A Major New Review Found No RCT Evidence Cannabis Works for PTSD. Thirty-Eight States Still List It as a Qualifying Condition.

The largest systematic review of cannabis and mental health RCTs just landed. The policy response has been silence.

A landmark study published this month in The Lancet Psychiatry — the most comprehensive randomized controlled trial analysis of medicinal cannabis and mental health ever conducted — found no significant evidence that cannabis effectively treats anxiety, depression, or post-traumatic stress disorder.

Those are the three most common reasons Americans hold a medical cannabis card.

The review, led by Dr. Jack Wilson at the University of Sydney’s Matilda Centre, analyzed 54 RCTs spanning 45 years of clinical data from 1980 to 2025, encompassing nearly 2,500 participants worldwide. Its finding on the conditions that have driven the expansion of state medical programs is unambiguous: for PTSD, anxiety, and depression, the authors found no significant treatment effect.

“We found no evidence any form of cannabis is effective in treating anxiety, depression, or post-traumatic stress disorder,” Wilson told the press after publication.

Thirty-eight states and the District of Columbia currently list PTSD as a qualifying condition for medical cannabis. Anxiety and depression appear on qualifying lists in dozens more. None of these programs are built on RCT evidence — because, as the Lancet review makes plain, that evidence doesn’t exist.

What the Study Actually Found

The review is not anti-cannabis. It’s a careful scientific accounting of what randomized controlled trial data actually supports.

On PTSD, anxiety, and depression — conditions where demand for cannabis has driven years of program expansion — the evidence base is either negligible or nonexistent at the RCT standard. Zero trials met the bar for depression. The anxiety and PTSD findings were statistically insignificant across the available trials.

The researchers did find limited, low-quality evidence that cannabis may help with certain conditions: seizure disorders (where FDA-approved Epidiolex is already a benchmark), spasticity in multiple sclerosis, some chronic pain applications, and — notably — cannabis use disorder itself. But the lead author was direct about quality: “The overall quality of evidence for these other conditions… was low. In the absence of robust medical or counseling support, the use of medicinal cannabis in these cases are rarely justified.”

These findings aren’t isolated. A Cochrane review published in January found “no clear evidence” for THC’s effectiveness in treating chronic pain. A JAMA review from November, led by researchers at UCLA, UCSF, and NYU, concluded the evidence “does not support the use of cannabis or cannabinoids for most conditions for which it is promoted.”

The research base is not ambiguous. It’s consistently thin, and it’s been thin for a long time.

The Policy Gap

Here is the quiet contradiction at the center of American cannabis law: the federal government restricts cannabis research on the grounds that it’s a Schedule I substance with no accepted medical use — and state governments have spent two decades authorizing medical use based on anecdote, advocacy, and political momentum rather than the clinical trial data Schedule I status has suppressed.

That circular logic has consequences. Researchers who wanted to run the RCTs that might have proven or disproved these claims have spent years navigating DEA research licensing bottlenecks, restricted cannabis supplies from the National Institute on Drug Abuse, and an administrative infrastructure that treats cannabis research as inherently suspect. The Senate passed the Cannabis Research Act (S.1876) 92-4 in February — the only major cannabis the full legislative tracker with that level of bipartisan support in a decade. It’s been sitting in the House since.

Meanwhile, states have continued to add qualifying conditions. About 27 percent of Americans aged 16-65 who use cannabis medically say they do so primarily for mental health symptoms, according to population data cited in the Lancet review. They’re making that choice in a legal framework that was designed to look like medicine without being required to behave like it.

Dr. Wilson is careful about causation. “Though our paper didn’t specifically look at this, the routine use of medicinal cannabis could be doing more harm than good by worsening mental health outcomes — for example, a greater risk of psychotic symptoms and developing cannabis use disorder — and delaying the use of more effective treatments.”

That’s not a fringe position. It’s the conclusion of the most methodologically rigorous review ever conducted on the subject.

What Happens Now

Almost certainly: nothing in the short term.

The political architecture of state medical cannabis programs was never primarily scientific. States added conditions through legislative votes or ballot initiatives. The constituency of cardholders is now enormous — and politically organized. Removing PTSD from a qualifying list is the kind of move that ends careers in state legislatures.

At the federal level, the DEA’s ongoing Schedule III rescheduling process has operated entirely on a different axis — focused on abuse potential comparisons and administrative law, not therapeutic efficacy. Rescheduling cannabis to Schedule III would not require states to revise their qualifying condition lists. It would ease research barriers, which would allow the clinical trial gap to slowly close over the next decade. That’s worth doing. But it’s not a solution to the problem the Lancet review names.

The FDA, for its part, has approved cannabis-derived medications for specific conditions with RCT evidence: Epidiolex for rare seizure disorders, Marinol and Syndros as synthetic THC for specific nausea and appetite applications. The FDA pathway exists. It’s just slow, expensive, and doesn’t map neatly onto state medical program logic.

Who’s Behind This

Dr. Jack Wilson (lead author, University of Sydney): Conducted the review through the Matilda Centre for Research in Mental Health and Substance Use, funded by Australia’s National Health and Medical Research Council. No industry ties disclosed. His position is straightforwardly scientific: the evidence that exists does not support the current scope of medical cannabis authorization for mental health.

Kevin Sabet (Smart Approaches to Marijuana): Published an op-ed in STAT News this week citing the Lancet findings to argue against state medical programs broadly. Sabet is a prohibitionist — his position on cannabis policy extends well beyond this study’s scope — but his citation of the data is accurate. The political use of accurate evidence is worth distinguishing from the evidence itself.

State medical cannabis regulators: No state health department has publicly responded to the Lancet findings. The political incentive to do so is low. The constituency pressure to maintain current qualifying conditions is high.

The DEA: The agency’s Schedule I research barriers are the single largest structural reason the RCT evidence base is thin in the first place. DEA rescheduling would reduce those barriers. The review implicitly makes the case for it.

The Bottom Line

Thirty-eight states built medical cannabis programs on the premise that cannabis treats PTSD. The largest randomized controlled trial review ever conducted on the question says the evidence for that doesn’t exist. The federal policy that suppressed that research for decades also deserves to answer for the gap. Both things are true simultaneously — and neither cancels the other out.