Psilocybin and OCD: what the small studies show, and what they don’t

Obsessive-compulsive disorder (OCD) involves intrusive, unwanted thoughts (obsessions) and repeated behaviours or mental rituals (compulsions). A Toronto research team notes that up to 60% of people with OCD do not respond to the standard treatments (Ledwos et al., 2025). That gap is why researchers have looked at psilocybin, the compound in magic mushrooms.

The research is real but small. This page goes through the human studies with results, the case reports, the animal work and the registered trials, and says where each falls short. It is not medical advice, it does not say what anyone should do, and it covers supervised research only.

Does psilocybin help OCD?

It is not known yet. Every human study so far has been small, and most had no placebo group. The two that compared psilocybin with an active placebo, at the University of Arizona and at Yale, reported that OCD scores fell more with psilocybin than with a comparison drug. Together they randomized 43 people, and the Yale results have not yet been peer reviewed.

So the honest summary is a consistent early signal, not a proven effect. A 2024 review of the field reached a similar view: rodent studies and small human studies point the same way, but the mechanisms are unclear and the current evidence has problems that ongoing trials aim to address (Collins, 2024). Health Canada describes the evidence for psilocybin across conditions, including OCD, as “currently limited” (Health Canada).

What human studies have reported results?

We found five studies of psilocybin in people with OCD that have reported results, two of them only as preprints. The table sets them side by side. Scores come from the same clinician-rated scale (the Yale-Brown Obsessive Compulsive Scale, or Y-BOCS), but the designs differ so much that the numbers should not be compared across rows. We leave out dose amounts on purpose.

The 2006 study: nine people

The first modern study ran from November 2001 to November 2004. Nine people with OCD and no other current major psychiatric disorder received up to four single doses, in rising order, with a very low dose slipped in at random and double-blind. Each session lasted eight hours in a clinic, followed by an overnight stay on an inpatient unit (Moreno et al., 2006).

All nine had a marked fall in OCD scores, of 23% to 100%, during at least one session, and improvement generally lasted past 24 hours. But the analysis found a significant effect of time and no significant effect of dose. One person had a brief rise in blood pressure; no other significant adverse effects were seen (Moreno et al., 2006). The authors describe it as a “modified double-blind” study; later reviewers call it a small open-label trial (Collins, 2024). Either way, there was no placebo group.

The 2026 Arizona trial: 15 people, eight sessions

The same group returned with a randomized trial at the University of Arizona. Fifteen people, who had lived with OCD for 18.8 years on average and had severe symptoms, were split into three groups of five: a higher psilocybin dose, a lower dose, or lorazepam, a sedative used as an active placebo. After four weekly double-blind sessions, everyone received four more higher-dose sessions. Enrolment ran from spring 2019 to spring 2022, with a 17-month pause for COVID-19 (Moreno et al., 2026).

There were no serious adverse events, no psychotic symptoms and no significant change in suicide severity scores. Psilocybin, but not placebo, lowered OCD scores. At the end of eight weeks, 73.3% had responded (a fall of at least 35%) and 40% were in remission, and the effect faded but remained substantial at six months (Moreno et al., 2026). The authors note the small sample, a heavy schedule that may have favoured motivated people with more resources, and that blinding was not formally assessed.

The Yale trial: 28 people, one session

This trial randomized 28 adults whose OCD had not responded to treatment to one session of psilocybin or niacin, a vitamin that causes flushing and is used as an active placebo, with unstructured support. Two days later, scores on an acute OCD scale had fallen from about 24 to about 14 with psilocybin and had not changed with niacin. At one week, 69.2% (9 of 13) on psilocybin had responded, against none of 14 on niacin. The benefit persisted through 12 weeks in the psilocybin group (Kelmendi et al., 2026, preprint).

Three cautions. It is a preprint, not yet peer reviewed. The niacin group crossed over to open-label psilocybin after about a week, so the 12-week results have no comparison group. And there was one serious adverse event; the trial record posted on 16 October 2026 lists it as suicidal ideation in 1 of 14 people in the psilocybin group (ClinicalTrials.gov NCT03356483). Interviews with 12 participants found that acute effects often felt “partial”, which the researchers think OCD symptoms may have interfered with (Ching et al., 2025).

London and Johns Hopkins

At Imperial College London, 19 adults received a lower dose, then a higher one four weeks later, without knowing the order. One week after the higher dose, OCD scores were lower than after the lower dose, mostly in compulsions. That difference faded over the following three weeks, and no effect on depression scores was found. Nobody had a serious adverse event. The authors list the small sample and the lack of randomization as limits (Pellegrini et al., 2025).

At Johns Hopkins, 37 people were randomized to start two psilocybin sessions straight away or after an eight-week wait; 30 completed follow-up. The immediate group’s scores fell more than the waitlist group’s, and the improvement held a month after the second session (Tiwari et al., 2026, preprint). It is open-label with no placebo, and it is not yet peer reviewed. The registry lists 35 people enrolled, not 37 (ClinicalTrials.gov NCT05546658).

What do case reports say, and why are they weak evidence?

Interest in psilocybin for OCD began with individual stories. These are the ones we could confirm:

• A 1987 report in the American Journal of Psychiatry, titled “Relief of obsessive-compulsive symptoms by LSD and psilocin” (Leonard and Rapoport, 1987).
• A 2014 report of a person who used psilocybin on their own to reduce OCD symptoms for several years. The author states plainly that they do not endorse it and calls for proper research (Wilcox, 2014).
• A 2022 report of one patient with treatment-resistant OCD treated in a clinical study at Yale and followed for a year, whose score fell from 24 to between 0 and 2 (Kelmendi et al., 2022).
• A 2026 report of identical twins, one with severe OCD who took small amounts of psilocybin mushrooms on their own. The affected twin reported fewer symptoms, but a test of mental flexibility still showed deficits compared with the unaffected twin (Drange et al., 2026).

Case reports can raise a question but cannot answer it. A single person’s course may reflect natural ups and downs, other treatment, expectation or chance, and striking successes are more likely to be written up than failures. Self-treatment cases add unknowns about what was taken. None of these reports shows that psilocybin works or is safe for OCD.

What has animal research found?

Animal studies cannot show that psilocybin treats OCD in people. Mice do not have OCD; researchers use behaviours such as excessive grooming or burying marbles as rough stand-ins. With that label clearly attached:

• Animal work: in ordinary lab mice, psilocybin reduced marble burying. Blocking the receptor usually linked to psychedelic effects did not stop this, and a single injection’s effect did not last (Singh et al., 2023).
• Animal work: in 50 mice bred to groom excessively (SAPAP3 knockout mice), grooming rose by about 119% over 21 days with a dummy injection but fell by about 15% with psilocybin and 19% with a mushroom extract. In mice that responded, the effect lasted up to seven weeks (Brownstien et al., 2025).
• Animal work: a second team using the same mouse strain found lasting reductions in compulsive behaviour in male knockout mice and less grooming in females, for up to a week, and no effect on anxiety-like behaviour (Gattuso et al., 2025).

These findings help explain why trials were funded, but mouse behaviour is a long way from human obsessions. For what is known about how psilocybin acts on the brain, see does psilocybin rewire the brain?

Which trials are registered, and have they reported?

A search of ClinicalTrials.gov on 19 October 2026 for psilocybin and OCD returned 10 records. The table lists the ones testing treatment. A registration shows that a trial was planned, not that it worked, and dates and sizes are the sponsors’ own.

The Canadian trial at CAMH

The only Canadian site we found is the Centre for Addiction and Mental Health (CAMH) in Toronto. Its published protocol describes a 12-week open-label study of ten adults with treatment-resistant OCD, each receiving one psilocybin session with psychological support, plus brain imaging and EEG. The study received ethics approval from CAMH and a No Objection Letter from Health Canada (Ledwos et al., 2025). The registry lists it as completed on 4 September 2026, with no results posted yet (ClinicalTrials.gov NCT06299319). With no control group, its results will show feasibility and safety rather than settle whether psilocybin works. For other studies in the city, see psilocybin clinical trials in Toronto.

What does this evidence not show?

• It does not show that psilocybin is an effective, safe or approved treatment for OCD. The two placebo-controlled trials together randomized 43 people.
• It does not separate the drug from expectation. Psilocybin’s effects are hard to hide, and the Arizona team did not formally test whether participants could tell which group they were in.
• It does not show how long any change lasts. Controlled comparisons ran for days to weeks; longer follow-ups had no comparison group.
• It does not cover the people the studies left out, such as those with a personal or family history of psychosis or mania, recent suicide attempts or active substance use disorders in the Arizona trial.
• It does not show who might be harmed. Studies this small cannot detect uncommon risks, and the Yale trial recorded a serious adverse event.
• It does not test self-treatment or microdosing. Trial participants were screened, prepared and supervised. Nothing here applies to taking mushrooms outside such a setting.

Is psilocybin an approved OCD treatment in Canada?

No. Health Canada’s page, last modified 20 October 2026, states that production, sale and possession of magic mushrooms are illegal in Canada and that there are no approved therapeutic products containing psilocybin in Canada or elsewhere. It names OCD among the conditions being researched and says clinical trials are the most appropriate way to advance research with unapproved drugs such as psilocybin (Health Canada).

What is still unknown?

• Whether the early results hold up in larger, multi-site, peer-reviewed randomized trials.
• How much of any change comes from the drug versus support, expectation and the natural course of OCD.
• Whether one session or several is better. The Arizona trial found that more cumulative psilocybin went with bigger drops, but that was an after-the-fact analysis.
• Why some people respond and others don’t. In the Yale trial, stronger “mystical-type” experiences went with lower OCD scores later, while challenging experiences did not (Shnayder et al., 2026).
• How psilocybin fits with established OCD treatments, and what it means for people who would have to stop other medicines to take part.

For now, psilocybin for OCD is a question being tested, with small, encouraging signals and no settled answer. For the wider picture, see what the research on psilocybin’s benefits shows, our look at psilocybin depression trials, magic mushrooms and anxiety, and the basics of what magic mushrooms are.

Shroomverse sells mushroom products. This article is general information about clinical research on psilocybin and OCD. It is not medical, legal or purchasing advice, and it does not recommend any product.