Researchers' New Model Predicts Number Of People Currently Contagious With COVID-19

Feb 5, 2021
Originally published on February 5, 2021 10:16 am
Copyright 2021 NPR. To see more, visit https://www.npr.org.

STEVE INSKEEP, HOST:

We are able today to better understand the pandemic's spread in the United States. A new calculation tells us roughly how many Americans are spreading the virus each day. It's been common to quote a different number, the number of new infections reported every day. Some days that is a really big number. Yet, big as it is, health officials have known for months that that number is an understatement. Many people are infected but not reported. Even more people infected in the past remain infectious. Now a research team at Columbia University has come up with a way to calculate how many people are spreading.

NPR global health correspondent Nurith Aizenman has seen this. Hey there, Nurith

NURITH AIZENMAN, BYLINE: Hi, Steve.

INSKEEP: So what's the new calculation?

AIZENMAN: OK, quick version - the true number of active cases, people who are still infectious, is likely 10 times the number that health officials give us every day.

INSKEEP: Wow.

AIZENMAN: This really matters, of course, because if you think about the decisions people make - how safe is it to move around in their area? Should there be lockdowns? What's the best vaccine strategy? - all of this depends on knowing to what degree the virus is actually spreading in the population.

INSKEEP: Although this requires them to make calculations based on, among other things, cases that aren't reported, information they don't have. So how did researchers do their estimate?

AIZENMAN: OK, this comes to us from a team at Columbia University led by Jeffrey Shaman. They have built a mathematical model. It's not out publicly yet, but they've shared it with NPR. It starts out by estimating, for each day of the outbreak so far, how many people actually became infectious. Then they compare that to the number who got tested and counted as a confirmed case. What they find is that over the entirety of the pandemic nationwide, only 1 in 5 cases have been reported. And currently, it's about 1 in 4.

INSKEEP: Wow. OK. So that's a lot of people. But you just said take the daily case number, multiply it by four, we get a more accurate number of cases. But then you said the number of people who are out there spreading is 10 times higher.

AIZENMAN: Right. Multiplying by four is useful, but it won't give you a true sense of how bad things are. It just tells you how many people became infectious that day. But people stay contagious for about three or four days. So if I'm counted as a new case on Monday, on Tuesday and Wednesday, I would still be shedding virus, but I wouldn't be counted in the new case tally for those days.

INSKEEP: So if I want to understand the threat level on a given day, I need to take this already public number of new reported infections and multiply it by 10. That's what I would have to do.

AIZENMAN: Right. You want to know how many people whose infections started earlier are still shedding the virus. And that's what the researchers estimated next, and that's where they find that 10 times figure. Here's the lead researcher Jeffrey Shaman.

JEFFREY SHAMAN: So the numbers amplify greatly. When we look at confirmed cases, we're really only seeing the tip of the iceberg as to how many people are currently walking around and unfortunately putting other people at risk with their infections.

INSKEEP: Can you take us through some numbers then? How many people does this add up to?

AIZENMAN: Shaman estimates that on the very worst day so far, December 30, 1% of the U.S. population was actively shedding virus. That's 3.36 million people. Transmission has slowed in the last several weeks, but it's still really high, still above what it was during the summer surge. Shaman estimates that as of last Saturday, 1.25 million people were shedding virus.

INSKEEP: This is day after day after day, which would seem to suggest that by now an awful lot of Americans have already had COVID-19.

AIZENMAN: Exactly. Our listeners can go to npr.org to get Shaman's figures for each state, also more technical details on how the team put this whole model together. But for example, in North Dakota, Shaman estimates that more than 50% of the population has now been infected, which might even be enough to start reaching some degree of herd immunity there. But in many other states, the share of infection is lower. Nationwide, Shaman estimates about 120 million people have now been infected, which is just over a third of the U.S. population. Shaman says the takeaway here is that physical distancing and masking will need to continue until many more people are vaccinated.

SHAMAN: If we let up now, given how much infection is out there, we're going to make it so that more people are going to get the virus before they get the vaccine.

AIZENMAN: He doesn't recommend any kind of return to normal until this summer.

INSKEEP: NPR's Nurith Aizenman, thanks.

AIZENMAN: Glad to do it. Transcript provided by NPR, Copyright NPR.