Test big or stay home: Why the US needs millions of coronavirus tests per day

By one estimate, America may need 35 million Covid-19 tests per day for people to return to work.

We now know you can’t effectively fight the coronavirus pandemic without widespread testing to find out who has the disease. Developed countries that have managed to keep their case counts and deaths tolls low or bring them way down — including Iceland, Germany, and South Korea — have generally tested a greater proportion of their population than the United States.

It’s especially critical for finding those people who may be spreading the virus without showing symptoms.

Rani Molla and Dylan Scott/Vox

With the most confirmed cases and confirmed fatalities in the world and a devastating spike in unemployment, the US situation is stunningly costly and the response inadequate. The status quo is untenable.

Christina Animashaun/Vox

To truly end the crisis, keep deaths low, and allow the economy to recover, testing has to happen on an enormous scale.

How enormous?

Vox’s Ezra Klein has gone through some of the major proposals from several think tanks and researchers mapping the routes out of the pandemic. All of them involve testing huge numbers of people. One of the lower-end benchmarks estimates the US will need 750,000 tests per week. The high-end proposal, from Nobel laureate economist Paul Romer, starts at 22 million tests per day and goes up. And not just one test per person, but repeated testing over time.

Test millions. Test early. Test late. Test over and over. Test until the whole damn pandemic is over.

Following through on this would require a massive increase in government funding, huge numbers of workers trained to administer the tests, and the coordination of raw materials, manufacturing, and delivery across the entire global economy. It’s a costly, brute-force approach. But it may the only way out of the expensive and blunt measures already deployed to control the pandemic, from social distancing to shelter-in-place orders.

However right now, the US is struggling to test even 100,000 people per day, so getting to the scale some researchers suggest would demand a huge ramp up in capacity. Yet, they say, it’s possible to do it. And despite the immense financial and social costs of a national testing system, it may yet be the cheapest path to reopening the economy.

Why testing is so critical to controlling the Covid-19 pandemic

SARS-CoV-2, the virus that causes Covid-19, is almost perfectly optimized to infect huge swaths of a population.

For one, it can spread directly and easily between people in close contact. Yet the symptoms of Covid-19 can be confusing, varying person to person, making it tricky to identify suspected cases. Meanwhile, asymptomatic carriers — up to half of the total number infected — can spread the virus unwittingly for weeks, triggering outbreaks in their wake.

The virus can also turn dangerous and deadly in some people, particularly high-risk populations. But it can also be dangerous in a small percentage of otherwise healthy people.

While most can fight off the infection on their own, an epidemic like the one the US is facing with hundreds of thousands of infected people means tens of thousands will need to be hospitalized. Those patients in urgent or critical care can then spread the virus to health workers — a particular problem because of the shortage of personal protective gear — who end up sidelined and further strain the capacity of the health system to care for the sick.

And since it’s a new virus, there is no cure, vaccine, or widespread immunity to SARS-CoV-2.

Controlling the spread of the pandemic then demands finding the infected and isolating them until they can no longer spread the disease, alongside broader measures like social distancing. With an untold number of asymptomatic carriers, the only option to find out who truly has the virus is to test.

Mario Tama/Getty Images

Meanwhile, a lack of testing hampers the response to the virus. Health officials can’t preempt outbreaks in new regions. The threat then silently persists, infecting, killing, and draining resources.

The main kind of tests needed are those that look for active infections, like RT-PCR tests that can detect the virus’s genetic material. There are also cheaper and faster serological,or antibody, tests that can identify past infections. But to control ongoing spread, screening for active infections is critical.

Why testing on a massive scale could help the US economy recover

It’s hard to overstate the enormity of the challenge of ending a pandemic.

Gregory Gray, an infectious disease epidemiologist at Duke University, explained that while the virus is spreading rapidly, the major epidemics seen so far in several countries are still only infecting a tiny sliver of the overall population. Even if one were to presume that the roughly 1.5 million cases diagnosed around the world so far represent a quarter of the total number of the infected, those 6 million infections leave more than 99 percent of the population still at risk of eventual SARS-CoV-2 infection.

“No matter whether you live in an area where transmission has been high (Hubei [China], Italy, New York City) or relatively low, there remain many susceptible people,” Gray said in an email. The reproduction rate of the SARS-CoV-2 virus in a population, a variable known as R0, also remains high in many parts of the world.

When the R0 is greater than 1, the outbreak continues. When less than 1, the outbreak will end. A recent paper the Centers for Disease Control and Prevention calculated that the virus had a median R0 of 5.7 during the outbreak in Wuhan, China, up from prior R0 estimates ranging between 2.2 and 2.7.

The magnitude of this problem — and the threat of having tens of thousands or more severely ill people requiring hospitalization at the same time — is why so many parts of the world are being forced to deploy drastic, onerous measures, like lockdowns and aggressive social distancing.

But replacing these tactics would require something just as massive to halt the spread of the virus, let alone gain ground against it, explained Danielle Allen, director of Harvard University’s Edmond J. Safra Center for Ethics.

So testing has to occur on a vastly larger scale if it is going to help relax or replace social distancing. Right now, there are still not enough tests in the US to meet all the clinical needs, let alone widespread population testing. The health sector is focusing on testing patients who are symptomatic and people who think they may have been exposed. That leaves a huge segment of the population untested with potential spread.

To end the pandemic, the US instead needs to actively seek out potential spreaders, particularly the asymptomatic spreaders. And that requires tests. A lot of them.

The Safra Center put out a series of white papers on the various components of ending the pandemic. The paper looking at testing for Covid-19 suggests that between 5 million and 20 million tests per day will be necessary, between 2 and 8 percent of the US population, coupled with contact tracing to target the tests. (The Safra Center is planning to publish a more in-depth report on the recovery process this week.)

Paul Romer, a professor of economics at New York University, wants even more: 20 to 25 million tests per day, plus more frequent testing for frontline workers exposed to the virus.

He has run his own simulations of the impact of testing on the spread of Covid-19, finding that an ongoing testing regime reduces the number of people that need to be isolated, allowing the remainder to return to work. That’s true even if the test isn’t perfect at identifying spreaders. Part of the appeal of an aggressive test-and-isolate approach is that it leaves less of the burden of fighting the disease to individuals, unlike other tactics like reporting contacts, wearing masks, and maintaining social distance.

But it does require testing repeatedly, on the order of the entire US population every two weeks.

Craig F. Walker/The Boston Globe via Getty Images

“Health care professionals, police officers, EMTs — you may need to actually test them every single day to catch an infection early enough to make sure they are not spreading it to their colleagues,” Romer said. “What I would say right now is a good target would be 35 million per day. You can test everybody every two weeks, and then you’ve got 10 million frontline occupations that you can test every single day.”

Such a level of testing would effectively bring the R0 below 1, according to Romer. And that will ultimately cause the pandemic to fade out.

In a follow-up message, Romer noted that some infectious disease specialists have told him even 35 million tests per day may not be enough. He said that he is willing to scale up the number of tests in his proposal to whatever it would take because would still be cheaper than the cost of persistent or sporadic shutdowns of the economy.

Making and deploying the millions of tests needed would require radical investments and interventions

Testing millions of people a day until the pandemic is over is not as simple as buying more tests. It would require a national mobilization on the scale of a world war.

“There is no historical example of this without command and control,” Allen said.

To coordinate this effort, Allen and her colleagues have proposed creating a Pandemic Testing Board along the lines of the War Production Board created by the US government in World War II. The board would pool expertise from the private sector, government, and universities to iron out all the wrinkles in getting the vast number of tests needed. That could involve procuring supplies directly from manufacturers. The board could also use the Defense Production Act to compel companies to manufacture the requisite hardware, from personal protective equipment for health workers to the chemical reagents needed to run the tests.

The board would also come up with recommendations for training a workforce to run the testing apparatus, from collecting samples, to running the tests, to tracking the spread of the virus.

Win McNamee/Getty Images

The tests themselves would also need to improve. The most common RT-PCR genetic tests for the SARS-CoV-2 virus can test dozens of specimens at once, but can take hours to run at an in-house lab at a hospital or clinic. These tests can take days to get results if the facility has to ship the tests to an outside lab. The Food and Drug Administration recently approved a genetic test from Abbott Laboratories that can deliver a result in as little as 5 minutes. But the system can only run one sample at a time and Abbott is currently aiming at running 50,000 tests per day, far short of the millions researchers say are needed, so further scaling up is necessary.

An international research team recently presented a new genetic test for Covid-19 that they say could be “scaled to analyze millions of samples per day using existing sequencing infrastructure.”

Romer noted that the United States would also need more laboratories to run these tests; academic biology labs and corporate biotech labs could be marshaled for the effort. He proposed a billion-dollar prize to be offered to any lab that can process 10 million tests per day, an incentive that could spur an increase in processing capacity in a couple months.

Getting to 35 million tests per day would cost around $100 billion, according to Romer. It’s a steep price tag, but a fraction of the $350 billion or so in monthly economic losses due to the ongoing lockdowns and social distancing measures in the United States.

In fact, Romer envisions a federal agency with a $100 billion annual budget whose job it is to fight Covid-19 and prepare for future pandemics. The goal is to be ready to squelch an outbreak without shutting down the economy again.

“I think it would be enormously costly if we were to say that our plan is to relax and then when things get bad lockdown again,” Romer said. “We need to give people confidence that we’re going back to normal.”

There are social aspects that have to be accounted for as well. For people that test positive, the government should take steps to make it easy for them to self-isolate. In South Korea, officials sent comfort packages with food and cleaning supplies to people under quarantine, for example.

Testing the whole country is also a towering administrative challenge. The country that has tested the greatest share of its population and is on track to test everyone is Iceland, with 10 percent of its 360,000 residents tested so far. So testing a country of 330 million people on a regular basis may seem insurmountable.

But Allen pointed out that such a testing regime could be administered more easily at local levels. “This country has 3,000 counties. That makes our counties on average slightly smaller than Iceland,” she said. “I think it becomes conceptually not that hard to imagine having 2 to 8 percent of the population tested on a daily basis.” That will require channeling federal money to local governments in many cases.

Together, these efforts form a logistical orchestra that needs to happen a hundred million times multiple times per week until the pandemic tapers off, which could be more than a year from now. Yet such coordination would allow people to safely begin to return to work, relieving some of the heaviest pressure of the pandemic.

Test. Identify. Isolate. Repeat.

There’s almost no way to get around the need for vastly more testing for Covid-19 in the United States

Clearly, building a national pandemic response strategy around mass testing is not simple, easy, or cheap. But it may end up being simpler, easier, and cheaper than just about any other approach that’s been tried.

The recent Centers for Disease Control and Prevention paper modeled how different tactics would work to curb the spread of Covid-19. The researchers reported that “active surveillance, contact tracing, quarantine, and early strong social distancing efforts are needed to stop transmission of the virus.”

A combination of these measures have helped places like Taiwan, South Korea, and Singapore limit the spread of the virus. But many of these areas are now seeing a rise in new infections, mainly imported from abroad, as they relax these measures to try to return to normal. Some places have had to restore lockdown measures. So no one is out of danger from Covid-19 just yet.

Ed Jones/AFP via Getty Images

Ed Jones/AFP via Getty Images

Truly ending the pandemic then requires a complex strategy, since many of the tactics that have been deployed so far are inadequate on their own. And it would require far more effort to deploy them in a large, dispersed country like the US.

For instance, take contact tracing, the method of figuring out who an infected person encountered and could have infected. That can take the form of interviews with infected people or location tracking with mobile devices.

But a full regime of contact tracing in the US would still require mass testing as well as information collection to limit the Covid-19 outbreak to levels of places like Hong Kong and South Korea. Many of the places that deployed contact tracing did so at early stages of the pandemic, when the prevalence rate of the infection was still low, which limited the numbers of contacts that needed to be traced. That made it cheaper and faster to conduct contract tracing.

In the US, with a much higher prevalence of the virus, such tracing would still require testing for a much larger segment of the population than is currently available.

“Even if we were to use those methods, you would need vastly more tests,” said E. Glen Weyl, who coauthored the Safra Center’s paper on mass testing and works as a political economist at Microsoft. “And if you don’t want to use those methods, or you’re afraid that they won’t get enough take-up, then you would need truly enormously larger numbers of tests.”

Weyl estimated that a completely random testing system without contact tracing would require about 100 million tests a day. Contact tracing could help economize the use of tests, but testing would still be needed for millions of people per day. So scaling up testing capacity is a no-regrets tactic for combating the pandemic.

Meanwhile, the current social distancing regime in the United States is not only a huge economic and social drain, it also has holes and inconsistencies. That means it could take several months if not longer to bring the prevalence of the virus down within the population on its own.

Spencer Platt/Getty Images

“40 percent of the workforce — four-zero percent — is in essential sectors, so our social distancing is being applied to maybe half of the people in the country,” Weyl said. “That slows things down but there is a huge population where the disease is still almost certainly spreading because we just can’t avoid it.”

Testing for antibodies could be useful for research purposes to identify who was previously infected in order to trace the spread of the virus. People with antibodies could also donate their blood serum to help treat infected people or to develop new therapies. And if it is confirmed that surviving an infection confers robust, lasting immunity to the virus, a positive antibody test could help identify people who can safely end their social distancing. That would be particularly useful for frontline workers who are desperately needed back on their feet.

But immunity testing is not an easy ticket to restarting the economy. The number of people who have survived the virus so far are nowhere near enough to reopen all the shuttered shops and offices. A scenario where there are enough immune workers to restart the economy would imply spread of the disease through the population, and with it, the inevitable spike in hospitalizations and deaths. And that is exactly the scenario most public health officials are trying to avoid.

So any plan to resuscitate the economy and society will have to incorporate a major testing element. “It is just not realistic for us to greatly reduce the prevalence rates in anytime soon without the application of [mass] testing or some regime like that,” Weyl said.

The fight against Covid-19 will continue for at least a year until a vaccine is widely available, so regular testing for the virus will have to become part of the milieu for the United States. What was once normal is still a far-off prospect.

“I just don’t think never-SARS-CoV-2-infected people are going to quickly enjoy our previous way of life any time soon, no matter where you live,” said Duke University’s Gray.

The recovery will be slow, and the overall situation — infections, fatalities, job losses — will likely get worse before it gets better. But half-measures will only prolong the problem. Go big or stay home.

Support Vox’s explanatory journalism

Every day at Vox, we aim to answer your most important questions and provide you, and our audience around the world, with information that has the power to save lives. Our mission has never been more vital than it is in this moment: to empower you through understanding. Vox’s work is reaching more people than ever, but our distinctive brand of explanatory journalism takes resources — particularly during a pandemic and an economic downturn. Your financial contribution will not constitute a donation, but it will enable our staff to continue to offer free articles, videos, and podcasts at the quality and volume that this moment requires. Please consider making a contribution to Vox today.