The study by Tohen and colleagues addresses a field of clinical practice that has traditionally posed a great deal of therapeutic challenge. ^{Reference Tohen, McDonnell, Case, Kanba, Ha and Fang1} Evidence of potential therapeutic response in initial trials are therefore welcome and the authors are right to call for further research to assess the efficacy of olanzapine, while cautioning in relation to the high non-adherence rates observed with this medication.

The authors also attempt to explore the degree of recovery experienced by individuals within their trial. It is correct that this concept is addressed, even in early trials such as this. By considering concepts such as recovery, clinical trials can provide information that allows clinicians and service users to make truly informed decisions in relation to treatment options. Calls for the inclusion of recovery-oriented outcomes in clinical trials into various disorders have been made. ^{Reference Slade and Hayward2,Reference Bateman3}

However, in this study the authors appear to make the mistake of conflating the concepts of recovery and symptom remission. The concept of recovery is generally recognised as being more than simple remission of symptoms, instead involving a deeper acceptance of disorder and personal adaptation to experience. In this journal, a narrative review by Leamy et al described five main themes of recovery that are representative of this concept; they are the sense of: connectedness, hope, identity, meaning and empowerment. ^{Reference Leamy, Bird, LeBoutillier, Williams and Slade4}

Measures such as the Montgomery–Åsberg Depression Rating Scale (MADRS) are valuable in their sensitive detection of change in the symptoms of depressive disorders but they do not address the core concepts of recovery. ^{Reference Montgomery and Åsberg5} Simple definition of recovery as a sustained period of symptom remission (MADRS ⩾12 for ⩽4 weeks) as in this paper is therefore inadequate.

The development of suitable recovery-oriented outcome measures for inclusion in clinical trials is urgently required to allow us to develop an evidence base that considers all aspects of treatment and allows us to provide service users with the information they require to make informed treatment decisions.