Major Undercount In COVID Cases Makes Our Tracking Data Less Useful

17:26 minutes

an abstract hazy image of multiple line graphs superimposed on top of each other. it's chaotic, some are hazy
Credit: Shutterstock

For many, it’s become routine to pull up a chart of COVID-19 case counts by state or county. Though imperfect, it’s been a pretty good way to assess risk levels: Follow the data.

But recently, that data has become even more imperfect, and less useful at determining individual risk. Thanks to a variety of factors, case counts are now so inaccurate that a COVID surge could be missed entirely. 

“We are really flying blind,” said epidemiologist Katelyn Jetelina, assistant professor at the University of Texas School of Public Health and the author of the newsletter, Your Local Epidemiologist.

How Did We Get To This Point?

Currently, for every 100 COVID-19 cases in the United States, only seven are being officially recorded, according to projections from the Institute for Health Metrics and Evaluation. As a point of comparison, during the Delta wave 43 out of 100 cases were recorded, and during the Omicron wave the figure was 26 out of 100 cases

The reasons behind the current undercount are due in part to the unintended consequences of good public health policies, like increased vaccinations and the availability of at-home tests, both of which lead to fewer cases being included in official CDC data. Mild cases are more common now, thanks to vaccines and changing variants. 

“People may just not get tested because they just have the sniffles,” said Jetelina. 

Donate To Science Friday

Invest in quality science journalism by making a donation to Science Friday.


Others may forgo testing altogether. The virus can spread asymptomatically from there.

“We just haven’t done the groundwork as a nation to systematically capture these cases,” said Jetelina. 

The majority of at-home tests are not tracked in official case counts. And, beginning last month, medical providers were no longer able to submit claims for COVID tests for uninsured patients

“In other words, we removed testing incentives,” she said. “Unfortunately, this means that testing disparities will soon follow. Just like we saw in the beginning of the pandemic, the poorest neighborhoods will have even more depressed case numbers than [wealthier] neighborhoods.”

Adding to that confusion, the CDC changed its risk level calculations in February. Overnight, the color-coded map of the United States went from mostly red to mostly green. Risk levels and mask guidance are now determined based on the likelihood of hospitals being overwhelmed by COVID cases. These risk levels do not accurately capture risk levels for individuals to catch covid, said Jetelina.

“This has a major limitation though,” she said. “If we just rely on hospitalizations, it won’t truly reflect the risk and transmission in that community for people to inform their behavior.”

How You Can Judge Individual Risk

That doesn’t mean it’s impossible to get a broad sense of what’s going on in your community. But, Jetelina urges people not to use static numbers or case thresholds to determine when to loosen up on pandemic restrictions. 

Instead, she recommends paying attention to trends, like test positivity rates, percent change of cases, or wastewater data. However, not every county collects COVID-19 wastewater data and it’s not a substitute for comprehensive clinical testing. 

“Wastewater can contribute to our holistic views of a virus’ spread across a community, but not necessarily a threshold in which you [decide if you] should wear a mask or not wear a mask,” she said.

If these trends are increasing rapidly in your area, it’s time to increase your COVID-19 precautions, said Jetelina. 

The same tried and true principles will work no matter the variant: mask up in crowded indoor spaces and get tested before seeing elderly and immunocompromised friends and family. 

With a pandemic where so much has changed, public health advice hasn’t. 

Segment Guests

Katelyn Jetelina

Katelyn Jetelina is an adjunct professor in the UTHealth School of Public Health, and author of the Your Local Epidemiologist newsletter.

Segment Transcript

IRA FLATOW: This is Science Friday. I’m Ira Flatow. At some point during this pandemic, you’ve probably pulled up one of those charts, you know, showing how many COVID cases were in your area. Maybe you even looked at something called the seven-day rolling average. Though imperfect, it’s been a pretty good way to assess COVID risk levels– follow the data.

But now, our case counts are inaccurate enough that they may actually be missing a surge. And on top of that, the CDC changed its risk level calculations in February. So why now, two years into the pandemic, is our data so spotty? And what should you do to calculate risk?

Joining me now to help us better understand what’s going on and how you can still understand COVID risk despite it all is my guest Katelyn Jetelina, assistant professor at the UT Health School of Public Health based in Dallas, Texas, and author of the newsletter Your Local Epidemiologist. Welcome to Science Friday.

KATELYN JETELINA: Hi, thanks for having me.

IRA FLATOW: Let’s start off. Why are our COVID infection rates not as accurate as they could be or are they once were? I mean, one of the reasons could be, as Dr. Fauci pointed out, is that we are all self-testing more than we used to and not reporting the results. Could that be right?

KATELYN JETELINA: That’s right. You know, we saw during the Omicron wave that the use of antigen tests more than tripled compared to the Delta wave. And while this is fantastic news– it’s a tool that’s been chronically underutilized– we just haven’t done the groundwork as a nation to systematically capture these cases. I do know that some local jurisdictions have implemented systems for communities to report at-home antigen results. But uptake has been less than optimal.

IRA FLATOW: OK, so what are some of the other reasons these rates are not being reported or they’re not as accurate?

KATELYN JETELINA: We also just have less testing period. Last month, around March 22, providers were just no longer able to submit claims for tests for uninsured patients. So in other words, we removed testing incentives. Unfortunately, this means that testing disparities will soon follow. Just like we saw in the beginning of the pandemic, the poorest neighborhoods will have even more depressed case numbers than high neighborhoods.

The other reason is asymptomatic cases. With vaccines, with infection-induced immunity, there are just going to be more mild cases than ever. And because of this, there will just be more, one, asymptomatic spread. Two, people may just not get tested because they just have the sniffles. Or three, they just find no concern of being infected, so won’t test.

IRA FLATOW: Do we know how big of a gap there is, then, between the true infection rates and what’s being reported in the figures?

KATELYN JETELINA: We have a rough sense. I will say, you know, in the UK, they’ve started hinting that official reported cases can’t be trusted. One researcher looked at comparing the official case records in the UK to survey prevalence. And they found a dramatic decoupling starting in January 2022 because of removing freely available options for testing.

In the United States, I will say that estimates are even more rough, because we just don’t have such a strong public health surveillance system as the UK. But the latest numbers that I have seen are from the Institute of Health Metrics and Evaluation over in Seattle. You know, in the past day or two, they estimate that for every 100 cases in the United States, only six or seven are being officially recorded in our surveillance system.

The gap between reported cases and, quote unquote, “true cases” has dramatically widened over time. During the Delta wave, there was an estimated 43% of cases reported. During Omicron, about 26% of cases recorded. And like I said, right now, we have about 7% of cases reported, which is abysmally low. We are really flying blind.

IRA FLATOW: So we could be in, already, a COVID wave and just not know it because we don’t have the data.

KATELYN JETELINA: That’s true. I mean, what we can look at is trends. So our trends will still be pretty accurate. So for example, in the Northeast, we know that trends are exponentially increasing for cases. The thing that this changes is that instead of, for example, I don’t know, 1,000 new cases per day, it’s about 14 times higher than that. And so really it’s the level of transmission in our community is much higher than our official statistics are reporting.

IRA FLATOW: Now, we typically do follow Europe, right? A few weeks later. And they’ve recently had a surge.

KATELYN JETELINA: That’s exactly right. So the United States, throughout the pandemic, has almost mirrored exactly what happens in Europe. And the only exception, and really interestingly, is during the Alpha wave, which was at the same time period, April of last year, that the UK was hit really hard with Alpha variant and only a few states like Michigan was hit hard in the United States. So we’re not sure what’s going to happen with the BA.2. The United States may get hit hard in the coming weeks, but we may not. We’re kind of at the mercy of time right now.

IRA FLATOW: And when you say the mercy of time, would it also be the mercy of the weather and the season, because it’s getting warmer and people are outside more?

KATELYN JETELINA: That’s right. So there’s a few things that may be going to our advantage in the States. One is the weather. You’re right. It is getting warmer. People are outside. And coronaviruses typically thrive in the winter.

The other thing that is a little more nuanced and not really reported is that our first Omicron wave in the United States was actually driven by subvariant BA.1.1, which is different than what was hit in the first Omicron wave in Europe, which was BA.1. And so this slight difference– BA.1.1 is a little more transmissible than the original Omicron. And because of that, we may just not have as much more wood to burn in the United States as Europe did with BA.2.

IRA FLATOW: Sounds like software technology numerology has moved over to epidemiology.

KATELYN JETELINA: Yeah, it’s a little difficult to follow. And like I said, the nuance is small, but it may make a big impact when we’re talking about population-level statistics.

IRA FLATOW: Well, let’s talk about this imperfect data. Where are we seeing COVID cases rising? If you say follow the trends, what are the trends?

KATELYN JETELINA: Well, I will say in the past two months, we’ve been consistently nosediving in cases. But currently, the US is actually showing a modest 10% increase at a national level. And relative to Delta and Omicron waves, we’re at really low recorded case numbers, which will help soften the impact if BA.2 is found a threat here in the States.

On a regional level, the Northeast is showing the most case acceleration. Rhode island is the leader, with about 100% increase in cases, followed by Maryland, DC, New Jersey, New York. There’s also, interestingly, some other random states that I’m paying attention to, including Mississippi and Oregon. These obviously aren’t in the Northeast. And the reason I’m interested in it, because they may be the initial seeds that will spark those regional hot spots.

IRA FLATOW: You know, one of the reasons we hear about why the case numbers are important is that some experts argue it’s because of hospitalization, right? Hospitalization numbers are more important, as they are what put a strain on the health care system, and not the milder cases.

KATELYN JETELINA: That’s true. There’s been a lot of focus on hospitalizations and just using that metric to determine individual-level behavior. In fact, that’s why the CDC changed its whole transmission map a few months ago, was so we can focus on just not straining the hospital systems. I do think that this has a major limitation, though, because people are still trying to, and rightfully so, avoid infections and/or break transmission chains for the vulnerable. And if we don’t have valid data, robust case data, and just rely on hospitalizations, it really won’t truly reflect the risk and transmission in that community for people to inform their behavior.

IRA FLATOW: Good point, because there still are lots of people who are not immunized. And I’m thinking of the youngest people in our population.

KATELYN JETELINA: That’s right, the youngest. I mean, I have two kids under five, and we still don’t have vaccines for them. The other reason is we have long COVID. We know that vaccines reduce the chance of long COVID, but they’re not also perfect. And long COVID isn’t something that the average Joe wants, either. And so paying attention to transmission levels is still very important right now.

IRA FLATOW: Yeah, and I think there are some people who are feeling self-conscious now about wearing a mask in public, because so few people are. And I was on, you know, mass transit the other day, where there is still a rule about wearing your mask, and I would say that at least half the people were not wearing a mask.

KATELYN JETELINA: Yeah. I mean– yeah. There is. There’s this kind of herd mentality that influences behavior like wearing masks.

I will say I hope that, with time, we can normalize the use of masks. This is done over in Asia all the time. If someone has a cold or the flu, they wear a mask out of respect. And so I think we really need to try and move that direction. Unfortunately, masks were incredibly politicized here in the States. And so people also see them as signals and messages, not just for health, which I think is disappointing.

IRA FLATOW: Yeah. Yeah. With the holidays coming up this weekend, as an epidemiologist, will you be watching, let’s say, five, 10 days after the holiday season to see if there’s been a spike?

KATELYN JETELINA: I always think it’s interesting to see the impact of the holidays. The challenge is that it’s close to impossible to weed out the differences between the impact of behavior like the holidays with the impact of the viral dynamics of just BA.2 catching ground. I would assume the interaction of the both of those would really springboard us into more transmission and more exponential growth. But again, it may not. We kind of just have to see where this virus takes us.

IRA FLATOW: One of the ways of seeing where the virus takes us and tracking COVID-19 rates is analyzing wastewater. Is that a good stopgap measure in lieu of good testing data?

KATELYN JETELINA: Yeah, so wastewater detection is very useful in predicting and indicating early waves. It’s thankfully implemented in quite a few cities, but unfortunately not all of them. So it’s not a reliable metric for all areas.

The other problem with wastewater is it’s meant to be used in addition to and not a substitute for clinical testing data. And the reason for that is because there’s really no consensus on a direct comparison between wastewater concentration, or RNA concentration, and clinical case numbers. Wastewater is really only used to indicate trends. And so wastewater can, again, contribute to our holistic views of a virus’s spread across a community, but not necessarily a threshold in which you should wear a mask or not wear a mask.

IRA FLATOW: Speaking of wearing a mask or not wearing a mask, and people trying to assess their risk, the incompleteness of the data makes it hard for individuals to do that, right? Let’s listen to what Dr. Fauci said in a recent interview with ABC’S This Week.

ANTHONY FAUCI: What’s going to happen is that we’re going to see that each individual is going to have to make their calculation of the amount of risk that they want to take in going to indoor dinners and going to functions.

IRA FLATOW: You’re on your own, basically, is what he’s saying.

KATELYN JETELINA: Yeah, unfortunately, this has been a theme in the United States throughout the pandemic, is this individual-level response. And I was quite surprised when Fauci said that, honestly, because he’s taken a very population-level approach to the pandemic until now.

And it makes it almost impossible, nearly impossible, for individuals to navigate, because this landscape does continue to shift. There aren’t reliable data for them to rely on. And it’s really difficult for individuals to navigate this risk because this landscape does continue to shift. Our case numbers are crumbling, and there’s really no other metric people can use to determine their behaviors or to even navigate this impossible landscape.

And so while I guess I kind of understand the individualistic approach, you know, infectious diseases violate the assumption of independence. And we are in this together, and we really need population-level strategies to continue and that guidance to continue, because we are still in a pandemic.

IRA FLATOW: Just a reminder, this is Science Friday from WNYC Studios. I’m sure Dr. Fauci would say that even though you’re on your own, there are still things you can do, like still wear your mask indoors, avoid getting in close situations, good ventilation, right? I mean, there are still those many-layered things that one can do.

KATELYN JETELINA: That’s true. There’s lots of layers that people can do. And you mentioned some great ones– using antigen tests to break transmission chains, wear masks, advocate for better ventilation and filtration in workplaces and in schools. I think the challenge that people have, though, is, when should you wear a mask, and when can you take it off? This pandemic is a marathon. It’s not a sprint. And so we desperately need some sort of guidance on a population level on what individuals should and should not do.

IRA FLATOW: So then, what should a person do? I mean, what’s the best advice you can give us?

KATELYN JETELINA: Yeah, so like we said in this, you know, raw case counts aren’t necessarily accurate. So I would suggest, and I would urge people not to use static numbers or case thresholds like 50 cases per 100,000 to determine behavior. Instead, I would use trends– wastewater trends if you have them in your area, test positivity rate trends, or even case trends. And if these trends are increasing in your county, and if they’re increasing fast, it’s time to dial up your COVID-19 layers. So you wear a mask, or you be sure to leverage antigen testing to break transmission chains before seeing the vulnerable.

IRA FLATOW: Dr. Jetelina, is this something that we’re just going to have to live with and continually adjust our expectations because this virus is not going anywhere real soon?

KATELYN JETELINA: That’s right. You know, this virus is not going anywhere real soon. Unfortunately, I do think that this landscape will continue to change, because we know the virus will continue to change. Hopefully, it will get easier and easier with time, because we will have more people boosted. We’ll have a lot more antiviral treatment. But we never know what the virus is going to throw to us next, and so we need to be proactive and really pay attention to how things are changing on the ground.

IRA FLATOW: Thank you very much for taking time to talk with us today.

KATELYN JETELINA: Yeah, thanks for having me again.

IRA FLATOW: Dr. Katelyn Jetelina, assistant professor at the UT Health School of Public Health based in Dallas, and author of the newsletter Your Local Epidemiologist.

Copyright © 2022 Science Friday Initiative. All rights reserved. Science Friday transcripts are produced on a tight deadline by 3Play Media. Fidelity to the original aired/published audio or video file might vary, and text might be updated or amended in the future. For the authoritative record of Science Friday’s programming, please visit the original aired/published recording. For terms of use and more information, visit our policies pages at http://www.sciencefriday.com/about/policies/

Meet the Producers and Host

About Shoshannah Buxbaum

Shoshannah Buxbaum is a producer for Science Friday. She’s particularly drawn to stories about health, psychology, and the environment. She’s a proud New Jersey native and will happily share her opinions on why the state is deserving of a little more love.

About D. Peterschmidt

D. Peterschmidt is a producer, host of the podcast Universe of Art, and composes music for Science Friday’s podcasts. Their D&D character is a clumsy bard named Chip Chap Chopman.

About Ira Flatow

Ira Flatow is the host and executive producer of Science FridayHis green thumb has revived many an office plant at death’s door.

Explore More