Cannabis Use and Clinical Outcomes in First Episode Psychosis: Results From a 2‐Year Follow‐Up Study

Aim: This study aimed to investigate the association between cannabis use prior to enrollment in a First Episode Psychosis (FEP) program and prognosis, including severity of psychiatric symptoms, type of antipsychotic treatment prescribed, and clinical recovery. Methods: A longitudinal cohort study was conducted on patients aged 18-35 diagnosed with affective and non-affective psychosis who were referred to the FEP Program in Ferrara, Italy, between 2012 and April 2024. Sociodemographic and clinical data were collected at program entry and every six months for up to 24 months. Cannabis use was categorised dichotomously based on reported or charted use within the month prior to program entry. The Health of the Nation Outcome Scale (HoNOS) was used to monitor clinical outcomes (assessed through reductions of the total HoNOS score, employment status, and type of antipsychotic prescription) and recovery (total HoNOS < 8). Bivariate analyses compared sociodemographic characteristics and clinical outcomes between people with cannabis use and people without cannabis use at baseline and at 1- and 2-year follow-ups, considering both baseline cannabis status and current cannabis use status. A Mixed Model for Repeated Measures, was used to compare the average HoNOS scores and the recovery rates between the two groups at several endpoints (baseline, 12 months and 24 months). Results: The total sample included 173 patients, predominantly men (74.5%), diagnosed with non-affective psychosis (83.8%), with a median age at onset of psychotic symptoms of 22.9 years. At admission, individuals reporting cannabis use (48%), compared to those without cannabis use, were predominantly male (87.9% vs. 62.2%, p < 0.001), had a significantly shorter duration of untreated psychosis (32.8 vs. 52.2 weeks, p = 0.05) and were more likely to receive combined oral and depot antipsychotic treatment (16.9% vs. 5.6%, p = 0.05), no difference in baseline HoNOS scores was observed. At 1- and 2-year follow-ups, no significant differences in overall psychopathology were found between groups under either cannabis use condition. However, baseline cannabis exposure was associated with lower rates of student status at both 1-year (11.8% vs. 24.3%) and 2-year (6.3% vs. 23.2%) follow-ups. At the 1-year follow-up, participants with cannabis use were more likely to receive a combination of long-acting injectable (LAI) and oral antipsychotics (21.1% vs. 7.3%, p = 0.05) compared to those without cannabis use. No significant differences were found between the two groups in terms of HoNOS scores or recovery rates at any time point (baseline, 12 months, 24 months). Conclusions: Cannabis use prior to FEP program entry was not associated with significant differences in clinical outcomes or recovery, despite being admitted with a shorter DUP and being predominantly male. Nevertheless, cannabis use was associated with more intensive antipsychotic treatment, both at admission and at 1 year. These findings underscore the importance of early detection of cannabis use and integrated care addressing both psychiatric symptoms and substance use. Future randomised controlled trials should explore whether reducing cannabis use can decrease antipsychotic burden and improve long-term outcomes.