The letter articulates a regimen of increased testing, allowing the federal government to categorize various counties by risk.
“This is what we envision,” it reads. “Our expanded testing capabilities will quickly enable us to publish criteria, developed in close coordination with the Nation’s public health officials and scientists, to help classify counties with respect to continued risks posed by the virus. This will incorporate robust surveillance testing, which allows us to monitor the spread of the virus throughout the country. Under these data-driven criteria, we will suggest guidelines categorizing counties as high-risk, medium-risk, or low-risk.”
What that categorization actually means isn’t clear. Will “high-risk” counties be closed off from entry, like the area in New Rochelle, N.Y., that was locked down after it emerged as the epicenter of a number of coronavirus cases? Will “low-risk” counties not require that visitors avoid nursing homes?
The point of the distancing measures that are in effect is to limit the spread of the virus. Most of the counties in which there are the most cases currently confirmed are ones in which there are large cities. The New York City area, for example, has more confirmed cases than anywhere else in the country. A map of the number of cases by county reveals other cities that are similarly affected: Detroit, Denver, San Jose, Seattle, Dallas, Atlanta, New Orleans.
But those aren’t necessarily the places where the outbreaks are the most significant. If we control for population, identifying the number of confirmed cases per 10,000 residents in a county, different areas emerge as more heavily affected: Idaho, North Dakota, West Texas, Arkansas.
The change stands out more if we animate the two maps.
In this sense, the problem is more geographically distributed. Is a county with a low population and few cases low-risk? Is a county with no cases that’s adjacent to a county with a high density of cases high-risk?
All of this comes in the context of a dynamic that’s changing rapidly. We noted on Thursday that the number of confirmed cases in every state is climbing exponentially, even in more rural states such as Nebraska. There, the number of confirmed cases is rising more slowly than in other states, but it’s still seeing its case count double every four to five days.
Two weeks ago, Nebraska had about as many confirmed cases as Michigan. Now Michigan has more than 30 times as many, thanks to its rapid increase in cases, especially in Detroit.
But two weeks is, as you may know, a significant time period in the context of the coronavirus. It is believed to be the maximum period of incubation for the virus, meaning that some cases now being reported involved people who were infected weeks ago. That’s why Trump’s recommendation for increased social distancing was initially for a period of about 15 days. It was both an effort to contain the virus’s spread and to get a better sense for where the virus had already spread.
That brings us back to Trump’s plan. The figures above are only confirmed cases. There are tens of thousands of infections that haven’t been confirmed, because the people with the virus are asymptomatic, aren’t yet symptomatic or haven’t been confirmed as having the virus through testing. The president has repeatedly hailed the number of tests being conducted at this point, but it remains the case that most states have tested relatively few people.
The Covid Tracking Project is tallying the daily testing in each state and Washington, D.C., and publishing the results. In states such as Washington and New York, which saw early surges in recorded cases, there’s been far more testing. In many states, though, testing continues to be fairly slow, even as testing overall has increased.
Again, notice Nebraska on this chart.
The state has conducted only about 71 tests for every 100,000 residents. In New York, there have been about 532 tests per 100,000 residents. But it’s not only heavily affected states like New York that have much higher rates of testing. North Dakota has conducted about 233 tests per 100,000 residents.
Scaling up testing is important and will, to Trump’s point, give a better sense of where the virus is most problematic. Adding testing will also presumably cause rapid increases in the number of confirmed cases in a lot of places. And of course, with each day that passes, there are a number of new infections which may not manifest symptoms until the middle of April anyway.
What Trump’s letter implies is that a robust testing regimen and subsequent assessments of the country are required to relax the distancing restrictions — a goal he says he’d like to see in effect to some extent in 16 days. That seems unlikely. The question, then, is which gives: the testing regimen or the timeline?