Photo by Lola M. Chavez

When COVID-19 began to crawl across the United States, Americans watched as it transformed the landscape of their communities: Essential stores became lined with tape marks to enforce physical distancing and social media became awash with videos – from comic to institutional – on how to wash one’s hands. Physical distancing and hand-washing are critical tools to stop the spread of COVID-19. Then on April 3, the Center for Disease Control and Prevention issued a recommendation to add one more tool: cloth face coverings.

While surgical and N95 masks should be reserved for medical personnel, the CDC advised, individuals in public spaces where physical distancing may be difficult should wear cloth face coverings to inhibit the spread of the virus. Within weeks, cities issued facial covering requirements with varying degrees of severity. San Francisco began asking residents to wear them in lines and in stores, but last month, it instituted one of the more stringent mandates, requiring individuals (with a few exceptions) to wear coverings when outside or within 30 feet of people who do not live in their households.

What changed? 

Research and interviews with several epidemiologists indicate that more than any specific piece of research on masks, it was the evolving understanding of the virus that made masks newly relevant. And, that understanding is still evolving.

“One of the biggest improvements has been our understanding of asymptomatic transmission of COVID-19,” Marie Stoner, an epidemiologist at Research Triangle International, a nonprofit research institute based in North Carolina. “This finding has been informed by more and more studies that have come out documenting asymptomatic transmissions and asymptomatic people in the community.” Stoner pointed to several studies also cited by the CDC for this finding.

In its recommendation for cloth face coverings, the CDC lists seven different studies published between mid-February and May as the basis for its new mask policy. As opposed to research on the effectiveness of cloth coverings, each study focuses on asymptomatic transmission. 

The first study, published in the New England Journal of Medicine in early March, reported that a German businessman appeared to have contracted COVID-19 from a visiting business colleague who was not showing symptoms when they were in contact before both tested positive for COVID-19. Another study cited by the CDC found that the viral load of SARS-CoV-2, the virus that causes COVID-19, in an asymptomatic patient is similar to those of symptomatic patients, “suggest[ing] the transmission potential of asymptomatic or minimally symptomatic patients.”

“Because we now know that people can transmit the virus without showing symptoms, this impacts how we must respond to prevent transmission,” Stoner said. “It is important to limit contact and wear masks because you or someone else might be positive for COVID-19 but not yet showing symptoms.”

Because the virus is thought to spread primarily through droplets produced when infected individuals cough, sneeze, or speak, masks, in theory, are meant to catch these droplets. There are also questions regarding the potential spread of the virus through smaller, aerosolized particles that would make the virus airborne.

While research shows that asymptomatic carriers can spread COVID-19, on June 8, the head of WHO’s emerging diseases and zoonosis unit, Dr. Maria Van Kerkhove, cast doubt on how often that occurs. Talking to reporters at a news conference in Geneva, she said that based on current data, asymptomatic spread “seems to be rare.” UPDATE: On Tuesday Van Kerkhove “walked back” on that assertion, the NYTimes reported.

As the body of research on asymptomatic spread continues to grow, there is, so far, little systematic study on the effectiveness of cloth coverings in inhibiting COVID-19’s spread.

Dr. Michael Osterholm, director of the Center for Infectious Disease Research and Policy and a Regents Professor at the University of Minnesota, expressed concern about the lack of research regarding cloth face coverings in a special podcast episode produced by the center. He’s not against cloth coverings, but he’s waiting for the evidence to catch up to the advice.

“Never before in my 45-year career have I seen such a far-reaching public recommendation issued by any governmental agency without a single source of data or information to support it,” Osterholm said in response to the CDC’s lack of evidence on the effectiveness of cloth masks. “This is an extremely worrisome precedent of implementing policies not based on science-based data or why they were issued without such data.”

Osterholm criticized the “increasing number of poorly conducted and inadequately reviewed studies getting published in rapid succession.” Remarking on the evidence that we do have regarding cloth masks, Osterholm found it to be “limited and indirect.”

Dr. Lee Riley, division head of Infectious Diseases and Vaccinology at UC Berkeley, believes that a combination of the low rates of COVID-19 in Asian countries and experiments on the filtering ability of different materials prompted a gradual change in the scientific community’s attitude towards facial coverings.

He surmised that mask recommendations began as the CDC and public health officials observed how quickly the pandemic was controlled in Asian countries where wearing masks is a regular practice.

Citing a video from an applied chemistry professor in Japan who tested the efficacy of three types of “masks,” Riley described how even a covering made from three paper towels folded into six layers was successful in filtering out 80 percent of particles in the air—though there were no SARS-CoV-2 viruses in the air and this test was not a peer-reviewed experiment. In the video, Dr. Tomoaki Okuda of Keio University measured different materials’ ability to block airborne particles using a Scanning Mobility Particle Sizer.

Osterholm cautioned against such reasoning and experiments as stand-ins for hard evidence. In his podcast, he points to widespread testing, extensive contact tracing programs and quarantine measures in Asian countries as important measures that reduced the virus’s spread. He also addressed such lab experiments, noting the lack of evidence for smaller aerosolized particles the size of those potentially exhaled by the infected.

He believes masks could play a role in slowing the epidemic but wants to ask additional scientific questions regarding the virus’s spread so the public can make informed choices. One of the most important questions, he said, is how infectious the virus is in the air and how time and dose play a role in infection.

“My whole challenge has not been about whether you wear a mask or not,” Osterholm said in an interview on Monday. “I’ll throw the kitchen sink at this if it will help. The challenge is, ‘How well do they work?’ so that the public knows what level of protection they’re getting.”

Two of the epidemiologists interviewed for this article cited a new study, published on June 1, as important evidence regarding the effectiveness of masks in preventing person-to-person transmission.

Dr. Derek Chu at McMaster University in Canada, along with researchers from universities worldwide, conducted the research in the June study on behalf of the international research team, COVID-19 Systematic Urgent Review Group Effort. The study found that “face mask use could result in a large reduction in risk of infection.” 

Again, however, the study does not evaluate the cloth coverings that most people are using. It did, however, combine results of 172 studies from 16 countries and 6 continents to do a meta-analysis, which means it looked at the data to establish trends. The study found that certain types of institutional-grade masks, as well as physical distancing and eye protection, could reduce the risk of infection. Respirators and N95s were found to be most effective, but disposable surgical masks and similar products (such as reusable 12–16-layer cotton masks) also seemed to reduce risk.

Its findings supported “physical distancing of 1 m or more” and advised that “Optimum use of face masks, respirators, and eye protection in public and health-care settings should be informed by these findings and contextual factors.” It also cautioned that more study is needed: “Robust randomized trials are needed to better inform the evidence for these interventions, but this systematic appraisal of currently best available evidence might inform interim guidance.”

On June 5, just days after the study was published, WHO updated its guidelines on masks recommending for the first time that face coverings be worn in public when physical distancing is difficult. Those over the age of 60, it now advises, should wear medical masks, while others should wear non-medical masks.

The recommendation for non-medical masks came with a caveat: there is not any “high quality or direct scientific evidence” to support their use, but the recommendation noted that research on asymptomatic transmission, observational evidence regarding mask use by the general public, individual values and the difficulty of physical distancing led WHO to update its guidelines. It’s unclear how WHO’s Van Kerkhove’s new revelations on asymptomatic transmission will impact these guidelines. 

For now, its guidance noted that non-medical masks “should only be considered for source control (used by infected persons) in community settings and not for prevention.”

In contrast to the CDC’s recommendation for cloth masks, WHO recommends a three-layered mask composed of an inner absorbent layer (such as cotton) to retain droplets, an outer layer of hydrophobic material (polypropylene, polyester or their blends) to filter and a middle layer made with cotton or hydrophobic material.

WHO also warns that promoting mask use could result in a “false sense of security, leading to potentially lower adherence to other critical preventive measures such as physical distancing and hand hygiene” and cites the potential for increased self-contamination as people manipulate the masks. Epidemiologists continue to advise that masks be worn in conjunction with other practices, such as physical distancing and regular hand-washing. 

Despite CDC and WHO recommendations, there remain plenty of state public health officials who have yet to require masks.

Facial coverings are currently mandated in some capacity in all but two Bay Area counties, though Solano and Santa Clara counties both strongly recommend their use. The term “face covering” is often loosely defined and includes everything from medical-grade masks to bandanas. 

At the national level, 25 states have some kind of facial covering requirement, not including local guidance, according to CNET. These requirements vary in severity. Many pertain to businesses and employees doing essential work, but some apply to individuals in public.

The efficacy of masks outdoors is also unclear.

As the weather warms up, beaches, parks and other outdoor spaces are becoming increasingly populated. Protests have also filled the streets with demonstrators. Epidemiologists believe the likelihood of infection is reduced outdoors as wind dilutes virus droplets in the air. 

Though being outdoors may decrease risk, many epidemiologists say individuals should still be wary of person-to-person spread when in close contact with others. Moreover, they warn, face coverings, especially cloth coverings, are not a substitute for other critical measures such as physical distancing and good sanitation.

For those confused by changing guidelines, Dr. Arthur Reingold, department head of epidemiology and biostatistics at UC Berkeley, had one message: “For a completely new infectious agent, like the novel coronavirus that causes COVID-19, the reality is that there’s an incredibly fast evolution of knowledge. Virtually every day, new information becomes available that might change our understanding of things. It would be foolish if we didn’t change our recommendations as we acquire more knowledge . . . I personally do put faith in what comes from the CDC. There are a lot of really smart, hardworking, well-intentioned people there doing their best to inform the public about how to stay safe.”

The challenge, Osterholm said, is that “we owe the public a lot more information than either ‘everybody masks’ or they don’t.”

We sort through the studies and talk to the experts so that you don’t have to. If you haven’t already, support our reporting today.

Follow Us

FREELANCER. Madison Alvarado was raised in the Bay Area and moved to San Francisco after attending undergrad at Duke University. She fell in love with reporting in high school, and after a brief hiatus is eager to continue learning and growing as a storyteller. She has been covering UCSF's Grand Rounds since the summer of 2020.

Join the Conversation


Please keep your comments short and civil. Do not leave multiple comments under multiple names on one article. We will zap comments that fail to adhere to these short and very easy-to-follow rules.

Your email address will not be published. Required fields are marked *

  1. Excellent analysis and fair reporting. In addition to the analysis of the protection factor of masks is the question of use. As a social worker, I have maintained contact in the community since Covid made it’s ugly debut. My immersion in the community has not diminished since this all began and unless I am meeting someone who is vulnerable, I have not worn a mask. In addition, I encounter multiple people in multiple environments every day and have come to the conclusion that cloth masks and bandanas will offer very little protection because of the way they are used:. Off and on, cross contamination, wide gaps, constant touching, worn beneath the nose, or constantly adjusted because they cause fogging of eyeglasses. In summation, I agree strongly with Dr. Osterrholm that the evidence for mandating wearing masks should be evident.

  2. I listened with great interest to the podcast from Dr Osterholm on cloth masks and science – and I am equally of the view that there is little to no evidence supporting their widespread use. However, latest reporting on TV keeps citing a recent ‘study’ indicating a reduction from 17 to 3 percent by wearing a mask. I have read the Lancet article and don’t see it actually making the determination claimed but find it hard to actually work out what conclusion their facts actually support…. can you offer any comment that I can use when arguing against people citing this study?

    1. Hi Craig,

      The 17 to 3 percent statistic comes from a press release (which can be found here: for the Lancet study I discuss in the article. Although the full study doesn’t to cite these specific numbers, the graphic included in the press release for it (which you can see in finer detail here: seems to be where the 17 to 3 % statistic comes from.

      Although I cannot speak to the analysis that led to this figure from the Lancet, the study is peer-reviewed and was brought to my attention by several epidemiologists that I spoke to. What may be leading to the misunderstanding is the language “face mask” that is used in the graphic. A lot of people have come to understand this to include cloth masks, but in this study face mask is more narrowly defined.

      The press release for the study reads: “Evidence from 10 studies (across all three viruses, including 2,647 participants) also found similar benefits for face masks in general (risk of infection or transmission when wearing a mask was 3% vs 17% when not wearing a mask). Evidence in the study was looking mainly at mask use within households and among contacts of cases, and was also based on evidence of low certainty [1]. For healthcare workers, N95 and other respirator-type masks might be associated with a greater protection from viral transmission than surgical masks or similar (eg, reusable 12-16 layer cotton or gauze masks). For the general public, face masks are also probably associated with protection, even in non-health-care settings, with either disposable surgical masks or reusable 12-16 layer cotton ones.”

      From what I understand based on who I have spoken to, the study was conducted in a highly scientific manner and seems to be legitimate. The problem is in the confusion over what exactly “face mask” means when different people discuss it. I hope this helps.

  3. The reporting in the NYT is beyond useless. The WHO didn’t walk back anything. Anyone who listed to the entire press conference would understand the following rather logical concepts:

    Presymptomatic transmission — someone who is infected but doesn’t have symptoms yet (asymptomatic at the moment, but will eventually develop symptoms) is a main path of viral transmission.

    Asymptomatic transmission — someone who is infected but never develops symptoms is believed to have a very, very low probability of infecting someone else.

    Of course, you can only tell the difference between asymptomatic and presymptomatic in retrospect.

  4. If somebody is coughing or sneezing, I wouldn’t describe them as “asymptomatic.” Of course that person should wear a mask, or quarantine to whatever extent possible.

    Although the scientific literature is usually clear in what exactly was tested, most non-technical articles tend to conflate all kinds of masks, from N95 masks (which are effective at blocking COVID), surgical masks (less so) to bandanas (probably just a placebo for asymptomatic people). Thanks for publishing an article that presents a more nuanced view.

  5. Thanks Madison, this must have taken a lot of work, and congrats to ML for a sober, balanced and informative treatment of a complex and visceral issue. Unlike the shaming/cheerleading pieces which have been run in major outlets like NYT and WaPo and of course, the Chron. In the podcast referred to above, which is refreshing in both thought and language, Osterholm says at one point that “as famed historian John Barry, author of The Great Influenza, the epic history of the 1918 pandemic has taught me, the cities that most successfully contained the 1918 pandemic had leaders who told the truth about what was happening and admitted when they didn’t know.” San Francisco government and health authorities should pay attention. Those truths, a rare commodity in the Age of Trump, may be difficult, unsettling or scary, but they are better than the City’s latest masking order which requires a panel of Talmudic scholars to interpret. The recent WHO guidelines would be a good place to start. Unlike our local authorities and social media propagandists like Jeremy Howard, WHO does not simply say masking won’t do any harm, or it’s “better than nothing.” Beside effectiveness and compliance problems, there are well-known downsides to wearing a mask. WHO recognizes the downsides and lists them. Mask-shamers on Next Door and elsewhere would be well advised to check out the entire Osterholm podcast and the WHO guidance. Thanks again ML gang — much more helpful than simply saying “shut up and put on a mask.”

  6. Anti-maskers are truly the new anti-vaxxers. And publishing articles like this will both-sides us to death.

    1. No, and that’s a disingenuous false equivalence. Anti-vaxxers support their position either through ignoring science or by cherry-picking pseudoscience, precisely the kind of behavior exhibited by London Breed & Grant Colfax. As Dr. Osterholm points out, the efficacy of masks (Cloth Face coverings or non N95 Masks) has yet to generate even a baseline of scientific consensus.

  7. So… what about face SHIELDS? They are incredibly comfortable, and a JAMA study from late April showed they reduce transmission to you by 94%. I actually wear these outdoors now (although I have a cloth mask as well for when I am closer than 6ft to someone) as they provide eye covering, while masks do not. There is no current science on source control for these, but the mathematical models suggest they may at least as effective as masks.

    Plus, they are easier to keep clean, reducing the risk of the virus remaining on the cloth.

    I would really really really like to see SF epidemiologists and health officials weigh in on whether using Face Shields outdoors at distances greater than 6ft is acceptable and whether they are likely to reduce transmission rates.

    If you try one of these, you won’t go back. I’m keeping up on the science, and it turns out they may even be better than face masks in reducing your likelihood of catching COVID-19.

    I’d really like to see Mission Local do a story on the usage of these and whether they may be an acceptable alternative.

    1. Of note, Jeremy, Dr. Jake Scott, the Stanford infectious disease physician who we’ve been talking to about COVID-19, is a huge proponent of face shields.

      I haven’t asked him yet about using face shields AND masks, as many doctors do.



      1. Joe, can you ask him about his opinion specifically of using face shields INSTEAD of masks, outdoors, when you are further than 6 feet away from someone?

        Because I hate wearing masks…I will wear them indoors and on crowded streets, because I can’t guarantee I will be able to pill my mask on in time if someone comes closer than 6ft. But outdoors, having a mask on all the time unless I am beyond 30ft is really too much. Why can’t joggers jog with a face shield, for example? They are much more comfortable, and if they prevent transmission as well as a mask, why can’t joggers or people exercising use them instead of a mask?

        Furthermore, the mandate says that a mask you wear has to fit securely… I don’t know if the people who crafted this mandate thought about this, but many guys in SF (including me) have a long beard. It is physically impossible for me to wear any mask and have it fit securely. Should I shave my beard to comply with the mandate? Not gonna happen. I’ve seen a ton of guys walking down the street with a mask semi-placed over their beard because it just wont fit. It’s worthless at that point for preventing transmission.

        Would city health officials recommend a face shield instead of a mask for guys with beards?

        1. Jeremy — 

          I will ask these questions when next I have the opportunity.

          Thank you for the suggestion,


    1. The counter argument to this are four words: “New Orleans” and “Mardi Gras”

      As in, New Orleans saw it’s large spike in cases occur almost immediately after Mardis Gras. Now, unless everyone who got this in NO during Mardi Gras got it by being in packed bars along Bourbon Street with a super spreader, rather than catching it from someone actually walking ON Bourbon Street who had it, there’s good reason to doubt the relevance of the Chinese study. Mardis Gras and New Orleans is a better example of the risks of having lots of people outdoors in a confined area, and it seems to show that it’s still easily possible to catch it outdoors in that situation.

      The people who are “suspicious” of the media because of your Chinese study would do well to look at what happened to New Orleans.

      1. Funny, I was there in NOLA all week with six of my friends, shoulder to shoulder with packed crowds every night for 7 days straight and not a single one of us caught it. Our livers on the other hand did not feel very well, but no covid

    2. Hi Paulo,

      I am aware of the Wuhan study you are referencing and considered including it in this article; however, the study has not been peer-reviewed yet and many scientists expressed concern with the widespread citation of studies that have not undergone any kind of vetting before being shared via Twitter and other major media outlets. It is also important to note that the Wuhan study was in winter, a time when there people are more likely to be indoors rather than outdoors.

      That being said, many of the epidemiologists I talked to said that the chances of getting infected are lower outdoors, so it is entirely possible that the Wuhan study will hold up under peer review.

      Thank you for your comment,