Older Adults with Overlapping Caregiving Responsibilities and Care Needs in a U.S. National Community-Based Sample

Abstract

Background:

Community-dwelling older adults often serve as caregivers despite having their own health concerns and disabilities, yet little is known about their care needs.

Methods:

Cross-sectional analysis including community-dwelling U.S. adults over age 60 years who self-identified as caregivers in the National Social Life, Health and Aging Project in 2015–2016. Caregiving was defined by self-reported assistance of another adult with day-to-day activities due to age or disability; overlapping care-receiving was defined by simultaneous receipt of help for at least one activity of daily living (ADL) or independent ADL (IADL). Multivariable logistic regression models examined attributes associated with overlapping care-receiving among older caregivers, adjusted for caregiver characteristics (age, gender, spousal caregiving, self-reported physical and mental health, and cognitive function).

Results:

Among the 444 caregivers, the mean age was 67.8 (SD 0.29) years, 55.8% were women, 78.1% were non-Hispanic White, 54.7% self-identified as primary caregivers, and 30.7% were caring for a spouse. Thirty two percent of older caregivers were caregiving while themselves receiving assistance with at least one ADL or IADL. Thirty four percent of caregivers reported <$50,000 in household assets and 10% did not answer the question. Given prior research that supports that most nonrespondents fall into the low-income group, subjects were combined. Analyses with and without nonrespondents did not substantially change the results. Compared to caregivers who were not simultaneously receiving care, caregivers reporting overlapping care-receiving had greater odds of being older (AOR 1.30, 95% confidence interval (CI) [1.14, 1.48] per each 5-year age increase), caregiving for a spouse (AOR 1.93, 95% CI [1.20, 3.13]), having limited household assets (AOR 2.10, 95% CI [1.17, 3.80], for <$50,000 compared to ≥$50,000), and having poor or fair self-reported physical health (AOR 2.94, 95% CI [1.43, 6.02]).

Conclusions:

Over 30% of older adult caregivers report simultaneously receiving care for their own daily activities. Older caregivers who receive care are more likely to be older, spousal caregivers, and have limited assets and worse physical health. Targeted strategies are needed to support older caregivers who are uniquely vulnerable due to their overlapping care needs.

Introduction

Currently 48 million Americans are caregivers, up from 40 million in 2015—a rise primarily driven by a significant increase in caregiving for older adults.[1, 2] Caregivers fulfill a critical role in enabling older adults to remain in the community and age at home.[3] At the same time, older adults themselves comprise a significant proportion of individuals in caregiving roles, with nearly one in five U.S. caregivers being age 65 or older.[1]

Compared to their younger counterparts, older caregivers report higher levels of physical strain and burden related to caregiving.[4, 5] Moreover, for older caregivers, caregiving is often complicated by the simultaneous challenge of managing their own health care needs. Prior research has shown that over half of older caregivers report having two or more chronic conditions, and one third report experiencing a physical or mental disability.[6, 7] Given the demands of their role, older caregivers may neglect their own health needs, placing themselves at risk for poor health outcomes and care recipients at risk for having unmet care needs.[8–10]

Studies of caregiving often presume that caregivers are functionally independent; however, older caregivers may have some degree of functional impairment even as they fulfill caregiving roles. The existing literature has neglected the extent to which older caregivers are care recipients themselves, who need or receive care for activities of daily living or independent activities of daily living. We examined the understudied area of overlapping caregiving and care receiving status among a national community-based sample of older adults. In doing so, we aim to disrupt the frequently siloed classification of “caregivers” and “care recipients” to instead characterize older adults who serve in these roles simultaneously.

Methods

Setting and Participants

A cross-sectional data analysis was conducted using data from the National Social Life, Health and Aging Project (NSHAP). NSHAP is a longitudinal, population-based cohort study that aims to understand health and social factors impacting the well-being of older community dwelling adults living in the United States.[11] The first round of NSHAP data collection (2005–2006) involved multistage probability sampling of U.S. adults aged 57 to 85 years; Round 2 (2010–2011) involved surviving Round 1 respondents and a sample of their co-residing spouses/partners. Round 3 (2015–2016) involved surviving Round 1 and 2 participants along with newly recruited participants from the Baby Boom cohort, with their co-residing spouses/partners. This analysis focused on NSHAP Round 3 participants aged 60 or older, who self-identified as caregivers and answered questions related to care-receiving status from 2015 to 2016.[12]

Caregiving and Care-Receiving Measures

Caregiving status was assessed by asking participants, “Are you currently assisting an adult who needs help with day-to-day activities due to age or disability?” Participants were then asked about their relationship to the care recipient and whether they were the primary caregiver (e.g., the primary person responsible for assisting with day-to-day activities).

NSHAP interviewers asked whether participants experienced difficulty with performing any of seven activities of daily living (ADLs) and any of seven independent activities of daily living (IADLs), excluding any difficulties expected to last less than three months. Participants rated the degree of difficulty with each activity as “no difficulty,” “some difficulty,” “much difficulty,” or “unable to do,” and participants who reported having at least “some difficulty” for any activity were then asked whether anyone ever helped them with the activity. Overlapping caregivers were defined as caregiving participants who responded yes to receiving help with at least one ADL or at least one IADL.

Caregiver Sociodemographic and Clinical Characteristics

Participants self-reported sociodemographic characteristics including age, gender, race, Hispanic ethnicity, marital status, educational attainment, and household assets. Overall physical and mental health were self-reported using a single-item validated measure asking respondents to rate each type of health on a five-point scale from “excellent” to “poor;” responses were dichotomized as “poor or fair” vs. “good, very good, or excellent” for these analyses.[11] Participants’ cognitive function was assessed using an 18-item survey adaptation of the Montreal Cognitive Assessment (MoCA-SA), in which respondents having MoCA-SA scores higher than 22 are considered to have a normal cognitive assessment, scores 18–22 are consistent with mild cognitive impairment, and scores less than 18 indicate a positive screen for dementia.[13, 14]

Statistical Analysis

Descriptive statistics were used to summarize sociodemographic, clinical, and caregiving-specific characteristics for caregivers in the sample. Chi-square and t-tests were used to compare the characteristics of caregivers who reported overlapping care-receiving versus those not receiving care. All analyses incorporated NSHAP-recommended survey design variables and analytic survey weights to account for differential nonresponse and the complex sampling design.[12] Multivariable logistic regression models were developed to examine associations between sociodemographic, clinical, and caregiving-specific characteristics and overlapping caregiving status. Covariates were included in the model based on strength of association and significance (p <0.05) in bivariate analyses (age, assets, caregiving for a spouse, overall physical health, overall mental health, and MoCA-SA score) as well as a priori (gender) based on likelihood of association with overlapping caregiving. Participants who were missing household asset data (N=45) were grouped with participants reporting <$50,000 in household assets, based on prior NSHAP research indicating that individuals with lower assets had higher non-response rates than those with higher assets.[15] For all models, only participants who responded to questions regarding care received were included. A two-sided p value < 0.05 was statistically significant. All analyses were conducted using SAS v.9.4 (SAS Institute, Cary, NC).

Results

Among the 4,777 NSHAP Round 3 participants, 3,303 were age 60 years or older, including 2,634 who answered questions about their caregiving or care-receiving status. Of these participants, 444 (16.9%) self-identified as caregivers and were included in analyses. The mean age of these 444 older caregivers was 67.8 (SD 0.29) years, 55.8% (n= 253) were women, and 78.1% (n= 318) were non-Hispanic White. Thirty two percent (n=140) of all caregivers reported overlapping care-receiving by virtue of simultaneously receiving assistance with at least one ADL or IADL. Sixty three percent (n= 83) of overlapping caregivers were receiving assistance with at least one ADL. The most commonly reported activities requiring assistance among caregivers were driving during the night (68.7%, n=92), walking a block (50.6%, n=66), light housework (36.8%, n=51), and preparing meals (25.0%, n=42) (Figure 1).

Figure 1.

Prevalence of receiving assistance with activities of daily living (ADLs) or independent activities of daily living (IADLs) among older caregivers (N = 140). Percentages based on non-missing values. Data were missing for: preparing meals (n=2); taking medication (n=1); managing money (n=5); shopping for food (n=2); light housework (n=1); using the telephone (n=2); daytime driving (n=5); nighttime driving (n=6)

Among caregivers reporting overlapping care-receiving, 59.3% (n=85) were women, 80.9% (n=114) were married/partnered, and mean age was 70 years; 81.1% (n=106) were non-Hispanic White, 13.5% (n= 24) were Black, and 4.5% (n=9) were Hispanic. Thirty two percent (n=53) had a high school degree or less than high school level education and 60.5% (n=73) had less than $50,000 in household assets. Over half (n=80) of older caregivers who reported overlapping care-receiving were serving as a primary caregiver, with 44.3% (n=68) caring for a spouse, 18.2% (n=16) caring for a parent, and 8.5% (n=10) caring for a child or grandchild. Thirty-six percent (n=52) self-reported having poor physical health and 16.9% (n=19) poor mental health; 37.2% (n=56) had a MoCA-SA score less than 22 (Table 1).

Table 1.

Demographic and Clinical Characteristics of Caregivers Over Age 60 Years, by Care-Receiving Status^a,b

Participant Characteristic	All Caregivers N=444	Caregiving + care-receiving N = 140	Caregiving, not care-receiving N = 304	P
Gender
Female	253 (55.8%)	85 (59.3%)	168 (54.4%)	.410
Age in years
Mean (SD)	67.8 (0.3)	69.9 (0.5)	67.0 (0.3)	<.001
Age 60–69	224 (68.6%)	54 (58.8%)	170 (72.4%)	.001
Age 70–79	144 (21.9%)	46 (24.7%)	98 (20.8%)
Age 80–89	71 (8.9%)	39 (16.1%)	32 (6.1%)
Age ≥90	5 (0.6%)	1 (0.4%)	4 (0.7%)
Race
White	318 (78.1%)	106 (81.1%)	212 (77.0%)	.352
Black or African American	71 (13.0%)	24 (13.5%)	47 (12.8%)
Hispanic, non-Black	45 (6.5%)	9 (4.5%)	36 (7.3%)
Other	8 (2.4%)	1 (0.9%)	7 (2.9%)
Caregiver role
Serving as primary caregiver	256 (54.7%)	80 (54.8%)	176 (54.6%)	.977
Caregiving for spouse	166 (30.7%)	68 (44.3%)	98 (25.6%)	<.001
Caregiving for parent	96 (32.4%)	16 (18.2%)	80 (37.7%)
Caregiving for child or grandchild	34 (7.0%)	10 (8.5%)	24 (6.5%)
Caregiving for any adult other than above	135 (29.9%)	39 (29.0%)	96 (30.2%)
Educational attainment				.110
High school or less	143 (26.0%)	53 (31.6%)	90 (23.8%)
Vocational/some college/associate’s degree	177 (42.0%)	57(43.5%)	120 (41.3%)
Bachelor’s or more	124 (32.1%)	30 (24.9%)	94 (34.9%)
Household Assets				<.001
<$50,000c	194 (45.0%)	73 (60.5%)	121 (39.0%)
≥$50,000	250 (55.0%)	67 (39.5%)	183 (61.0%)
Marital status				.531
Married or living with a partner	354 (79.0%)	114 (80.9%)	240 (78.1%)
Self-reported health
Poor or fair physical health	97 (20.5%)	52 (36.2%)	45 (14.4%)	<.001
Poor or fair mental health	40 (9.3%)	19 (16.9%)	21 (6.5%)	<.001
MoCA-SA
Mean score (SE)	24.1 (0.2)	22.83 (0.4)	24.62 (0.2)	<.001
MoCA-SA score 22 or lower	143 (25.1)	56 (37.2%)	87 (20.5%)	<.001

^aActual frequencies are as reported. Percentages weighted for NSHAP sample distribution and presented based on non-missing values.

^bData were missing for: race and ethnicity (n=2); serving as primary caregiver (n=5); caregiver relationship (n=13); household assets (n=45); physical health (n=1); mental health (n=25).

^cGiven prior research that supports that most nonrespondents fall into the low-income group, subjects with missing household assets data (n=45) were combined with those with <$50,000 in household assets (n=149).

In unadjusted analyses, caregivers who reported simultaneously receiving assistance with daily activities were older than caregivers not receiving assistance (mean (SD) yrs., 69.9 (0.5) vs. 67.0 (0.3), p<.001) and more likely to be caregiving for a spouse (44.3% vs. 25.6%, p<.001). Caregivers receiving support reported having fewer household assets than caregivers who were not receiving assistance (60.5% vs. 39.0% with household assets <$50,000, p<.001). Compared to caregivers not receiving assistance, caregivers who reported overlapping receipt of assistance were more likely to report poor or fair overall physical health (36.2% vs. 14.4%, p<.001) and mental health (16.9% vs. 6.5%, p<.001), and they had lower MoCA-SA scores (mean score (SE), 22.83 (0.4) vs. 24.62 (0.2), p<.001) (Table 1).

In multivariable logistic regression models, caregivers who reported overlapping receipt of assistance had greater odds of being older (adjusted odds ratio (AOR) 1.30, 95% confidence interval (CI) [1.14, 1.48]) per each 5-year age increase), caregiving for a spouse (AOR 1.93, 95% CI [1.20, 3.13]), having limited household assets (AOR 2.10, 95% CI [1.17, 3.80] for household assets <$50,000), and having worse self-reported overall physical health (AOR 2.94, 95% CI [1.43, 6.02]) (Table 2).

Table 2.

Adjusted Odds of Care Receiving Associated with Demographic and Health Characteristics, Among Older Caregivers

	Odds Ratio (95% Confidence Interval)a
Age (per 5-year increase)	1.30 (1.14, 1.48)
Women (vs. men)	1.68 (0.88, 3.23)
Caregiving for spouse (vs. caregiving with other relationship to care-recipient)	1.93 (1.20, 3.13)
Cognitive function (per 1-point increase in MOCA-SA score)	1.07 (0.99, 1.15)
Poor or fair self-rated physical health (vs. good to excellent)	2.94 (1.43, 6.02)
Poor or fair self-rated mental health (vs. good to excellent)	2.20 (0.97, 4.99)
Household assets <$50,000 (vs. ≥$50,000)b	2.10 (1.17, 3.80)

^aOdds ratios and 95% confidence intervals are directed from multivariable logistic regression models, adjusting for age, gender, spousal caregiving, cognitive function, physical health, mental health, and household assets.

^bGiven prior research that supports that most nonrespondents fall into the low-income group, subjects with missing household assets data (n=45) were combined with those with <$50,000 in household assets (n=149).

Discussion

In this national community-based sample of older adults, more than 30% of older caregivers were themselves receiving assistance with at least one ADL or IADL. Caregivers who simultaneously received assistance were older, more likely to be caring for a spouse, had fewer financial assets, and reported worse overall physical health compared to caregivers not receiving assistance. These findings suggest that programs and policies designed to support older community caregivers need to address the unique challenges faced by those who are simultaneously managing their own care needs. Rather than making binary distinctions between “caregivers” and “care-recipients,” clinicians, researchers, and policymakers should acknowledge the fundamental overlap and potentially dynamic nature of these roles among older populations.

Adults age 50 or older provide the majority of caregiving for adults in the U.S., with one in five caregivers being over the age of 65.[1] Prior research has reported that older caregivers assist with a broad range of daily activities and tend to provide more hours of care compared to younger caregivers, while frequently managing their own declining health.[16] They are more likely than younger caregivers to rate their health as fair or poor and report higher levels of physical strain related to caregiving.[4, 9, 17, 18] Consequently, older caregivers may find themselves having to manage their own, considerably greater care needs while caregiving.

Strikingly, a significant number of older caregivers were receiving assistance with basic ADLs including getting dressed and getting in and out of bed; yet, they persisted in their caregiving roles. While few studies have described caregivers’ functional needs, our findings are consistent with a previous Center for Disease Control and Prevention survey reporting that 36% of caregivers over age 65 have a disability causing activity limitations.[19] Based on prior research, older adults may find themselves simultaneously providing and receiving care due to their unique caregiving context. Older caregivers are more likely to co-reside with their care recipients than younger caregivers, which may provide proximity and convenience that facilitates caregiving.[6] However, given that co-residing has been associated with more caregiving duties than distance caregiving, this may lead to higher caregiving-related burden.[20] Older caregivers also frequently provide care to a spouse, a relationship that can be tied to a strong sense of obligation and duty. However, spousal caregiving has also been associated with higher risk of experiencing physical, mental and emotional strain.[21–23]

Our study has limitations. Although participants were drawn from a national sample of community-dwelling older results, the majority of respondents identified as non-Hispanic White. Future research should address the overlap between caregiving and care-receiving in more racially and ethnically diverse populations. Our study could not provide detailed information about the caregiving environment, such as specific types of caregiving activities, or availability of supportive resources (e.g. respite care, home health, additional caregivers). Future studies should explore the broader support networks that enable functionally impaired or dependent older adults to serve as caregivers. We also could not confirm whether respondents were paid or unpaid caregivers, although the majority had a familial relationship to their care recipient, suggesting they were family caregivers.

To our knowledge, this is the first study to examine the characteristics of older adult caregivers simultaneously providing and receiving care. Our findings suggest that older caregivers who are themselves receiving assistance for daily activities represent a vulnerable group that requires greater recognition and support. Older adults in overlapping caregiving and care-receiving roles may benefit from referrals for supportive resources, such as respite care for temporary caregiving relief and physical assistance to support functional needs.[24, 25] Targeted approaches are needed to help clinicians systematically identify caregivers and document the assistance they require as well as characterize the broader caregiving network. These strategies may reveal unmet caregiving needs and prompt clinicians to direct crucial resources to these older caregivers. With such changes, we can begin to challenge the limiting distinctions between “caregiver” and “care recipient” to better support vulnerable older adults in the community who fulfil these overlapping and evolving dual roles.

Key Points.

• Over 30% of older caregivers were simultaneously receiving assistance with their own daily activities.

• Caregivers who were simultaneously receiving assistance had worse physical health and fewer financial assets than other older caregivers.

Why Does this Paper Matter?

Programs designed to support older community caregivers should address the unique challenges faced by those who are simultaneously managing their own care needs.