African American grandparents raising grandchildren: A national profile of demographic and...

Social workers, health care providers, and mental health care providers working in large inner-city clinics in the late 1980s reported a marked increase in the number of missed appointments, stress-related conditions, and exacerbation of previously controlled hypertension and diabetes among their middle-age and older female patients (Davis, 1993; Miller, 1991; Minkler & Roe, 1993). On further investigation it became apparent that many of these patients had recently assumed custody of grandchildren or greatgrandchildren.The 1990 census revealed that what social workers and services providers were finding mirrored a much larger national trend. The 1980s indeed saw a 44 percent increase in the number of children living with grandparents or other relatives. In a third of these homes, neither parent was present (Saluter, 1992), typically making the grandparent the sole or primary caregiver.

Although grandparent caregiving includes all racial and ethnic minority groups, it is particularly prevalent in African American families. By the mid-1990s, 13.5 percent of African American children were living with grandparents or other relatives (Lugaila, 1998), compared with 6.5 percent of Hispanic children and 4.1 percent of white children. Almost 30 percent of African American grandmothers and about 14 percent of African American grandfathers reported having had primary responsibility for raising a grandchild for at least six months at some point in their lives (Szinovacz,1998), compared with 10.9 percent of all grandparents (Fuller-Thomson, Minkler, & Driver, 1997).

This article examines differences in physical and mental health status between African American grandparents raising grandchildren and those who are not involved in this caregiving role. This discussion builds on our earlier study of grandparent caregivers, using a national data set, but it is unique in that it explores the physical and mental health of the growing subpopulation of African American grandparents who are raising grandchildren.

REVIEW OP THE LITERATURE

Increase in Grandparent-Headed Households

The rapid growth in the number of children formally placed with relatives can be traced in part to federal and state laws and policies, beginning in 1979, which had the effect of encouraging or requiring that a preference be given to next-of-kin in the placement of foster children. In many of the most populated parts of the United States, fully one-half of the children in out-of home placements are in "kinship care" or formal placements with relatives (Brooks, Webster, Berrick, & Barth, 1998). But policies promoting kinship care do not explain the concomitant increase social workers also have seen in the number of children who have been informally "going to live with grandma"-a trend that has continued into the late 1990s (Harden, Clark, & Maguire, 1997; Minkler,1999). A variety of social factors have contributed to the increase. Prominent among these are drug abuse (particularly the epidemic of cocaine use in the 1980s), the rise in households headed by single parents, teenage pregnancy and youth unemployment (Burnette, 1997; Harden et al., 1997), HIV/ AIDS (Joslin & Harrison, 1998), and a sixfold increase in the rate of female incarceration from 1980 to 1995 (U.S. Department of Justice, 1997).

Grandparent Caregiving among African Americans

Several factors help explain the disproportionate number of African American children living in the care of grandparents. Historically, the extended family was the primary West African family structure at the time of slavery (Scannapieco & Jackson, 1996). The immediate needs of children who were separated from their parents during slavery promoted continued reliance on extended family caregiving (Sudarkasa,1981; Wilson,1989). In the United States, during the first half of the 20th century, poverty, oppression, racism, and the consequent lack of opportunity in the South led to great migration of African Americans to the North in pursuit of employment. It was commonplace for grandparents to remain in the South to care for children while the middle generation established themselves in Northern cities (Sudarkasa, 1981; Wilson, 1989). Subsequently and continuing through the 1960s, urban parents often would send their children in summer to grandparents and other extended family in the South. These visits allowed the new generation to remain closely connected with their grandparents and other relatives in the South and to be exposed to cultural trraditions.

In contrast, as Burton and Dilworth-Anderson (1991) have noted, contemporary grandparent caregiving often occurs in response to crises such as incarceration of children's mothers, HIV/AIDS, and substance abuse. These reasons for caregiving frequently carry with them a stigma for the whole family, including grandparent caregivers. Extended families, churches, and other supports that historically have helped African Americans cope with racism and other adversities (Gibbs, 1991) are in place for many grandparent caregivers and will be discussed later; however, the stigma attached to AIDS and drug abuse also has led to social isolation from peers and not infrequently from churches in the African American community (Generations United,1998; Poe, 1992).

Health Problems Associated with Grandparent Caregiving

Although caring for one's grandchildren brings many rewards, including "keeping the family together" (Burton, 1992; Jendrek, 1994; Minkler & Roe, 1993; Poe, 1992), it also has been associated with potentially serious physical and mental health problems. Key among these is depression, which, as Walker and Pomeroy (1996) noted, is one of "the more enduring effects of caregiving" (p. 247). Recent analysis of the National Survey of Families and Households (NSFH) multicultural sample found that 25 percent of all grandparent caregivers had clinically relevant levels of depression (Minkler, Fuller-Thomson, Miller, & Driver,1997). Depression may develop because of difficulties in balancing multiple work, family, and social responsibilities at a time when many grandparents had hoped to have more time to themselves. The increased demands on caregivers' time and finances may be particularly disheartening when contrasted with noncaregiving grandparents' increasing freedom and leisure time.

Earlier qualitative studies among African American grandparents raising grandchildren (Burton, 1992; Minkler & Roe, 1993; Poe, 1992) have suggested that such depression also maybe triggered by sorrow that results from distressing circumstances surrounding the onset of care (such as the substance abuse, incarceration, or death of the adult child). Walker and Pomeroy (1996) suggested that symptoms of depression, in fact, may reflect the caregivers' "normal distress" and grief (p. 248). Whether depression or grief, however, the elevated rates of psychological distress reported among African American grandparent caregivers are noteworthy.

Adverse physical health outcomes and functional health limitations also have been associated with primary caregiving for a grandchild, including exacerbation of pre-existing chronic conditions, comorbidity, declines in self assessed health, and limitations in one or more activities of daily living (Burnette, 1999b; Miller, 1991; Minkler & FullerThomson, 1999a; Minkler & Roe, 1993; Strawbridge, Wallhagen, Shema, & Kaplan,1997). To untangle the unique effects of caregiving on health, it is essential to take into account the fact that older African Americans have, on average, more physical health problems because of, in part, racism, oppression, and poverty.In fact,AfricanAmerican women experience greater morbidity and mortality than all other groups of women (Gaston, Barrett, Johnson, & Epstein,1998). Consequently, African American caregivers must be compared with noncaregiving African Americans, rather than with the general population of noncaregivers.

The great majority of studies on grandparent caregivers to date have used small nonrepresentative samples, and hence their results cannot be generalized (for example, Burnette, 1999b; Burton,1992; Minkler & Roe, 1993; Shore & Hayslip, 1994). Studies that have used representative data sets (for example, Chalfie, 1994; Fuller-Thomson et al., 1997; Minkler & Fuller-Thomson, 1999a, 1999b; Minkler et al., 1997; Szinovacz, 1998) have not focused their research on African American grandparent caregivers. With the use of a nationally representative subsample of African Americans, we sought to address this gap and answer the question, "How do African American caregiving grandparents differ with respect to demographic and physical and mental health characteristics from African American grandparents who are not raising a grandchild?"

METHODS

Two cycles of the NSFH were conducted, the first from 1987 through 1988 and the second from 1992 through 1994. In the first cycle 13,008 people were interviewed, 2,390 of whom were African American. African Americans and other ethnic minority groups were oversampled in the NSFH to allow subanalysis by race. To adjust for this oversampling, as well as oversampling related to nontraditional families and recently married people and to deal with the problem of nonresponse, a weighting variable was constructed by the NSFH. This weighting represents a sample that is demographically representative of the continental United States. (For a more in-depth discussion of the NSFH, see Sweet, Bumpass, & Call, 1988.) In the second cycle 10,008 people were reinterviewed, including 1,723 or 72 percent of the original African American respondents. The analyses reported here are based on the second cycle of data collection.

The original multiracial NSFH sample had 3,477 grandparents in the second cycle of data collection, 173 of whom were caregivers. (For detailed discussions of custodial caregiving in the general population, see Fuller-Thomson et al., 1997; Minkler & Fuller-Thomson,1999a; Minkler et al., 1997). The subsample here is restricted to African American grandparents who were raising a grandchild during the 1990s (n = 78) and a comparison sample of African American grandparents who had never been primary caregivers for a grandchild (n = 485). Custodial grandparents were defined as those who replied in the affirmative to the question"For various reasons, grandparents sometimes take on the primary responsibility for raising a grandchild. Have you ever had the primary responsibility for any of your grandchildren for six months or more?" Grandparents who reported that they had begun or ended caregiving during the 1990s were the subgroup of primary interest in this investigation. To avoid confounding the analysis, grandparents who previously had provided care but who were no longer doing so during the 1990s were excluded.

Because we restricted the analyses to the African American subsample of the original survey, the weighting variable required modification. To allow for appropriately weighted comparisons in the African American subsample, we divided the weighting variable by the mean value of that variable for this group (Statistics Canada, 1996). This technique allows the overall sample size of African Americans to remain constant, but it "takes into account the unequal probabilities of selection" (p. 28). Caregivers and noncaregivers were then compared and contrasted on demographic and physical health and mental health characteristics. Chi-square tests were used for nominal and ordinal variables, and independent t tests were used for interval- and ratio-level variables.

Depression was measured using a modified 12item version of the Center for Epidemiological Studies Depression Scale (CES-D) (Radloff,1977). The full CES-D is a self report measure of current symptoms of depression, developed for use in large-scale community-based studies. The CES-D scale has excellent concurrent validity, good internal consistency, and acceptable test-retest reliability (Devins & Orme, 1985; Radloff, 1977). The correlation between the full CES-D and the 12item CES-D is 0.88 (Fuller-Thomson, 1995). The traditional cut-point that suggests clinically relevant levels of depressive symptoms in the full CESD is a score of 16 or higher. Although the 12-item CES-D used in this study has a much smaller range (0-36) than the full 20-item CES-D (0-60), we conservatively chose to retain the full CES-D cutpoint of 16. Because we retained this cut-point, the possibility of identifying respondents as depressed when they would not be so identified on the full CES-D was minimized (false positives), but the true level of depression may have been underestimated (false negatives).

RESULTS

African American custodial grandparents of the 1990s tended to begin caregiving when the grandchildren were very young. More than half (53.6 percent) began when the grandchildren were younger than one year old, and an additional 20.5 percent undertook care when the grandchildren were between the ages of one and five; 17.5 percent began caregiving when the child was between five and 10 years old, and the remaining 8.4 percent undertook care when the child was 11 years old or older.

The findings of this study indicate that there are marked differences between caregiving grandparents and noncaregiving grandparents in the African American community (Table 1). Threefourths of caregiving grandparents were widowed, divorced, separated, or never married, compared with one-half of noncaregivers. Caregivers were also more likely to be female (77 percent compared with 55 percent), to have more grandchildren (on average, 8.2 grandchildren compared with 5.3 grandchildren), and to be poorer. Almost half lived below the poverty line, compared with one-quarter of noncaregiving grandparents.

African American caregivers were significantly more likely than their noncaregiving peers to have limitations in four of the five activities of daily living (ADL) investigated (Table 2). Caregivers had more problems moving around inside the house and doing day-to-day tasks. The levels of limitation were quite substantial, with 29 percent of caregivers reporting "a lot of limitation" climbing a flight of stairs and two of five caregivers indicating that they had a lot of limitation walking six blocks. Approximately two-thirds of caregivers had at least one limitation, and the mean number of limitations was two. No significant differences were found between African American caregivers and noncaregivers with respect to self reported health status or with their reported ability to bathe, dress, and provide other personal care.

African American caregivers also had more symptoms of depression compared with their noncaregiving peers, with more than one-third of caregivers reporting clinically relevant levels of depression, compared with one-fifth of noncaregivers.

DISCUSSION

There are several limitations to this study. First, the relatively small sample size of African American grandparent caregivers in the NSFH prohibited us from taking advantage of the longitudinal character of the data and explore causation. As a result, we were able to report only on associations and had no information on whether these factors were causally related to grandparent caregiving status and, if so, in which direction the relationship flowed. Second, the NSFH did not allow us to distinguish between caregivers raising one grandchild and those raising two or more grandchildren. Smaller, nonrandom studies in New York (Joslin & Brouard, 1995), Oakland, California (Minkler & Roe, 1993), and elsewhere have suggested that many African American grandparents may be raising more than one grandchild, particularly in inner-city neighborhoods. Because the financial, emotional, and physical consequences of raising several children may be even more substantial than those associated with raising one, further study in this area is warranted. Third, the NSFH did not provide information on whether grandparents had legal custody of the grandchildren. This information is critical because grandparents raising grandchildren in the formal foster care system are eligible for significantly greater benefits than those without the "kinship care" designation. Furthermore, some grandparents without legal custody may avoid obtaining any government assistance because they fear that the child will be taken into nonfamilial foster care.

Finally, the NSFH data set did not enable an examination of why individuals become grandparent caregivers, and it did not allow us to understand the reason for the elevated risk for depression and ill health among African American grandparents compared with their noncaregiving peers.

IMPLICATIONS FOR SOCIAL WORK RESEARCH AND PRACTICE

Despite these limitations our study was able to outline a health profile of African American grandparents raising grandchildren, and that profile was troubling. Caregivers had, on average, two limitations of their ADL, twice as many as were experienced by noncaregiving grandparents. More than 50 percent had trouble climbing stairs or walking six blocks. Because 75 percent of the grandparents undertook custodial care when the child was younger than five years old, when the physical aspects of child-rearing typically are most strenuous, the consequences of these limitations may be particularly problematic.

The extent of physical limitations among African American caregiving grandparents has a number of implications for social work practice and research. Social workers who have caregiving grandparents as clients need to be attentive particularly to current health problems and the possibility that the extensive demands of caregiving may exacerbate these conditions. Devices for assistance, home modifications, and in-home support services should be made easily accessible to grandparent caregivers in need. Similarly, respite services should be made far more available to grandparents, and their child care component should include relief from some of the more physically demanding tasks of caregiving, such as bathing young children and taking them on outings.

Grandparents who are in the poorest health and those raising children they perceive as having physical or behavioral problems may be among the least likely to seek and receive counseling and other help for themselves (Burnette, 1999a; Shore & Hayslip, 1994). More targeted outreach to such grandparents and the creation of"one-stop shopping" centers, where grandparents can receive mental and physical health care and services for their grandchildren and themselves, may be critical for effective service delivery (Generations United, 1998; Minkler, 1999). An Afrocentric neighborhood-based health promotion center (Elliott Brown, Jemmott, Mitchell, & Walton, 1998) holds considerable promise for this population of older African Americans. These holistic services agencies could offer direct services, information, and referral for mental and physical health needs of caregiving grandparents, support groups, exercise and wellness workshops; coordinate babysitting exchanges; and provide legal advice on custody and access issues. African American grandparents' willingness to use social services and health centers and the cultural appropriateness of services they receive are enhanced by hiring staff who live in the community (Harvey & Rauch, 1997).

The high levels of depressive symptoms for Af rican American grandparent caregivers in this study (37 percent) underscores the need for highquality and culturally sensitive psychotherapeutic interventions, including support groups designed to reinforce and build on the strengths of African American grandparents and to address their unmet needs. Detroit's Project GUIDE ("Assisting Intergenerational Families," 1993), a comprehensive program of services, support, and cultural enrichment for African American grandparents and the grandchildren in their care, which grew out of the drug epidemic, provides an excellent example of an intervention that meets many of these criteria.

Social workers assisting grandparent-headed households also should stay abreast of pre-existing resources such as the grandparent information center (GIC) of the American Association of Retired Persons (AARP). The GIC, for example, maintains a database of approximately 500 grandparent caregiver support groups in all 50 states (personal communication with M. Hollidge, director, AARP GIC, June 4, 1999) and can put social work professionals and grandparents themselves in touch with programs and services in their geographic areas. If further research indicates that the grandparents' depression is primarily a grief reaction from loss associated with caregiving, the focus of social work interventions should be on coping with grief rather than managing long-term depression (Walker & Pomeroy, 1996). More effestive involvement of African American churches in the provision of support and assistance to grandparents raising grandchildren also is needed. As several observers have noted, the shame often associated with two major causes of the increase in grandparent caregiving-substance abuse and HIV/AIDS-has resulted in the tendency for many African American churches to shy away from greater involvement in this area (Generations United, 1998). Social workers can play an important role in bringing visibility to the work of churches that have helped address the needs of grandparent caregivers in these situations and in helping to find culturally appropriate and comfortable ways for other churches to follow their example.

The fact that fully half of the grandparent caregivers in this study were living below the poverty line underscores the importance of studying and responding to grandparent caregiving within a broad sociostructural framework. Indeed, many of the causes of the increase in grandparent caregiving nationally (for example, incarceration of one or both parents, the rise in single parentheaded households, and the epidemics of substance abuse and AIDS) are tied in fundamental ways to the continued and often interconnected problems of poverty and racism in our society (Gibbs,1991). The rapid increase in incarceration rates of women, for example, has affected disproportionately low-income African American women and their families (Dressel & Barnhill, 1994). Similarly, the disproportionate rates of single parent-headed households among African Americans reflect, in part, the continued loss of men's relative economic advantage as breadwinners-a loss that has been far more pronounced in the African American community as a result of institutionalized racism and oppression (Ozawa, 1994). Serious efforts to address the factors contributing to the rise in intergenerational households headed by grandparents cannot be made without a concomitant commitment to confronting these underlying problems.

Of immediate salience in this regard are the potential effects of the 1996 welfare reform legislation on low-income grandparents raising grandchildren. As Mullen (1997) has pointed out, the legislative changes incorporated in the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PL. 104-193) (which removed the entitlement status of Aid to Families with Dependent Children and replaced the latter with the Temporary Assistance to Needy Families [TANF] program), "were never designed with grandparent-headed households in mind:' Despite this oversight, a growing proportion of TANF recipients are

grandparents and other relatives, particularly in the inner city. TANF regulations that establish five-year lifetime benefit limits, impose a work requirement after two years, and require teenage mothers to live at home as a condition of receiving aid may have the effect of increasing the number of grandparents who become primary caregivers to their grandchildren (Minkler, Berrick, & Needed, 1999; Mullen, 1997). As frontline workers, social workers are among the first professionals to become aware of the ramifications of TANF for their clients. Policymakers need to be apprised of these ramifications, with suggestions on how to improve the policies so they take less of a toll on our nation's grandparents. Social workers also are well situated to alert researchers of the need for further investigation into and data collection on neglected caregiving issues (Gaston et al.,1998). Both qualitative and quantitative longitudinal studies on the effect of TANF on African American grandparents are critical as we attempt to assess the consequences of welfare reform for the growing number of intergenerational households headed by grandparents.

As suggested earlier, further social work research should attempt to examine also the legal custody status of the children being provided care by African American grandparents and its effect (mediated through differential access to resources and so forth) on grandparents' health and well-being. Finally, studies are needed to determine the extent to which the elevated depression rates found in grandparent caregivers may in fact be a function of grief as has been found with other forms of caregiving (Walker & Pomeroy,1996). Taking a cue from Walker and Pomeroy's study of the experience of caregivers of people with dementia, instruments (for example, The Grief Experience Inventory [Sanders, Mauger, & Strong, 1985] and the Beck Depression Inventory [Beck, Steer, & Garbin, 1988]) that enable the delineation of the role of grief in influencing depression scores should be used.

CONCLUSION

In focusing on grandparents as surrogate parents to their grandchildren, this study inadvertently may have reinforced the common misconception that grandparent caregiving is "a black issue" and that this role is in fact common in the African American community. As Hunter and Taylor (1998) pointed out, "research on black grandparents has historically been couched within the public discourse on black families and social policy" (p. 70). In contrast to the majority of research on white grandparents, studies of black grandparenthood tend to stress the roles of grandparents within a context of family crisis. Much further research is needed on the 71 percent of African American grandmothers and 86 percent of grandfathers (Szinovacz, 1998) who never serve as surrogate parents to their grandchildren. The results should be publicized widely to help counter these misconceptions. At the same time, however, the needs of those African American grandparents raising grandchildren are deserving of continued attention from researchers and increased attention from policymakers and social work practitioners. As Gibbs (1991) explained, issues such as grandparent caregiving in African American communities "can be better understood if they are conceptualized as an interaction between historical patterns of adaptation, current social policies, environmental stress, and coping strategies utilized by black family members" (p. 328). Davis, Aguilar, and Jackson ( 1998) suggested that social workers can play a key role in "raising substantive questions about social justice" (p. 83) and in advocating for change in policies that may adversely of fect women who make low wages and their families. As the findings of this study make clear, such advocacy should include work with and on behalf of the growing number of grandparents, many of them African American, who are raising their grandchildren.