Ibogaine Dosage: Why There Is No Standard Dose

Search for ibogaine dosage and the first page of results will hand you at least four incompatible answers, several of them on government and academic domains. They are not competing estimates of one quantity. They are answers to different questions, presented as though they were the same answer, and the gap between the largest and the smallest is more than a factor of ten.

Understanding which question each number answers is the only useful thing that can be said about ibogaine dosage today.

Four numbers, four different claims

None of these four is a recommendation, and no two of them are answering the same question.

20 mg, given to healthy volunteers. The only published study in which ibogaine was given to healthy people under controlled conditions was a pharmacokinetics study: a single oral 20 mg dose in 21 healthy subjects, one group of them pre-treated with a CYP2D6 inhibitor to see what happened to clearance. It was designed to measure how the body handles the drug, not to treat anyone. A companion analysis of the same 20 mg doses looked for stimulant and mood effects and found minimal influence on any of them.

Roughly 0.87 mg/kg, calculated. A 2016 review in Drug and Alcohol Dependence asked the question directly in its title and worked the problem from the animal data. Limited toxicology suggested lethality in rodents by the oral route at around 263 mg/kg. Applying the safety factors ordinarily used to move from animals to humans, and allowing for variation between and within species and for susceptible people within a population, the authors arrived at an approximate initial dose of 0.87 mg/kg. They stated plainly that this was substantially lower than the doses then being administered, and warned that morbidities and mortalities would continue unless practitioners reconsidered. It has never been tested in a person.

10 mg/kg, in an observational study. The Dutch team that monitored fourteen patients in a university medical centre used a single 10 mg/kg dose, which they described as sitting at the lower bound of the range used in previous research. Half of those patients crossed the QT threshold at which arrhythmia risk is considered clinically serious, which is the mechanism by which ibogaine kills. That is what a dose at the lower bound of prior research did to a screened, monitored group.

1,540 mg, on average, in a clinic. In an observational study of 30 people with opioid dependence treated at a clinic, the mean total dose was 1,540 mg of ibogaine hydrochloride. The standard deviation was 920 mg.

That last figure deserves a sentence of its own. A standard deviation of 920 mg around a mean of 1,540 mg means the doses given inside a single study, at a single site, under one team, varied more than most of the published estimates differ from one another. Whatever governed those decisions, it was not a protocol.

The trial that was supposed to settle it

The absence of a dose is not a permanent condition of the field. Somebody set out to fix it, and then the answer did not appear.

Study DMX-IB 201 was registered as a Phase 1/2a trial whose official title states its purpose without ornament: to determine the maximum tolerated dose or treat-to-target dose for the evaluation of efficacy and safety. Its design put a single ascending dose escalation stage in healthy volunteers first, and only then a randomised, double-blind, placebo-controlled stage in opioid-dependent patients seeking medically supervised withdrawal, at the dose the first stage selected.

That is exactly the study whose absence every other paragraph on this page describes. It ran at two clinical research units in the United Kingdom, in London and Manchester. It enrolled 116 people across its two stages. It recorded both primary completion and study completion on 16 January 2024.

The result has still not been reported in a form anyone can check. More than two and a half years after the last participant, the registry entry carries no results, and as of 30 August 2026 we have found no peer-reviewed report of the trial in the indexed literature.

Why a number would not transfer between people anyway

Even a validated milligram figure would be a weaker instrument than it looks, because the same amount does not produce the same exposure in two people.

Ibogaine is converted to its active metabolite noribogaine largely by the liver enzyme CYP2D6, and why that enzyme varies between people is a page of its own. The activity of that enzyme varies substantially between people by inheritance, and it is reduced by a number of common medicines. In the healthy-volunteer study, CYP2D6 phenotype correlated strongly with the amount of ibogaine circulating, and six days of pre-treatment with an inhibitor changed the picture markedly: the drug was absorbed faster, remained detectable out to 72 hours, and had an elimination half-life of just over ten hours.

A dose is a means to an exposure, and here the relationship between the two is person-specific and can be moved by a prescription somebody has been taking for years. How that plays out with individual drugs is a page of its own, and antidepressants are the case where it bites hardest.

There is a second variable before the first one even applies. A milligram of what? Isolated ibogaine hydrochloride, total alkaloid extract and raw root bark are different substances with different alkaloid content, and the human studies above used the first of them. What is actually in material bought online is frequently not what the label says, which makes any figure meaningless before metabolism is reached.

What the reviews say when asked directly

The systematic reviews do not treat the missing dose as a detail.

A review of adverse events between 2015 and 2020 found highly heterogeneous results in terms of the kind of product used and the known dosages, and closed by calling for phase I clinical trials that could describe the safety of different dosages of ibogaine with standardised products. Both halves of that phrase name something that does not exist: the dose-ranging data, and the standardised product to range it in.

A review of the clinical literature to December 2020 found 24 studies covering 705 individuals who received ibogaine or noribogaine, across designs ranging from case report to randomised controlled trial, of which there were two. It concluded that rigorous study designs within medical settings are necessary to warrant safe application. Seven hundred people across the whole published record is not a dose-response dataset.

What would have to happen

A standard dose is not an unreachable object. It requires four things, and the field currently has none of them together.

A standardised product, so that a milligram means the same thing twice.

A dose escalation in healthy volunteers with published results, which has been done and not reported.

A dose-response relationship for the harm as well as the benefit, so that the QT effect can be placed against the amount rather than described alongside it.

A way of adjusting for the metabolic variation between people, which at minimum means knowing CYP2D6 status before dosing rather than after.

Until then, the honest description of any figure a clinic gives you is that it is the number that clinic uses. It may be careful, it may be derived from experience, and it is not a clinical standard, because there is no body that has set one and no published trial from which one could be set.

If you are being given a figure, the questions worth asking are what it is based on, what happens if you clear the drug slowly, and what is being monitored afterwards. What screening should cover and who should never take it at all are more useful to a decision than any dose, because they are the parts of this subject where the evidence actually supports a position.