73.
*Karlin, N. J., & Retzlaff, P. D. (1995). Psychopathology in caregivers of
the chronically ill: Personality and clinical syndromes. Hospice Journal,
10(3), 55–61.
*Kiecolt-Glaser, J. K., Dura, J. R., Speicher, C. E., Trask, J., & Glaser, R.
(1991). Spousal caregivers of dementia victims: Longitudinal changes in
immunity and health. Psychosomatic Medicine, 53, 345–362.
*Kiecolt-Glaser, J. K., Glaser, R., Shuttleworth, E. C., Dyer, C. S.,
Ogrocki, P., & Speicher, C. E. (1987). Chronic stress and immunity in
family caregivers of Alzheimer’s disease victims. Psychosomatic Med-
icine, 49, 523–535.
*King, A., Oka, R. A., & Young, D. R. (1994). Ambulatory blood pressure
and heart rate responses to the stress of work and caregiving in older
women. Journals of Gerontology. Series A, Biological Sciences and
Medical Sciences, 49, M239-M245.
*Kinney, J. M., & Stephens, M. A. (1989). Hassles and uplifts of giving
care to a family member with dementia. Psychology and Aging, 4,
402–408.
Kinney, J. M., Stephens, M. P., Franks, M. M., & Norris, V. K. (1995).
Stress and satisfaction of family caregivers to older stroke patients.
Journal of Applied Gerontology, 14, 3–21.
Kosberg, J. I., & Cairl, R. E. (1986). The Cost of Care Index: A case
management tool for screening informal care providers. Gerontolo-
gist, 26, 273–285.
Kramer, B. J. (1997). Gain in the caregiving experience: Where are we?
What next? Gerontologist, 37, 218–232.
*Kramer, B. J., & Lambert, J. D. (1999). Caregiving as a life course
transition among older husbands: A prospective study. Gerontolo-
gist, 39, 658–667.
*Lawton, M. P., Moss, M., Hoffman, C., & Perkinson, M. (2000). Two
transitions in daughters’ caregiving careers. Gerontologist, 40, 437–448.
*Lee, J. A. (1997). Balancing elder care responsibilities and work: Two
empirical studies. Journal of Occupational Health Psychology, 2, 220–
228.
*Lee, J. A., Walker, M., & Shoup, R. (2001). Balancing elder care
responsibilities and work: The impact on emotional health. Journal of
Business and Psychology, 16, 277–289.
*Levesque, L., Cossette, S., & Ducharme, F. (1993). La sante´ mentale de
conjoints aˆge´s Quebe´cois: Comparaison entre aidants naturels et non
aidants [The mental health of elderly spouses in Quebec: A comparison
of caregivers and noncaregivers]. Canadian Journal of Community Men-
tal Health, 12, 111–125.
*Lichtenberg, P. A., Manning, C. A., & Turkheimer, E. (1992). Memory
dysfunction in depressed spousal caregivers. Clinical Gerontologist, 12,
77–80.
*Loomis, L. S., & Booth, A. (1995). Multigenerational caregiving and
well-being: The myth of the beleaguered sandwich generation. Journal
of Family Issues, 16, 131–148.
*Lorensini, S., & Bates, G. W. (1997). The health, psychological and social
consequences of caring for a person with dementia. Australian Journal
on Aging, 16, 198–202.
*Lutgendorf, S. K., Garand, L., Buckwalter, K. C., Reimer, T. T., Hong,
S.-Y., & Lubaroff, D. M. (1999). Life stress, mood disturbance, and
elevated Interleukin-6 in healthy older women. Journals of Gerontology.
Series A, Biological Sciences and Medical Sciences, 54, M434-M439.
Lutzky, S. M., & Knight, B. G. (1994). Explaining gender differences in
caregiver distress: The roles of emotional attentiveness and coping
styles. Psychology and Aging, 9, 513–519.
*McGuire, L., Kiecolt-Glaser, J. K., & Glaser, R. (2002). Depressive
260
PINQUART AND SO
¨
RENSEN

symptoms and lymphocyte proliferation in older adults. Journal of
Abnormal Psychology, 111, 192–197.
*McMillan, S. C., & Mahon, M. (1994). The impact of hospice services on
the quality of life of primary caregivers. Oncology Nursing Forum, 21,
1189–1195.
Miller, B., & Cafasso, L. (1992). Gender differences in caregiving: Fact or
artifact? Gerontologist, 32, 498–507.
Miller, B., Campbell, R. T., Farran, C. J., & Kaufman, J. E. (1995). Race,
control, mastery, and caregiver distress. Journals of Gerontology. Series
B, Psychological Sciences and Social Sciences, 50, S374-S382.
Moen, P., Robison, J., & Dempster-McClean, D. (1995). Caregiving and
women’s well-being: A life course approach. Journal of Health and
Social Behavior, 36, 259–273.
*Morris, D. L., Wykle, M. L., & Lu, J. (2001, November). A comparison
of self-care practices on the health of elderly caregivers and non-
caregivers. Paper presented at the 54th annual meeting of the Geronto-
logical Society of America, Chicago.
*Mort, J. R., Gaspar, P. M., Juffer, D. I., & Kovarna, M. B. (1996).
Comparison of psychotropic agent use among rural elderly caregivers
and noncaregivers. Annals of Pharmacotherapy, 30, 583–585.
Neugarten, B. L., Havighurst, R. J., & Tobin, S. S. (1961). The measure-
ment of life satisfaction. Journal of Gerontology, 16, 134–143.
Ory, M. G., Yee, J. L., Tennstedt, S. L., & Schulz, R. (2000). The extent
and impact of dementia care: Unique challenges experienced by family
caregivers. In R. Schulz (Ed.), Handbook of dementia caregiving:
Evidence-based interventions for family caregivers (pp. 1–32). New
York: Springer.
Overton, R. C. (1998). A comparison of fixed-effects and mixed (random-
effects) models for meta-analysis: Tests for moderator variable effects.
Psychological Methods, 3, 354–379.
*Owens, S. J., & Swensen, C. H. (2000, August). Loneliness in older adult
spousal caregivers. Paper presented at the 108th Annual Convention of
the American Psychological Association, Washington, DC.
Pearlin, L., & Schooler, C. (1978). The structure of coping. Journal of
Health & Social Behavior, 19, 2–21.
*Perez-Cahill, D. (1998). The impact of multiple caregiving roles on
well-being: A longitudinal study of middle-aged adults. Unpublished
doctoral dissertation, University of Massachusetts.
*Picot, S. J., Zauszniewski, J. A., Debanne, S. M., & Holston, E. C. (1999).
Mood and blood pressure response in Black female caregivers and
noncaregivers. Nursing Research, 48, 150–161.
*Picot, S. J., Zauszniewski, J. A., & Delgado, C. (1997). Cardiovascular
response of African American female caregivers. Journal of National
Black Nurses Association, 9, 3–21.
Pinquart, M. (1998). Das Selbstkonzept im Seniorenalter [Self-concept in
old age]. Weinheim, Germany: Beltz.
Poulshok, S. W., & Deimling, G. T. (1984). Families caring for elders in
residence: Issues in the measurement of burden. Journal of Gerontol-
ogy, 39, 230–239.
*Pruchno, R. A., & Potashnik, S. L. (1989). Caregiving spouses: Physical
and mental health in perspective. Journal of the American Geriatrics
Society, 37, 697–705.
Radloff, L. S. (1977). The CES-D scale: A self-report depression scale for
research in the general population. Journal of Applied Psychological
Measurement, 1, 385–401.
Raudenbush, S. W. (1994). Random effect models. In H. Cooper & L. V.
Hedges (Eds.), Handbook of research synthesis (pp. 301–321). New
York: Sage.
*Reese, D. R., Gross, A. M., Smalley, D. L., & Messer, S. C. (1994).
Caregivers of Alzheimer’s disease and stroke patients: Immunological
and psychological considerations. Gerontologist, 34, 534–540.
*Richardson, R. C., & Sistler, A. B. (1999). The well-being of elderly
Black caregivers and noncaregivers: A preliminary study. Journal of
Gerontological Social Work, 31, 109–117.
*Robinson, K. M., & Kaye, J. (1994). The relationship between spiritual
perspective, social support, and depression in caregiving and noncare-
giving wives. Scholary Inquiry for Nursing Practice, 8, 375–389.
*Robinson-Whelen, S., Kim, C., McCallum, R. C., & Kiecolt-Glaser, J. K.
(1997). Distinguishing optimism from pessimism in older adults: Is it
more important to be optimistic or not to be pessimistic? Journal of
Personality and Social Psychology, 73, 1345–1353.
*Robinson-Whelen, S., Tada, Y., McCallum, R. C., McGuire, L., &
Kiecolt-Glaser, J. K. (2001). Long-term caregiving: What happens when
it ends? Journal of Abnormal Psychology, 110, 573–584.
*Rose-Rego, S. K., Strauss, M. E., & Smyth, K. A. (1998). Differences in
the perceived well-being of wives and husbands caring for persons with
Alzheimer’s disease. Gerontologist, 38, 224–230.
Rosenthal, R. (1991). Meta-analytic procedures for social research. Bev-
erly Hills, CA: Sage.
*Roth, D. L., Haley, W. E., Owen, J. E., Clay, O. J., & Goode, K. T.
(2001). Latent growth models of the longitudinal effects of dementia
caregiving: A comparison of African American and White family care-
givers. Psychology and Aging, 16, 427–436.
*Russo, J., Vitaliano, P. P., Brewer, D. D., Katon, W., & Becker, J. (1995).
Psychiatric disorders in spouse caregivers of care recipients with Alz-
heimer’s disease and matched controls: A diathesis-stress model of
psychopathology. Journal of Abnormal Psychology, 104, 197–204.
*Scanlan, J. M., Vitaliano, P. P., Zhang, J., Savage, M., & Ochs, H. D.
(2001). Lymphocyte proliferation is associated with gender, caregiving,
and psychosocial variables in older adults. Journal of Behavioral Med-
icine, 24, 537–559.
*Scharlach, A., & Runkle, C. (1994). Health conditions and service utili-
zation of adults with elder care responsibilities. Journal of Aging and
Health, 6, 336–352.
Schneider, J., Murray, J., Banerjee, S., & Mann, A. (1999). EUROCARE:
A cross-national study of co-resident spouse carers for people with
Alzheimer’s disease: I. Factors associated with carer burden. Interna-
tional Journal of Geriatric Psychiatry, 14, 651–661.
*Schulz, R., Beach, S. R., Lind, B., Martire, L. M., Zdaniuk, B., Hirsch, C.,
et al. (2001). Involvement in caregiving and adjustment to death of a
spouse. Journal of the American Medical Association, 27, 3123–3129.
*Schulz, R., Newsom, J., Mittelmark, M., Burton, L., Hirsch, C., &
Jackson, S. (1997). Health effects of caregiving: The caregiver health
effects study: An ancillary study of the Cardiovascular Health Study.
Annals of Behavioral Medicine, 19, 110–116.
Schulz, R., O’Brien, A. T., Bookwala, J., & Fleissner, K. (1995). Psychi-
atric and physical morbidity effects of dementia caregiving: Prevalence,
correlates, and causes. Gerontologist, 35, 771–791.
*Seltzer, M. M., & Li, W. L. (2000). The dynamics of caregiving: Tran-
sitions during a three-year prospective study. Gerontologist, 40, 165–
178.
*Shaw, W. S., Patterson, T. L., Semple, S. J., Grant, I., Yu, E. S., Zhang,
M., et al. (1997). A cross-cultural validation of coping strategies and
their associations with caregiving distress. Gerontologist, 37, 490–504.
*Shaw, W. S., Patterson, T. L., Semple, S. J., Ho, S., Irwin, M. R., Hauger,
R. L., & Grant, I. (1997). Longitudinal analysis of multiple indicators of
health decline among spousal caregivers. Annals of Behavioral Medi-
cine, 19, 101–109.
*Sistler, A. B., & Blanchard-Fields, F. (1993). Being in control: A note on
differences between caregiving and noncaregiving spouses. Journal of
Psychology, 127, 537–542.
So¨rensen, S., Pinquart, M., & Duberstein, P. (2002). How effective are
interventions with caregivers? An updated meta-analysis. Gerontolo-
gist, 42, 356–372.
Steinmetz, S. K. (1988). Elder abuse by family caregivers: Processes and
intervention stragegies. Contemporary Family Therapy: An Interna-
tional Journal, 10, 256–271.
261
DIFFERENCES BETWEEN CAREGIVERS

Stone, R. G., Cafferata, L., & Sangl, J. (1987). Caregivers of the frail
elderly: A national profile. Gerontologist, 27, 616–626.
*Strawbridge, W. J., Wallhagen, M. I., Shema, S. J., & Kaplan, G. A.
(1997). New burdens or more of the same? Comparing grandparent,
spouse, and adult-child caregivers. Gerontologist, 37, 505–510.
*Teel, C. S., & Press, A. N. (1999). Fatigue among elders in caregiving and
noncaregiving roles. Western Journal of Nursing Research, 21, 498–
520.
Tennstedt, S. L., McKinlay, J. B., & Sullivan, L. M. (1988, November).
Informal care for frail elders: The role of secondary characteristics.
Paper presented at the annual scientific meeting of the Gerontological
Society of America, San Francisco.
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,
622–631.
*Vitaliano, P. P., Russo, J., & Niaura, R. (1995). Plasma lipids and their
relationship with psychological factors in older adults. Journals of
Gerontology. Series B, Psychological Sciences and Social Sciences, 50,
P18-P24.
*Vitaliano, P. P., Russo, J., Scanlan, J. M., & Greeno, C. G. (1996). Weight
changes in caregivers of Alzheimer’s care recipients: Psychobehavioral
predictors. Psychology and Aging, 11, 155–163.
*Vitaliano, P. P., Russo, J., Young, H. M., Becker, J., & Maiuro, R. D.
(1991). The screen for caregiver burden. Gerontologist, 31, 76–83.
Wallhagen, M. I. (1992–1993). Perceived control and adaptation in elder
caregivers: Development of an explanatory model. International Journal
of Aging and Human Development, 36, 219–237.
*Wallsten, S. M. (2000). Effects of caregiving, gender, and race on the
health, mutuality, and social supports of older couples. Journal of Aging
and Health, 12, 90–111.
*Wallsten, S. M., & Snyder, S. S. (1990). A comparison of elderly family
caregivers’ and noncaregivers’ perceptions of stress in daily experiences.
Journal of Community Psychology, 18, 228–238.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and vali-
dation of brief measures of positive and negative affect: The PANAS
scales. Journal of Personality and Social Psychology, 54, 1063–1070.
*Wells, Y. D., & Kendig, H. L. (1997). Health and well-being of spouse
caregivers and the widowed. Gerontologist, 37, 666–674.
*Wright, D. L., & Aquilino, W. S. (1998). Influence of emotional support
exchange in marriage on caregiving wives’ burden and marital satisfac-
tion. Family Relations, 47, 195–204.
*Wright, L. K. (1994). AD spousal caregivers: Longitudinal changes in
health, depression, and coping. Journal of Gerontological Nursing,
20(10), 33–48.
Yates, M. E., Tennstedt, S., & Chang, B. -H. (2000). Contributors to and
mediators of psychological well-being for informal caregivers. Journals
of Gerontology. Series B, Psychological Sciences and Social Sci-
ences, 54, P12-P22.
Yee, J. L., & Schulz, R. (2000). Gender differences in psychiatric morbid-
ity among family caregivers: A review and analysis. Gerontologist, 40,
147–164.
*Yonan, C. A., Hebel, J. R., & Baumgarten, M. (2000, November).
Depression in bereaved caregivers: A five-year prospective study. Paper
presented at the 53rd annual meeting of the Gerontological Society of
America, Washington, DC.
Zarit, S. H., Reever, K. E., & Bach-Petersen, J. (1980). Relatives of the
impaired elderly: Correlates of feelings of burden. Gerontologist, 20,
649–655.
Zarit, S. H., Stephens, M. -A., Townsend, A., & Greene, R. (1998). Stress
reduction for family caregivers: Effects of adult day care use. Journals
of Gerontology. Series B, Psychological Sciences and Social Sci-
ences, 53, S267-S277.
Appendix
Effect Sizes and 95% Confidence Intervals of Single Studies Ordered by the Size of the Effects
The articles by Vitaliano, Russo, and Niaura (1995) and Vitaliano et al.
(1996) are, in part, based on the same data set, but they report results for
different outcome variables or points of measurement. The same applies to
Dura, Haywood-Niler, and Kiecolt-Glaser (1990); Dura, Stukenberg, and
Kiecolt-Glaser (1990); Kiecolt-Glaser et al. (1991); and Schulz et al.
(1997; Schulz et al., 2001), as well as Grant et al. (2002); Irwin et al.
(1991); Shaw, Patterson, Semple, Grant, et al. (1997); and Shaw, Patterson,
Semple, Ho, et al. (1997). Differences in the levels of stress between
caregivers and noncaregivers that were reported in different articles were
only coded once. If more than one measure of a dependent variable was
used (e.g., self-rated and clinician-rated depression), we report both effect
sizes. N
cg
⫽ number of caregivers; N
noncg
⫽ number of noncaregivers;
g
stress
/g
dep
/g
swb
/g
health
/g
self-efficacy
⫽ differences in perceived stress/depres
-
sion/subjective well-being/health/self-efficacy between caregivers and
noncaregivers; C.I. ⫽ confidence interval. (a) dementia caregivers; (b)
caregivers for frail elderly; (c) Parkinson caregivers; (d) caregivers for
cancer patients; (e) clinician-rated depression (Hamilton Rating scale); (f)
self-rated depression; (g) clinical interview-based diagnosis of depression;
(h) White caregivers; (i) Black caregivers; (j) life satisfaction; (k) positive
affect; (l) subjective health; (m) objective health; (n) male caregivers; (o)
female caregivers; (p) first time of measurement; (q) second time of
measurement; (r) third time of measurement; (s) spouses; (t) adult children;
(u) United States caregivers; (v) Chinese caregivers; (w) stroke caregivers;
(x) first study; (y) second study.
262
PINQUART AND SO
¨
RENSEN

Figure A1. Differences in stress between caregivers and noncaregivers (effect-sizes g, and 95% confidence
intervals).
(Appendix continues)
263
DIFFERENCES BETWEEN CAREGIVERS

Figure A2. Differences in depression between caregivers and noncaregivers (effect-sizes g, and 95% confi-
dence intervals).
264
PINQUART AND SO
¨
RENSEN

Figure A3. Differences in subjective well-being between caregivers and noncaregivers (effect-sizes g, and 95%
confidence intervals).
(Appendix continues)
265
DIFFERENCES BETWEEN CAREGIVERS

Figure A4. Differences in physical health between caregivers and noncaregivers (effect-sizes g, and 95%
confidence intervals).
266
PINQUART AND SO
¨
RENSEN

Received July 24, 2001
Revision received August 22, 2002
Accepted September 3, 2002 䡲
Figure A5. Differences in self-efficacy between caregivers and noncaregivers (effect-sizes g, and 95%
confidence intervals).
267
DIFFERENCES BETWEEN CAREGIVERS