FiveThirtyEight

When lives are on the line, a dose of hope can easily turn into a cultural battle. 

In the 1990s, doctors and patients searching for better treatments for breast cancer got excited about a new, controversial therapy. It showed big promise in early studies but lacked the kind of conclusive, solid proof that makes medicine mainstream. The treatment — called high dose chemotherapy plus autologous bone marrow transplant, or HDC-ABMT — basically concentrated doses of chemo drugs in the bloodstream like orange juice in a can, delivering high doses — and dangerous side effects. One of the early studies showed 40 percent improvements in three-year survival rates compared to standard chemo. 

But insurance companies didn’t want to pay for it, because they didn’t believe there was enough evidence to prove it was effective and safe. The results, while promising, were coming from trials that lacked the statistical guardrails that help scientists sort coincidence from causality. Frustrated oncologists and their patients fought back with a media blitz and millions in congressional lobbying, eventually winning a series of lawsuits against the insurance companies. More than 41,000 patients received HDC-ABMT. “It became a highly politicized innovation,” said Wendy Lipworth, a professor of health ethics at the University of Sydney in Australia. 

And then it turned out that the insurance companies had been right all along. By 2001, five high-quality, randomized controlled trials had found no evidence that HDC-ABMT did anything. The treatment was toxic and dangerous, and it didn’t help

This kind of fools’ gold is what Lipworth and other scientists are thinking about now, as the world still searches for good treatments for COVID-19. Potential cures like hydroxychloroquine and metformin have cycled through moments of initial promise, slavering media hype, and deeply politicized scientific uncertainty. The latest miracle drug is ivermectin, an anti-parasitic used by veterinarians and doctors treating parasitic worms. Pushed for a year by grassroots groups, media influencers and some governments, the drug lost some of its luster this month when a large trial that had offered some of the best evidence in its favor was withdrawn after a medical student found serious flaws and possibly outright forgery in the data. Despite that, advocates — some of whom have connections to governments and major national newspapers — are still pushing hard. Other studies are ongoing, but it’s not clear how to sift data from hype in the meantime.

Treatments like ivermectin — and the way we treat them — are representative of a much bigger conundrum in medicine. It’s good that patients have more knowledge about the possible solutions that exist and a bigger say in their own treatment, experts said. It’s good to scour the libraries of existing drugs for new uses that might have been overlooked. It’s good to be skeptical of the priorities of insurance companies and pharmaceutical giants. And yet, sometimes, all those good intentions can backfire. So how is the public supposed to know when it’s fighting the good fight for an outsider idea that needs attention … and when it’s just tilting at windmills? 

The research methods that lead scientists to explore drugs as treatments for diseases they weren’t originally approved to treat are common and uncontroversial. This kind of work is happening all the time, said Carolyn Bramante, a professor of medicine at the University of Minnesota and the principal investigator on an ongoing randomized clinical trial testing the efficacy of ivermectin and other heavily hyped COVID-19 treatments. Metformin, a diabetes drug, was first identified as a possible COVID-19 treatment by researchers who had been using computer modeling techniques to find existing drugs that might be able to kill cancer cells, Bramante said. When the pandemic happened, that team had all their research up and running and could turn their attention to SARS-COV-2, the virus that causes COVID-19. They looked for proteins that could stop the virus’s life cycle, she told me. Once they’d found several targets, they used natural language processing programs to scour databases of drug profiles, looking for medications that were known to have an impact on those particular kinds of proteins. It wasn’t much different than what they’d already done for cancer.

But there’s a difference between identifying a drug that might have an effect on a target that might matter to destroying a virus, and finding a drug that definitely kills a virus. The reality is that most potential drugs turn out to not be effective at treating the thing everyone was hopeful they would, even if the drugs looked effective in early clinical trials, which are often small studies that aren’t randomized and don’t have a placebo comparison. In fact, said Jonathan Kimmelman, director of the biomedical ethics unit at Canada’s McGill University, that’s why there is this vast array of existing drugs to test as treatments, to begin with. “A lot of drugs that get to Phase II [trials] never demonstrate safety and efficacy and get abandoned. They end up in the graveyard of drugs.”

This is why it’s not surprising to scientists to find drugs like metformin, ivermectin, hydroxychloroquine, or fluoximine that are worth testing out … and why it’s not surprising that there’s no consistent, compelling evidence that they actually work. It’s also why scientists are uncomfortable getting very excited about the results of any single study, let alone a smaller one, with no randomization or placebo. They know the odds and they’ve been burned too many times by false hope.

But what has been surprising to the researchers I spoke to is the extent to which these drugs — and the hope people have put in them — have been politicized. Ivermectin, in particular, has become a bit of a populist, anti-establishment cause celebre. It first started getting attention as a COVID-19 drug in April 2020, after the publication of a study where it seemed to slow the growth of SARS-COV-2 in a test tube. From those humble beginnings, it has been championed by Tucker Carlson, promoted in invited testimony to Congress, and embraced by the governments of Brazil and India. The journalist Matt Taibbi wrote an article suggesting that a discussion of ivermectin’s potential was being censored by reporters and social media companies who didn’t want to challenge official positions.

Some of these boosters have ulterior motives for wanting to believe in the hope of ivermectin. As a guest on Carlson’s show told the audience, if ivermectin cures COVID-19, then nobody would need vaccines.

Layperson advocacy isn’t always a bad thing, said Dr. Jeremy Faust, a specialist in emergency medicine at Brigham and Women’s Hospital in Boston. At the height of the AIDS crisis patients pushed for more (and faster) testing of more drugs, more accountability from the government and the health care industry, and more say in their own treatment. Today, we, the public, know more about what’s out there, we know better how to ask for what we want and we know more about the very legitimate reasons to question the industries tasked with our well-being. But our knowledge of how to evaluate evidence hasn’t grown at the same rate. “It’s like owning a Ferrari, but you don’t know how to drive,” C. Michael Gibson, a professor of medicine at Harvard University, said. 

COVID-19 took all those good and bad aspects of layperson advocacy and magnified them, Faust said. He and other scientists and doctors told me they hadn’t seen medicine be this virulently politicized before. This could be because of rising anti-establishment politics in the United States. Cynicism has been growing — and trust has been on the decline — for nearly five decades now, a trend that’s been linked to a new kind of political polarization: between the people who view most aspects of American life through an us vs. them conspiratorial populist model and those who don’t. 

Through that perspective, it’s easy to latch onto the idea that someone is suppressing access to drugs that can cure pandemic illness. But the reality is that the data is rarely certain. Take ivermectin, for instance. A systematic review of research, published this week, found that results were all over the map — some positive, some negative. And none of those results were producing strong enough signals in the data to be confident in their own conclusions. There are several larger, better-designed studies in process, which should provide answers that are more clear. But the very fact that people are still setting up randomized, controlled trials means nobody knows whether it works or not, Lipworth said. “You’re doing it because you don’t know. It’s only ethical to do it because you don’t know. If you did, it wouldn’t be ethical to randomize to another group,” she told me.

I spoke with two of the researchers who are currently running randomized controlled trials of ivermectin and other, similarly politicized drugs for treating COVID-19. Neither said they felt pressured to do their studies because of politicized advocacy. And there’s even more good news in the fact that, when past studies have turned up negative findings, they’ve still been published, Bramante said. That’s not always true. Under non-pandemic circumstances, there’s not a lot of incentive or glory in telling everyone something doesn’t work, she said. That lack of negative findings produces a kind of bias in the scientific literature that can mean we don’t always have a full picture of how effective new treatments might be. So it’s heartening to Bramante to see negative findings coming out. It represents real transparency — and a real contrast to pharmaceutical industry norms.

But how do scientists know when enough studies have been done? And how can the public know when it’s time to give up on a promising drug and move on to a new hope? Unfortunately, as Faust told me, “there’s no one cookbook for how to vet an idea and how to say when we’re done.” Scientific consensus is a process of just doing a bunch of studies and then looking at the results in aggregate. An emergency situation like COVID-19 throws that system into disarray, forcing scientists toward smaller, faster, less-certain studies and adding confusion for the anxious public. 

But the experts did have some advice for people trying to vet medical claims in a time of hype and fake news. First, remember that most drugs we test will fail, even if some of the early studies look successful. Second, never decide too much based on one study. Third, it’s the quality of the data — and how certain a study’s conclusions are — that really matters.

Fourth, think about incentives and behaviors of entities in the real world, and whether those really hold up to the claims of a conspiracy theory. For example, the idea that drug companies would suppress a treatment for COVID-19 because it’s an older, off-patent drug they can’t make money off of doesn’t really hold up to scrutiny, Kimmleman said. Drug companies make tiny modifications to chemistry and delivery of old drugs and re-patent them all the time, he pointed out. That’s what AstraZeneca did around the turn of the century, when it evaded an expiring patent on the profitable heartburn drug Prilosec by introducing Nexium, a drug that was basically just different enough from Prilosec to get a new patent. If anything, the conspiracy theory underestimates the pharma giants’ skill at turning medicine into money. 

Finally, Faust suggested, remember the “reverse Erin Brokovich” problem. “Any outside the box thinking is important. I encourage that from my colleagues and from anybody,” he said. “But just because outside-the-box thinkers are sometimes right doesn’t mean they usually are.”

CORRECTION (Aug. 24, 2021, 11:10 a.m.): An earlier version of this article mischaracterized an article on ivermectin by Matt Taibbi. His article did not claim ivermectin “works” as a treatment for COVID-19 but rather explored some evidence that it might. Taibbi also included a number of reasons to explain why, in his view, online conversation about ivermectin was being censored. Originally we had mentioned only the idea that pharmaceutical companies and/or governments wanted to suppress a cheap cure for COVID-19.


Filed under