Using Dynamic Risk to Enhance Conditional Release Decisions in Prisoners to Improve Their Outcomes.

Advances in criminal risk assessment have increased sufficiently that inclusion of valid risk measures to anchor assessments is considered a best practice in release decision-making and community supervision by many paroling authorities and probation agencies. This article highlights how decision accuracy at several key stages of the offender's release and supervision process could be further enhanced by the inclusion of dynamic factors. In cases where the timing of release is discretionary and not legislated, the utilization of a validated decision framework can improve transparency and potentially reduce decision errors. In cases where release is by statute, there is still merit in using dynamic risk assessment and case analysis to inform the assignment of release conditions, thereby attending to re-entry and public safety considerations. Finally, preliminary results from a recent study are presented to highlight the fact that community supervision outcomes may be improved by incorporating changes in dynamic risk into case planning and risk management, although this work requires replication with larger populations reflecting diverse groups of offenders. Nonetheless, these decision strategies have implications for both resource allocation and client outcomes, as outlined here. Copyright © 2016 John Wiley & Sons, Ltd.

Rehabilitation group coparticipants' risk levels are associated with offenders' treatment performance, treatment change, and recidivism.

OBJECTIVE: Exposure to antisocial others within treatment group sessions may have negative impact. We extend prior research with adolescents by examining rehabilitation group composition among adult male incarcerated offenders. METHOD: Data were gathered from institution files of rehabilitation completers (N = 1,832; M age = 33.5; 19% Aboriginal, 68% Caucasian), including general, substance, violent, and sex offenders. Capacities for treatment (including motivation, learning ability, and inhibitory control) were gathered from intake assessments. At the beginning and end of rehabilitation, providers rated program performance. Risk for recidivism and postrelease recidivism were gathered from official files, up to 3 years following release. RESULTS: Group effects accounted for up to 40% of variance in program outcomes. Group features (average group participant risk to reoffend and risk score diversity) significantly interacted with treatment capacities to explain program outcomes. Most models revealed a dampening effect whereby the positive association between capacities and outcome was reduced in groups of higher risk and more risk diverse coparticipants. Group composition typically accounted for 30-38% of variance between groups, but total variance in outcome explained was generally small. Higher average group risk predicted postrelease recidivism among family violence offenders. CONCLUSIONS: Coparticipants should be considered when researching and providing group programs to adult offenders, with specific attention toward how positive outcomes may be attenuated in the presence of criminogenic coparticipants.