ARI SHAPIRO, HOST:
On March 31, the White House Coronavirus Task Force presented the nation with a sobering number - 100,000. That's how many Americans they projected would die at minimum from the virus. Dr. Anthony Fauci warned us to be ready.
(SOUNDBITE OF ARCHIVED RECORDING)
ANTHONY FAUCI: As sobering a number as that is, we should be prepared for it. Is it going to be that much? I hope not. And I think the more we push on the mitigation, the less likelihood it would be that number.
SHAPIRO: Well, less than two months later, that likelihood has now become a near certainty. Even as we inch closer to 100,000 American lives lost to the virus, experts say the true death count is almost certainly higher than what we've been able to record. Here to explain more is Maggie Koerth of the data journalism website FiveThirtyEight.
Welcome to ALL THINGS CONSIDERED.
MAGGIE KOERTH: Hi. Thank you for having me.
SHAPIRO: So in your reporting, you talk to people whose job it is to count the number of dead from ground floor-level staff at local hospitals to national officials at the CDC. Why do they think some COVID-19 deaths have gone uncounted?
KOERTH: We think that it's being undercounted because of a few things. First off is just logic. You know, early on in the pandemic, you had these situations where hardly anyone was getting tested. But there were definitely people that were sick, and there are definitely people that were dying. And so we know that a lot of people got missed.
There's also the fact that we have seen data that suggests there are more deaths happening that we're not seeing. For instance, if you look at the number of deaths that have been attributed to, like, influenza or pneumonia over the course of this summer in hot spots like New York, you'll see that it is a much higher rate than has been in previous years and is, in fact, higher than would be accounted for by just the flu going around.
And you see kind of similar overages with all-cause death. So looking at, like, death from every cause, no matter what it is, is also much higher than it has been in previous years. So we can make some assumptions based on that that there are COVID deaths or COVID-related deaths that we are missing.
SHAPIRO: You tell one story in detail of a man named Bob Duffy who died early in the pandemic with symptoms of the disease, and he had contact with people who were known to have COVID-19. Yet Bob was never tested, and COVID-19 is not listed on his death certificate. What can we learn from his story?
KOERTH: Well, I think one of the things we learn is how when you're dealing with something that's new and that people don't really understand very well, the rules around how people get counted change. You know, early on in the pandemic - and certainly when Bob died - it was still fairly normal to only count people who had been tested and shown to be positive for COVID-19.
But just two weeks after Bob's death, there was a CDC Web seminar for doctors that was sort of talking about - OK, how do we start trying to capture some of these people who probably died from COVID that we can put down probable cause of death even if we don't know for certain? And the circumstances of Bob's death, today we would count him as a probable COVID-19 death.
SHAPIRO: President Trump and others have suggested that COVID-19 deaths may be overcounted, not undercounted as you report. What do the experts that you talked to say about that?
KOERTH: The experts said that there are probably some people who are being counted as COVID deaths who shouldn't be - that that probably does happen. But ultimately, because of that data where we can see that all-cause deaths and influenza-like deaths are much higher than they have been in past years this time of year, ultimately, they think that undercounting is the overall result.
SHAPIRO: I recently learned that we still don't know how many people died in Hurricane Katrina. Even 15 years later, the estimates vary widely. And so if the same ends up being true of COVID-19 and we never have an exact count, what difference does that make? Why does it matter?
KOERTH: Maybe the biggest place that this matters, I think, is psychologically. When I spoke with Bob Duffy's family, it was really hard for them to not know what he had died of - to know that he wasn't going to be counted even though they thought that he should be. One of the things that we've sort of run into with reporting on COVID is that there's a lot of unknowns. And that's an uncomfortable place for people to sit. We want science to tell us exactly what's going on. We want science to tell us what to do.
And sometimes the best we can get is an estimation. And we have to kind of figure out how to adapt to that, how to live with that. And that's hard. I think it matters to get a estimation, though, because we need to have some way of setting policy. And so getting the best counts that we can - even if we know those counts are not perfectly exact and accurate - that gets us closer to the truth. And getting closer to the truth, in this kind of situation, is sometimes the best you can do.
SHAPIRO: That's Maggie Koerth, senior science writer for FiveThirtyEight.
Thanks for joining us.
KOERTH: Yes. Thank you. Transcript provided by NPR, Copyright NPR.