Author reply to Letter to the Editor regarding ‘Homicide associated with psychotic illness: What global temporal trends tell us about the association between mental illness and violence’

We thank Bastiampillai and colleagues for their letter commenting on our paper (Simpson et al., 2021) on homicide associated with psychosis. We agree with their view that more widespread clozapine use is likely to be associated with lower rates of violence for people with psychosis. There is certainly evidence that pharmacology can reduce aggression, and clozapine has a particular anti-aggressive effect, but as yet we don’t know by what mechanism. That said, there are valid reasons why services have struggled to used it more widely – side effects and tolerability especially, but also the need for closer clinical follow-up to ensure compliance and regular blood tests.

Investment has been made in trying to find the next generation clozapine with fewer side effects, but it has so far not yielded results. Until new agents as effective but more tolerable than clozapine are found, clozapine should not be overlooked as part of the overall strategy to reduce violence for people with psychosis on the basis of current evidence. As Bastiampillai et al. note, there is building evidence that adherence to effective psychopharmacology regimes reduces violence, not only with clozapine. So we should not wait for new medications, but use the existing ones better, in association with comprehensive recovery-oriented care.

Homicide associated with psychosis may represent particular manifestation of symptom-driven behaviour. We know it is proportionately most common in the first episode of a psychotic illness (Nielssen et al., 2011) and strongly associated with illness factors in terms of when it occurs in the life course (Simpson et al., 2015). We would expect that the rate of psychosis-associated homicide to therefore be reduced by early intervention, and comprehensive stable service structures that include effective medication algorithms, including a clear path to clozapine for those who are treatment resistant. As homicide is a rare complication of acute psychosis, we are not arguing that these tragic events are in the main a problem of a failure of risk assessment. Rather, achieving effective treatment that reduces the prevalence of acute psychosis will pay many dividends in terms of a reduction of all violence, including serious violence. We need to continue to look for that evidence.

We would expect homicides by people with psychosis to be reduced by having services that are comprehensive, reduce the duration of untreated psychosis and provide treatment with clozapine to those with treatment resistance. Violence more generally associated with psychosis is a heterogeneous problem with overlapping issues of substance use, and socio-economic disadvantage, which is likely only to be addressed with multilayered interventions.

Yours sincerely,

Alexander Simpson, Stephanie Penney and Roland Jones.