Analysis of / response to: COVID-19 Superspreader Events in 28 Countries: Critical Patterns and Lessons

The article above, pointed out to me by many good sources, does one of the things we should be doing orders of magnitude more of than we are doing. It attempts to physically model Covid-19, and figure out the ways it spreads and how relatively dangerous they are. Then, based on that, it reasons out the wise policy responses and wise personal behaviors.

My analysis backs up the article’s conclusions. There are ways out, but they seem implausible.

There are three plausible vectors; that part of this article is matched by what I’ve seen everywhere.

Two of them are airborn:

According to the binary model established in the 1930s, droplets typically are classified as either (1) large globules of the Flüggian variety—arcing through the air like a tennis ball until gravity brings them down to Earth; or (2) smaller particles, less than five to 10 micrometers in diameter (roughly a 10th the width of a human hair), which drift lazily through the air as fine aerosols.

And it is on this crucial scale that our knowledge is thinnest. Despite the passage of four months since the first known human cases of COVID-19, our public-health officials remain committed to policies that reflect no clear understanding as to whether it is one-off ballistic droplet payloads or clouds of fine aerosols that pose the greatest risk—or even how these two modes compare to the possibility of indirect infection through contaminated surfaces (known as “fomites”).

This seems super important because we are all choosing how paranoid to be about each of these three vectors, and in what ways. 

Each of the three vectors has different implications. Here’s the article’s take on them.

  1. If large droplets are found to be a dominant mode of transmission, then the expanded use of masks and social distancing is critical, because the threat will be understood as emerging from the ballistic droplet flight connected to sneezing, coughing, and laboured breathing. We would also be urged to speak softly, avoid “coughing, blowing and sneezing,” or exhibiting any kind of agitated respiratory state in public, and angle their mouths downward when speaking.

  2. If lingering clouds of tiny aerosol droplets are found to be a dominant mode of transmission, on the other hand, then the focus on sneeze ballistics and the precise geometric delineation of social distancing protocols become somewhat less important—since particles that remain indefinitely suspended in an airborne state can travel over large distances through the normal processes of natural convection and gas diffusion. In this case, we would need to prioritize the use of outdoor spaces (where aerosols are more quickly swept away) and improve the ventilation of indoor spaces.’

  3. If contaminated surfaces are found to be a dominant mode of transmission, then we would need to continue, and even expand, our current practice of fastidiously washing hands following contact with store-bought items and other outside surfaces; as well as wiping down delivered items with bleach solution or other disinfectants.

Right now we’re doing some mix of these three things, none of them especially consistently or well. 

Large Droplets: Six Foot Rule is Understandable, But Also Obvious Nonsense

For large droplets, there is essentially zero messaging about angling downwards or avoiding physical actions that would expel more droplets, or avoiding being in the direct path of other people’s potential droplets. 

Instead, we have been told to keep a distance of six feet from other people. We’ve told them that six feet apart is safe, and five feet apart is unsafe. Because the virus can only travel six feet.

That’s obvious nonsense. It is very clear that droplets can go much farther than six feet. Even more than that, the concept of a boolean risk function is insane. People expel virus at different velocities, from different heights, under different wind conditions and so on. The physics of each situation will differ. The closer you are, the more risk.

Intuitively it makes sense to think about something like an inverse square law until proven otherwise, so six feet away is about 3% of the risk of one foot away. That’s definitely not right, but it’s the guess I feel comfortable operating with.

Alas, that’s not the message. The message is 72 inches safe, 71 inches unsafe.

Unlike the previous case of obvious nonsense, there is a reasonable justification for this one. I am sympathetic. You get about five words. “Always stay six feet apart” is a pretty good five words. There might not be a better one. Six feet is a distance that you can plausibly mandate and still allow conversations and lines that are moderately sane, so it’s a reasonable compromise.

It’s a lie. It’s not real. As a pragmatic choice, it’s not bad.

The problem is it is being treated as literally real.

Joe Biden and Bernie Sanders met on a debate stage. The diagram plans had them exactly six feet apart.

In an article, someone invites the author, a reporter, to their house to chat. Says he’s prepared two chairs, six feet apart. “I measured them myself,” he says.

Lines have markings six feet apart everywhere.

The parking lot of a Las Vegas hotel marks off spaces six feet apart for homeless people to sleep in, while the hotel is closed. Then of course they sleep end to end within the spaces, so they’re actually one foot apart or less, but then what did you expect.

And so on. People really are trying to make the distance exactly six feet as often as possible.

This isn’t remotely a straw man situation. This is society sacrificing bandwidth to get a message across. 

Again, I get it. The problem is we are also sacrificing any ability to convey nuance. We are incapable, after making this sacrifice, of telling people there is a physical world they might want to think about how to optimize. There is only a rule from on high, The Rule of Six Feet. 

Thus, we may never be able to get people to talk softly into the ground rather than directly looking at each other and loudly and forcefully to ‘make up for’ the exact six foot distance, which happens to be the worst possible orientation that isn’t closer than six feet. 

In theory, we can go beyond this. You get infected because droplets from an infected person travel out of their face and touch your face.

Thus, a line is remarkably safe if everyone faces the same way, modulo any strong winds. The person behind you has no vector to get to your face. And we can extend that. We can have one sidewalk where people walk north, and another on the other side of the street where people walk south. If you see someone approaching from the other direction, turn around and walk backwards while they ensure the two of you don’t collide. If necessary, stand in place for that reason. Either way, it should help – if this is the mechanism we are worried about. 

It would be interesting to go through the SSEs (super spreader events) listed, and see which ones involved people facing only one way, if any. 

Yes, it’s annoying to not face other people, but you absolutely can have a conversation while facing away from each other. It’s a small price to pay.  

In similar fashion, it seems a small price to pay to shut the hell up whenever possible, while out in public. Talking at all, when around those outside your household, can be considered harmful and kept to a bare minimum outright (and also it should be done while facing no one). 

Masks and goggles/glasses also obviously help block such infections, and we should all be forever furious at those who lied to us and pretended otherwise. At this point I’m assuming we’ve won that battle, at least among anyone reading this.

Aerosol Droplets: Embracing the Great Outdoors

The article covers this one well, as the implications are mostly straightforward.

In sufficiently dense places even being outdoors might not be good enough. At the height of the pandemic in New York City, it seemed likely that the air in recently crowded areas even outdoors was dangerous, even if you managed to stay six feet away slash behind other people. There’s a limit where diffusing the air goes from ‘if we diffuse then everything’s fine’ to ‘if we diffuse then everything’s not fine’ and it’s not obvious where that line might be. Also possible that all outdoor travel is at least tiny risk, at that point. Indoors we now care a lot about ventilation and about who has been in a place. 

Surfaces: Are They Even Real?

I don’t know that we know. We could find out easily enough, if we were willing to be a real civilization that understood that people can make choices and trade-offs, and that was capable of making choices and trade-offs and actually doing things. All you have to do is run physical experiments to find out whether people actually get infected from surfaces, at what rates and under what conditions. Then either we ramp up the sanitation messages, or we can stop worrying. 

Alas, we are not a functional civilization in this sense. So we don’t know, and articles like this one, written by non-professionals because someone had to and no one else would, are our best source of educated guesses. We are all doing our own research.

The precautionary principle here says that one continues to sanitize, so we do, even though I think it’s probably unnecessary. A probably unnecessary action that might be very necessary remains necessary.

Identified Super Spreader Events are Primarily Large Droplet Transmission

The article makes a strong case that in identified super spreader events the primary mode of transmission is large droplets. And that large droplets are spread in close proximity, by people talking (basically everything) or singing (several choir/singing practices) frequently or loudly, or laughing (many parties) and crying (funerals), or otherwise exhaling rapidly (e.g. the curling match) and so on.

There is a highly noticeable absence of SSEs that would suggest other transmission mechanisms. Subways and other public transit aren’t present, airplanes mostly aren’t present. Performances and showings of all kinds also aren’t present. Quiet work spaces aren’t present, loud ones (where you have to yell in people’s faces) do show up. University SSEs are not linked to classes (where essentially only the professor talks, mostly) but rather to socializing. And so on; see the full text of the original, near the end, for full details.

Also strong is the concrete example of a restaurant where one individual infected many others and the direction of air flow seems convincingly to be the determining factor of who was at risk.

I buy the core thesis. Identified super spreader events, where lots of people get infected, are primarily fueled by large droplet transmission. The pattern is too consistent to be anything else.

The easiest way for that to be the case is for large droplets to be the primary means of transmission. But that doesn’t have to be the case. What about the alternative possibilities?

Are Unidentified Super Spreader Events Different?

The sample bias in identified events is not subtle. That doesn’t mean we know its magnitude or direction. What would we expect to cause events to be identified?

Events where it is easy to track down participants are going to be included, whereas events where it is hard to track down participants are going to be excluded. 

For an event to count, we’ll need to track down participants to first confirm that an SSE took place, and then to figure out how many people likely were infected. If you can’t do those things, it won’t be counted, even if many were infected.

You’d also have to realize that you should try to do this in the first place.

Even if it is possible to track participants down, if you don’t know to start doing so in the first place, you won’t. So there needs to be an obvious pattern, that is noticed and pointed out, that allows the tracking down to even begin.

Thus, if a subway car was an SSE, would we ever know? It’s not like you can send out a general call for car 5 of the 9:35 red line between stops 9 and 21. You might be able to figure out which subway car or which bus a given person was on, but often you won’t be able to even if you’re talking to them. I don’t think people would even try to track these types of things down. 

This explains some of the absent types of things, but not others. 

Overall the pattern still holds. 

Are Super Spreader Events Different from Regular Infections?

This is a much bigger issue. We’ll take surfaces first, then small droplets.

Suppose a third of infections were via surfaces (which I don’t believe).

Is it plausible that those infections could be distributed and diffuse, rather than creating super spreader events?

In theory, it’s possible. Under this model, people are constantly touching things, and then other people touch those things, and any one interaction is low probability of infection, but there’s a lot of them and they add up over time.

There’s a power law on how often things get touched. A doorknob might be touched once per minute. Your package was touched maybe three times, period. Thus, one doorknob touch is two or three orders of magnitude more dangerous than one package if everyone touches the package in the same place. Since they don’t, it’s more like four or even five orders of magnitude. 

The fall in exposure to surfaces with countermeasures would also seem very, very dramatic with intervention, because you have to actually touch your face before washing your hands in order for it to count. When you touch commonly touched surfaces you know you’ve done it. This has to fall almost entirely on the few people not paying attention. But again, that’s plausibly still a big deal, and doesn’t answer the original question of whether this is something worth guarding against in the first place. 

Still, there’s an upper bound here. Surfaces don’t cause SSEs. We’d probably know it from things like doorknobs and elevator buttons if they did. It’s possible that each person touching an object ends up with a large part of its viral load somehow, which would in turn make subsequent people safer and prevent true SSEs that didn’t have conflated potential causes. Maybe. Or perhaps it requires extensive touching of a surface on both ends, which again makes the infections more diffuse in location and time. But if that is required, it would be that much harder for this to be that big a vector.

Any one piece of missing evidence is easy to dismiss. But the absence of evidence keeps piling up for surfaces as a major vector. 

Small droplets as a constant small risk is the other possibility. It makes sense that they don’t cause SSEs while still perhaps causing a lot of infections. They’re constantly there but not acute, so one is never at a super high risk at any given time and place from them, but they’re there a lot because they linger for a long time. In the cases where people do linger in large groups for a while, such that the risk might compound from lots of different infected people continuing to put out small droplets over time that accumulate, there would almost always be a huge confounding with possible large droplets. So even if that happened, we would likely not have noticed it happening.

It still seems like a long shot. It’s an especially long shot given that contact tracing has been shown to essentially work in multiple places. If small droplets are a major cause, and they linger for a while, contact tracing will break down. Thus, it’s likely that this too is a minor factor in any situation where infection density is sufficiently low for contact tracing. Maybe that changes under mass social distancing plus mass infection, resulting in a meaningful risk from miasma in for example parts of New York City, at least for a while.

Focus Only On What Matters

So, yes. I think it’s probably large droplets. 

Focus on wearing a mask, on not facing anyone not in your household, on avoiding talking or anyone else who is talking. Aim down whenever possible. And so on. 

That doesn’t mean the other causes aren’t worth avoiding. But unless I’m missing something big, we should be focusing the bulk of our efforts on large droplets, plus direct physical contacts, as the primary source of infection.

We shouldn’t pay zero attention to packages and other surfaces. We shouldn’t pay zero attention to small droplets. Better to be safe, even if all you get in most worlds is peace of mind. You feel safe, you know you did everything you could, and so forth.

As individuals trying to be responsible for ourselves and others, it makes sense to use ‘an abundance of caution’ in such spots. I approve.

But if I was running an army that was fighting for survival, and I had limited resources, I’d devote essentially no resources to those efforts. 

Or if I was trying to save a global economy, and I had limited resources, I’d do likewise. I wouldn’t interfere with efforts on other lines, but I also wouldn’t sweat them.

The thing, from the beginning, is that only the big exposures, and the big mistakes, matter.

Within those big risks, small changes matter. They matter more than avoiding small risks entirely.

A single social event, like a funeral, birthday party or wedding, might well by default give any given person a 30%+ rate to infect any given other person at that event if the event is small, and a reasonably big one even if large. You only need one. Keeping slightly more distance, speaking slightly less loudly, and so on, at one such event, is a big risk reduction.

Note that within-household transmission rates are not that much higher than that and there are studies saying it is lower! Simply being around a person is much less dangerous than the other methods being important would imply.

Whereas a ‘close contact’ that doesn’t involve talking or close interaction probably gives more like (spitballing a guess, but based on various things) an 0.03% rate of infection if the other person is positive, and likely with a lower resulting viral load. Certainly those contacts add up, but not that fast. Thus, a subway car full of “close contact” might give you 10 of them per day, most of whom are not, at any given time, infectious. If this model is correct.

That’s not to minimize the risks one takes there. The big risk is model error. We might be wrong about what’s happening. Thus, my best estimate of risk is different from the risk level I’m going to use when deciding what to do. That is as it should be. That’s how we stay alive. 

More thinking like this, please. 












New Comment
12 comments, sorted by Click to highlight new comments since: Today at 3:43 PM

Should I conclude from this that is the optimal protective gear? (Assuming I'm too lazy to put on a hazmat suit.)


I mean, I assume that would work...

Every country should be going crazy doing studies and RCTs to understand (a) most likely causes of transmission, (b) PPE and ways of organizing work/public spaces to minimize risk. Under lockdown, it should be much easier to work out the cause of infection. There are also enormous numbers of hospitals and large care homes all around the world (E.g. janitors and cleaners would seem higher risk if surfaces/aerosol is a major mode of transmission.)

We also need much more work on outdoor/indoor transmission, as this would be a super cheap intervention if it helped. Is there any study of infections by job, say in Germany where the volume of tests is high? Under lockdown, most infections will be from work. So compare indoor vs. outdoor jobs. (What about houses with gardens/decks vs. not in the same area, in a place where the weather is nice in March/April?)

For human interaction while avoiding droplet transmission: we need an app for this. You see a stranger 10m away and want to talk to them. The app would ping their phone via bluetooth and initiate a call. So you can see/gesture to the person but the app enforces a safe distance. Everyone interacting with the public (stores, public transport) should use this app. (Over time, the app could factor in indoor/outdoor and even estimate the safety of indoor spaces based on ventilation).

At many superspreader events (e.g. Korean call center) there are a bunch of people who seem to have very similar exposure to the virus. Yet a substantial proportion (50% in the call center case) don't get infected. This will partly be due to randomness in droplets. But I'd a substantial part of this is variance in infectability. Are some people less infectable in general, or just relative to a particular person? (Younger people seem less infectable in general, and I've heard the suggestion that antibodies to other coronaviruses may provide weak immunity to covid). Natural (weak) immunity would also help explain why if someone in your household is infected you have only a ~20% chance of being infected by them). Someone should use the SSE case studies and try to tabulate this.

Ever played Halo? Imagine a shotgun-like cone of death emanating from everyone's mouth (including, potentially, yours) that attenuates every meter, and try to avoid them as much as possible when you're out for a walk or shopping.

I agree with this post, indeed I've been singing a similar tune for quite a while.

I've been working a bit with one of the digital contact tracing groups, who in turn are working with Google and Apple on their upcoming smartphone OS update for Bluetooth contact tracing. I keep encouraging the group to encourage Google and Apple to use the microphone to detect talking / coughing / etc. and raise the threshold for "contact" if they're absent. I'm very far removed from the decision-makers on this and I doubt anything will come of it, but anyway, I'm fighting the good fight :)


Could we get any info about the risk of spreading through talking by looking at infection rates among members of the deaf community who use sign language? It seems they would be much less likely to spread the virus through communication. Not sure if there would be a strong enough signal to noise ratio to draw any conclusions, but may be worth looking into.

[EDIT: Probably not a valid counterexample; see steve2152's comment below]

Subways and other public transit aren’t present [in the list of superspreader events]

There was that bus in China, which also suggests that recirculated air might transfer aerosols (since many people sat in between the spreader and those who became infected).

FYI, that article was later retracted without explanation.


Nature article giving some evidence for aerosol transmission. More specifically, what it gives evidence of is that in some circumstances you can find aerosolized SARS-CoV-2 where there are infected people, which doesn't seem very surprising. It doesn't say anything about how effectively that causes infection, or about the relative importance of this mode of transmission compared with others. It also has some discussion of the sizes of aerosol particles and how they got that way, and of what circumstances make it more likely for there to be non-negligible amounts of SARS-CoV-2 in the air.

The least obvious things there, to me: Toilets are pretty bad (lots of people, each there for a while, small space). In hospitals, one source of SARS-CoV-2 in the air (in smaller aerosol particles -- do these stay around longer?) may be from PPE after it's been taken off. In the public areas they looked at, only the most densely used ones had substantial amounts of SARS-CoV-2 in the air. [EDITED to add:] "Least obvious" does not mean "very not-obvious"; most of these are pretty unsurprising. I don't know that I'd have guessed the thing about discarded PPE, though.

Um, direction of airflow, by definition, doesn't affect the ballistic transmission of anything. On the other hand, the longer something hangs in the air, the more it's affected by the direction of airflow, and that applies all the way down to gas molecules.

Singing or breathing hard seems likely to increase droplets of all sizes right down to submicron.

My prediction is that (1) it will be 2021 Summer before we see and hear the next festival, and (2) everyone will be required to wear face masks. Lastly, (3) Ravers won't realize any difference.

Appreciate you making the call for more nuance.

What do you make of the South Korea call center, where no one on other floors was infected via the crowded lobby? How do you square that with the scenario you described of recently crowded areas even outdoors during peak outbreak times NYC being a major hazard?

Still very much feels a lot of what made NYC bad was families sheltering in tight quarters, especially in cultures that are very tight knit, with lots of embraces.