The objective of the current research was to examine the relation between non-steroidal anti-inflammatory drugs (NSAID) use and risk of psychosis. To this end we performed a longitudinal case-control study using prescription data from a Dutch health insurance company. Men aged 25 years or over and women aged 30 years or over were excluded to prevent inclusion of non-incident cases. This resulted in eighty-two cases and 359 randomly selected controls from the same population. The overall relative risk of incident antipsychotic use for NSAID users, adjusted for age and prescription frequency, was 0.80 (95% CI: 0.48-1.33). After stratification for gender the risk of psychosis was significantly lower (59%) in male NSAID users only. The relative risks for male and female subjects were 0.41 (95% CI: 0.17-0.97) and 1.31 (95% CI: 0.65-2.64), respectively. These results suggest that in men NSAIDs may lower the risk of psychosis.