Facilitators of Home and Community-Based Service Use by Urban African American Elders

Abstract

Objectives: To examine the factors associated with urban African American elders’ utilization of home and community-based services and explore whether these factors differ by category of service. Methods: Data came from a representative sample of 1,099 African American older adults living in Detroit. Logistic regression models were used to explore the predisposing, enabling, and need factors associated with any service use and five categories of service use: in-home care, household services, functional care, out-of-home services, and financial/legal services. Results and Discussion: Findings suggest that previous studies using the Andersen model may not be generalizable to an urban African American population. Service use was related primarily to enabling resources, particularly in terms of financial resources, not driving a car, and social resources and support. Future research should further examine the facilitators and barriers to service use by urban African Americans to improve their access to care.

Keywords

service utilization behavioral model African Americans

Introduction

Research on the use of home and community-based services (HCBS) by African American older adults suggests that the influential factors in this group may differ from those influencing service use by Whites. Some studies report that African American older adults are more likely to use HCBS than their White counterparts (Henton, Hays, Walker, & Atwood, 2002; White-Means & Rubin, 2004). These findings are primarily limited to in-home health care and have been attributed to higher prevalence levels of disability and chronic illness among African Americans (Dunlop, Manheim, Song, & Chang, 2002). Other studies have found that African Americans are less likely to use formal HCBS than White older adults (Calsyn & Winter, 1999; Houde, 1998; Wallace, Fields, Witucki, Boland, & Tuck, 1999). Researchers attribute these lower rates of service use by African Americans to a variety of factors, including a lack of awareness of available services (Calsyn & Winter, 1999), distrust of formal health providers (Musa, Schulz, Harris, Silverman, & Thomas, 2009), beliefs that community agencies are not knowledgeable of and responsive to their needs (Scharlach, Kellam, Ong, Baskin, Goldstein, & Fox, 2006), norms of familial responsibility for care (Bradley et al., 2002; Connell & Gibson, 1997), and fewer economic resources, particularly in relation to the use of HCBS that are not covered by Medicare or Medicaid (Dunlop et al., 2002).

Framed by the Behavioral Model of Health Services and Access to Care (Andersen, 1995) and using data from a representative sample of older adults living in Detroit, Michigan, this study addresses gaps in the literature for urban African American elders by examining the factors associated with their utilization of HCBS and exploring whether these factors differ by category of service. Facilitators and barriers to service utilization among older African Americans may potentially differ from those of older Whites and thus should be examined separately. With the exception of a study of urban African Americans two decades ago (Richardson, 1992) research on HCBS utilization includes predominantly White participants (e.g., Houde, 1998; Mitchell & Krout, 1998) and therefore little is known about African Americans, such as those living in urban neighborhoods.In addition, different factors may play different roles depending on the type of HCBS examined. For example, health status may have the greatest influence on the use of home health care, while other economic and social factors may be associated with the utilization of out-of-home services, such as congregate meals and senior centers (Johnson & Wolinsky, 1996). Few studies, however, have explored whether predictors of service use vary between types of HCBS (for exceptions see Johnson & Wolinsky, 1996; Mitchell & Krout, 1998). It is important to understand potential differences in the facilitators of different types of HCBS because, as highlighted in the Older Americans Act, which is a major source of assistance for older adults, a wide array of programs contributes to older adults maintaining their health, independence and dignity, and avoiding unnecessary institutionalization (Administration on Aging, 2012).

Background and Significance

Nearly all older adults want to remain in the community and avoid institutionalization (Feldman, Oberlink, Simantov, & Gursen, 2004), and evidence shows that the use of HCBS can delay or prevent nursing home placement (Gaugler, Kane, Kane, & Newcomer, 2005; Jette, Tennstedt, & Crawford, 1995). With the increasing availability of HCBS (Shirk, 2006) and lessened ability of families to meet the assistance needs of older adults (Spillman & Pezzin, 2000), it is important to examine the factors associated with the use of HCBS by older adults.

Current demographic changes call attention to the need to understand the particular facilitators of HCBS use among urban African American elders living in cities such as Detroit. First, the percentage of the older adult population that is African American is expected to increase in the coming decades (He, Sengupta, Velkoff, & DeBarros, 2005). By 2050 a projected 12% of older adults will be African American compared to 9% today, while 58% will be non-Hispanic White, down from 80% today (Federal Interagency Forum on Aging-Related Statistics, 2012). Second, evidence suggests that older African Americans, particularly those at a low socioeconomic status, are in greater need of assistance than White older adults. Despite the general decline in disability among older adults, racial and ethnic disparities remain largely unchanged (Schoeni, Martin, Andreski, & Freedman, 2005). For example, the proportion of African Americans with limitations in activities of daily living (ADLs) is higher than that of Whites (Dunlop, Song, Manheim, Daviglus, & Chang, 2007). These disparities in disability are purportedly the result of variations in income and education levels between these two racial groups (Fuller-Thompson, Nuru-Jeter, Minkler, & Guralnik, 2009; Schoeni et al., 2005). Third, African Americans have typically had lower rates of nursing home placement than Whites (Akamigbo & Wolinsky, 2007; Wallace, Levy-Storms, Kington, & Andersen, 1998), but African Americans have accounted for an increasing percentage of the nursing home population in recent years (Feng, Fennell, Tyler, Clark, & Mor, 2010). It is possible that this also reflects differences in socioeconomic status, with older Whites better able to afford options such as assisted living, and older African Americans entering nursing homes because they are covered by Medicaid (Feng et al., 2010). In addition, African Americans and individuals with low socioeconomic status are more likely to live in nursing homes with fewer financial and staffing resources and greater regulatory deficiencies (Angelelli, Grabowski, & Mor, 2006; Grabowski, 2004). Fourth, older adults tend to underutilize HCBS, with 30% of disabled community-dwelling elders reporting unmet needs for care (Zarit, Shea, Berg, & Sundstrom, 1998). Unmet needs may be even higher for African American elders living in urban neighborhoods in cities such as Detroit. For example, using data from 2000 and 2002, the Detroit Area Agency on Aging identified increasing the availability and accessibility of health and supportive services as a high priority (Detroit Area Agency on Aging and Detroit Senior Citizens Department, 2004). Finally, African American households continue to be concentrated and segregated in inner city neighborhoods (Goldsmith & Blakely, 2010). For example, in 2000, more than 81% of the population of Detroit was African American, compared to a little over 14% for the state of Michigan (United States Census Bureau, 2000).

Cultural issues may also be associated with service use among African Americans, such as the belief that HCBS reflect the dominant culture and are therefore uninformed of and unresponsive to the needs of African Americans (Scharlach et al., 2006). Past negative experiences with members of other cultures may reduce African American elders’ willingness to access services provided by those who are culturally different (Yeatts, Crow, & Folts, 1992). Furthermore, African Americans’ interest in receiving community-based services is impacted by familism (Bradley, et al., 2002), defined as “reliance on family for support, obligation towards family members, and use of relatives as referents” (Magana, 1999, p. 466). Previous research has documented that African Americans believe that family members have the primary responsibility for providing care to aging relatives regardless of the particular needs or circumstances (Scharlach et al., 2006). These feelings of obligation, however, have been linked to higher levels of psychological distress among caregivers (Rozario & DeRienzis, 2008), suggesting that facilitating access to HCBS for African American elders could also benefit their family members.

Theoretical Framework

This study was guided by the Behavioral Model of Health Services and Access to Care (Andersen, 1995), the predominant theoretical framework in explorations of HCBS use among older adults (see Bookwala et al., 2004; Borrayo, Salmon, Polivka, & Dunlop, 2002; Henton et al., 2002; Mitchell & Krout, 1998; Mui & Burnette, 1994; White-Means & Rubin, 2004). According to this model, service use is a function of predisposing characteristics, enabling resources, and need for care (Andersen, 1995). Predisposing characteristics include demographic measures (e.g., age and gender), indicators of social structure (e.g., education), and personal beliefs about health and health services. Enabling resources provide individuals with the means to obtain and make use of services, and include income, health insurance, access to transportation, and social resources. The need for care can be both perceived (e.g., self-rated physical health) and evaluated (e.g., diagnosed health conditions).

Research Questions

This study addresses two limitations in the existing literature on HCBS use by older adults. First, while researchers have explored racial and ethnic differences in service use (Bradley et al., 2002; Henton et al., 2002; Houde, 1998; Mitchell & Krout, 1998; Mui & Burnette, 1994), samples typically include a small number of African American respondents. Race is typically entered in multivariate models as a predisposing characteristic, which allows researchers to determine whether being African American is associated with service use but does not uncover the impact of predisposing, enabling, and need factors for African Americans in particular. When Wallace and colleagues (1999) compared HCBS utilization of African American and White older women, however, they found significant differences between these groups in terms of predictors of service use. For example, the total number of services used by African American women was related primarily to their health conditions, while enabling resources such as being widowed or living with others were significantly associated with service use among White women (Wallace et al., 1999). Second, few studies have investigated whether the impact of predisposing, enabling, or need factors vary by the specific type of HCBS, whether for the general population or specifically for African American older adults. Researchers who have constructed conceptually meaningful categories of HCBS report that predictors can differ for services depending on the specific needs they address (e.g., ADL limitations vs. social interaction) and the location of their delivery (e.g., in-home vs. out-of-home). Mitchell and Krout (1998), for example, created categories of HCBS based on the extent to which they are discretionary. They found that need variables largely account for whether individuals use the least discretionary services (e.g., medical services), while predisposing and enabling variables play a role in determining whether older adults use more discretionary services (e.g., home-delivered meals). Similarly, Johnson and Wolinsky (1996) determined that in-home service use is due primarily to health needs, while the use of out-of-home services is related mostly to factors that could be classified as predisposing or enabling.

The purpose of this study was therefore to address the following research questions:

What predisposing, enabling, and need factors are associated with the use of home and community-based services among a sample of urban African American older adults?

How do these predisposing, enabling, and need factors vary between different categories of home and community-based services for this sample?

Method

Data and Sample

We explored the above research questions through secondary analysis of data from the Detroit City-Wide Needs Assessment of Older Adults collected by the Center for Urban Studies for the Institute of Gerontology and the Center for Healthcare Effectiveness of Wayne State University (Chapleski, Massanari, & Herskovitz, 2002). The needs assessment sample was a representative sample of noninstitutionalized persons aged 60 years or older, and was designed to reflect those eligible for Older Americans Act programs to help the city plan more effectively for their service needs. Details about the data collection procedures for the Detroit needs assessment are reported elsewhere (Chapleski et al., 2002). Briefly, data were collected during 2001 via telephone interviews with a stratified random digit dialing sample of 1,310 older adults and via in-person interviews with 100 older adults living in census tracts with low telephone coverage. The stratified sample targeted city-designated neighborhood area clusters, and we used poststratified sampling weights in the present analyses so that that all areas of the city of Detroit were represented in the research analyses in proportion to the total population of eligible respondents.

Because we were interested in uncovering the factors associated with service use among urban African American elders, we restricted analyses to the 1,126 respondent who were African American. We also deleted 22 cases that were missing data on service use, our outcome variable of interest. The final sample, reflecting poststratified sampling weights, was n = 1,099.

Measures

Table 1 presents the characteristics of the sample for all predictor and outcome variables prior to missing data imputation, and the number of missing cases for each variable.

Table 1.

Characteristics of the Sample (N = 1,099 Community-Dwelling Older Adults Age 60 and Older Living in Detroit in 2001).

Characteristics	Distribution
Predisposing
Female	71.9
Age	71.6 (.24)
	Range 60-97
Education
Less than high school (ref)	42.3
High school graduate or General education development	23.0
College and over	34.7
Enabling
Low income (Below 125% poverty line)	16.3
Housing problems (e.g., Insects or rodents, inadequate heating, excessive noise)	1.6 (0.1)
	Range 0-5
Neighborhood problems (e.g., Heavy crime, abandoned buildings, trash and litter)	1.9 (0.1)
	Range 0-5
Medicaid insurance	32.2
Driving primary transportation	60.3
Lives alone	41.2
Anticipated assistance
Short-term help	94.1
Long-term help	82.1
Emergency help	96.7
Need
Chronic health conditions	0.7 (0.03)
	Range 0-5
Self-rated health (Poor to excellent)	3.1 (0.38)
	Range 1-5
Health limits physical activity (Not at all to a lot)	1.7 (0.03)
	Range 1-3
Service use
Any service	62.7
Categories of service
In-home care (Home health care)	15.0
Functional care (Chore/homemaker, home-delivered, meals)	14.9
Household-Related care (Home repair, utility bill assistance)	18.9
Out-of-home care (Congregate meals, senior center, transportation, health screening, education, employment)	46.6
Financial/legal services	10.8

Note. Table entries reflect poststratified sampling weights. Percentages are shown for categorical variables. Means with standard errors in parentheses and range below are shown for continuous variables.

Independent Variables

Based on the Andersen model, we selected predictor variables reflecting predisposing characteristics, enabling resources, or indicators of need.

Predisposing characteristics

The three predisposing demographic characteristics were gender (male = 0, female = 1), age in years, and educational attainment (high school graduate, some college or higher, and less than a high school diploma as the reference group).

Enabling

Enabling resources that facilitate or impede the use of services included indicators of financial resources, measures of health insurance, mobility, and social resources. Financial resources at the individual level included a dichotomous measure indicating whether the respondent lived in a low-income household. We constructed this variable by dividing annual household income (reported as one of 12 categories ranging from less than US$5,000 to more than US$50,000) by number of individuals in the household and then determining whether this number was less than 125% of the poverty rate for the year 2000 (Dalaker, 2001). The needs assessment included 10 questions about the existence of common housing problems (e.g., inadequate heat in winter, nonworking, or leaking toilet) and 9 questions about the existence of common neighborhood problems (e.g., heavy traffic, abandoned buildings), which could both be indicators of limited financial resources at the individual and neighborhood level. We measured these as two count variables and top-coded each to range from 0 to 5 because they had a nonnormal distribution. The specific items included in each of these measures were selected for inclusion in the needs assessment because of their proposed relationship to health, life satisfaction, and the ability to age in place (Chapleski et al., 2002). Medicaid coverage was measured as a dummy variables (0 = no, 1 = yes). We included a measure of whether or not the respondent reported driving as their primary mode of transportation (0 = no, 1 = yes). Social resources included a dichotomous variable indicating whether the respondent lived alone (0 = no, 1 = yes). We measured anticipated instrumental assistance from family and friends using three dichotomous variables indicating someone is available for short term, long term, and emergency help (0 = no, 1 = yes).

Need

We selected three aspects of the respondent’s physical health and well-being to indicate potential need for HCBS: total number of health conditions, self-rated health, and limitations in activities. The total number of health conditions was measured by whether the respondent had any of five common serious chronic conditions affecting the elderly (Federal Interagency Forum on Aging-Related Statistics, 2012): chronic bronchitis or emphysema, heart problems, stroke, diabetes, and cancer. We transformed the individual measures into a count variable ranging from 0 to 5. Self-rated health came from a single-item Likert-type scale measure in the needs assessment (1 = poor, 2 = fair, 3 = good, 4 = very good, 5 = excellent). The needs assessment also asked respondents to rate the degree to which their health limited their ability to engage in moderate activities, such as vacuuming, moving a table, bowling, or playing golf (1 = not limited at all, 2 = limited a little, 3 = limited a lot).

Service Use

The needs assessment included 13 questions (0 = no, 1 = yes) regarding whether respondents had used any of the following services within the past 12 months: senior center, chore/homemaker services, congregate meals, home-delivered meals, home repair, utility bill assistance, home health care, free transportation, health screening, educational services, employment services, legal services, and financial services. The interview included examples of specific services to clarify the different types for respondents (e.g., home health care examples included visiting nurses or health aides, chore/homemaker service examples included light housekeeping or cleaning). Due to the low rates of service use among this sample, which limited our ability to examine the factors associated with use separately for each type of service, we constructed six outcome measures. The first outcome was a dichotomous measure indicating whether the respondent had used any of the 13 services (0 = none, 1 = at least one). The remaining five outcome measures were constructed to reflect a modified version of HCBS categories devised by Bookwala et al. (2004): in-home care (home health care), functional care (chore/homemaker services, home-delivered meals), household-related services (home repair, utility bill assistance), out-of-home services (congregate meals, senior center, free transportation, health screening, education services, and employment services), and a separate category for financial and legal services. We coded each of these five outcomes as a dichotomous measure of whether the survey participant had used any of the services included in each category (0 = none, 1 = at least one). While we initially conducted an exploratory factor analysis to create service categories, we found only modest correlations between service types and decided that a data-driven approach was not appropriate for the present analyses. We therefore chose to use conceptually developed categories informed by the prior work of Bookwala and colleagues (2004).

Data Analysis Procedures

To obtain accurate statistics and standard errors, all analyses used SVY commands in Stata Statistical Software Release 12 to take into account the needs assessment sampling design. After examining the patterns of missing data and concluding that the data were missing at random, we addressed missing data among predictor variables using the multiple imputations by chained equations (MICE) procedure. This procedure constructed five complete data sets and then adjusted for the effects of missing data on statistical inference. We used logistic regression models to estimate the effects of predisposing, enabling, and need variables for the six dichotomous outcome variables (i.e., any service use, in-home care services, functional care services, household services, out-of-home care services, financial/legal). We assessed goodness of fit using the Hosmer–Lemeshow test, which indicated that the overall model fit for each of the five models was acceptable. Tolerance and variance inflation factor (VIF) results indicated multicollinearity was not a concern with predictor variables. We used an α of .05 for statistical tests.

Results

Predictors of Any Service Use

As shown in Table 2, among the predisposing characteristics, age was significantly associated with any service use. Among enabling resources, those reporting a higher number of neighborhood problems, had Medicaid coverage, or lived alone had an increased odds of using any services, while those who reported driving as their primary mode of transportation had a decreased odds. None of the variables reflecting health needs were significantly associated with any service use.

Table 2.

Logistic Regression of Predisposing, Enabling and Need Factors and Service Use. (N = 1,099 Community-Dwelling Older Adults Age 60 and Older Living in Detroit in 2001).

	Any service use		In-home		Functional		Household		Out-of-home		Financial/legal
	OR	SE	OR	SE	OR	SE	OR	SE	OR	SE	OR	SE
Predisposing
Female	1.30	0.19	1.02	0.22	0.98	0.22	1.30	0.24	1.26	0.20	1.27	0.30
Age	1.02*	0.01	1.02	0.01	1.04**	0.01	1.00	0.01	1.01	0.01	0.98	0.01
Education (No High school ref)
High school graduate	1.02	0.18	1.45	0.33	1.15	0.28	1.32	0.27	0.95	0.18	1.07	0.28
College and higher	1.05	0.17	0.97	0.24	1.01	0.25	1.13	0.23	1.35	0.24	1.07	0.29
Enabling
Low income	1.16	0.24	1.08	0.25	1.07	0.25	1.16	0.27	1.23	0.26	0.99	0.31
Housing problems	1.03	0.04	0.99	0.05	1.09	0.06	0.99	0.05	1.01	0.03	1.03	0.06
Neighborhood problems	1.08*	0.04	1.05	0.06	0.97	0.05	1.16**	0.06	1.08*	0.04	1.21***	0.07
Medicaid	1.60**	0.24	1.77**	0.35	1.47	0.29	1.58**	0.28	1.33	0.22	2.76***	0.60
Driving primary transportation	0.63**	0.10	0.70	0.15	0.37***	0.08	1.16	0.22	0.64**	0.11	1.08	0.26
Lives alone	1.35*	0.20	1.02	0.21	1.70**	0.33	0.83	0.15	1.52**	0.23	1.03	0.24
Short-term help	0.61	0.21	0.37*	0.15	1.52	0.67	0.79	0.31	0.97	0.36	0.97	0.51
Long-term help	0.83	0.20	1.21	0.37	0.64	0.19	0.81	0.19	0.78	0.18	0.84	0.25
Emergency help	1.70	0.70	0.72	0.37	1.13	0.63	1.59	0.85	1.51	0.62	3.85	0.39
Need
Chronic health conditions	1.11	0.09	1.73***	0.18	1.35**	0.14	0.83	0.09	0.99	0.09	0.99	0.12
Self-rated health	0.96	0.07	1.05	0.10	0.99	0.09	1.07	0.09	1.01	0.08	0.86	0.09
Health limits physical activity	1.11	0.10	1.34*	0.16	1.34*	0.16	1.12	0.12	0.80*	0.09	1.16	0.16
Goodness-of-fit Hosmer–Lemeshow	1.14		0.57		0.64		0.43		0.56		1.17

Note. *p ≤ .05. **p ≤ .01. ***p ≤ .001. Table entries reflect poststratified sampling weights.

Predictors of Categories of Service Use

Table 2 also includes the logistic regression models for the five categories of HCBS.

In-Home Care

None of the predisposing variables were statistically significant with in-home care use. Among enabling variables, Medicaid coverage had a positive association with in-home care. Those who reported that they had someone to turn to for short-term assistance had reduced odds of in-home care use. Among all need variables, the number of chronic health conditions and health limitations on moderate activities were both significantly associated with in-home care use.

Functional Care Services

The only predisposing characteristic associated with the use of this category of service was age, with each additional year slightly increasing the odds of receiving functional care services. The odds were lower for those with the enabling resources of driving as their primary mode of transportation, while those who lived alone had higher odds of using functional care services. In terms of need variables, number of chronic health conditions and limitations in moderate activities were significant predictors of functional care service use.

Household Related Services

No predisposing variables were associated with use of household-related services. In terms of variables reflecting enabling predictors, number of neighborhood problems and Medicaid coverage were associated with increased odds of household services use. None of the need variables was associated with the use of household services.

Out-of-Home Services

Predisposing variables did not significantly affect the use of out-of-home services. For enabling resources, respondents who had more neighborhood problems or lived alone had greater odds of using this category of services. Those who reported that they drive had reduced odds of using out-of-home services. In the need category, those who reported that their health limited their activities had reduced odds of using out-of-home services.

Financial/Legal Services

Only enabling characteristics were significantly associated with financial and legal services, with neighborhood problems and Medicaid increasing the use of these services.

Discussion

Using a representative sample of elders living in Detroit, the aims of our study were: (a) to examine the predisposing, enabling, and need characteristics associated with the use of HCBS among African American older adults and (b) to explore whether these characteristics differed by the type of HCBS used. While our sample did not allow us to compare differences between Whites and African Americans, it did provide the opportunity for a closer examination of facilitators of HCBS use for older African Americans, who historically have not been the focus of service use research. Furthermore, according to Bradley and colleagues (2002), much of the research on how race influences service use has focused on traditional long-term care services that address health and personal care assistance, rather than the wide range of HCBS examined in the current study, which we selected to reflect the variety of services provided through the Older Americans Act.

In terms of the first aim, with a few exceptions (discussed below) for this sample of African American elders, the use of different types of HCBS was related primarily to the lack of enabling resources. Previous research has reported that need factors (e.g., health conditions, functional status) typically play the largest role in determining whether older adults use services (Borrayo et al., 2002; Mui & Burnette, 1994; Wolinsky & Johnson, 1991). When compared with findings from other studies, the majority of which included predominantly White samples (e.g., Borrayo et al., 2002; Houde, 1998), our results potentially signal that the association between predisposing, enabling and need factors, and HCBS utilization differs for older adults who are African American. Of particular importance were potential indicators of limited financial resources, not driving a car, and the absence of social resources and support.

First, while there was no significant association between low household income and service use, three potential proxies for limited financial resources appeared to serve as facilitators in several of the regression models. Those who reported more neighborhood problems, such as litter, abandoned buildings, heavy traffic, and crime, were more likely to use any services, as well as household services and financial/legal services. While perceived environment is not typically included in studies of service use, recent conceptualization of the behavioral model has called attention to the role community characteristics can play in terms of predisposing, enabling, and need for services (Andersen, 2008). Prior research has found that neighborhood problems are associated with negative individual outcomes, including depression, physical functioning, and unhealthy behaviors (Echeverria, Diez-Roux, Shea, Borrell, & Jackson, 2008; Yen, Yelin, Katz, Eisner, & Blanc, 2006). Those who live in neighborhoods with more problems may therefore have an increased need for HCBS that was not captured by the need variables we included in our models (e.g., self-rated health). That is, while we categorized neighborhood problems as a potentially enabling resource, the observed relationship may be because more neighborhood problems leads to poor physical and mental health, which in turn leads to more service use.

Another explanation, as suggested by the positive association between Medicaid and service use, is that more neighborhood problems signifies increased access to HCBS. Older adults who are not low income get little assistance with paying for HCBS, because private insurance and Medicare primarily cover acute care services (Komisar & Thompson, 2007). Those who meet the income and asset requirements for Medicaid, however, may receive HCBS offered through waiver programs, although access can be limited by enrollment caps and restricted availability (Komisar & Thompson, 2007). Furthermore, many of the HCBS in our study may be funded by the Older Americans Act, including senior centers, transportation, and home-delivered meals. While Older Americans Act services are available to all individuals age 60 or older, states are required to target services to persons with the greatest social or economic needs.

Second, those who reported driving as their primary mode of transportation were less likely to use any services, functional care, or out-of-home services, meaning that those who depended on alternative forms of transportation (e.g., friends, public transportation, walking) were more likely to use these services. While it is not possible to determine whether those who reported that they do not drive had to give up driving versus never drove a car in the first place, or the specific reasons why they do not drive a car, this finding may also indicate an increased need for HCBS that we did not capture in our need variables. Older adults consistently report that they plan to continue driving for as long as possible (Kostyniuk & Shope, 2003), and typically reduce or stop driving only when health problems (e.g., cognitive limitations, declining vision) impair their ability to safely operate a vehicle (Lynott et al., 2009). Older adults in our sample who were still driving may be less likely to use services because they were less likely to need services. Results could also reflect decreased access to the community, since older nondrivers make 15% fewer trips for medical appointments and 65% fewer trips for social, community, or religious activities than their driving counterparts (United States Government Accountability Office, 2004).

Third, the potential absence of social resources, as indicated by living alone, increased the likelihood of service use (including any services, functional care, and out-of-home services), while perceived instrumental support, as indicated by the belief that someone would help for a short period of time, decreased the odds of use of in-home services. These results suggest support for the hierarchical compensatory model (Cantor & Little, 1985), which proposes that older adults do not use formal services unless they either do not have informal sources of support, or their particular assistance needs exceed the ability of informal services. That is, older adults turn to their social resources before seeking professional help. Existing research on the relationship of social resources to support with formal care is somewhat contradictory. For example, some studies report that formal care substitutes for informal care, (Van Houtven & Norton, 2004; Johnson & Wolinsky, 1996) while others report no evidence of an inverse relationship between formal and informal support (Li, 2005; Penning, 2002). Given that African Americans typically rely on family members for care (Connell & Gibson, 1997), it is possible that social isolation was a motivating factor for service use in this sample, and future research should explore the relationship between informal support and formal support among older African Americans.

In terms of the second aim, many previous studies have not differentiated services by their location (e.g., in-home or out-of-home) and the needs they aim to address (e.g., health, functional care, household), but our findings suggest that the factors associated with HCBS use differ by service type among African American elders in Detroit. In many ways, these findings were not unexpected. For example, health need, indicated by the number of chronic illnesses and limitations in moderate activities, was positively associated with the use of in-home services and functional care, but not with other categories of services (with the exception of limitations in moderate activities’ negative association with out-of-home services). These findings and those of Johnson and Wolinsky (1996) indicate that the use of more traditional forms of long-term care may not differ between African Americans and other racial/ethnic groups, although it was not possible for us to test this with our sample. Household-related services (e.g., utility bill assistance) as well as financial and legal services address economic needs and therefore use was associated with potential proxies of financial limitations (e.g., Medicaid, neighborhood problems). Out-of-home services, including senior centers, transportation, and employment services, address both social and economic needs, and indeed measures of both (e.g., living alone and neighborhood problems) increased the likelihood of using services in this category.

Findings of the current study should be interpreted in light of its limitations. First, because fewer than 200 respondents in this representative sample of Detroit elders were White, it was not possible to conduct subanalyses to examine differences in the facilitators and barriers to HCBS use between African Americans and Whites. Because previous research suggests that different factors influence service use in these two racial groups (Bradley et al., 2002; Dunlop et al., 2002; Musa et al., 2009), we decided not to combine African American and White respondents in the same model. Future research should address this limitation by including adequate numbers of White and African American respondents. Second, prior studies using the behavioral model typically report the percent of variance explained by predisposing, enabling, and need factors. It is not possible in Stata 12, however, to compute the pseudo-R2 for logistic models when using SVY commands to reflect the sampling design. Third, the survey data did not include questions about certain factors that may be related to service use (e.g., limitations in activities of daily living), and used self-report, cross-sectional data. Fourth, the needs assessment did not include proxy respondents, and therefore may have excluded older adults who are frail, disabled, or chronically ill. Finally, findings should be compared with research on urban African American populations living in other cities than Detroit.

Despite these limitations, this study calls attention to the facilitators and barriers to HCBS use among a sample of urban African American elders. Understanding the factors contributing to the use of HCBS by this population is particularly important given the increased focus on keeping elders in the community and out of institutions, which is in response to a confluence of factors, including: (a) the Supreme Court’s Olmstead decision in 1999, which called for an end to unnecessary institutionalization of individuals with disabilities (Olmstead v. L.C., 1999); (b) the need for local, state, and federal governments to curtail rising long-term care costs, particularly in nursing homes, which cost an average of over US$87,000 annually per resident in 2011 (MetLife Mature Markey Institute, 2011); and (c) the large numbers of older adults who express a desire to age in place (Feldman et al., 2004). Furthermore, in cities such as Detroit, high rates of morbidity and mortality among older adults suggest a greater need for health and supportive services, while the closure of 16 nursing homes in the past 13 years indicates fewer options for institutional care (Detroit Area Agency on Aging, 2012). It is therefore critical that these elders receive a wide range of HCBS to address their health, economic, and social needs. Our findings suggest that targets for intervention differ depending on the particular economic, social, and health needs of older African Americans, and that these differences are not apparent when looking at all the types of services together.

Footnotes

Acknowledgements

We gratefully acknowledge valuable assistance from Dave Childers at the University of Michigan Center for Statistical Consultation and Research. We also thank Kristen Gustavson, Jennifer Price Wolf, and Letha Chadiha for valuable comments on previous drafts.

Author’s Notes

Preparation of this paper was supported by NIA grant T32-AG00017.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Institute on Aging (grant number T32-AG000117).

References

Administration on Aging. (2012). Older Americans Act and aging network. Retrieved from http://www.aoa.gov/AoARoot/AoA_Programs/OAA/introduction.aspx

Akamigbo

A. B.

Wolinsky

F. D.

(2007). New evidence of racial differences in access and their effects on the use of nursing homes among older adults. Medical Care, 45, 672-679. doi:10.1097/MLR.0b013e3180455677

Andersen

R. M.

(1995). Revisiting the behavioral model and access to medical care: Does it matter? Journal of Health and Social Behavior, 36, 1-10.

Andersen

R. M.

(2008). National health surveys and the behavioral model of health services use. Medical Care, 46, 647-653. doi:10.1097/MLR.0b013e31817a835d

Angelelli

Grabowski

D. C.

Mor

(2006). Effect of educational level and minority status on nursing home choice after hospital discharge. American Journal of Public Health, 96, 1249-1253. doi:10.2105/AJPH.2005.062224

Bookwala

Zdaniuk

Burton

Lind

Jackson

Schulz

(2004). Concurrent and long-term predictors of older adults’ use of community-based long-term care services: The Caregiver health effects study. Journal of Aging and Health, 16, 88-115. doi:10.1177/0898264303260448

Borrayo

E. A.

Salmon

J. R.

Polivka

Dunlop

B. D.

(2002). Utilization across the continuum of long-term care services. Gerontologist, 42, 603-612.

Bradley

E. H.

McGraw

S. A.

Curry

Buckser

King

K. L.

Kasl

S. V.

Andersen

(2002). Expanding the Andersen model: The role of psychosocial factors in long-term care use. Health Services Research, 37.5, 1221-1242. doi: 10.1111/1475-6773.01053

Calsyn

R. J.

Winter

J. P.

(1999). Predicting specific service awareness dimensions. Research on Aging, 21, 762-780. doi:10.1177/0164027599216003

10.

Cantor

Little

(1985). Aging and social care. In Binstock

R. H.

Shanas

(Eds.), Handbook of aging and the social sciences (2nd ed., pp. 745-781). New York, NY: Van Nostrand Reinhold.

11.

Chapleski

E. E.

Massanari

R. M.

Herskovitz

2002. Facing the future: 2002 City of Detroit needs assessment of older adults. Detroit, MI: Wayne State University.

12.

Connell

C. M.

Gibson

G. D.

(1997). Racial, ethnic, and cultural differences in dementia caregiving: Review and analysis. Gerontologist, 37, 355-364. doi:10.1093/geront/37.3.355

13.

Dalaker

(2001). Poverty in the United States: 2000. U.S. Census Bureau, Current Population Reports (Series P60-214). Washington, DC: U.S. Government Printing Office. Retrieved from http://www.census.gov/prod/2001pubs/p60-214.pdf

14.

Detroit Area Agency on Aging and Detroit Senior Citizens Department. (2004). Dying before their time: The startling truth about mortality and Detroit area seniors. Retrieved from http://www.daaa1a.org/DAAA/media/DBTT%20Synopsis.pdf

15.

Detroit Area Agency on Aging. (2012). Dying before their time II: The startling truth about mortality and Detroit area seniors. Retrieved from http://www.daaa1a.org/DAAA/media/Dying%20Before%20Their%20Time%20II%20-%202012%20Final%20Report.pdf

16.

Dunlop

D. D.

Manheim

L. M.

Song

Chang

R. W.

(2002). Gender and ethnic/racial disparities in health care utilization among older adults. Journals of Gerontology, Social Sciences, 57B, S221-S233. doi:10.1093/geronb/57.4.S221

17.

Dunlop

D. D.

Song

Manheim

L. M.

Daviglus

M. L.

Chang

R. W.

(2007). Racial/ethnic differences in the development of disability among older adults. American Journal of Public Health, 97, 2209-2215. doi:10.2105/AJPH.2006.106047

18.

Echeverria

Diez-Roux

A. V.

Shea

Borrell

L. N.

Jackson

(2008). Associations of neighborhood problems and neighborhood social cohesion with mental health and health behaviors: The multi-ethnic study of atherosclerosis. Health & Place, 14, 853-865. doi:10.1016/j.healthplace.2008.01.004

19.

Federal Interagency Forum on Aging-Related Statistics. (2012). Older Americans 2012: Key indicators of wellbeing. Retrieved from http://www.agingstats.gov/agingstatsdotnet/Main_Site/Data/2010_Documents/Docs/OA_2010.pdf

20.

Feldman

P. H.

Oberlink

M. R.

Simantov

Gursen

M. D.

(2004). A tale of two older Americas: Community opportunities and challenges. AdvantAge Initiative 2003 National survey of adults aged 65 and older. New York, NY: Center for Home Care Policy and Research, Visiting Nurse Service of New York.

21.

Feng

Fennell

M. L.

Tyler

Clark

M. A.

Mor

(2010). Shifts in the racial/ethnic composition of nursing home residents in the United States, 2000-2007. Paper presented at the 2010 International Conference on Aging in the Americas, Austin, TX. Retrieved from http://www.utexas.edu/lbj/caa/2010/presentations/FengZ1.pdf

22.

Fuller-Thompson

Nuru-Jeter

Minkler

Guralnik

J. M.

(2009). Black-White differences in disability among older Americans: Further untangling the role of race and socioeconomic status. Journal of Aging and Health, 21, 677-698. doi:10.1177/0898264309338296

23.

Gaugler

J. E.

Kane

R. L.

Kane

R. A.

Newcomer

(2005). Early community-based service utilization and its effects on institutionalization in dementia caregiving. Gerontologist, 45, 177-186. doi:10.1093/geront/45.2.177

24.

Goldsmith

Blakely

E. J.

(2010). Separate societies: Poverty and inequality in U.S. cities. Philadelphia, PA: Temple University Press.

25.

Grabowski

D. C.

(2004). The admission of Blacks to high-deficiency nursing homes. Medical Care, 42, 456-464. doi:10.1097/01.mlr.0000124307.17380.df

26.

Sengupta

Velkoff

V. A.

DeBarros

K. A.

(2005). 65+ in the United States: 2005 (U.S. Census Bureau Current Population Reports). Washington, DC: U.S. Government Printing Office.

27.

Henton

F. E.

Hays

B. J.

Walker

S. N.

Atwood

J. R.

(2002). Determinants of Medicare home healthcare service use among Medicare recipients. Nursing Research, 51, 355-362.

28.

Houde

S. C.

(1998). Predictors of elders’ and family caregivers’ use of formal home services. Research in Nursing & Health, 21, 533-543. doi:10.1002/(SICI)1098-240X(199812)21:6<533::AID-NUR7>3.0.CO;2-I

29.

Jette

A. M.

Tennstedt

Crawford

(1995). How does formal and informal community care affect nursing home use? Journal of Gerontology: Social Sciences, 50B, S4-S12.

30.

Johnson

R. J.

Wolinsky

F. D.

(1996). Use of community-based long-term care services by older adults. Research on Aging, 8, 512-537. doi:10.1177/089826439600800403

31.

Komisar

H. L.

Thompson

L. S.

(2007). National spending for long-term care. Washington, DC: Georgetown University. Retrieved from: http://ltc.georgetown.edu/pdfs/whopays2006.pdf.

32.

Kostyniuk

L. P.

Shope

J. T.

(2003) Driving and alternatives: Older drivers in Michigan. Journal of Safety Research, 34, 407-414. doi:10.1016/j.jsr.2003.09.001

33.

L. W.

(2005). Longitudinal changes in the amount of informal care among publicly paid home care recipients. Gerontologist, 45, 465-463. doi:10.1093/geront/45.4.465

34.

Lynott

Haase

Nelson

Taylor

Twaddell

Ulmer

Stollof

E. R.

(2009). Planning complete streets for an aging America. Washington, DC: AARP Public Policy Institute.

35.

MetLife Mature Market Institute (2011). Market survey of long term care costs. Retrieved from https://www.metlife.com/assets/cao/mmi/publications/studies/2011/mmi-market-survey-nursing-home-assisted-living-adult-day-services-costs.pdf

36.

Mitchell

Krout

J. A.

(1998). Discretion and service use among older adults: The behavioral model revisited. Gerontologist, 38, 159-168. doi:10.1093/geront/38.2.159

37.

Mui

A. C.

Burnette

(1994). Long-term care service use by frail elders: Is ethnicity a factor? Gerontologist, 34, 190-198. doi:10.1093/geront/34.2.190

38.

Musa

Schulz

Harris

Silverman

Thomas

S. B.

(2009). Trust in the health care system and the use of preventive health services by older Black and White adults. American Journal of Public Health, 99, 1293-1299. doi:10.2105/AJPH.2007.123927

39.

Older Americans Act of 1965, 3 U.S.C. § 305 (1965).

40.

Olmstead v. L. C., 527 U.S. 581 (1999).

41.

Richardson

(1992). Service use among urban African American elderly people. Social Work, 37, 47-54.

42.

Rozario

P. A.

DiRienzis

(2008). Familism beliefs and psychological distress among African American women caregivers. Gerontologist, 48, 772-780. doi:10.1093/geront/48.6.772

43.

Scharlach

A. E.

Kellam

Ong

Baskin

Goldstein

Fox

P. J.

(2006). Cultural attitudes and caregiver service use. Journal of Gerontological Social Work, 47, 133-156. doi:10.1300/J083v47n01_09

44.

Schoeni

R. F.

Martin

L. G.

Andreski

P. M.

Freedman

V. A.

(2005). Persistent and growing socioeconomic disparities in disability among the elderly: 1982-2002. American Journal of Public Health, 95, 2065-2070. doi:10.2105/AJPH.2004.048744

45.

Shirk

(2006). Rebalancing long-term care: The role of the Medicaid HCBS waiver program. National Health Policy Forum Background Paper. Retrieved from http://www.nhpf.org/library/background-papers/BP_HCBS.Waivers_03-03-06.pdf

46.

Spillman

B. C.

Pezzin

L. E.

(2000). Potential and active family caregivers: Changing networks and the “sandwich generation.” Milbank Quarterly, 78, 347-374.

47.

United States Census Bureau. (2000). Profile of general demographic characteristics: 2000. Retrieved from http://factfinder2.census.gov/faces/tableservices/jsf/pages/productview.xhtml?src=bkmk

48.

United States Government Accountability Office. (2004). Transportation-disadvantaged seniors: Efforts to enhance senior mobility could benefit from additional guidance and information. Washington, DC: Author.

49.

Van Houtven

C. H.

Norton

E. C.

(2004). Informal care and health care use of older adults. Journal of Health Economics, 23, 1159-1180. doi:10.1016/j.jhealeco.2004.04.008

50.

Wallace

D. C.

Fields

B. L.

Witucki

Boland

Tuck

(1999). Use of home and community-based services by elderly Black and White females. Journal of Women and Aging, 11, 5-20.

51.

Wallace

S. P.

Levy-Storms

Kington

R. S.

Andersen

R. M.

(1998). The persistence of race and ethnicity in the use of long-term care. Journals of Gerontology, Social Sciences, 53B, S104-S112. doi:10.1093/geronb/53B.2.S104

52.

White-Means

S. I.

Rubin

R. M.

(2004). Is there equity in the home health care market? Understanding racial patterns in the use of formal home health care. Journal of Gerontology: Social Sciences, 59B, S220-S229. doi:10.1093/geronb/59.4.S220

53.

Yeatts

D. E.

Crow

Folts

(1992). Service use among low-income minority elderly: Strategies for overcoming barriers. Gerontologist, 32, 24-32. doi:10.1093/geront/32.1.24

54.

Yen

Yelin

Katz

Eisner

Blanc

(2006). Perceived neighborhood problems and quality of life, physical functioning, and depressive symptoms among adults with asthma. American Journal of Public Health, 96, 873-879.

55.

Zarit

S. H.

Shea

D. G.

Berg

Sundstrom

(1998). Patterns of formal and informal long term care in the United States and Sweden ( AARP Andrus Foundation Final Report). Pennsylvania: Pennsylvania State University.