In 40 people with cocaine use disorder, psilocybin nearly doubled abstinence odds

No medication has been proven effective for cocaine use disorder — a gap that has persisted despite decades of pharmacological research into the condition. A trial published this week in JAMA Network Open, conducted at a major medical research center in the Deep South, tests psilocybin against that gap directly, in a population the researchers specifically describe as underrepresented in psychedelic research to date [s1].

The design

This randomized, quadruple-blind, placebo-controlled trial recruited individuals with cocaine use disorder who were motivated to quit and without significant psychiatric comorbidities, between May 2015 and August 2023 [s1]. Participants were randomized 1:1 to a single oral dose of psilocybin (25 mg per 70 kg of body weight) or an active placebo (100 mg diphenhydramine, an antihistamine with sedating effects meant to help preserve blinding) [s1]. All participants received manualized cognitive-behavioral psychotherapy delivered roughly one month before and one month after an all-day investigational drug session [s1]. Outcomes — percentage of cocaine-abstinent days, rates of complete abstinence, and time to first cocaine lapse — were tracked through 180 days after treatment ended, using a structured recall interview confirmed by urine testing [s1].

Of the 40 participants, 82.5% were men, median age was 50, 82.5% were Black, and 65% had an annual income of $20,000 or less [s1]. Four participants were lost to follow-up, leaving 36 who completed the full 180-day assessment [s1].

What it found

Psilocybin recipients had a significantly higher percentage of cocaine-abstinent days than placebo recipients (β = 28.95, 95% CI 18.22–39.67, p < .001) [s1]. They also had substantially greater odds of achieving complete abstinence from cocaine over the follow-up period (odds ratio 18.37, 95% CI 1.92–2468.17, p = .007) and a lower risk of relapsing to cocaine use over time (hazard ratio 0.28, 95% CI 0.13–0.60, p = .001) [s1]. No serious adverse events occurred [s1].

Why that odds ratio's confidence interval is so wide

An odds ratio of 18.37 with a confidence interval spanning from 1.92 to 2468.17 is a genuinely wide range — nearly a 1,300-fold spread between its lower and upper bounds — which reflects how few participants achieved complete abstinence in a trial of just 40 people; when an outcome is rare in one or both groups, odds ratios can become statistically unstable and produce very wide intervals even when the underlying association is real and statistically significant. The lower bound, 1.92, still indicates the finding clears conventional statistical significance, but the true effect size could plausibly sit anywhere across that enormous range, and the point estimate of 18.37 itself should be read with real caution given how it was generated.

Why the study population is part of the story

The trial's authors specifically frame their population — 82.5% Black participants, two-thirds with household income under $20,000 — as "underrepresented and vulnerable" [s1], a notable departure from the demographic makeup of most published psychedelic trials, which have skewed heavily white and higher-income. Cocaine use disorder itself carries disproportionate impact in lower-income and Black communities in the US, partly a legacy of decades of disparate drug policy enforcement, making a trial recruiting specifically from this population a meaningful test of whether psilocybin's benefit, established mostly in different demographic contexts for other conditions, extends to the population most affected by this particular disorder.

What this doesn't establish

Forty participants is a small trial for a condition with no existing effective medication — a genuinely notable gap this trial addresses, but one trial of this size, however promising, does not establish psilocybin as an effective cocaine use disorder treatment on its own. The trial recruited people already motivated to quit, which may not represent the broader population of people with cocaine use disorder, many of whom aren't actively seeking treatment. The 1,300-fold-wide confidence interval on the complete-abstinence odds ratio, discussed above, is a real statistical caution flag even though the result is significant.

What the authors say

The study's authors describe psilocybin as appearing "safe and efficacious" in this population and explicitly call for further research to replicate and expand these findings [s1] — language appropriately cautious given the trial's size, even as the effect sizes reported are substantial.

What to watch

Whether a larger, multi-site trial replicates this finding with a narrower confidence interval, and whether it extends to people with cocaine use disorder who aren't already motivated to seek treatment. Psilocybin is not approved for cocaine use disorder or any addiction indication; this article describes investigational drug trial results and is not medical advice.

Sources

1. Psilocybin in the Treatment of Cocaine Use Disorder: A Randomized Clinical Trial — JAMA Network Open, 1 May 2026