Do Masks Work? What the Cochrane Review Can and Can’t Tell Us

A new review of studies on masking has revived an old COVID-19 discussion point: How effective are masks at preventing infection?

Last month, Cochrane published a new review on what’s known about masking, now three years into the COVID pandemic. The review is an update to Cochrane’s last report on the physical interventions meant to reduce the spread of respiratory viruses, published in 2020. It highlights what the authors describe as “research gaps” concerning the effective ness of masks, and the overall “uncertainty about the effects of face masks.”

The review concluded: “Wearing masks in the community probably makes little or no difference to the outcome of laboratory-confirmed influenza/SARS-CoV-2 compared to not wearing masks.” That conclusion, however, doesn’t definitively prove that masks do not help—though mask critics have latched on to the findings and used them to question CDC guidance.

“Overall, this review doesn’t move me one way or the other because there isn’t enough there for me to really believe that there’s evidence showing that masks don’t work,” Peter Chin-Hong, MD, a professor of medicine at the University of California, San Francisco, told Health. “The public should not use this as a definitive statement that masks don’t work.”

Rather, the review calls attention to the limitations of the research—including how difficult it is to use randomized control trials to study the effectiveness of masks—and what this lack of evidence means when it comes to making decisions about public health, said Dr. Chin-Hong, who was not affiliated with the new Cochrane review.

Masking remains a choice based, in part, on personal risk; and learning about the newest research available in the Cochrane review as well as the debate surrounding the findings, can further help inform your choice.

What Is the Cochrane Review?

Cochrane Reviews are systematic reviews of health care and policy research assembled by members of the Cochrane Collaboration, a not-for-profit organization and an international network of researchers. Their findings are based on the results of studies that meet specific quality criteria, namely, randomized controlled trials (RCTs) which are often viewed as the “gold standard” for measuring the effectiveness of a new intervention or treatment.

This update to the 2020 Cochrane Review adds 11 new RCTs and CRTs, bringing the total number of studies examined to 78. The authors then combined the results of the multiple studies to get a summary estimate, a process called meta-analysis. 

Limitations of the Cochrane Review on Masks

Critics have several issues with the Cochrane Review’s methodology. For example, just six of the 78 trials included were conducted during the COVID-19 pandemic. The others focus on other respiratory illnesses, like the flu. In a February subcommittee hearing, CDC director Rochelle Walensky listed this element as one of the review’s limitations.

The authors acknowledge that these studies were “conducted in the context of lower respiratory viral circulation and transmission compared to Covid-19.”

The Cochrane Review also combines studies where face masks or respirators were worn continuously with studies where these tools were used inconsistently. David Fisman, MD, MPH, a professor at the University of Toronto, Dalla Lana School of Public Health, likened this to comparing apples to oranges—a combination that can lead to a misleading conclusion.

“Taking a respirator on and off is sort of like opening your umbrella in the middle of a rainstorm,” Dr. Fisman told Health. “It’s not going to work very well.”

The Cochrane Review does acknowledge that “adherence with interventions was low in many studies.” This, in part, is why some experts don’t think RCTs are the best way to study whether or not masks and respirators are effective. Scientists aren’t following a person around to see if they are actually wearing the intervention consistently—there is a lack of certainty in methodology.

“In controlled settings, where people are masking all the time perfectly, you do find an effect,” Dr. Chin-Hong said. “It’s just that many of these studies haven’t been randomized controlled, and it’s really difficult to do a randomized control study. In a hospital, for example, you can’t ethically randomize somebody in a higher setting to not wear a mask.”

“It’s fair to say that we need more and better studies on masking, but there likely will not be many randomized control trials forthcoming,” said William Schaffner, MD, a professor of infectious diseases at the Vanderbilt University School of Medicine in Nashville. This is partly to do with inherent issues with compliance. Nevertheless, other types of studies have convinced Dr. Schaffner that masking is a sensible action.

“Other research does tend to point in the direction of at least some protection,” Dr. Schaffner told Health. “Nobody thinks masking is the complete and total answer. That’s also a false expectation. But they are an additional layer of protection.”

Noting the Difference Between Masks

In the Cochrane Review, one section looks at how well surgical masks prevent the spread of illness, while another section compares surgical masks with N95 respirators. The authors conclude that wearing surgical masks “makes little or no difference” compared to not wearing a mask, while “there were no clear differences” between surgical masks and respirators.

However, beyond other methodology concerns, there are issues with expecting surgical masks to perform in a comparative way to respirators, noted Dr. Chin-Hong. Respirators like N95s and KN95s are designed to prevent airborne infections. Meanwhile, surgical masks are meant to block germ-filled droplets and splatters. Laboratory and observational studies consistently show that respirators outperform surgical masks when it comes to protecting the wearer from COVID-19.

The Cochrane Review also does not include a random control trial comparing wearing an N95 with no mask.

“That’s a key question,” Dr. Schaffner said. “Maybe that study hasn’t happened, but I’m curious why they didn’t acknowledge in the review that this is a very important question to answer.” 

In a January 2023 update, the World Health Organization recommended masks and added that these tools should be used “irrespective of the local epidemiological situation, given the current spread of COVID-19 globally.”

What Does Wearing a Mask Do?

Dr. Schaffner advises people who are at a higher risk of severe COVID-19 to strongly consider wearing a respirator in public spaces. This group includes people over the age of 65, pregnant people, and people who smoke. In addition, people with underlying medical conditions, like diabetes, heart conditions, or primary immunodeficiencies, are also more vulnerable. 

This advice extends to one-way masking, or masking in situations where you are one of the few people wearing protection. While masks work best when everyone is masked, lab and observational studies show they still offer a measure of protection. “If you are putting on a mask, you are doing a great job protecting yourself,” Dr. Fisman said.

But masks are still only one layer of protection. A person's vaccination status as well as the ventilation in any specific setting can also influence whether or not a person becomes infected with COVID—further underscoring why it can be difficult for some studies to parse out the role of masks in mitigating harm.

“It’s hard to really disentangle the effects of the mask itself because you’re doing multiple things at the same time,” Dr. Chin-Hong said. “But if you use a high-quality mask, it does reduce your risk of getting infective.”

Though masks and mask studies aren’t perfect, these interventions—alongside other COVID-19 guidance—have proved to be “very effective at preventing disease” transmission, Dr. Fisman said.

“At the end of the day, people will do what they want, and science is not going to move some people one way or the other,” concluded Dr. Chin-Hong. “But a mask does give you a big bang for the buck, and not just for COVID-19. If you are high risk, you probably want to not err on the side of not using it.”