What We Know About Masks and the New Coronavirus

People crossing street in face masks.
“Under conditions when the infections are spreading fast,” Siddhartha Mukherjee says, “the evidence suggests that mask wearing really decreases the chance of infecting others and getting infected yourself.”Photograph by Kena Betancur / Getty

During the first months of the coronavirus pandemic, the World Health Organization and the Centers for Disease Control and Prevention advised that masks were unnecessary for healthy people going about their daily lives, and should be reserved for patients, health-care workers, and others in close contact with the infected. But, on Friday, the C.D.C. changed its recommendations, saying that all Americans should wear cloth face coverings in public, while emphasizing that surgical and N95 masks should still be reserved for front-line workers. The change was sparked, in part, by new research showing that people without symptoms can transmit the virus, and that masks can reduce the amount of viral particles released. (Amy Davidson Sorkin recently wrote about this subject for The New Yorker.)

To talk about what Americans should be doing to prevent transmission of the coronavirus, I recently spoke by phone with Siddhartha Mukherjee, a stem-cell biologist and New Yorker contributor. His books include “The Gene: An Intimate History” and “The Emperor of All Maladies: A Biography of Cancer,” which won the Pulitzer Prize. During our conversation, which has been edited for length and clarity, we discussed the changing C.D.C guidance, the need for more research on the effectiveness of masks in preventing the spread of the new coronavirus, and why Americans should wear masks regardless.

Should everyone be wearing masks right now when they leave their house?

I think so. I think the issue is there are various kinds of masks. There are the expensive and highly efficient N95 respirators. Doctors have to get fitted for them. They are hard to get. Not everyone should be wearing those, and no one should be hoarding those. Those should be reserved for medical personnel. But then there is a simpler, so-called surgical mask, which is just a filter across your face, which prevents you from secreting respiratory droplets, and prevents others from getting respiratory droplets from you. These are cheap. In the hospital, we used to just use one and throw it away. But under these circumstances, we need a good decontamination protocol. So if you are talking about the simpler ones, then yes. Everyone should be wearing them, because the risk is low, and the return may be high.

You alluded to it, but is this more about protecting people from getting the coronavirus or preventing people who have it from spreading it?

It’s about both. It is very much about the second, about people who have the coronavirus who may be coughing and sweating and shedding the coronavirus. But I also think it is about people potentially getting coronavirus secretions from coughing and so forth. We can’t quantify how much of it is either. But I can tell you that the crucial number we want to reduce is this magic number called R0. That is the number of people that an individual person with the virus can potentially infect. For the coronavirus, that number appears to be between two and three. And that is a very dangerous number. It means the infection will spread exponentially. So one of the main things we need to do to control the infection is to bring that number down to as low as possible. Masks and social distancing and hand hygiene, all together, have the capacity to do exactly that and bring that down, both because you get the effect on the person who is infecting other people, as well as the potential reduction of infection of someone who is receiving the respiratory droplets.

Are there any situations where we don’t need them? What about just an outdoor walk where you aren’t going into a store?

Certainly, if you are not going to come into contact with people, or respiratory droplets, you don’t need them. I am not a social scientist, but I do think there is something that is powerful about the universal signalling if everyone wears a mask. For instance, there are people we assume are immune to the coronavirus, because they already have had infection and recovery. Now they probably don’t need to wear a mask because they are neither shedding virus nor are they likely to get infected. But there is something very powerful about everyone doing it as a group. My plea would be that of course there are people who find it uncomfortable or unnecessary, but that we try to do this as a society during this period, because if everyone does do it, it becomes something we do as a whole society, and you don’t want a stigma around people who don’t want to do it.

A stigma around people who don’t want to do it, or do want to do it?

It goes in both directions. If we decide that we should wear masks, we should really do it as a group.

I was going to say that there is a very important piece here that is part of the article I wrote for The New Yorker. Health-care workers who are in the front lines are likely getting hit by very high loads of virus because they are treating symptomatic patients. They need more protection than someone with a much more lower-dose exposure that you might get at a grocery store or a bodega or on the subway. In the famous story, it’s the dose that makes the poison. In most respiratory viruses—and we don’t know this about the coronavirus—there is a relationship between dose and severity of response. And we would suspect a similar relationship for this virus as well. That is why health-care workers need more protection. They need to be fitted with masks. They have a high degree of exposure to high concentrations of virus, while a casual encounter has a low degree of exposure to a low concentration of virus, and the protection needed under those circumstances is different.

What are the best arguments against wearing masks, and why don’t you think they measure up?

My fear is that when we started this conversation—when the nation started this conversation—the worry was that there weren’t enough N95 respirators, and there was a worry people would hoard them. And my fear is that to hide that misjudgment—that there weren’t enough, and they weren’t prepared—the government made a second misjudgment, which is that these masks are not protective. But that doesn’t make sense. If they are protective for doctors, why would they not be protective for the public?

Would the answer be that doctors are exposed to much worse, as you just said, and so it is negligible for the public?

It might be limited with the public, but it’s a non-zero number. So my suspicion is that, initially, there were non-alarmists who wanted to say, “Let’s not hoard these things that are necessary for medical personnel, and let’s make sure the medical personnel are adequately protected.” The problem with that is that people are smart, and they saw through that logic very quickly, and then began to distrust those who had been saying that in the first place, and that just creates a bad cycle. And the fact that now we are going backward and saying that the public should be wearing masks just reinforces the idea that one day we are making this decision, and the next day we are making that decision.

So, going back to your question about the argument against the public wearing masks, I think that there was a mixture of concern that the public would hoard and that it would create alarm. But, to this point, we are in alarming circumstances, and the risk is non-zero. Let’s say in the worst-case scenario that masks only offer fifty-per-cent protection. Well, fifty per cent is better than zero. And it seems to me, more and more, that the C.D.C. and the White House have come around to the idea that people should be wearing masks.

The other arguments I have read against them were about people scratching and touching their faces more, or about the difficulty of cleaning them, which introduces another risk.

If we ask the public to wear masks, whatever masks they are, we need a simple decontamination protocol. Because what we don’t want is people to wear masks, and feel safe, whereas, in fact, they are collecting respiratory droplets. So we need a simple decontamination protocol.

What about cleaning masks? The C.D.C. says that they should be routinely washed, and that a washing machine “should suffice.”

You should follow the C.D.C. recommendations, but I’m not aware of a lot of empirical data about whether washing degrades the fibres, and it would be very helpful if someone produced that data. I am agnostic, to be honest. If washing works, it works. I am totally fine with washing. I would follow the C.D.C. recommendations.

Just to emphasize: I do not have empirical data that simple masks work for the coronavirus. But we do know they work for other respiratory viruses. So what we are trying to do is extrapolate. There is no empirical evidence that I’m aware of that going to the grocery store with a mask will protect you more than anything else from the coronavirus. But with other viruses, including ones that behave similarly, it helps.

What sort of face coverings should we be using if we don’t have access yet to professionally made masks?

The quick answer is that I don’t know and we don’t know. I would vastly prefer that if we are going to ask the public to use masks, the C.D.C. or someone should give us very clear guidelines on what works and what doesn’t work. There is a study from Cambridge University in which they tested a number of household materials, ranging from cloth to linen, against the size of a typical respiratory droplet. The study suggested that T-shirts and pillowcases worked fine. Again, we are operating with little data. Surgical masks are the most effective in this study. Then a vacuum-cleaner bag and then a dish towel. Then a cotton-blend T-shirt. So that is probably a reasonable item to use in a double layer.

The problem is that if we have a free-for-all, and say you can make your own, then there is no control over it. Some people will make things that are really good, and some people will make things that are terrible. And we will think they are all identical and move right along. It would be very helpful if we could standardize this and have a decontamination protocol for public use.

Has the science changed on masks at all recently? Or has this always been what it has been?

There is more science now. There are more papers for SARS, which show that, in the hospital, a simple surgical mask, in conjunction with hand hygiene and social distancing, vastly decreased the infection rate. The problem with science is that it can’t catch up with what people are doing on the side. If everyone says they are going to use their own material, then science can’t catch up with it, because we don’t know what you are using. So there is more science, but the science has to go hand in hand with the social shifts. And it would seem to me that, rather than the President saying to cover your mouth with a scarf, having a scientific, centralized way to distribute it, produce it free of charge, and make everyone wear it, and have a decontamination protocol would be crucial.

Make everyone wear it, or suggest that they do?

I would suggest it, and if it really makes a dent, then put out a public-service announcement saying everyone should wear it. You can’t force people to adopt behaviors in this country, although you can make it very uncomfortable for people who show up without masking because of social pressures. But I don’t think it should be compulsory. I do think that, given the seriousness of what is going on, given the paralysis in the economy, and the desire for people to get back to work, I think there should be some sense of social responsibility, and if it really decreases the chance of infecting others and yourself, and if people agree universally to wear it, I think we could potentially convince the vast majority of the people to wear a mask.

Does this contradict what you said earlier about not wanting to shame people who don’t wear it? Could we get to a point where we should be shaming people who don’t?

I think shame is not the best mechanism for getting people to wear it. I think the better mechanism is for everyone to understand that there is a sense of social responsibility that they have, and adopting that social responsibility is part of what we do as a country to prevent economic collapse. If we don’t get back to work by June or July—hopefully, there will be medicines by then. But there will not be vaccines, and if we want to get back to work by June or July, I strongly suggest that we find mechanisms by which we can get back to work with masks.

Lots of evidence has shown that masked countries are doing better than unmasked ones in terms of spread. Do you attribute that to the masks?

It’s very hard, because there is no trial and we have to believe the evidence at face value. There was a meme going around Twitter showing that countries with masks were doing better, but then someone superimposed that same meme and instead of masks they wrote bubble tea and no bubble. It is suggestive, but it is an unknowable, because no one did the experiment, and the scientist in me has to say that it’s an unknowable because there is no direct evidence that the countries wearing masks were more successful in decreasing the rise in infections.

And, in South Korea, say, it could be about something else, like testing.

It could be about testing. Yes.

Is mask wearing something that would generally be healthy and cut down on other diseases and flu deaths? Is this a societal change we should be making or thinking about apart from the coronavirus?

We know, again, from a vast amount of scientific evidence that viruses that are secreted through respiratory secretions and cough droplets, including influenza and SARS, are decreased in their exposure when people wear masks. So, under conditions when the infections are spreading fast, the evidence suggests that mask wearing really decreases the chance of infecting others and getting infected yourself. And the return on the so-called R0 of the virus may be quite large. So I think in conditions such as now, when there is exponential spread of the virus, I think that would be the right thing to do.

A Guide to the Coronavirus