Roughly 1 in 8 U.S. adults takes an antidepressant. The peer-reviewed evidence on how well those medications actually work — including the trials that were quietly never published — is more sobering than most patients ever hear. This is what the data shows, why the incentives kept it quiet, and what the science is now pointing toward.
Live at psychedelic-connect.com/the-antidepressant-gap/
Here is a number that should be on the front page of every health section in America, and almost never is. In STAR*D — the largest, longest, government-funded study of
antidepressants ever conducted — researchers followed 4,041 real patients through medication after medication, doing everything the standard playbook prescribes. After a
year, the share who had gotten well and stayed well was about three percent. Three. Not thirty. Not thirteen. Three. That figure deserves a careful reading, because the honest version is actually more damning than the careless one. It does not mean only 3% of people ever felt better — many did, at least for a while. It means that of everyone who walked in the door, only
about 3% reached full remission and were still there twelve months later, without relapsing or dropping out. People climbed out of the hole. Then, over and over, they slid back in. Nearly 30 million American adults are taking these medications right now. A great many are being helped, and nothing here is an argument to stop taking anything. But a very large group is doing everything right and still not getting all the way better — and they have rarely been told how common their experience is, or why.
First, the Number Everyone Misreads
The distinction between “felt better once” and “stayed well” is the whole story. In STAR*D, about 37% reached remission after their first medication — roughly one in three (“response”: symptoms drop by at least half; “remission”: symptoms essentially gone). But of everyone who started, only about 3% were still well a year later without relapsing or dropping out.
Source: STAR*D (Rush et al., 2006; reanalysis Pigott et al., 2023). “3%” = 108 of 4,041 enrolled who reached remission and sustained it. It is not “only 3% improved.”
So the real indictment isn’t that antidepressants never work. It’s that the improvement so often doesn’t last — and that the standard of care has quietly accepted a revolving door of switching, relapsing, and switching again as normal.
How Weak the Evidence Really Is
STAR*D measured staying well. A separate body of research asks the blunter question: how far do antidepressants beat a sugar pill at all? In 2008, a team led by Erick Turner published a study in the New England Journal of Medicine that should have changed the conversation permanently. They pulled the complete set of antidepressant trials registered with the FDA — including the ones the public never saw.
- Appeared positive in published journals: 94%
- Actually positive in the FDA’s complete record: 51%
- Trials never published at all (~3,449 patients): 31%
Source: Turner EH, et al., New England Journal of Medicine, 2008 (74 FDA-registered trials). This is a documented finding about publication bias — which studies reached print — not an accusation against any single medication.
A landmark meta-analysis by Irving Kirsch pooled the complete FDA dataset and found an average improvement of just 1.8 points on the Hamilton depression scale over placebo — below the 3-point threshold the UK’s guideline body considers clinically meaningful. The drug only cleared that bar for the most severely depressed patients, and even then, the authors concluded, it was because placebo stopped working in severe cases — not because the medication worked better.
Follow the Incentives, Not the Conspiracy
It is tempting to reach for a villain here — but the truth is both more mundane and more powerful than a conspiracy. You don’t need anyone to be evil for this outcome. You just need the incentives pointing the way they point.
A medication a patient takes every day, potentially for decades, is one of the most valuable products a company can own — a protected molecule, a renewable prescription, a revenue stream that refills itself every month. A treatment a person undergoes a handful of times and then may not need again is, by that same logic, a far worse business. And a medicine that grows in the ground — a mushroom, a shrub’s root bark — can’t be patented the way a novel compound can, which means no one has a billion-dollar reason to fund the trials that would prove it works.
None of that makes antidepressants bad medicine, and it doesn’t make a mushroom magic. “Natural” is not a synonym for “safe” or “effective” — ibogaine in particular carries real cardiac risk and demands medical supervision. But it does explain why the most-prescribed treatment for one of the most common conditions in the country rests on a thinner evidence base than patients assume, while a genuinely promising alternative spent decades starved of the funding that would let it prove itself.
What the Science Is Finally Finding
Over the last decade, teams at Johns Hopkins, NYU, Stanford, and Imperial College London have tested whether psychedelic-assisted therapy can reach the people conventional care leaves behind. The early results earned the field FDA Breakthrough Therapy designations and Phase 3 trials — and they look nothing like a 1.8-point edge.
- MDMA-assisted therapy — no longer met PTSD criteria · Phase 3, 2021: 67% (vs. 32% placebo+therapy)
- Psilocybin-assisted therapy — depression remission · Johns Hopkins, 2021: 54%
- Psilocybin — smoking abstinence at 6 months · Johns Hopkins, 2014: 80%
- Standard antidepressant care — still well after 1 year · STAR*D: ~3%
Sources: MDMA — Mitchell et al., Nature Medicine 2021, Phase 3 (n=90). Psilocybin depression — Davis et al., JAMA Psychiatry 2021 (n=27). Psilocybin smoking — Johnson/Garcia-Romeu 2014 pilot (n=15). Early psychedelic trials are small; figures describe research populations, not any individual’s result — this is context, not a scoreboard.
To keep the ledger balanced: the largest analysis ever conducted (Cipriani et al., The Lancet, 2018 — 522 trials, over 116,000 patients) did find all 21 antidepressants studied outperformed placebo, though the effects were modest. Both things are true at once.
Why a Few Sessions Might Outlast a Daily Pill
The mechanism is still being worked out, and we won’t overstate it. Antidepressants are taken continuously to manage symptoms. The psychedelic model under study aims instead to produce change that persists after the medicine clears — and researchers point to measurable spikes in neuroplasticity, the brain’s capacity to rewire itself, as one candidate explanation. The honest headline was never “antidepressants failed.” It’s that a very large group is under-served by the current standard — and for the first time in a generation, rigorous research is pointing somewhere new.
Where This Leaves You
If you or someone you love has worked through medication after medication and still feels stuck, hold two truths at once. First: that experience is common, it is documented, and it is not a personal failing. Second: the alternatives now being studied are genuinely promising but still emerging, carry real medical and legal weight, and are not right for everyone. Ibogaine and 5-MeO-DMT in particular require careful medical screening, including cardiac evaluation, and should only ever be approached in a properly supervised setting.
Not sure where to start? Speak with an advisor who understands both the research and the landscape. Independent, confidential guidance, with no pressure and no obligation. Call (866) 435-7057 or visit psychedelic-connect.com.
Medical disclaimer. This editorial is for education only and is not medical advice, diagnosis, or treatment, and it is not a recommendation to start or stop any medication. Psychedelic substances carry real medical and legal risks, are not appropriate for everyone, and remain illegal in many jurisdictions. Always consult a qualified physician before making treatment decisions. If you are in crisis, call or text 988 (Suicide & Crisis Lifeline), available 24/7.
Sources
- Rush AJ, et al. “Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report.” American Journal of Psychiatry, 2006. (n=4,041)
- Pigott HE, et al. “A reanalysis of the STAR*D study’s patient-level data.” BMJ Open, 2023.
- CDC/NCHS. “Prescription Medication Use for Depression: United States, 2023.” NCHS Data Brief No. 528, 2025. (11.4% of adults)
- Turner EH, et al. “Selective Publication of Antidepressant Trials and Its Influence on Apparent Efficacy.” New England Journal of Medicine, 2008.
- Kirsch I, et al. “Initial Severity and Antidepressant Benefits: A Meta-Analysis of Data Submitted to the FDA.” PLoS Medicine, 2008.
- Cipriani A, et al. “Comparative efficacy and acceptability of 21 antidepressant drugs…” The Lancet, 2018. (522 trials; 116,477 participants)
- Mitchell JM, et al. “MDMA-assisted therapy for severe PTSD: a randomized, phase 3 study.” Nature Medicine, 2021. (n=90)
- Davis AK, et al. “Effects of Psilocybin-Assisted Therapy on Major Depressive Disorder.” JAMA Psychiatry, 2021. (n=27)
- Johnson MW, Garcia-Romeu A, et al. “Psilocybin in the treatment of tobacco addiction.” Journal of Psychopharmacology, 2014. (n=15)
