The confusion around hydroxychloroquine is one of the most enduring parts of the coronavirus pandemic. There’s no reliable medical evidence showing that the drug, which medical practitioners have used for years as an antimalarial and treatment for chronic conditions like lupus, reduces the severity of COVID-19 cases or lowers the likelihood of death in people who have the condition. Ultimately, the public health consensus is that hydroxychloroquine isn’t the COVID-19 treatment that will get us out of this crisis. But, for a time, no one could blame you for thinking otherwise. On March 28, the FDA granted an emergency use authorization for doctors to give hydroxychloroquine to some hospitalized COVID-19 patients who couldn’t participate in clinical trials. Less than two weeks later, the National Institutes of Health (NIH) announced the start of a large clinical trial to evaluate hydroxychloroquine for COVID-19 patients. Then June arrived. On June 15, the FDA reversed course, revoking its emergency authorization and writing that the known and potential benefits of the hydroxychloroquine no longer outweigh the “potential serious side effects.” A few days later, the NIH halted their clinical trial because “the study drug was very unlikely to be beneficial to hospitalized patients with COVID-19.” The NIH’s most up to date COVID-19 treatment guidelines (which the Centers for Disease Control and Prevention refers to on its page about COVID-19 treatment) recommend against using hydroxychloroquine for treating this illness. And yet, the myth of hydroxychloroquine as our COVID-19 savior persists.
As the pandemic drags on, we’re all getting desperate for a cure. At press time, there have been more than 8.9 million confirmed COVID-19 cases in the United States and more than 227,000 confirmed deaths, along with tens of thousands of additional deaths related to the circumstances of the pandemic (like people not wanting to seek medical care for fear of catching the illness). The science just isn’t there to show hydroxychloroquine is the answer to COVID-19. So why did the world go wild for it?
How COVID-19—and politics—disrupted how we do science.
Before we dive into why hydroxychloroquine keeps making the news, here’s a quick primer on how experts research the safety and efficacy of potential COVID-19 treatments.
The only way to know if a treatment is safe and really works on COVID-19 is through rigorous testing. Researchers may deem a treatment “successful” if it meets certain metrics, such as whether it reduces the likelihood of hospitalization or death.
Randomized controlled trials are considered the gold standard in research to ensure treatment safety and efficacy. They’re often “blind” in some way so that participants (and sometimes even researchers) don’t know who is receiving the drug or a placebo, since a person’s belief that they’re receiving treatment can affect how they respond to it. When a study is double blind, meaning neither the researchers nor the participants know who is getting the drug or a placebo, that helps prevent methodological flaws like the experts subconsciously giving sicker patients the experimental treatment. And the trials are randomized to ensure characteristics that could affect a person’s response to treatment (sex, weight, degree of illness, etc.) are equal between groups.
Another major type of study is observational, meaning researchers give patients a treatment and watch what happens. Observational studies are easier and cheaper than randomized controlled trials but more likely to contain biases or flaws in some way. “It’s possible but not easy” to come to solid scientific conclusions from observational studies, Eleanor J. Murray, Sc.D., an assistant professor of epidemiology at Boston University School of Public Health, tells SELF.
Another important part of the study process is peer review and publication. This helps make sure a study is as accurate and as truthful as possible. Even then, “I don’t think we should assume that any particular study is without flaws,” Ivan Oransky, M.D., co-founder of Retraction Watch and vice president of editorial at Medscape, tells SELF. “Peer review can be an important filter, but it also misses a lot. I think the real mistake is treating any single study as definitive.”
As Dr. Oransky points out, “We all want drugs and other treatments to work. Thanks to confirmation bias, we’ll tend to see the positive when we’re faced with an existential threat like the novel coronavirus,” he continues. “Rigorously performed studies keep us honest by reducing the risk that our data are skewed or even wrong, and scrutiny before and after publication makes life-threatening errors less likely.”
The glaring issue here is that pretty much nothing is normal during this pandemic—including how experts research and use different kinds of medical treatment. There’s never 100 percent consensus in science, but COVID-19 has completely upended the medical and societal systems that would normally raise the alarm about hydroxychloroquine not actually being effective. At the start of the pandemic, when hospitals were overwhelmed, doctors were willing to try just about anything in an attempt to save critically ill patients. Hydroxychloroquine “was cheap, it was available, it had some laboratory data that supported it, and there were some anecdotes out of China that people felt like it was doing something,” Vinay Prasad, M.D., M.P.H., a hematologist-oncologist and an associate professor of medicine at the University of California San Francisco who studies cancer drugs, health policy, and clinical trials, tells SELF. Under normal circumstances, meeting these criteria would have led doctors to study a drug for a specific purpose—not automatically use it. “But desperate times call for desperate measures…and so people did jump on board with hydroxychloroquine pretty rapidly,” says Dr. Prasad. What’s more, the FDA had already approved hydroxychloroquine for various uses, like treating lupus, so doctors are essentially free to use it off-label. “This is a unique situation,” says Dr. Prasad. “There is no oversight. Doctors can really do whatever they want.”
In the rush to latch on to any hopeful evidence, some news outlets then blew hydroxychloroquine’s potential promise out of proportion, like by not giving proper context when discussing a pre-print study. This added to the confusion in public understanding of hydroxychloroquine. With pre-print research, scientists haven’t published their study in a journal but have made their research available online because the peer-review process takes time. Given how quickly the coronavirus situation is changing, pre-print research published on sources like bioRxiv and medRxiv has been integral to helping the world quickly discover important information about this disease. The goal is for pre-print papers to eventually get peer reviewed and published in a journal, but when it seems like they have answers to COVID-19 mysteries, there can be real value in sharing them before the peer-review and publication process is done. With that said, it’s necessary for anyone discussing these studies—whether that’s journalists or scientists—to be clear about the fact that this kind of research hasn’t yet undergone peer review or publication, because it’s not guaranteed that even the most promising pre-print research will hold up under that level of scrutiny. “That doesn’t always happen for lots of reasons,” says Murray. Sometimes it may be because a paper winds up having a massive flaw in design or execution. “Good science is hard and takes time, so it’s actually somewhat predictable that the earliest research will be the least well-conducted and the most likely to turn out to be wrong, especially when so many groups were rushing to ‘solve’ this whole COVID thing,” says Murray.
Another major factor that made the hydroxychloroquine vetting process abnormal: Some politicians and high-profile figures gave the drug far too large of a platform. Brazilian President Jair Bolsonaro has claimed hydroxychloroquine would treat him when he tested positive for the virus. In May, American President Donald Trump said he was taking the medication preventively in case he contracted the infection. Later that month, he said he’d stopped taking the drug. Then, on October 2, Trump announced he’d gotten COVID-19, which sparked renewed interest in the drug on social media (this in spite of the fact that the President’s doctors never indicated that he was taking HCQ as part of his treatment). Some followers of these public figures, along with various adamant scientists, continue to publicly support the drug as a potential treatment even though scores of public health experts and agencies insist hydroxychloroquine has no significant value against the coronavirus.
“I do think society got wrapped up in the promise of hydroxychloroquine purely because of a few prominent individuals pushing it as a cure despite the lack of evidence,” says Murray. “My sense is that almost anything could have been promoted in the same way as hydroxychloroquine was and would have captured the public imagination similarly.”All that said, the strongest science doesn’t support the idea that HCQ treats or prevents COVID-19. “That’s the thing about medicine that’s tough…There are many more failures than successes because biology is really hard,” says Dr. Prasad. “Scientists responded to the public pressure to investigate this drug, found that it was not useful, and moved on,” says Murray. But even after science doesn’t seem to show significant value of a potential treatment, a few supporters may remain. “I think in this case they’re likely to exist forever because this particular drug has been tied to political beliefs and identities,” says Dr. Prasad.
So what does the science actually say?
First, let’s discuss why some proponents of hydroxychloroquine think it can help treat COVID-19 in the first place.
Earlier in the pandemic, many epidemiologists and infectious disease experts thought hydroxychloroquine might make it more difficult for the coronavirus to enter a person’s cells and replicate, as SELF previously reported. Many experts also wondered if the drug could change a person’s immune system response to be more effective against the virus.
Some doctors and researchers were especially interested in trying to treat COVID-19 by using hydroxychloroquine combined with zinc or azithromycin. Zinc helps the body fight off bacteria and viruses; when it’s present in cells, it appears as though it may be able to block the SARS-CoV-2 virus from replicating. Some evidence suggested chloroquine (of which hydroxychloroquine is a derivative) could help zinc do this job. This is why some believe zinc does “the dirty work against the virus,” Harvey Risch, M.D., Ph.D., a cancer epidemiologist and professor at the Yale School of Public Health and the Yale School of Medicine, tells SELF.
Most people in the U.S. are not deficient in zinc, according to the Mayo Clinic. But some groups of people who are more likely to develop severe cases of COVID-19—such as older adults—are also more likely to get less than the recommended daily amount of zinc, according to the National Institutes of Health (NIH). As a result, some researchers have speculated that people with COVID-19 who are at high risk for severe cases may avoid serious complications if they receive hydroxychloroquine combined with a zinc supplement early on in their treatment. However, as the NIH notes, “There are insufficient data to recommend either for or against the use of zinc for the treatment of COVID-19.”
As for hydroxychloroquine and azithromycin: Azithromycin is an antibiotic that treats bacterial infections like pneumonia, which people with viral respiratory infections (including COVID-19) often develop. Because past research suggests that azithromycin is effective in treating people with Zika and Ebola viruses, and that the drug may help prevent severe respiratory tract infections in people with viral infections, some researchers hypothesize that if hydroxychloroquine were effective against COVID-19, adding azithromycin to the mix could help better treat severe coronavirus cases.
Let’s start with the studies hydroxychloroquine proponents often use to bolster their case. Unfortunately, the major studies in this camp have limitations that cast doubt on their findings that hydroxychloroquine may treat COVID-19.
Risch is the author of a review and a letter to the editor, both published in the American Journal of Epidemiology, arguing that doctors should treat high-risk COVID-19 patients with hydroxychloroquine and azithromycin as soon as possible after infection. As evidence of his point, Dr. Risch cites results from research like a controlled non-randomized study in Brazil. The study, which hasn’t been peer reviewed, was released in April and included 412 people who had “flu-like” symptoms for three days on average and accepted treatment with both hydroxychloroquine and azithromycin. The 224 people who declined the treatment made up the control group. The study found that those who got the treatment were less likely to be hospitalized for their symptoms (1.9 percent vs. 5.4 percent for controls). But, crucially, in addition to not being peer reviewed, randomized, or blinded, the study does not specify if any of the participants had actually tested positive for COVID-19. This is a significant limitation—it’s impossible to know how many people in the study even had COVID-19 rather than another illness with similar symptoms. Dr. Risch also highlights a piece of unpublished observational research, released in July, of 712 people in New York with confirmed COVID-19 infections. The study was a retrospective case series, meaning it used pre-existing data for the analysis (rather than recruiting participants specifically for the study). Of the 712 people involved, 141 began taking hydroxychloroquine with azithromycin and zinc for five days about four days after their symptoms started. Another 377 untreated people made up the control group. Ultimately, the researchers found that people in the treatment group were less likely to be hospitalized (2.8 percent vs. 15.4 percent) or die (1 vs. 13 deaths) within 28 days of treatment. There were multiple limitations with this study, though, like the researchers excluding some patients then adding others without explaining why. Also, since the research was retrospective, there weren’t details on the control group’s age, risk factors, COVID-19 symptom severity, or where they got treatment. The researchers did adjust for the fact that there were nearly three times as many people in the non-treatment group. But, unlike with randomized controlled trials, people with roughly the same characteristics weren’t randomly split evenly between treatment and placebo groups to have a comparable playing field.
After the publication of these pieces of research, a collective of Yale faculty members, including in epidemiology but also across other disciplines, released a statement saying, “The evidence thus far has been unambiguous in refuting the premise that HCQ is a potentially effective early therapy for COVID-19.” The Yale School of Public Health also released a statement emphasizing that the FDA had revoked its emergency use authorization for the drug in treating COVID-19.
Even a July International Journal of Infectious Diseases retrospective study, which many hydroxychloroquine proponents cite in support of the drug, has its issues. The study looked at 2,541 adults who were hospitalized for COVID-19 at Henry Ford Hospital in Southeast Michigan. Some received hydroxychloroquine, some received azithromycin, some received both, and some took neither medication. The researchers found that people who were treated with hydroxychloroquine were 66 percent less likely to die than those who received neither treatment, while those who received hydroxychloroquine and azithromycin had a 71 percent lower risk of mortality than untreated people. One major limitation to these findings: People who took hydroxychloroquine or hydroxychloroquine with azithromycin were about twice as likely to also receive steroids than people who received neither drug. That makes it tough to determine which drug was responsible for the noted benefit. That’s important because a growing amount of high-quality clinical trials has suggested that steroids may reduce the risk of death from COVID-19, leading the World Health Organization to update its guidance in early September to encourage steroids as a treatment for severely ill patients.
The studies that cast doubt on hydroxychloroquine as an effective COVID-19 treatment tend to have a more rigorous design than the ones suggesting the drug may be helpful against the disease. That, however, doesn’t mean they don’t also have their own limitations.
For example, Murray points to a randomized controlled trial published this July in The New England Journal of Medicine. It involved 665 people in Brazil with mild to moderate COVID-19 symptoms (504 participants had positive COVID-19 tests). “It’s not necessarily the perfect trial, but it’s absolutely much better evidence than the other studies on this topic,” Murray says. Researchers randomly assigned participants to receive a placebo, hydroxychloroquine alone, or hydroxychloroquine with azithromycin. The study authors found that those who received treatment were less likely to improve their “clinical status” than those who got the placebo. Those who received any kind of hydroxychloroquine-involved treatment were also not less likely to get put on mechanical ventilation or die than those who got the placebo. As Murray suggests, though, there are limitations to the study. It wasn’t blinded, for example. It was also funded in part by Brazilian pharmaceutical company EMS Pharma, which sells hydroxychloroquine in its generic form.
Another randomized controlled trial published in July, this one in the Annals of Internal Medicine, looked at 423 non-hospitalized people who either tested positive for COVID-19 or had symptoms after coming into contact with someone who tested positive. About half randomly received hydroxychloroquine and half got a placebo. This study was double-blind, meaning neither the participants nor the doctors knew who received treatment or a placebo. The people who received hydroxychloroquine had no significant difference in symptom severity over 14 days compared to the placebo group. A follow-up analysis found that zinc supplementation also had no effect. One limitation, though, was that only 58 percent of people actually got tested for COVID-19 due to testing shortages.
Even with these limitations, at this point, it seems clear that less scientifically solid studies suggest there may be a significant benefit to using hydroxychloroquine against COVID-19 while more rigorously designed studies aren’t finding one. This is why the general medical consensus is that hydroxychloroquine isn’t the answer to the coronavirus pandemic.
On top of the lack of evidence that hydroxychloroquine is a viable COVID-19 treatment, trying to use it this way also comes with some health risks. While hydroxychloroquine is relatively safe, it can be very toxic in higher doses, as SELF previously reported. A slight risk of potentially fatal cardiac arrhythmia is one of a few known side effects. “There are basically no medications people can take with no side effects ever,” says Murray. “This one does have some serious side effects that we need to watch out for.” Researchers continue to actively look into this. An opinion paper published in July 2020 in the Lancet warned that hydroxychloroquine-related heart issues may increase during the pandemic for several reasons, including that many COVID-19 patients already have multiple risk factors.
There are other potential risks of giving people an unproven drug. Increased demand for hydroxychloroquine has led to shortages that affect “people for whom hydroxychloroquine is an important and life-saving medication,” says Murray. There’s additional concern about unnecessary antibiotic resistance: If everyone with COVID-19 takes azithromycin, “it’s not going to be available to treat other infections down the line,” says Murray.
Ultimately, the strongest evidence we have doesn’t support that this medication is an effective, universal treatment for COVID-19. “I think there’s a lot of incentive to rush to put something out that looks like it might work…[but] the easiest way to do science is to not consider all of the things that could go wrong and to not ask really clear questions in a really detailed and specific way,” Murray tells SELF. “A lot of people go into medicine because they want to help people. And every day we don’t find something to treat this, huge numbers of people are dying. Everyone would love that this works, but it doesn’t mean that it does.”
The situation with coronavirus is evolving rapidly. The advice and information in this story is accurate as of press time, but it’s possible that some data points and recommendations have changed since publication. We encourage readers to stay up to date on news and recommendations for their community by checking with their local public health department.