Delay in the initiation of treatment in individuals with first-episode psychosis has been associated with poorer long-term outcomes. ^{Reference Barnes, Leeson, Mutsatsa, Watt, Hutton and Joyce1} El-Adl et al report on general practitioner (GP) experiences of patients with a first psychotic episode. ^{Reference El-Adl, Burke and Little2} However, I have a number of concerns about the reported results.

The low reported incidence of new cases per year within the authors’ locality (n = 100) was demonstrated by the majority (68%) of GPs seeing only one or two such individuals per year. I find it difficult to see, given these low cell counts, how GPs could answer questions about initiating treatment (10%, 25%, 50% and 75% of the time) and thus conclude that GPs are unlikely to start treatment before referring to secondary care services.

The information requested from the GPs regarding engagement of patients with first-episode psychosis and causes of delayed referral are based on these low patient numbers and would be subject to recall bias on behalf of the GP. Getting the patients’ views on barriers to mental health services would certainly have helped triangulate the data.

I was also concerned that the data published were 5 years old and as such the current generalisability of these results could be questioned.

With the National Institute for Health and Clinical Excellence schizophrenia guidelines recently updated ³ and early intervention/crisis resolution teams the norm rather than exception, El-Adl et al echo the view that active engagement with our primary care colleagues is paramount in ensuring these patients receive both a responsive and effective service.