The pilot study by Ditton-Phare et al, ^{Reference Ditton-Phare, Sandhu, Kelly, Kissane and Loughland1} which was not published at the time of writing our paper, is a most welcome addition to the field. Their focus on how to communicate about diagnosis and prognosis is particularly helpful, given that there can be a reluctance to disclose a diagnosis of psychosis for fear of causing harm. ^{Reference Moran, Oz and Karnieli-Miller2} As Ditton-Phare et al describe, they evaluated trainees' skills ‘in vitro’ using role plays with actors. While assessing trainees' skills interacting with actors (or simulated patients) is a useful tool in training, there are ‘qualities of the psychiatrist–patient encounter that may be resistant to simulation’. ^{Reference Brenner3} Our study ^{Reference McCabe, John, Dooley, Healey, Cushing and Kingdon4} differed in testing an intervention in the natural clinic setting with patients. Hence, as far as we are aware, this is the first study to enhance communication with patients with psychosis.