Citation

Fazel S, Grann M. Am. J. Psychiatry 2006; 163(8): 1397-1403.

Affiliation

Department of Psychiatry, University of Oxford, Warneford Hospital, Oxford OX3 7JX, UK. seena.fazel@psych.ox.ac.uk

Comment In:

Am J Psychiatry 2006;163(12):2193; author reply 2194

Copyright

(Copyright © 2006, American Psychiatric Association)

DOI

10.1176/appi.ajp.163.8.1397

PMID

16877653

Abstract

OBJECTIVE: This study aimed to determine the population impact of patients with severe mental illness on violent crime. METHOD: Sweden possesses high-quality national registers for all hospital admissions and criminal convictions. All individuals discharged from the hospital with ICD diagnoses of schizophrenia and other psychoses (N=98,082) were linked to the crime register to determine the population-attributable risk of patients with severe mental illness to violent crime. The attributable risk was calculated by gender, three age bands (15-24, 25-39, and 40 years and over), and offense type. RESULTS: Over a 13-year period, there were 45 violent crimes committed per 1,000 inhabitants. Of these, 2.4 were attributable to patients with severe mental illness. This corresponds to a population-attributable risk fraction of 5.2%. This attributable risk fraction was higher in women than men across all age bands. In women ages 25-39, it was 14.0%, and in women over 40, it was 19.0%. The attributable risk fractions were lowest in those ages 15-24 (2.3% for male patients and 2.9% for female patients). CONCLUSIONS: The population impact of patients with severe mental illness on violent crime, estimated by calculating the population-attributable risk, varies by gender and age. Overall, the population-attributable risk fraction of patients was 5%, suggesting that patients with severe mental illness commit one in 20 violent crimes.

Language: en