Cannabis use has demonstrated an association with earlier onset of psychosis. Investigation of sociodemographic and clinical characteristics in association with cannabis use in adults with first-episode psychosis (FEP) has resulted in inconsistent findings. The clinical profile of cannabis users amongst adolescents with FEP remain widely understudied.

To investigate the frequency of cannabis use, and its association with sociodemographic and clinical characteristics in adolescent-versus adult-onset FEP.

Naturalistic cross-sectional study of 1363 FEP cases aged 14-35 (136 with adolescent-onset psychosis) referred to Early Intervention Services for psychosis in London (UK) (2003-2009). Sociodemographic and clinical data (age of psychosis onset, symptom domains, substance misuse, insight, violence, global functioning, and duration of untreated psychosis [DUP]) were collected at entry to EIS.

Cannabis users were more likely to be male (78.2%), White (43.0%) and unemployed (72.0%). No significant difference was found in cannabis abuse/dependence frequencies between adolescents (28.4%) and adults (24.7%). Cannabis abuse/dependence was associated with an earlier onset of psychosis by 2 years (p<0.001), greater manic and positive symptoms (p<0.001), increased violence (p=0.011), and poorer functioning (p=0.013) and insight (p=0.003). For adolescents, cannabis abuse/dependence was associated with greater positive symptoms, poorer functioning and longer DUP.

Cannabis use shapes the clinical presentation of FEP individuals. Similar frequencies of cannabis use between age groups suggest that substance misuse services should be provided to all, aiming to reducing consumption. Greater vigilance amongst clinicians would enable earlier detection of psychosis in substance misusing adolescents, to reduce DUP and minimise associated poor outcomes.