A Psilocybin Trial Reported 55% Abstinence. Then 93% of Patients Guessed Their Group
A French trial of psilocybin for alcohol use disorder reported 55 percent abstinence at 12 weeks. The control group came in at 11 percent. It also reported that 93.3 percent of patients correctly guessed which group they were in. We came to the trial through a column at Clinical Trial Vanguard, which focused on a different question. The blinding number is the one worth your attention.
Reading trial data well is a clinical skill. Lowering your overhead is a business one. HealingMaps’ GPO, the first built for the ketamine industry, gets member clinics 40 percent or more off medical supplies, 15 to 20 percent off malpractice and liability premiums, discounted LegitScript certification, and 20 to 30 percent off HIPAA compliant payment processing. Joining is free with no obligation. See what your clinic qualifies for →
| Key Takeaway | Detail |
|---|---|
| The trial | Pilot randomized controlled study in a French inpatient addiction program, published in Addiction |
| The sample | 350 screened, 30 included, randomized 2 to 1 |
| The doses | Two sessions of 25 mg (n=20) or 1 mg (n=10), three weeks apart, added to standard care |
| The primary outcome | Feasibility, not efficacy |
| The efficacy signal | 55 percent abstinent at 12 weeks versus 11 percent, p = 0.043 |
| The catch | 93.3 percent of patients and 86.7 percent of investigators guessed the assignment correctly |
What the Trial Actually Did
The study ran at a single French inpatient addiction program and appeared in Addiction. Patients had severe alcohol use disorder plus depressive symptoms, and had completed detox 14 to 60 days earlier.
Of 350 patients screened, 30 were included. Twenty received two 25 mg psilocybin sessions and ten received 1 mg, three weeks apart, on top of standard care.
The primary outcome was feasibility. The authors set out to learn whether this could be run at all, not whether it works.
The Numbers People Will Quote
At 12 weeks, 11 of 20 patients in the 25 mg group were abstinent, which is 55 percent. In the control group it was 1 of 9, or 11 percent.
Drinking days and craving frequency also moved further in the 25 mg group, with p values of 0.038 and 0.045.
One number went the other way. Relapse was 35 percent versus 50 percent, and that difference was not statistically significant.
Then Almost Everyone Guessed
Blinding failed. 93.3 percent of patients and 86.7 percent of investigators correctly identified the assignment.
A 25 mg psilocybin session is not a subtle experience. A 1 mg session is not much of one. Patients knew, and so did the people assessing them.
That matters because expectation drives outcomes in addiction treatment. When a patient knows they received the real thing, and the assessor knows too, the measured difference includes that knowledge.
The Control Group Did Not Hold
Four control participants refused the second session after working out their assignment. The paper reports that difference at p = 0.019.
One of them took MDMA on their own. The control group result also rests on nine patients, because one was lost to follow up.
Differential dropout like that bends a comparison. The people most disappointed by their assignment are the ones who leave, and they do not leave at random.
Safety Looked Unremarkable
Adverse events were reported in 50 percent of the 25 mg group and 60 percent of the control group.
One participant in the 25 mg group had a myocardial infarction three days before the second session. The investigators judged it unrelated to treatment, and that patient did not receive the second dose.
This Is the Field’s Recurring Problem
Unblinding is not a flaw in this one study. It is the structural issue in psychedelic research, which we covered when FDA confronted it directly.
Trial designers are responding in different ways. Definium built a low dose arm aimed at the unblinding problem. The VA trial gives psilocybin to the control group too.
FDA spent much of its September hearing on exactly these questions, which we covered in our look at the label fight.
What a Clinician Should Take From It
Thirty patients cannot establish efficacy, and the authors do not claim otherwise. Their conclusion is that the approach is feasible, acceptable and safe in this population.
The useful habit is to read the blinding line in every psychedelic trial before reading the effect size. If nearly everyone guessed, the effect size carries expectation inside it.
That caution is not unique to psilocybin. Evidence has gone both ways on whether the psychedelic experience drives the benefit, including work suggesting ketamine may help alcohol addiction through other means.
One Footnote on the Second Paper
A separate preprint from the same trial reports reduced egocentric bias on a perspective taking task after 25 mg.
It is a secondary analysis of these same 30 patients, it has not been peer reviewed, and the effect appeared in reaction times rather than accuracy.
Read it as a hypothesis for a larger trial, not as a second piece of evidence.
The Bottom Line
A 55 percent abstinence rate in a 30 patient pilot is encouraging and nothing more. The same paper tells you why, which is to its credit.
Patients looking for supervised care can start with our directory of verified ketamine clinics.
This article is for informational purposes and is not medical advice. Psilocybin is not approved by the FDA, and the trial described was a pilot whose primary outcome was feasibility.
