I would like to comment on the editorial by De Leo (Reference De Leo2002) which came to the conclusion that little is new in suicide prevention. Since nothing was mentioned about pharmacotherapeutic advances in suicide, I would like to take the opportunity to discuss recent information concerning the role of novel antipsychotics in the reduction of suicidality.

Suicide rates in schizophrenia are about 13 times greater than in the general population, and make a substantial contribution to the overall suicide statistics in the UK. Suicide rates in schizophrenia were unaffected by the advent of conventional neuroleptics. This was not because these drugs are ineffective, rather that they also come with adverse events that put patients at risk for suicide — most particularly akathisia and depression. However, there is now evidence that atypical antipsychotics — most particularly clozapine — may have antisuicidal potential. This was first hinted at by a mirror-image study by Meltzer & Okayli (Reference Meltzer and Okayli1995), which suggested an 86% reduction in suicidality. Subsequently, a large epidemiological study (Reference Walker, Lanaza and ArellanoWalker et al, 1997) including data on completed suicides showed that deaths from suicide in clozapine users occurred at a rate of 39 per 100 000 patient-years compared with 222 per 100 000 patient-years in former users of clozapine. Our own UK clozapine study (Reference Munro, O'Sullivan and AndrewsMunro et al, 1999) confirmed this result. There are also suggestions from pivotal studies of olanzapine that suicidality is also reduced in users of this drug (Reference Tran, Hamilton and KuntzTran et al, 1997).

All these observations have their limitations, which led Novartis, in collaboration with the US Food and Drug Administration (FDA), to embark on a randomised controlled trial of clozapine v. olanzapine in the reduction of suicidality in schizophrenia (the InterSePT study), the results of which have recently been reported (Reference Meltzer, Alphs and GreenMeltzer et al, 2003). Overall there was a 25% reduction in all key measures for suicidality in favour of clozapine. This has recently led the Psychopharmacology Advisory Committee to the FDA to recommend that this body approves suicidality in schizophrenia (not restricted to treatment resistance) as a new indication for clozapine. It is disappointing that the National Suicide Prevention Strategy for England and Wales has little to say about the role of new treatments in suicide prevention. However, in a recent modelling study of ours (Reference Warner, Knapp and DugganWarner et al, 2003), which also took into account drop-out rates and treatment failure rates, we calculated that one-quarter of the target for suicide reduction in all patients in contact with mental health services could be achieved by the broader use of clozapine in treatment resistance. If clozapine were to be approved for suicidality, 50% of all patients with schizophrenia would be technically eligible. Again, calculating in drop-outs and failures an even more substantial proportion of the national target could be met. Much is made of the rates of thromboembolism and agranulocytosis with this drug. However, in comparison with overall reduction in all-cause mortality as well as the reduction in suicidality with treatment with clozapine, such caution is not supported by the epidemiological evidence for the overall advantage of this drug (Reference Walker, Lanaza and ArellanoWalker et al, 1997).