This Article Abstract Full Text (PDF) Services Download to citation manager Citing Articles Citing Articles via HighWire PubMed PubMed Citation

Systems of Social Support in Families Who Care for Dependent African American Elders

^a Center on Minority Aging, University of North Carolina at Chapel Hill
^b Department of Human Development and Family Studies, University of North Carolina at Greensboro

Correspondence: Sharon Wallace Williams, PhD, Center on Minority Aging, University of North Carolina at Chapel Hill, 730 Airport Road, CB 3465, Chapel Hill, NC 27599-3465. E-mail: sharon_williams{at}unc.edu.

Decision Editor: Laurence G. Branch, PhD

	Abstract

TOP Abstract Systems of Support Predictors of Social Support Relationships Between Social... Research Questions Methods Results Discussion References

 
Purpose: This study examined connections (linking, compensatory, or none) between three systems of social support (informal, church, and formal). Predictors of each system were also examined. Design and Methods: A community sample of 187 caregivers who provided care to older African American participants in the Duke Established Populations for Epidemiological Studies of the Elderly was used. Multiple regression analyses were used to examine relationships between systems of social support as well as to determine predictors of each social support system. Results: Findings provide evidence for linking connections between systems of support: Informal and formal supports were linked, and church support and formal support were linked. The relationship of need as positively related to social support was limited to the care recipient's number of instrumental activities of daily living limitations being associated with formal support. Level of education was also associated with use of formal support. Implications: Cohesive family networks and network size are important factors that help determine what support caregivers receive. African American caregivers may be at risk for negative health outcomes because they are less likely to use formal support as care recipients' activities of daily living limitations increase.

Key Words: Formal support • Church support • Informal support

It is well documented that family members provide most of the informal care when community-dwelling dependent elders need help (Angel and Angel 1997; Cotler 1996; George 1987). Although replacement of this informal care by formal care is not a likely outcome, supplementing informal support with formal support can offset caregiving costs and perhaps reduce or delay the need for institutionalization of the older family member (Bass, Noelker, and Rechlin 1996; Kosloski and Montgomery 1995). Despite such benefits, care recipients and caregivers consistently use few formal support services (Caserta, Lund, Wright, and Redburn 1987; Cox and Monk 1993; Lawton, Brody, and Saperstein 1989). Similar to other caregivers, African American caregivers also use low levels of formal support. However, empirical evidence has suggested that the church may provide support to dependent African American elders and their caregivers (Caldwell, Chatters, Billingsley, and Taylor 1995; Caldwell, Greene, and Billingsley 1994; Taylor, Ellison, Chatters, Levin, and Lincoln 2000).

When caregivers, particularly African American caregivers, do seek and use church or formal support in connection with the care they provide, little is known about what background characteristics and conditions help determine whether this connection is a positive (linking) or negative (compensatory) relationship. The limited information that is available has suggested that a combination of caregiver and care recipient conditions (i.e., physical health, depression, cognitive impairment) are important determinants of social support (Bass and Noelker 1987; Houde 1998; Miller and McFall 1991). In addition to caregivers' and care recipients' conditions, background characteristics such as age, education, race, and living arrangement are also documented as important determinants of informal, church, and formal social support (Logan and Spitze 1994).

Understanding the needs of both caregivers and care recipients may also help explain if, and in what combination, they use one or more social support systems. The need concept from Andersen and Newman 1973 behavioral model of health services use provides guidance for examining the use of social support for primary caregivers. Need most often consists of physical and /or mental limitations and is a consistent predictor of formal service use by caregivers and care recipients (Houde 1998; Krause 1990; Wallace, Levy-Storms, Kington, and Andersen 1998; Wolinsky 1994).

In addition to need, findings have shown that familial context (family cohesion and family networks) is also important in the receipt of informal, church, and /or formal social support by African American families (Belgrave, Wykle, and Choi 1993; Dilworth-Anderson and Anderson 1994; Dilworth-Anderson, Williams, and Cooper 1999; Taylor and Chatters 1986b). Although some research (Hatch 1991; Martin and Martin 1978) has shown that African American families have large and extended cohesive networks to provide support to needy members, an emerging body of literature has reported that researchers can no longer assume that these large and extended networks still exist in African American families (Jarrett and Burton 1999; McDonald and Armstrong 2001; Roschelle 1997). Jarrett and Burton 1999 and Roschelle 1997, in particular, have suggested that characteristics such as age distribution, education, and living arrangements of those in the network can influence the type of support given and received. Roschelle also suggested that, unlike in the past, more African American families today rely on varied combinations of support that include the family and formal services.

The purpose of this study was to examine the multidimensional system of social support families use to care for dependent elders. We conceived the system of support as consisting of three parts or systems: (a) informal, (b) church, and (c) formal social support. These systems of social support are examined separately in this study to determine the relationship background characteristics, needs of care recipients, needs of primary caregivers, and network structure have on predicting primary caregivers' use of each system of social support. Second, we examined whether a linking, compensatory, or no relationship existed between the three systems of social support. When systems are linked (positively related), higher levels of informal social support are associated with higher levels of church social support and /or formal social support. When they compensate one for the other (negatively related), lower levels of informal support are associated with higher levels of church and /or formal support. It is also possible that the systems are not related at all.

	Systems of Support

TOP Abstract Systems of Support Predictors of Social Support Relationships Between Social... Research Questions Methods Results Discussion References

 
Informal Social Support
Care and support from family members, as well as from friends and neighbors, are typically referred to as informal social support. The structure (typically size) of the informal network is often used as an indicator of social support (Burton et al. 1995). Generally, a larger support network is considered preferable to a smaller network. However, use of size as an indicator of informal support is problematic; size may facilitate access, but it does not automatically translate into support (Lockery 1992).

Although perceived support and received support are also used as indicators of informal support, only perceived support is examined in this study. Received social support refers to support that is actually obtained, and perceived support refers to the perception of support believed to be available (Dunkel-Schetter and Bennett 1990; B. R. Sarason, Sarason, and Pierce 1990). Although both perceived and received support are negatively associated with burden, depression, and other negative outcomes, perceived support is more consistently documented to positively affect mental and physical health (Fredman, Daly, and Lazur 1995; George and Gwyther 1986; Wailing, Seltzer, and Greenberg 1997).

Church Social Support
The African American church's tradition of serving as a major source of support for individuals and families in African American communities (Caldwell et al. 1995; Lincoln and Mamiya 1990) provided guidance for including the church in this study as a source of support to caregivers. Studies have indicated that African American adults are frequently affiliated with churches and that they frequently participate in prayer and religious activities (Billingsley 1999; Lincoln and Mamiya 1990; Taylor and Chatters 1986a). Taylor and Chatters also found that church support was second to support from the family for African Americans who coped with a crisis.

In a research study on the local church as a form of support for elderly African Americans, Steinitz 1981 reported that churches provided emotional support. Taylor and Chatters 1986b also documented that support from the church is typically emotional support. They further reported that families and church members provided different types of assistance, with church members more likely to provide emotional support and family members more likely to provide instrumental support. Given the importance of the church for African Americans, the inclusion of church support in this study contributes to a more complete understanding of how African American families make use of the church in providing care to older family members.

Formal Social Support
Although help from formal services has been theorized by some researchers as culturally unacceptable in African American families and help from the church may be viewed as a more acceptable form of support (Cantor and Mayer 1978; Netting, Thibault, and Ellor 1988; Taylor and Chatters 1986a), we included formal support as a part of the system of social support used by some African American caregivers. Formal support is help provided by professionals, paid helpers, or companies who provide caregiving help. It is assistance governed by contractual rather than affiliative norms (Eustis and Fischer 1991; Miller, McFall, and Campbell 1994). Operationalization of formal support typically determines if the care recipient and/or caregiver use specific services. Typical services include home health, adult day care, support groups, transportation services, and referral services.

It is generally thought that the use of formal support can reduce some of the physical and mental stressors associated with providing care (Mui, Choi, and Monk 1998; Zarit, Gaugler, and Jarrott 1999). However, several researchers have found that the use of formal support is positively associated with institutional placement for the care recipient. It has been suggested that this positive association reflects that formal support is often used too late to be helpful in terms of lessening stress and other challenges of caregiving (Gottlieb and Johnson 2000; Miller and McFall 1991).

	Predictors of Social Support

TOP Abstract Systems of Support Predictors of Social Support Relationships Between Social... Research Questions Methods Results Discussion References

 
Background Characteristics
Typically, formal education levels are positively related to social support in general, and more specifically to the individual social support systems examined in this study. Through increased access to various sources of information, financial resources, and /or personal ability, individuals with higher levels of education may be more aware of informal and formal support opportunities (Lu 1995; Richardson 1992). Equally important, as stated by Lu 1995, individuals with high levels of education may have social network members who also have access to varying sources of information, financial resources, and/or personal ability. Although less is known about level of education and church support, it is reasonable that the same posited mechanisms for education and informal or formal support work to form a positive association between a higher level of education and use of church support.

Findings have shown that older adults typically have smaller support networks from which to obtain support (Ajrouch, Antonucci, and Janevic 2001; Rook and Schuster 1996). In contrast to the negative association between age and network size, age is positively associated with church and formal support. Most of the findings on church support and age are among older adults in general and not focused on caregivers of any age. However, these findings have consistently documented that older adults have higher levels of religiosity and affiliations with churches and other spiritual institutions (Levin, Chatters, and Taylor 1995; Picot, Debanne, Namazi, and Wykle 1997). Research findings have been more specific regarding the positive relationship between age and use of formal support. These findings have shown that older caregivers and care recipients use higher levels of formal support than their younger counterparts (Bass and Noelker 1987; Miller and McFall 1991). Although need rather than age may account for this association, older adults are the main users of formal support services.

According to findings reported by Miller and colleagues 1996, receipt of Medicaid was one of the most significant predictors of increased use of formal services among older care recipients. Medicaid appears to facilitate access to paid services by connecting poorer individuals who are not able to purchase services with formal support. Other researchers (Bass and Noelker 1987; Headen 1993) have also documented that the care recipient's receipt of Medicaid is related to the use of formal support.

Research that has included the living arrangement of care recipient and caregiver in examining use of social support has shown that less support is received from outside the informal support system when they live together (Norgard and Rodgers 1997; Tennstedt, Crawford, and McKinlay 1993). Conversely, more support is provided by the informal caregiver. In their study examining the relationship between the living arrangement of the caregiver and that of care recipient, Tennstedt and colleagues 1993 concluded that coresidence may be the most powerful factor in structuring the caregiving situation.

Need
The most consistent predictor of formal service use is need (Logan and Spitze 1994; Miller et al. 1996). Need, most often measured in terms of the physical and /or mental limitations of the care recipient, is associated with an increase in informal and formal support. Researchers have documented higher levels of activities of daily living (ADL) and instrumental ADL (IADL) limitations as the most important predictors of formal service use (Thornton, White-Means, and Choi 1993; Wallace et al. 1998). Some evidence has suggested that providing care to a cognitively impaired family member may increase the use of formal support use (Bass, McClendon, Deimling, and Mukherjee 1994; Kemper 1992). This research has suggested that behaviors such as wandering and aggression are more difficult to manage than many physical disabilities. Although the positive association between the care recipient's physical and mental limitations is well documented, less is known regarding that between the caregiver's health and use of social support.

In a study that examined the need of both the care recipient and the caregiver, Noelker and Bass 1989 found that the caregiver's level of stress, burden, and activity restriction were predictive of in-home services. In a later study, Houde 1998 also documented that the caregiver's health was an important predictor of formal service use. She further concluded that both care recipient and caregiver characteristics should be considered when examining use of formal support. In a study specific to adult day care, Wallace, Snyder, Walker, and Ingman 1992 concluded that caregiver variables were as important as care recipient need in understanding the use of adult day care.

Network Structure.
Although some earlier studies on social support equated structural characteristics such as network size, marital status, and number of children with social support, it is now recognized that structure is not equal to social support and that the two may not be highly correlated (Dunkel-Schetter and Bennett 1990; Lockery 1992). Mui and Burnette 1994 found that the number of informal helpers did not predict in-home service use. Conversely, Seeman and Berkman 1988 found that social support, or the perception of social support, increases with larger network sizes.

Sociohistorical evidence has suggested that it is through a strong sense of family cohesion and connectedness that African American families have been able to support dependent family members (Franklin 1997; Martin and Martin 1978). In a research study of family competence and adjustment to disability in a sample of African Americans, Alston and McCowan 1995 found that family cohesion was positively related to adjustment. They further inferred that the strong family cohesion in these African American families suggested available family members to assist in managing tasks and adjustment to disability. Syrjala, Chapko, Vitaliano, Cummings, and Sullivan 1993 suggested that families high in cohesion may mobilize the resources necessary to adjust to health challenges (bone marrow transplant in their study). Research has also suggested that strong cultural values and beliefs about helping and giving support encourage cohesiveness among African American families (Dilworth-Anderson et al. 1999; Jarrett and Burton 1999; Taylor and Chatters 1986b).

	Relationships Between Social Support Systems

TOP Abstract Systems of Support Predictors of Social Support Relationships Between Social... Research Questions Methods Results Discussion References

 
Findings regarding the relationship between informal support and formal support are mixed. Some evidence has suggested a positive relationship (Luckey 1994; Taylor, Neighbors, and Broman 1989), and other evidence has suggested a negative relationship (Houde 1998; Stoller and Cutler 1993). Luckey 1994, in a study of support provided by different generations of African American kin, found that roles and tasks performed by grandchildren, nieces, or nephews assisted elderly care recipients as well as primary caregivers in interacting with formal services. Other researchers (Auslander and Litwin 1990; Bass and Noelker 1987) have also documented that informal support is important in affirming the need for formal assistance, providing referrals, and linking care recipients and caregivers to formal systems. Conversely, studies that documented a negative relationship between informal and formal social support found that lower levels of informal support were significantly related to nursing home use and paid help in the community (Miller and McFall 1991; Stoller and Cutler 1993).

Often positioned as a conduit between families and formal services (Lincoln and Mamiya 1990), African American churches, as vehicles of social support, can serve as gatekeepers and connectors for members and needed community-based services (Taylor and Chatters 1986a). For example, ministers serve as conveyers of health-related information about formal organizations that assist elderly people and those who care for them (Levin 1986; Wood and Parham 1990). More specifically, the church can function with formal and informal social support systems to (a) act as bridges, linking individuals and families to formal services, and/or (b) limit the use of formal services (Hendrickson 1986; Netting et al. 1988).

	Research Questions

TOP Abstract Systems of Support Predictors of Social Support Relationships Between Social... Research Questions Methods Results Discussion References

 
Two central questions guided this study:

1. What are the relationships between four sets of predictors (background characteristics of care recipients and caregivers, need of the care recipients, need of the caregivers, and network structure) and each system of support (informal, church, and formal)?
   
2. How are the three systems of social support related to one another? Specifically, can use of informal support predict use of church support or formal support, and can use of church support predict use of formal support?
   

In this study, we subdivided into four sets variables identified in the social support literature as potentially predictive of the use of informal, church, and /or formal support by caregivers in order to model the relative contribution of each set to each of the three systems of support. The first set (background characteristics of care recipients and caregivers) identifies factors that are part of the context within which caregivers make decisions to use (or not to use) different systems of support. The second and third sets, the need of both care recipients and caregivers, are related to higher levels of social support and may help determine which system or combination of systems of support is used. The fourth set, the structure of the caregiving network and family closeness, can further help explain how structural features of the caregiving network are related to the different systems of support. Last, an examination of how different systems of support (informal, church, and formal) are related to each other is needed to assess whether there is a linking or compensatory connection between the systems of support. We therefore hypothesized the following:

1. Higher levels of informal support are associated with higher levels of education, living separately from the care recipient, higher levels of care recipient and caregiver need, higher levels of family cohesion, and larger network size.
   
2. An increased likelihood of use of church and formal support is associated with higher levels of education, living separately from the care recipient, higher levels of care recipient and caregiver need, higher levels of family cohesion, and larger network size.
   
3. Higher levels of informal support link caregivers with use of church and formal support, and use of church support links caregivers with use of formal support.
   

	Methods

TOP Abstract Systems of Support Predictors of Social Support Relationships Between Social... Research Questions Methods Results Discussion References

 
Study Participants
The study participants were 187 primary caregivers for older family members with functional and /or cognitive impairment (care recipients, in this study). Caregivers were selected in a four-stage process.

Stage 1.
Care recipients were selected from the African American sample members of the Duke Established Populations for Epidemiological Studies of the Elderly (EPESE; Cornoni-Huntley, Blazer, Service, and Farmer 1990). The Duke EPESE is a longitudinal study with a stratified random sample of 4,162 (54% African American; n = 2,261) noninstitutionalized older persons aged 65 or older living in five contiguous counties in the Piedmont area of North Carolina (Cornoni-Huntley et al. 1990).

Stage 2.
Because we wanted to use the latest available information on the potential respondents, the sample of care recipients was restricted to African American EPESE participants meeting criteria for functional and /or cognitive impairment at the follow-up interview 6 years after baseline. Criteria for selection were inability to perform two or more basic ADL (Branch, Katz, Kniepmann, and Papsidero 1984) and /or a score of 3 or more on the Short Portable Mental Status Questionnaire, a brief screen of cognitive functioning (Pfeiffer 1975). This reduced the sample size to 497.

Stage 3.
Of the 497 African American elderly persons in the EPESE study who met the above criteria, 31% had died, 9% had been institutionalized, 4% could not be contacted, 3% refused to participate, and 6% had no caregiver when this study began in 1995. This resulted in a sample of 234 elderly African Americans who were eligible for the study. The EPESE study (1992–1993) provided information on sociodemographic characteristics, physical health status, level of cognitive impairment, and ability to perform activities of daily living for this group of 234 elderly people.

Stage 4.
Letters were sent to all eligible participants (n = 234). Some of the elderly participants could be contacted directly (n = 124), and others were contacted through a proxy (n = 110). The letters informed them of our study and how we obtained their names, addresses, and phone numbers. A follow-up telephone call was made to schedule a time for a short telephone interview. This screening telephone interview with the elderly participants or a proxy respondent was conducted to determine if they had a primary person who helped provide them with care and support. Once identified, this person was contacted by phone to determine specific caregiving roles and responsibilities, and an in-person interview was scheduled. Ten percent of the 234 caregivers refused to participate, 2% could not be contacted, and 2% of the care recipients had died or been institutionalized. As a result, our sample consisted of 202 caregivers. Fifteen of these caregivers provided limited specialized support and were not included in this study. The other 187 were primary caregivers and had main responsibilities and/or provided the most care for the care recipient. In-person interviews were conducted with all 187 primary caregivers included in this study.

Measures: Dependent Variables
Informal Social Support.
The "perceived availability" score from the short version of the Social Support Questionnaire (I. G. Sarason, Sarason, Shearin, and Pierce 1987) was used to operationalize informal social support. Caregivers were asked to list the people whom they counted on (a) to be dependable, (b) to help them feel relaxed, (c) to totally accept them, (d) to care about them, (e) to help them feel better, and (f) to console them. The number of people listed (up to nine) for each question was summed and averaged. Scores ranged from 1 to 9, with a higher score indicating greater perceived availability (B. R. Sarason et al. 1990). A test–retest correlation of .90 and a Cronbach's alpha of .90 were previously reported for the short form of the Social Support Questionnaire (I. G. Sarason et al. 1987). The Cronbach's alpha for this study was .85.

Church Social Support.
Church support was defined as use of any of 10 listed possible types of assistance caregivers received from the church. Because of the small percentage of caregivers who used any type of church support, the variable was dichotomized into used versus did not use church support. The list of possible church supports included the following: help you during times when you are ill, pray for you, give advice, give encouragement, bring you food or gifts, give you financial assistance, send a spiritual leader or minister to visit, come to visit, provide you with transportation, and help care for care recipient. If caregivers reported they used any 1 or a combination of any of the 10 types of assistance, this was coded 0, indicating use of church support. If caregivers did not report using any of the 10 types of assistance, this was coded 1, indicating did not use church support.

Formal Social Support.
Formal support was operationalized as the use by caregivers of any of 16 formal support services during the 6 months preceding the interview. These services included Meals on Wheels, congregate meals, counseling, information about services and referral, social or recreational services, legal services, financial advice or financial planning, adult daycare, homemaker, transportation, shopping help, respite care, home health, nursing care, support groups, and/or treatments or therapy (occupational, speech, etc.). As with church support, only a small percentage of caregivers reported use of formal support. The use of formal support was therefore recoded as a dichotomous variable. If caregivers reported they used any one or a combination of any of the 16 support services in the past 6 months, this was coded 0, indicating use of formal support. If caregivers did not report using any of the 16 support services, this was coded 1, indicating did not use formal support.

Measures: Independent Variables
Background Characteristics.
Background characteristics tested as independent predictors of social support use included caregiver's educational level, care recipient's age, whether the care recipient received Medicaid, and care recipient's living arrangement. Caregiver educational level was categorized as less than high school, high school, 1–3 years college, or college degree or greater. Less than high school was the referent category. Age of the care recipient was measured in years. Receipt of Medicaid by care recipient was coded as 1 = yes and 0 = no. Living arrangement was coded 1 when the care recipient and caregiver lived together and 0 when they did not.

Care Recipient Need.
The set of variables measuring care recipient need included cognitive status and physical dependency measures. Cognitive status was measured by the Short Portable Mental Status Questionnaire (Pfeiffer 1975). Sample questions included "What is the date today?" "What day of the week is it?" and "Who is president of the U.S. now?" Scores ranged from 0 to 10, with a score of 0 indicating no impairment and a score of 10 reflecting the highest level of impairment. The Cronbach's alpha for this measure was previously reported as .82 (Pfeiffer 1975) and was also .82 for this study.

The level of physical limitation among care recipients was assessed in terms of the number of limitations in basic ADL and IADL from the Duke Older Americans Resources and Services Questionnaire (Fillenbaum 1988). Primary caregivers indicated the care recipient's level of impairment in six ADL areas and seven IADL areas. The caregiver indicated whether the care recipient was unable to perform the task (coded as 2), able to perform the task with help (coded as 1), or able to perform the task without help (coded as 0). Higher scores indicated greater physical limitations. Reported reliability coefficients were .84 for the ADL scale and .87 for the IADL scale (Fillenbaum 1988). Cronbach's alpha levels of .77 and .82, respectively, were found for the ADL and IADL scales in this study.

Primary Caregiver Need.
The set of variables measuring primary caregiver need included an emotional well-being and general health measure. Emotional well-being was measured by dichotomizing the global severity index of the Brief Symptoms Inventory (Derogatis and Spencer 1982). The Brief Symptoms Inventory subscales had Cronbach's alphas ranging from .71 to .85. Cronbach's alphas on the nine subscales for this study ranged from .71 to .90. Caregivers were classified as distressed if they had a T score of 63 or greater on two or more of the nine subscales.

The general health subscale of the short form of the Rand Health Survey was used to measure the caregiver's physical health (Stewart, Hays, and Ware 1988). The general health component had five questions. Questions included "I seem to get sick a little easier than other people" (responses: definitely true, mostly true, mostly false, and definitely false) and "In general, would you say your health is excellent, very good, good, fair, or poor?" Scores were converted to percentile scores ranging from 0 to 100, with a higher score indicating a more favorable health state. McHorney, Ware, Lu, and Sherbourne 1994 documented a reliability coefficient for general health of .93. The alpha level for primary caregivers for our sample was .91.

Network Structure.
Two measures of network structure were examined. Network size was operationalized as the total number of people the primary caregiver listed as providing help. The second measure was the family cohesion subscale of the Family Environment Scale (Moos 1974), which included nine true/false items concerning the level of commitment and help provided by family members. Responses indicating high levels were coded 1, and the alternative responses were coded 0. Scores were summed, with higher scores indicating higher family cohesion. The reported Cronbach's alpha level on this scale was .74 (Moos 1974). The Cronbach's alpha level for our study was .80.

Statistical Analysis
Three hierarchical regression equations (linear and logistic) were used to analyze the three systems of social support (informal, church, and formal). We expected that similar factors as well as unique variables would predict each outcome. The first equation analyzed the continuous outcome of informal social support. The next two equations analyzed dichotomous outcomes (the caregiver's use of church support and the caregiver's use of formal support). To assess the relative contribution of each set of independent variables, the initial four sets of predictors (background characteristics, care recipient need, caregiver need, and network structure) were entered into each equation. Therefore, in each of the three equations, a background characteristics set (Model 1) was entered first to control for the relationship on the dependent variable. Evidence, some of which was discussed earlier, has shown that need is a major determinant of social support; therefore, the care recipient's need was entered as Model 2 and the caregiver's need was entered next as Model 3. Network structure was then added (Model 4) to assess the relationship of network size and family cohesion on each of the dependent variables.

In examining use of church support, an additional model (Model 5) added informal support. This permitted examination of the relationship between informal support and church support. The third equation (formal support) included the initial four models and Model 5 (informal support) and added a sixth model (church support). This permitted examination of the relationships between both informal and formal support and between church and formal support. For each of the three equations used in this study to examine informal, church, and formal support, a backward elimination procedure was used with a significance level of .05 to limit the number of nonsignificant variables. Variance inflation factors were used to rule out multicollinearity problems among the variables. All analyses were conducted with Version 3.2 of the JMP Statistical Program from SAS (SAS, 1997). A significance level of .05 was used to determine statistical significance.

	Results

TOP Abstract Systems of Support Predictors of Social Support Relationships Between Social... Research Questions Methods Results Discussion References

 
Profile of Sample
The caregivers were typically middle-aged (M = 55.5 years, SD = 13.6) daughters (49%) who provided care for their mothers. Eighty-two percent of the primary caregivers were female (see Table 1 ). Twenty-four percent of the caregivers reported a high school degree, 32% reported some college or more, and 44% reported less than a high school degree. Forty-three percent of the caregivers were married (or living as married), and they most often provided care to parents (62%), mothers in particular. Fifty-eight percent of the primary caregivers lived with the care recipient.

Table 4 shows results of the hierarchical linear regression models of the relationship between sets of independent variables and the three systems of social support. Parameter estimates (betas) are reported for the linear regression models used to predict informal support, and odds ratios (OR) are reported from the logistic regression models used to predict church support and formal support. Confidence intervals (CI) at the 95% level are not included in Table 4 but are provided in the text for the betas and OR estimates. Also shown in Table 4 are the R² values and change in the R² values at each step, the chi-square values, and the change in chi-square values at each step. The change in R² at each step represents the variance in the informal social support accounted for by the set of variables entered at that step. The change in chi-square at each step represents the difference between the previous chi-square value and the new chi-square value. A statistically significant change indicates that the addition of the new set significantly contributed to the overall fit of the model.

Church Support
The logistic equation used to predict church support shows that from the background characteristics set (Model 1) only receipt of Medicaid met the criterion for inclusion in the equation for church social support. This resulted in a significant chi-square of 8.25 (df = 1, N = 178, p .004). Caregivers of persons receiving Medicaid were less likely (OR = .27, 95% CI = .09–.67, p = .009) to use church support than caregivers of persons who were not receiving Medicaid (Table 4 ). Results also show that neither the care recipient need set nor the caregiver need set of variables (Models 2 and 3) entered into the equation. However, similar to the equation that analyzed informal social support, entry of the network structure set (Model 4) of variables yielded a significant chi-square change of 16.42 (df = 1, N = 178, p < .005). As network size increased, caregivers were 1.5 times more likely to use church support (OR = 1.53, 95% CI = 1.23–1.94, p = .0002). Although informal support entered into the equation, the chi-square change was not significant at the .05 level of significance. We obtained a whole model chi-square of 27.49 (df = 3, N = 178, p .0001).

Formal Support
Similar to the analysis on church support, we used logistic regression to predict use of formal support. The regression equation used to predict formal support shows that a significant chi-square of 10.42 (df = 2, N = 180, p .01) was obtained for the background characteristics set (Model 1) of variables. Within this set, 1–3 years of college was significant. Those with 1–3 years of college were nearly four times more likely to be users of formal support (OR = 3.83, 95% CI = 1.7–8.9, p = .001) than those without a high school degree. Findings in this equation also show that the care recipient's number of ADL limitations and IADL limitations (Model 2) entered the equation and resulted in a significant chi-square change of 6.18 (df = 2). Although the number of ADL was not significant, the number of IADL was significant at the .05 level. Caregivers were more likely (OR = 1.17, 95% CI = 1.03–1.34, p = .017) to use formal support when they cared for care recipients with higher levels of IADL limitations.

Only family cohesion from the network structure set of variables (Model 4) met our criterion for inclusion in the analysis to predict formal support; however, it did not result in a significant chi-square change. Significant predictors of use of formal support at this stage in the analysis included two levels of education and care recipient's IADL. Caregivers with 1–3 years of college and those with a high school degree were more likely to use formal support than those without a high school degree. Also, as the care recipient's number of IADL limitations increased, caregivers were more likely to use formal support.

The last two sets of variables (Models 5 and 6) that entered into the analysis were informal support and church support. Both sets resulted in significant chi-square changes. Results show that caregivers with higher levels of informal support were more likely to use formal support (OR = 1.30, 95% CI = 1.06–1.60, p = .009). Findings also show that caregivers who used church support were nearly three times more likely to use formal support (OR = 2.86, 95% CI = 1.24–6.68, p = .013). With all the sets entered, two levels of education, along with number of ADL and IADL limitations, family cohesion, informal support, and church support were significant predictors of formal social support. Paradoxically, caregivers who provided care to care recipients with higher levels of ADL were less likely to use formal support than those who cared for less physically impaired elders. The chi-square for the whole model was 31.64 (df = 7, N = 180, p < .0001).

	Discussion

TOP Abstract Systems of Support Predictors of Social Support Relationships Between Social... Research Questions Methods Results Discussion References

 
This study of 187 African American caregivers of impaired elderly family members examined potential predictors (background characteristics of care recipients and caregivers, needs of care recipients and caregivers, and structure of the caregiving network) of three systems of social support. Connections between the three systems of social support (informal, church, and formal) were also examined.

In regard to certain background characteristics that help determine caregivers' use of each system of support, current findings highlight the importance of education in seeking and locating social support. As in other studies, higher education levels in this study were associated with use of formal social support (Luckey 1994; Miller et al. 1996; Stone, Cafferata, and Sangl 1987). This greater usage of formal services may exist because caregivers with more education are more aware of services and /or more resourceful. Thus, they are able to better identify, access, and use services to address the challenges of caregiving.

Findings regarding use of Medicaid differ from the literature. Receipt of Medicaid was associated with church social support. However, this association was negative, with caregivers of elders receiving Medicaid being less likely to use church support. Living arrangement was not associated with any of the three support systems examined in this study. This finding differs from what most researchers report. Living arrangement is typically documented as one of the most important variables in explaining use of social support for both caregivers and care recipients, particularly formal support (Miller et al. 1996; Penrod, Kane, Kane, and Finch 1995; Tennstedt et al. 1993). The absence of significance for living arrangement in the present study may reflect the historical presence of multigenerational households in African American families (Coontz 2000; Dilworth-Anderson et al. 1999; Martin and Martin 1978). Living arrangements for these caregivers and care recipients may have been in existence before dependency of the care recipient, and therefore were not related to informal, formal, or church support.

Although we hypothesized that greater need would be associated with higher levels of support and a combination of different systems of support, this was only found for formal support for the caregiver. When the care recipient had greater IADL limitations, the caregiver was more likely to use formal support. In contrast, although the care recipient's number of ADL limitations was also associated with use of formal support for the caregiver, we found that caregivers for elders with more ADL limitations were less likely to use a formal support service. Perhaps caregiving tasks associated with higher functional limitations of the care recipient left less time for involvement with formal support. This finding may reflect sociohistorical issues regarding lack of access to and limited quality of support services given to African Americans in this society.

Similar to findings reported by Lyons, Zarit, and Townsend 2000, increased need of the care recipient was associated with receipt of less support. In both cases, higher levels of need, behavioral problems in their study, and increased number of ADL in this study were associated with caregivers being less likely to use formal support. These caregivers may be missing help that could offset caregiving tasks and lessen demands on them. In addition, this lack of support may have implications for increased negative emotional (role strain, depression) and physical health (lower general health, increased morbidity) outcomes (Bass et al. 1996; Schulz et al. 1997; Stephens, Franks, and Townsend 1994).

Also consistent with our hypotheses and the reported literature (Burton et al. 1995; Oxman and Hull 1997), as network size increased, an association was found in this study with higher levels of informal social support and with being more likely to use church support. Caregivers in more cohesive networks also reported higher levels of informal social support, but they were less likely to use formal support. We believe this result suggests that some African American families become insulated in their caregiving, which may have a sociohistorical basis. African Americans have historically been denied access to formal support, and the support that was provided was often inferior (Franklin 1997). Over time, individuals and families learned to adapt by creating networks, usually in the extended family, that could absorb the needs of dependent family members (Billingsley 1999; Martin and Martin 1978). However, as noted earlier, recent evidence has shown that contemporary African American families are finding it increasingly difficult to claim the strength and resilience of the kin network of the past (Jarrett and Burton 1999; McDonald and Armstrong 2001; Roschelle 1997).

Although families with higher levels of cohesion were less likely to use formal support, other findings from this research on primary family caregivers of impaired elders highlight the interdependence of social support systems in caregiving families. As hypothesized, results show a linking relationship between higher levels of informal support and the increased likelihood that caregivers will use formal support. Thus, similar to other studies, caregivers who perceived more informal support were more likely to use formal support. Taylor and colleagues 1989, in a sample of African Americans who were confronted with a serious personal problem, also documented informal helpers (relatives and friends) as a link or bridge to social service providers. Other researchers have also noted that family members mediate and coordinate linkages between informal and formal support (Chappell 1985; Luckey 1994). Similarly, this study reported a linking relationship between informal and formal support. The hypothesized linking relationship between informal support and church support was not supported. A linking relationship was found, however, between church and formal support; the church served to link caregivers with formal support. This ability of the church to link its members with formal support services has been found by other researchers (Caldwell et al. 1995; Taylor and Chatters 1986a, Taylor and Chatters 1986b; Walls and Zarit 1991). This finding has implications for assisting churches in their efforts to support caregivers by linking them to formal services in the communities.

However, additional research is needed to further examine the other ways in which informal, church, and formal support services interface beyond linking and compensatory arrangements. Lyons and associates 2000 and Lyons and Zarit 1999 suggested that further research is needed to assess other types of interfaces (formal services only, isolates) and also to determine whether the ways in which informal and formal caregivers perform caregiving tasks are linked by supplementary or complementary relationships. We believe that their suggestions and findings, along with the findings in this study, point out the complex and varied ways in which families meet the demands of caregiving.

Although this study has numerous strengths, there are important limitations. First, the cross-sectional design of our analyses did not allow for examining the links and connections of the different support systems over time. Therefore, future analyses, using a longitudinal design, would allow for examining changing conditions in the sociodemographic status of caregivers, the needs of both caregivers and care recipients, and how each system of support responds to these changes. Also, as suggested by Berry, Zarit, and Rabatin 1991, future analyses need to compare varying types of formal services to assess if there are differential relationships between families and specific formal services. Second, although our distress measure was a global assessment of emotional well-being, a more direct measure of stressors related to caregiving, such as role strain, may better assess the relationship of stress to the use of support services. Last, this study's findings are based on responses by a single primary caregiver at one point in time and do not indicate whether (or how) other caregivers in the network may replace and assist the primary caregiver. As suggested by Gaugler, Kane, and Langlois 2000, when multiple family caregivers are involved in the caregiving process, strategies are needed for measuring responses from all caregivers. With these limitations in mind, we believe that the strengths of this study, reflected in the representative sample used, the testing of instruments for cultural relevance, and the conceptual grounding of this study, provide valuable and needed information on African American families who provide care to dependent elders.

The Forum

Book Reviews

Practice Concepts

	Acknowledgments

 
This study was funded by National Institute on Aging Grant RO1 AG12268-01. We would like to thank Dan Blazer (principal investigator) and the Duke Established Populations for Epidemiologic Studies of the Elderly, National Institute on Aging (NIA) Contract N02-AG-1-2102, for providing additional support. Support was provided to Sharon Wallace Williams by the Claude D. Pepper Older Americans Independence Center of Wake Forest University (through Grant P60 AG10484) and by a National Institute on Aging training grant (T32 AG00182). We would like to thank Gerda Fillenbaum and Baila Miller for their invaluable comments on drafts of this article.

Received for publication March 7, 2001. Accepted for publication November 8, 2001.

	References

TOP Abstract Systems of Support Predictors of Social Support Relationships Between Social... Research Questions Methods Results Discussion References

 

• Ajrouch K. J., Antonucci T. C., Janevic M. R., 2001. Social networks among Blacks and Whites: The interaction between race and age. Journal of Gerontology: Social Sciences 56B:S112-S118. [Abstract/Free Full Text]
• Alston R. J., McCowan C. J., 1995. Perception of family competence and adaptation to illness among African Americans with disabilities. The Journal of Rehabilitation 61:27-32.
• Andersen R. M., Newman J. F., 1973. Societal and individual determinants of medical care utilization in the United States. Milbank Memorial Fund Quarterly 51:95-124.
• Angel R., Angel J., 1997. Who will care for us? Aging and long-term care in multicultural America New York University, New York.
• Auslander G. K., Litwin H., 1990. Social support networks and formal help seeking: Differences between applicants to social services and a nonapplicant sample. Journal of Gerontology: Social Sciences 45:S112-S119.
• Bass D. M., McClendon M. J., Deimling G. T., Mukherjee S., 1994. The influence of a diagnosed mental impairment on family caregiver strain. Journal of Gerontology: Social Sciences 49:S146-S155.
• Bass D. M., Noelker L. S., 1987. The influence of family caregivers on elders' use of in-home services. Journal of Health and Social Behavior 28:184-196. [Medline]
• Bass D. M., Noelker L. S., Rechlin L. R., 1996. The moderating influence of service use on negative caregiving consequences. Journal of Gerontology: Social Sciences 51B:S121-S131. [Abstract]
• Belgrave L. L., Wykle M. L., Choi J. M., 1993. Health, double jeopardy, and culture: The use of institutionalization by African Americans. The Gerontologist 33:379-385. [Abstract]
• Berry G. L., Zarit S. H., Rabatin V. X., 1991. Caregiver activity on respite and nonrespite days: A comparison of two service approaches. The Gerontologist 31:830-835. [Abstract]
• Billingsley A., 1999. Mighty like a river: The Black church and social reform Oxford University Press, New York.
• Branch L. G., Katz S., Kniepmann K., Papsidero J., 1984. A prospective study of functional status among community elders. American Journal of Public Health 74:266-268. [Abstract/Free Full Text]
• Burton L., Kasper J., Shore A., Cagney K., LaVeist T., Cubbin C., German P., 1995. The structure of informal care: Are there differences by race?. The Gerontologist 35:744-752. [Abstract]
• Caldwell C. H., Chatters L. M., Billingsley A., Taylor R. J., 1995. Church-based support programs for elderly Black adults: Congregational and clergy characteristics. Kimble M. A., McFadden S. H., Ellor J. W., Seeber J., , ed.Aging, spirituality, and religion: A handbook 306-324. Fortress, Minneapolis, MN.
• Caldwell C. H., Greene A. D., Billingsley A., 1994. Family support programs in Black churches: A new look at old functions. Kagan S. L., Weissbourd B., , ed.Putting families first: America's family support movement and the challenge of change 137-160. Jossey-Bass, San Francisco.
• Cantor M. H., Mayer M., 1978. Factors in differential utilization of services by urban elderly. Journal of Gerontological Social Work 1:57-62.
• Caserta M., Lund D., Wright S., Redburn P., 1987. Caregivers to dementia patients: Use of community services. The Gerontologist 27:209-213. [Medline]
• Chappell N. L., 1985. Social support and the receipt of home care services. The Gerontologist 25:47-54. [Medline]
• Coontz S., 2000. Historical perspectives on family studies. Journal of Marriage and Family 62:283-298.
• Cornoni-Huntley, J., Blazer, D. G., Service, C., & Farmer, M. E. (1990). Introduction. In J. Cornoni-Huntley, D. G. Blazer, M. E. Lafferty, D. F. Everett, D. B. Brocik, & M. E. Farmer (Eds.), Established populations for epidemiologic studies of the elderly, Vol. 2: Resources data book (p. 17; NIH Publication No. 90–495) Washington, DC: National Institute on Aging.
• Cotler M., 1996. Editor's introduction: Chronic illness and current perspectives: What are the problems for patients, families, and other caregivers?. American Behavioral Scientist 39:647-654. [Abstract]
• Cox C., Monk A., 1993. Hispanic culture and family care of Alzheimer's patients. Health and Social Work 18:92-101.
• Derogatis L. R., Spencer P. M., 1982. The Brief Symptom Inventory: Administration and procedures manual Johns Hopkins University School of Medicine, Clinical Psychometric Research Unit, Baltimore.
• Dilworth-Anderson P., Anderson N., 1994. Dementia caregiving in Blacks: A contextual approach to research. Light E., Niederehe G., Lebowitz B., , ed.Stress effects on family caregivers of Alzheimer's patients 385-409. Springer, New York.
• Dilworth-Anderson P., Williams S. W., Cooper T., 1999. Family caregiving to elderly African Americans: Caregiver types and structures. Journal of Gerontology: Social Sciences 54B:S237-S241. [Abstract]
• Dunkel-Schetter C., Bennett T. L., 1990. Differentiating the cognitive and behavioral aspects of social support. Sarason B. R., Sarason I. G., Pierce G. R., , ed.Social support: An interactional view 267-296. Wiley, New York.
• Eustis N. M., Fischer L. R., 1991. Relationships between home care clients and their workers: Implications for quality of care. The Gerontologist 31:447-456. [Abstract]
• Fillenbaum G. G., 1988. Multidimensional functional assessment of older adults: The Duke Older Americans Resources and Services Procedures Erlbaum, Hillsdale, NJ.
• Franklin D., 1997. Ensuring inequality: The structural transformation of the African American family Oxford University Press, New York.
• Fredman L., Daly M. P., Lazur A. M., 1995. Burden among White and Black caregivers to elderly adults. Journal of Gerontology: Social Sciences 50B:S110-S118. [Abstract]
• Gaugler J. E., Kane R. A., Langlois J., 2000. Assessment of family caregivers of older adults. Kane R. L., Kane R. A., , ed.Assessing older persons: Measures, meaning, and practical applications 320-359. Oxford University Press, New York.
• George L. K., 1987. Easing caregiver burden: The role of informal and formal supports. Ward R. A., Tobin S. S., , ed.Health in aging: Sociological issues and policy directions 133-158. Springer, New York.
• George L. K., Gwyther L. P., 1986. Caregiver well-being: A multidimensional examination of family caregivers of demented adults. The Gerontologist 26:253-259. [Medline]
• Gottlieb B. H., Johnson J., 2000. Respite programs for caregivers of persons with dementia: A review with practice implications. Aging and Mental Health 4:119-129.
• Hatch L. R., 1991. Informal support patterns of older African American and White women. Research on Aging 13:144-170. [Abstract]
• Headen A. E., Jr. 1993. Economic disability and health determinants of the hazard of nursing home entry. Journal of Human Resources 28:80-110.
• Hendrickson S. L., 1986. Churches as geriatric health clinics for community based elderly. Journal of Religion and Aging 2:13-24.
• Houde S. C., 1998. Predictors of elders' and family caregivers' use of formal home services. Research in Nursing and Health 21:533-543.
• Jarrett R. L., Burton L. M., 1999. Dynamic dimensions of family structure in low-income African American families: Emergent themes in qualitative research. Journal of Comparative Family Studies 30:177-187.
• Kemper P., 1992. The use of formal and informal care by the disabled elderly. Health Services Research 27:421-451. [Medline]
• Kosloski K., Montgomery R. J., 1995. The impact of respite use on nursing home placement. The Gerontologist 35:67-74. [Abstract]
• Krause N., 1990. Illness behavior in later life. Binstock R. H., George L. K., , ed.Handbook of aging and the social sciences 227-245. Academic Press, San Diego, CA.
• Lawton M. P., Brody E. M., Saperstein A. R., 1989. A controlled study of respite service for caregivers of Alzheimer's patients. The Gerontologist 29:8-16. [Abstract]
• Levin J. S., 1986. Roles for the Black pastor in preventive medicine. Pastoral Psychology 35: (2) 94-103.
• Levin J. S., Chatters L. M., Taylor R. J., 1995. Religious effects on health status and life satisfaction among black Americans. Journal of Gerontology: Social Sciences 50B:S154-S163. [Abstract]
• Lincoln C. E., Mamiya L., 1990. The black church in the African American experience Duke University Press, Durham, NC.
• Lockery S. A., 1992. Caregiving among racial and ethnic minority elders: Family and social supports. Stanford E. P., Torres-Gil F. M., , ed.Diversity: New approaches to ethnic minority aging 113-122. Baywood, Amityville, NY.
• Logan J. R., Spitze G., 1994. Informal support and the use of formal services by older Americans. Journal of Gerontology: Social Sciences 49:S25-S34.
• Lu L., 1995. Correlates of social support: Personal characteristics and social resources. Counseling Psychology Quarterly 8:173-182.
• Luckey I., 1994. African American elders: The support network of generational kin. Families in Society: The Journal of Contemporary Human Services 75:82-89.
• Lyons K. S., Zarit S. H., 1999. Formal and informal support: The great divide. International Journal of Geriatric Psychiatry 14:183-196. [Medline]
• Lyons K. S., Zarit S. H., Townsend A. L., 2000. Families and formal service usage: Stability and change in patterns of interface. Aging and Mental Health 4:234-243.
• Martin E. P., Martin J. M., 1978. The Black extended family University of Chicago Press, Chicago.
• McDonald K., Armstrong E., 2001. De-romanticizing Black intergenerational support: The questionable expectation of welfare reform. Journal of Marriage and Family 63:213-223.
• McHorney C., Ware J. E., Lu R., Sherbourne C. D., 1994. The MOS 36-item short-form health survey (SF-36): III. Tests of data quality, scaling assumptions, and reliability across diverse patient groups. Medical Care 32:40-66. [Medline]
• Miller B., Campbell R. T., Davis L., Turner S., Giachello A., Prohaska T., Kaufman J. E., Li M., Perez C., 1996. Minority use of community long-term care services: A comparative analysis. Journal of Gerontology: Social Sciences 51B:S70-S81. [Abstract]
• Miller B., McFall S., 1991. The effect of caregiver's burden on change in frail older persons' use of formal helpers. Journal of Health and Social Behavior 32:165-179. [Medline]
• Miller B., McFall S., Campbell R. T., 1994. Changes in sources of community long-term care among African American and White frail older persons. Journal of Gerontology: Social Sciences 49:S14-S24.
• Moos R. H., 1974. Family environment scale. Moos R. H., Insel P. M., Humphrey B., , ed.Combined preliminary manual: Family, work, and group environment scales 1-15. Consulting Psychologists Press, Palo Alto, CA.
• Mui A. C., Burnette D., 1994. Long-term care service use by frail elders: Is ethnicity a factor?. The Gerontologist 34:190-198. [Abstract]
• Mui A. C., Choi N. G., Monk A., , ed.Long-term care and ethnicity 1998Auburn House, Westport, CT.
• Netting F. E., Thibault J. M., Ellor J. D., 1988. Spiritual integration: Gerontological interface between the religious and social service communities. Journal of Religion and Aging 5:61-75.
• Noelker L. S., Bass D. M., 1989. Home care for elderly persons: Linkages between formal and informal caregivers. Journal of Gerontology: Social Sciences 44:S63-S70.
• Norgard T. M., Rodgers W. L., 1997. Patterns of in-home care among elderly Black and White Americans. Journal of Gerontology: Social Sciences Psychological Sciences 52: (Special Issue) 93-101.
• Oxman T. E., Hull J. G., 1997. Social support, depression, and activities of daily living in older heart surgery patients. Journal of Gerontology: Psychological Sciences 52B:P1-P14. [Abstract]
• Penrod J. D., Kane R. A., Kane R. L., Finch M. D., 1995The Gerontologist, 35, 489–497.and composition of the caregiver support system. Who cares?. The size scope:
• Pfeiffer E., 1975. A short portable mental status questionnaire for the assessment of organic brain deficit in elderly patients. Journal of American Geriatrics Society 23:433-441.
• Picot S. J., Debanne S. M., Namazi K. H., Wykle M. L., 1997. Religiosity and perceived rewards of black and white caregivers. Education Research 37:89-101.
• Richardson V., 1992. Service use among urban African American elderly people. Social Work 37:47-54. [Medline]
• Rook K. S., Schuster T. L., 1996. Compensatory processes in the social networks of older adults. Pierce G. R., Sarason B. R., Sarason I. G., , ed.Handbook of social support and the family 219-248. Plenum, New York.
• Roschelle A. R., 1997. No more kin: Exploring race, class, and gender in family networks Sage, Thousand Oaks, CA.
• Sarason B. R., Sarason I. G., Pierce G. R., 1990. Traditional views of social support and their impact on assessment. Sarason B. R., Sarason I. G., Pierce G. R., , ed.Social support: An interactional view 9-25. Wiley, New York.
• Sarason I. G., Sarason B. R., Shearin E. N., Pierce G. R., 1987. A brief measure of social support: Practical and theoretical implications. Journal of Social and Personal Relationships 4:497-510. [Abstract]
• SAS Institute Inc1997. JMP Statistical Discovery Software (Version 3.2) Author, Cary, NC.
• Schulz R., Newsom J., Mittelmark M., Burton L., Hirsch C., Jackson S., 1997. Health effects of caregiving: The caregiver health effects study: An ancillary study of the Cardiovascular Health Study. Annals of Behavioral Medicine 19:110-116. [Medline]
• Seeman T. E., Berkman L. F., 1988. Structural characteristics of social networks and their relationship with social support in the elderly: Who provides support. Social Science and Medicine 26:737-749.
• Steinitz L. Y., 1981. The local church as support for the elderly. Journal of Gerontological Social Work 4:44-53.
• Stephens M. A., Franks M. M., Townsend A. L., 1994. Stress and rewards in women's multiple roles: The case of women in the middle. Psychology and Aging 9:45-52. [Medline]
• Stewart A. L., Hays R. D., Ware J. E., 1988. The MOS short-form general health survey. Medical Care 26:724-735. [Medline]
• Stoller E. P., Cutler S. J., 1993. Predictors of use of paid help among older people living in the community. The Gerontologist 33:31-40. [Abstract]
• Stone R., Cafferata G., Sangl J., 1987. Caregivers of the frail elderly: A national profile. The Gerontologist 27:616-626. [Medline]
• Syrjala K. L., Chapko M. K., Vitaliano P. P., Cummings C., Sullivan K. M., 1993. Recovery after allogeneic bone marrow transplantation: Prospective study of predictors of long-term physical and psychosocial functioning. Bone Marrow Transplant 11:319-327. [Medline]
• Taylor R. J., Chatters L. M., 1986a. Church-based informal support among elderly Blacks. The Gerontologist 26:637-642. [Medline]
• Taylor R. J., Chatters L. M., 1986b. Patterns of informal support to elderly Black adults: Family, friends, and church members. Social Work 31:432-438.
• Taylor R. J., Ellison C. G., Chatters L. M., Levin J. S., Lincoln K. D., 2000. Mental health services in faith communities: The role of clergy in Black churches. Social Work 45: (1) 73-87. [Medline]
• Taylor R. J., Neighbors H. W., Broman C. L., 1989. Evaluation by black Americans of the social service encounter during a serious personal problem. Social Work 34:205-211.
• Tennstedt S. L., Crawford S., McKinlay J. B., 1993. Determining the pattern of community care: Is coresidence more important than caregiver relationship?. Journal of Gerontology: Social Sciences 48:S74-S83.
• Thornton M. C., White-Means S. I., Choi H. K., 1993. Sociodemographic correlates of the size and composition of informal caregiver networks among frail ethnic elderly. Journal of Comparative Family Studies 24:235-250.
• Wailing L., Seltzer M., Greenberg J. S., 1997. Social support and depressive symptoms: Differential patterns in wife and daughter caregivers. Journal of Gerontology: Social Sciences 52B:S200-S211. [Abstract]
• Wallace S. P., Levy-Storms L., Kington R. S., Andersen R. M., 1998. The persistence of race and ethnicity in the use of long-term care. Journal of Gerontology: Social Sciences 53B:S104-S112. [Abstract]
• Wallace S. P., Snyder J. L., Walker G. K., Ingman S. R., 1992. Racial differences among users of long-term care: The case of adult day care. Research on Aging 14:471-495. [Abstract]
• Walls C. T., Zarit S. H., 1991. Informal support from Black churches and the well-being of elderly Blacks. The Gerontologist 31:490-495. [Abstract]
• Wolinsky F. D., 1994. Health services utilization among older adults: Conceptual, measurement, and modeling issues in secondary analysis. The Gerontologist 34:470-475. [Abstract]
• Wood J. B., Parham I. A., 1990. Coping with perceived burden: Ethnic and cultural issues in Alzheimer's family caregiving. Journal of Applied Gerontology 9:345-355.
• Zarit S. J., Gaugler J. E., Jarrott S. E., 1999. Useful services for families: Research findings and directions. International Journal of Geriatric Psychiatry 14:165-181. [Medline]

This article has been cited by other articles:

				C. W. Peek, T. Koropeckyj-Cox, B. A. Zsembik, and R. T. Coward Race Comparisons of the Household Dynamics of Older Adults Research on Aging, March 1, 2004; 26(2): 179 - 201. [Abstract] [PDF]

				M. C. Nightingale Religion, Spirituality, and Ethnicity: What it Means for Caregivers of Persons with Alzheimer's Disease and Related Disorders Dementia, October 1, 2003; 2(3): 379 - 391. [Abstract] [PDF]

				P. Dilworth-Anderson, I. C. Williams, and B. E. Gibson Issues of Race, Ethnicity, and Culture in Caregiving Research: A 20-Year Review (1980-2000) Gerontologist, April 1, 2002; 42(2): 237 - 272. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Services Download to citation manager Citing Articles Citing Articles via HighWire PubMed PubMed Citation