The psychedelic safety conversation has spent years trying to escape two equally lazy slogans.
One says psychedelics are basically harmless because they are natural, ancient, spiritual or currently fashionable.
The other says psychedelics are dangerous because they are drugs, full stop, thank you for attending the 1987 assembly.
A new paper in JAMA Network Open makes both positions harder to defend.
Researchers analyzed a nationally representative survey of U.S. adults who reported recent use of psilocybin, LSD, MDMA, ibogaine or 5-MeO-DMT. The study did not ask whether psychedelics are “good” or “bad.” It asked people what health outcomes they themselves attributed to their recent use.
That distinction matters.
The headline numbers are uncomfortable
Among people who had used each substance in the previous three months, the proportion reporting at least one of 13 adverse effects was:
- 75.6% for psilocybin
- 83.6% for LSD
- 81.4% for MDMA
- 94.8% for ibogaine
- 92.8% for 5-MeO-DMT
Those are not tiny numbers.
Ibogaine and 5-MeO-DMT stood out most strongly. For example, 44.2% of respondents in the ibogaine group and 51.3% in the 5-MeO-DMT group attributed heart palpitations, chest pain or chest pressure to their use. Thoughts of death or suicide were reported by 59.3% of the ibogaine group and 36.6% of the 5-MeO-DMT group.
If you stop reading there, however, you can manufacture a frightening headline very quickly.
Please do not.
This study does not prove that the drugs caused those outcomes
The study is cross-sectional. It relies on people recalling experiences and deciding for themselves whether a particular symptom was caused by the substance they took.
That is useful information, but it is not the same thing as a controlled clinical trial establishing causality.
The groups also differed in important ways before anyone tried to compare them. Participants varied in psychiatric history, patterns of other substance use, frequency of use and whether the drug was taken in a therapeutic or clinical environment.
And the ibogaine and 5-MeO-DMT groups were small: 74 and 95 people respectively reported past-year use, with even smaller numbers reporting use in the previous three months. Small samples produce much wider uncertainty around the percentages.
The researchers themselves are explicit about these limitations.
In other words: this paper does not demonstrate that 59.3% of people who take ibogaine will develop suicidal thoughts. It reports that, within a small group of recent ibogaine users in this survey, that proportion said they had experienced and attributed such thoughts to the substance.
Those are very different statements.
The internet is traditionally excellent at removing that sentence from the conversation.
But “not causal” does not mean “ignore it”
There is another bad habit in psychedelic discourse: treating every limitation as a reason to dismiss an inconvenient finding.
That would also be a mistake.
Real-world safety surveillance matters precisely because clinical trials do not represent every person, every dose, every setting or every combination of substances found outside a research center.
Ibogaine already has recognized cardiac concerns, including effects on cardiac electrical activity and the risk of arrhythmias. The fact that this survey also found comparatively high self-reported cardiac symptoms among ibogaine users is therefore not meaningless simply because the study cannot prove pharmacological causation.
Likewise, an extremely intense 5-MeO-DMT experience may generate psychological or physiological responses very different from what occurs with psilocybin mushrooms at a moderate dose.
“Classical psychedelic” is a useful scientific grouping. It is not a magic spell that makes every substance interchangeable.
The paper also found benefits
This is where the study gets more interesting than the panic-or-praise format.
Participants reported positive outcomes too.
Mental-health improvement attributed to the substance was most common among psilocybin respondents, at 51.2%. Physical-health improvement was reported most often among 5-MeO-DMT respondents, at 28.5%.
That does not establish therapeutic efficacy either.
It does show why the safety conversation becomes distorted when people insist on choosing one side of the ledger.
The same population can report meaningful benefit and meaningful harm.
The same substance can be transformative for one person, destabilizing for another and simply strange for a third.
That should not be controversial. Somehow, in psychedelics, it still occasionally is.
Psilocybin looked safer here, but “safer” is not “safe”
Psilocybin produced the lowest overall proportion reporting at least one adverse effect among the five substances studied.
It also had the largest user base in the survey, meaning the estimated total number of Americans experiencing some adverse outcomes could still be substantial even if the individual rate is lower.
This is the difference between relative risk and population impact.
A common substance with a lower rate of problems can affect more people overall than a rare substance with a much higher rate.
The paper estimates past-year use at roughly 8.1 million U.S. adults for psilocybin, compared with about 560,000 for ibogaine and 740,000 for 5-MeO-DMT.
So “which drug had the highest percentage?” is not the only useful question.
We should also ask: how many people are using it, in what circumstances, with what screening, and what happens when things go wrong?
The safety conversation is getting more adult
For a field trying to become medicine, this is progress.
Medicine is not built by proving that your favorite molecule is morally good.
It is built by identifying benefits, contraindications, dose effects, vulnerable populations, interactions, adverse events and the conditions under which risk changes.
That is less romantic than announcing a revolution.
It is also what a mature field looks like.
The researchers conclude that their findings support stronger safety surveillance, targeted harm reduction and balanced public communication about both risks and benefits.
“Balanced” is doing a lot of work there.
It means resisting prohibition-era fearmongering without replacing it with wellness-era mythology.
It means admitting that a profound experience can be valuable without pretending intensity itself is evidence of safety.
And it means taking adverse experiences seriously without automatically turning them into arguments against all psychedelic research.
The useful conclusion is not “psychedelics are dangerous”
The useful conclusion is that psychedelic safety is substance-specific, person-specific and context-specific.
Psilocybin is not ibogaine.
Ibogaine is not 5-MeO-DMT.
A screened clinical participant is not the same population as someone combining substances in an uncontrolled setting.
And a survey finding is not the same thing as a causal estimate.
The interesting part of this new JAMA paper is not that it gives us a clean answer.
It is that it makes the clean answers look less believable.
That is probably healthy.
Sources
- Simonsson O, Bradley M, Black JC, et al. Adverse Health Outcomes of Psilocybin, LSD, MDMA, Ibogaine, and 5-MeO-DMT. JAMA Network Open. Published September 29, 2026. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2854616
- JAMA Network media release: https://media.jamanetwork.com/media-releases/10.1001/jamanetworkopen.2026.36756
Key concepts
Brief definitions for terms used in this article.
psilocybin
A naturally occurring compound that the body converts to psilocin; its effects and risks depend on many factors.
More reading tools
No relevant article link has been confirmed yet. Browse the Library to continue exploring.
Your Reading Shelf
Saved only in this browser.