Jail recidivists with serious mental illness and substance use disorders were treated in an in-custody setting and then randomly assigned to either a high fidelity Integrated Dual Disorders Treatment program (103 participants) or to service as usual (79 participants). Outcomes were tracked an average of 18 months from program entry at the termination of the initial incarceration. A reduction in jail days from baseline to study period was significant for both groups. The pre to post reduction for arrests and total convictions was significant in the experimental group but not the control group. However, during the study period, differences between experimental and control groups in arrests, convictions and jail days were not statistically significant. Experimental participants had lower study period psychiatric inpatient and crisis utilization and greater outpatient utilization than did control group participants. The groups did not differ with regard to total institutional days. Experimental group attrition was relatively high.