Aim – To test the hypothesis that stressful life events, in particular intrusive events, would cluster before the first onset of psychosis. Method – Forty-one patients with first episode psychosis were interviewed with the Life Events and Difficulties Schedule to assess events in the year before the onset of illness. Events were rated contextually on timing, independence, threat, and intrusiveness. Within-patient group clustering was evaluated, and the patient group was compared with two general population surveys from London. Results – Events were very frequent in the year before the onset of psychosis. Stressful and intrusive events were more common in the three months before onset compared with the previous nine months, although few analyses reached conventional statistical significance. Threatening and intrusive events were much more frequent in patients than in the comparison groups. First episode psychosis patients were strikingly more likely to report an intrusive event in the three months before onset (OR=17.1, 34% v 3%), and also showed an excess between three months and one year before onset (OR=8.1, 11% v 3%). Conclusions – Events may trigger many cases of first episode psychosis. The effect emerges several months before onset, particularly if the events are intrusive. This has both theoretical and clinical implications.

Declaration of Interest: DR was supported by a UK Medical Research Council studentship. The authors have no conflicts of interest in writing this paper.