Performance measures and racial/ethnic disparities in the treatment of substance use disorders

A limited literature on racial/ethnic disparities in the treatment of substance use disorders suggests that quality of treatment may differ based on client's race/ethnicity. This study examined whether (a) disparities exist in the probability of treatment engagement, a performance measure for substance use disorders, and (b) treatment engagement is associated with similar reductions in likelihood of arrest for Whites, Blacks, Latinos, and American Indians. Adult clients who began an outpatient treatment episode in 2008 in public sector specialty treatment facilities in Connecticut, New York, Oklahoma, and Washington made up the sample (N = 108,654). Administrative treatment data were linked to criminal justice data. The criminal justice outcome was defined as an arrest within a year after beginning treatment. Engagement is defined as receiving a treatment service within 14 days of beginning a new outpatient treatment episode and at least two additional services within the next 30 days. Two-step Heckman probit models and hierarchical time-to-event models were used in the analyses. Black clients in New York and American Indian clients in Washington had significantly lower likelihood of engagement than White clients. As moderators of engagement, race/ethnicity had inconsistent effects across states on the hazard of arrest. Racial/ethnic minority groups may benefit from additional treatment support to reduce criminal justice involvement. States should examine whether disparities exist within their treatment system and incorporate disparities reduction in their quality improvement initiatives.