Back to Annual Meeting Page		American Public Health Association 133rd Annual Meeting & Exposition December 10-14, 2005 Philadelphia, PA

Abstract #114949

Thomas M. Brady, PhD, Medstat, 4301 Connecticut Avenue, N.W., Suite 330, Washington, DC 20008, (202) 719-7871, thomas.brady@thomson.com

Background: A significant body of research demonstrates that racial and ethnic minorities receive a lower quality and intensity of health care, even when factors such as education, income, and the severity of presenting symptoms are taken into account (Institute of Medicine, 2002). The purpose of this study is to explore the effects of race and ethnicity on the utilization of addiction treatment. Retention, expressed as the length of stay (LOS) in treatment, was the dependent variable. Studies have previously identified race and ethnic differences in treatment retention however the research has employed small samples (n<300) from single treatment providers (Milligan, 2004; McCaul, 2001). Retention is often used as an analytic endpoint because research suggests longer retention in treatment is associated with improved outcomes such as gains in employment and decreases in criminal behaviors. Method: Admission (1999-2001) and discharge data (2001) of the Treatment Episode Data Set (TEDS) were merged. TEDS data come from facilities that receive some public funding. In 2001, 22 states provided 555,090 admission/discharge records of treatment episodes for analysis. Inpatient, residential and free-standing detoxification records (n=154,149) were excluded because they generally do not resemble substance abuse treatment episodes. Cox's proportional hazard regression was used to model LOS. The multivariable analysis controlled for age, sex, education, employment status, type of primary drug use, and frequency of use. The analysis also controlled for treatment modality (outpatient drug free, outpatient methadone, residential and inpatient facility) as well as year of admission to treatment. Results: The average LOS was 64.1 days for Latinos, 56.1 days for American Indian/Alaskan Natives, 73.2 days for Asian Americans, 66.8 days for African Americans and 71.8 days for Whites. These data vary a great deal by frequency of drug use, treatment modality and year of admission to treatment. After statistical adjustment, the data suggest that race was associated with retention. The odds were that Latinos (Wald Test=270.431, p < 0.001) and African Americans (Wald Test=13.228, p < 0.001) were more likely to exit treatment earlier than Whites. Implications: Race/ethnic differences in the utilization of addiction treatment services may persist after adjustment for some socioeconomic and drug use severity factors. It is not clear however if race and ethnic differences may be attributable to unobservable events such as poverty or health system factors such as insurance or source of referral. Addiction treatment policy must address barriers to treatment retention for some minorities.

Learning Objectives: