Predicting violent reconvictions using the HCR-20

Abstract

Risk assessment of future violent acts is of great importance for both public protection and care planning. Structured clinical assessments offer a method by which accurate assessments could be achieved. To test the efficacy of the Historical, Clinical and Risk Management Scales (HCR-20) structured risk assessment scheme on a large sample of male forensic psychiatric patients discharged from medium secure units in the UK. In a pseudo-prospective study, 887 male patients were followed for at least 2 years. The HCR-20 was completed using only pre-discharge information, and violent and other offending behaviour post-discharge was obtained from official records. The HCR-20 total score was a good predictor of both violent and other offences following discharge. The historical and risk sub-scales were both able to predict offences, but the clinical sub-scale did not produce significant predictions. The predictive efficacy was highest for short periods (under 1 year) and showed a modest fall in efficacy over longer periods (5 years). The results provide a strong evidence base that the HCR-20 is a good predictor of both violent and non-violent offending following release from medium secure units for male forensic psychiatric patients in the UK.