EDINBURGH -- The link between psychosis and violence is considerably stronger in women than men, a review of the available evidence suggests.
"I think there is quite good evidence that psychosis seems to confer a disproportionate risk of violence in women as compared to men. Psychosis is a very important risk factor for violence among women--more so than in men," Dr. Pamela J. Taylor said at the meeting.
She characterized gender differences in psychosis-related violence as an understudied field. Some of the most illuminating research in this area, she added, can be credited to Dr. Seena Fazel of the University of Oxford (England). He and his colleagues conducted a meta-analysis of 20 studies that compared the risk of violence in a total of 18,423 people with schizophrenia and other psychoses vs. 1.7 million nonpsychotic controls in the general population of the 11 countries involved. In all, 10% of the people with psychosis had committed violent crimes, compared with 1.6% of the general population.
Dr. Fazel concluded that men with psychosis were fourfold more likely to commit a violent act than were men in the general population. In contrast, an analysis of six studies totaling 5,002 women with psychosis showed that they were 7.85-fold more likely to commit violent acts than were women in the general population, noted Dr. Taylor, professor of forensic psychiatry at Cardiff (Wales) University.
The association between psychosis and violence was particularly striking with regard to homicide. In the five studies that included data on homicide, the odds that a person with psychosis would commit homicide was 19.5-fold greater than in the general population. Nevertheless, the investigators stressed, only 1 in 300 people with psychosis had killed someone (PLoS Med. 2009;6:el000120).
Comorbid substance abuse substantially increased the risk of violence by people with psychosis. Indeed, the Oxford group found that the odds of violence was roughly fourfold greater in individuals with psychosis and comorbid alcohol or drug abuse than in those with psychosis only. Moreover, the odds of violence in people with psychosis and comorbid substance abuse was no different from that in individuals with substance abuse without psychosis.
In another major study, Dr. Fazel and his colleagues used Sweden's comprehensive national hospital admission and judicial registries to scrutinize all violent crimes committed in the country during 1988-2000. The investigators sought to calculate the proportion of violence attributable to persons with schizophrenia or other severe mental illness.
The population-attributable risk turned out to be 5.2%. In other words, 1 in 20 violent crimes in Sweden was committed by someone with severe mental illness.
Of note, the population-attributable risk was significantly greater in women than men across all age groups. …