Caring for Aging Parents in the Last Years of Life

Abstract

Background:

Adult children provide a large portion of end-of-life caregiving for older adults and make up the majority of caregivers for adults with dementia. Yet research has been limited to the hours of care that primary caregivers provide, neglecting the other ways adult children provide caregiving support. This study aims to describe the caregiving support adult children provide to their parents at the end of life and characterize differences by race and ethnicity and dementia status.

Methods:

We conducted a retrospective study using survey responses from the Health and Retirement Study between 2002 and 2018. The sample population (n=8,040) included decedents aged 65 with at least one living adult child at their time of death. Caregiving support was defined as providing financial support, providing help with basic or instrumental activities of daily living (ADLs or IADLs), or coresiding with the care recipient. Respondents were stratified by self-identified race and ethnicity as Hispanic, non-Hispanic White, or non-Hispanic Black. Respondents were further stratified by dementia and marital status.

Results:

Black and Hispanic respondents without dementia were more likely to report receiving financial help from (28.0% and 25.9%) or coresiding with their adult children (38.9% and 49.7%) compared to White respondents (15.0% receiving financial help and 23.3% coresiding) (p<0.05). Among respondents with dementia, 47.1% of both Black and Hispanic respondents reported coresiding with their adult children, compared to only 24.6% of White respondents (p<0.05). Notably, married Black and Hispanic respondents reported significantly higher rates of all support types compared to married White respondents (p<0.05).

Conclusions:

The majority of older adults at the end of life receive some form of care and support from their adult children, with Black and Hispanic older adults receiving particularly high rates of care and support from their adult children regardless of dementia or marital status.

Introduction

An estimated 41.8 million adults care for someone aged 50 or older, and 48% of these are caring for a parent.¹ At the end-of-life, adult children not only make up the majority of informal caregivers, but also provide care of greater amounts and complexity.^2,3 This is particularly true for older adults with dementia, whose primary caregivers are predominantly adult children.⁴ Despite this, end-of-life caregiving research has prioritized spouses and focused on primary caregivers, defined as those providing the most hours of care (typically functional help). This neglects the diverse ways adult children provide care, such as providing help with finances, household chores, personal tasks, or care coordination, regardless of whether they are narrowly identified as primary caregivers. Additionally, the end of life is associated with high medical costs to families of older adults,⁵ but general financial support during the end of life is underexamined. With more Americans spending the end of life at home, older adults are increasingly living with their adult children which has been documented to have a negative impact on adult children’s financial stability and physical health.^6,7

Prior work has shown that race and ethnicity shape the experiences of family caregivers. Black older adults on average require higher hours of care and more intense care, likely as a consequence of financial, health, and other disparities perpetuated by systemic racism.⁸ With caregiving research focused primarily on hours of help, the disproportionate effects of caregiving on Black and Hispanic families has been underexamined. This study aims to examine the many roles of adult children in providing end-of-life caregiving support and assess racial and ethnic group differences.

Methods

We conducted a retrospective descriptive analysis using data from the Health and Retirement Study (HRS) biennial surveys from 2002–18. We included decedents 65 or older at the time of death, identified by a post-death exit interview completed by a friend or family member (n=9,902). We excluded any respondents lacking at least one living, adult child at the time of their final interview (n=1,696). To measure probable dementia status we used an algorithm validated for HRS data.⁹ Respondents missing data necessary to assess dementia status were excluded (n=529). Self-reported race and ethnicity included Hispanic, non-Hispanic Black, non-Hispanic White, and non-Hispanic Other. Due to data privacy restrictions on reporting cell sizes <11, we excluded respondents identifying as non-Hispanic Other and those missing race or ethnicity measures from analysis (n=129).

For each respondent, we utilized data from three interviews, two pre-death interviews taken two years apart and one post-death “exit interview”. Demographic information was taken from the final pre-death interview which occurred on average 14 months prior to death. All survey data prior to the exit interview was self-reported unless a respondent was unable to answer due to illness or cognitive impairment. In these instances, a knowledgeable proxy responded. We assessed three categories of caregiving support provided by adult children: financial support, help with daily activities, and coresidence. We constructed a composite measure of ‘any support’, defined as receipt of one or more of the three measures of caregiving support. The HRS questionnaire asks if since the last interview, the respondent received financial help totaling $500 or more from any children or grandchildren, with an affirmative answer triggering follow up questions to identify which child(ren) or grandchildren and the amount of money given. Using this question, we defined financial support as reporting financial help from any child at any interview. HRS defines financial help as, “money, help paying bills, or covering specific types of costs such as those for medical care or insurance, schooling, down payment for a home, rent, etc,” and explicitly excludes shared household expenses. To measure functional help, HRS asks if the respondent currently receives help with Activities of Daily living (ADLs) and Instrumental Activities of Daily Living (IADLs). HRS asks about each ADL (ambulating, dressing, bathing, eating, toileting) and IADL (meal preparation, shopping, telephone use, money management) separately, then asks for the identity of helpers for each. We defined help with daily activities as receipt of help from any child with any ADLs or IADLs at the time of any interview. For assessing coresidence, HRS asks if each living child currently lives with the respondent or not. We defined coresidence as reporting residing with one or more children at any interview.

We assessed each support type for the overall sample then stratified results by race and ethnicity, dementia status, and marital status. For a descriptive analysis of group differences, one-way and two-way ANOVA tests were performed to compare proportions between groups.

Study Results

Our sample consisted of 8,040 decedents, with an average age at death of 82.7 years. 79.3% identified as non-Hispanic White, 54.0% were female, 43.2% were married, and 35.7% had dementia (Table 1). Black and Hispanic respondents were more likely to be in the lowest quartile of net worth (55.2% and 55.3%) compared to White respondents (27.8%), p<0.05. Compared to White respondents who received Medicaid (15%), Black respondents were over twice as likely to receive Medicaid (40%) and Hispanic respondents were over three times as likely to do so (47%), p<0.05.

Table 1.

Demographic and socioeconomic characteristics of decedents age 65 and older by race and ethnicity, 2002–18

	N.H. White (n=6,378)	N.H. Black (n=1,083)	Hispanic (n=579)
Age at last interview	82	80	80
Time from final interview to death (months)	14.7	14.5	14.1
Female	53.8%	56.1%	52.7%
Married	44.7%	33.9%	43.5%
Education
<12 Years	30.29%	60.91%	73.8%
12–14 Years	50.8%	30.9%	20.6%
≥15 Years	18.9%	8.2%	5.7%
Quartile of Net Worth among 65+a
Lowest	27.8%	55.2%	52.3%
2	24.9%	31.5%	30.6%
3	25.6%	10.5%	12.4%
4	21.7%	2.8%	4.7%
Avg. Net Worthb	$419,611	$80,740	$99,471
Proxy Respondent	27.1%	32.8%	37.7%
Probable Dementia	33.6%	43.2%	44.4%
Medicaid recipient	14.6%	39.7%	46.7%
Lives alone	25.9%	22.8%	19.3%
Number of Household Membersa
0	31.9%	26.9%	22.2%
1	51.6%	39.2%	35.9%
2	10.5%	17.1%	15.3%
≥3	5.9%	16.9%	26.6%
Nursing Home Resident	19.0%	15.0%	12.8%
4+ comorbidities	23.2%	22.2%	20.3%
Recieves ADL Help	40.0%	45.6%	48.9%
Any Paid Help	7.7%	7.3%	15.9%
Total Number of Helpersa
0	46.2%	42.0%	37.3%
1	21.1%	21.1%	23.3%
2	18.6%	16.9%	20.6%
≥3	14.1%	19.9%	18.8%

^aTotals for categorical variable breakdowns may not add to 100% due to rounding.

^bNet worth quartiles are based on the net worth of all Health and Retirement Study respondents ≥65 included in waves from 2002– 18.

Dementia Status

66.5% of respondents without dementia and 80.8% of respondents with dementia reported receiving some kind of caregiving support from their adult children, p<0.05 (Figure 1). Black and Hispanic respondents were more likely to receive any support regardless of dementia status, but the difference was only significant for the non-dementia group. Among respondents without dementia, 72.4% of Black respondents and 74.8% of Hispanic respondents reported receiving any kind of support, compared to 65.0% of White respondents (p<.05). For respondents with dementia, 80.1% of White respondents, 80.8% of Black respondents, and 86.0% of Hispanic respondents reported receiving any kind of support.

Figure 1.

Percent of older adults receiving caregiving support from adult children in their last 3 years of life by dementia status and race and ethnicity. Financial support includes providing money, help paying bills, or help with other costs that wouldn’t otherwise be shared by the adult child. Activities of Daily Living (ADL) and Instrumental Activities of Daily Living (IADL) include activities such as bathing, toileting, dressing, grooming, managing finances, using the phone, and shopping. Error bars represent 95% confidence interval. Asterisks indicate p<0.05 in comparison to White older adults.

Among respondents without dementia, 28.0% of Black respondents and 25.9% of Hispanic respondents received financial support, compared to 15.0% of White respondents (p<0.05). Coresidence for respondents without dementia shows a similar trend. 38.9% of Black respondents and 48.7% of Hispanic respondents reported living with an adult child, compared to 23.2% of White respondents (p<0.05). There was no significant difference in proportions of help with daily activities for White (57.8%), Black (58.9%), and Hispanic (57.6%) respondents without dementia.

Among individuals with dementia, there was no significant difference in proportions of financial support for White (20.6%), Black (29.3%), and Hispanic (21.7%) respondents. Proportions of respondents receiving help with daily activities were similarly high for White (78.6%), Black (77.7%), and Hispanic (83.9%) respondents. Black and Hispanic respondents reported significantly higher proportions of coresidence with an adult child (47.1% for both) compared to White respondents (24.6%), p<0.05.

Marital Status

Proportions of support received by unmarried respondents were similarly high for all race and ethnicity groups, though coresidence remained significantly higher for unmarried Black (46.4%) and Hispanic (51.2%) respondents compared to White respondents (31.6%), p<0.05 (Figure 2).

Figure 2.

Percent of older adults receiving caregiving support from adult children in their last 3 years of life by marital status and race and ethnicity. Financial support includes providing money, help paying bills, or help with other costs that wouldn’t otherwise be shared by the adult child. Activities of Daily Living (ADL) and Instrumental Activities of Daily Living (IADL) include activities such as bathing, toileting, dressing, grooming, managing finances, using the phone, and shopping. Error bars represent 95% confidence interval. Asterisks indicate p<0.05 in comparison to White older adults.

Proportions of each support type were significantly higher for married Black and Hispanic respondents. Over 20% of married Black and Hispanic respondents received financial support from their children, compared to only 10.7% of married White respondents (p<0.05). Similarly, 34.5% of married Black respondents and 43.7% of married Hispanic respondents reported living with adult children, compared to 16.5% of married White respondents (p<0.05). Over half of married Black and Hispanic respondents reported receiving help with daily activities from their children, compared to 44.4% of White married respondents (p<0.05).

Discussion

This study is novel in simultaneously examining three types of caregiving support provided by adult children (financial support, coresidence, and ADL/IADL help), capturing the integral role of adult children in end-of-life care. Including coresidence is particularly important, as coresidence entails financial, social, and health impacts that are not always captured by other measures. Our study shows that adult children provide integral end-of-life caregiving support that may not be captured by studies focused only on primary caregivers providing functional help.

Our study has several limitations. The majority of respondents identified as non-Hispanic White, with Hispanic respondents representing only 7.2% of our sample. It must also be acknowledged that the ethnic designation “Hispanic” encompasses a wide variety of diverse cultural backgrounds. The lack of sufficient numbers of respondents identifying as “non-Hispanic Other” meant that our analysis excludes the experience of many Americans. The HRS questionnaire likely underestimates the financial impact of caregiving by excluding shared household expenses from its definition of financial help. In the case of coresidence, an adult child may be paying part or all of the shared grocery and household expenses. This may be why the majority of respondents who were living with an adult child did not report receiving financial support. The financial support question is also uniquely retrospective, so that this measure spans five years while the other outcomes span three years. Our data also does not include the financial burden of caregiving’s impact on participation in the workforce. Adults caring for their parents at the end of life are more likely to take lower paying jobs with flexible hours, are less likely to return to work after a leave of absence, and receive lower wages if they do return.¹⁰

We found that Black and Hispanic older adults were much more likely to live with their adult children than their White counterparts, regardless of dementia or marital status. These differences are influenced by factors ranging from personal and family preferences to differences stemming from lifetime exposure to systemic racism. Wealth disparities limiting access to privately hired home care or assisted-living facilities may be a contributing factor, as over half of Black and Hispanic older adults in the sample had a net worth in the bottom quartile of adults 65 and older. Additionally, residential segregation and segregation by insurance status have led to disparities in affordability and quality of nursing homes, assisted-living facilities, and paid home care for Black and Hispanic older adults, likely contributing to low utilization of these services and reliance on coresidence with adult children.^11,12

Cultural norms also play a role in increased coresidence. Many Hispanic cultures value family collectivism and communal care for family members throughout life.¹³ These normative beliefs may encourage increased levels of informal caregiving support like coresidence, but they may also contribute to caregivers not seeking out support for fear of judgment.^13,14 While Hispanic families are more likely to report psychological benefit from caregiving experiences, they are also more likely than White caregivers to report social isolation, lack of support, and symptoms of depression.^14,15 Surveys of Black end-of-life caregivers show higher rates of reporting positive aspects of caregiving and higher psychological well-being than both White and Hispanic caregivers.^6,7,15

Policies to support family caregivers are limited and may not match the needs of adult children providing end-of-life caregiving support. Paid family leave policies are only available in nine states and Washington, D.C.¹⁶ Even these states offer only partial salary reimbursement, and six require unpaid waiting periods, limiting utility for the episodic nature of end-of-life care. Federal policies are even more lacking. The National Family Caregiver Support Program (NFCSP) funds support services for family caregivers, but there is high variability in services and accessibility by state, with many states capping hours of respite care and limiting eligibility by income. While the NFCSP authorizes funds to be used to reimburse family caregivers for out-of-pocket expenses, fewer than half of states offer this service.¹⁷ Services also remain underutilized by caregivers with low income or education, who are disproportionately Black and Hispanic. Multiple efforts are underway at federal, state, and local levels to improve caregiving support policies.^11,18

Adult children are integral sources for end-of-life caregiving support. Black and Hispanic adult children provide high levels of caregiving support, especially coresidence, putting them at increased risk of caregiver burden and financial burden. This study lays the groundwork for future work needed to characterize the economic and social impact of end-of-life caregiving on adult children, including measuring the scope and intensity of caregiver support and use of respite and support services. A better understanding of the vital roles of adult children as end-of-life caregivers will better enable policymakers to create support programs that are commensurate to the high levels of care adult children provide.

Key points:

• Adult children provide a high level of caregiving support to their aging parents at the end of life.

• Black and Hispanic older adults are significantly more likely to live with their adult children or receive financial support from their children when compared to White older adults.

• Married Black and Hispanic older adults receive significant amounts of caregiving support from their adult children despite the assumption that spouses are the typical primary caregivers

Why does this paper matter?:

Millions of adult children provide end-of-life care for their older adult parents every year, yet current policies to support end-of-life family caregivers are limited and unequally distributed. While much research has explored the functional help adult children provide older adults, little attention has been paid to other ways adult children may be supporting their parents at the end of life. We found that adult children provide not only high rates of functional help, but also coresidence and financial support, two types of caregiving support that have profound social and economic impacts on caregivers and their families. Black and Hispanic adult children provide particularly high rates of caregiving support regardless of their parent’s dementia or marital status, putting them at increased risk of caregiver burden and financial stress. Our study demonstrates that current state and federal caregiver support programs are not commensurate with the high levels of caregiver support adult children provide their parents at the end-of-life.