Content uploaded by William E Haley
Author content
All content in this area was uploaded by William E Haley
Content may be subject to copyright.
Compound Caregiving:
When Lifelong Caregivers Undertake Additional Caregiving Roles
Elizabeth A. Perkins and William E. Haley
University of South Florida
Objective: Lifetime parental caregivers of adults with intellectual disabilities (ID) may also become
caregivers to other family members. This study investigated caregiver experiences of compound care-
giving (i.e. additional caregiving roles) and its association with caregiver quality of life. Participants:
Ninety-one older caregivers living with their adult son/daughter with ID were interviewed. Mean age of
the caregivers was 60 years and their sons’/daughters’ mean age was 29 years. Main Outcome Measures:
Compound caregiving status, physical and mental health, life satisfaction, depressive symptomatology,
and desire for alternative residential placement for their co-residing son/daughter. Results: Thirty-four
(37%) reported being current compound caregivers to an additional care recipient, predominantly a
mother, father, or spouse. Caregivers averaged 39 hours per week fulfilling their primary caregiving
tasks, an additional 12 hours per week on the compound caregiving role, and the median duration of
compound caregiving was 3 years. Compared with the non-compound caregivers, the compound
caregivers had increased desire to place their son/daughter into residential care, though no group
differences were apparent in life satisfaction, depressive symptomatology, physical health, or mental
health. The most problematic issues reported by compound caregivers were having little personal time
and a lack of adequate help from others. Conclusion: Compound caregiving was often experienced, and
may galvanize these lifetime caregivers to start making future plans for their sons/daughters. Future
research is warranted to refine more homogeneous groupings of compound caregivers, who may be at
greater risk for adverse outcomes.
Keywords: intellectual disabilities, developmental disabilities, aging, parents, family
Disability is increasingly recognized as a public health concern
(Lollar & Crews, 2003). This has resulted in increased awareness
of the importance of including persons with disabilities in popu-
lation-based assessments, and highlighting this population when
considering policy development, and assurance of services (Lollar
& Crews, 2003). Disability, however, is never experienced in a
vacuum; rather it is a product of how an individual with a disability
interacts with the biological, psychological, social, cultural, phys-
ical, and political environments (Stiers, 2008). Often, the primary
caregiver of an individual with a disability becomes a critical agent
between the individual and those previously stated environments.
Their contribution and assistance to the goal of rehabilitation and
healthcare professionals to maximize the biopsychosocial func-
tioning of the individual with a disability is substantial. Not sur-
prisingly, the contribution of caregivers and caregiver health out-
comes has become increasingly acknowledged as major public
health concern in of itself (Talley & Crews, 2007). Caregiving is
a diverse endeavor in terms of the ages, developmental levels,
mental health needs, and physical health needs of caregivers and
care recipients, and due to such diversity a greater emphasis of the
lifespan context in which caregiving occurs has recently been
advocated (Talley & Crews, 2007).
Older caregivers of adults with intellectual disabilities (ID) have
long been recognized as being a unique group of the caregiving
population (e.g. Heller & Factor, 1993). They face distinctive
circumstances and challenges from caregivers of persons that have
developed illnesses, or disabilities that have arisen from accident
or trauma. Challenges include the extended duration of their care-
giving role, lifespan health issues that arise from their own and
their care recipients’ often complex aging process, and anxiety for
the future well-being of their care recipient after they die or
discontinue their caregiving role (Haley & Perkins, 2004; Perkins
& Moran, 2010). However, there are several mitigating factors that
may positively impact this unique caregiving experience. The
normative nature of caring for one’s own child and feelings of
mastery and expertise are notable (Haley & Perkins, 2004). A
parental caregiver is a stable role dynamic that does not require the
role adjustments a new caregiver may face. Furthermore, lifelong
caregiving experience may result in the caregiver gaining skills
and self-efficacy in the caregiving role. The fact that many care-
Elizabeth A. Perkins, Department of Child and Family Studies, Univer-
sity of South Florida; and William E. Haley, School of Aging Studies,
University of South Florida.
This study was conducted in partial fulfillment of a doctoral dissertation
at the School of Aging Studies, University of South Florida. The authors
would like to acknowledge Matthew Janicki (University of Illinois at
Chicago), Cathy McEvoy, Sandra Reynolds, and Brent Small, (University
of South Florida) for their contributions as members of this doctoral
committee. The authors would also like to acknowledge their debt of
gratitude to all the caregivers that participated in this study.
Correspondence concerning this article should be addressed to Elizabeth
A. Perkins, PhD, RNMH, Florida Center for Inclusive Communities,
Department of Child and Family Studies, University of South Florida,
13301 Bruce B. Downs Blvd., MHC 2113A, Tampa, FL 33612. E-mail:
eperkins@mail.usf.edu
Rehabilitation Psychology
2010, Vol. 55, No. 4, 409– 417
© 2010 American Psychological Association
0090-5550/10/$12.00 DOI: 10.1037/a0021521
409
givers actively choose to continue to co-reside with their son/
daughter beyond a time when other residential alternatives are
available, attests to the possible reciprocal benefits derived in such
relationships (Heller, Miller, & Factor, 1997).
An aspect that could adversely affect well-being in older care-
givers is their experiences as sandwich caregivers. Sandwich care-
giving refers to those people (usually aged between 40 – 65 years)
who undertake caregiving responsibilities for one or both of their
aging parents while still having parenting duties to their own
children (Chisolm, 1999). Some of the major factors that generally
make simultaneous parenting/caregiving an issue is the financial
drain of a child, errand running, and maintaining responsibilities
for non-coresiding children due to divorce or school attendance
(Loomis & Booth, 1995). These circumstances may actually not be
too taxing in many cases, and, indeed, there is considerable con-
troversy between researchers as to whether the sandwich caregiv-
ing generation is really as “beleaguered” or frantically overloaded
as previously thought (Aldwin & Levenson, 2001; Loomis &
Booth, 1995).
Conversely, there has been more recent speculation that sand-
wich caregiving is a common and sometimes difficult issue, and
the potential duration of being a sandwich caregiver is increasing
due to the combined effect of baby boomers having delayed
child-rearing and their own parents’ increased life expectancy
(Rogerson & Kim, 2005). Rogerson and Kim also state that baby
boomers are actually having a “stretched period of caregiving”
compared with their own parents, and the same factors (delayed
child-rearing and increased life expectancies) will also result in the
children of baby boomers being sandwich caregivers for even
longer.
Though there remains argument regarding the relative impact of
sandwich caregiving overall, one scenario that has not been ade-
quately explored is unique to caregivers of adults with ID. Com-
pound caregivers are those parents who are already providing
considerable caregiving responsibilities for their son/daughter with
ID, who subsequently becomes a caregiver for an additional family
member (Perkins, 2010).
What makes these compound caregivers significantly different
from traditional “sandwich caregivers” is that they are already
lifelong caregivers when the additional caregiving role is adopted.
Furthermore, a recent case study has highlighted that periods of
compound caregiving may occur several times, and may occur not
just for parents, but parents-in-law, and siblings, in some cases
(Perkins, 2010). The same case study also reports that such periods
can be extremely stressful, partly because many difficult decisions
about prioritizing caregiving tasks had to be made. Although we
have found no previous empirical studies of the association of
compound caregiving with caregiver quality of life, it seems likely
that additional primary stressors and role demands could lead to
problems such as role overload (Pearlin, Mullin, Semple, & Skaff,
1990), and diminish the caregiver’s quality of life.
Previous caregiving studies have, understandably, focused on
the caregiving experience and tasks in terms of dyadic relation-
ships between caregiver and a single care recipient. There has been
little attention paid to assess whether the primary caregiver had
responsibility for other care recipients either concurrently, or in the
past, or indeed, whether they are anticipating a likelihood of
additional care recipients in the future. This previous omission,
though understandable, deserves further exploration especially as
compound caregiving is likely to be an increasingly common
scenario for caregivers of adults with ID, given the increased life
expectancy of persons with ID, and the general population. Per-
sons with mild ID are rapidly approaching parity with the general
population, and individuals with moderate/severe ID can expect to
live into their late 60s and late 50s respectively (Bittles et al.,
2002). In previous years parents routinely survived their adult
children with ID whereas now the reverse is more likely to occur,
and an increasingly aging population of adults with intellectual
disabilities will now routinely survive their parents. Therefore,
many caregivers in these circumstances are destined to be care-
givers for their entire lifetime. Todd and Shearn (1996) refer to
them as “perpetual parents” in recognition of their lifespan care-
giving commitment. As parental caregivers continue to provide the
majority of care to adults with ID (Braddock, Hemp, & Rizzolo,
2008), investigation of the potential experience (with increasing
age) of being a caregiver to other family members is pertinent.
The purpose of the present study was two-fold. Our first goal
was to conduct a preliminary estimate on the frequency of com-
pound caregiving in a sample of older caregivers of adults with ID.
We then investigated whether compound caregivers differ in qual-
ity of life outcomes, including depression, life satisfaction, phys-
ical and mental well-being, as well as in the desire to place their
relative with ID into an alternative residential facility. We hypoth-
esized that the additional responsibilities of caregiving for a sec-
ond care recipient would be associated with poorer quality of life
outcomes in terms of poorer physical health, poorer mental health,
lower life satisfaction, increased levels of depressive symptom-
atology. We also hypothesized that the greater desire for alterna-
tive residential placement may arise from the additional caregiving
responsibilities, by prompting feelings of role overload and in-
creasing a caregiver’s awareness of the possibility of succumbing
to illness themselves.
Methods
The present study was a cross-sectional design using a conve-
nience sample. Data was collected during semi-structured inter-
views utilizing survey instruments. Caregivers were also invited to
share additional comments regarding their caregiving experiences.
Participants were interviewed in their own home if they were
located in the Tampa Bay area, otherwise the interview was
conducted by telephone.
Participants
The study used a convenience sample of parent caregivers, since
it is extremely difficult to gather representative samples as many
aging caregivers are unlikely to be known to disability support
agencies (Janicki, McCallion, Force, Bishop, & Lepore, 1998). For
entry into the study parental caregivers were aged 50⫹years, their
care recipient (son/daughter) with ID was 18⫹years, and they had
continuously resided together. The research protocol was approved
by the Institutional Review Board at the University of South
Florida. Participants were recruited by a variety of approaches
including distribution of flyers to parent support groups, provider
agencies of habilitation/employment services, postings to list-
serves and websites, and parent-to-parent referral. The special
410 PERKINS AND HALEY
interest in compound caregiving was not mentioned in the flyers or
throughout the recruitment process.
Demographic and Independent Variables
The following variables were included to provide detailed de-
scriptive and demographic information of the caregivers, and their
care recipients, and to enable thorough comparison of sample
characteristics between caregiver and compound caregiver groups.
Caregiver Demographic and Descriptive Information
Each participant was asked to provide details regarding their age
(in years), gender, income (combined household income), educa-
tion (years in education), ethnicity, and current comorbidities.
Comorbidity in the caregiver was assessed using the self-report
comorbidity scale developed by Bayliss, Ellis, and Steiner (2005).
This scale has a checklist of 23 items that details chronic medical
conditions and determines both the presence of the condition, as
well as a 5-point Likert-type scale to report the level of interfer-
ence each of these conditions had on regular daily activities. The
participants were specifically asked whether a medical doctor has
ever diagnosed them with each particular health condition. Exam-
ples include congestive heart disease, diabetes, osteoarthritis, os-
teoporosis, cancer, and stroke. A total count of comorbid condi-
tions (Cronbach’s alpha was .63), as well as comorbidity
interference scores was obtained (Cronbach’s alpha was .61).
Care Recipient Demographic and
Descriptive Information
Demographic details and characteristics of the care recipient
were also obtained from the caregiver. Age, gender, intellectual
disability level (i.e. mild, moderate, severe, and profound), and ID
etiology were noted. Challenging behaviors were recorded using
the 8-item Maladaptive Behavior subscale, of the Scales of Inde-
pendent Behavior-Revised (Bruininks, Woodcock, Weatherman,
& Hill, 1996). For this study, the total number of challenging
behaviors present was noted, as was the total sum score of the
severity of each behavior as a problem specifically perceived by
the caregiver. Cronbach’s alpha for the total sum of challenging
behaviors was .80 and .90 for the severity score. Functional ability
of the care recipient was measured using the Activities of Daily
Living (Katz, Ford, & Moskowitz, 1963), and Instrumental Activ-
ities of Daily Living scales (Lawton & Brody, 1969). Activities of
Daily Living (Katz, Ford, & Moskowitz, 1963) were assessed by
6 items with yes/no responses to questions that determine whether
functions such as bathing, dressing, and feeding oneself, are able
to be done independently. Yes responses score 1 point, no re-
sponses score zero. The total score can range from 0 to 6, with
lower scores indicating that greater assistance is required with
activities of daily living. Cronbach’s alpha was .92 in this study.
The Instrumental Activities of Daily Living scale (Lawton &
Brody, 1969) is comprised of 8 items that assessed the level of
independence in a variety of areas including use of telephone,
shopping, using transport, ability to manage finances etc. The total
scores can range from 0 to 8, with a lower score indicating less
independence. Cronbach’s alpha for IADL’s was .68. Weekly
number of caregiving hours was calculated by including all tasks
and activities that the caregiver needed to do in support of their
son/daughter with ID. This included assistance with personal hy-
giene, toileting, preparing meals, supervising those who were
unable to be left alone in the household, providing transport to
school, competitive or supportive work environments, providing
transport to activities, and running errands, etc.
Compound Caregiver Status
Current compound caregiving status was determined by care-
giver response to the question “Do you currently have caregiving
tasks and responsibilities to another family member, other than
your child?” This definition of compound caregiving was purpose-
fully broad, as to capture the full range of possible multiple
caregiving roles that may be undertaken. The relationship of the
compound caregiving recipient, as well as the duration and weekly
hours spent in this role was noted. The major health issue and
previous history of compound caregiving was recorded. Caregivers
were also asked if they anticipated any caregiving responsibilities
for a specific family member in the future. Compound caregivers
were also asked to report the level of difficulty they had in a list of
eight potential problems that they might encounter due to manag-
ing multiple caregiving responsibilities. Problems included “lack
of adequate help from others,” “unable to do the things that I
enjoy,” and “prioritizing caregiving demands.” Content validity of
the items were demonstrated by their derivation from issues pre-
viously identified in a case study on compound caregiving (Per-
kins, 2010). Caregivers were asked to rate each problem from 1 to
5, with 1 representing “not a problem” through to 5 representing “a
very serious problem.” Cronbach’s alpha was .86.
Outcome Variables
Four variables were used as outcome measures to assess care-
giver quality of life; global physical health, global mental health,
depressive symptomatology, and life satisfaction. In addition, we
assessed the caregiver’s personal desire to seek placement of their
son/daughter with ID into an alternative residential setting.
Global Physical and Mental Health
Overall physical and mental health of the caregiver was mea-
sured using the Medical Outcomes Study-Short Form (SF-36).
This is a 36-item instrument and has been widely validated as a
summary measure of health-related quality of life (Ware & Sher-
bourne, 1992). The SF-36 has two major subscales, the physical
component summary (PCS) and mental component summary
(MCS), which are the scales used to assess global physical health,
and global mental health in this study. Scores are reported in a
standardized t-score metric, i.e. M⫽50, standard deviation ⫾10.
For both PCS and MCS, a mean of 50 is interpreted as average
health status for the overall U.S. population. Cronbach’s alpha for
the PCS was .91, and .87 for the MCS.
Depressive Symptomatology
Depressive symptomatology was assessed by the 20-item ver-
sion of the Center for Epidemiologic Studies-Depression Scale
411
COMPOUND CAREGIVING
(Radloff, 1977). Participants were asked to report how often var-
ious feelings or behaviors were experienced during the past week,
and their responses were rated on a 4-point likert scale. The
categories are presented in the following order: “rarely or none of
the time,” “some or a little of the time,” “much of the time,” and
“most or all of the time.” Scores can range from 0 (indicating no
depressive symptoms) to 60 (indicating severe depressive symp-
toms). Scores of 16 or higher on the Center for Epidemiologic
Studies-Depression Scale (CES-D) are typically viewed as clini-
cally significant and evidence of probable depression (Andresen,
Malmgren, Carter, & Patrick, 1994). Cronbach’s alpha was .89 in
the present study.
Life Satisfaction
The Life Satisfaction Index–Zisthe13-item short form version
of a measure designed to indicate levels of self-perceived morale
and general life satisfaction (Wood, Wylie, & Schaefor, 1969).
Participants were asked whether they agree, disagree or unsure one
way or the other, to statements including “I am just as happy as
when I was younger” and “Most of the things I do are boring or
monotonous.” Scores range from 0 –26 with higher scores indicat-
ing greater life satisfaction. Cronbach’s alpha in the present study
was .77.
Desire for Residential Placement
This variable was an adaptation of the Desire-to-Institutionalize
Scale by Morycz (1985). Caregivers were asked “In respect to
seeking alternative residential placement for your son/daughter,
which of the following best describes you.” A continuum of
responses ask the caregiver the level to which they have either
considered, discussed with their son/daughter the possibility, or
with other family members, through to actively seeking it, and
making steps to finding alternative residential placement. Scores
range from 1– 6, with 6 indicating greatest desire to place. In order
to make the scale suitable and sensitive to care recipients with ID,
the wording of items was altered so as to not cause potential
offense to caregivers (e.g. “institutionalization” was replaced with
alternative residential accommodation, “patient” with son/daugh-
ter).
Data Analysis
All variables were checked for normality of distribution using
skewness and kurtosis indicators, and all variables were found to
be in acceptable range, and required no transformation. There were
no missing data in any of the variables utilized in this study.
Descriptive analyses were conducted of all independent and de-
pendent variables. Independent sample ttests were conducted to
compare sample characteristics of current compound caregivers
with noncompound caregivers. For categorical variables, Pear-
son’s chi-square tests were used. To specifically assess quality of
life differences between the two groups, independent sample ttests
were conducted in global health functioning (MCS/PCS), depres-
sive symptomatology, life satisfaction, and desire for residential
placement. An alpha level of .05 was used for all tests of statistical
significance.
Results
Caregiver Characteristics
A total of 91 caregivers participated in the study. As shown in
Table 1, the caregivers averaged just over 60 years of age but there
was considerable range. The majority of the sample (78 caregiv-
ers), were recruited from Florida. Due to using internet sources,
caregivers from other states also participated, including five from
New Jersey, four from Georgia, and one each from Maryland,
Oklahoma, South Dakota, and Nevada. The caregivers were pre-
dominately White and female. Those who identified in the non-
white category included four caregivers who were African Amer-
ican/Black, two who were Hispanic/Latino, and one Pacific
Islander. Overall, the caregivers were highly educated, and quite
affluent.
With respect to compound caregiving status, over a third of the
sample was currently undertaking compound caregiving duties.
Furthermore, two thirds were either presently or previously a
compound caregiver. Regarding the possibility of becoming a
future compound caregiver for a specified family member (regard-
less of their present compound caregiving status), approximately
one third responded that this was a likely scenario.
Care Recipient Characteristics
There were 55 male and 36 female care recipients, and their
mean age was 29 years, with considerable variation. Primary
etiologies of intellectual disability were Down syndrome (28),
unknown etiology (18), autism spectrum disorders (15), others
(14), cerebral palsy (12), fragile X (4). Within the category of
“others” there were two individuals with tuberous sclerosis, one
care recipient with Prader-Willi syndrome, and another with Wil-
liams syndrome. Overall, this was quite a diverse sample in terms
of etiology. The majority of care recipients were individuals with
moderate intellectual disabilities; 27 were individuals with mild
ID, 44 with moderate ID, 12 with severe ID, and eight with
profound ID. The means, standard deviations, and ranges for all
care recipient demographic details are also detailed in Table 1.
Compound Caregiver Characteristics
Table 2 presents the major types of relationships between com-
pound caregivers and their care recipients. The largest categories
of compound caregiving recipients were for mothers, fathers, and
spouses. Table 3 also presents the major health issue of the
compound caregiving care recipient, the most common being
Alzheimer’s disease.
In terms of duration of caregiving, there were several compound
caregivers in the sample that had been compound caregivers for
many years. These caregivers were either caring for a second child
with intellectual disability (3), a sibling with intellectual disability
(1), or a sibling with a chronic mental disorder (2). These outliers
inflated the mean duration of caregiving to 77 months, with a range
from 3 – 504 months. However, the median duration was 36
months, and the modal response was 24 months.
412 PERKINS AND HALEY
Comparisons Between Compound and
Noncompound Caregivers
Using independent sample t tests (see Table 3), and Pearson’s
Chi square analyses, compound caregivers’ demographic vari-
ables, caregiving variables, and care recipient characteristics, were
compared with noncompound caregivers. Overall, compound care-
givers were very similar to their noncompound caregiving coun-
terparts, with three exceptions; combined caregiving hours, care-
giver ethnicity, and care recipient gender. With respect to the time
spent in their primary caregiving role, there was no significant
difference (⬃39 hours for both groups), but when the additional
compound caregiving hours were added, compound caregivers
weekly tally increased to around 52 hours which was significantly
greater than noncompound caregivers ⬃39 hours, t(89) ⫽
⫺2.30, p⬍.05, Cohen’s dwas 0.49, indicating a moderately
large effect (Cohen, 1988). The compound caregivers were all
white and significantly differed from noncompound caregivers
2
(1, N⫽91) ⫽5.23, p⬍.05. Regarding gender of their
primary care recipients (i.e. son/daughter with ID), the gender
distribution of compound care recipients was fairly equal, but
for noncompound caregivers, caregiving for their sons predom-
inated
2
(1, N⫽91) ⫽4.06, p⬍.05. Chi square analyses of
caregiver gender and income, revealed no significant differ-
ences between compound caregivers and noncompound care-
givers.
All variables were also analyzed by method of interview (i.e.
telephone versus in-person) to investigate whether the two groups
of participants differed on major study variables. Only education
was significantly higher in the telephone versus the in-person
interview participants and desire to place was higher in telephone
participants. However, group differences between the telephone
and in-person interviews were no longer significant on desire to
place when an ANCOVA using education as the covariate was
conducted.
Table 2
Compound Caregivers’ Care Recipient Relationship and Major
Health Issue (N⫽34)
N%
Relationship
Mother 13 (38.22)
Father 4 (11.76)
Spouse 4 (11.76)
Sibling 3 (8.82)
Aunt/uncle 3 (8.82)
2nd Child with intellectual disability 3 (8.82)
Mother-in-law 2 (5.88)
Grandchild with medical needs 1 (2.94)
Major health issue
Alzheimer’s disease 7 (20.58)
Elderly frail 4 (11.76)
Advanced macular degeneration 4 (11.76)
Cardiovascular disease 4 (11.76)
Intellectual disability 4 (11.76)
Parkinson’s disease 2 (5.88)
Cancer 2 (5.88)
Chronic mental disorder 2 (5.88)
Hip fracture/replacement 2 (5.88)
Stroke 1 (2.94)
Diabetes 1 (2.94)
Postoperative convalescence 1 (2.94)
Table 1
Sample Caregiver and Care Recipient Characteristics (N⫽91)
Mor % SD Range
Caregiver characteristics
Demographic
Age (years) 60.8 8.5 50–92
Education (years) 15.1 2.4 12–22
Gender (Female) 91%
Race/ethnicity (White) 92%
Household income (⬎$50,000) 78%
Health
Total comorbidities 5.1 2.9 0–13
Comorbidity interference 9.2 7.9 0–35
Caregiving
Total caregiving hours per week 39.4 21.3 7–88
Compound caregiver now (Yes) 37%
Compound caregiver ever (Yes) 66%
Anticipated future caregiving (Yes) 34%
Care recipient characteristics
Demographic
Age 29.7 9.5 18–54
Gender (Female) 40%
Health and functional abilities
Activities of daily living (ADL) 4.6 2.1 0–6
Instrumental ADLs 2.3 1.7 0–6
Intellectual disability level 2.0 0.8 1–4
Challenging behaviors (total) 2.9 2.5 0–8
Challenging behaviors (severity)
a
6.1 6.0 0–26
a
N⫽63.
413
COMPOUND CAREGIVING
Descriptive variables on caregiver quality of life and desire to
place are presented in Table 4. Compound caregivers did not
statistically differ from noncompound caregivers in four of the
outcome measures; life satisfaction, depressive symptomatology,
physical and mental health. However, their desire to place their
primary care recipient with intellectual disability was significantly
higher than their noncompound caregiving counterparts, t(89) ⫽
⫺3.11, p⬍.01, Cohen’s dwas 0.67, a moderately large effect
(Cohen, 1988). The top 2 items of the scale, i.e. “I am likely to
seek residential placement,” or “I have made actual steps towards
placement,” revealed that 32% of compound caregivers had re-
sponded that they were currently at that level, compared with only
10% of the noncompound caregivers.
With respect to managing multiple caregiving roles, the eight
items are ranked in order of mean scores (1–5) as reported by the
current compound caregivers, and are presented in Table 5. Care-
givers identified lack of time to pursue personal interests, inade-
quate help from others, feeling stressed and emotionally drained,
and being physically fatigued, as being the most problematic
issues, from the list presented.
Discussion
In this preliminary investigation of compound caregiving activ-
ities in lifelong parental caregivers, we found that 37% of the
caregivers in our sample reported being current compound care-
givers. Study recruitment materials did not specifically mention
this particular aspect of the study, i.e. that there would be questions
regarding caregiving for others in addition to their primary care-
giving role, so we do not believe that our sampling method had any
clear bias in oversampling compound caregivers. This figure is
consistent with a recent National Alliance for Caregiving (NAC)/
AARP survey finding that 34% of general caregivers were taking
care of more than one person (NAC/AARP, 2009). We also found
that 66% were either a current or previous compound caregiver,
and irrespective of their current compound caregiving duties, 34%
Table 3
Demographic and Descriptive Variables in Compound Versus Noncompound Caregivers
Compound caregiver
(N⫽34)
Noncompound caregiver
(N⫽57)
Mor % SD M or % SD
Caregiver characteristics
Demographic
Age (years) 58.8 7.9 61.1 8.9
Education 14.7 2.29 15.4 2.53
Gender (Female) 91% 91%
Race/ethnicity (White) 100%
ⴱ
86%
Household income (⬎$50,000) 82% 75%
Health and caregiving
Total comorbidities 4.79 2.96 5.42 3.01
Comorbidity interference 8.68 8.66 9.52 7.47
Total caregiving hours 38.66 20.82 39.84 21.89
Caregiving hours ⫹CCG hours
a
51.60
ⴱ
26.34 39.84 21.88
Care recipient characteristics
Demographic
Age 28.38 8.91 30.51 9.80
Gender (Male) 47%
ⴱ
68%
Health and functional abilities
Activities of daily living (ADLs) 4.88 1.93 4.47 2.12
Instrumental ADLs (IADLs) 2.64 1.91 2.21 1.52
Intellectual disability level 1.94 .91 2.05 1.52
Challenging behaviors (total) 3.44 2.57 2.68 2.42
Challenging behavior (severity) 5.62 6.73 6.47 5.62
a
CCG ⫽Compound caregiving.
ⴱ
p⬍.05 (2-tailed).
Table 4
Comparison of Compound Versus Noncompound Caregivers in Quality of Life and Desire to Place
Compound caregiver (N⫽34) Noncompound caregiver (N⫽57)
MSD Range MSD Range
Life satisfaction (Life Satisfaction Index) 17.05 5.81 3–24 17.58 7.03 3–26
Depressive symptomatology (CES-D) 10.94 9.30 0–40 12.61 9.65 0–40
Physical health (PCS of SF-36) 44.51 11.28 19.00–65.89 43.09 10.07 17.07–60.42
Mental health (MCS of SF-36) 47.66 11.22 17.16–63.40 48.67 11.53 18.14–64.48
Desire to place 3.61
ⴱⴱ
1.72 1–6 2.49 1.63 1–6
ⴱⴱ
p⬍.01 (2-tailed).
414 PERKINS AND HALEY
also anticipated future caregiving responsibilities for a specified
family member. Overall, it appears that many parental caregivers,
if not currently a compound caregiver, either have been a com-
pound caregiver in the past, or is anticipating becoming one at
some future date.
The second research question was to determine whether com-
pound caregivers had poorer quality of life outcomes and a greater
desire to place compared with their noncompound caregiving
counterparts. Contrary to our predictions, in the present study,
compound caregiving was not found to be associated with poorer
outcomes in terms of life satisfaction, depressive symptomatology,
physical or mental health. Furthermore, the two groups were very
similar regarding primary caregiver characteristics, and the char-
acteristics of their primary caregiving recipients. No differences
were evident in their ages, comorbid health status, or the level of
daily interference their comorbid health issues presented. It is also
important to note that differences between the primary care recip-
ients’ characteristics were not found. This suggests compound
caregivers did not become compound caregivers merely because
their primary caregiving duties were less time consuming or the
characteristics of their son/daughter were significantly different.
We found no evidence that compound caregiving had a major
impact on the well being of the caregiver.
There may be several explanations why compound caregivers’
physical and mental health, depressive symptomatology, or life
satisfaction did not differ from the noncompound caregivers. Older
caregivers have a lifetime of caregiving experience, and become
experts at their caregiving role (Haley & Perkins, 2004). There-
fore, an additional caregiving role may not be too onerous and may
actually be a straightforward adaptation for these highly experi-
enced caregivers, and in some much welcomed. A lifetime of
caring for an individual with special needs may certainly better
equip the compound caregivers with the knowledge, skills, and
empathy required to do so. In addition, other researchers often find
that, once caregivers have gone past a threshold of caregiving
duties, additional hours of caregiving are not associated with
greater caregiver distress. For example, Roth, Perkins, Wadley,
Temple, and Haley (2009), in a study of over 5,000 caregivers,
found that additional caregiving hours beyond a 20 hour per week
threshold were not associated with reduced caregiver quality of life
or increased caregiver depression.
Our hypothesis that compound caregivers were more likely to
have a greater desire to seek alternative residential placement for
their primary care recipient was supported. Many more compound
caregivers endorsed the items of “likely to seek” or “have made
actual steps” towards alternative residential placement compared
with the noncompound caregivers. The very act of being a com-
pound caregiver may have given these parents the urgency of
addressing “what will happen to my son/daughter if I get sick”
type of scenarios. It could also be argued that they are experiencing
more strain, which have not yet manifested in the other quality of
life outcomes, but has resulted in their being more willing to cease
their primary caregiving duties in acknowledgement of their own
mortality. The majority of compound caregivers were looking after
their parents, spouses, or in laws with chronic health conditions, so
such questions may be more pressing in their own mind in recog-
nition of their own aging process. Therefore compound caregiving
status may galvanize the discussion and prompt the decision mak-
ing process about the future residential arrangements of the pri-
mary care recipient. Many families are reluctant to make future
plans and have not determined with their son/daughter with ID
about where and with whom they may wish to reside, and who
would become their next primary caregiver (Heller, Caldwell, &
Factor, 2007). In one notable study less than 50% of aging care-
givers had made such arrangements (Freedman, Krauss, & Seltzer,
1997). Compound caregiving may be a very beneficial catalyst to
solidify such details not from increased stress but from increased
impetus to finalize such arrangements.
Compound caregivers in this study did report difficulties in
managing their multiple caregiving tasks. One compound care-
giver stated, “My biggest problem is how do I integrate my son
into all the demands of my roles.” Another acknowledged palpable
guilt and remarked, “I feel bad that my time is taken up with my
many caregiving duties—it stops me from being able to encourage
my son to do more.” Another issue that arose was the effect of the
compound care recipients’ illness upon their adult child with ID,
“My daughter gets very upset at seeing the aging process of her
grandparents. She doesn’t comprehend the hallucinations her
grandfather has due to dementia.” Clearly, compound caregiving
when the primary care recipient has intellectual disabilities, can
create additional complexities in the caregiver’s role.
Greater attention to compound caregivers who have had multi-
ple episodes of compound caregiving may be warranted. In a
recent case study, the caregiver had four separate episodes of
compound caregiving duties (Perkins, 2010), with the fourth care-
giving episode resulting in considerable distress. Recalling how
she felt during the latest episode, the caregiver stated “When I
think back now, I realize that I was apprehensive about acknowl-
edging I needed help, let alone asking for it.”
In the present study, questions were asked regarding the current,
Table 5
Problematic Issues for Compound Caregivers (N⫽34)
Problem Mean Range
Unable to have the time do things that I enjoy (e.g., hobbies, social events, vacations etc). 3.44 1–5
Lack of adequate help from others. 2.94 1–5
Feeling stressed and emotionally drained. 2.94 1–5
Being physically tired. 2.85 1–5
Prioritizing caregiving demands (i.e., managing what tasks you need to do, and when you can do it). 2.41 1–5
Finding someone to care for my son/daughter when I’m caring for the other care recipient. 2.00 1–5
Unable to maintain my own job. 2.00 1–5
Having to care for the very person who used to help me look after my own son/daughter with intellectual disabilities. 1.94 1–5
Note. Responses were on a 5-point Likert scale: 1 ⫽not a problem to 5 ⫽a very serious problem.
415
COMPOUND CAREGIVING
and up to two previous episodes of compound caregiving. The
sample actually had six participants that were currently compound
caregivers, and who had on two previous occasions been com-
pound caregivers to other family members in the past. There were
also five triple caregivers, i.e. had caregiving duties to their son/
daughter with intellectual disabilities, and two other family mem-
bers concurrently. Some caregivers also had more than one child
with intellectual disabilities.
The present study has a number of limitations that need to be
acknowledged. Firstly, this was not a random sample, and the
convenience sampling methodology can lead to biases in the
results. It may be that caregivers willing to participate were health-
ier, better educated and affluent, and more knowledgeable and
resourceful than caregivers in general. Significant differences in
compound caregiver wellbeing may emerge in a more ethnically
and economically diverse sample than was used in the present
study. The parent-to-parent referral that was used to recruit some
participants may also have biased the sample to include more
caregivers who greatly benefit from durable support networks with
other parents. However, since previous research has also indicated
that caregivers who are more depressed are likely to participate in
studies if they are offered the alternative of in-home interview
(Dura & Kiecolt-Glaser, 1990), we did make every effort to make
participation convenient with our method to reduce this form of
error. While random samples of aging caregivers of adults with ID
would be ideal, it is extremely difficult to accomplish, and in order
to increase sample size, we utilized two methods of interview
rather than restricting participation from willing caregivers from a
variety of locations. In addition, we recruited participants in two
very distinct methods: some via direct recruitment (with face to
face interviews), and others via internet recruitment (with tele-
phone interviews). We did so because of the difficulty of recruiting
enough participants via one method alone. However, we found
some minor differences in participants by type of recruitment and
assessment. Future research should standardize these factors if
possible.
Regarding ascertaining the frequency of compound caregiving,
we acknowledge that this question can only be definitively an-
swered by epidemiological data. However, our findings suggest
that compound caregiving deserves greater research attention, with
more ethnically and economically diverse samples of compound
caregivers, compound caregivers with multiple experiences with
compound caregiving, and compound caregivers with greater ho-
mogeneity of the compound care recipient (e.g. care recipients
with Alzheimer’s disease). Our project was also cross-sectional,
and longitudinal changes associated with compound caregiving
episodes would be optimal particularly as it would track caregivers
that change status between being a compound and noncompound
caregiver. This would allow investigation of whether residual
effects of the compound caregiving manifest after the compound
caregiving episode has ceased. Despite these limitations, our find-
ings provide preliminary evidence that compound caregiving roles
are undertaken by many caregivers, and thus warrants further
investigation. These caregivers seem to be rather resilient despite
the extra time and effort they devote to their multiple caregiving
roles. Resilient individuals are noted to be more flexible to chang-
ing demands (Tugade & Fredrickson, 2004), and exhibit greater
emotional stability during times of adversity (Bonanno, Papa, &
O’Neill, 2001), thus compound caregivers’ previous caregiving
experiences may have buffered their resiliency to weather their
current circumstances. Indeed, some caregivers may become com-
pound caregivers because they feel able and willing to adapt to a
new caregiving challenge.
Further research as to why compound caregivers are more likely
to seek placement is also warranted, and whether the level and type
of assistance they receive from others is a predictive factor. Over-
all, the issue of compound caregiving is one that faces many older
parental caregivers of adults with ID, and they may undertake a
compound caregiving role for several years, and on more than one
occasion. However, compound caregiving may also be an emerg-
ing issue for other groups of long-term caregivers, such as care-
givers of persons with chronic mental health issues, physical
disabilities, mobility issues, traumatic brain or spinal cord injury,
or stroke survivors, where similar challenges are also likely to
arise. Another interesting issue is the changing nature of the
caregiver/care recipient roles—with increasing age there may be
role reversals in these relationships. It is also possible the recip-
rocal benefits of the primary caregiver/care recipient relationship
are closely associated with compound caregiving duties. For ex-
ample, some primary care recipients (i.e. the son/daughter with
intellectual disabilities) may have a very active and supportive
caregiving role to the compound care recipient, and therefore
provides much appreciated assistance to the compound caregiver.
In conclusion, managing multiple caregiving roles does present
considerable challenges. This study emphasizes the need for care-
ful consideration of all caregiving duties and responsibilities, in
addition to the primary caregiving role, when caregiver support
and respite services are being planned. Compound caregivers are
certainly deserving of the attention of policy makers and service
planners as the time they devote to caregiving duties both in terms
of weekly hours, and the duration of the additional compound
caregiving role, is substantial. Ultimately, professionals endeavor
to assist a person with a disability to maximize their independence,
promote self-determination and assist that person to live the life
that they wish to lead. Caregivers are an integral part of this
process. Current caregiver supports, interventions, and service
provision requires a prompt reevaluation in order to recognize and
assess the needs of compound caregivers to ensure that a fair and
equitable distribution of their caregiving roles and responsibilities
is promoted.
References
Aldwin, C., & Levenson, M. (2001). Stress, coping, and health at midlife:
A developmental perspective. In M. Lachman (Ed.), Handbook of
midlife development (pp. 188 –216). New York: John Wiley.
Andresen, E. M., Malmgren, J. A., Carter, W. B., & Patrick, D. L. (1994).
Screening for depression in well older adults: Evaluation of a short form
of the CES-D. American Journal of Preventive Medicine, 10, 77– 84.
Bayliss, E. A., Ellis, J. L., & Steiner, J. F. (2005). Subjective assessments
of comorbidity correlate with quality of life health outcomes: Initial
validation of a comorbidity assessment instrument. Health and Quality
of Life Outcomes, 3, 51. doi:10.1186/1477-7525-3-51
Bittles, A. H., Petterson, B. A., Sullivan, S. G., Hussain, R., Glasson, E. J.,
& Montgomery, P. D. (2002). The influence of intellectual disability on
life expectancy. Journal of Gerontology: Medical Sciences, 57A,
M470 –M472. doi:10.1093/gerona/57.7.M470
Bonanno, G. A., Papa, A., & O’Neill, K. (2001). Loss and human resil-
416 PERKINS AND HALEY
ience. Applied and Preventive Psychology, 10, 193–206. doi:10.1016/
S0962-1849(01)80014-7
Braddock, D., Hemp, R., & Rizzolo, M. (2008). The state of the states in
developmental disabilities 2008. Denver, CO: University of Colorado,
Department of Psychiatry and Coleman Institute for Cognitive Disabil-
ities.
Bruininks, R. H., Woodcock, R. W., Weatherman, R. F., & Hill, B. K.
(1996). Scales of Independent Behavior-Revised. Itasca, IL: Riverside.
Chisholm, J. F. (1999). The sandwich generation, Journal of Social Dis-
tress and the Homeless, 8, 177–180. doi:10.1023/A:1021368826791
Cohen, J. (1988). Statistical power analysis for the behavioral sciences
(2nd ed.). Hillsdale, NJ: Erlbaum.
Dura, J. R., & Kiecolt-Glaser, J. K. (1990). Sample bias in caregiving
research. Journal of Gerontology: Psychological Sciences, 45, P200 –
P204. doi:10.1093/geronj/45.5.P200
Freedman, R. I., Krauss, M. W., & Seltzer, M. M. (1997). Aging parents’
residential plans for adult children with mental retardation. Mental
Retardation, 35, 114 –123.
Haley, W. E., & Perkins, E. A. (2004). Current status and future directions
in family caregiving and aging people with intellectual disability. Jour-
nal of Policy and Practice in Intellectual Disabilities, 1, 24 –30. doi:
10.1111/j.1741-1130.2004.04004.x
Heller, T., Caldwell, J., & Factor, A. (2007). Aging family caregivers:
Policies and practices. Mental Retardation and Developmental Disabil-
ities Research Reviews, 13, 136 –142. doi:10.1002/mrdd.20138
Heller, T., & Factor, A. (1993). Aging family care givers: Changes in
burden and placement desire. American Journal on Mental Retardation,
98, 417– 426.
Heller, T., Miller, A. B., & Factor, A. (1997). Adults with mental retar-
dation as supports to their parent: Effects on parental caregiving ap-
praisal. Mental Retardation, 35, 338 –346.
Janicki, M., McCallion, P., Force, L. T., Bishop, K., & Lepore, P. (1998).
Area agency on aging outreach and assistance for households with older
carers of an adult with a developmental disability. Journal of Aging &
Social Policy, 10, 1–24. doi:10.1300/J031v10n01_02
Katz, S., Ford, A. B., & Moskowitz, R. W. (1963). The index of ADL: A
standardized measure of biological and psychosocial function. Journal
of the American Medical Association, 185, 914 –919.
Lawton, M. P., & Brody, E. M. (1969). Assessment of older people:
Self-maintaining and instrumental activities of daily living. The Geron-
tologist, 9, 179 –186. doi:10.1093/geront/9.3_Part_1.179
Lollar, D. J., & Crews, J. E. (2003). Redefining the role of public health in
disability. Annual Review of Public Health, 24, 195–208. doi:10.1146/
annurev.publheath.24.100901.140844
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. doi:10.1177/019251395016002001
Morycz, R. K. (1985). Caregiving strain and the desire to institutionalize
family members with Alzheimer’s disease: Possible predictors and
model development. Research on Aging, 7, 329 –361. doi:10.1177/
0164027585007003002
National Alliance for Caregiving & AARP. (2009). Caregiving in the US -
executive summary. Retrieved from http://www.caregiving.org/data/
CaregivingUSAllAgesExecSum.pdf
Pearlin, L. I., Mullin, J. T., Semple, S. J., & Skaff, M. M. (1990).
Caregiving and the stress process: An overview of concepts and their
measures. The Gerontologist, 30, 583–594. doi:10.1093/geront/30.5.583
Perkins, E. A. (2010). The compound caregiver: A case study of multiple
caregiving roles. Clinical Gerontologist, 33, 248 –254.
Perkins, E. A., & Moran, J. A. (2010). Aging adults with intellectual
disabilities. Journal of the American Medical Association, 304, 91–92.
Radloff, L. (1977). The CES-D Scale: A self-report depression scale for
research in the general population. Applied Psychological Measurement,
1, 385– 401. doi:10.1177/014662167700100306
Rogerson, P. A., & Kim, D. (2005). Population distribution and redistri-
bution of the baby-boom cohort in the United States: Recent trends and
implications. Proceedings of the National Academy of Sciences, 102,
15319 –15324. doi:10.1073/pnas.0507318102
Roth, D. L., Perkins, M., Wadley, V. G., Temple, E., & Haley, W. E.
(2009). Family caregiving and emotional strain: Associations with psy-
chological health in a national sample of community- dwelling middle-
aged and older adults. Quality of Life Research, 18, 679 – 688.
Stiers, W. (2008). Message from the President. Retrieved from http://
www.div22.org/
Talley, R. C., & Crews, J. E. (2007). Framing the public health of
caregiving. American Journal of Public Health, 97, 224 –228. doi:
10.2105/AJPH.2004.059337
Todd, S., & Shearn, J. (1996). Time and the person: Impact of support
services on the lives of parents of adults with intellectual disabilities.
Journal of Applied Research in Intellectual Disabilities, 9, 40 – 60.
doi:10.1111/j.1468-3148.1996.tb00097.x
Tugade, M. M., & Fredrickson, B. L. (2004). Resilient individuals use
positive emotions to bounce back from negative emotional experiences.
Journal of Personality and Social Psychology, 86, 320 –333. doi:
10.1037/0022-3514.86.2.320
Ware, J. E., & Sherbourne, C. D. (1992). The MOS 36-item short-form
health survey (SF-36). I. Conceptual framework and item selection.
Medical Care, 30, 473– 83. doi:10.1097/00005650-199206000-00002
Wood, V., Wylie, M. L., & Schaefor, B. (1969). An analysis of a short
self-report measure of life satisfaction: Correlation with rater judgments.
Journal of Gerontology, 24, 465–9.
Received April 22, 2010
Revision received August 24, 2010
Accepted September 8, 2010 䡲
417
COMPOUND CAREGIVING
