Cannabis Use Disorder Carries Greater Psychiatric Risk in Adolescents Than Adults, Johns Hopkins Study Finds
By Dr. Benjamin Caplan, MD | Board-Certified Family Physician, CMO at CED Clinic | Evidence Watch
A large propensity-matched study from Johns Hopkins found that cannabis use disorder was associated with lower psychiatric risk than other substance use disorders in adults, but higher psychiatric risk in adolescents. The age-dependent reversal is clinically significant, suggesting that adolescent brains may be uniquely vulnerable to cannabis-related psychiatric harm compared to other drugs, and that risk communication should be tailored accordingly.
Cannabis Use Disorder Carries Greater Psychiatric Risk in Adolescents Than Adults, Johns Hopkins Study Finds
A propensity-matched cohort study suggests age fundamentally shapes how cannabis use disorder compares to other substance use disorders in psychiatric risk, but the available summary is a media interview rather than the primary evidence, requiring caution when interpreting magnitude and methodology.
#74
Strong Clinical Relevance
Directly informs age-stratified risk communication and pediatric screening practices for cannabis use disorder, though the observational design limits causal inference.
Adolescent Psychiatry
Substance Use Disorders
Neurodevelopment
Propensity-Matched Cohort
Cannabis policy and clinical guidance have historically treated cannabis-related psychiatric risk as uniform across age groups, yet the developing adolescent brain and the adult brain respond very differently to substance exposure. With cannabis legalization expanding and youth use remaining a persistent concern, clinicians urgently need comparative evidence that distinguishes cannabis-specific risk from the broader harms of substance use disorders. Understanding whether cannabis use disorder poses different psychiatric risks in teenagers versus adults is essential for designing age-appropriate screening tools, refining harm-reduction messaging, and guiding family conversations in both primary care and specialty settings.
The relationship between cannabis use disorder and serious psychiatric diagnoses has been studied extensively, but most prior work has either pooled all ages together or focused narrowly on psychosis risk. This Johns Hopkins-led study used the TriNetX Research Network, a large multi-institutional electronic health record database, to perform a propensity-matched retrospective cohort analysis comparing patients with a documented cannabis use disorder diagnosis against patients with other (non-cannabis) substance use disorders. By matching on demographics and clinical characteristics, the investigators aimed to isolate the relative psychiatric burden attributable specifically to cannabis use disorder. Separate analyses were conducted for adult and pediatric populations, and the study further stratified by whether patients carried a single SUD diagnosis or multiple concurrent substance use disorders.
Among adults with cannabis use disorder alone, rates of schizophrenia, major depressive disorder, and psychotic disorders were lower than among adults with other substance use disorders, while anxiety disorder risk was slightly elevated. In pediatric patients, the pattern reversed strikingly: cannabis use disorder was associated with higher rates of schizophrenia, depression, and anxiety compared to other SUDs. Adults with cannabis use disorder combined with another SUD also showed lower psychiatric risk than adults with multiple non-cannabis SUDs. The lead author, Ryan C. Nicholson, MD MPH, explicitly cautioned against interpreting these findings as evidence that cannabis is safe, framing the results as evidence that risk is relative and context-dependent. The study’s observational design precludes causal conclusions, and the media summary does not provide effect sizes, confidence intervals, or sample sizes, all of which are necessary for a full appraisal of clinical significance. The primary article in the American Journal of Psychiatry should be consulted before clinical or policy application.
This study gets something important right: it refuses to treat all substance use disorders as equivalent and instead asks the comparative question that clinicians and parents actually need answered. The finding that cannabis use disorder appears less psychiatrically harmful than other SUDs in adults aligns broadly with what many of us observe in practice, where alcohol and stimulant use disorders tend to carry devastating psychiatric comorbidity. But the pediatric reversal is the real headline here. It reinforces a principle I have held firmly for years: the adolescent brain and the adult brain are not interchangeable when it comes to cannabis risk. What concerns me is the gap between a propensity-matched EHR study and the complexity of real-world cannabis exposure, where potency, frequency, route, and timing of initiation all matter profoundly and none are captured in ICD codes.
In my own practice, I counsel families of adolescents differently than I counsel adult patients. For teens, I am unambiguous: the psychiatric risks of regular cannabis use are real, potentially lasting, and not mitigated by legality or perceived social acceptability. For adults, I focus on individual context, co-occurring conditions, and whether cannabis use is part of a broader pattern of substance use or a targeted therapeutic intervention. This study gives me stronger footing for that age-differentiated approach, even as I await the full quantitative data to sharpen the conversation further.
For clinicians working with substance use disorders, this study sits at a critical juncture in the evolving evidence base. Previous research has firmly established that adolescent cannabis use is associated with increased schizophrenia risk, and this work now contextualizes that finding by showing the risk may be proportionally greater when compared to other substance use disorders in the same age group. This comparative framing is particularly useful for clinicians practicing in settings where polysubstance use is common, because it suggests that among adolescents, cannabis-specific interventions deserve at least as much clinical attention as interventions targeting other substances. For adult patients, the data suggest that cannabis use disorder, while not without psychiatric consequences, may not warrant the same level of alarm relative to alcohol or opioid use disorders when it comes to schizophrenia and psychotic outcomes specifically.
From a pharmacological and safety standpoint, clinicians should note that this study examined diagnostic categories rather than specific cannabis products, doses, or routes. Today’s high-potency concentrates and frequent daily use patterns may carry different risk profiles than what is captured in EHR diagnostic codes assigned years ago. Drug interaction considerations remain relevant: adolescents on SSRIs, antipsychotics, or stimulant medications who also
Have thoughts on this? Share it:
