Predicting Institutional Violence in Offenders with Intellectual Disabilities: The Predictive Efficacy of theVRAGand theHCR‐20

Abstract

The HCR-20 and the VRAG have excellent predictive efficacy in offenders with an ID. A structured clinical judgement based on the HCR-20 was especially predictive.

Predictive validity of the HCR‐20 for inpatient aggression: the effect of intellectual disability on accuracy

“…Fitzgerald et al . () found that twice as many people with ID in medium‐secure care engaged in violence over a 6‐month period compared with a control group of patients without ID (80 vs. 40%); similar results were reported by Dickens et al . () who found that 72.1% of the ID sample had engaged in aggression and/or self‐harm compared with 47.2% of patients in the mental health pathway.…”

“…Like Fitzgerald et al . (), the overall predictive validity of the HCR‐20 total score did not significantly differ between the ID and comparison groups. However, there were significant differences in the performance of the C5 subscale and SJ.…”

“…Unlike the sole previous study on this topic that used a control group of mentally disordered offenders without an ID (Fitzgerald et al . ), our analysis controlled for important potential confounding variables including gender, age, co‐morbid mental illness, security level and length of stay; as a result our findings are likely to be the most robust thus far. Like Fitzgerald et al .…”

“…Fitzgerald et al . () found that the SJ and historical scale score were significantly better predictors of inpatient physical aggression in people with an ID compared with a control group of mentally disordered offenders without an ID. With the exception of the clinical scale, the remaining AUC values for the prediction of inpatient physical aggression and severe physical aggression were also larger in the ID group, although not significantly so.…”

“…Fitzgerald et al . () found that the overall summary judgment (SJ) was a significant predictor of inpatient physical aggression in an ID sample, producing very large effect sizes. However, none of the HCR‐20 total or subscale scores were significantly predictive of physical aggression, and only the HCR‐20 total and risk management subscale were significant predictors of severe physical aggression.…”