Rates of violent crimes decrease during periods with antipsychotics and mood stabilisers compared to periods without

ABSTRACT FROM: Fazel S, Zetterqvist J, Larsson H, et al. Antipsychotics, mood stabilisers, and risk of violent crime. Lancet 2014;384:1206–14.

What is already known on this topic

The benefits of antipsychotics and mood stabilisers in the treatment of not only schizophrenia¹ and bipolar disorder,² but also major depression³ and borderline personality disorder,⁴ have been established with regard to relapse prevention and readmission rates. Knowledge about other important outcomes of pharmacotherapy is incomplete, specifically when it comes to violent behaviour which has emerged as a recurrent problem linked to periods with severe mental illness.⁵

Methods of study

In this population-based study (2006–2009), linked Swedish national registers were used to identify 82 647 patients (40 937 men and 41 710 women) who were prescribed antipsychotics or mood stabilisers, their psychiatric diagnoses and subsequent convictions for violent crime (2657 men (6.5%) and 604 women (1.4%)). Each patient was used as their own control by doing within-individual analyses comparing the frequency of convictions for violent crimes (according to Sweden's national crime register) during periods when they were prescribed these medications versus the frequency when they were not receiving these medications. This method allowed the researchers to control for all confounders that remained constant within each individual during the study period.

What does this paper add

• This is the second population-based study by these authors⁶ that compares within-individual periods for psychiatric patients on psychotropic medication with periods without medication.

• The findings enhance the evidence base for antipsychotics and mood stabilisers as active beyond what concerns relapse prevention into new episodes of illness and symptom relief, since periods on medication were related to significant reductions in the rate of violent crime. Compared with periods when participants were not on medication, violent crime fell by 45% in patients receiving antipsychotics (HR 0.55, 95% CI 0.47 to 0.64) and by 24% in patients prescribed mood stabilisers (0.76, 0.62 to 0.93).

• There were differences between drug classes, with a larger reduction in violence for antipsychotics, while mood stabilisers only reduced the rate of violent crime in patients with bipolar disorder.

• Rate of violence reduction for antipsychotics was stronger in patients who were prescribed higher doses.

• Reductions in violent crime were also recorded for depot medication (HR adjusted for concomitant oral medications 0.60, 95% CI 0.39 to 0.92).

Limitations

• How antipsychotics and mood stabilisers affect violent crime or whether the reduction was mediated by some other effects was not covered by this study, and thus it cannot be used to assess causal effects.

• This study was not able to investigate disease phase (ie, prodromal, first episode, acute or chronic) due to the nature of the data set, and thus to control for to what extent disease phase was associated with outcome. So no conclusions concerning the effect of medication on disease phase and how it was related to violence could be drawn, while there is evidence pointing to that first episode of psychosis is especially crucial with regard to violence.⁵

• Violent convictions were used as the primary outcome, and it is a well-known fact that convictions only capture a limited part of all violent acts (violence towards family members and healthcare professionals, eg, is often undetected).

What next in research

Besides replication studies, using official records and register-based data from different countries to establish whether this treatment effect is a similar pattern cross-culturally, a further step would be to carry out prospective clinical follow-up studies of samples with severe mental disorders and known criminality or aggressive behaviours from for example forensic psychiatric settings. Such studies would make it possible to control not only for periods with medication, but also for whether patients during the study period spent time in other settings (eg, correctional) or were subjected to risk management interventions. In addition to register-based outcome measures, self-rated and family-reported information could be included to further establish whether patients during periods of medical treatment are less prone to commit violent acts.

Do these results change you practices and why?

The answer to this question is yes, especially with regard to forensic psychiatric settings where patients have showed a propensity to act with irritability and violence. This study confirms the stabilising effects of antipsychotics for severe mental disorders and mood stabilisers for bipolar disorders, not least with regard to concurrent symptoms of irritability and aggression. These results stress the importance of establishing and maintaining a therapeutic alliance with the patient to increase adherence to treatment, since the benefits of efficacious medication is not limited to clinical symptoms, but also to a more adaptive functionality with reduced propensity to react with aggression and violence.