Does Medical Cannabis Work for Mental Health Conditions?
A systematic review and meta-analysis in The Lancet Psychiatry found no benefit of cannabinoids for anxiety, PTSD, psychotic disorders, anorexia nervosa, or opioid use disorder, and no randomized trial evidence at all for depression. The review pooled 54 randomized controlled trials with 2,477 participants, searching the literature published between January 1, 1980 and May 13, 2025. This is a synthesis of existing trials, not a new trial. It did find signals of benefit for four conditions: cannabis use disorder, insomnia, tic or Tourette’s syndrome, and autism spectrum disorder.
This matters because interest in medical cannabis for mental health has grown fast. Many patients and providers have hoped cannabinoids could offer a new path for conditions like depression and anxiety. The pooled trial data does not support that hope, though it does point somewhere else.
What the Research Shows
The review included 54 randomized controlled trials with 2,477 participants, 1,713 (69%) male and 764 (31%) female, with a median age of 33.3 years. The authors assessed risk of bias with the Cochrane Risk of Bias 2.0 tool and graded certainty with the GRADE framework.
Compared with placebo, a combination of cannabidiol and delta-9-tetrahydrocannabinol reduced cannabis withdrawal symptoms (SMD -0.29, 95% CI -0.57 to -0.02) and weekly grams of cannabis used (-1.00, -1.69 to -0.30) in people with cannabis use disorder. Any cannabinoid type increased sleep time in people with insomnia, both by electronic device recording (0.54, 0.14 to 0.95) and by sleep diary (0.55, 0.01 to 1.09). Tic severity fell in tic or Tourette’s syndrome (-0.68, -1.03 to -0.34), and autistic traits fell in autism spectrum disorder (-0.36, -0.66 to -0.07).
The null results were specific. There were no significant effects on outcomes associated with anxiety, anorexia nervosa, psychotic disorders, post-traumatic stress disorder, or opioid use disorder. For ADHD, bipolar disorder, obsessive-compulsive disorder, and tobacco use disorder there were too few data to pool. For depression there was an absence of randomized controlled trial evidence entirely.
One finding ran the wrong way: cannabinoids increased cocaine craving in people with cocaine use disorder (0.69, 0.22 to 1.15) compared with placebo.
On safety, cannabinoids carried higher odds of all-cause adverse events (OR 1.75, 95% CI 1.25 to 2.46), a number needed to treat to harm of 7. There were no higher odds of serious adverse events or of study withdrawal.
Dr. Kumar’s Take
I think this review is important precisely because it refuses to give a simple answer. Cannabinoids are not a general mental health treatment. They did nothing for anxiety or PTSD, and for depression, the condition patients ask me about most, there is not a single randomized trial to weigh. That is worth saying plainly when marketing claims are getting bolder.
At the same time, I would not read this as a blanket negative. There were measurable effects on cannabis withdrawal, tics, autistic traits, and sleep time in insomnia. The authors are clear that the certainty of that evidence is generally low, and 24 of the 54 trials carried a high risk of bias. So these are leads worth following, not prescriptions.
The safety number I keep coming back to is the number needed to treat to harm of 7. For every seven people treated, roughly one experiences an adverse event they would not have had on placebo. Serious adverse events were not elevated, which is reassuring, but for a treatment with low-certainty benefit, that ratio is not favorable.
Key Patterns Across Studies
The pattern here is not “cannabinoids fail.” It is that effects are condition-specific and the evidence base is thin. Where benefit appeared, it appeared in a narrow set of conditions: cannabis use disorder, insomnia, tic or Tourette’s syndrome, and autism spectrum disorder. Where it did not appear, it did not appear at all.
The quality picture is consistent across those findings. Twenty-four of the 54 trials (44%) had a high risk of bias, and certainty of evidence for most outcomes was low. That applies to the positive results as much as the null ones.
Gaps in the Evidence
The largest gap is depression, where the review found no randomized controlled trial evidence to analyze. Beyond that, ADHD, bipolar disorder, obsessive-compulsive disorder, and tobacco use disorder had too few data to meta-analyse.
Where trials do exist, their quality limits what can be concluded. The authors state that there is a crucial need for more high-quality research, and that given the scarcity of evidence, routine use of cannabinoids for mental disorders and substance use disorders is currently rarely justified. I would not treat this review as settling the question in either direction.
It is also worth noting that this review focused on cannabinoids as the primary treatment for diagnosed mental health and substance use disorders. It did not address recreational use, pain management, or other medical uses of cannabis. Those are separate questions with their own bodies of evidence.
Practical Takeaways
- If you are considering medical cannabis for anxiety or PTSD, this pooled trial evidence found no benefit. Talk to your doctor about treatments with stronger evidence, such as cognitive behavioral therapy or established medications.
- For depression specifically, there is no randomized controlled trial evidence here to support cannabinoid treatment.
- Do not stop or change your current mental health treatment based on claims about cannabis without discussing it with your healthcare provider first.
- If you have cocaine use disorder, be aware that cannabinoids increased cocaine craving in these trials.
- Ask about adverse events. Cannabinoids raised the odds of any adverse event (number needed to treat to harm of 7), though serious adverse events were not increased.
Related Studies and Research
If you are interested in evidence-based approaches to mental health, these related articles may be helpful:
- L-theanine supplementation for mental disorders: systematic review explores another supplement that has been studied for mental health conditions.
- CBT vs all other depression treatments: massive meta-analysis of 409 trials highlights the strong evidence behind cognitive behavioral therapy.
- Internet-based CBT self-help: long-term depression benefits proven shows how accessible therapy options can deliver lasting results.
- Potential health benefits of cold-water immersion review covers another alternative approach that has been studied for mood and mental health.
FAQs
Can cannabinoids help with anxiety or depression?
This review found no significant effects on outcomes associated with anxiety. For depression, it found an absence of randomized controlled trial evidence, so there is nothing in this pooled analysis to support cannabinoid treatment for it. If you are experiencing either condition, established options like therapy and proven medications have much stronger evidence behind them.
Which conditions did cannabinoids actually help?
Four. In cannabis use disorder, a cannabidiol and delta-9-tetrahydrocannabinol combination reduced withdrawal symptoms (SMD -0.29) and weekly grams of cannabis used (-1.00). In insomnia, any cannabinoid type increased sleep time by device recording (0.54) and sleep diary (0.55). In tic or Tourette’s syndrome, tic severity fell (-0.68). In autism spectrum disorder, autistic traits fell (-0.36). The authors rate the quality of this evidence as generally low.
How safe were cannabinoids in these trials?
Cannabinoids were associated with higher odds of all-cause adverse events (OR 1.75, 95% CI 1.25 to 2.46), which works out to a number needed to treat to harm of 7. There were no higher odds of serious adverse events or of participants withdrawing from studies. In cocaine use disorder specifically, cannabinoids increased cocaine craving (0.69, 0.22 to 1.15) compared with placebo.
Bottom Line
Across 54 randomized controlled trials and 2,477 participants, cannabinoids showed no significant effect for anxiety, anorexia nervosa, psychotic disorders, PTSD, or opioid use disorder, and there was no randomized trial evidence for depression at all. They did reduce cannabis withdrawal and use in cannabis use disorder, increase sleep time in insomnia, reduce tic severity in tic or Tourette’s syndrome, and reduce autistic traits in autism spectrum disorder, all on evidence the authors grade as generally low certainty. They also raised the odds of any adverse event, with a number needed to treat to harm of 7, though not of serious adverse events. The authors’ own conclusion is the one I would give a patient: there is a crucial need for more high-quality research, and routine use of cannabinoids for these conditions is currently rarely justified.
