Professional service use for a serious personal problem: Comparing older African Americans, Black Caribbeans, and non-Hispanic Whites using the National Survey of American Life

Abstract

Objectives

Examines combinations of professionals visited for a serious personal problem.

Methods

The sample includes those aged 55 and older (n=862) from the National Survey of American Life (NSAL). Latent class analysis was used to identify groups of respondents based on types of professionals visited. Multinomial logistic regression was used to identify factors associated with group membership.

Results

Classes included health provider plus clergy, physician plus mental health provider, and limited provider use. Whites were more likely than African Americans to fall into the health provider plus clergy and physician plus mental health provider classes. Those with physical and emotional problems were more likely to be in the health provider plus clergy and physician plus mental health provider classes, respectively.

Discussion

Most respondents were in the limited provider use class suggesting that for many of problems, minimal professional help is utilized. Physicians and clergy were important across all three classes.

Introduction

Studies consistently find that older adults and racial/ethnic minorities are less likely to receive treatment for mental disorders than younger age groups or Whites. When they do seek help, older adults are more likely to visit a primary care doctor than a mental health specialist (Administration on Aging, 2001; Perron et al., 2009; Wang et al., 2005; Wetherell & Unützer, 2003). However, people experience a wide range of emotional/mental health problems that do not necessarily meet the level of psychiatric diagnosis (e.g., depression, anxiety). Less attention in the literature focuses on help-seeking for serious personal problems (e.g., interpersonal problems, the death of a loved one, or economic problems), whether it is for the problem itself or the psychological distress that may emerge from the problem.

Examining help-seeking for serious personal problems contributes to our overall understanding of the larger picture of mental health care, as stressful life experiences can cause significant psychological distress and lead to more serious problems if not addressed. Help-seeking for these types of life problems may focus on the presenting issue itself (e.g., help with funereal arrangements after the death of a loved one) and/or the negative sequelae of the experience (e.g., grief, depressive symptoms, anxiety). Accordingly, it is important to understand where people seek help for these problems as it may have implications for the type of care they receive. Finally, previous help-seeking research has focused largely on the use of individual professionals (e.g., family doctor, psychiatrist) or health sectors (e.g., general medical, mental health specialty) without consideration of the ways in which people may utilize multiple providers for the same problem.

This study addresses gaps in the current literature on help-seeking by examining the use of professional services for a serious personal problem within a racially and ethnically diverse sample of older adults. In particular, we examine three research questions: 1) What types of professionals do older adults visit for a stressful personal problem? 2) Are there homogeneous groups of older adults based on the types of professionals they visit? and 3) What social and demographic factors are associated with those groupings, with a focus on racial/ethnic differences across groups?

Background

Older adults identify a wide range of stressful life events (in addition to their own physical illness) including the death or illness of a loved one as well as a variety of nonmedical events (e.g., financial problems, interpersonal issues) (Hardy, Concato, & Gill, 2002). Among older adults who seek mental health services, stressful events such as bereavement, financial loss, or changes in their physical health are especially prominent (Phillips & Murrell, 1994). While younger age groups tend to report more stressful events compared to older adults (Hatch & Dohrenwend, 2007), evidence suggests that older adults are more vulnerable to the negative effects of stressful life events, even at lower levels of exposure (Cairney & Krause, 2008). Specifically, stressful life events among older adults can contribute to changes in problem drinking behaviors (Perreira & Sloan, 2001); difficulties with ADLs and IADLs, and poor self-rated health (Hardy et al., 2002); and higher levels of depression and anxiety (Beekman, Penninx, Deeg, & et al, 2002; Hardy et al., 2002).

Older adults report seeking help for stressful personal problems from both professionals and family and friends (Woodward, Chatters, Taylor, Neighbors, & Jackson, 2010). Regardless of the source of stress or the nature of the problem, older adults are more likely than other age groups to seek help from a family doctor (primary care provider) than they are from a mental health specialist (Mackenzie, Pagura, & Sareen, 2010; Phillips & Murrell, 1994; Simning et al., 2010). This has raised some concerns about the receipt of appropriate treatment for older adults with diagnosable disorders. On the other hand, for less serious problems primary care providers may provide what Begum et al (2013) refer to as “alternative pathways to care” through “medical surveillance” (i.e., recognizing and facilitating help seeking for a mental health problem in the course of addressing other problems) and “opportunistic help seeking” (i.e., patients using routine doctor visits to raise other concerns).

Other research on patterns of help-seeking indicate that social service agencies are the most common sources of help for economic problems (Neighbors & Jackson, 1984); clergy are used for the death of a loved one (Neighbors & Jackson, 1984); and physicians and emergency rooms are contacted for a physical health problem (Neighbors & Jackson, 1984; Peat, McCarney, & Croft, 2001).

Bereavement, in particular, is one area where physicians and clergy are recognized as important sources of care (Chatters et al., 2011; Ghesquiere, Shear, & Duan, 2013; Nagraj & Barclay, 2011; Taylor, Woodward, Chatters, Mattis, & Jackson, 2011) even for persons with access to specialized grief services through hospice (Bergman & Haley, 2009). The bereavement care provided by primary care physicians varies greatly with some practices having very structured proactive plans when patients experience the death of a loved one and others being reactive and responding only when the patient approaches them (Nagraj & Barclay, 2011). However, outcomes research suggests that grief-related support from physicians may be less effective than help from religious leaders and grief-specific support groups (Ghesquiere et al., 2013). Religious leaders may be particularly helpful in managing grief and grief-related depression because of their personal relationship with the bereaved (Ghesquiere et al., 2013) and because religion and religious beliefs help the bereaved find a sense of meaning in the loss (Neimeyer, 2000).

Research related to help-seeking for financial problems is much more limited and does not focus specifically on older adults. Early efforts in this area suggest that assistance for financial problems is largely restricted to help-seeking from informal support networks (Gourash, 1978) or from the social service sector (Neighbors & Jackson, 1984). More recent research on problem gambling and associated mental health issues (e.g., Evans & Delfabbro, 2005) focuses on the most severe forms of financial problems. Overall, there is little research examining help-seeking efforts and the psychological distress associated with non-gambling related financial problems.

Clergy as a source of assistance in Black communities

Although Blacks in the United States are typically seen as a homogenous group and treated as such in much research (Whitfield, Allaire, Belue,& Edwards, 2008), the growing number of immigrants from the Caribbean alone dramatically increases the diversity of this population. Black Caribbean immigrants in the U.S. population number 1.5 million according to the 2000 U.S. census. This represents a significant percentage of growth in the U.S. Black population and a sizable component of the Black population in select urban centers (McKinnon, 2001). Attention to ethnicity within the Black population can help us better understand potential ethnic-specific differences in the help-seeking behaviors of older adults.

For example, substantial literature documents the important role of clergy and religious coping for both African Americans and Black Caribbeans. The majority of African Americans and Black Caribbeans report that prayer is an important means of coping with stress (Chatters, Taylor, Jackson, & Lincoln, 2008). Further, when faced with a stressful personal problem, both groups are more likely to seek help from clergy than from other professionals (Chatters et al., 2011; Taylor et al., 2011). At the same time, however, there are significant differences. Early research indicated that African Americans who contacted clergy first were less likely to seek help from other professionals (Neighbors, Musick, & Williams, 1998). In contrast, Black Caribbeans who sought help from clergy saw other professionals as well, particularly for bereavement, physical health problems, and emotional problems (Taylor et al., 2011). Identified roles for ministers in mental health care includes direct support for problems such as bereavement, substance abuse, marital problems and family conflict, financial distress, and unemployment (Mattis et al., 2007; Neighbors et al., 1998; Taylor, Ellison, Chatters, Levin, & Lincoln, 2000; Young, Griffith, & Williams, 2003), as well as counseling and referral services for psychological distress (Bentz, 1970; Blank, Mahmood, Fox, & Guterbock, 2002; Neighbors et al., 1998; Young et al., 2003). Based on existing research we would expect that Black Caribbeans would be more likely than African Americans to rely on a variety of service providers.

In sum, existing research suggests that primary care doctors and clergy are important sources of help for older adults facing a serious problem that involves significant psychological distress. This study extends this prior research by using latent class analysis to identify homogeneous groups of older adults based on the combinations of professional service providers visited. This will permit a better understanding of the variety of professionals from whom older adults seek help. Following this, we examine predictors of latent class membership with a particular focus on race/ethnicity and type of problem. The analysis uses a national probability sample of older African Americans, Black Caribbeans, and non-Hispanic Whites to examine differences among older Black Americans, as well as Black-White differences in latent class membership. This depth of investigation is important because although between-group (i.e., Black vs. White) differences can confirm the existence of racial/ethnic disparities in help-seeking, understanding within group differences (i.e., African American vs. Black Caribbean) can help shed light on the mechanisms contributing to those disparities. Based on previous research we hypothesize that physicians and clergy will be an important combination of professionals used for a subset of respondents. In addition, African Americans and Black Caribbeans will be more likely than Whites to rely on clergy.

Design and Methods

Sample

This sample used data from the National Survey of American Life: Coping with Stress in the 21^st Century (NSAL), a national multi-stage probability design survey (Jackson et al., 2004). The study used sampling frames and a four-step sampling process common to the University of Michigan Survey Research Center’s national sample design. This process includes a primary stage sampling of U.S. Metropolitan Statistical Areas and counties, a second stage sampling of area segments, a third stage sampling of housing units within the selected area segments, and finally a random selection of eligible respondents from the sample housing units.

The African American sample is the core sample of the NSAL and consists of 64 primary sampling units. Fifty-six of these primary areas overlap substantially with existing Survey Research Center National Sample primary areas. The remaining eight primary areas were chosen from the South in order for the sample to represent African Americans in the proportion in which they were distributed nationally. The African American sample is a nationally representative sample of households located in the 48 contiguous states with at least one Black adult 18 years or over who did not identify ancestral ties in the Caribbean.

The NSAL includes the first major probability sample of Black Caribbeans ever conducted. The Black Caribbean sample was selected from two area probability sample frames: the core NSAL sample as well as an area probability sample of housing units from geographic areas with a relatively high density of persons of Caribbean descent. Respondents were asked to self-identify their race. Those who self-identified as black and of West Indian or Caribbean descent, reported that they were from one of a list of Caribbean countries presented by the interviewers, or indicated that their parents or grandparents were born in a Caribbean country were coded as Black Caribbean. For a more detailed discussion of the NSAL sample and instrumentation see Heeringa et al. (2004) and Jackson et al. (2004).

The field work for the study was conducted from 2001 to 2003 by the Institute for Social Research Survey Research Center, University of Michigan in cooperation with the Program for Research on Black Americans. Interviews were face-to-face within respondents’ homes and respondents were compensated for their time. The overall response rate was 72.3% with response rates of 70.7% for African Americans, 77.7% for Black Caribbeans, and 69.7% for non-Hispanic Whites.

A total of 6,082 face-to-face interviews were conducted with persons aged 18 or older, including 3,570 African Americans, 891 non-Hispanic Whites, and 1,621 Blacks of Caribbean descent. Roughly three quarters (77.5%) of NSAL respondents indicated that they had experienced a serious personal problem. The analytic sample for this study is comprised of the 862 of these respondents who are aged 55 and older and reported experiencing a serious personal problem, 456 of whom were African American, 239 White, and 167 Black Caribbean.

Measures

The section of the NSAL questionnaire used for this study was designed to examine issues related to help-seeking for a serious personal problem. Respondents were asked to report the most serious personal problem they had experienced in their lives that had caused them a significant amount of distress, to the “point of nervous breakdown”. They were next asked to describe the nature of the problem. The researchers then coded these responses into five categories for type of problem consistent with those used in previous research (Neighbors & Jackson, 1984; Neighbors, Music, & Williams, 1998): physical (e.g., poor health, accident), emotional (e.g., depression, unhappiness, self-doubt), interpersonal (e.g., difficulties with close family and friends, divorce), economic (e.g., poor or declining financial status, loss of assets), and death of a loved one.

Respondents were presented with a list of professional service providers and asked to indicate all of the people on that list they had talked to about their problem. For this study, dichotomous indicators were created to represent use of 7 provider categories: 1) psychiatrist; 2) other mental health professional such as a psychologist, psychotherapist, social worker, mental health nurse or counselor; 3) a family doctor; 4) any other doctor, such as a cardiologist; 5) any other health professional such as a nurse, physician’s assistant, or chiropractor; 6) a religious or spiritual advisor such as a minister, priest, rabbi, or pastor; 7) or any other healer such as an herbalist, spiritualist, naturalist or faith healer.

Other measures included age, sex, marital status (currently married, previously married, never married), years of education, household income, and insurance coverage (none, private, public).

Analysis

Latent class analysis (LCA) was used to identify homogeneous groups of respondents based on the 7 dichotomous items reflecting the different types of professionals visited in their lifetime. Mplus v. 6 was used for LCA models including weights and complex survey design variables (Muthén & Muthén, 2001). In an effort to identify the smallest number of classes needed to account for patterns of service providers visited, the number of latent classes was determined iteratively, beginning with a one-class model and testing models of increasing number of classes using robust maximum likelihood estimation. The optimal number of classes was determined based on a variety of goodness-of-fit statistics (e.g., the Bayesian Information Criterion (BIC), Akaike Information Criterion (AIC), Lo-Mendell-Rubin’s adjusted likelihood ratio test (LRT), entropy measures) as well as the extent to which classes were distinct and substantively meaningful (Nylund, Asparouhov, & Muthén, 2007).

Respondents were assigned to the class for which they had the highest probability of membership. For each class, the prevalence of respondents and the probability of visiting a particular service provider are reported. Conditional probabilities of .70–1 were considered to be a high probability of visiting a specific type of provider, .40–.69 were considered a moderate probability, and less than .40 was considered a low probability (Collins & Lanza, 2010; Thorpe, Thorpe, Kennelty, & Pandhi, 2011).

Following LCA model estimation, multinomial logistic regression was used to analyze the factors predicting latent class membership with a particular focus on racial/ethnic differences and the type of problem reported. These analyses were performed with the survey commands in Stata 12.0 (StataCorp, 2011), which accounted for the complex multistage clustered design of the NSAL sample. All percentages reported are weighted.

Results

Distribution of all study variables by race/ethnicity is presented in Table 1. A higher proportion of non-Hispanic Whites reported an interpersonal problem (28.9%), and a higher proportion of older African Americans reported a bereavement issue (33.6%). More Black Caribbeans reported an emotional problem (26.9%) compared to African Americans (10.9%) or Whites (16.1%). Roughly two out of three (62.1%) of respondents sought assistance from at least one professional helper. Overall, the highest proportion of respondents sought assistance from family doctors (34.9%), followed by clergy (31.5%). Significant bivariate findings in professional service use indicated that a higher proportion of older non-Hispanic Whites and Black Caribbeans contacted a psychiatrist or other mental health professionals compared to African Americans and more Whites contacted and other health professionals compared to both African Americans and Black Caribbeans.

Table 1.

Distribution of study variables.^a

	Total		African Americans		Black Caribbeans		Non-Hispanic Whites		X2/F
	% (M)	N (S.D.)	% (M)	N (S.D.)	% (M)	N (S.D.)	% (M)	N (S.D.)
Gender
Male	44.5	322	42.4	170	49.4	68	45.4	84	0.4
Female	55.5	540	57.6	286	50.6	99	54.6	155
Age	66.38	8.39	66.22	8.38	64.51	8.06	66.52	8.6	1.31
Income	39686	42142	35498	40646	44520	33092	41539	49425	0.95
Education	12.44	3.3	11.83	3.51	11.91	3.33	12.75	2.7	2.52
Marital Status
Unmarried	53.6	570	62.6	324	46.1	96	49.6	150	5.18**
Married	46.4	292	37.4	132	53.9	71	50.4	89
Insurance Status
No Insurance	5	64	7.5	38	7.2	14	3.7	12	7.91***
Public coverage	21.5	251	33.7	174	26.2	38	15.4	39
Private coverage	73.6	547	58.8	244	66.6	115	80.9	188
Type of Problem
Physical	20.1	174	22.4	94	22.8	36	18.8	44	2.31*
Interpersonal	24.8	180	17.6	86	14.5	36	28.9	58
Emotional	14.7	96	10.9	43	26.9	16	16.1	37
Bereavement	26.8	255	33.6	157	18.8	34	23.8	64
Economic	12.1	123	13.6	59	11.7	36	11.4	28
Other	1.5	16	2.3	6	5.2	7	1.0	3
Professional Service Use
Any Professional	62.1	462	52.4	240	42.2	70	67.5	152	10.85***
# of Professionals	1.18	1.13	0.89	1.08	0.81	1.03	1.34	1.24	6.82**
Type of professional
Psychiatrist	13.1	76	8.6	36	13.8	10	15.2	30	3.09*
Other mental health	16.9	85	8.4	37	13.8	10	21.2	38	6.52**
Family Doctor	34.9	264	31.8	146	18.2	40	37.1	78	2.06
Other doctor	11.5	79	8.5	35	8.5	15	13.1	29	1.66
Other health professional	9.5	46	3.9	14	6.1	10	12.3	22	4.79*
Religious/spiritual	31.5	254	26.8	130	20.9	36	34.2	88	2.74
Healer	0.9	7	8.6	5	0	0	10.6	2	0.07

^aPercents are weighted; frequencies are unweighted. M= Mean, S.D. = Standard Deviation

Percents and N’s are presented for categorical variables; Means and Standard Deviations are presented for continuous variables.

Rao-Scott X² is used with categorical Variables and F test is used with continuous variables.

^*p < .05;

^**p < .01;

^***p < .001

Results of fitting latent class models

A three-class solution was chosen based on measures of model fit and because it was the most conceptually meaningful. Figure 1 depicts the differences in service providers visited across the three classes. Class 1, labeled “health plus clergy”, contained 5.1% (n=36) of respondents. The defining characteristic of this class was visiting health professionals or clergy for a serious personal problem. Members of Class 1 had a 100% probability of visiting a family doctor or another doctor. They also had a high probability of visiting a religious or spiritual leader (.88), a moderate probability of visiting another health professional (.63), a low probability of visiting a psychiatrist (.17) or another mental health professional (.10), and no probability of visiting another healer.

Figure 1.

Latent class profile plot of service providers visited

Class 2, labeled “doctor plus mental health”, contained 14.4% (n=71) of the respondents. Overall, this class reported visiting a doctor or a mental health professional for a serious personal problem. Members of this group had a high probability of visiting a family doctor (.71) or another mental health professional (.81) and a moderate probability of visiting a psychiatrist (.52). They had a low probability of visiting a religious/spiritual leader (.34), another health professional (.23), another doctor (.04), or another healer (.01).

Finally, class 3, labeled “limited provider use”, contained the majority of respondent (80.5%, n=755). Members of this class had a low probability of visiting all providers with the largest probability being for a religious/spiritual leader (.269) and a family doctor (.228).

Predictors of class membership

Table 2 presents the distribution of classes by race/ethnicity and type of problem. While the majority of all three racial/ethnic groups were in the limited provider class, a higher proportion of African Americans (90%) were in this class compared to Black Caribbeans (80.7%) or Whites (76%). More Whites (17.8%) were in the doctor plus mental health class followed by Black Caribbeans (13.6%) and African Americans (7.4%). A smaller proportion of African Americans were in the health plus clergy class (2.6%) compared to either Black Caribbeans (5.7%) or Whites (6.3%).

Table 2.

Distribution of classes by race/ethnicity and type of problem^a

	Health plus clergy		Doctor plus mental health		Limited provider use		X2	p
	%	n	%	n	%	n
Race/ethnicity
African American	2.6	13	7.4	31	90.0	412	4.54	0.003
Black Caribbean	5.7	9	13.6	9	80.7	149
White	6.3	14	17.8	31	76.0	194
Type of Problem
Physical	19.0	27	13.6	16	67.4	131	7.12	<.001
Interpersonal	0.2	1	12.5	13	87.3	166
Emotional	3.4	2	37.1	22	59.5	72
Bereavement	2.4	4	8.8	15	88.8	236
Economic	2.1	2	2.1	3	95.9	118

^aPercents are weighted; frequencies are unweighted.

A higher proportion of those with an interpersonal (87.3%), bereavement (88.8%), or economic problem (95.9%) fell in the limited provider use class. Nineteen percent of those with a physical problem fell in the health plus clergy class and 13.6% were in the doctor plus mental health class. Thirty-seven percent of those with an emotional problem were in the doctor plus mental health class.

Multinomial logistic regression was used to assess predictors of class membership, with the limited provider use class as the reference category (Table 3). Whites were more likely than African Americans to fall into both the health plus clergy and the doctor plus mental health classes. There were no significant differences in class membership between Black Caribbeans and either African Americans or Whites. Results for type of problem are reported with both a physical problem and an emotional problem as the reference category since these are the types of problems most commonly studied in previous research. Compared to those reporting a physical problem (as reference category), respondents who reported all other types of problems were less likely to be in the health plus clergy class. Those with an emotional problem were more likely to be in the doctor plus mental health class and those with an economic problem were less likely to be in this class compared to the limited provider use class. Compared to those with an emotional problem (as reference category), respondents reporting an interpersonal problem were less likely to be in the health plus clergy class and all problem types were associated with less likelihood of being in the doctor plus mental health class. In terms of other significant variables, those with more years of education and who were married were more likely to fall in the doctor plus mental health class. Respondents with public insurance coverage were less likely to fall in the health plus clergy class compared to the limited provider use class. Gender, age, and income were not significantly associated with class membership.

Table 3.

Multinomial logistic regression for class membership (limited provider use category as reference)

	Health plus clergy		Doctor plus mental health
	RRR	95% CI	RRR	95% CI
Black Caribbean vs. African American	2.39	.30 – 18.97	1.63	.35 – 7.60
White vs. African American	4.43*	1.31 – 15.01	2.75**	1.32 – 5.75
White vs. Black Caribbean	1.85	.21 – 16.10	1.69	.34 – 8.31
Male	2.62	.97 – 7.10	0.48	.20 – 1.13
Age	0.95	.90 – 1.00	1.02	.96 – 1.07
Income	0.94	.83 – 1.06	1.00	.96 – 1.04
Education	1.01	.85 – 1.20	1.17**	1.05 – 1.31
Married	1.57	.53 – 4.66	2.16*	1.11 – 4.24
Insurance Status (reference: none)
Public coverage	.31**	.13 – .74	0.86	.09 – 8.62
Private coverage	0.46	.08 – 2.48	0.38	.04 – 3.68
Physical problem as reference
Interpersonal	.01***	.00 – .06	0.75	.19 – 2.96
Emotional	.16*	.03 – .88	3.41*	1.16 – 10.05
Bereavement	.10**	.02 – .41	0.62	.24 – 1.58
Economic	.04**	.01 – .31	.12**	.03 – .54
Emotional problem as reference
Physical	6.20*	1.13 – 33.94	.29*	.10 – .86
Interpersonal	.04*	.00 – .50	.22**	.08 – .59
Bereavement	0.59	.13 – 2.77	.18*	.05 – .67
Economic	0.27	.03 – 2.46	.03**	.00 – .29

^*p<.05,

^**p<.01,

^***p<.001

Discussion

This study examines the types of professionals older adults visit for serious personal problems. Using latent class analysis, we add to existing knowledge in this area by documenting different combinations of providers used and the characteristics of older adults who fall in these groups. In addition, we examine both between (Black vs. White) and within (African American vs. Black Caribbean) group differences in these patterns.

The majority of older adults in this study fell into the limited provider use class suggesting that for many of life’s stressful problems, little or no professional help is utilized. This is consistent with previous research indicating a strong reliance among older adults on informal support (family, friends), either alone or in combination with professional services (Woodward et al., 2010). The finding is also consistent with work indicating that older adults are less likely to perceive a need for professional help for mental health issues (Mackenzie et al., 2010). The probability of talking to any one type of professional was quite low for this class suggesting less perceived need for professional services and the possibility that those services were complemented by informal support.

Overall, our hypotheses were partially supported. Consistent with previous research, older adults’ help-seeking preferences for physicians and clergy is clearly evident. Even in the limited provider use class, older adults had the highest probability of visiting a family doctor or a religious/spiritual advisor. Both of these professionals played a role in each class, although clergy were less prominent in the doctor plus mental health class. The importance of both physicians and clergy suggest that they each play different roles in relation to the identified problem. The health plus clergy class, for example, was particularly important for those for with physical health problems. While physicians address the physical health problem itself, clergy may help older adults deal with the emotional response to the problem. This might include activities as diverse as personal counseling for the potential loss of physical functioning resulting from chronic illness, accessing other community services such as Meals-on-Wheels, or mobilizing a church-based informal support network to provide meals, transportation, and other help during a health crisis. This study is not capable of determining the timing of visits to service professionals. However, it may be the case that clergy may also encourage older adults to visit their doctor for a physical health problem that they might otherwise ignore. This is consistent with Pescosolido’s Network Episode Model which emphasizes the complex reality of help seeking (Pescosolido, 1991). In this model there is no clear path to help seeking which combines both informal and professional supports in a variety of ways. While individuals sometimes make informed choices about where and when to seek help, they often describe a process of “muddling through” (Pescosolido, Gardner, & Lubell, 1998).

Older Whites were more likely than African Americans to use professional services overall, with a higher probability of being in both the health plus clergy and the doctor plus mental health classes. This is consistent with previous research related to overall service use. It is, however, counter to our hypothesis that African Americans and Black Caribbeans would rely more on clergy than Whites. Previous research finds both a greater reliance on religious/spiritual advisors among African Americans for a variety of problems (e.g., Bohnert et al., 2010) and more interactions with clergy among African Americans overall (Krause, 2008). This analysis found that older Whites and African Americans were comparable in their reliance on both a family doctor or a religious/spiritual advisor. This suggests that the difference in help-seeking between the two groups is perhaps not the use of any one provider, but the combination of professionals.

Some of this may be related to differences in the type of problem experienced. A higher proportion of African Americans, for example, reported the death of a loved one, a problem for which they may be more likely to only visit a religious/spiritual advisor. It is also possible that there is an interaction between race/ethnicity and the type of problem due to group differences in understandings and beliefs about the causes and appropriate roles for professionals in relation to different problems. These analyses, however, were not included because small samples sizes in classes 1 and 2 made the confidence intervals for some significant results very large (and unreliable). One benefit of this study was the ability to examine ethnic differences within the Black population (comparing African Americans and Black Caribbeans) in relation to the help-seeking behaviors of older adults. Contrary to our expectations, study findings indicated no difference in class membership between Black Caribbeans and either African Americans or Whites despite bivariate differences in both type of problem and the proportion reporting visiting individual providers. As with other findings in this study, this suggests that looking at combinations of providers may give a more nuanced understanding of both between and within group differences in help-seeking. Understanding within group differences is increasingly important as the population becomes more diverse. While Black Caribbeans are the largest and fastest growing subgroup among Blacks in the United States, Black communities also include immigrants from other areas of the world such as Africa and potentially differ in other ways such as geography within the U.S. (e.g., South, Northeast) and residence (e.g., urban, rural). Continuing to examine within as well as between groups differences can add to our understanding of racial/ethnic differences in help-seeking.

Finally, an interesting finding that was not part of the original focus of the study was that older adults who had public insurance were less likely to be in the health plus clergy class compared to the limited provider use class. One possible reason for this may be that Medicare coverage, the public insurance by which most older adults are covered, does not include many of the other providers included in this study. In fact, the health and mental health service systems deliver care provision in relation to diagnosable problems only, thus there may be some variation by problem type. Furthermore, because of the structure of the Medicare and Medicaid programs, respondents aged 55 to 64 with public insurance are more likely to be covered by Medicaid, which is needs-based, while Medicare is nearly universal for those aged 65 and older. Unfortunately, these data do not allow differentiation of insurance coverage at this level, limiting our ability to separate age differences from differences in coverage or to examine some of the complexities of coverage such as dual eligibility for both Medicare and Medicaid.

Limitations

This study has a number of limitations which should be considered. First, cross-sectional data limits our ability to understand the nature of help-seeking for personal problems as a process. It is impossible to determine whether respondents visited professionals concurrently or consecutively and, if consecutively, in what order. Further understanding of the entry points into professional care (i.e., visiting a clergy first or as a referral from a family doctor) would further add to our understanding. Related to this, we are unable to determine if the reason for the professional visit was specifically for the problem itself (e.g., financial help), to alleviate psychological distress associated with the problem, or for a combination of reasons.

Second, respondents were asked to talk about a problem that may have occurred at any point in their lifetime. Most responses were retrospective reports of a problem that happened more than a year before the interview, and some problems may have been experienced when the respondent was younger than age 55. Although allowing a retrospective approach enabled us to consider aspects of a problem most salient to the older adult, the reliability of the details about the problem is subject to potential recall bias. Further, the results are limited to examining help-seeking behaviors in the one example the respondent selected, which may or may not be representative of overall help-seeking patterns.

Another limitation is the relatively small number of older adults who fell into the health plus clergy and doctor plus mental health classes. From a substantive perspective, this suggests that for the majority of serious personal problems, limited professional help is needed or that there is substantial unmet need. Although we do not have a measure of severity of problem, we might assume that persons seeking additional professional help experienced more severe problems. From a statistical standpoint, the small number of cases for some categories may not only influence the reliability of some results, but also limits the ability to examine interactions between race and problem type which would have added valuable information to our analysis.

Conclusion

This study enhances our understanding of racial and ethnic differences in help seeking for serious personal problems among older adults by examining differences among Black Americans as well as between Blacks and Whites. Observed racial differences in this study were limited with the only difference being between Whites and African Americans.

Our use of latent class analysis to examine groupings of service providers confirms that primary care doctors and religious/spiritual leaders are important sources of help for older adults. In addition, although limited to the one serious personal problem identified by the respondent, this study expands the realm of help-seeking efforts beyond the typical issues of long-term care, physical health, and diagnosable mental disorders. Help-seeking and services for the treatment and care of physical and mental health problems are undoubtedly important concerns for older adults. However, older adults face stressors in other life domains as well (financial, interpersonal, bereavement). Understanding patterns of responses to and consequences of these stressors, and the factors associated with these differences, provides greater insight into the full spectrum of help-seeking behaviors of diverse groups of older adults.

This study provides the foundation for future research to examine these patterns more in-depth. For example, future studies might more closely examine the relationship between use of physicians and clergy in different circumstances, in particular, the order and timing of visits to different providers and the extent to which individuals discuss other help they are receiving with each provider. Just as efforts to integrate mental health specialty care into primary care practices has shown benefits in terms of access and quality of mental health care, improving communication between clergy and primary care may be beneficial, particularly in those communities where the church and clergy are an important source of support and access to other services may be limited.