Despite President Trump’s recent assertion that COVID-19 testing is “overrated,” expanded testing will be crucial for managing the coronavirus pandemic in the coming months. The current inability of states to fully count the number of people infected means they are missing opportunities to isolate infectious patients and trace their contacts to prevent further spread. Many states are reopening their economies with an incomplete picture of the number of COVID-19 cases because they haven’t tested enough. Comprehensive testing can provide an early warning that case numbers may be growing and alert policymakers of the need to deploy critical resources to interrupt spread and prevent a return to pre-shutdown disease levels.
But in our efforts to increase testing, we have become fixated on metrics that don’t address the real deficiencies. Trump has boasted that the United States has conducted more tests than any other country. While true, it’s also irrelevant. On a per capita basis, the number of tests conducted in the United States is lower than Iceland, Estonia, Italy, Norway, and Switzerland, among others. The United States does exceed each of these countries in one area: cases of COVID-19 per capita. America should test more people per capita than most other countries because it has more infections to find.
Estimates vary about the amount of testing the United States should be doing, ranging from the potentially achievable 500,000 tests per day to more than 20 million per day. The administration’s 2 percent target is one such guess, but it’s arbitrary. Why not 5 percent or 20 percent or any other number? The problem is this: The total number of tests conducted is a blunt and potentially misleading measure of whether testing levels are adequate.
The total number of tests doesn’t show whether they’re being performed in the right places or on the right people. For example, urban and rural areas have different testing needs based on population density: total testing numbers may seem high in cities, but they may obscure the fact that testing is inaccessible to certain residents or isn’t reaching those at greatest risk.
The United States needs a thoughtful strategy rooted in data and a combination of metrics that show how well infections in communities are being identified. As communities reopen — and assess whether it is safe to stay open — we should pay careful attention to positivity, the percentage of tests that come back positive, which is tied to the level of illness in a community. Ideally, this percentage should be low. If positivity is high, it indicates that states are testing only the sickest individuals and missing those with milder symptoms who may nevertheless spread the virus.
‘In many states, reported testing statistics probably include people who have been tested more than once.’
The World Health Organization has recommended that countries maintain for two weeks a positivity of less than 5 percent before they begin to relax social distancing restrictions. But in many US states, including several of those in the process of reopening, positivity is worryingly higher. If a state’s positivity begins to rise, it may portend an eventual growth in worrisome metrics like hospitalizations and deaths. States should respond by further increasing testing.
We also need better data, shared quickly. Though states are now reporting the numbers of tests performed — an improvement over earlier months — this number may not equate to numbers of people tested. In many states, reported testing statistics probably include people who have been tested more than once. There are also reports that states are including serology tests in their testing totals; such tests are not used to diagnose infection and may artificially inflate a state’s COVID-19 testing number. This potential for double-counting and inclusion of irrelevant test results distorts the data and calls into question this benchmark by which to gauge progress toward improved testing capabilities.
Finally, testing data is currently available mostly at the state level, obscuring whether testing is widely distributed within states or whether there are geographic holes in access to tests. We also can’t fully account for testing that occurs in private sector labs or identify which testing platforms are being used and what capacity truly exists. To expand testing capacity, we need to better understand which testing strategies work and which are subject to bottlenecks.
The recent stimulus package that passed in April includes $25 billion to expand COVID-19 testing and calls for the development of a strategic testing plan. These efforts must define the level of testing that is required to accurately characterize the occurrence of COVID-19 in the United States, inventory current limitations in testing, and establish a plan of action to overcome them. Instead of leaving states to fend for themselves, the federal government must lead in eliminating bottlenecks in the testing pipeline, such as shortages in testing supplies and reagents. It must also define how the expansion of testing will be monitored and outline the data requirements to ensure that progress is being made.
Without good testing data, we’ve been driving this pandemic response road blind, not knowing where we’re going or how fast. As more and more states reopen, it’s crucial that they have the capacity to identify and isolate COVID-19 cases rapidly in order to prevent case numbers from once again accelerating to pre-shutdown levels. But simply increasing the number of tests conducted does not ensure that we are on the path to better identifying coronavirus cases in the United States. To do that, we need smarter metrics, better data, and federal leadership.
Jennifer B. Nuzzo is the lead epidemiologist for the Johns Hopkins COVID Testing Insights Initiative and an associate professor at the Johns Hopkins Bloomberg School of Public Health. Lauren M. Sauer is the director of operations with the Johns Hopkins Office of Critical Event Preparedness and Response and an assistant professor in the Department of Emergency Medicine of the Johns Hopkins School of Medicine.