Substance Use is a Robust Predictor of Adolescent Recidivism

Abstract

How well does substance use predict adolescent recidivism? When the Cox proportional hazards model was applied to officially recorded first rearrest of 505 juvenile offenders, a best-fitting complex multivariate model indicated that: (a) parent reports that youths “often” use substances more than doubles first rearrest risk, (b) averaged youth and parent substance use reports predict recidivism better than a single source, (c) parent or youth denial of youth substance use predicts recidivism, (d) age at first arrest does not predict recidivism, (e) non-White/non-Asians have a 79% higher recidivism risk than peers, (f) parent-reported delinquency predicts recidivism with declining accuracy, and (g) substance use robustly predicts recidivism despite prior reported delinquency, gender, ethnicity, age, follow-up time, or data source. Findings are related to host-provocation theory.