Three independent 2026 surveys agree psilocybin is the most-used psychedelic among U.S. adults, with past-year estimates from 7.8 million (NSIHT 2025, Denver Health) to 11.06 million (RAND Health Quarterly 2026) to roughly 8 million (NSDUH 2024, published in the American Journal of Psychiatry). Those are three different survey designs, not a trend line. All three measure unsupervised, mostly non-clinical use and none measures clinical outcomes. The best real-world outcome data — a 346-person uncontrolled Oregon cohort in JAMA Network Open, August 2026 — found large within-group symptom improvements alongside serious behavioral adverse reactions in 1.2% of sessions, and facilitators recorded only one of those four events. The lead author says the results should not be used to claim psilocybin treats depression or anxiety. Meanwhile the only Phase 3 program near an FDA filing has no peer-reviewed publication, its headline effect is a 3.6-point MADRS difference, and psilocybin remains Schedule I federally.
The short answer
Yes — three national surveys released between April and October 2026 independently put psilocybin ahead of MDMA, LSD and ketamine among U.S. adults. All three measure self-reported, mostly unsupervised use. None of them measures whether it worked, and psilocybin is still Schedule I federally.
DoubleBlind's October 12, 2026 piece gets the direction right: psilocybin is America's psychedelic of choice. The surveys agree, and they agree despite using very different methods, which is the strongest thing you can say about any survey finding.
The part worth your attention is what those surveys are and are not. They are population counts of people who say they used a drug in the past year. They are not evidence of benefit, they contain no information about dose or potency, and almost none of the use they capture happened inside a licensed program or a trial.
So you end up with two curves moving at different speeds. Use is climbing fast. The evidence base — the actual controlled human data — is climbing slowly, and the single best real-world study published this year is an uncontrolled cohort whose own lead author says it cannot show psilocybin treats anything.
That gap is the story. Everything below is detail on it.
Three surveys, three numbers, and why that is not a trend line
NSIHT 2025 says 7.8 million adults; RAND's probability-panel survey says 11.06 million; the federal NSDUH 2024 says 2.8% of people 12 and older, roughly 8 million. The spread is methodological, not a correction or a trajectory.
NSIHT 2025, run by Denver Health's Rocky Mountain Poison & Drug Safety and released October 7, 2026, found about 14 million U.S. adults — roughly 1 in 19 — used at least one of 13 tracked psychedelic, hallucinogenic or dissociative substances in the past year. Psilocybin led at 7.8 million adults (2.9%), up from 2.1% in 2024, ahead of MDMA (4.2 million), prescription ketamine or esketamine (3.3 million) and LSD (2.6 million). NSIHT surveys more than 10,000 adults a year who used at least one of 17 such substances, using enriched sampling of recent users fused with a large general-population survey. It is funded by NIDA (Award No. R01DA064612), with SAMHSA collaborating. https://www.denverhealth.org/news/2026/10/magic-mushrooms-more-popular-ever
RAND published the peer-reviewed version of its 2025 Psychedelics Survey in *RAND Health Quarterly* (Priest, Kilmer, Senator and Setodji, June 26, 2026). Using the NORC AmeriSpeak probability panel — n=10,122, fielded September 9 to October 1, 2025, AAPOR RR3 of 26.3%, margin of error ±1.33 points — it estimated 11,059,000 past-year psilocybin users (4.3%, 95% CI 9.40M–12.72M), again the highest of any psychedelic. https://pmc.ncbi.nlm.nih.gov/articles/PMC13308765/
The federal NSDUH asked psilocybin-specific questions for the first time in 2024. Yang, Eun and Palamar published the analysis in the *American Journal of Psychiatry* (epub April 22, 2026; n=58,633, ages 12+): 2.8% past-year use, about 8 million people, with higher odds among 18-to-25-year-olds, males, White respondents, and people reporting a past-year major depressive episode. https://pubmed.ncbi.nlm.nih.gov/42014961/
Different sampling frames, different age ranges, different question wording. Do not read 7.8M → 8M → 11M as a line going up. (Note too that DoubleBlind's figures drift slightly from the published ones: it cites 225 million psilocybin use-days and 10 million microdosers where the peer-reviewed RAND report says about 216 million and about 9.5 million. Use the published numbers.)
The one dataset built to show change over time is the multisource study in *Annals of Internal Medicine* (2025, Denver Health/RMPDS authors with SAMHSA co-authors), pooling five national datasets: adult lifetime use rose from 10.0% (~25 million) in 2019 to 12.1% (31.3 million) in 2023, while poison center calls from 2019 to 2023 rose 201% among adults, 317% among adolescents and 723% among children under 12. UVA researchers reported in the *Journal of Adolescent Health* that psilocybin calls for ages 13–19 went from 152 in 2018 to 464 in 2022. Toxicologist Christopher Holstege's own caveat matters: calls are not prevalence, because "not everyone is calling us." And the surveillance hole is real — only three psilocybin poisonings appeared in ED and outpatient records from 2015 to 2021.
Most of this is not therapy, and most of it is small doses
About 80% of psilocybin, MDMA and LSD users in NSIHT 2025 cited fun, enjoyment or celebration as at least one reason for use. RAND estimates about 9.5 million adults microdosed psilocybin, MDMA or LSD in the past year — but microdosing accounts for only around 8% of total psilocybin consumed.
The framing that psilocybin's rise is a mental-health phenomenon doesn't survive contact with the survey data. In NSIHT 2025, roughly 80% of psilocybin, MDMA and LSD users cited fun, enjoyment or celebration as at least one reason for using. Forty-five percent of psilocybin users also used cannabis — relevant, because combining the two changes the experience in ways nobody has characterized well (/guides/psilocybin-and-cannabis). Nearly a quarter of psilocybin users, 1.8 million people (22.7%), were new within the year.
On self-reported outcomes: nearly half of psilocybin users said their mental health got somewhat or much better, and 4.6% said it got worse. Comparable perceived improvement was 47.3% for prescription ketamine/esketamine and 43.1% for LSD. That is perceived change with no comparison group, no blinding and no baseline measurement — closer to a satisfaction rating than an outcome.
Side effects were common and mostly physical. Among people who had used in the prior three months, at least one side effect was reported by 72.6% of psilocybin users, 77.2% for MDMA, 82.1% for LSD and 68.8% for prescription ketamine/esketamine. The most frequent were headache, nausea or vomiting, and fatigue.
RAND's microdosing numbers are the ones most often repeated and least often caveated. About 9.5 million adults (3.7%) microdosed psilocybin, MDMA or LSD in the past year; among past-year users, 69% of psilocybin users microdosed at least once. Of roughly 216 million psilocybin use-days, about 102 million (47%) involved microdosing — yet the authors estimate microdosing accounts for only about 8.2% of total psilocybin consumed, under stated assumptions. The unweighted samples are small (psilocybin n=372, MDMA n=163, LSD n=90), the authors note microdosing is self-reported and "can be difficult to define," and neither NSDUH nor NSIHT asks about it at all. Our own read on the evidence base: /guides/microdosing.
The best real-world data is an uncontrolled Oregon cohort
Korthuis et al., JAMA Network Open, August 19, 2026: a single-arm prospective cohort of 346 adults at Oregon service centers found large within-group symptom reductions at three months and serious behavioral adverse reactions in 1.2% of sessions. No control group means no causal claim.
This is the study that should anchor any honest conversation about what regulated psilocybin looks like in practice. Korthuis, Cook, Gregoire and colleagues enrolled 346 adults across 24 of Oregon's 26 active service centers with 83 facilitators (mean age 49.5; 52.5% women; 89.0% White; 44.2% psychedelic-naive; mean dose 31.9 mg, range 2–50 mg; 90.2% retention at three months). https://jamanetwork.com/journals/jamanetworkopen/fullarticle/10.1001/jamanetworkopen.2026.30608
At three months, within-group: moderate-to-severe depression on the PHQ-9 fell from 42.2% to 16.5% (RR 0.40, 95% CI 0.32–0.50); anxiety on the HADS from 45.1% to 13.2% (RR 0.29); PTSD on the PCL-6 from 48.0% to 16.8% (RR 0.35); well-being rose 3.21 points on the S-WEMWBS; life satisfaction went from 63.0% to 82.3%.
Now the harms, because this is the part the coverage skips. Serious behavioral adverse reactions requiring medical attention occurred in four sessions (1.2%). All four participants were psychedelic-naive and were seeking treatment for mental health symptoms. There were no serious medical adverse reactions. Facilitators recorded only one of the four events — which tells you something about how much to trust facilitator-reported dashboard counts anywhere. Self-reported harm was 1.6% at one week, 1.9% at one month, 2.3% at three months. Past-30-day HPPD symptoms were reported by 10.0% at one month and 8.7% at three months, with one participant in significant distress. Thoughts of dying or suicide rose from 1.0% at one week to 1.9% at three months.
The authors state outright that the uncontrolled design "precludes causal inference." Only about 5% of Oregon clients were sampled, psychiatric diagnoses were unknown, baseline symptoms were relatively low, and the findings do not generalize to unsupervised use. A September 2026 correction fixed a coding error and changed "statistically significant" to "clinically significant" regarding the serious adverse reactions. Funding came from SAMHSA, the OHSU Foundation, NIH/NIDA and NCATS; co-author Luoma disclosed personal fees from the OHSU Foundation, ownership of the Portland Institute for Psychedelic Science, and personal fees from Atai Life Sciences.
Korthuis told OPB that the results should not be used to claim psilocybin effectively treats depression or anxiety. When the lead author says that, believe him. https://www.opb.org/article/2026/08/19/psilocybin-oregon-use-ohsu-study-results/
Screening is the part that actually travels
Korthuis named lithium, active suicidality or psychosis, and pregnancy as reasons not to take psilocybin. Every Oregon center in a 2026 mystery-shopper study asked about psychiatric history and medications — and the median session cost $1,900 with no insurance accepted.
If 11 million people are using psilocybin and roughly none of them are in a trial, then screening is the only transferable finding in this entire evidence base.
Adams, Nicol, Cabassa and Xu published a mystery-shopper study of Oregon service centers in *Health Affairs Scholar* on October 12, 2026 (doi 10.1093/haschl/qxag231). Scripted calls reached 15 of 22 active centers. Median session cost: $1,900. No center accepted insurance. No center allowed unsupervised or take-home use. All of them asked about psychiatric history, medications and contraindications; nearly every one flagged psychoactive medications that should not be combined with psilocybin; and none advised callers on antidepressants to stop without first consulting a physician. That last detail is the right answer, and it is worth copying.
On specific interactions, here is where the evidence actually sits:
- SSRIs — one human RCT, healthy volunteers only. Becker and colleagues (*Clinical Pharmacology & Therapeutics*, 2022) pretreated 14 days with escitalopram before 25 mg psilocybin. Escitalopram *reduced* bad drug effects, anxiety and adverse cardiovascular effects without blunting positive mood, and did not alter psilocin pharmacokinetics. Healthy subjects, short pretreatment; the authors call for work in clinical populations. https://pmc.ncbi.nlm.nih.gov/articles/PMC9299061
- Lithium — anecdote tier, and alarming anyway. Nayak et al. (*Pharmacopsychiatry*, 2021) analyzed 62 online reports of psychedelic use alongside lithium from Erowid, Shroomery and Reddit: 47% involved seizures, 39% involved medical attention. None of 34 lamotrigine reports involved seizures. This is self-reported internet data, not a clinical study, and the authors conclude "provisionally." It is also the only signal anyone has, and Korthuis independently names lithium as a reason not to proceed. /guides/psilocybin-and-lithium
- Seizure threshold — n=1. Blond and Schindler (*Frontiers in Neurology*, 2023) captured 32 "long episodes" in one day on intracranial electrocorticography in a 31-year-old man with refractory temporal epilepsy after about 20 mg psilocybin, against a baseline of 0–2. He did not recognize them as seizures. One case cannot establish causation. https://pmc.ncbi.nlm.nih.gov/articles/PMC10343433/
For MAOIs and anything else in your medicine cabinet, the honest instruction is a prescriber, not a forum: /guides/psilocybin-drug-interactions and /guides/safe-trip. We don't publish dosing guidance beyond /dosage.
The legal reality has not moved as fast as the use curve
Psilocybin remains Schedule I under federal law and no psilocybin drug is FDA-approved. Oregon is the only state operating supported adult use; Colorado is licensing; New Mexico has opened no patient enrollment.
Federally, psilocybin is Schedule I at 21 CFR 1308.11(d)(29) (DEA Code 7437), with psilocyn at (d)(30). https://www.law.cornell.edu/cfr/text/21/1308.11
On April 24, 2026, the FDA granted Compass Pathways a rolling NDA review and selected COMP360 for the Commissioner's National Priority Voucher program for treatment-resistant depression, which offers a shortened one-to-two-month review *after* an NDA is filed. Compass's own language warns the voucher "may not actually lead to a faster FDA review or approval process," that the program does not change safety and efficacy standards, and that approval efforts "may be unsuccessful"; CEO Kabir Nath said any NDA "must still meet FDA's established standards." Usona Institute received a voucher the same day for psilocybin in major depressive disorder and said much the same thing. As of this writing we have found no evidence any psilocybin approval has occurred. https://ir.compasspathways.com/News--Events-/news/news-details/2026/Compass-Pathways-Announces-FDA-Granted-NDA-Rolling-Review-Request-and-Awarded-Commissioners-National-Priority-Voucher/default.aspx
And here is the thing nobody says loudly enough: there is still no peer-reviewed publication of the Phase 3 data. Everything public comes from press releases, SEC filings and conference reporting. COMP005, the randomized, double-blind, placebo-controlled trial of a single 25 mg dose across 32 U.S. sites with 258 participants dosed, reported a mean MADRS treatment difference of −3.6 points at week 6 (95% CI −5.7 to −1.5, p<0.001). The company's own filing calls the data preliminary, notes the FDA had not reviewed them, and warns that "full results and safety data may not be consistent with the preliminary results." https://www.sec.gov/Archives/edgar/data/1816590/000181659025000085/cmps-20250623.htm
3.6 points is statistically clean and clinically modest. COMP006's 26-week data (announced July 7, 2026) put about 39% of the 25 mg arm at a ≥25% MADRS reduction by week 6, sustained through week 26, versus 25% in COMP005 after a single dose. The 52-week COMP005 Part C figure of roughly 13 points from baseline is open-label topline with no control arm. An investigator in the trial, Boadie Dunlop, noted granular data beyond the press release were not yet available.
At the state level: Oregon is the only state running supported adult use, under ORS 475A, with mandatory client data reporting since January 1, 2025 (/legal-status/oregon). Colorado splits licensing between the Department of Revenue's Natural Medicine Division and DORA, issuing its first healing center license on March 31, 2025 (/legal-status/colorado, and /vs/oregon-vs-colorado-psilocybin-therapy). New Mexico's Medical Psilocybin Act took effect June 20, 2025, with the Department of Health saying the program is "set to be implemented by December 31, 2026" — no patient enrollment has opened (/legal-status/new-mexico).
Public opinion is further behind than the use numbers suggest. RAND's February 24, 2026 report found about 23% of U.S. adults support legal psilocybin use, versus roughly 10% for LSD and MDMA and 65% for cannabis — and among supporters, only 42% think adults should be able to use it for any reason.
PsyBear takeaways
Psilocybin is the most-used psychedelic in America and the least-understood at the scale it is being used. Popularity is not evidence, perceived improvement is not an outcome, and the harm signal in the best cohort we have was undercounted by the people running the sessions.
1. The surveys agree on rank, not on magnitude. 7.8 million, 8 million and 11.06 million come from three different designs. Cite whichever you like, but do not chain them into a trend.
2. "Nearly half said their mental health improved" is not a clinical finding. No baseline, no control, no blinding. In the same survey, 4.6% said it got worse and 72.6% of recent users reported at least one side effect.
3. The only real-world outcome study is uncontrolled, and its authors say so. The Oregon cohort's 1.2% rate of serious behavioral adverse reactions is the number to carry forward — along with the fact that all four affected participants were psychedelic-naive and treatment-seeking, and facilitators logged only one of the four. Any state dashboard built on facilitator reporting is probably undercounting.
4. HPPD showed up at 10% at one month. Mostly mild, one participant in significant distress. That is in a supervised, screened, facilitator-attended setting — not a festival. Worth knowing before you decide: /guides/long-term-effects-of-psilocybin.
5. Screening and medication review are the transferable findings. Lithium, active suicidality or psychosis, and pregnancy are Korthuis's own exclusions. Every Oregon center asked about psychiatric history and medications; none told a caller to quit their antidepressant on the phone. Match that standard or don't proceed: /guides/trip-preparation, /guides/safe-trip.
6. Legal access is narrow and expensive. Median $1,900 per session in Oregon, no insurance, no take-home. Supervised state programs exist in two states; New Mexico has enrolled nobody yet.
7. The Phase 3 case is thinner than the headlines. A −3.6-point MADRS difference, unpublished in any peer-reviewed journal as far as we can find, with the sponsor itself cautioning that final results may differ. A national priority voucher speeds review; it does not lower the bar.
Psilocybin remains Schedule I under U.S. federal law outside licensed state programs and authorized research. Nothing here is medical advice or an endorsement of unsupervised use, and nothing here supports a claim that psilocybin treats depression, anxiety or PTSD. If you take lithium, an SSRI, an MAOI or any psychoactive medication, or if you have a personal or family history of psychosis or bipolar disorder, talk to a prescriber before anything else. Start at /guides/psilocybin-drug-interactions.
Key Takeaways
Three independent 2026 surveys agree psilocybin is the most-used psychedelic among U.S. adults, with past-year estimates from 7.8 million (NSIHT 2025, Denver Health) to 11.06 million (RAND Health Quarterly 2026) to roughly 8 million (NSDUH 2024, published in the American Journal of Psychiatry). Those are three different survey designs, not a trend line. All three measure unsupervised, mostly non-clinical use and none measures clinical outcomes. The best real-world outcome data — a 346-person uncontrolled Oregon cohort in JAMA Network Open, August 2026 — found large within-group symptom improvements alongside serious behavioral adverse reactions in 1.2% of sessions, and facilitators recorded only one of those four events. The lead author says the results should not be used to claim psilocybin treats depression or anxiety. Meanwhile the only Phase 3 program near an FDA filing has no peer-reviewed publication, its headline effect is a 3.6-point MADRS difference, and psilocybin remains Schedule I federally.
FAQ
- How many Americans used psilocybin in the past year?
- It depends on which survey you read, and the three major 2026 estimates were built differently. NSIHT 2025 (Denver Health, released October 7, 2026) put it at 7.8 million adults, or 2.9%. RAND's probability-panel survey, published in RAND Health Quarterly in June 2026, estimated 11,059,000 adults (4.3%, 95% CI 9.40M–12.72M). The federal NSDUH 2024, analyzed in the American Journal of Psychiatry, found 2.8% of people aged 12 and older, roughly 8 million. All three ranked psilocybin ahead of MDMA, LSD and ketamine.
- Is psilocybin legal in the United States now?
- No. Psilocybin remains Schedule I under federal law at 21 CFR 1308.11(d)(29), with psilocyn at (d)(30), and no psilocybin drug is FDA-approved. Oregon is the only state operating a supported adult use model under ORS 475A. Colorado is licensing healing centers and facilitators, having issued its first healing center license on March 31, 2025. New Mexico's Medical Psilocybin Act took effect June 20, 2025, but no patient enrollment has opened.
- How often do serious problems happen in a supervised psilocybin session?
- In the 346-person Oregon cohort published in JAMA Network Open on August 19, 2026, serious behavioral adverse reactions requiring medical attention occurred in four sessions, or 1.2%. All four participants were psychedelic-naive and were seeking treatment for mental health symptoms, and there were no serious medical adverse reactions. Notably, facilitators recorded only one of those four events. Past-30-day HPPD symptoms were reported by 10.0% of participants at one month and 8.7% at three months.
- Does the Oregon study prove psilocybin treats depression?
- No. It was a single-arm prospective cohort with no control group, and the authors state the design "precludes causal inference." Lead author P. Todd Korthuis told OPB directly that the results should not be used to claim psilocybin effectively treats depression or anxiety. Only about 5% of Oregon clients were sampled, psychiatric diagnoses were unknown, baseline symptoms were relatively low, and the findings do not generalize to unsupervised use.
- What does the research say about mixing psilocybin with antidepressants or lithium?
- For SSRIs there is one human RCT in healthy volunteers: Becker et al. (Clinical Pharmacology & Therapeutics, 2022) found 14 days of escitalopram before 25 mg psilocybin reduced bad drug effects, anxiety and adverse cardiovascular effects without altering psilocin pharmacokinetics — but that was healthy subjects, not patients. For lithium, the only data is anecdotal: Nayak et al. (Pharmacopsychiatry, 2021) reviewed 62 online reports, of which 47% involved seizures and 39% involved medical attention. Korthuis names lithium, active suicidality or psychosis, and pregnancy as reasons not to take psilocybin. Talk to a prescriber.
- How much does a legal psilocybin session cost?
- A mystery-shopper study of Oregon service centers published in Health Affairs Scholar on October 12, 2026 reached 15 of 22 active centers and found a median session cost of $1,900, with no center accepting insurance. No center allowed unsupervised or take-home use, and all asked about psychiatric history, medications and contraindications.