Searching for stats on spread without symptoms

Infectious disease expert Jeanne Marrazzo, MD, FACP, discusses asymptomatic COVID-19, face coverings, outdoor versus indoor transmission, and tracing spread with ACP Hospitalist.


Asymptomatic transmission of SARS-CoV-2 is a hot topic. On June 8, a World Health Organization (WHO) official said that it was very rare for truly asymptomatic people to spread the virus. The next day, she clarified her statement, noting that how many COVID-19 infections have come from people without symptoms is still uncertain.

There has also been public debate about the relative risks of indoor and outdoor gatherings and how well face coverings protect against transmission. To learn more about what the existing research says on all these issues, ACP Hospitalist recently spoke with Jeanne Marrazzo, MD, FACP, director of the division of infectious diseases at the University of Alabama in Birmingham.

Q: The WHO has caused some confusion recently. How much do we know about asymptomatic transmission's responsibility for spread of the virus?

A: We do not have all the data we need. I think the best overview was published in the Annals of Internal Medicine. It was a narrative review because at this point, you really can't do a systematic analysis of all the data. Stuff is just coming out so quickly. And it's also very diverse, so it's really hard to compare one study to another. … They ended up coming to an estimate of between 20% and 40% of infection occurring in probably what is a combined phase of presymptomatic and asymptomatic.

Q: What's the significance of the distinction between presymptomatic and asymptomatic?

A: When we first started understanding and seeing this virus, we were really convinced that everybody more or less developed symptoms, even if they were really mild … [but] there's no way that this infection could have spread like wildfire if there isn't a fair amount of transmission in that presymptomatic or asymptomatic person. We know many people who say they haven't been around people who are ill, and yet they end up having the virus and getting sick from the virus. They often get it, I think, from people who are in that presymptomatic phase.

If you look at some of the studies that have been done looking at transmission potential, when they culture the virus in people right around the time they start to get symptomatic and then follow them through, it's very clear that you're very infectious even when you're not really overtly symptomatic. You can just be a little bit sick, and your viral loads are really high. There's some nice studies from Germany where they measured the viral quantity in people's nasal secretions during the early part of their illness into when they started to get sicker. What was remarkable about it was really by day three, their viral quantities were amazingly high, even relative to where they went later. What that says is that people get the virus, then they really start to churn it out and be infectious with large quantities of infectious virus in the first three to four days pretty efficiently. Either they're presymptomatic and they become symptomatic or they remain truly asymptomatic and they just get through it. We really don't know.

Q: Turning to another hot topic on transmission, what do we know about how much of a difference it makes to be outside versus inside?

A: When you look at really discrete clusters, where it's very clear there has been very intense transmission, they have occurred in indoor spaces or very close quarters. So the best examples are churches, those choirs in Skagit Valley north of Seattle, really impressive transmission; the church in Arkansas that was written up by the CDC last month, unbelievable rates of transmission; weddings, barbecues where people were really close together. Yes, the barbecue was outside, but the thing about the barbecue is you are connected by lots of physical contact. If it's not direct physical contact, where you're hugging people or shaking people's hands, you're sharing utensils and you're sharing food. You can't just say “outside,” you have to say: What are you doing outside? Are you sharing things? Are you touching people? Are you in sustained close contact where you could be breathing towards each other for 15 minutes or more? ... You just have to take a nice big, deep inspiration while somebody's sitting across from you and spewing virus out, and it doesn't matter whether you're inside or outside. That said, if you're inside, there's a lot less room for that cloud to dissipate.

When I walk my dog, if I don't see anybody, I'm not wearing a mask. If I can't go across the street, and I have to walk close to somebody, I'll put my mask on, just to be very safe.

Q: Speaking of wearing masks, what do we know about that? If everyone in the churches was wearing a mask, and doing it properly, would that have worked to prevent transmission?

A: It depends on the kind of mask, and it depends on how well they were wearing it. You said properly. Not on their chin. Or their ears—there's some pretty funny memes going around. Provisionally, yes. Because if you look at places that have instituted widespread mask regulations (I'm thinking mostly of China, of course, because people actually did comply), they were able to contain things dramatically and very, very, very quickly.

They also, of course, did physical distancing by isolating people in their homes, so it can be a little bit hard to tease out what the contribution of the physical distancing is versus the mask-wearing. You can find a lot of pointy-headed academic arguments about that, which I have absolutely no patience for. If you're physically distanced, you probably don't need a mask, but the masks really, really, really help when you have to be within a certain distance of each other. Six feet is a good estimate, but it could be more than six feet if it's somebody who's a super sneezer or there's some forced exhalation, like if somebody's running and really, really breathing really hard.

With proper mask-wearing, even if they're not all the highest quality, i.e., N95 masks, you can definitely make an impact. For example, most of the health care workers are wearing cotton masks or just plain surgical masks, and with that, plus N95s for people who are caring for patients with COVID-19, we have essentially cut health care worker infections to really rare occurrences. That's why it's so frustrating to hear people give us this false choice of either we shut down or we don't. We have to just cautiously re-emerge in a different way with careful mask-wearing and thinking carefully about our physical distancing, and then I think we would be OK.

Q: State officials have discussed testing and tracing as a major component of reopening. How does asymptomatic transmission fit with that?

A: It's very confusing because the concept of testing and tracing got introduced very early on. And in the beginning, when there were small clusters of outbreaks like Northern California, Seattle, you could go in there and really quickly test people, isolate the positives, quarantine people who were exposed, and get a handle on it. That's exactly what they did with SARS. The challenge now is there is no way you could possibly do that with the number of infections that we are seeing. In New York or in Washington state, they can start thinking about doing testing and isolating positive cases now. In Alabama, we had 1,000 cases reported yesterday. How could you possibly even know where to start?

There becomes a tipping point where it goes from a good thing to do to where the virus is so spread out everywhere that it almost doesn't matter until it runs its course a little bit. In that case, widespread testing to give people the tools they need to protect themselves and protect the people around them is good, but I don't think you're going to be able to have the luxury of swooping in there and doing more of a real intense public health intervention.

Q: Do you think we'll better understand asymptomatic spread soon?

A: Yes, I do. The way to do it is going to be to have a widespread, reliable way to measure antibodies. That's been a real challenge, because the serology tests have been not all as reliable as we'd like. Ideally, you would have an antibody test that you believed really meant somebody was infected, and you would test 1,000 people, and then you would really intensively find out from them: Were they sick? When were they sick? How sick were they? You can also do it prospectively, of course, when you start doing widespread testing and follow people to see if they do get sick. But that takes a lot more because you really have to try to catch people as they are actively infected. So I think we will get more information. I also think there are some animal models that are looking a little bit promising.