“You Can Get Prepared, But It’s Still Scary Walking Out The Door”: Exploring the Role of Dynamic Re-entry Factors in Release Planning for High-Risk Offenders

The task of preparing high-risk prisoners for the multitude of challenges they will face once released is vital to their chances of successful re-entry. Recent research in New Zealand has found that developing good quality plans for life after prison is associated with reduced rates of reoffending after release – but how? One suggestion is that release plans help to ameliorate risks in offenders’ release environments. However, research examining how these risk factors are affected by the quality of release plans is scarce. This thesis investigates whether release planning has an indirect relationship with recidivism through its influence on dynamic risk and protective factors in re-entry, as measured by a risk management tool used by Community Probation Services in New Zealand: the Dynamic Risk Assessment for Offender Re-entry (DRAOR; Serin, 2007). A coding protocol to assess the quality of release plans was developed and retrospectively applied to a sample of 303 high-risk male parolees. Outcomes of interest were “short-term recidivism” (within 100 days of release) and “longer-term recidivism” (within one year of release) across four different indices. Results indicated that parolees who did not reoffend within the first 100 days of release had significantly better quality release plans than those who did reoffend. Better quality release plans also predicted greater stability in acute risk factors, and greater improvements in overall DRAOR scores, within the first 100 days of release. Logistic mediation analyses confirmed that release planning had an indirect relationship with both short-term and longer-term recidivism through its influence on DRAOR total scores. Together, these findings suggest that release planning may facilitate successful re-entry by reducing the impact of acute triggers or destabilisers in the release environment, thus protecting against a potential relapse. Theoretical and practical implications of these findings are discussed, along with limitations of the study and suggested directions for future research.