New data indicate that people with psychiatric illness are almost half as likely to commit a violent crime when taking antipsychotic medication (such as clozapine or risperidone) compared with when they are not using such medication. The reduction in violent crime was smaller with use of mood stabilising drugs (such as lithium or carbamazepine) and only apparent among male patients with bipolar disorder.
Schizophrenia, bipolar disorder, and related psychiatric disorders affect up to 2% of the general population. Antipsychotic and mood stabilising medications are commonly used in the treatment of these conditions and are known to reduce the risk of relapse and re-hospitalisation for psychiatric illness. However, there are few data relating to the effect of treatment on real-world adverse outcomes, such as reducing violent behaviour.
A study, led by Dr Seena Fazel of Oxford University, reviewed Swedish national health registry records of over 80,000 patients prescribed antipsychotic or mood stabilising medication between 2006 and 2009 for psychiatric diagnoses and any subsequent criminal convictions. The team found that 6.5% (2657) of the men, and 1.4% (604) of the women had been convicted of a violent crime. When comparing rates for times when prescribed medication was being taken with times when it was not, violent crime fell by 45% in patients receiving antipsychotics and 24% in patients taking mood stabilisers. Mood stabilisers, however, were only found to reduce violent crime in male patients with bipolar disorder. Although antipsychotics and mood stabilisers are often prescribed in combination, the study showed no evidence of this further reducing violent crime.
Dr Fazel explained “Patients with psychiatric disorders are at risk of perpetrating violent acts, as well as being victims... our study provides evidence of potentially substantial reductions in risk of violence, and suggests that violence is to a large extent preventable in patients with psychiatric disorders.”
Professor Sheilagh Hodgins of Université de Montréal, Canada, and Karolinska Institutet, Stockholm, Sweden, commented “The study illustrates again that de-identified data from national registers that were established for administrative reasons can be used by epidemiologists to identify potential strategies to reduce health-related social problems.”