Theoretically Informed Meta-Analysis of the Risk for General and Violent Recidivism for Mentally Disordered Offenders

• NCJ Number: 247399
• Journal: Aggression and Violent Behavior Volume: 19 Issue: 3 Dated: May/June 2014 Pages: 278-287
• Author(s): James Bonta; Julia Blais; Holly A. Wilson
• Date Published: June 2014
• Length: 10 pages
• Publication Format: Article
• Publication Type: Research (Applied/Empirical)
• Language: English
• Country: United States of America

Annotation

This article presents the results of a meta-analysis of studies that used the risk/need domains from the General Personality and Cognitive Social Learning perspective to predict general and violent recidivism for mentally disordered offenders.

Abstract

Mentally disordered offenders (MDOs) pose a significant problem for law enforcement and corrections personnel. Incarcerated MDOs create problems for corrections officials who are tasked with ensuring the safety of all prisoners and staff members. This review examined studies that used the risk/need domains from the General Personality and Cognitive Social Learning (GPCSL) perspective to predict general and violent recidivism in MDOs. The review identified 126 studies that examined 96 unique samples of MDOs. The review looked at whether the eight risk/need domains from the GPCSL perspective were better predictors of recidivism among MDOs than were clinical variables used for predicting violent recidivism. The eight domains from the GPCSL perspective are criminal history, antisocial personality pattern, procriminal attitudes and cognition, procriminal companions, family/marital, education/employment, substance abuse, and leisure/recreation. The clinical variables include psychosis, schizophrenia, mood disorder, intelligence, prior admissions, psychiatric treatment history, personality disorders, and antisocial personality or psychopathy. The review found that the eight domains were better predictors of general and violent recidivism among MDOs than were the clinical variables, and that four of the eight factors were better at predicting violent recidivism compared to general recidivism. Among the clinical variables, only three were identified as being effective at predicting recidivism: intelligence, antisocial personality, and psychopathy. These findings validate the use of the eight domains from the GPCSL perspective as an effective tool for risk assessment with MDOs. Tables and references