Oregon Tracks Psilocybin Safety for Three Days. The New Data Found Problems at Three Months
Oregon requires its licensed psilocybin service centers to report safety events that happen within three days of a session. A new study followed 346 participants for three months. The problems it found mostly showed up after that three day window had closed.
Follow up structure is one of the few safety controls a clinic fully controls. 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 study | Published in JAMA Network Open on August 19, 2026, led by Oregon Health & Science University |
| The sample | 346 participants at 24 of Oregon’s 26 licensed service centers, supervised by 83 facilitators |
| Mental health results | Moderate to severe depression fell from 42.2 percent at baseline to 16.5 percent at three months |
| Serious reactions | Four people, or 1.2 percent, needed medical attention. All four were psychedelic-naive |
| The late signal | Reports of thoughts of dying or suicide went 3 at one week, 2 at one month, then 6 at three months |
| The reporting gap | Oregon’s state mandate captures only events within three days of a session |
What the Study Actually Did
This was a prospective cohort study run by Oregon Health & Science University. The lead author is Dr. P. Todd Korthuis. It appeared in JAMA Network Open on August 19, 2026.
Researchers enrolled 346 people between November 2024 and March 2026. Participants came from 24 of the 26 licensed psilocybin service centers operating in Oregon. Eighty-three licensed facilitators supervised the sessions.
Retention was unusually good. Ninety-three percent completed the one week follow up. Ninety percent were still in at three months. The mean dose was 31.9 mg.
The Mental Health Results Were Strong
Depression scores dropped sharply. Mean PHQ-9 went from 9.3 at baseline to 4.2 at one month. It sat at 5.2 at three months.
The categorical numbers are easier to picture. Moderate to severe depression fell from 42.2 percent of participants to 16.5 percent. Moderate to severe anxiety fell from 45.1 percent to 13.2 percent. PTSD symptoms fell from 48.0 percent to 16.8 percent.
Life satisfaction rose from 63.0 percent to 82.3 percent. On the efficacy side, this is a good looking dataset.
Four Serious Reactions, All First Timers
Four participants, or 1.2 percent, had adverse behavioral reactions requiring medical attention. Every one of them was psychedelic-naive.
The cases are described individually in the paper. One man over 65 had a violent reaction at a 50 mg dose. He was taken to a hospital during the session. Three others reported emergency department visits, new medication, or persistent suicidal ideation at one and three months.
That pattern is the most directly usable finding in the study. Prior psychedelic experience was the dividing line. Roughly 44 percent of the sample was psychedelic-naive, and all four serious cases came from that group.
The Harm Signal Arrives Late
Here is the number that deserves attention. Reports of new thoughts of dying or suicide were 3 at one week. They fell to 2 at one month. At three months there were 6.
Participants who described the experience as harmful followed the same shape. That figure was 1.6 percent at one week, 1.9 percent at one month, and 2.3 percent at three months.
Nearly everything else in the study improves over time. These two measures move the other way. Whatever is driving them is not visible in the days after a session.
Oregon Only Looks for Three Days
The authors state the problem directly in their limitations. State mandates for collecting safety events cover only events occurring within three days of services. They write that this underestimates actual safety events.
Read that against the timeline above. Oregon’s official safety record cannot see the adverse events this study found at one month and three months.
That is not a criticism of the service centers. It is a design flaw in the surveillance rule. Any state copying the Oregon model inherits it.
Lingering Visual Symptoms Were Common and Mostly Mild
Ten percent of participants reported symptoms consistent with hallucinogen persisting perception at one month. At three months the figure was 8.7 percent.
The distress numbers matter more than the prevalence. Only one person reported significant distress at one month. Nobody did at three months.
So these symptoms were frequent but largely tolerable. That framing beats dismissing them or alarming people. It matches what the new clinics treating lasting psychedelic side effects report seeing.
Satisfaction Was High at the Same Time
The safety findings sit alongside strong satisfaction data. Ninety-seven percent rated the services high quality. Ninety-two percent said they experienced benefit.
Almost 59 percent ranked the session among the ten most meaningful experiences of their life. Nearly 89 percent said they felt safe during the experience.
Both things are true at once. Most people did well, and a small group did not, and the second group took months to become visible.
Who Was in This Study
The sample is not the general public. It was 86.7 percent white and 76.6 percent college educated. Twenty-nine percent had household income above 200,000 dollars.
It also represents roughly 5 percent of everyone who received state-regulated psilocybin services in Oregon during that window. People who agree to a year of follow up surveys are not a random draw.
There is no control group either. The design is single arm and observational, so it cannot establish that psilocybin caused the improvements.
What This Means for Any Clinic Running Sessions
The transferable lesson is about follow up architecture, not about psilocybin. If your worst outcomes surface at month three, your last patient contact cannot be week two.
Most infusion practices have no three month touchpoint. Neither does most of the psychedelic services industry. A single scheduled check in at 90 days would have caught every late signal in this study.
Screening deserves a second look too. First session patients carried all four serious reactions here. That is a reason to staff a first session differently than a fourth.
What Happens Next
The cohort is being followed to 6, 9 and 12 months. This paper reports the three month results only. The longer data will show whether the late signal keeps climbing or settles.
Oregon remains the country’s longest running regulated program, and more than 20,000 people have passed through it. Other states are drafting from its template, including Minnesota.
The clinical evidence is moving in parallel. A federal phase 3 trial in veterans began enrolling this month. Readers weighing a legal session can start with our guide to psilocybin retreats and legal access.
This article is for informational purposes and is not medical advice. Psilocybin remains a Schedule I substance under federal law. Oregon’s program operates under state law only.
