Methods: Four data-bases were searched using the keywords: diversion, diverted, youth court, restorative justice, victim offender mediation, drug court, mental health court, first time offender, experiment, juvenile justice. 471 article abstracts were reviewed, 29 studies met the following inclusion criteria: experimental or quasi-experimental design comparing at least 1 active treatment condition against a minimal intervention or standard justice processing condition, published 1980 or after, served youths referred to the juvenile justice system, diverted to alternative services prior to adjudication. A graduate research assistant and the first author coded the studies. Interrater reliability was assessed (>90% agreement) and discrepancies were resolved through consensus. Heterogeneity in effect sizes (Q=63.13) supported a random effects model.
Results: The 29 studies yielded 58 comparisons involving 19,301 youths. Forty-five comparisons involved recidivism as a dichotomous measure. Across these studies, the effect of diversion programs on recidivism was not significant (OR=.83, 95%CI=.43-1.58). Effect sizes were moderated by two sample characteristics. Family interventions (k=4) were associated with statistically significant reductions in recidivism (OR=.57, 95%CI=.40-.82), and treatment studies that assessed specific manualized interventions with high levels of researcher involvement (k=6) were associated with recidivism reduction regardless of intervention type (OR=.54, 95%CI=.33-.87). Case management (k=18), individual treatment (k=11), youth court (k=6), and restorative justice programs (k=6) were not associated with statistically significant reductions in recidivism.
Conclusions and Implications: On average, diversion programs fail to deliver their promised outcomes. Results of this study suggest that an overreliance on generic case management approaches, limited family involvement, and infrequent utilization of manualized interventions may contribute to these disappointing findings. Recent research in evidence-based approaches in the treatment of delinquency suggest that the provision of specific interventions with high fidelity delivered in as-usual treatment settings can sustain effects demonstrated in more controlled settings. Findings of this study point to the need for clinical support mechanisms to sustain fidelity to manualized, family-based diversion models. The current study also underscores the importance of family involvement, which may be a particularly important moderator in the collectivist Arab community context. The findings will inform the ongoing work of local Jordanian stakeholder working groups in developing and implementing the first alternatives to institutionalization for children in conflict with the law in the Kingdom.
