Gender Disparities in the Receipt of Home Care for Elderly People With Disability in the United States FREE

One of the most important health policy issues that we face now in this century is how to provide support to elderly people with disabilities living in the community. Because of projected demographic shifts in the US population over the next 50 years, families, health care practitioners, and policymakers will be confronted with a marked increase in the number of elderly people living in the community who must cope daily with disabilities associated with aging and chronic disease.¹ Two additional social and organizational trends will further increase the need for home care support for elderly people with disabilities. First, there is a projected trend toward reduced nursing home use that will increase the number of more severely disabled people living in the community. Second, changes in the pattern of living arrangements will increase the number of elderly people living alone and thus reduce the availability of informal care.² These trends portend an increasing number of disabled elderly people living in the community vulnerable to unmet support needs. In response, many health care organizations are developing programs to provide a myriad of community support services.

The concern about the adequacy and quality of support in the community for disabled elderly people is particularly salient to women for several reasons. First, women make up a disproportionate number of disabled elderly people in the community because they tend to live longer than men. Second, disabled women living in the community may be particularly vulnerable to unmet need because many of them live alone with limited resources. Third, even disabled women in married households may be vulnerable to unmet need because they may be more likely than men to be in a caregiver role themselves.^3- 4

Unfortunately, we know very little about gender differences in the receipt of informal (generally unpaid) and formal (generally paid) home care for disabled elderly people in the United States. Few studies have examined patterns and determinants of informal and formal home care by gender and the results of these studies have conflicted. Some studies have suggested that women receive somewhat more care than their male counterparts,^5- 7 but the generalizability of these studies is limited by either small nonpopulation-based samples or unique settings. By contrast, results from one study using a national population-based community sample in 1989 suggested that disabled women may have received somewhat less informal care than their male counterparts.⁸

In contrast to the paucity of articles on gender differences in receiving care, there are more articles addressing gender differences in giving care. This literature strongly suggests that women are much more likely to be caregivers than men.^3- 4 The dominance of women in the caregiving role has been attributed to several factors including the fact that fewer women have been engaged in paid employment and that women may more easily assume the role of caregiver to a disabled spouse or parent because of traditional social role functioning.⁹

We used a large national survey representative of noninstitutionalized elderly people living in the United States to address gender differences in the receipt of informal and formal home care. The research questions were: (1) Does receipt of informal care differ by gender after controlling for other factors? (2) If so, what are the reasons for these disparities? and (3) Does formal care substitute for informal care, and thus, diminish the gender disparity in receipt of home care?

We used data from the first wave (1993) of the Asset and Health Dynamics Among the Oldest Old (AHEAD) study, an ongoing longitudinal survey of a nationally representative cohort of US elderly people born in 1923 or earlier (N = 7443). Funded by the National Institute on Aging, the study was designed to examine health transitions in old age and their effect on individuals, families, and society.¹⁰ Importantly, both spouses within married households received a full interview. In addition to measures of the health and functional status of the elderly survey respondents, data on the number of hours of care provided by both paid and unpaid caregivers in the home were collected. Seventy-two percent of those aged 70 to 79 years were interviewed by telephone, whereas 70% of those aged 80 years or older were interviewed in person (baseline response rate was 80%). Approximately 19% of respondents in our study sample required a proxy (most often a family member) to help complete the survey. Response rate did not differ significantly by mode of questioning and neither mode of questioning nor proxy status affected our results in this study.

Our study sample included 3109 respondents who: (1) were aged 70 years or older; (2) lived in the community; and (3) were disabled, defined as reporting during the prior month that they had difficulty or were receiving help with 1 or more activity of daily living (ADL), ie, eating, transferring, toileting, dressing, bathing, walking across a room, or instrumental activity of daily living (IADL), ie, preparing meals, grocery shopping, making telephone calls, taking medications, managing money.

The principal dependent variables were the number of weekly hours of informal and formal home care. Respondents were classified as having received informal home care if they received in-home assistance with any ADL or IADL in the prior month from a relative (paid or not), or unpaid nonrelative with no organizational affiliation.¹¹ Formal home care was defined as help with any ADL or IADL in the prior month from a paid nonrelative or from someone affiliated with an organization, whether paid or not.¹¹ Respondents who did not normally perform an IADL, such as shopping, were not classified as having an IADL impairment nor as having received help for that IADL. The number of weekly hours of informal and formal home care were calculated using the average number of days per week (in the prior month), and the average number of hours per day that respondents reported receiving help from caregivers.¹²

Independent variables of greatest interest were gender, living arrangement (unmarried and living alone, unmarried and living with others, and currently married) and level of disability (2 variables indicating the number of current ADL impairments and the number of current IADL impairments).^13- 15 About three quarters of unmarried elderly people living with others were residing with either children or grandchildren. We also included the most common medical comorbidities and an indicator of the presence of dementia (derived from an administered test of cognition)¹⁶ because these clinical factors may be associated with need for home care support. We controlled for other covariates previously shown to be associated with home care.^17- 19 These included predisposing and enabling factors: age (continuous), ethnicity, education, and net worth (Table 1).

Because a substantial proportion of respondents used no informal or formal home care in the month prior to the administration of the survey and because the distribution of hours among users was highly skewed, we examined caregiving using a 2-part model.^17,20 In the first part, logistic regression was used to estimate the independent association of gender with any use of informal home care in the prior month controlling for other factors. In the second part, ordinary least squares regression was used to estimate the association of gender with the natural logarithm of informal home care hours per week for individuals who used any services. The results from both models were combined to calculate adjusted mean weekly hours of informal home care for different subgroups of respondents. We then repeated these analyses and calculations with formal care use and total care use (formal and informal care combined) as the dependent variable.

All analyses were weighted for differential probability of selection and adjusted for the complex sampling design of the AHEAD survey.^10,21 We tested for significant interactions between all covariates and regression diagnostics were performed to check for model assumptions. All analyses were performed using STATA statistical software, Release 6.0 (College Station, Tex).

Table 1 shows characteristics of the study population for all variables. About two thirds of the study sample were women. Compared with men, women reported receiving fewer weekly hours of informal care but reported receiving a greater number of hours of formal care. Women were older and reported lower net worth than men. Women were much more likely to be living alone and much less likely to be living with a spouse. Women reported more ADL impairments and more chronic medical conditions.

Figure 1 shows the mean weekly hours of informal care by gender and living arrangement after adjusting for all other covariates. Overall, women received fewer hours of informal care (15.7 hours; 95% CI, 14.5-16.9) compared with men (21.2 hours; 95% CI, 19.7-22.8). Two factors related to living arrangement appeared to explain the overall gender disparity in the use of informal care. First, many more disabled women lived alone than men (45.4% vs 16.8%, respectively; Table 1) and disabled elderly people living alone received many fewer hours of informal care than those who were married or living with others. Second disabled married women received many fewer hours of informal care (14.8 hours; 95% CI, 13.7-15.8) than disabled married men (26.2 hours; 95% CI, 24.6-27.9). Disabled married women received about 80% more hours of informal care than disabled women living alone (14.8 vs 8.2 hours, respectively), whereas disabled married men received 230% more hours than disabled men living alone (26.2 vs 7.9 hours, respectively).

Table 2 shows that these gender differences were observed for both the percentage of people who received any informal care and the intensity of care (number of weekly hours) among users after controlling for all other factors. Among all disabled elderly people, 55.1% of women vs 60.8% of men received any informal care (P<.001) and women received 28.3 hours of care per user vs 35.6 hours received by men (P<.001). Among the married disabled population, 53.8% of women vs 64.6% of men received any informal care (P<.001), and women received 26.8 hours of care per user vs 39.9 hours received by men (P<.001). Interactions between gender and living arrangement were significant for both the probability of any informal care use and the mean number of hours per user.

Table 3 shows the distribution of informal care (percentage using any care and mean hours among users) by type of helper, disability level and gender for elderly people who were married and disabled. Among those with 1 or 2 ADL or IADL impairments 27.2% of women vs 48.6% of men (P<.001) received any informal care and women received 10.0 weekly hours vs 19.6 hours received by men (P<.001). This gender disparity resulted from disabled women receiving much less spousal care than their male counterparts. For example, 17.4% of women vs 40.3% of men received any informal care from a spouse alone (P<.001) and female receivers of spousal care on average received 9.8 weekly hours vs 18.9 hours received by men (P = .003). The gender disparity in informal care diminished as the level of disability increased, primarily because children provided increasing number of hours of care to their disabled mothers. For example, among those with 3 to 5 ADL or IADL impairments, 54.6% of women vs 65.1% of men received any informal care from a spouse alone (P = .04). However, 19.6% of women vs 8.7% of men received any informal care from children alone (P = .005). These gender differences in the involvement of children as caregivers were even more marked for elderly people with 6 to 11 impairments.

Indeed, our results as a whole show that children played a dominant role in the care of disabled women whereas wives played the dominant role in the care of disabled men. Among all disabled women, 44.6% reported receiving informal care from 1 or more children vs 22.8% of disabled men (P<.001) and more than 80% of these children were daughters, daughters-in-law, or granddaughters. By contrast, only 11.1% of disabled women reported any informal care from a spouse compared with 43.8% of disabled men (P<.001). Children played an important caregiving role even among married disabled women since 23.0% reported receiving informal care from 1 or more children vs 13.2% of married disabled men (P<.001). By contrast, 38.0 % of married disabled women reported receiving any informal care from their spouse compared with 58.0% of married disabled men (P<.001).

The gender disparity in the receipt of informal home care for married disabled elderly people was not limited to IADL impairments such as shopping or preparing meals but rather, it was observed consistently across levels of both ADL and IADL impairments. For instance, 49.8% of men with 1 ADL impairment and 69.6% with 2 ADL impairments received any informal ADL support vs 35.1% of women with 1 ADL impairment and 54.3% with 2 ADL impairments (P<.001).

Two factors may explain why husbands of disabled women play a much lesser role in caregiving than wives of disabled men. First, husbands of disabled women may be less able to provide care because they may be older and more disabled than wives of disabled men. Indeed, on average, disabled women lived with husbands who were 1 year older; while disabled men lived with wives who were on average 4.6 years younger. However, the disability level of husbands of disabled women was similar to that of wives of disabled men; 42.4% of the husbands of disabled women reported 1 or more impairments (mean [SD] number of impairments, 3.0 [2.6]) compared with 37.8% of the wives of disabled men (3.1 [2.7]). To examine whether spousal age or spousal disability level influenced the gender disparity in home care use among married disabled respondents, we performed secondary analyses regressing spousal home care support on respondent gender, age, and disability level with and without the addition of spousal age and spousal disability level. These analyses showed that spousal disability, not spousal age, was inversely associated with spousal support but that the gender disparity in weekly hours of spousal home care support was unchanged after controlling for these spousal factors.

Another explanation for the gender difference in informal care in married households is that husbands may be less prepared to fulfill the social role function of caregiving even without the presence of disability.^3- 4 Indeed, our results suggest that many married women with disability remained in a dominant caregiving role even if they themselves reported severe disability: 20% of married disabled women who reported 2 or more ADL impairments were providing informal home care to their husbands whereas only 8% of similarly disabled married men were providing informal care to their wives.

Figure 2 shows adjusted weekly hours of formal care by gender and living arrangement. Formal care did not substantially diminish the large gender disparity in use of informal care because formal care use was much lower compared with informal care use, and the absolute difference in formal care use across gender was very small. The probability of receiving any formal care was about one fifth that of informal care. About 10% of disabled elderly people received some formal care (13.2% women vs 11.0% men, P = .13), but more than half received some informal care. Adjusted weekly hours of formal care were 2.8 hours (95% CI, 2.5-3.1) for women and 2.1 hours for men (95% CI, 1.7-2.4). The gender difference in adjusted weekly hours of total home care, which combines informal and formal care, was significant for all respondents (for women 19.9; 95% CI, 18.2-21.5 vs for men, 26.9; 95% CI, 24.8-29.1) and among married disabled elderly people (for women, 18.6; 95% CI, 17.1-20.1 vs for men, 31.0; 95% CI, 28.7-33.3 hours).

Our results show that there was a large gender disparity in the receipt of informal home care for elderly people with disability in the United States. Controlling for disability and other factors, disabled women received about one third fewer hours of care than their male counterparts. This was mainly due to 2 factors. First, nearly half of disabled women lived alone vs 17% of men, whereas nearly three quarters of disabled men lived with a spouse. Second, married disabled women received many fewer hours of care compared with their married male counterparts.

Many disabled women lived alone with very limited financial resources available to obtain outside help. Our results show that in 1993 more than 25% of disabled women lived alone and had a net worth of less than $30 000 (vs about 6% of disabled men). This represented about 1.3 million women in the United States. A surprising finding in our study was that married women with disability received many fewer hours of care than their male counterparts. Disabled married women received about 80% more informal home care hours than disabled women living alone while disabled married men received about 230% more care than disabled men living alone. Our results suggest that this gender disparity was likely related to sociocultural factors regarding the caregiving role rather than physical limitations of the husbands of disabled women because husbands and wives of disabled married respondents reported similar levels of impairment. Furthermore, controlling for spousal age and spousal impairment did not change the gender disparity in receipt of home care. The dominance of women as caregivers for disabled elderly people may be because fewer women are engaged in paid employment and because women may more easily assume the role of caregiver to a disabled spouse or parent because of traditional social role functioning.^3- 4,9

Paid home care did not substantially reduce the large gender disparity in home care observed in this study because it accounted for a relatively small proportion of total care and the gender difference in paid care was less than 1 hour per week. Changes in 1989 in Medicare payment policies for paid home care resulted in markedly increased expenditures for these services with growth of 20% per year, from $5 billion in 1990 to $18 billion in 1997.²² This increase in the availability of paid home care during and after the study period may have had an impact on the gender difference in the use of formal home care that we found in this study. However, in another study we found that the growth in paid home care between 1993 and 1995 was almost entirely restricted to disabled elderly people living with others and that no significant change was observed among elderly people living in married households or alone.²³

Our results show that children, largely daughters, daughters-in-law, and granddaughters, were the dominant caregivers of disabled women whereas wives were the dominant caregivers of disabled men. This is because three quarters of disabled women lived alone or with a child and almost all of their informal care was provided by children. By contrast, three quarters of disabled men lived with spouses who provided most of their informal care. Even children of married disabled women played an important caregiving role especially for their mothers with more severe disability. By contrast, as disability increased among married men, wives largely assumed the increased care needs even when these women were themselves disabled. Indeed, we found that 20% of married women with 2 or more ADL impairments vs 8% of married men were caregivers to their spouse.

Several aspects of the study merit comment. First, the gender differences that we found might have been mainly related to IADL support for grocery shopping or preparing meals, which is traditionally provided by women, and may have been overreported in the survey. We do not believe this to be the case because the gender difference was consistently observed across both ADL and IADL support categories. Furthermore, question formats in the survey sought to exclude IADL support unassociated with a health problem. Another limitation is that our data are for only 1 year. Although the AHEAD survey is longitudinal, the subsequent 2 waves (1995 and 1998) dropped questions pertaining to hours of informal care provided by spouses. Thus, we could not include these waves in our analyses.

The gender difference in the receipt of informal care and the role of family members in the care of disabled elderly people have enormous implications for home care program policies in the United States. Biological, social, and organizational trends in the United States will result in a marked increase in the number of elderly people living with disability in the community.²⁴ The consequences of these trends on the quality of life of elderly people and their families differ markedly by gender. Our results show that women and their extended families are much more vulnerable to the consequences of these trends than men. Many more women than men are disabled, and care may be less available for disabled women because they are much more likely to be living alone with very limited financial resources. Less obvious to policymakers and clinicians is the fact that many married women with disability may be vulnerable to unmet need because of limited caregiving from their husbands and because they themselves must often remain a central caregiver in the family. Because disabled elderly women rely heavily on children for support, especially female children, the family burden and stress associated with caring for a disabled woman should be the subject of further study. Programs providing home care support to elderly people need to consider these large gender disparities and consequences on the burden on the family when developing and targeting intervention strategies in the community.