The big thing I got wrong about COVID-19

As an emerging disease researcher, I learned that with new viruses you are never working with good information. The thing about emerging diseases is that you’ve never seen them before. You’re going to make mistakes. So when this thing first started, I was counting on being wrong — either because the early data would be wrong, or my instincts would.

Photo by Anshu A on Unsplash

In mid-January, I was telling my friends to wash their hands because flu was a bigger threat to them than COVID-19. At the time, this was correct. But I thought it would stay correct; I was wrong, and I was wrong for reasons that had nothing to do with the disease.

When SARS-CoV-2, the agent that causes this disease, was spreading in Wuhan, I followed it closely. The rapid growth in case numbers concerned me, but because the case numbers were small compared to the population of Wuhan, I thought that this could be something that we could make go away.

I was worried at the time that it would not go away, though; with every emerging disease, there is some proportion of subclinical infections. This means infections that are not severe enough to get onto the radar of the public health officials keeping track of the disease. I’m used to many emerging diseases looking more deadly than they really are because subclinical cases are often not detected at all. In the case of COVID-19, I thought this might be true too, but I was also worried that if we were only looking at symptomatic, serious cases, we might miss people who were spreading the disease.

These people could then travel the world and spread the disease.

I was worried until I heard the WHO was saying that asymptomatic patients carrying the virus were rare, and also that they did not contribute substantially to the spread of the disease.

It turns out, WHO must have been wrong. Several recent analyses have shown that many of the infections that occur are asymptomatic, and that these asymptomatic infections contribute to the spread of the disease.

For example, this recent paper looks at the “bottle” population on the Diamond Princess, which is about as close as we will get to a field trial of how the virus spreads. As it turns out, just about one-sixth of the passengers aboard tested positive. But what’s more interesting is that of those who tested positive, half of them did not report symptoms, approximately. Half! Now, we can’t just assume that for every symptomatic case anywhere, there is another, asymptomatic case. The ratio could be more asymptomatic cases than that. But we do know that in this case, there were at least twice as many COVID-19 carriers as was apparent from the assessment of symptoms.

I don’t think this is the actual Diamond Princess. Let’s call this a stunt double. Photo by Jamie Morrison on Unsplash

Now, of course, the Diamond Princess passengers were heavily scrutinized. It’s possible that some of the “symptomatic” cases would have been subclinical in a land-based setting where there wasn’t so much focus on who was sick vs. who wasn’t. That sore throat I’ve had on and off for the last week? That could be my COVID infection. I won’t know for a long time. I’d bet more than a few people reading this have the same story.

The other paper that has got me thinking is this one:

It says that China failed to detect 85% of cases, so that’s almost 10 times as many subclinical infections as clinical ones. Instead of 80,000 infections, China had about 800,000. What’s more, this paper reports that without the asymptomatic transmission factored in, they would have expected almost 80% fewer infections in China than were actually seen. So in other words, these asymptomatic patients have a lot of role to play.

Asymptomatic patients don’t cough, so they aren’t producing the quantities of respiratory droplets that I would expect to be required to transmit this. That means that there must be some aerosol component to the transmission of this disease, allowing the virus to spread over more than just short distances. I don’t think that it can waft on the air for a long trip, but it might make it within 6 feet of someone who is carrying it, while a respiratory droplet might make it only 3 feet, for example.

I think this illustrates the short-distance aerosols you might get from an asymptomatic patient. Photo by Dmitry Ermakov on Unsplash

What this means to me is that from one little piece of bad evidence — that asymptomatic cases don’t contribute much to spread, according to the WHO — we made a lot of bad assumptions. That it took close contact to transmit, for example. This meant we were implementing social distancing later than we should have. It meant that just washing your hands wasn’t enough. It might even mean that advice about wearing masks wasn’t right. Though I don’t know which direction that advice might have been wrong, either.

But overall, it tells me that in agreeing with what the WHO said, and even in trusting data that they released before peer review, I and other experts made a bad bet. It goes to show that we are often working with bad data, and I think it’s important for us to be our own Monday-morning quarterbacks and learn from those errors. I’ll do it here by mentioning that I’m just sharing my thoughts, concerns, and opinions; this is what my gut tells me from the data I’ve seen recently. I could be wronger than anyone. But by second-guessing myself, I’m hoping I can be more right. We should all do it.

It may not save the people already infected, but it can do a lot of good for those yet to be exposed.

Virologist, author, damn fool. Also found at and Opinions my own, impressions yours.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store