Since the spring, many public health experts have been calling for rapid, at-home, over-the-counter Covid-19 tests for people without symptoms. The Food and Drug Administration (FDA) finally authorized the first one Tuesday, potentially ushering in a new approach to combating the pandemic, which is projected to continue taking thousands of US lives each day through the winter and early spring.
These home tests would work alongside the standard, lab-based PCR tests that have been the backbone of testing in the US but have often been too slow at catching infections before they spread. Rapid at-home tests, like the recently greenlit one from Ellume, could alert people they have the virus before they start feeling sick — and even if they never get symptoms at all.
The new Ellume Covid-19 Home Test uses a nasal swab mixed with a specialized fluid to generate results on a small device that connects via Bluetooth to a smartphone app. “By authorizing a test for over-the-counter use, the FDA allows it to be sold in places like drug stores, where a patient can buy it, swab their nose, run the test, and find out their results in as little as 20 minutes,” FDA Commissioner Stephen Hahn said in a statement.
The Ellume product joins two other rapid at-home tests, both of which require prior approval or a prescription and are for people with suspected Covid-19 cases: the BinaxNOW COVID-19 Ag Card Home Test made by Abbott, which got an EUA Wednesday, and the Lucira COVID-19 All-In-One Test Kit, approved in November.
This rush of new approvals is exciting to many who have been following these tests for months. “This is great news to put testing in the hands of people who need it,” says Gigi Kwik Gronvall, an immunologist and senior scholar at the Johns Hopkins Center for Health Security, who has been leading the center’s test tracking.
It “shows that we’re getting to another level of access to timely testing, and hopefully that trend is going to continue,” says Mark McClellan, director of the Center for Health Policy at Duke University who served as FDA commissioner under President George W. Bush. More than two dozen similar tests are reportedly in the works, using a variety of methods — looking for viral proteins or scraps of genetic code — to tell people if they have high levels of the virus in their system.
“But I think people shouldn’t get too far ahead of themselves,” McClellan says. “The supply of these tests is not in line with what you would need for routine testing for lots and lots of Americans.” Ellume, for example, is expecting to be able to roll out 100,000 tests per day in January, scaling up to 1 million tests per day by June. To test 330 million people in the US twice a week, as some experts are recommending, we’d need about 94 million tests a day.
And even at scale, these new at-home tests cannot solve our pandemic problems on their own. We’ll continue to need to wear masks, maintain social distancing, and get vaccines as soon as we can. “Relying on testing and the personal responsibility of people to take these tests on a regular basis is not going to be the only thing to get us out of this mess,” Kwik Gronvall says.
Here’s why, and what we — and the Biden administration — need to do next.
At-home tests face major hurdles, but Ellume cleared many of them
Most Covid-19 tests conducted in the US so far have been PCR tests. These require a laboratory and specialized equipment, along with trained personnel, to multiply viral material from a sample up to a detectable level. This makes them quite (if not perfectly) reliable in diagnosing infections.
But it has also made them prone to major delays, as laboratories got backlogged during spikes. This meant results have sometimes taken a week or more to return, making these tests impractical for helping people not spread the virus.
“Essentially all of the testing that is happening in this country right now is without purpose” when there are multiple-day delays in returning results, Michael Mina, an epidemiologist at the Harvard T.H. Chan School of Public Health and advocate for rapid testing, said in a call with reporters earlier this month.
The whole idea behind at-home — or “at-anywhere” — tests is that they don’t require a lab and a professional to run them. They provide quick results so that people can isolate themselves immediately, if necessary.
But there have been major concerns about sidelining advanced equipment and trained staff while determining whether someone is infected. “A first law of testing is that everyone wants a lab test to be cheap, fast, and accurate,” says Geoffery Baird, chair of laboratory medicine and pathology at University of Washington Medicine. But, he says, of those three requirements, “You only get two. … What normally happens when people are making that trade-off is that they give up accuracy.”
One issue he and others worry about is the opportunity for user error. “The performance of all these tests is always worse out in the field than when it’s data submitted to the FDA,” Baird says. He also oversees the toxicology lab at the county hospital, where they field questions about over-the-counter home tests for drugs, and he sees a lot of issues from people trying to use those tests by themselves. People can misunderstand the instructions or collect the sample incorrectly, for example.
Two other challenges with testing at home, especially with a test people buy over the counter, are getting people connected to a medical professional for appropriate care or follow-up testing, and getting test results reported to public health authorities.
First, nuances in an individual’s situation might warrant different responses to a test result than a simple positive or negative might tell them, Baird notes. For example, someone with Covid-19 symptoms who receives a negative antigen test result might still be recommended for follow-up PCR testing to help verify they really don’t have the virus. On the other side, someone who tests positive but doesn’t have any symptoms or known exposures might also need follow-up, especially if they live in an area where there isn’t much virus circulating. These are all considerations a health care provider could walk someone through perhaps better than an app can.
Second, many experts have been advocating for low-tech, simple paper strip-based tests — akin to a pregnancy test — but that might mean that results are off the grid, keeping public health officials in the dark about where and how much the virus was spreading. “This is a concern — how are you going to get accurate counts of positive cases at the same time as giving people access to testing,” Kwik Gronvall says.
Ellume went higher-tech with its new test, ensuring results — transmitted via Bluetooth to the linked smartphone app — would be automatically transferred (with the person’s zip code and date of birth; name and email are optional) to public health officials. But the technology also nudged the price tag up from the $1 to $5 that Mina has proposed to $30. This would make frequent testing unlikely for someone earning, for example, the federal minimum wage ($7.25 per hour) and paying for it out of pocket.
Abbott’s newly greenlit at-home test took a different route. It kept the test itself low-tech, but it requires interaction with a professional “certified guide” through an online medical service, who can advise people on best next steps and also report test results to public health channels. This means the test (which itself costs $5) actually retails for $25 to cover the cost of the professional’s time. Abbott’s test also requires people to meet the criteria of being in the first seven days of having symptoms (which also means not everyone can take it), and it takes extra time for the kit to be shipped.
People have also raised concerns over how these tests work. The method that many companies are pursuing, and the one the Ellume and Abbott tests use, is to seek out specific proteins on the virus cells called antigens (not to be confused with antibodies, which are proteins the body makes to find an infection and can indicate if someone was exposed to the virus in the past). But without expensive equipment, these tests have generally needed high virus levels to produce a positive result that could be read at home. Compared to PCR tests, many antigen tests have missed some of the positive cases.
The Ellume test, in its small study of about 200 people, was able to jump up to close to PCR levels of sensitivity — detecting 96 percent of the people with symptoms that PCR also found — and of specificity, returning a negative result for all the people who also got a negative PCR test. Perhaps even more crucially, for those without symptoms, it picked up 91 percent of infections, and delivered correct negative results to 96 percent of people who didn’t have the virus. To get FDA authorization, the Abbott and Lucira tests had to achieve similar levels of accuracy.
But many experts see the accuracy difference with these antigen kits less as a shortcoming and more as a new way to approach testing.
The power of less sensitive tests
From a medical diagnostic perspective, a traditional PCR test can tell if someone has even a very small amount of the SARS-CoV-2 virus. This is incredibly helpful to doctors who are trying to figure out the best way to treat an ill patient, for example, as people who are very sick with Covid-19 actually often have smaller amounts of the virus left in their system. But PCR’s slower turnaround means that it doesn’t always help us find people who are most at risk of spreading the virus to others.
In fact, it rarely does. According to Mina, less than 5 percent of tests currently being done are returning results while people are still infectious. “Lab PCR testing probably should never have been a public health tool; it’s a medical tool,” Mina said. “That needs to be reserved for medical use, for people who are sick, for people who are in the hospital.”
There are two intertwined reasons for this. First, unlike previous severe coronaviruses, such as SARS and MERS, Covid-19 spreads most readily just before and just as someone develops symptoms. As soon as someone catches the virus, it begins multiplying in their system, usually making them infectious around day two to four. However, people often don’t develop symptoms until day four or five after exposure. That means many people spread the virus in the 24 to 48 hours before they feel sick. (And this is to say nothing of the 40 percent or so of people who have the virus — at similarly infectious levels — and never develop symptoms.)
Offering these rapid, at-home tests to people without symptoms, as the Ellume authorization has, could play a powerful role in monitoring and slowing the virus’s spread.
The second reason these quick tests could help substantially is that time is of the essence, and PCR testing has been very slow. Let’s say someone waits until they have had symptoms for a couple of days, gets a PCR test, and then waits four days for the results. It’s likely that by the time a positive result is returned, that person isn’t even infectious anymore. “Even a 24- to 48-hour delay in testing, which right now is rapid for PCR tests, it’s just not going to stop spread,” Mina said.
That’s why he and others have been advocating for the quick at-home antigen tests. Although less sensitive at-home tests might not pick up the very small levels of the virus that PCR tests can, they should be able to detect higher levels, which is exactly when people are most likely to spread the virus, whether or not they have symptoms.
“Antigen tests have been much maligned and much misunderstood — and have a really, really important role to play in this pandemic,” says Ashish Jha, health policy expert and dean of the Brown School of Public Health.
Kwik Gronvall agrees: “They have some really great advantages as far as giving you actionable information in a timely manner.”
Many experts point out that with a 90 percent rate at catching infections, that’s 10 people out of 100 that might be spreading the virus unknowingly (assuming they are still infectious). When you scale up to millions of tests, if 1 million people tested are positive, 100,000 of those infections could be missed. Which sounds pretty bad.
But this is why these tests are not a replacement for other scientifically backed public health practices. Rather, they are another layer in the “Swiss cheese” model of using many imperfect methods concurrently to stop viral spread. Many questioners of rapid tests have pointed to the outbreaks at the White House that occurred while people were being tested frequently.
So this example doesn’t actually present a case against these tests, but illustrates how not to use them. Instead of testing everyone and allowing everyone with a negative result to waltz into large events — or into their grandfather’s living room — without a mask or social distancing, these tests should alert people with a positive result to self-isolate and get in touch with their health care provider. Those with a negative result could continue with carefully masked and distanced activities. “Testing doesn’t make things safe by itself; it adds protection to things you’re already doing,” Baird says.
Besides, most of those hypothetical 100 people in the testing scenario probably wouldn’t have otherwise gotten tested then. “The alternative is what we have now, which is out of 100 people with the disease almost no one is getting a test,” Jha says. “So would you like to pick up 90 out of 100 or would you like to pick up zero out of 100?”
“I’d love a perfect test, but these are compromises that don’t, in my mind, substantially affect the value of these tests,” Jha notes.
How we could use rapid at-home testing to slow the pandemic
For all of their strengths and all of the enthusiasm surrounding them, rapid at-home Covid-19 tests are not meant to become the new solution for those of us waiting for our vaccine. Nor are they meant to be the only form of testing we do.
Even the most vocal advocates for widespread at-home testing are simply pushing for these tests to become an additional testing tool, alongside PCR and other techniques. “You want to have the ability to do PCR testing for verification,” Jha says. “No one is saying throw out PRC testing, and none are saying do antigen testing and don’t worry about anything else.”
Above all, these tests need to be part of a comprehensive testing strategy for the incoming Biden administration, Jha says.
This will be especially important in the early days when these tests have limited supply. “We should be deploying them in nursing homes, schools, health care facilities, high-risk workplaces like meatpacking plants, and for other essential workers like police officers, firefighters, and EMTs,” Jha says.
For now, among the newly authorized at-home rapid tests, “we’re talking about maybe 20 million to 30 million tests per month, which is not anywhere near enough to do large-scale testing,” McClellan says. So although “test capacity is increasing a lot, from the standpoint of maximizing public health impact for the next three to six months … we’ve got to prioritize [rapid testing] to settings where it’s going to have the most impact.”
The Trump administration has done this to some extent with previously approved rapid antigen tests (for use by trained professionals), now purchasing some 50 million Abbott BinaxNOW professionally administered tests each month and sending many directly to long-term care facilities and other priority locations. This ensures “that the tests go not just to people who can afford to buy it but the places where we need it the very most,” McClellan notes.
But if the Ellume and other at-home tests are simply released to pharmacies and other marketplaces for anyone to buy, Jha thinks the limited number of tests are most likely to be bought up by companies to test their employees and by people who are looking to travel or see others in person. So he says he doubts that, for now, these tests will actually do much to change daily life.
As more people in high-priority groups receive the vaccine, however, more of the testing might be allocated away from these folks and to the general public, McClellan notes. And, in the meantime, we should all be doing our part to reduce demand for the limited supply of tests by staying at home and not gathering, Baird says.
But to see a big impact, we are going to need not only strategic allocation but also, eventually, a lot more tests. “In an ideal world, you’d be testing everybody in America twice a week,” Jha says. He also recognizes that this level of testing isn’t likely to be feasible (or even necessary if many people continue to stay home and take adequate precautions). So he proposes that “if we want to save lives and bring our economy back, we need 10 million to 20 million tests a day.” (So far we have yet to cross 2 million tests given per day in the US.)
This is because, regardless of the method of testing, Kwik Gronvall notes, “testing is a moment in time, and it can’t be the only thing that you do to prevent transmission.”
How we can get rapid at-home tests for everyone, often
Getting more rapid tests authorized, made, and distributed is contingent on the Trump and Biden administrations making it a priority. From his conversations with members of the Biden advisory and transition teams, Jha says the incoming administration “is really focused on ramping up testing across a variety of modalities, and that includes rapid testing.” That is, he says, because they understand that even after more people can get a Covid-19 vaccine, the virus will still be circulating in the US for a while — and abroad even longer. We also don’t yet know if people who have gotten the vaccines can still carry and spread the virus easily, even if they themselves don’t get sick.
Jha suggests four things the Biden administration should do right away to improve the testing situation:
- Use the funds for testing from the CARES Act
- Ask Congress for more money to manufacture and distribute these tests
- Review the FDA process for approving more rapid tests
- Help states come up with a testing strategy for most effectively deploying tests (much as the CDC has done with vaccines)
Getting more rapid at-home tests to more people on a regular basis also depends on Congress to fund these efforts. Mina has been lobbying members of Congress to dedicate $1 billion of the potential $900 billion pandemic stimulus package to bringing more rapid tests to market (which could come from the $16 billion in the bill earmarked for vaccines, testing, and tracing).
“It sounds like a lot of money,” he said. But “it’s less than 0.05 percent of what this virus is likely to cost Americans over the next few months.” He recommends a $200 million boost each to five companies that already have these tests developed to scale up manufacturing. From there, he suggests $5 billion to $10 billion earmarked for manufacturing and distribution in 2021.
With this backing, Mina says, “we could find ourselves in an entirely different position” than the current tragic death tallies. Even with the vaccine rolling out, one major model puts the expected new toll at more than 101,000 additional US deaths by February 1 (if universal masking isn’t adopted; 78,000 if it is). This means testing will remain critical.
McClellan agrees. If the Biden administration rapidly invests even more money in scaling up manufacturing of these tests, the US could have even more available for early 2021. “So I think it’s not unreasonable to expect that schools who want it and other priority settings that really want it should be able to get it,” he says.
But that also depends on the FDA being able to greenlight more of these tests. McClellan notes that rapid, at-home testing “is a high priority for the FDA.”
Mina and others have called for a new pathway to get these tests to market that doesn’t hold them to the same diagnostic criteria as PCR tests. And McClellan says the FDA is discussing options for authorizing Covid-19 tests less sensitive, for example, than Ellume’s. One way to do this would be to instruct that such tests are to be used frequently — multiple times a week — to increase the odds that they will catch an infection.
So long as the tests are thoughtfully authorized and deployed, if we can add relatively widespread, rapid testing to a more comprehensive plan to fight the pandemic, Jha says, “I see a pretty bright spring. Not normal, not spring of 2019, but a lot better than the spring of 2020.” Americans won’t be going to packed concerts, and they will continue to need to wear masks and continue distancing, but many essential functions will be more solid. “There will be little to no justification for not having schools open,” he says.