clock menu more-arrow no yes
A hand holding a small packet labeled Ivermectin Tablets USP. In the background is the aisle of a drugstore.
Ivermectin, a drug touted as a Covid-19 treatment, has become subsumed into our forever culture war.
Mike Stewart/AP

The dubious rise of ivermectin as a Covid-19 treatment, explained

Evidence suggests ivermectin is not a Covid-19 “miracle drug.” How did it get so popular?

On June 22, Joe Rogan’s wildly popular podcast dropped a new episode in which the host and guests Bret Weinstein and Dr. Pierre Kory talk about the pandemic — and a drug that they said would defeat it.

The drug is called ivermectin, and their message was that it was a stunningly good Covid-19 treatment — “good enough to end the pandemic at any point you wanted,” said Weinstein, an evolutionary biologist and now podcaster associated with the intellectual dark web movement. But they argued that the powers that be are trying to keep word from getting out, preferring instead to push a profitable Big Pharma vaccine.

The weeks that followed the episode saw a massive surge of interest in ivermectin as an alternative Covid-19 treatment. The CDC reports that 88,000 prescriptions were written in a single week in mid-August, up from 15,000 in the week before Rogan’s podcast and 3,600 a week before the pandemic began.

That surge of interest is despite the fact that the evidence base on ivermectin is shockingly shoddy. The studies finding massive effects are “probably fraudulent,” says Gideon Meyerowitz-Katz, an epidemiologist at the University of Wollongong in Australia who has looked into the ivermectin literature extensively. Higher-quality studies suggest it may have mild benefits or may do nothing at all — and there’s not enough evidence yet to rule out that it causes any harm.

But the debate has, in many ways, moved past the evidence entirely. Ivermectin — as its starring role on a Joe Rogan episode suggests — has become subsumed into our forever culture war.

The drug’s fans have wildly overstated its benefits and assailed overwhelmed doctors and nurses who refuse to give it to patients. Ivermectin enthusiasm on social media has propelled some into anti-vaccine, anti-Western medicine conspiracies, with some posters even going so far as to warn members against going to the emergency room — because “they fear nurses are killing them on purpose.”

Partisans on the other side have made their own sloppy (if decidedly less egregious) claims in the effort to debunk the ivermectin mania, ridiculing the drug as “horse paste” when it’s actually an anti-parasitic medication for humans, taken by hundreds of millions each year. (It’s also a veterinary drug that can get rid of parasites in animals.)

The furor has also ensnared the researchers studying ivermectin. Ed Mills, a researcher at McMaster University in Canada, told me he’d been flooded with abusive emails when his study (“larger than all the other ivermectin trials put together,” he told me) found that ivermectin’s purported benefits were too small to be detectable. That study (not yet published) is part of a several-thousand-person randomized controlled trial, the gold standard for evaluating Covid-19 treatments.

But it’s not clear those results — or the many other results in the pipeline from high-quality research — will end the public debate about ivermectin. The mania feels beyond the reach of reason at this point.

Meanwhile, even as many Americans resort to a drug with a sketchy evidence base to fight Covid-19, the US estimates 15 million vaccine doses have gone to waste as the country struggles to persuade more people to get the vaccine, for which the case is much, much stronger.

That about sums up the depressing state of America’s battle against Covid-19 in 2021.

The extremely thin evidence for ivermectin as a Covid-19 treatment, explained

First, the basics: Ivermectin is an anti-parasite medication discovered in 1975. It’s one of the world’s front-line drugs against some devastating parasitic infections, especially river blindness, which is transmitted by biting flies and can lead to skin problems and loss of eyesight.

About 250 million people take ivermectin each year. Side effects are generally mild, though nausea, itching, and rashes are common. (Claims have circulated on social media that it causes male infertility, based on a Nigerian study from 2011, but larger studies have not detected this effect.)

And, yes, ivermectin is also a veterinary drug, for deworming dogs and livestock. Because the human version of the drug is prescription-only, making it hard to access without a doctor who agrees that you need it, some people interested in taking ivermectin for Covid-19 have turned to purchasing it from livestock stores, giving rise to the “horse paste” memes and jokes.

Veterinary formulations are harder to dose correctly; information nationwide on ivermectin overdoses and complications is limited, but the South Texas Poison Control Center, for example, which has seen 260 ivermectin-related calls so far this year (up from 191 in all of 2019), says that incidents are overwhelmingly from people attempting to use veterinary formulations to treat Covid-19. The side effects from such misuse include nausea, allergic reactions, seizures, and potentially death. Serious side effects from taking a dose safe for humans are rare.

How did ivermectin enter the Covid-19 treatment conversation? Almost as soon as Covid-19 hit, researchers started looking to repurpose existing drugs against the disease. It’s not unheard of for such repurposed drugs to work: Most drugs act in the body through many different mechanisms, so they can work even against diseases quite different from the ones they were initially developed for.

Many drugs that are not antivirals still have antiviral properties: For instance, a drug like fluvoxamine, which shows promise as a Covid-19 treatment, is an antidepressant, but it seems to modulate the inflammatory response that causes lung damage in Covid-19 patients. Some early studies seemed to suggest that ivermectin — cheap, FDA-approved, relatively safe when taken as directed — might be one such lucky drug, possessing antiviral properties in addition to the anti-parasitic properties it was known for.

But the evidence thus far for its effectiveness against Covid-19 is extremely thin.

Early in the pandemic, ivermectin, along with several other drugs repurposed to fight Covid-19, showed promise in small trials. But it’s very common in such trials for promising results to occur by chance. And while extremely high doses of ivermectin work against Covid-19 “in vitro”— that is, in petri dish samples in laboratories — that tells us practically nothing about how well the drug works against the disease in live patients.

Indeed, the doses of ivermectin that kill Covid-19 in in vitro settings are higher than are achievable in the human body.

So for much of last year, ivermectin was in a common category with many other repurposed drugs like fluvoxamine, dexamethasone, metformin, and hydroxychloroquine: There were some small studies showing promise, and more rigorous research was needed to figure out which of those small studies was for real. “There was genuine interest in the medical community in ivermectin,” Mills, the McMaster researcher, told me. He started enrolling patients for his ivermectin trial last fall.

Over the last year, those studies of ivermectin have been accumulating — and a pattern has emerged. Careful, large, well-conducted studies tend to find modest benefits or no statistically significant benefits for Covid-19 patients who took ivermectin. “The confidence intervals span both modest benefit and modest harm,” says Meyerowitz-Katz. In other words, the results that the studies have found are small enough that they’d be plausible if the drug works — and also plausible if it actively has negative effects.

Mills’s own study, presented at an NIH roundtable in August and still awaiting publication, found “no important clinical benefit,” he told me. The results were sufficiently unpromising that the study would have been terminated sooner if not for the furor around the drug. “The data safety person said, ‘This is now futile and you’re offering no benefit to patients involved in the trial,’” Mills told the New York Times.

The allegations of fraud in ivermectin research, explained

Meanwhile, even as rigorous studies deflated the ivermectin hype, other studies have made headlines claiming outsize, fairly extraordinary benefits — but there’s reason to conclude that those studies are wrong. And in some cases, they’re allegedly fraudulent.

One of the most prominent studies finding positive results for ivermectin was a study from Egypt with lead researcher Ahmed Elgazzar of Benha University. “It was one of the first papers that led everyone to get into the idea ivermectin worked,” researcher Eduardo López-Medina told Nature.

It found extraordinary results for ivermectin, and even though it had not undergone peer review, it was widely cited and was incorporated into various efforts to estimate ivermectin’s benefits. But it was criticized from the get-go for unclear methodology and for not publishing the underlying data the researchers used to find their conclusions (publishing such data is generally good practice to make sure other researchers can do their own vetting of a study).

Then some researchers noticed bigger problems: Most of the introduction to the paper was plagiarized, the numbers in its tables didn’t add up, and the experiment as described would have been very difficult to conduct. Soon, it was removed by the preprint platform that had hosted it. (Elgazzar maintains that his study is legitimate and says the removal occurred without his permission.)

Another big study that has formed part of the case for ivermectin is a study out of Argentina by the University of Buenos Aires’s Hector Carvallo and several co-authors. The paper, published in November 2020, claims that ivermectin prevented health care workers from getting Covid-19 in the first place. The paper describes the drug, along with a nasal algae supplement, as 100 percent protective against infection, while a control group saw up to 56 percent of health care workers get sick.

There haven’t been many studies of ivermectin as a potential prophylactic (a drug you take to prevent infection). Carvallo and company’s stunning finding formed the backbone of much ivermectin advocacy in the US. The study was widely shared: Kory, the head of the ivermectin advocacy organization Front Line Covid-19 Critical Care Alliance and Rogan’s guest on that June podcast, cited it in a hearing before Congress.

But experts on scientific fraud don’t believe Carvallo conducted his study as described. The data bears telltale signs of being manufactured, like numbers occurring in distributions that don’t occur naturally. Information from the study registration about which participants enrolled doesn’t match the information from the published study. The numbers in the tables don’t add up. Key data is missing. I reached out to Carvallo to ask for clarifications on some of these anomalies, and I was directed to a PDF that did not contain the data I had asked for, and then a spreadsheet that also didn’t have it.

As part of a BuzzFeed investigation, journalists contacted one of the hospitals where Carvallo and co-authors claim the study was conducted. The hospital said the study was not conducted there. Carvallo says it did happen there but without the hospital administration knowing.

I asked Meyerowitz-Katz about these two problematic studies, and he suggested the issues with pro-ivermectin research went beyond them. While he emphasized that there’s legitimate high-quality research going on, he says the inconsistencies and potential fraud in pro-ivermectin research are widespread.

Meyerowitz-Katz told me that as his team has reviewed papers about ivermectin’s benefits, “more and more studies appear to be fake or if they did happen, they didn’t happen in the manner described in the paper. We’ve got a bunch of studies that we haven’t gone public with. Some of the studies where we have very serious concerns about fraud are in very high-quality journals.”

These dubious results then turn up in meta-analyses, which are studies of studies that summarize what’s known about a given topic. Because what we have so far are many small, underpowered studies — studies that have sample sizes too small for researchers to confidently detect the effect they’re studying — a meta-analysis would in theory be very useful.

But when some of the literature appears fraudulent, as in this case, it makes it nearly impossible to get meaningful and helpful results from a meta-analysis. One prominent, widely cited meta-analysis found that ivermectin may work quite well — but the Elgazzar paper alone accounted for 15 percent of the effect.

Another meta-analysis studying ivermectin prophylaxis looked at just three studies. Two of them were Elgazzar and Carvallo.

A month after Elgazzar was retracted, the authors of that first meta-analysis released a repeat of their analysis without Elgazzar and it still looks promising for ivermectin — but it still incorporates other studies that Meyerowitz-Katz and other scientific forensics experts suspect did not occur at all or did not occur as described.

A supporter of Brazil’s president, Jair Bolsonaro, holds a large box of box of ivermectin during a rally to show support to his government in Brasilia, Brazil.
Andressa Anholete/Getty Images

Another problem is that many studies, especially early ones from last year, looked at a combination of ivermectin and many other treatments — ivermectin with doxycycline, for example, or ivermectin with dexamethasone. (Dexamethasone has been shown to reduce mortality on its own).

These combination treatments are compared to control groups that didn’t receive any treatments. That makes it very difficult to tell whether the ivermectin did anything or the other substances did — and studies where only ivermectin varies between treatment groups tend to find much smaller effects.

To be clear, there’s some high-quality research on ivermectin, including some that finds positive results. But meta-analyses that incorporate only methodologically high-quality studies comparing just ivermectin rather than combinations of many drugs, and about which there are no significant fraud concerns — as is done in the recent, comprehensive, 100-page analysis of ivermectin as a Covid-19 treatment in the Cochrane Database of Systematic Reviews — tend to find that there is simply not enough high-quality evidence to conclude ivermectin does anything at all.

Beyond a scientific debate

During a June 26 debate on the YouTube channel/podcast Rebel Wisdom, hosted by former BBC filmmaker David Fuller, ivermectin defenders and opponents debated the drug. “What evidence would persuade you that ivermectin didn’t work?” Fuller asked ivermectin defender and medical researcher Tess Lawrie, the director of The Evidence-Based Medicine Consultancy in Bath, UK.

“Ivermectin works,” she responded. “There’s nothing that would persuade me.”

That, to put it mildly, is not an attitude conducive to evidence-based medicine. And it’s an attitude that has come to define the public discussion over ivermectin.

To be clear, the research community studying therapeutics has been careful, in their public statements, not to rule ivermectin out, and to emphasize that current evidence is inconclusive — which is not the same as saying it definitely doesn’t work. The truth is that high-quality research suggests that benefits, if there are any at all, will be small, and that treatments that work better exist and are just as cheap. In the meantime, there’s high-quality ongoing research that should help clear up the remaining uncertainty about ivermectin.

For people trying to push back against Covid-19 misinformation and the ivermectin fad, it’s enough to point out the truth — no need to exaggerate the case against this yet-unproven drug or resort to condescension toward people falling for potentially fraudulent research and experts leading them astray.

“It’s not about laughing at the poor people who are taking it,” Mills, the McMaster University researcher, told me. The most culpable parties aren’t those who believed apparently fraudulent studies, it’s those who conducted, published, and boosted them.

But research can only clear up uncertainty where uncertainty exists — it can’t do anything about certainty that the drug works regardless of the evidence. As ivermectin’s role in the public spotlight has intensified, the debate about it has in many ways split into two. One is a serious debate among researchers and individuals, trying to understand whether the drug works at all and in what doses and under which circumstances, if any, it would pass a cost-benefit analysis.

The other is a fact-free arena in which research is sometimes cited but never grappled with — and is often greeted with hostility. Carlos Chaccour, a researcher at the Barcelona Institute for Global Health, told Nature that he has been called “genocidal” because he was studying ivermectin rather than just advocating for it.

When an anti-vax/anti-mask activist was hospitalized in Chicago with Covid-19, Lin Wood — a lawyer who represented Donald Trump in his lawsuits baselessly challenging the 2020 election — posted a widely shared video of himself calling the hospital and threatening the phone operator with murder charges because the doctors treating the activist wouldn’t treat her with ivermectin.

The complexities reflected in high-quality research are not well represented by the online communities that purport to describe the scientific evidence around ivermectin — and false and misleading information about the drug can often be found in attractive packaging.

The website c19ivermectin.com, for example, has an attractive user interface and is incredibly detailed, chronicling hundreds of studies of the drug. But it presents the estimated effect size based on all the studies out there, never mind that many of them appear fraudulent, and its visualizations effectively weigh every study equally — so a 100-person study that looked at a combination of many therapies including ivermectin will get a data point just like a many-thousand-patient RCT studying just ivermectin against a control group.

The end result is something that looks very scientific. But it’s all built on a backbone of studies that won’t report their data, that in at least some cases may have made that data up, that study ivermectin in combination with many other drugs, and that therefore aren’t any more enlightening when all meta-analyzed together.

For the researchers doing the actual work of figuring out what can help in the fight against Covid-19, ivermectin fandom has become a hindrance. “From a physician’s perspective, you’re not a fan of a drug,” David Boulware, a practicing physician and infectious disease researcher who has studied ivermectin and is currently running a randomized clinical trial to test its benefits, told me. “We’re actually trying to investigate it because we want an answer.”

Figuring out which drugs work against Covid-19 is one of the most critical problems facing humanity. And it makes sense for people to be confused and frustrated by messaging from public health officials, which often hasn’t been very good, or to take it upon themselves to do their own research when the medical establishment has made its own missteps.

But the evidence-free state of public ivermectin advocacy isn’t just hurting people, it’s also derailing the larger goal of ending this pandemic. There are drugs we can be pretty confident work better than ivermectin, and scared, sick people ought to be able to learn about those instead of being deluged with “miracle drug” claims. And researchers need to be able to conduct studies without hearing that the case for ivermectin is so obvious that no further research is needed.

“I’ve been compared to Joseph Mengele and the Nazis, I’ve been told I’m going to hell, all of that stuff,” Boulware said. Health care providers are being threatened; pro-ivermectin Facebook groups are indulging in wild conspiracy theories.

“Can I suggest you include a psychologist as one of your interviewees?” Mills asked me in the course of my reporting. “This is not a medical evidence issue anymore.”

Explainers

How you’ll know when Covid-19 has gone from “pandemic” to “endemic”

Down to Earth

The animals that may exist in a million years, imagined by biologists

Recode

Maybe losing the AI race to China isn’t such a bad idea

View all stories in Future Perfect

Sign up for the newsletter Sign up for The Weeds

Get our essential policy newsletter delivered Fridays.