Over 15 years ago, a 17-year-old male was charged with murder after  a 23-year-old woman, his friend and confidante, was stabbed repeatedly and subsequently sexually assaulted post-mortem. Alcohol and benzodiazapine abuse were acute factors at the time of the incident. Suicidal urges and a dissociative state were also factors. After hearing the evidence of a number of psychiatrists, the prosecutor eventually joined the defence and invited the jury to render a verdict of not guilty by reason of insanity.

The subject’s history revealed well-documented severe neonatal asphyxia, with subsequent temper tantrums, rage attacks, suicidal urges, and learning problems, as well as some substance abuse. Impulsivity was a common behavioural problem. Thorough psychological and electrophysiological assessments suggested organicity but were contradictory regarding behavioural correlations. Consequently, the patient was remanded to a maximum-security facility where he was diagnosed with antisocial and narcissistic personality disorder. Annual reviews by the provincial review board, required by Canadian law, determined that he should remain in maximum security on the recommendation of the hospital. A neurological evaluation ordered by the review board revealed no abnormalities.

Manuscript received and accepted September, 2000.
¹Associate Professor, Faculty of Social Work and Institute for Medical Sciences; Academic Coordinator, Centre for Applied Social Research, University of Toronto, Toronto, Ontario.
²Member, Law Society of Upper Canada.
³Professor and Head, Division of Forensic Psychiatry, Faculty of Medicine, University of Ottawa, Ottawa, Ontario.
Address for correspondence: Dr C Regehr, Centre for Applied Social Research, University of Toronto, 246 Bloor Street West, Toronto, ON  M5A 1S1
e-mail: cheryl.regehr@utoronto.ca

At the review board hearing in 1990, the research department at the hospital indicated that the patient met the criteria for psychopathy as measured by the Psychopathy Checklist (PCL) (1). He had been scored on the instrument via a chart review pursuant to a contract with the Ministry of Health—described as both a research and a program initiative—to score all forensic patients in Ontario. The research resulted in the development of the Statistical Risk Appraisal Guide (SRAG) (2), which was later refined to become the Sex Offender Risk Assessment Guide (SORAG) (3) and the Violence Risk Appraisal Guide (VRAG) (4). These instruments relied on static historical factors for the prediction of a broad range of violent behaviour. Contrary to the instructions in the manual accompanying the PCL-R,  the patient was not interviewed as part of the assessment (5). Further, the patient was not informed that the study was being carried out, nor was he given the opportunity to consent or refuse participation. The hospital staff  relied on the findings of the test to confirm their diagnosis and make their recommendations to the board.

In 1992, the board ordered that the patient be transferred to a medium security facility. Soon after the patient’s arrival at the new facility, the hospital received a SRAG (2) prepared by the research staff of the original hospital. Their research placed the patient in the highest category of risk for a community violent reoffense. It was suggested that among mentally disordered offenders, 100% of individuals in this category would reoffend within 7 years of release.