Age of onset of Cannabis use and cognitive function in first-episode non-affective psychosis patients: Outcome at three-year follow-up. Abstract BACKGROUND: In recent years, the effects of cannabis use on cognitive functions in patients with psychosis have been widely studied. Recently, special emphasis has been placed on the impact of age at the onset of consumption on cognition in these patients. METHOD: 349 patients with a first episode of non-affective psychosis were studied. Patients were classified as cannabis users and non-users. Users were divided, according to their age when they began using cannabis, into: early-onset (age < 16) and late-onset (age ≥ 16) users. Differences between groups at baseline were studied based on sociodemographic, clinical and cognitive variables. The groups were longitudinally (3-year) compared on cognitive variables. RESULTS: Out of the 349 patients included in this study, 38.7% (N = 135) were cannabis users. Of them, 39.3% (N = 53) were early-onset and 60.7% (N = 82) were late-onset cannabis users. No baseline differences were found between the early-onset and late-onset groups on cognitive domains. Longitudinally, only patients who had withdrawn from cannabis use during follow-up showed a significant improvement in verbal memory. CONCLUSION: Our results did not show differences between the early-onset group and the other two groups in long-term cognitive performance, even if they kept consuming cannabis during the first three years of disease progression. Further studies are needed to elucidate the true relationship between early-onset cannabis use and cognitive function in patients with a first episode of psychosis.