Article Correctness Is Author's Responsibility: Why Aren’t Cities Widely Testing for Coronavirus?

Updated: 2020-03-11

On March 1, on a return flight to her home in Washington, D.C., after a five-day work trip in Bangkok, Maggie McDow came down with something.

The symptoms — aches and fatigue, tightness in her chest — come and go. “One minute I think I’m getting better and the next I’m having trouble breathing and can barely lift my head off the pillow,” McDow wrote in an email on Saturday from her home in D.C.’s Forest Hills neighborhood, where she has been on self-imposed quarantine since her return. “I don’t know if this is Covid-19, but it [is] definitely different [than] anything I’ve had before.”

Because she traveled through an airport in South Korea — where some 7,400 people have been diagnosed with Covid-19 as of Monday — McDow was concerned. A colleague who traveled with her came down with the same symptoms, she says, but in that colleague’s home in rural Indonesia, tests for the coronavirus were not yet available. To McDow’s surprise, getting tested in D.C. would be difficult, too.

Last week, McDow bounced around between her doctor, George Washington University Hospital, and the D.C. Department of Health. On Friday afternoon, she checked in at the hospital’s emergency room. No luck: Even after she tested negative for the flu, she said that the city’s health department declined to run her test for coronavirus. The reason? Since she never left the airport in Seoul, she was deemed at low risk for infection. (The Department of Health did not return requests for comment; George Washington University Hospital referred CityLab to the city.)

McDow described her Kafkaesque journey through the medical system in a Facebook post that was quickly shared more than 20,000 times. On Monday night, she finally got tested for Covid-19: negative. It’s a relief in more ways than one, since she can now see her doctors, who couldn’t treat her before, for whatever is ailing her. She can see her daughters again, too.

McDow’s story has parallels with that of Robin Shulman, who wrote in The New York Times about making three trips to the emergency room before she was given a coronavirus test. (It was also negative.) Chris Hayes, the MSNBC host, reports the same trend in San Diego. “In order to get a test the patient has to have had close contact [with] someone [with] a confirmed infection or travelled to an infected area,” he writes. “But that's totally insufficient!”

As President Donald Trump has repeatedly stated, coronavirus cases in the U.S. are still fewer than those being reported in France, Germany, or Spain, to say nothing of the many thousands more cases in hot spots like Italy or Iran. Yet the reported number of cases in the U.S. may simply reflect the fact that so few tests are being administered. Given the experience of other nations, it’s almost certainly the case that test results in the U.S. fail to convey the true severity of the problem.

“The numbers of ‘new’ cases reported daily in the U.S. are not new,” writes Marc Lipsitch, director of the Center for Communicable Disease Dynamics at Harvard University, in a post on Twitter. “They are newly discovered as we start to test more. Testing is still completely inadequate, and actual case numbers are much larger than the numbers we’re hearing because most cases never get tested.”

This coronavirus map of Europe and North America from Johns Hopkins University’s now well-known Covid-19 tracker, in other words, in part reflects national differences in how those areas are testing and monitoring the disease.

A screenshot from the Johns Hopkins University’s tracker of Covid-19 cases on March 10. (Center for Systems Science and Engineering (CSSE) at Johns Hopkins University)

America’s testing regime presents a problem as authorities struggle — maybe in vain, at this point — to keep local outbreaks in the U.S. from becoming regional epidemics like the one that has led authorities in Italy to put the entire nation under quarantine. Decisions at the state level may be contributing to bottlenecks that are exacerbating a crisis that began with a botched test rollout from the U.S. Centers for Disease Control and Prevention. The CDC removed testing totals from its website on March 2.

Dan Hanfling, the vice-president for In-Q-Tel, a nonprofit venture capital firm for U.S. intelligence and defense communities, says that the lack of testing isn't so much a decision to ration tests as much as a lack of access to them. Hanfling — who is also a clinical professor of emergency medicine at George Washington University and co-chair of the National Academy of Sciences Forum on Medical and Public Health Preparedness for Catastrophic Events — says that the limited number of laboratories able to process tests and the initially strict criteria for testing were also limits on capacity.

“More kits are coming to the local and state health departments, and both Quest and LabCorp announced on Monday that they now have the ability to run commercial tests for those who desire them,” Hanfling says.

On Saturday, D.C. Mayor Muriel Bowser gave a press conference to announce the city’s first positive case of Covid-19, one with apparently no connection to international travel, suggesting the dreaded indication of social spread. (She also fielded a question about McDow, who says she was surprised to learn from a press conference that she was feeling better.)

Two days later, 39 patients total have been tested for for Covid-19, according to a website where the city is posting coronavirus data. That’s in a city of more than 700,000 and the nation’s capital, where both the extent of international travel and the national security context might suggest that an outbreak could be a particular concern. Yet the city’s health lab can process up to 50 tests per day, D.C. Department of Forensic Science director Jenifer Smith told WAMU.

Other cities aren’t saying as much as the District. Cities in Texas are only reporting positive test results, for example. A spokesperson for the City of Austin said that they’re following protocols established by the Texas Department of State Health Services and the CDC; local officials didn’t respond to a follow-up question about whether Austin was allowed to provide more details about their coronavirus testing procedures.

In Washington’s King County — which is dealing with the largest known outbreak of Covid-19 in the country as of Tuesday — the capacity for coronavirus testing is much greater. Both Seattle and King County health officials say they do not typically conduct testing themselves, but the virology department at the University of Washington alone is processing more than 400 tests per day. Still, it’s unknown how many tests are being conducted overall, or if patients in Seattle are being refused tests. Local officials referred CityLab to the state about why that information isn’t public.

In the absence of official U.S. testing and tallying efforts, a group of coders, analysts, scientists, journalists, and others are working to follow coronavirus testing across the country through an open-sourced database called the Covid Tracking Project. The site aggregates and publishes data on state-level testing efforts, updating the numbers daily. The group is upfront about the limits of what is possible to learn; for example, authorities in California, New York, and Washington State are not regularly publishing data on testing.

It’s hard to know how big a problem Texas faces. So far the Lone Star State is acknowledging 12 positive cases total. Maybe that means that the coronavirus isn’t messing with Texas and its 29 million residents yet — but the global experience with the spread of this bug points to a different conclusion. Twelve cases could simply indicate a small denominator: If very few people are tested, then very few people will test positive.

Compare this stingy trickle of coronavirus data with the official response from the city-state of Singapore, which published a detailed coronavirus dashboard that allows anyone to track the status of anyone who has come down with Covid-19 across a densely settled expanse of 6 million people. The World Health Organization praised Singapore’s “all-of-society, all-government approach” to finding people with the virus, tracking their social contacts, and publishing details about the places where they live or visited in order to contain the disease.

That panopticon appears to have kept the coronavirus bottled up. The Straits Times even has a graphic feature showing how each of Singapore’s 166 confirmed cases are linked; the rate of new cases has flattened and is now declining. China and South Korea have also seen their rates of new infections dip. Their interventions came early: Guangzho implemented strict social distancing protocols when the region had just seven confirmed cases (and zero deaths), according to Lipsitch. Wuhan went on lockdown when they had 495 confirmed cases (and 23 deaths). The U.S. is now further along than those places were — but with far fewer tests in the field to guide public decisions.

Health experts say that efforts to trace the contacts of infected individuals in order to contain the virus are now overwhelming health departments in the U.S.; mass testing only works in the early stages of a contagion outbreak. Now, it’s critical to lower the demand for medical attention before the nation’s healthcare system is overwhelmed (a real possibility). The way to flatten that demand curve is to make the painful decisions associated with social distancing. For the New York suburb of New Rochelle, that means a one-mile “containment” zone that will be administered by the National Guard. Soon, for a host of North American cities, that may mean mass closures of schools, workplaces, public gatherings, and other high-risk sites and events.

As local authorities begin to make those hard calls, questions about testing are starting to look more like woulda, coulda, shoulda. As of Tuesday afternoon, the Covid Tracking Project’s findings show 4,449 total tests nationwide, with 566 positive results and 409 cases pending. That looks like a reassuringly tiny percentage of people in a nation of 329 million are walking around with this bug. Keeping it that way, though, likely looks less like testing centers and more like proclamations. The federal government failed to do more to get tests to the public in time, and bottlenecks at the state and local levels proceeded from there — notwithstanding Trump’s repeated claims that everyone with symptoms “gets a test.”

“What we are learning from this experience is the following: If you cannot see an epidemic developing, and cannot diagnose it, you will not be able to treat it and control the spread of disease,” Hanfling says. “What is really needed is a complete reset on how we think about diagnosing emerging infectious diseases. The technology to develop rapid, point-of-care effective diagnostics — think pregnancy tests — exists now.”

He adds, “What is missing is the commitment to build those capabilities such that they become the manner by which future epidemics are managed. See it, diagnose it, treat it, contain it.”