Court-Mandated Restorative Justice for Domestic Violence Offenders: A Randomized Controlled Trial

Background and Purpose: The evidence on intervention programs aimed at reducing domestic violence (DV) through court-mandated treatments is generally weak and mixed, at best. Most offenders who are convicted of misdemeanor DV crimes in the US are mandated to attend one of the nation’s 2500 Batterer Intervention Programs (BIP). BIPs bring groups of DV offenders together in treatment to hold them accountable for their crimes. Recent research raises serious questions about the effectiveness of BIPs due to high attrition rates, little evidence of attitudinal and behavioral change, and inconsistent contact with victims. A number of states are now offering other treatment approaches for DV crimes; some states, including Utah, require these alternative programs to be offered after a period of BIP treatment. One alternative, Circles of Peace (CP), a restorative justice treatment approach, was developed as a response to an emerging interest in alternative treatments that included victim participation (when the victim chooses to participate). The CP model, the first of its kind in the US to use restorative justice principles to treat those arrested for DV crimes, is flexible, culturally sensitive, and works with the criminal justice system to bring about healing and transformation. Experiments on the efficacy of alternative approaches such as CP are scarce. This study tests a hybrid restorative justice program (BIP followed by CP) in Salt Lake City, Utah with offenders and their victims (if the victim chose to participate), with a focus on both recidivism as well as harm reduction measures. 

Methods: Pretest-posttest, randomized controlled field trial to assess the effect of a traditional Batterer Intervention Program (BIP-only) mandated by the court for misdemeanor DV crimes (18 weeks) against a court-mandated BIP (12 weeks) plus a restorative justice program (CP) (6 weeks) (BIP plus CP), using a 24-month follow up period on arrest counts as well as severity scores (N=222). Poisson regression models were used to estimate the treatment effect, along with Cohen’s d as a measure of the magnitude of the intervention, compared to control conditions.

Results:  The BIP plus CP intervention resulted in a 53% and 52% reduction in chance of new arrests and severity, respectively [(Exp(b)=.470, 95% CI .312, .707); (Exp(b)=.480, 95% CI .385, .600)], within 24 months post-random assignment. For both measures, the magnitude of the difference ranges between a medium effect size for the count-based model (d=-.493, 95% CI -.763, -0.244), to a large effect size for the harm-based model (d=-.859, 95% CI -1.137, -.582).

Conclusions and Implications: A court-mandated hybrid restorative justice intervention (BIP plus CP) is more effective than a BIP-only approach and can better address the range of DV cases mandated to treatment. Issues with implementation associated with lack of victim participation in treatment with their offender mask a potentially more efficient overall intervention, which future research needs to address more robustly. At the very least, the experiment shows that restorative justice is a viable treatment in what historically has been considered an inappropriate forum for a restorative justice approach.