has ibogaine never been tested in university drug trials?
“hasn't been, by the way, subjected to university drug trials,”
a guest, author and journalist said on Huberman Lab, 2026-09-28, at 1:47:50. hear it in context →
first, the part that holds
the most prominent university work on ibogaine, the stanford veterans study, is not a drug trial: stanford measured veterans before and after treatment, but the ibogaine itself was given at a private clinic near tijuana, and the paper says "stanford played no role in ibogaine administration." no randomized, placebo-controlled trial run by a university has shown that ibogaine itself works, and a 2026 review of thirty years of research says the evidence is not yet enough to support clinical use.
what we found when we went and looked
the flat version of the claim does not hold. a university medical center in the netherlands (radboudumc) gave ibogaine itself to 14 people with opioid use disorder. that study was approved by an ethics committee and registered with eudract (2014-000354-11), and it found an average peak qtc prolongation of 95 ms, with half of participants above 500 ms. in 2026 the university of são paulo published an open-label pilot giving ibogaine to 9 adults with alcohol use disorder. the university of otago ran a double-blind, placebo-controlled trial of noribogaine, ibogaine's active metabolite, in 27 people on methadone. what the speaker gets right is narrower: these were small safety, pharmacokinetic and pilot studies, and the widely cited stanford work was observational, not a trial.
assertion by assertion
| assertion | verdict | what the record says |
|---|---|---|
| no university has run a human drug study that gives people ibogaine | false | a descriptive open-label study was run in the "department of psychiatry in a university medical center, the netherlands" (radboudumc): 14 people with opioid use disorder got a single dose of 10 mg/kg ibogaine hydrochloride. it was approved by an ethics committee (cmo 2014/081) and registered under eudract 2014-000354-11. in 2026 the university of são paulo published an open-label pilot giving ibogaine to 9 adults with alcohol use disorder. limitation: both studies were small and uncontrolled. they measured safety and pharmacokinetics, not whether ibogaine works. [2][3] |
| the best-known university ibogaine research (the stanford veterans study) was a university drug trial | confirmed | this supports the speaker. the nature medicine paper calls itself a "prospective observational study" of 30 special operations veterans. the veterans "independently scheduled themselves" at a clinic in mexico, and "stanford played no role in ibogaine administration." the paper says "controlled clinical trials to assess safety and efficacy are needed." limitation: stanford ran the measurement under a registered protocol (nct04313712), so it is university research, just not a trial in which the university gave the drug. [1] |
| no university-run controlled trial has shown ibogaine works | confirmed | a 2026 review in the journal of clinical psychopharmacology covered 24 studies and 38 case reports from 1990 to february 2025. it says "no double-blind RCT to date has demonstrated that ibogaine or noribogaine can effectively treat opioid use disorder" and that most positive findings come from uncontrolled studies done outside clinical settings. limitation: this is about efficacy for one condition. it does not mean no controlled study of any kind exists. |
| ibogaine-class compounds have never gone through a placebo-controlled university trial | false | the university of otago and the university of auckland, with industry partners, ran a randomized, double-blind, placebo-controlled single ascending-dose trial of noribogaine (60, 120 or 180 mg) in 27 people moving off methadone (clinical pharmacology in drug development, 2016). it found dose-dependent qtc prolongation and no significant difference in time to resuming opioid substitution therapy. limitation: noribogaine is ibogaine's metabolite, not ibogaine itself. |
both directions
for the speaker: the speaker's intuition fits the shape of the evidence. the stanford work that brought ibogaine to wide attention is observational: veterans arranged their own treatment in mexico, stanford did not give the drug, and the authors call their results "initial open-label findings" that need controlled trials. the university of miami work we read (a 2018 case series of 191 people) also describes open-label inpatient treatment rather than a randomized trial. a 2026 literature review says no double-blind randomized trial has shown ibogaine or noribogaine works for opioid use disorder and that current evidence is "insufficient to support its clinical use." if "university drug trials" means large controlled efficacy trials like the ones behind approved psychiatric medications, we found none for ibogaine itself.
against: the absolute wording, "hasn't been subjected," is contradicted by the record. a dutch university medical center gave ibogaine to patients in an ethics-approved, eudract-registered study, and its safety findings are some of the most important data on the drug: an average peak qtc prolongation of 95 ms, 7 of 14 people above 500 ms, and prolongation lasting more than 24 hours in 6 of 14. the university of são paulo gave ibogaine to patients in a 2026 pilot. new zealand universities ran a placebo-controlled trial of its active metabolite. ibogaine has been tested in universities. what is missing is a large controlled trial showing that it works.
what this cannot say
we read pubmed and pmc records through ncbi's e-utilities, because the pubmed and pmc web pages blocked automated fetching, and we did not have access to clinical trial registries (clinicaltrials.gov is not on our allowed list). so we cannot say which university-run ibogaine trials are currently registered, recruiting or funded as of october 2026, and we may have missed recent or unpublished studies. we also did not read the original record behind a reported 1990s fda-cleared phase 1 study at the university of miami, so we do not cite it. "university drug trial" has no fixed definition, and our verdict depends on reading it as any registered, ethics-approved study in which a university gives the drug to people. under a stricter reading (randomized efficacy trials of ibogaine itself), the speaker would be close to right. confidence: high that university-administered human studies of ibogaine exist; moderate that no completed university-run randomized efficacy trial of ibogaine itself has been published.
sources, each one fetched and read
- Stanford School of Medicine / VA Palo Alto investigators, "Magnesium–ibogaine therapy in veterans with traumatic brain injuries," Nature Medicine 2024;30(2):373-381, doi:10.1038/s41591-023-02705-w, PMC10878970, NCT04313712 What we read there: "we report a prospective observational study ... in 30 male SOVs"; participants "had independently scheduled themselves for MISTIC at Ambio Life Sciences in Mexico"; "Stanford played no role in ibogaine administration"; "Controlled clinical trials to assess safety and efficacy are needed to validate these initial open-label findings." Fetched 2026-10-06 (HTTP 200). https://pmc.ncbi.nlm.nih.gov/articles/PMC10878970/
- Radboudumc / IrisZorg investigators, "The pharmacokinetics and pharmacodynamics of ibogaine in opioid use disorder patients," Journal of Psychopharmacology 2024;38(5):481-488, PMC11102648 What we read there: open-label study, 14 people with opioid use disorder, single dose of 10 mg/kg ibogaine hydrochloride, given in Nijmegen; ethics approval by CMO Nijmegen ref. 2014/081; EudraCT 2014-000354-11. Fetched 2026-10-06 (HTTP 200). https://pmc.ncbi.nlm.nih.gov/articles/PMC11102648/
- Radboudumc / IrisZorg investigators, "Safety of ibogaine administration in detoxification of opioid-dependent individuals: a descriptive open-label observational study," Addiction 2022;117(1):118-128, PMID 33620733 What we read there: "DESIGN: A descriptive open-label observational study. SETTING: Department of psychiatry in a university medical center, the Netherlands." n=14; mean maximum QTc prolongation 95 ms (range 29-146); 50% exceeded 500 ms; prolongation over 450 ms lasted beyond 24 hours in 6 of 14. Fetched 2026-10-06 (HTTP 203). https://pubmed.ncbi.nlm.nih.gov/33620733/
speakers are described by role, not named. one sentence is quoted under fair use, with attribution and a link to the original at the timestamp; we do not re-host audio or video. nothing on this page is a reason to start, stop or change a medication. if you have the file behind anything marked unverifiable, the research board is where people can answer. adam