Evaluation of the HCR-20 Violence Risk Assessment Scheme in a Belgian forensic population

Abstract

Violence risk assessment constitutes a major concern in forensic psychiatry, psychology and related fields. Numerous instruments like the Hare Psychopathy Checklist (PCL-R, Hare, 1991) and the Violence Risk Appraisal Guide (VRAG) have been developed in the aim of improving precision in the prediction of violence. This study assesses the reliability index of the HCR-20 Violence Risk Assessment Scheme and evaluates its relationship with the PCL-R the Buss and Perry's self-report Aggression Questionnaire (AQ) and the type of offenses officially registered. All participants (n=86) are male adult offenders detained in a Belgian high-security forensic hospital. The mean IQ and age were respectively, 81,03 and 36,71. Items common to these instruments were omitted in order to avoid a circular effect. Results showed that the HCR-20 was significantly related to the PCL-R: the historical factor was strongly correlated to the PCL-R factor 2 while the clinical factor was strongly correlated to the PCL-R factor 1. The HCR-20 was significantly related to the AQ: the historical factor was strongly correlated to the AQ Physical Aggression factor and to the Anger factor. As regards to the HCR-20 relations with offenses, its Historical factor was related to both violent (assault and battery, robbery) and non-violent offenses (drug offenses, theft). Finally, the HCR-20 was also positively related to "instrumental" and rather premeditated homicide but negatively related to psychotic homicide. The overall findings are congruent with -published data on the convergent validity of the HCR-20.