Risk Assessment in Forensic Patients with Schizophrenia: The Predictive Validity of Actuarial Scales and Symptom Severity for Offending and Violence over 8 – 10 Years

Abstract

Assessment of risk of violence is essential in the management of patients with schizophrenia admitted to secure hospitals. The present study was conducted to test the validity of actuarial measures and psychotic symptoms in the prediction of further violence and offending in this group. The H-10 scale of the HCR-20, Violence Risk Appraisal Guide and Psychopathy Checklist-Revised were scored retrospectively. Symptom severity was rated at interview and persistence from notes. Outcome was measured using criminal records and recorded incidents of aggression over an 8-10 year period. Seventy-six percent of patients were involved in more than 1800 incidents defined as physical contact with a victim or damage to property, and 28% in a serious incident defined as injury to a victim requiring hospital treatment, a contact sexual incident or fire setting. Fifteen percent of patients were convicted of any offense and 5% of a violent offense. The risk scales had moderate to high predictive accuracy for offenses and violent offenses but failed to predict incidents or serious incidents. Symptom severity and persistence predicted incidents but not offenses. Violence within this population is common. Actuarial measures of risk assessment are valid predictors of offending and violent offending but psychotic symptoms are more relevant to the prediction of violent incidents. Assessments of likely inpatient aggression must emphasize symptoms.

The authors wish to thank the patients and staff of the State Hospital and many other hospitals for their participation in this study, including the medical records officers for their assistance. In addition, we would like to thank the Scottish Prison Service and the Scottish Criminal Records Office for their help.

The authors wish to thank the patients and staff of the State Hospital and many other hospitals for their participation in this study, including the medical records officers for their assistance. In addition, we would like to thank the Scottish Prison Service and the Scottish Criminal Records Office for their help.