Ethnic Disparities in Unmet Need for Alcoholism, Drug Abuse, and Mental Health Care

We analyzed data from Healthcare for Communities, a national survey funded by the Robert Wood Johnson Foundation. The Healthcare for Communities survey reinterviewed participants in the Community Tracking Study (17) about 14 months after their initial interview. The Healthcare for Communities sample was a stratified random sample of 14,985 of the 30,375 adult telephone respondents in the Community Tracking Study; it was oversampled for psychological distress (i.e., shown by those scoring above a cutoff point on a two-item screen of the mental health inventory from the Medical Outcomes Study); use of prior specialty alcoholism, drug abuse, and mental health treatment; and family income below $20,000. We obtained 9,585 eligible responses (for a 64% rate of response). Data were weighted for the sampling design and for responses on each survey to represent the noninstitutionalized adult U.S. population. The weighted Healthcare for Communities sample closely matches the 1997 U.S. household population in sociodemographic characteristics (18). The study design has been described elsewhere (19).

As expected, the ethnic groups differed significantly in demographic characteristics and insurance type (Table 1). Relative to whites, Hispanics and African Americans had lower mean incomes and less education and were younger and less likely to be married. Hispanics and African Americans were more likely than whites to be uninsured or to be covered by Medicaid, while whites were more likely to be covered under Medicare or unmanaged or partially managed private insurance. Compared to whites, African Americans had higher rates of probable mental disorders (13.4% versus 20.1%, respectively; all data were nationally weighted and unadjusted) and of perceived need for substance abuse treatment (1.1% versus 2.3).

The percentage receiving any alcoholism, drug abuse, or mental health treatment was 31.9% for whites, 28.1% for African Americans, and 30.1% for Hispanics (Table 2). While the overall effect of ethnicity was not significant for any indicator of access, both minority groups had point estimates lower than those of whites for each access indicator, and African Americans had significantly lower use than whites when one-tailed tests were used on adjusted models for any use of alcoholism, drug abuse, or mental health treatment; use of primary care alcoholism, drug abuse, or mental health treatment; and use of specialty care alcoholism, drug abuse, or mental health treatment.

Among those with perceived need for care for alcoholism, drug abuse, or mental health disorders, minorities were more likely to report unmet need (12.5% of whites, 25.4% of African Americans, and 22.6% of Hispanics; data were nationally weighted). These differences were significant in unadjusted models, and the difference between African Americans and whites was significant in adjusted models (Table 2). Furthermore, Hispanics tended to have more delays in care than whites in adjusted and unadjusted models (e.g., 22.7% versus 10.7%).

Among those with perceived or clinical need, whites were more likely to be receiving active treatment (37.6%) than either African Americans (25.0%) or Hispanics (22.4%), and both adjusted and unadjusted comparisons were significant. In addition, Hispanics were less satisfied than whites with every component of health care. These comparisons were significant for all unadjusted models and for satisfaction with mental health and substance abuse care in adjusted models.

We found consistent ethnic differences, all in the same direction: less access to care, poor quality of care, and greater unmet need for alcoholism, drug abuse, and mental health treatment for Hispanics and African Americans in comparison to whites. With such consistency, formal correction for multiple statistical comparisons was deemed too conservative, and the overall pattern of findings for each minority group in comparison with whites was considered most relevant. For African Americans, the pattern included less access to care and greater unmet need for care among those with alcoholism, drug abuse, or mental health needs and also a lower rate of active treatment among those in need, relative to whites. For Hispanics, the pattern included more delays in receiving care, lower satisfaction with care, and lower rates of active treatment among those in need. We had somewhat less precision for comparing access for Hispanics versus whites than for African Americans versus whites, but the observed differences in access were relatively small. The ethnic differences in unmet need for care and for quality of care, however, seem large. For example, the percent of those with unmet need for alcoholism, drug abuse, and mental health care was twice as high for African Americans as for whites, and the percentage of those in need who were receiving active treatment was nearly 50% less for Hispanics than for whites. Had we applied formal correction for multiple comparisons, differences in unmet need for care and for active treatment would still have been significant. Furthermore, these findings were robust after control for other individual characteristics, including indicators of socioeconomic status.

Our findings complement those of Young et al. (16), who reported that African Americans have lower rates of guideline concordant care for depressive and anxiety disorders than whites. Our findings suggest, however, that some types of quality-of-care problems extend to Hispanics and involve treatment for alcoholism, drug abuse, and mental health conditions. Examples of new policies that might address these unmet needs are programs that extend insurance coverage to the near-poor who are in need of care and implementation of quality-improvement programs for major psychiatric disorders in community-based health care settings. For example, effective quality improvement programs have been developed for patients with depressive disorders in primary care settings (30–32), but few such programs have been evaluated or implemented for minorities or for those with other psychiatric disorders. Developing programs to improve access to care and quality of care for minorities should be a high priority for psychiatrists, other mental health specialists, and general medical clinicians.

Our study has important limitations, including modest sample sizes of minority groups, an absence of measures of acculturation, reliance on brief screening measures for psychiatric disorders, and use of self-reports regarding utilization. We had only moderate rates of response among persons who previously participated in a national survey, so nonresponse combined across surveys could have biased our results. We applied nationally representative weights to all phases of the study to control for attrition.

Overall, our findings document ethnic disparities in access to care, unmet need for care, and quality of alcoholism, drug abuse, and mental health care and emphasize the importance of public education and interventions in medical and psychiatric practice to broadly improve the quality of care for ethnic minorities.

Received Oct. 6, 2000; revision received April 2, 2001; accepted June 13, 2001. From RAND; the UCLA Neuropsychiatric Institute and Hospital and the UCLA Clinical Scholars; and the University of California, Davis, Sacramento. Address reprint requests to Dr. Wells, RAND, 1700 Main St., P.O. Box 2138, Santa Monica, CA 90407-2138; [email protected] (e-mail). Funded by grants from the Robert Wood Johnson Foundation, Health Care for Communities (031280), the NIMH Research Center on Managed Care for Psychiatric Disorders (MH-01170), and the UCLA Faculty Scholars Program (MH-00990). The authors thank Fuan Yue Kung, M.S., for programming support and Lingqi Tang, Ph.D., for statistical advice.