Cannabis and bipolar disorder show a strong link in one of the largest studies of its kind. Researchers followed 463,396 teenagers in Northern California. Those who reported using cannabis in the past year were about twice as likely to be diagnosed with bipolar disorder later on. Their risk of psychosis was higher still.
The study appeared in JAMA Health Forum in February 2026. A team led by Dr Kelly Young Wolff at Kaiser Permanente Northern California carried it out. It lands as cannabis becomes easier to buy and more widely accepted. In 2023, 11.2 per cent of US adolescents aged 12 to 17 reported past year use, the paper notes.
How Researchers Tracked Cannabis and Bipolar Disorder Over Time
Every teenager aged 13 to 17 completed a confidential questionnaire during routine paediatric checkups between 2016 and 2023. One question asked whether they had used marijuana in the past year. At the start, 26,345 of them, or 5.7 per cent, said yes.
The team then followed the group until age 25, or until the end of 2023. Diagnoses came from health records, and the team counted only those made by clinicians. Because teens answered again at later visits, changes in their use counted too. The researchers also adjusted for sex, race and ethnicity, neighbourhood deprivation, insurance type, and alcohol and other substance use.
Bipolar Risk Roughly Doubled for Teens Who Used Cannabis
The numbers on cannabis and bipolar disorder were stark. Teenage cannabis use carried an adjusted hazard ratio of 2.01 for bipolar disorder, which means about double the risk. Psychotic disorders came in at 2.19. Depressive disorders reached 1.34 and anxiety disorders 1.24.
Those figures rest on thousands of diagnoses: 4,061 bipolar, 4,105 psychotic, 62,137 depressive and 73,096 anxiety. The authors say the bipolar result fits earlier research. A meta-analysis they cite found 2.63 times greater odds of bipolar disorder emerging among people who used cannabis. They also stress that any past year use showed the link, not only daily use.
Timing matters for cannabis use and bipolar risk. Bipolar disorder had the longest gap between first reported use and diagnosis, at 2.3 years on average. Cannabis use came first.
Depression and Anxiety Signals Faded With Age
Age made a clear difference for the more common disorders. For depressive disorders, the hazard ratio was 1.78 among 13 to 15 year olds. It fell to 1.35 at ages 16 to 17 and 1.21 at ages 18 to 20. By ages 21 to 25, the link was no longer statistically significant.
Anxiety followed the same path, dropping from 1.47 in the youngest group to 0.95 in the oldest. The authors read this as a sign that adolescence is an especially vulnerable window. THC acts on cannabinoid 1 receptors, and the adolescent brain carries them in high numbers. The paper adds that typical cannabis flower in northern California now exceeds 20 per cent THC.
Could Something Else Explain the Cannabis and Bipolar Disorder Findings?
The researchers tested that question several ways. When they accounted for earlier psychiatric diagnoses, the bipolar hazard ratio stayed high at 1.73. When they removed every teenager with any psychiatric history at baseline, it reached 2.00. Finally, an unmeasured factor would need an association of at least 3.44 with both cannabis use and bipolar disorder to explain the finding away.
The authors still urge caution. No design like this can prove that cannabis causes bipolar disorder. Reverse causation remains possible, since some young people may turn to cannabis to cope with early symptoms before a diagnosis. Use was also self-reported, and the sample came from one insured population in a state where cannabis is legal.
What Cannabis Use and Bipolar Risk Mean for Prevention
The authors argue that the findings back early screening for cannabis at paediatric visits. They also call for stronger public health messaging and policies that limit youth exposure as legalisation spreads. Their list includes more prominent health warnings, stricter enforcement of sales to minors, and tighter rules on packaging, flavours and advertising.
Use was most common among older teens, among Hispanic, Black and White adolescents, and among those on Medicaid or living in more deprived neighbourhoods. The authors say prevention efforts should reach those groups first.
For parents, clinicians and lawmakers, the practical point is plain. Cannabis and bipolar disorder showed one of the two largest signals in the study, alongside psychosis. Delaying or preventing teenage cannabis use looks like a sensible place to start.
(Source: jamanetwork)