What has ibogaine done for PTSD?

The strongest human data on ibogaine and PTSD comes from thirty veterans treated in Mexico and assessed at Stanford. Their PTSD scores fell from 31.7 to 3.9 within days, and stood at 4.8 a month later [1]. The effect sizes are among the largest reported in psychiatric research, and the study had no control group [1].

What was measured

The instrument was CAPS-5, the Clinician-Administered PTSD Scale, which is the standard structured interview in this field rather than a questionnaire a person fills in alone [1].

Assessments were made two to three days before treatment, four to five days after, and one month after [1]. The study's primary measure was not PTSD at all but disability, using WHODAS-2.0, with the psychiatric scales recorded alongside it [1].

That ordering is worth noting. The headline most people encounter concerns PTSD, while the study was designed around functional disability in veterans with brain injuries.

Who took part

Thirty male US Special Operations Forces veterans were assessed, with a mean age of about 45 [1]. All had a history of traumatic brain injury, with a mean of 38.6 lifetime head injuries recorded, and a history of repeated blast and combat exposure across an average of 5.5 deployments [1].

At baseline, 23 met criteria for PTSD, 15 for major depression, 14 for an anxiety disorder and 15 for alcohol use disorder [1]. Nineteen had experienced suicidal thinking and seven had made an attempt [1].

This is a specific and severely affected group. A result in combat veterans with dozens of head injuries does not automatically describe what would happen in someone whose PTSD came from a car accident, an assault or childhood trauma. The population is part of the finding.

Participants self-scheduled treatment at a clinic in Mexico, funded by a veterans' nonprofit, and were referred to Stanford for assessment [1]. Stanford played no role in the dosing itself [1]. The people who sought this out were, by definition, motivated to try it.

The size of the change

CAPS-5 scores fell from 31.7 at baseline to 3.9 immediately after treatment, and were 4.8 at one month [1]. The effect sizes were 2.30 and 2.54 [1].

For scale, an effect size above 0.8 is conventionally described as large. Values above 2 are unusual in psychiatric trials of any kind.

The disability measure moved similarly. WHODAS-2.0 scores fell from 30.2 to 19.9 immediately after, and to 5.1 at one month, which the paper describes as no disability [1]. The effect size at one month was 2.20 [1].

Depression and anxiety scores fell alongside them [1]. All of the main results held after correction for multiple comparisons [1].

Numbers of this size invite two opposite errors. One is to dismiss them because the design was weak. The other is to treat them as settled because they are large. Neither follows: a large change in an uncontrolled study is a strong reason to run a controlled one.

What came with the dose

This is the part most often left out, and it changes what the result can be attributed to.

Participants received magnesium sulfate intravenously one to two hours before dosing and again about twelve hours after [1]. The ibogaine was given as a test dose of 2 to 3 milligrams per kilogram followed by further doses to a mean total of about 12 milligrams per kilogram [1].

Around the dose sat a substantial programme. Coaching with a licensed therapist before and after, group preparatory and ceremonial activities, a sweat lodge, massage, yoga, reiki, breathwork, meditation, and structured integration activities on the fourth day [1]. No psychotherapy took place during dosing itself [1].

The result belongs to that whole package, not to a molecule. Several of those components have their own evidence in trauma care, and the study was not designed to separate them. Anyone comparing a clinic that offers the drug alone with this study is comparing two different things.

Screening was also extensive, and included stopping CYP2D6 inhibitors, serotonergic drugs, calcium channel blockers, beta blockers, benzodiazepines, stimulants, corticosteroids and all psychiatric medications before treatment [1]. Coming off psychiatric medication is itself a change with effects of its own.

What the veterans described

A companion study analysed written answers from the same thirty men, collected on average about two weeks after treatment [2].

Four themes emerged. Re-encountering and re-appraising traumatic memories, often accompanied by love, forgiveness and acceptance. An observer-like or dissolved sense of self, described in terms of mystical connection. Emotional resolution arriving in surges. And bodily sensations that participants interpreted as the brain healing [2].

Participants described dream-like states and a sense of being in dialogue with the experience as a guide or teacher [2]. The sessions were self-guided, with participants lying on mats wearing eye shades [2].

A third analysis connects the experience to the outcome. Participants who reported a more intense mystical experience showed larger reductions in PTSD symptoms, both immediately and at one month [3]. That is an association in an observational study rather than a demonstrated cause [3].

If the degree of benefit tracks the character of the experience, then what happens during and around the session is part of the mechanism rather than scenery around a drug effect.

Did it last?

Follow-up at three, six, nine and twelve months was completed by 25 of the 30 participants [4]. The reductions in disability, PTSD, depression and anxiety held across the year, with effect sizes still above 2 at twelve months [4].

A survival analysis estimated the probability of sustained remission at twelve months at 84 per cent for PTSD, 66 per cent for depression and 61 per cent for anxiety [4].

That follow-up has not been peer reviewed. It is a preprint, and its figures may change before publication [4]. It is reported here because omitting it would be its own distortion, and because the distinction between a preprint and a published paper is one a reader can hold.

Five of the thirty did not complete the twelve-month follow-up [4]. In a cohort this small, what happened to those five matters to the interpretation, and uncontrolled follow-ups tend to retain the people who did well.

What the design cannot show

The study was prospective, open label, single arm, without blinding, without a control group and without a power calculation [1]. Each of those limits what can be concluded.

Without a control group, improvement cannot be separated from the passage of time, from regression to the mean in people assessed at their worst, or from expectation. PTSD symptoms fluctuate, and people typically seek treatment when things are bad.

Without blinding, both participants and assessors knew what had been given. Expectation effects are large in psychiatric research generally, and larger still when someone has travelled abroad, paid for treatment and undergone an intense multi-day experience.

Regression to the mean deserves naming rather than assuming. People enrol in studies when symptoms are at their worst, and symptom scales tend to drift back toward a person's average afterwards whatever is done. That drift is ordinarily modest, and a fall from 31.7 to under 5 is far larger than it would explain, but without a comparison group there is no way to measure how much of the change it accounts for.

The authors of the follow-up state plainly that randomised controlled trials are warranted [4]. That is the appropriate conclusion from their own data, and it is also an acknowledgement that the question is not yet answered.

None of this makes the result uninteresting. An effect size above 2 in thirty people, sustained for a year, is exactly the signal that justifies a harder test. It is not the harder test.

What a reader can take from this

Someone reading this because of their own PTSD is trying to answer a different question from the one the study answers, and the gap between them is worth naming.

The study answers whether thirty severely affected combat veterans improved after a package of care built around ibogaine. It does not answer whether any particular person will improve, nor whether the improvement would have come from a different intensive programme.

Three features of the cohort are worth holding alongside the numbers. The participants had repeated traumatic brain injuries, which is unusual among people with PTSD. They were screened thoroughly and came off their psychiatric medications beforehand. And they received a week of structured group work, bodywork and therapist coaching around the dose.

A person who resembles that cohort in some respects and not others is in genuinely uncertain territory, and no honest reading of this study removes that uncertainty.

The cardiac question also sits separately from the psychiatric one. A person whose PTSD is severe has no more cardiac margin than anyone else, and the screening that protects them is the same. Severity of need does not alter the EKG.

What the study does support is a narrower statement than most pages make. A group of people with long-standing, treatment-resistant symptoms reported large improvements that persisted for a year in most of those followed up, under conditions that included substantial medical and psychological support. That is a reason to take the question seriously, and it is not a result that transfers automatically to a different person in different conditions.

Where the research goes next

A controlled trial is the obvious next step, and it is genuinely difficult to run. Ibogaine produces unmistakable effects, including ataxia severe enough that patients cannot walk unaided [5], which makes blinding hard. An active comparator producing some altered state is the usual approach.

Two further questions matter as much. Whether the result holds in people whose PTSD did not come with repeated head injuries, since this cohort had both. And which parts of the package carry the effect, which would require comparing the drug with and without the surrounding programme.

The close read of the MISTIC study goes through the protocol in more detail, and the evidence overview places it beside the other human findings.

Frequently asked questions

How much did PTSD scores fall?

CAPS-5 scores fell from 31.7 at baseline to 3.9 immediately after treatment and 4.8 at one month, with effect sizes of 2.30 and 2.54 [1].

Has this been replicated?

Not in a controlled trial. The twelve-month follow-up covers the same thirty people rather than a new group, and it is a preprint that has not been peer reviewed [4].

Does ibogaine treat PTSD on its own?

The study cannot answer that. The dose came with magnesium, therapist coaching before and after, group ceremonial activities and structured integration [1], and the design was not built to separate them.

Did the benefits last beyond a month?

In the twelve-month follow-up, 25 of the 30 participants were reassessed and the reductions held, with estimated sustained remission at twelve months of 84 per cent for PTSD [4]. That follow-up is a preprint and has not been peer reviewed [4].

Is ibogaine safe for someone with PTSD?

The cardiac risk is unchanged by the reason for seeking treatment. Ibogaine prolongs the QT interval, and half the patients in one hospital cohort passed 500 milliseconds after a single dose [5]. The Stanford participants were screened for cardiovascular, liver, kidney, psychotic and neurological history before treatment [1].