Three years into the COVID pandemic, and there still isn’t a consensus on whether masks protect against the coronavirus. Proponents and skeptics both think common sense is on their side.
It’s simple physics, proponents say. COVID is airborne. Doctors wear masks when they treat patients. Plus, I wear a mask every time I leave the house, and I’ve never had COVID.
Not so fast, skeptics retort. The masks most people wear don’t stop viruses. States and countries with widespread mask use have seen large waves of cases. And besides, I wore a mask and got COVID anyway.
When faced with this kind of debate, many hope the answer could be found in formal scientific research. Which brings us to the recent Cochrane Review, which considered whether physical interventions—including masks—reduce the spread of respiratory viruses. Cochrane Reviews are widely considered the gold standard of evidence-based medicine.
“Wearing masks in the community probably makes little or no difference,” the review authors concluded of their work comparing masking with non-masking to prevent influenza or SARS‐CoV‐2. What’s more, even for health care workers providing routine care, “there were no clear differences” between medical or surgical masks versus N95s.
But as the saying goes, absence of evidence is not evidence of absence. The review doesn’t show that masks definitely do not reduce the spread of COVID—only that studies to date have not proven that they do.
“The Cochrane Review tells us two important things. First, there have been very few high-quality studies examining the effectiveness of masks during the COVID pandemic, and second, from the little high-quality data we do have, we don’t see large impacts of masking in preventing viral infections on the population level,” Jennifer Nuzzo, director of the pandemic center at Brown University School of Public Health, told Slate. “This doesn’t necessarily mean masks don’t protect individuals. But it could mean that the way they’re used at the population level is not effective. We need more randomized trials to understand why.”
The reason Cochrane Reviews are such a useful tool is because of the strict methodology the authors use to systematically evaluate and assess the quality of evidence about an intervention’s safety and effectiveness. Throughout the process, review authors work with a Cochrane Review Group and editorial team, and the findings undergo peer review. Thus, review findings are seen as a synthesis of the best evidence available.
In the case of masking, there’s been some consternation about the predispositions of the review authors. Tom Jefferson, a senior associate tutor at Oxford University, has spearheaded Cochrane Reviews of interventions to reduce the spread of respiratory viruses since 2006. But Jefferson has raised some eyebrows, as he has publicly expressed skepticism about masks. In a recent interview, he suggested that physical contact and fomites play a role in SARS-CoV-2 transmission. (Slate was unable to reach Jefferson for comment.) Senior review author John Conly of the University of Calgary—who has downplayed airborne COVID transmission, contrary to most experts, who think transmission is primarily airborne—is also the senior author of one of the key studies included in the review.
Before COVID, experts generally agreed that masking the entire population to prevent respiratory disease is unwarranted—that’s what the previous version of the review, published in 2020, showed. But this year’s version started adding COVID research to the mix. For this review, the authors used only randomized controlled trials, or RCTs, and cluster randomized trials, which randomize groups rather than individuals. Randomized trials are generally considered the most robust type of evidence because they minimize bias that can crop up in observational studies. As an example of bias, think about how states with mask mandates may have also imposed other COVID restrictions, or how people who wear masks are also likely to take other precautions.
The 2023 review adds 11 new studies—which together included nearly 611,000 participants—to the previous 67 studies analyzed. Six of the new studies were conducted during the COVID pandemic. The Danish DANMASK study, for example, found that recommending and providing masks did not reduce the COVID infection rate in the community, although overall use was low. A study in Bangladesh randomized some villages to a mask promotion intervention that included free masks, encouragement from community leaders, and reminders. The likelihood of getting COVID was 11 percent lower in villages given surgical masks, but the difference was not significant for cloth masks, and the benefit mostly accrued to older people. A third study (the one Conly co-authored) compared N95 use with surgical mask use among health care providers while at work and showed no clear benefit of N95s. No study pitted N95s against no masks at all.
The drawback of systematic reviews is that they’re only as good as the included studies: a classic case of “garbage in, garbage out.” And indeed, the studies in the Cochrane Review on masking had some issues. It was hard to eliminate bias, and low mask adherence was a problem. “The primary deficiency in the vast majority of studies is that they don’t measure mask wearing through direct observation, said Jason Abaluck, an economics professor at Yale who led the Bangladesh trial. “They simply ask people whether they wear masks. Self-reported mask-wearing doesn’t at all resemble actual mask-wearing.” A 2021 study in Kenya, for example, found that while 76 percent of people reported always wearing a mask in public, only 5 percent were observed doing so.
Comparisons of COVID case trajectories in different places show that while states and countries with strict mitigation policies or broad cultural acceptance of masking bought some time—often allowing more people to be vaccinated—most eventually saw large waves. If masks do make a big difference, it should be apparent by now on a large scale, but it’s not.
Some mask proponents think the review should have included other kinds of evidence. Masks, they argue, are more like helmets than novel medications—there’s less potential harm. Indeed, observational, case-control, and mechanistic studies typically make masks look much better. For example, a case-control study in California, published by the CDC, found that people who wore cloth masks, surgical masks, and N95 or KN95 respirators were 56 percent, 66 percent, and 83 percent less likely, respectively, to test positive for SARS-CoV-2. Unlike the Cochrane Review, several other systematic reviews that included evidence beyond RCTs have found that masks work well at the population level.
Mask defenders argue that the reason masks don’t appear to have a substantial effect in most RCTs is because people don’t always wear high-quality masks the right way all the time. During the pandemic, leaky cloth masks or “baggy blue” medical masks have been a common choice. Many people wear masks below their nose or take them off to eat or communicate more easily. That behavior doesn’t mean that the masks themselves don’t work. On the other hand, people aren’t mannequins in a laboratory—when it comes to COVID prevention, the only thing that matters is what they do in the real world.
Speaking of the real world, by now, most people have established opinions about masks—so much so that it’s unlikely the Cochrane Review will change many minds, never mind much behavior.
Given conflicting evidence from imperfect studies, mask skeptics say the burden of proof is on those who want to adopt the new intervention. “The problem with the absence of evidence argument is that it also applies to ludicrous things,” said Vinay Prasad, a professor in the department of epidemiology and biostatistics at the University of California, San Francisco. “No one has proven that dancing naked in the rain can’t ward off COVID, but we don’t advocate, and certainly don’t mandate it for years without proof that it helps.”
On the other side, mask proponents insist that in the face of uncertainty, it’s better to be safe than sorry. Even if you’re not worried about your own risk, masking protects other people who are more vulnerable. Why not, they ask, since the downsides of masking are negligible?
Ultimately, the decision about whether to mask comes down to personal feelings about risk tolerance, collective action, and the effects of masks—or COVID itself—on quality of life. People disagree on all three counts, so it’s unlikely we will ever come to a consensus. It doesn’t look like “the science” is going to be a tiebreaker anytime soon. More than showing whether or not masks work, the Cochrane Review finds that the kind of evidence gathered so far can’t really answer the question. Maybe that’s a good reason to let people decide for themselves.