As the Covid-19 epidemic regains momentum, a new study on ivermectin, this antiparasitic used against scabies, is of interest to the antivaccine sphere. Released on August 31, 2022 on the open-access platform Cureus, it has become popular again in recent days. From a sample of 88,012 people, a team of Brazilian researchers concluded that the regular use of ivermectin for prophylaxis of Covid-19 led to a 92% reduction in the mortality rate linked to Covid-19.
Access to this content has been blocked to respect your choice of consent
More information on the Cookie Management Policy page.
On social networks, this “explosive”, “peer-reviewed” study was taken up to affirm that ““ivermectin was 92% effective” against Covid-19. Other Internet users wonder why then the drug “is still not approved”.
But there is nothing explosive about this observational study: “It sucks. This is the zero level of pharmacoepidemiology”, indignant Mathieu Molimard, professor in the medical pharmacology department at the Bordeaux University Hospital and member of the scientific council of the site of the French Society of Pharmacology and Therapeutics (SFPT).
The reason for this anger? A “huge” immortal time bias, i.e. a factor that can skew the results. A “massive” bias, also noted by Dominique Costagliola, research director at Inserm, specializing in epidemiology. Behind this name hides a problem of temporality in the conduct of the study, which results from a poor methodological choice.
A group of 70,000 people excluded from the study
So, according to the definition of European Patients’ Academy on Therapeutic Innovation (Eupati), this immortal time bias means that participants in the treatment arm “will only be monitored if they do not have the symptoms that the treatment wants to prevent for the duration of the delay. Any participant for whom this is not the case will be excluded from the trial, which means that the treatment arm is selected on a different basis than the “placebo arm” or the comparison group without treatment, known as the control group.
In the case of the Brazilian study, conducted in Itajai, a city in the southeast of the country, the protocol is as follows: for someone weighing 60 kg, 24 mg of ivemectin was taken two days in a row every fortnightly for five months, between July and December 2020. Out of a population of 113,844 people to whom ivermectin had been administered, three groups are considered: irregular users, i.e. those who took less of 60 mg of ivermectin, i.e. 33,971 people, those who took more than 180 mg, i.e. 8,325 people (the treatment arm) and a group between 60 mg and 180 mg, comprising 71,548 people and which is not analyzed (without that we know why). The control group consists of 45,716 inhabitants, who did not receive ivermectin.
A “one-month” shooting window
However, to reach 180 mg, it is necessary to hold three and a half months at least, notes Mathieu Molimard. “If you caught Covid at two months, you are not in the group that was treated with ivermectin, since you did not have 180 mg, you are in the rejected group. The treated group of 8,325 people can only be analyzed “over the period from the fourth to the fifth month”. “Since the study lasts five months, and you have to have taken 180 mg, we only have one month of window of opportunity to analyze 8,000 people who are compared to five months of window of opportunity for the others” , he explains. The results are therefore biased, he says.
What should have been done to make the study work? “The 70,000 patients should have been taken into account in the results, examined throughout the study, underlines Mathieu Molimard. From D+1, if the patient is in the treatment arm and an event occurs, he remains there. And already, I think we correct almost everything. »
Conflicts of interest
Furthermore, the pharmacologist considers that the reviewing was not “serious”. The peer review began on August 22 and ended on August 27, far too short a time: “It’s at least 15 days,” he points out. “There is a real problem,” he warns. If the reviewing had been done, they would have been talking about immortal time bias. » Conflicts of interest authors are also to be raised: several are or have been consultants paid by Vitamedic, a manufacturer of ivermectin, or for the Médicos Pela Vida association, which promotes ivermectin as a treatment for Covid-19 and opposes the vaccination. Others have received grants from the COVID-19 Critical Care Alliance, an association which also defends the use of ivermectin in the prevention of Covid-19.
Cureus, which presents itself as an open access platform for medical publications, is also not part of the list of non-predatory scientific journals that the Faculty of Medicine of the Sorbonne draws up, and in which it is “considered acceptable to publish one’s work”.
“There is no need to continue research”
“With 92% reduction in mortality for ivermectin, the difference is too great to be true, adds Mathieu Molimard. It would show up in a clinical trial if it were true. However, this is not what the clinical trials are up to now. Published on August 18, 2022 in the New England Journal of Medecinea randomized trial of people taking metformin (an antidiabetic), ivermectin or fluvoxamine (an antidepressant), before going to the emergency room or hospitalizations linked to Covid, did not conclude that these three drugs against the virus.
“There is no need to continue research in the ivermectin and Covid field, because the doses to be reached in humans to reproduce the effects in vitro are unattainable and all the not too wobbly studies are negative”, explains, from his side, Dominique Costagliola, director of research at Inserm.
According to the French Society of Pharmacology and Therapeutics, there is no data to recommend the use of ivermectin to prevent or treat SARS-CoV-2 infection. “Regularly updated meta-analyses show no efficacy of ivermectin against Covid-19”, also underlines Dominique Costagliola.