New Project Aims to Address A Deadly Problem: Inconsistent COVID-19 Testing Data
Every day, KALW reports the number of new coronavirus cases in the Bay Area. But, we also know there aren’t enough tests available to confirm every case. Some people who experience symptoms are told by their doctors to just stay home and quarantine.
Alexis Madrigal is a Bay Area journalist who writes for The Atlantic. When he set out to report on coronavirus testing, he realized how inconsistent the numbers are across the country. Some states only report the number of tests that come back positive for the virus, some states report everyone who gets tested, and some states lag on reporting testing altogether.
Madrigal, along with a colleague at The Atlantic and a slough of volunteers, started a website called the COVID Tracking Project. It compiles all the coronavirus testing data from across the country and corrects for inconsistencies.
Madrigal explains what data we have, what data is missing, and what a lack of numbers means for fighting this pandemic.
Sarah Lai Stirland helped produce this interview. Click the play button above to listen.
The following is a transcript of the conversation. It has been edited for brevity and clarity.
KALW’s Marissa Ortega-Welch: Here we are in the middle of a pandemic. Why is this crowdsourced volunteer-run project the only source of this information?
Alexis Madrigal, The Atlantic: In every other country basically, the national health service does this; like the equivalent of our CDC puts out a map of all of Italy's provinces and what's going on in these different places. To be honest, when we - Robinson Meyer, my colleague at The Atlantic and I - started working on the testing story, we just sort of assumed that this data existed somewhere. In the U.S, it just didn't exist.
Ortega-Welch: You're doing this national project but you live in the Bay Area so I'm sure that you're following that data pretty closely. What do we know in terms of the data for the Bay and what's missing?
Madrigal: What's really interesting about the Bay is that we seem to be in fairly good shape relative to the rest of the country. It’s a little bit hard to tell though, right? And here's the reason why.
If you look at the testing situation for California over the last week, we've been averaging about 2,000 tests a day. We're the biggest state in the country and that is a tiny, tiny, tiny number of tests. Probably the state that's doing the best sort of relative job of testing is New Mexico and they basically have been doing the same amount of tests as the state of California. There’s what - a few million people in New Mexico? So we really don't know what's happening.
The only thing we know for sure is that a huge backlog has built up inside the state of California. By this, I mean, primarily at private testing labs contracted with the state of California. We just know that there's more pending tests than there are completed ones. So if those pending tests have a positive rate that's similar to those thus far completed, then California actually probably has the second largest state outbreak after New York. We don’t how those cases will then be distributed through the state on a county by county or regional basis, so it could be that the Bay is in tremendously good shape just by luck and by early action or it could be that we're actually hiding a problem that will soon show up in the form of hospitalizations.
Ortega-Welch: On that note, last week, six Bay Area counties announced that they would require labs to report all test results - both positive and negative test results - to state and local governments. I think a lot of people may have been surprised to realize that not all the test results had been reported before then!
Madrigal: The way we've just been talking about it is: we need the denominator. You know, you need to report all the tests because if you only have the positive [test results], that doesn't tell you the full story.
It's not just like, it'd be nice to know how many cases are in California. You need to be able to understand as quickly as possible where the hotspots are so you can direct resources to those places and really get moving. The other really, really important thing about testing though: you’ve probably heard a lot about “PPE” (personal protective equipment), this protective gear that healthcare workers wear. Basically the longer a test takes, the more PPE we burn through at the hospitals.
Ortega-Welch: Because when we don't know if somebody is sick or not, we have to continue to assume that they are sick?
Madrigal: Exactly. If [patients] have COVID-like symptoms, [healthcare workers] have to treat them as positive until they get a confirmed negative back. What that really means is this incredibly scarce resource - that all these people are trying to find ways of gathering to bring to hospitals - every day that goes by where we've got a test that's not coming back or we're not able to get tests, we're burning through more PPE.
Ortega-Welch: So you just told us why testing is important for understanding what’s going on within the hospital. For those of us that aren't in a hospital setting with the “shelter in place” order in effect, we are basically being told to assume that everybody has coronavirus and stay six feet away anyway. So why else is testing so important?
Madrigal: The thing is, less testing equals more aggressive social distancing measures. Basically the reason we're all having to shelter in place is because we have no idea how to actually surveill where [COVID19 cases] are. You think about places that have really had a lot of success with this in, in East Asia in particular, and a lot of what they're doing is having a more open economy, a more open society, but with much more, much higher levels of testing intensity.
Here, because we don't have testing, we're faced with this problem of having to really shut down society because we just don't know any better.
Ortega-Welch: A lot of news outlets now, KALW included, are giving daily updates on the coronavirus cases in our region. But you're saying that this data set is totally incomplete. Should we even be getting those updated numbers every day?
Madrigal: I've thought about this. I was on MSNBC over the weekend, and I was thinking about the way that they present this data as somehow conclusive and it's just clearly not. No one who has really thought about this for more than 30 seconds could conclude that that's the actual number of cases.
I think the right way to look at this is that every case you see is just a newly discovered case. Um, and it's just a subset of the total cases that will be out there. Um, and, and in many places, uh, people who have COVID-like symptoms are just told to stay at home, particularly if they're young. Uh, and they'll never be tested and we'll never know. I think the other, uh, numbers that many experts feel like are the truest indicators of the severity of an outbreak and it's sort of prevalence are the hospitalizations and the deaths.
Honestly, that’s basically been the problem throughout this whole thing for me is: if you don't do testing, the way you know how prevalent the disease is, is you count the bodies.