Measuring Comprehensive Service Provision for African Americans, Caucasians and Latinos in Substance Abuse Treatment

Background and Purpose: Current research has shown that members of diverse racial and ethnic groups have distinct service needs. Treatment programs have responded to these diverse needs by stressing the provision of comprehensive services. Nonetheless, little is known about what models of services are more effective for members of different racial and ethnic groups. This study seeks to understand service comprehensiveness by using a new approach to enhance measurement of receipt of comprehensive services. In addition, this study explores the benefits and limitations of using this new measure to explain outcomes in Latino, African American and Caucasian clients.

Methods: Data were collected from 1992 through 1997 for the National Treatment Improvement Evaluation Study (NTIES), a prospective, cohort study of substance abuse treatment programs and their clients. The analytic sample consists of 1,812 Blacks, 486 Hispanics and 844 Caucasians (1,123 women and 2,019 men). Rash measurement models (Wright & Masters, 1982) were used to create a measure for each race/ethnic group representing clusters of significant services associated with access, retention and coping skills. This innovative measurement approach allows for the examination of the ranking of each service per group, while it also explores their aggregate effect using a composite measure representing degree of individual comprehensive services received. This measure was finally used in multivariate analyses to explore the relationship of comprehensive services to post-treatment drug use.

Results: The data showed modest variability in the ranking of services received based on clients' racial/ethnic groups. The three groups reported a similar pattern using services such as child care, job skills and parenting training, mental health treatment and transportation. Transportation services were commonly obtained, while child care was rarely received by either member of these three groups. English as a second language, which may be a need associated with Latino immigrants, was received more frequently by Caucasian clients in this sample. The Rasch measures, representing degree of comprehensive services received, proved to be highly reliable. When included in separate multivariate analyses for each racial/ethnic group, the comprehensive service measure was significantly related to reduced post-treatment substance use for African Americans and Caucasians, but not for Latinos.

Conclusions and Implications: These results identify that the particular clusters of services received by Caucasian and African American clients are associated with a reduction in drug use. African American clients on average reported a larger reduction of drug use when they received more services. Using Rasch models to build a composite measure of services for individuals has the benefit of comparing individuals within and between race/ethnic groups. In this way, this measure can be used in rigorous analytic models to support the notion that not only comprehensiveness but also specific clusters of individual services may be influential in helping individuals from different groups reduce their post-treatment drug use. This approach may improve methodologies and inform program designs regarding patterns of services and their aggregate effect on treatment outcomes.