This PRISMA-compliant systematic review and meta-analysis examined the prevalence, correlates, and clinical impact of cannabis use (CU) and cannabis use disorder (CUD) in early-onset psychosis (EOP; onset <18 years), searching six databases through July 2025 across 40 studies (N=3,473; mean age 16.2±1.6 years; 59% male).
Pooled prevalence in EOP was 32.8% for current CU (95% CI 18.5–51.2%), 40.2% for lifetime CU (95% CI 31.3–49.7%), and 36.6% for lifetime CUD (95% CI 18.2–59.9%). Current CU was associated with male sex, higher hospitalization rates, less severe negative symptoms, and lower odds of a schizophrenia diagnosis; CU commonly preceded psychosis onset and was linked to longer duration of untreated psychosis.
- Study quality was mostly fair (67.5% of studies) with GRADE certainty only low to moderate. - Wide confidence intervals (especially for current CU and lifetime CUD) suggest substantial heterogeneity across studies. - No randomized controlled trials were available to assess intervention efficacy; evidence base is largely observational.
Roughly one-third of youth with early-onset psychosis currently use cannabis and over one-third meet criteria for CUD — screen for cannabis use routinely in this population. Integrate substance use care into early psychosis programs, and prioritize longitudinal and interventional research to guide treatment.
