JAMA ivermectin study deceived participants on study drug

JAMA ivermectin study deceived participants on study drug

The article “Effect of Ivermectin on Time to Resolution of Symptoms Among Adults With Mild COVID-19” was published in the Journal of the American Medical Association on March 4. The study subjects were deceived. They were told that the study drug was “D11AX22 Molecule”.

The lead author on the article, Eduardo López-Medina, confirmed that in a telephone conversation. He explained that it would have been difficult to recruit subjects to the study if they were told the study drug was ivermectin. López-Medina stated that this practice was disclosed during ethics and governmental reviews of the study protocol and that it had been approved.

It had been reported earlier that the title of the study protocol used the term “D11AX22 Molecule” instead of “ivermectin”. That raised the question of whether the study subjects had been informed that the study involved ivermectin.

This is a moment that requires moral clarity. The JAMA article needs to be retracted immediately.

Note that views expressed in this opinion article are the writer’s personal views and not necessarily those of TrialSite.


  1. People are understandably puzzled by the pseudorationale that Lopez-Medina gave for deceiving study participants into thinking they were testing a drug called D11AX22 because he didn’t explain it explicitly.  But as I will show you below,  if he had explained it explicitly, you would have seen how Uncontrolled and invalid his study was.

    The enrollment problem that Lopez-Medina admitted to in his phone conversation with Peter Yim (and later in a 3/23/21 online comment at JAMA Network) has to be understood in the context of the rampant self-medication of over-the-counter Ivermectin that was being strongly encouraged by medical and political authorities in Cali, Columbia where the study was performed.  If you honestly tell the patient that you are doing a study on Ivermectin, you also have to tell them that they canNOT self medicate with Ivermectin and that 50% are going to be given a placebo instead of Ivermectin.  This creates the recruitment problem that Lopez-Medina admitted to without fully explaining.  If the vast majority of the population already believes Ivermectin will protect them from COVID-19, Lopez-Medina knew that there was little chance of enrolling patients if they were told they might NOT get any Ivermectin during the study; none from the study and none from self-medication.

    Lopez-Medina claims that the only solution to this enrollment problem was to deceive study participants into thinking they will be testing a different drug D11AX22.  Anyone who thinks about this a little should be puzzled?  How is this going to help you, if you still have to forbid all study participants from self-medicating with Ivermectin.  Then it dawns upon you.  Ohh.  D11AX22 deception can only solve their enrollment problem, if it allows the participants to think they can still self-medicate with Ivermectin. And so far no one has directly challenged the authors and asked them WHETHER THEY EXPLICITLY EMPHASIZED TO STUDY PARTICIPANTS IN THE INFORMED CONSENT PROCESS THAT THEY COULD NOT SELF-MEDICATE WITH IVERMECTIN DURING THE 21 DAY STUDY PERIOD. Importantly, D11AX22 could also effectively deceive reviewers, editors, and readers into thinking that the Placebo group, which in the protocol is formally defined as those not receiving D11AX22 from the study, was really an Ivermectin Placebo group, completely unexposed to any Ivermectin whether from the study or from self-medication.  Unfortunately, for them. Peter Yim’s sharp eyes and dogged persistence unraveled this entire deception.

    Even if the study had been very explicit and specific about no self-medication with Ivermectin, the study was in no position to prevent participants in either arm of the study from self-medicating with an easily obtainable drug.  Hence, one would at least like to know if the study had any way of reliably determining if a patient self-medicated with Ivermectin so that the patient could be removed from the study’s analysis.  Unfortunately, the only reliable method, to measure plasma Ivermectin levels in all patients, was not done by the study.  That alone makes the study highly suspect.

    But what about an UNreliable way of assessing self-medication based on self-reporting in response to repeated specific questioning: have you taken any Ivermectin during the study period?  At least that would be better than nothing and provide some sense that the authors were not being willfully ignorant to make their study publishable.  Here is where the authors get very slippery, and the way they mince words and conflate things is very telling.
    In his 3/23/21 comment Lopez-Medina writes that: “As described in the article, patients were contacted daily for a structured interview that documented the use of any medications outside the study.”  However, if you comb through the article, study protocol, and supplementary tables, you see no evidence to support this entire statement.  You can find evidence for the first half before the “that” and for the second half after the “that”, but not for the full statement where both halves apply together.

    Regarding the first half of the statement, the article does indeed state that during the 21 day study period “…patients were contacted by study staff on days 2 through 5, 8, 11, 15 and 21 for a structured interview”.  What was in this interview?  The study protocol says the interview’s purpose was to “fill out a questionnaire” asking about (1) patient symptoms and (2) possible adverse effects often associated with Ivermectin use.  But there is no mention that this questionnaire “documented the use of any medications outside the study” during the study period, much less documented the specific use of Ivermectin.  And if they did document outside medication used during the study period, why isn’t it in any of the supplementary tables, which contain a plethora of patient information.

    Regarding the second half of the statement, the study “documented the use of any medications outside the study” but as described in eTable1 this was only for “Medications initiated between symptom onset and enrollment”.  There is no documentation anywhere of any medication initiated after enrollment during the 21 day study period.
    So it appears that the authors are conflating their enrollment interview, where they documented preenrollment initiation of medication, with their periodic phone interviews, where they apparently did NOT ask about any initiation of new medication, especially Ivermectin during the 21 day study period. To be absolutely clear about this the authors should be asked TO PUBLISH THE QUESTIONNAIRE USED IN THEIR PERIODIC STRUCTURED INTERVIEW.  That way we can see whether they did or did not specifically ask about initiation of Ivermectin self-medication during the study period.

    Without any solid evidence to the contrary, any disinterested reader has to conclude that (1) there was no explicit warning to study participants prohibiting them from self-medicating with Ivermectin ; and (2) there was no attempt to either objectively or subjectively monitor Ivermectin self-medication.  There is thus every reason to believe that JAMA published an UNcontrolled Ivermectin study in which the authors admit to potentially crippling enrollment and control problems due to rampant self-medication, and actually exacerbate the control problem by use of the D11AX22 deception to solve their enrollment problem.

    Someone needs to submit an online comment to JAMA asking Lopez-Medina:1) if he can document that the study emphasized during the enrollment process and throughout the study period that participants could not self-medicate with Ivermectin.2) to publish the questionnaire used in the periodic structured phone interviews during the 21 day study period.

    The lack of a direct response to such a comment would speak volumes and allow serious and honest medical scientists to dismiss this study as invalid.

  2. I feel there were multiple major issues with this Cali Colombia IVM. However, I don’t see the significance of not disclosing the name of the therapeutic to the participants. Wouldn’t this actually constitute another level of blinding to remove bias of preconceived beliefs of a well known drug? Maybe I’m missing something here.

  3. I’m not a scientist, please help me understand correctly. Is the problem that participants weren’t told IVM was the trial drug and, being available over the counter in Colombia at that time, it’s likely that one or both study groups were contaminated?
    Why would it have been difficult to recruit participants for an IVM study given the strong safety profile? Is it because participants wouldn’t want to be allocated to the control group?
    Was past or current Ivermectin usage an explicit participant exclusion criteria? I note a vague reference in the published version to removing participants who didn’t meet selection criteria post-hoc…. surely they weren’t questioned on IVM usage afterwards?

    1. I have no evidence that the control group was exposed to ivermectin as a result of using "D11AX22 Molecule" instead of "ivermectin".
      Yes, potential study subjects wouldn’t want to be allocated to the control group.
      Yes, recent prior ivermectin use is an exclusion criteria. Dr. Lopez said that the study subjects were questioned daily on whether they used ivermectin. He said that some subjects asked if the study drug was ivermectin. In those cases, they confirmed that the study drug was ivermectin.

  4. "He explained that it would have been difficult to recruit subjects to the study if they were told the study drug was ivermectin" – why exactly? Did it have a bad reputation? Or on the contrary the participants wouldn’t want to risk getting a placebo instead?

    1. No, ivermectin did not have a bad reputation. He said that people saw it as a good treatment option because it has few known adverse effects.

  5. Perhaps it would have been better to first have asked the lead researcher for a copy of the questionaire used to qualify participants — rather than to rail against the published study (and by extension, the lead researcher) and thereby alert everyone on Earth that there may have been a problem with the participant exclusion criteria.
    We seem to have evolved into a social media world where the first reaction to anything we disagree with is to place our potential adversary on the defense. This rarely results in getting to the facts — instead, it promotes obfuscation.
    My 2 cents — your mileage may vary.

    1. You may have misunderstood this report. There is not, as far as I know, a problem with the exclusion criteria.
      Dr. Lopez is not my adversary. I am reporting on a serious mistake that he made.