I was surprised to read the paper of Howes et al. ^{Reference Howes, Vergunst, Gee, McGuire, Kapur and Taylor1} No control group was investigated. This paper implies that any patient for whom a third antipsychotic is considered ought to be considered for clozapine, and if they are not started on clozapine, this constitutes an unacceptable delay.

Such a sweeping assumption cannot be verified without the inclusion of a group of preferably contemporaneous control patients of approximately the same age, gender, diagnosis and treatment duration, who were treated with a third antipsychotic drug, but who did not go on to receive clozapine. Merely the existence of such a group verifies that the delay is indeed theoretical. How are those who will go on to need clozapine be identified? One cannot predict the future, stating with confidence that all patients on a third antipsychotic will require clozapine eventually. The paper is silent on these consequential issues.

This obvious limitation is not mentioned by the authors in their critique of their methodology. Neither does Patel's editorial commentary remark on it. ^{Reference Patel2}

I emphasise that I agree with the authors, that delay in initiating clozapine does many patients a great disservice. I have been one of the UK's major users of and proponents of clozapine since 1990, and recently published on clinical use of clozapine in Advances in Psychiatric Treatment. ^{Reference Mortimer3} I still, regularly, encounter patients who should have been started on clozapine much longer than 4 years ago. The trouble is, the longer this treatment decision is ignored or put off, the more reluctance there seems to be. Nevertheless, there remain a number of alternative strategies to address those contributors to unsatisfactory response to antipsychotic drugs, such as poor adherence, substance misuse and intolerance, which may not necessarily succumb to clozapine treatment.