A meta‐analytic review of the efficacy of psychological treatments for violent offenders in correctional and forensic mental health settings.

Abstract

This meta‐analysis examined whether psychological treatments with adult violent offenders in correctional and forensic mental health settings are effective in preventing community recidivism and institutional (hospital/prison) misconduct. A total of 27 controlled studies containing 7,062 violent offenders were obtained via a comprehensive search strategy that yielded more than 13,000 records. Overall, treatments with violent offenders significantly reduced violent and general/nonviolent recidivism. The average effect for violent and general/nonviolent institutional misconduct did not attain statistical significance. Moderator analyses indicated numerous trends; however, most effects were nonsignificant following alpha‐level corrections. Findings regarding the impact of psychological treatments are promising and suggest that multimodal treatments are associated with the strongest treatment effects. However, the extant evidence base is limited by a small number of well‐controlled outcome studies and inconsistent/incomplete reporting of the evaluations. More high‐quality research is needed to investigate the effectiveness of violent offender treatment on outcomes and mechanisms of action, and to determine which treatment components are effective, in what combination, and for which offenders.

The next steps forward in determining “what works”.

“…In that analysis, a random‐effects model of violent recidivism produced a Cohen's d of −0.16, which corresponds to a very small to small effect size. Papalia et al () report an effect size of −0.21 when converted to d , which is 0.01 points over the small effect size threshold. Similarly, general recidivism changed from an effect size of −0.21 in 2007 to −0.23 presently, both of which are slightly above the small threshold.…”

“…Papalia et al () reported significant effects of treatment within community‐based settings, but found no significant changes for institutional misconduct. These findings highlight the common clinical wisdom that context matters.…”

“…Similarly, when treatment groups within institutions can only utilize a subset of important group processes, treatment effects are likely to suffer. Papalia et al’s () study statistically captures this contextual disparity, and clinicians may benefit from this reminder moving forward.…”

“…This shift has the potential for a particularly large impact with individuals responsible for repeated, serious crime, particularly violent offenses. Papalia et al (2019) provide much-needed data on whether the interventions delivered to these offenders are, in fact, reducing either the violent behaviors being targeted or reoffending more generally. However, before this question can be fully addressed, research in this domain must more carefully define who a violent offender is and, extending from that, who these interventions may help.…”

“…However, before this question can be fully addressed, research in this domain must more carefully define who a violent offender is and, extending from that, who these interventions may help. Papalia et al (2019) define violence broadly, including actual, attempted, or threatened harm identified in official records or through self-or informant-report. Unfortunately, the relatively small research base on the efficacy of treatment for violent offenders cannot support a meta-analysis with a narrower definition at this time.…”