Antipsychotic medication adherence, cocaine use, and recidivism among a parolee sample.

This study examined the independent and interactive associations between cocaine use and antipsychotic medication adherence in predicting 12 month criminal recidivism among a sample of mentally ill parolees (N = 200). Consistent with prior research, cocaine use (based on hair assays) was associated with more than a threefold increase (relative to non-cocaine users) in the likelihood of a parolee being returned to custody during the follow-up period. Although medication adherence (based on urine specimens) was not independently associated with a significant reduction in recidivism risk, the interaction between cocaine use and medication adherence was significant, revealing a disproportionate impact of medication adherence specific to cocaine users. Prediction models of recidivism based on self-reported measures of medication adherence and cocaine use revealed only marginally significant trends for cocaine use, no effect for adherence, and no significant interaction between these two predictors.