“I don’t want this job.”
It was late March, and Lauren Gardner, an engineer and epidemiologist at Johns Hopkins University, was questioning her decision to create an online dashboard for tracking COVID-19 cases and deaths around the world. She and her team had spent a long night in January scanning local news sources in China, intending to create a data set that other researchers could utilize. Before posting a link to their work on Twitter, Gardner had decided to visualize the data on a map in order to make it easier to parse. “Humans are horrible at statistics,” she says, and “presenting raw numbers is really tricky.”
The response was electric. In a matter of weeks, the user-friendly dashboard had attracted users not just in pandemic command centers around the world—from Italy’s health ministry to Connecticut’s governor’s office to the White House—but on social media, where fellow scientists, journalists, and armchair virus trackers followed its rising case counts with growing alarm. As of mid-June, the dashboard has garnered 650 million cumulative page views, making it one of the most popular sites in the world.
But for Gardner, codirector of Johns Hopkins’s Center for Systems Science and Engineering and a specialist in the role of mobility in spreading disease, the dashboard’s success only underscored the problems with our public health infrastructure, particularly in the United States. The Centers for Disease Control and Prevention’s data has been “very, very poor,” she lamented. The organization “is reliant on the states to get data to them, and not all states comply. . . It would be really nice if we could learn some lessons and come out of this with a better plan in terms of how we collect and provide and share data so that I don’t ever have to deal with this again.”
Even some of the world’s top medical experts are continuing to struggle to find accurate data about the novel coronavirus. In May, for example, the CDC was reporting that California had conducted 925,000 tests, while the California Department of Public Health was saying that it had conducted more than 1.1 million. CDC numbers differed from states’ own testing numbers by more than 150,000 in Florida, Massachusetts, and Texas, as well, according to the COVID Tracking Project.
“CDC should be putting all this data forward, including deaths by race, by location, by age,” says Andy Slavitt, who served as acting administrator of the Centers for Medicare and Medicaid Services under President Obama. “We’re not asking for perfection, just for transparency. Right now, medical providers and hospitals aren’t even providing data to CDC on what treatments are working.”
That has left entrepreneurial researchers such as Gardner with the painstaking task of finding and verifying local sources of coronavirus data, and then updating it as new numbers pour in. It’s a never-ending process, particularly on such a global scale, but Gardner is committed to keeping her dashboard up-to-date and available to all. Her team has grown from two graduate students to two dozen—plus professors, university staff, and support from mapping software provider Esri and Amazon Web Services. Since February, she has been waking up to hundreds of emails from around the world, some complimentary, and some complaining of inaccuracies. (Site visitors from France have been particularly exacting about perceived delays in her calculations, and after criticism on social media, she adopted the State Department’s naming conventions for contested states like Taiwan.) Gardner herself is only too aware of the problems. “This is just reported cases; it’s not true cases,” she says of the numbers displayed on her dashboard. “There are probably 10 to 20 times as many cases in the world as are reported in our dashboard. Almost surely.” She has begun to normalize the data so that visitors can understand COVID-19’s spread by more useful metrics, such as incidence rates, case-fatality ratios, and U.S. testing and hospitalization rates.
In the absence of strong, centralized public health leadership—and in the presence of digital platforms that thrive on misinformation—Gardner increasingly believes that it is the role of individual scientists to speak up, whether on TV or on Twitter. “While it is super time-consuming and distracting, I feel like it’s my responsibility to share things that I know, rather than just watching [celebrities like] Jenny McCarthy tell people they shouldn’t get vaccinated,” she says. “You have to put yourself out there.”
Now more than ever, she believes it is critical for scientists to use statistics—the right, contextualized statistics—to paint a picture that the public can understand. “Why isn’t there a national vaccination data set at a county level?” she asks. She recently built one, for measles.