Skip to main content

Predictive Validity of the HCR-20 and the Start for Assessing Risk of Insanity Acquittees on Conditional Release

23 April 2018


Risk assessment instruments are used to help determine treatment readiness and improvement and to inform decisions regarding placement in inpatient and outpatient civil commitment, parole, probation, and community supervision (Doyle & Logan, 2012; McKeown, 2010; Singh, Grann, & Fazel, 2011). Risk assessment measures are based on empirically validated static and dynamic risk factors (Wilson, Desmarais, Nicholls, Hart, & Brink, 2013). The present study investigated the static Historical scale and dynamic Clinical and Risk Management scales of the HCR-20 and the dynamic Strengths and Vulnerabilities scales of the START. None of the HCR-20 or START scales were significantly correlated with physical aggression, verbal aggression, or revocation of conditional release. However, the dynamic scales of the HCR-20 and START showed stronger correlations with verbal aggression compared to the static Historical scale of the HCR-20. Additionally, higher scores on the Vulnerabilities scale were associated with more incidents of verbal aggression. Finally, the HCR-20 scales had stronger correlations with outcomes for females than for men. The results also bring into question the use of the START with a female population. Assessment instruments used to inform decisions must be valid for the predicted outcomes.


File nameDate UploadedVisibilityFile size
Main File
19 Mar 2020
280 kB