Community mental health services to minority groups: Some optimism, some pessimism

Community Mental Health Services to Minority Groups Some Optimism, Some Pessimism STANLEY SUE

ABSTRACT: For many years, researchers and practitioners have found that minority-group clients who seek psychotherapeutic services receive discriminatory treatment from white therapists. Underlying this finding is the implicit assumption that the mental health delivery system should strive to provide equal and nondiscriminatory services for all clients. An analysis of the services received by minority clients in 17 community mental health facilities suggested that blacks received differential treatment and poorer outcomes than whites. However, Asian-American, Chicano, and native American clients who tended to receive treatment equal to that of white clients also had poorer outcomes as measured by premature termination rates. It is suggested that a time may well come when minority clients receive equal but unresponsive services and that primary attention should be placed upon the delivery of responsive services rather than upon the demonstration of inequities.

I have become increasingly concerned about the delivery of mental health services to minority groups—especially since the APA Task Force on Health Research (1976) has indicated (a) that good health is one of the top concerns of Americans and (b) that large segments of the population, particularly the disadvantaged and minority groups, lack access to adequate health care. These inadequacies are also apparent in mental health care. Because community mental health centers represent one of our major resources in the promotion of mental health, it is vital that these facilities respond to the needs of minority groups. All of us are aware of racism, but its effects on service delivery are both direct and subtle. Without belaboring the point, I do want to review some findings in this area as a background for discussing the Seattle Project, which examined the kinds of services received by ethnic groups. For several decades, various researchers have found that minority-group clients receive unequal 616 • AUGUST 1977 • AMERICAN PSYCHOLOGIST

University of Washington

and poor mental health services. For example, Yamamoto, James, and Palley (1968) examined the services received by 387 (65%) Caucasians, 149 (25%) blacks, S3 (9%) Chicanos, and 5 ( 1 % ) Asians admitted as consecutively new patients at the Los Angeles (California) County General Hospital Outpatient Clinic. The findings indicated that compared to Caucasian patients, minority-group members were discharged more quickly and were more often seen for minimal supportive psychotherapy rather than individual or group therapy. Such findings are, of course, not isolated, and individuals including Acosta and Sheehan (1976), Carkhuff (1972), Clark (1965), Padilla, Ruiz, and Alvarez (1975), Lerner (1972), Willie, Kramer, and Brown (1973), and Attneave (Note 1) have pointed to the difficulties that minority-group clients experience in receiving adequate mental health care. Schofield's (1964) oftencited YAVIS syndrome is also appropriate here. He found that therapists tend to prefer young, attractive, verbal, intelligent, and successful individuals for clients, several factors that place minority groups and the poor at a disadvantage in being preferred for therapy. This has led Goldstein and Simonson (1971) to suggest tongue in cheek that greater research and intervention be focused on the HOUND patient—the homely, old, unattractive, nonverbal, and dumb client.

An earlier version of this article was an invited address to the Center for Minority Group Mental Health Programs and the Community Mental Health Services Support Branch of the National Institute of Mental Health, Rockville, Maryland, May 1976. This research was supported by a grant from the Social Science Research Council and by Grant No. MH 22090-01 from the National Institute of Mental Health. Requests for reprints should be sent to Stanley Sue, Department of Psychology, University of Washington, Seattle, Washington 98195.

(I)

. POOR OUTCOME DIFFERENTIAL . , TREATMENT

'

(2)

' GOOD OUTCOME TYPE OF CLIENT

(3)

. POOR OUTCOME v

EQUAL TREATMENT

(4)

' GOOD OUTCOME

Figure 1. Models of service delivery and outcome. Figure 1 shows qualitatively different models of service delivery and presumed outcome. The first model assumes that the type of client (in this case, ethnic-group client) receives differential treatment which results in a poor outcome. Most of our traditional research efforts have been of this kind—namely, the demonstration that ethnic clients receive different (poorer) services and poorer outcome. The second model assumes that ethnic clients, by virtue of subcultural backgrounds, require differential services in order to facilitate a good outcome. In contrast to Model 1, the differences in treatment are intended to provide a "fit" between treatment technique and client's life-style rather than differences that are discriminatory and therapeutically inferior. Model 3 is based upon an extension of the arguments for the second model. Here, a minority client receives nondiscriminatory (equal) treatment, but the outcome is unfavorable because the fit between treatment and client's life-style is poor. Finally, the last model raises the possibility that equal services may be effective for some ethnic-group clients, and I would speculate that these clients have lifestyles that are similar to those of other mainstream Americans. Obviously, the models are quite molar, and I have not dealt much with differences within minority clients, nor with clearly defined treatment modalities and outcome. I shall return to these points later.

The Seattle Project The major question we entertained in the Seattle Project was the kind of model that best applied to the status of minority groups in this location. We were excited because it was possible to collect detailed information on nearly 14,000 clients seen in 17 community mental health centers over a 3year period in the greater Seattle area. Each facility was required by the state to collect the

information on standard face sheets and to keep records on all clients during their contacts. The information included client's ethnicity, age, sex, income, educational level, and marital status. Data were also available on diagnosis, type of program (inpatient, outpatient, day treatment, etc.), type of service (intake diagnosis,.individual therapy, group therapy, family therapy, marital counseling, etc.), and type of personnel at intake and during therapy (psychologist, psychiatrist, social worker, nurse, teacher, vocational rehabilitation counselor, physician, other professional, and other personnel). Finally, we had the number of contacts that clients had with the facilities. The data were exciting because we were not limited to only one mental health facility (in fact, we were dealing with a rather closed system of federal- and state-supported community mental health centers) and, as a result, gathered information in greater detail than we had ever seen reported previously in the literature. There were, however, limitations, such as the unknown reliability and validity of the data and the lack of information on the precise experiences and interactions of clients and staff. The results presented here represent a summary and elaboration of various reports (Sue, Allen, & Conaway, in press; Sue & McKinney, 1975; Sue, McKinney, Allen, & Hall, 1974) that we have made on each ethnic group. Specific details on the findings can be found in those reports. This presentation is the first integration of the results in order to provide a better gestalt. In fact, I have come to the conclusion that the gestalt is more than the sum of the individual reports. Let me begin by addressing myself to four specific questions in addition to the general issue of which of the four models seems applicable. First, do Asian-Americans, blacks, Chicanos, and native Americans differ in their utilization patterns? While this may seem like a trivial question, it is not. Even now, many studies report comparisons between whites and blacks or nonwhites (Kramer, 'Rosen, & Willis, 1973). Studies of this kind are understandable in view of past data-collection procedures that excluded specific ethnic categories or in view of the small numbers of nonblack minorities. If, however, we talk about race or ethnicity as a general variable, then specific ethnic groups must be delineated, and findings should show some consistency for each group. Even the concepts of native Americans and Asian-Americans encompass many different groups. The second question is, AMERICAN PSYCHOLOGIST • AUGUST 1977 • 617

Do ethnic-group clients exhibit other demographic differences from whites? Again, many previous studies (e.g., Malzberg, 1959) have made racial comparisons with little or no attempts to control variables such as social class, age, sex, marital status, etc.—factors known to correlate with type of treatment or with prognosis. In other words, we wanted to find out if ethnicity contributes to type of treatment and to outcome above and beyond other demographic characteristics. Third, what kinds of service modalities are received by minority groups? Is there evidence of differential treatment? Fourth and finally, what implications can be drawn, beyond the present study, for outcome, improvement of services, and a conceptualization of service delivery? What kinds of policy recommendations can be drawn? With these specific questions in mind, let us look at the results.

TABLE 2

Demographic Characteristics Native AsianBlacks Americans Americans Chicanos Whites

Characteristic Age (M) Education (M) Sex Male Female Monthly income None Below $100 $100-249 $250-499 $500-749 $750-999 $1,000 or more Unknown Marital status Single Married Separated Divorced Other Unknown

27.2** 93 t*«

24.4***

52%*** 48

31.5*** 9.7***

28.3 9.2***

28.7 10.5

50%

40%

50

60

56%* 44

42% 58

13%*** 10

11%** 6

27

30 19

8

7

2

2

2

2 23

13%** 6 30 19 9 2 2 19

6% 5 18

13

11%** 7 22 16 9 4 8 23 44% 30 8 10 3 5

44% 32 11 11 2 0

24

53%*** 17 12 13

3 3

8.4***

56% 23 8 11 1 1

22

15 9 8 17 42% 33 8 13 2 1

Note. Each ethnic group was compared to the white group. ( tests were employed for age and education, chi-squares for all other comparisons. *p < .001.

Residts UTILIZATION As indicated in Table 1, blacks and native Americans were heavily overrepresented in the community mental health centers. However, AsianAmerican and Chicano clients were several times underrepresented. The answer to the first question is that various ethnic groups may be quite different in their utilization of public mental health facilities. DEMOGRAPHIC DIFFERENCES

Comparisons between each ethnic group and the Caucasian group (a 10% random sample, or 1,190) in demographic characteristics are revealed in Table 2. Blacks and native Americans were TABLE 1

Utilization Patterns in Seattle-King County Patients in 17 CMHCs Group

White Black Native American Asian-American Chicano Total

significantly younger, and Asian-Americans were older, than whites. In terms of education and income, all of the ethnic groups were significantly lower than whites. Sex differences, were found, in that blacks and Chicanos had a higher proportion of males than whites did. Finally, blacks were less often married than whites. In Table 3, we see a summary of the significant demographic differences. It is apparent that from the very outset, minority-group clients represent very different kinds of individuals from white clients, particularly among black Americans. DIAGNOSIS Interestingly, no overall differences emerged in diagnosis for native Americans, Asian-Americans, and Chicanos when compared to whites (see Table 4). Blacks did show a significant overall differ-

County population

N

%

2V

11,904 959 152 100 83

90.2 7.3 1.2 .7 .6

1,076,216 40,597 7,391 27,912 20,952

91.7 3.5

13,198

100.0

1,173,068

100.0

%

.6 2.4 1.8

Note. 252 patients were listed as other or missing. CMHCs = com-

618 • AUGUST 1977 • AMERICAN PSYCHOLOGIST

TABLE 3

Summary of Demographic Characteristics Compared to Whites Characteristic Age Sex Education Income Marital

Blacks Younger More males Less Less Less married

Native Americans

AsianAmericans

Younger

Older

Less Less

Less Less

Chicanos More males Less Less

TABLE 4

Percentages of Diagnosis at Intake Diagnosis Mental retardation Organic brain syndrome Psychosis Neurosis Personality disorder Psychophysiological Transient situational Behavior disorder Other

Blacks**

Native Americans

AsianAmericans

1.8 2.4

2.7 1.4

1.7 1.7

1.8 1.8

2.3 1.5

13.8 11.6 33.3*

17.6 12.2 18.9

22.4 19.0 15.5

14.5 16.4 25.5

12.7 12.8 14.2

0 15.0*

Chicanos

0

0

0

17.6

' 13.8

14.5

Whites

.3 22.5

3.0

4.1

0

0

19.1*

25.5

25.9

25.5

29.8

507

74

58

55

794

3.9

Note. Comparisons were made by using chi-squares. *t < .01. **p < .001.

they were more likely to receive a diagnosis of personality disorder and other disturbance (nonpsychiatric) and less likely to be in the'category of transient situational disturbance than whites. Thus, from the perspective of diagnosticians, only blacks were more likely to be placed in different diagnostic categories. TYPES OF PERSONNEL SEEN

Tables S and 6 indicate the kind of personnel clients saw at intake (initial contact) and during therapy. Except for Chicanos, all other ethnic groups differed overall from whites in the kind of personnel seen at intake. Blacks significantly saw more other professionals, nonprofessionals, and

other personnel and fewer psychiatrists, psychologists, social workers, and nurses. When "personnel" was divided into professionals (psychiatrists, psychologists, and social workers) and nonprofessionals (including other personnel), blacks were more often than whites seen by nonprofessionals even after other demographic variables were controlled through partial correlations, r (1446) = .22, p < .001. Native Americans saw more social workers and other personnel and fewer nonprofessionals than whites did. A professional versus nonprofessional distinction was not significant. Asian-Americans were more likely at intake to encounter teachers, other professionals, and other personnel and less likely to see psychologists than whites were. Asian-Americans more often worked

TABLE 5

Percentages of Personnel at Intake Personnel Psychiatrist Psychologist Social worker Other physician Nurse Teacher Vocational rehabilitation counselor Other professional Nonprofessional Other

Note, *# **# *** t

Blacks***

2.1** 3.4** 33.1**

.5 2.8**

.8 .6 20.9** 26.9** 6.1**

949

Native Americans**

AsianAmericans***

6.7 6.0

3.1

51.3*

0 7.3 .7

2.1* 40.6

0 7.3 2.1*

Chicanos

Whites

6.0 8,4

7.2 9.5

34.9

40.4

0 4.8 0

.5 9.9 .2 .6 8.5

1.0

0

14.7* 2.7*

17.7** 20.8 5.2**

18.1 27.7

150

96

83

2.0 8.7

0

22.4

.7

1,177

Each ethnic group was compared to white group. Comparisons were made using chl-squares.