Type of Discharge and Risk of Recidivism Measured by the HCR-20: A Retrospective Study in a Dutch Sample of Treated Forensic Psychiatric Patients

Abstract

This retrospective study examined the predictive validity of the HCR-20, a violence risk assessment instrument. The HCR-20 as well as the Psychopathy Checklist-Revised (PCL-R) were coded on the basis of file information of 120 patients discharged from a Dutch forensic psychiatric hospital between 1993 and 1999 (average follow-up period 72.5 months). The patients were divided into four groups according to type of discharge: 1) discharge by the court in line with the hospital staff's advice and after a transmural phase; 2) discharge by the court in line with the hospital staff 's advice, but without a preceding transmural phase; 3) discharge by the court against the hospital staff's advice; and 4) readmission to another institution. Recidivism data (reconvictions) from the Ministry of Justice were related to the risk assessments. The base rate for violent recidivism was 36%, and 52% for general recidivism. The HCR-20 and PCL-R total scores demonstrated good predictive validity for violent recidivism (AUC = .82 and .75, respectively). The HCR-20 was a significantly better predictor of violent recidivism than unstructured clinical judgment stated in hospital staff 's advice to the court. In addition, the HCR-20 total score predicted significantly better than the PCL-R total score, although the difference in AUC values was no longer significant when the item ‘Psychopathy’ was removed from the HCR-20 total score.