The Problems (and possible solutions) of assessing risk, race and recidivism in long operating drug treatment courts

Abstract

Formal criminogenic risk tools can be an important control in assessing racial inequities in access to treatment courts and in evaluating both proximal and distal outcomes from those programs. To achieve this potential, however, it is important that risk tools themselves operate in a racially neutral fashion and that they operate consistently over the period assessed. Tools that are not properly calibrated by race and changes in the tools used over the life of a program are therefore significant evaluation concerns. Our paper is the first to assess the adequacy of an important risk-needs instrument, the LSI-R, across racial groups in a drug treatment court setting. The main contribution of the current study is not as a test of that instrument, which has been widely studied in other settings. Rather, because two different criminogenic risk tools were used over the study time period, we took this opportunity to explore the use of a readily constructible “proxy” measure of risk to support analysis of risk and race interactions over the life of the program.