We thank Professor Mortimer for raising some important issues. The first is about the absence of a control group. Our study was observational, designed to determine clinical practice in antipsychotic prescribing prior to clozapine relative to treatment guidelines, which is why there is no control group. Prescribing in patients who have treatment-refractory illnesses but have never received clozapine is an important issue but outside the scope of our study. The second issue was that the paper implies that any patient for whom a third antipsychotic is considered ought to be considered for clozapine. Our study specifically selected patients who went on to be prescribed clozapine, predominantly because the illness was treatment refractory (see p. 481). ^{Reference Howes, Vergunst, Gee, McGuire, Kapur and Taylor1} As such, it is relevant to patients with treatment-refractory illness, and in no way should be taken to imply that all patients who are being considered for a third antipsychotic should automatically be considered for clozapine. It is important to note that our design could not confirm the illness was treatment refractory at the point at which two antipsychotics had been used and, although treatment resistance can be evident from the first episode, ^{Reference Agid, Remington, Kapur, Arenovich and Zipursky2} it may potentially emerge later in some patients (this is discussed further in the paper, p. 483 ^{Reference Howes, Vergunst, Gee, McGuire, Kapur and Taylor1} ). For this reason, the delay is described as the maximum theoretical delay (see p. 482). ^{Reference Howes, Vergunst, Gee, McGuire, Kapur and Taylor1}

We agree with Professor Mortimer that it is currently not possible to predict which patients will have treatment-refractory illness or will respond to clozapine, although the development of pathophysiologically specific markers may enable this. ^{Reference Howes, Bose, Turkheimer, Valli, Egerton and Stahl3–Reference Demjaha, Murray, Kapur and Howes6} Likewise, we agree with Professor Mortimer that clozapine is not the only option for treatment resistance, although it has the best evidence base and this is reflected in its pre-eminence in treatment guidelines. However, we found that over a third of patients had received antipsychotic polypharmacy, and a third had received high-dose antipsychotic treatment prior to clozapine treatment. ^{Reference Howes, Vergunst, Gee, McGuire, Kapur and Taylor1} This indicates that poorly evidenced strategies are commonly used in preference to clozapine and, to echo Professor Mortimer, this is a disservice to our patients.