[Skip to Content]
[Skip to Content Landing]
Letters to the Editor
January 1998

Mental Disorder and Crime

Author Affiliations
  • 1Academic Department of Psychological Medicine, King's College Medical School and the Institute of Psychiatry, 103 Denmark Hill, Camberwell, London SE5 8AF, England
  • 2Perth, Australia
Arch Gen Psychiatry. 1998;55(1):86-87. doi:10.1001/archpsyc.55.1.86a

We read with interest the recent theme issue of Archives of General Psychiatry (June 1996) on the ever-controversial subject of the relationship between crime, violence, and psychosis. Hodgins et al1 found an association between psychiatric hospitalization and criminal convictions. Hodgins and colleagues are to be congratulated on their remarkable sample size. However, as both the authors and the accompanying editorial admit, none of the studies contained in the theme issue have avoided the problem of selection bias. Not all patients with psychiatric disorders are admitted to psychiatric hospitals, even those with a diagnosis of schizophrenia.2 One of the chief reasons for admission to hospital is actual or potential violence. In a British sample restricted to patients with first episodes of schizophrenia, 94 (37%) had been violent in the previous month, and 22% had had police contact.3 As one might expect, we were able to show that violent behavior was significantly associated with the decision to admit.2