In April 2026, the White House did something that would have been unthinkable a decade ago: it put the federal government behind psychedelics. President Trump signed an executive order directing the FDA to fast-track the review of compounds like psilocybin and MDMA, steering roughly $50 million toward expanding access for veterans and telling drug-enforcement agencies to loosen research restrictions [source: The White House, 2026]. Weeks later, the Departments of Veterans Affairs and Health and Human Services signed an agreement to coordinate psychedelic research for service members [source: Military Times, 2026]. Suddenly, drugs that remain illegal under federal law are being framed as a mental-health breakthrough.
And yet, less than two years earlier, the FDA had rejected the most advanced psychedelic therapy in the pipeline. That whiplash — federal enthusiasm running ahead of a federal rejection — is the story worth understanding. This is a piece about mental-health drug therapy and its regulation, not about weight-loss drugs, vaccines, or the social side of loneliness. The question is narrower and harder: when a psychedelic seems to help, how do we know it actually does?
Why the world is suddenly talking about psychedelic therapy
Psychedelic-assisted therapy pairs a controlled dose of a psychoactive drug — most often psilocybin (the active compound in "magic mushrooms") or MDMA (the drug also known as ecstasy) — with structured psychotherapy sessions before, during, and after the dose. The target conditions are among the hardest in psychiatry: treatment-resistant depression, meaning depression that has not responded to multiple standard antidepressants, and post-traumatic stress disorder (PTSD).
The interest is not just cultural. Millions of people do not respond to existing treatments, and veterans in particular face high rates of PTSD and suicide, which is why so much of the 2026 policy push is framed around them [source: Military Times, 2026]. When early trials reported large improvements in conditions that standard drugs often fail to touch, the field drew serious money, serious scientists, and serious regulatory attention. The task now is to separate what the trials measured from what enthusiasts hope they mean.
What the trials actually measured
Start with the numbers, not the vibes. For MDMA-assisted therapy in PTSD, two phase 3 trials produced the headline results. In the first, published in Nature Medicine, 67% of participants who received MDMA with therapy no longer met the diagnostic criteria for PTSD at the primary endpoint, compared with 32% in the placebo-plus-therapy group [source: Nature Medicine, 2021]. The confirmatory trial reported that 71.2% no longer had a PTSD diagnosis versus 47.6% on placebo, with 46.2% reaching remission compared with 21.4% [source: Nature Medicine, 2023]. Those are large gaps, and they earned MDMA a Breakthrough Therapy designation from the FDA back in 2017 [source: MAPS, 2017].
Psilocybin for depression tells a more measured story. In the largest published trial, a phase 2b study in the New England Journal of Medicine, a single 25 mg dose alongside psychological support produced a 37% response rate and a 29% remission rate at three weeks — better than the 10 mg and 1 mg comparison doses, but with roughly one in five patients still responding at week 12 [source: NEJM, 2022]. In 2025 and early 2026, the developer COMPASS Pathways reported the first phase 3 efficacy data for a classic psychedelic: a single dose beat placebo on the standard depression scale by about 3.6 points, and a two-dose regimen by about 3.8 points, both statistically significant [source: COMPASS Pathways, 2025] [source: COMPASS Pathways, 2026].
Here the distinction between "felt" and "measured" matters. A 3.6-to-3.8-point difference on a depression rating scale is real and statistically significant, but it is modest — smaller than the dramatic transformations that dominate media coverage. Clinicians genuinely disagree over whether it clears the bar of being clinically meaningful. The measured effect and the felt promise are not the same size.
Correlation, causation, and the blinding problem
The deeper issue is not whether patients improve — many clearly do — but whether we can attribute that improvement cleanly to the drug. Psychedelics create unmistakable subjective effects. You know, unavoidably, whether you have taken 25 mg of psilocybin or a sugar pill. That breaks the foundation of a randomized controlled trial: blinding.
The data on this are stark. A systematic review in JAMA Psychiatry found that psilocybin, LSD, and ayahuasca studies frequently reported blinding-failure rates above 90% among both participants and raters, and even MDMA trials with inert placebos exceeded 85% [source: JAMA Psychiatry, 2025]. A follow-up analysis went further, concluding that psychedelic-therapy trials "are effectively always open label," and that when compared under equal-unblinding conditions against openly labeled antidepressant therapy, the psychedelic advantage was not statistically significant [source: JAMA Psychiatry, 2026].
This is a textbook correlation-versus-causation problem. When people know they received the active drug, expectancy — hope, belief, the sense that something powerful is happening — inflates how much benefit they report [source: JAMA Psychiatry, 2024]. The improvement is genuinely reported, but part of it may flow from expectation and from many hours of intensive therapy rather than from the molecule itself. None of this proves psychedelics do not work. It means the cleanest question — how much of the benefit is the drug — remains genuinely unsettled.
The FDA's 2024 rejection of MDMA — and what it exposed
This is where company claims collided with regulatory scrutiny. In June 2024, an independent FDA advisory committee voted 2–9 that the available evidence did not show MDMA was effective for PTSD, and 1–10 that its benefits outweighed its risks under the proposed safeguards [source: STAT, 2024]. Two months later, the FDA issued a Complete Response Letter to the sponsor, Lykos Therapeutics, declining approval and asking for another phase 3 trial [source: AJMC, 2024].
The panel's objections were revealing. They centered less on the molecule than on how the trials were run: functional unblinding, gaps in safety data, questions about durability, and concerns about therapist conduct [source: STAT, 2024]. The problems compounded quickly. One day after the rejection, the journal Psychopharmacology retracted three papers on MDMA-assisted therapy, citing "protocol violations amounting to unethical conduct" at a Canadian study site [source: BioPharma Dive, 2024]. In the aftermath, Lykos laid off roughly 75% of its staff and its CEO stepped down [source: FierceBiotech, 2024]. In September 2025, the FDA released the rejection letter publicly under a new transparency initiative [source: Psychedelic Alpha, 2025].
The lesson is the third layer of this whole story: a developer announcing positive results is not the same as a regulator or peer reviewers accepting them. Promising trial data and an approvable drug are different things, and in 2024 the gap between them was the whole story.
Psilocybin's separate track
It would be a mistake to read the MDMA rejection as a verdict on all psychedelics. Psilocybin is on its own regulatory path. It holds Breakthrough Therapy designations — COMPASS Pathways for treatment-resistant depression in 2018, and the nonprofit Usona Institute for major depressive disorder in 2019 [source: Medscape, 2019]. Unlike MDMA, psilocybin now has phase 3 data, and both of COMPASS's late-stage trials hit their primary endpoints, with effects the company says lasted through 26 weeks after one or two doses [source: COMPASS Pathways, 2026].
But psilocybin inherits the same blinding problem, and its measured effect sizes are modest. Its trials also documented real risks: in the phase 2b study, adverse events occurred in 77% of participants, and instances of suicidal ideation were reported across dose groups [source: NEJM, 2022]. Being further along the evidence ladder than MDMA is not the same as being proven. The honest summary is that psilocybin is a promising, breakthrough-designated, phase-3-tested therapy that is still investigational — not an approved medicine.
A patchwork of access: states vs. federal law
Meanwhile, access is running ahead of federal approval through state programs — and it is easy to misread what these actually offer. Oregon, under a 2020 ballot measure, built the first state framework for supervised psilocybin services; its first licensed center opened in 2023, and by March 2025 more than 21,000 psilocybin products had been administered [source: CBS News, 2025]. Colorado followed with a 2022 measure, began licensing facilitators in January 2025, and held its first state-regulated session in Denver in June 2025 [source: Stateline, 2025] [source: Colorado Public Radio, 2025].
These are not pharmacies and not prescriptions. Both programs allow only supervised administration with a licensed facilitator; you cannot buy psilocybin to take home, and it is not being dispensed as an FDA-approved medicine. Crucially, they exist under state law only. Psilocybin and MDMA remain Schedule I under federal law, meaning they are federally illegal to manufacture, possess, or distribute — outside of these narrow state programs or an authorized clinical trial. The map of "legal access" is a patchwork, and its legal ground is genuinely contested.
Hype, evidence, and what this is not
So where does that leave a reader trying to make sense of the headlines? With a few honest distinctions. The evidence for psychedelic therapy is promising but unproven: real benefits are reported, effect sizes for depression are modest, the blinding problem makes causal attribution hard, and the single most advanced program was rejected by regulators in 2024. Federal enthusiasm in 2026 is a policy and funding decision, not a scientific verdict.
This article is not medical advice, and it is important to be explicit: outside of authorized clinical trials and a small number of supervised state programs, these substances are illegal and investigational. Psychedelics are not a do-it-yourself treatment. They are being studied precisely because they carry real psychological risks, require careful screening, and are delivered with intensive professional support in controlled settings. Self-medicating with unregulated substances is not what any of the trials in this article studied. Anyone struggling with depression, PTSD, or suicidal thoughts should seek qualified medical care and local crisis resources.
What to watch
Three things will tell us where this is heading. First, whether COMPASS's phase 3 psilocybin data survive peer review and FDA scrutiny — and whether regulators judge a roughly 3.6-to-3.8-point effect clinically meaningful once the blinding caveats are weighed. Second, whether MDMA gets a second act: a new, more rigorously run phase 3 trial that answers the 2024 objections, now backed by fresh federal and Department of Defense funding [source: Military Times, 2026]. Third, whether the state-by-state access experiments in Oregon and Colorado generate real-world safety data that either reassure or worry regulators.
The honest posture is neither dismissal nor hype. Psychedelic therapy has produced some of the most striking early results in modern psychiatry and some of its thorniest methodological problems, at the same time. Watch the effect sizes, watch the blinding, and watch what regulators and peer reviewers actually endorse — not just what gets announced.