Juvenile offender reentry in an urban setting: an evaluation of an intensive supervision program using a ‘time to failure’ model

The current study aimed to determine the extent to which a juvenile reentry program consisting of intensive supervision and collaboration between service providers was able to increase the amount of ‘time to failure’ and decrease the number of ‘days in custody’ for program participants when compared to a historical comparison group. This was done by using a quasi-experimental design, coupled with stratified propensity score matching to ensure group equivalency. Kaplan-Meier and Cox regression survival analyses along with t-tests found that survival likelihood was higher for the comparison group. It was also found that the rate of return to custody was substantially hastened for the treatment group, and the risk of failure climbed during the study period for both groups. The risk of failure also climbed when the treatment was combined with a court-mandated substance abuse intervention. The treatment group was however found to be more likely to receive education, and substance abuse and anger management treatments; as well as spend less time in custody upon failure. The above findings are likely due to treatment subjects being supervised more intensively, with non-compliance more likely to be recorded and acted upon – particularly with regard to failure to attend substance abuse treatment programs. Findings may also be due to a lack of adherence to principles central to the risk-need-responsivity approach to community supervision. Policy implications include the need to avoid unnecessary court mandated treatments; the need to further explore the use of collaborative approaches to supervision with a larger dosage to ensure access to essential treatments, while simultaneously excluding in this process the potentially harsher responses to non-compliance that seem to come with increased supervision; and the need for criminal justice agencies to collect more incident-based information for stronger evaluations of future programming.Ph.D.Includes bibliographical referencesIncludes vitaby Victoria Ann Sytsm