By Betsy Todd, MPH, RN, nurse epidemiologist and AJN clinical editor. Published April 20.

How is it that we in the U.S. make up 4% of the world’s population but account for more than 31% of global COVID-19 cases? Because at the start of the pandemic we weren’t prepared to test quickly and widely—and incredibly, after three months’ time, we still aren’t.

Coincidentally, both the U.S. and South Korea saw their first cases of COVID-19 on January 20. Two weeks later, South Korean scientists had perfected a diagnostic test for the virus, and infected people began to be identified and isolated.

Meanwhile, in the U.S. a series of problems and poor decisions held back test development. In early March, as the number of cases of COVID-19 exploded in parts of the country, Health and Human Services Secretary Alex Azar promised that by March 14, labs would be running four million tests per week. As of April 20, the total number of SARS-CoV-2 tests performed in the U.S. since the pandemic began was only 3.8 million, according to covidtracker.com. Our testing capacity is frighteningly low. Estimates of an adequate number of tests needed in the U.S. range from 3.8 to 10 million per week.

As of April 20, South Korea has 10,674 COVID-19 cases. The U.S., with just over six times as many people as South Korea (330 million vs. 51.5 million), has 760,570 cases—71 times as many infections (covidtracker.com). Other factors affect the number of infections in any country—including, notably, the disastrous lack of PPE in the U.S. that makes it nearly impossible to contain transmission within hospitals—but experts agree that in South Korea and several other Asian countries, the virus has been contained by the rapid and aggressive testing that makes possible the isolation of infected people.

Why is our lack of testing such a big problem? There are several reasons.

We don’t know the true number of COVID-19 cases in the U.S.

Although the U.S. already has more cases of COVID-19 than any other country, it has been suggested that the actual number of infections may be five or ten times the limited number we’ve been able to count. Every day, thousands of people are being told in EDs or via telehealth consultations that their symptoms point to COVID-19 infection. Those who are not seriously ill are sent home to recover—and never counted as actual cases.

Very few health care workers are being tested.

Many health care workers who have clearly been exposed to a patient with COVID-19—before that patient was diagnosed, or after diagnosis but while providing care with inadequate PPE—are being sent back to work. Almost certainly, some of these workers are infected and asymptomatically transmitting the virus.

We don’t know the true extent of asymptomatic transmission.

Asymptomatic infection appears to be unusually common with this virus. Just how many of us are walking around spreading virus, with no idea that we’re infected? Is asymptomatic transmission driving rapid spread? How long are asymptomatic people contagious? Do asymptomatic people shed virus for longer periods of time than people who had symptomatic infections but are no longer exhibiting symptoms? We can’t begin to answer these questions until millions more people have been tested.

Early in the pandemic, testing of passengers and crew aboard a cruise ship found that nearly half of those who tested positive were asymptomatic at the time of testing. A complicated model based on reported infections in China suggests that 86% of infections there were not recognized because those infected had only mild, limited, or no symptoms.

We don’t know whether the increasingly widespread use of masks by the public limits transmission.

We’ve never tried community-wide “droplet containment” as a strategy for preventing infections. If we could follow true case counts in an area before and after widespread masking, we might learn whether or not masks are helpful.

We can’t safely let up on stay-at-home and social distancing efforts without testing.

Drive-through testing during COVID-19 pandemic in Germany. By Henning Schlottmann

Test-based surveillance would give us more control during the pandemic. The trending of numbers and specific locations of cases could tell us when and where it’s safe to reopen at least some businesses, or allow small gatherings, or return to school or work.

Without testing, we have to rely on widespread social distancing and shutdowns for longer periods of time in order to manage the number of new infections without overwhelming hospital capacity. It’s estimated that despite the extraordinary (and incomplete) case count in the U.S., 90% or more of the U.S. population has not yet been infected. (Serological tests for antibodies, which may soon become more available, could clarify just how much of the population may have been infected without knowing it.)

Tom Frieden, the former CDC director, emphasizes that in order to manage social distancing and stay-at-home efforts, we need to test, isolate the infected, identify contacts of the infected person, and then quarantine those contacts. “And each of these four things needs to be going really, really well and at a massive scale.”

The authors of a new report from the Johns Hopkins Bloomberg School of Public Health agree:

“In order to relax community mitigation efforts and other measures to reduce COVID-19 transmission, it is essential to rapidly test all symptomatic cases of COVID-19, identify and isolate all positive cases, and conduct contact tracing for all close contacts of each and every case.”

Coronavirus testing capability in the U.S. is slowly catching up with testing in other developed countries, and rapid point-of-care testing as well as large-scale automated testing methods are starting to be deployed around the country. But it will still be several weeks, at least, before we have the testing capacity we need.