Risk assessment in offenders with intellectual disability: a comparison across three levels of security

In mainstream offender samples, several risk assessments have been evaluated for predictive validity. This study extends this work to male offenders with intellectual disabilities. Participants from high-, medium-, and low-security settings, as well as community settings, were compared on a range of risk assessments. The Violence Risk Appraisal Guide, HCR-20—Historical Scale, the Risk Matrix 2000-C (combined risk), and the Emotional Problems Scales—Internalising discriminated between groups, with participants from high security having higher scores than those in medium security, who had higher scores than those in the community. The Violence Risk Appraisal Guide, all HCR-20 scales, the Short Dynamic Risk Scale, and the Emotional Problems Scales (Internalising and Externalising) showed significant areas under the curve for the prediction of violence. The Static-99 showed a significant area under the curve for the prediction of sexual incidents. The discussion reviews the value of these various scales to intellectual disability services.