Cannabis use and subsequent onset of manic symptoms and bipolar disorder: A systematic review and meta-analysis.
Bipolar disorder is a disabling psychiatric condition affecting approximately 2% of the population, for which identifying modifiable risk factors prior to illness onset is a research priority. Cannabis use is highly prevalent, yet its role in the development of bipolar-spectrum outcomes remains unclear. Previous reviews have included few studies or combined distinct bipolar-spectrum outcomes, limiting outcome-specific inference.
To evaluate the association between cannabis exposure and subsequent incident BD and manic/hypomanic symptoms separately and examine potential moderating factors.
Registered (https://osf.io/s38x4/), PRISMA/MOOSE-compliant systematic review and meta-analysis of PubMed and Web of Science from inception to October 2025. We included longitudinal observational studies comparing cannabis-exposed and unexposed individuals without prior BD and assessing subsequent bipolar-spectrum outcomes. We meta-analysed studies with extractable data, calculating the unadjusted risk ratios (RRs) for incident BD and manic/hypomanic symptoms. Random-effects meta-analyses, heterogeneity analyses, subgroup analyses and meta-regressions were performed. Study quality was appraised using the Newcastle-Ottawa Scale (NOS).
Nineteen studies were included (total N = 6,960,517; mean age 23.8 ± 11.8 years; 49% female). Cannabis use was associated with increased risk of BD (RR = 2.46, 95% CI = 1.57-3.84; k = 8; N = 206.572) and manic/hypomanic symptoms (RR = 2.11, 95% CI = 1.07-4.18; k = 6; N = 57,243). Moderator analyses identified significant clinical and methodological sources of heterogeneity. NOS scores ranged from 6 to 9, indicating good to high methodological quality.
Cannabis use was associated with an increased risk of subsequent bipolar-spectrum outcomes. Routine evaluation of cannabis use alongside prevention strategies, to delay initiation and reduce use, may warranted.
To evaluate the association between cannabis exposure and subsequent incident BD and manic/hypomanic symptoms separately and examine potential moderating factors.
Registered (https://osf.io/s38x4/), PRISMA/MOOSE-compliant systematic review and meta-analysis of PubMed and Web of Science from inception to October 2025. We included longitudinal observational studies comparing cannabis-exposed and unexposed individuals without prior BD and assessing subsequent bipolar-spectrum outcomes. We meta-analysed studies with extractable data, calculating the unadjusted risk ratios (RRs) for incident BD and manic/hypomanic symptoms. Random-effects meta-analyses, heterogeneity analyses, subgroup analyses and meta-regressions were performed. Study quality was appraised using the Newcastle-Ottawa Scale (NOS).
Nineteen studies were included (total N = 6,960,517; mean age 23.8 ± 11.8 years; 49% female). Cannabis use was associated with increased risk of BD (RR = 2.46, 95% CI = 1.57-3.84; k = 8; N = 206.572) and manic/hypomanic symptoms (RR = 2.11, 95% CI = 1.07-4.18; k = 6; N = 57,243). Moderator analyses identified significant clinical and methodological sources of heterogeneity. NOS scores ranged from 6 to 9, indicating good to high methodological quality.
Cannabis use was associated with an increased risk of subsequent bipolar-spectrum outcomes. Routine evaluation of cannabis use alongside prevention strategies, to delay initiation and reduce use, may warranted.
Authors
Sanjuan-Ortiz Sanjuan-Ortiz, Querol Querol, Aymerich Aymerich, Laherran Laherran, Ditlevsen Ditlevsen, Navalón Navalón, Rojo-Bofill Rojo-Bofill, Young Young, Arango Arango, Sharma Sharma, Di Forti Di Forti, Murray Murray, Correll Correll, de Pablo de Pablo
View on Pubmed