OBJECTIVE: Several case reports and a recent study on coeliac disease (CD) and family history of schizophrenia indicate a link between CD and schizophrenia. The objective of our study was to determine the risk of non-affective psychosis in patients with CD in a national general population cohort. MATERIAL AND METHODS: We identified 14,003 individuals with a diagnosis of CD in the Swedish national inpatient register between 1973 and 2003. From the population register, Statistics Sweden then identified five reference individuals matched for age and calendar year at diagnosis, gender and county (n=68,125). Only individuals with more than one year of follow-up after the CD diagnosis was first recorded or a corresponding date in reference individuals were included in the analyses. The risk of subsequent non-affective psychosis in individuals with CD was estimated by Cox regression. RESULTS: CD was associated with a statistically significant increased risk of any non-affective psychosis (hazard ratio (HR)=1.55; 95% CI=1.16-2.06; p=0.003) (65 positive events in 14,003 individuals with CD and 216 positive events in 68,125 individuals without CD); this increased risk was largely due to the association with non-schizophrenic non-affective psychosis (HR=1.61; 95% CI=1.19-2.20; p=0.002: 56 positive events in individuals with CD and 180 among reference individuals). There was no statistically significant association with subsequent schizophrenia (HR=1.43; 95%=0.77-2.67; p=0.261: 14 positive events in individuals with CD and 50 among reference individuals). CONCLUSIONS: Individuals with CD may be at increased risk of non-affective psychosis.