Support the news
Dozens of blood tests are rapidly coming on the market to identify people who have been exposed to the coronavirus by checking for antibodies against it.
The Food and Drug Administration doesn't set standards for these kinds of tests, but even those that meet the government's informal standard may produce many false answers and provide false assurances. The imperfect results could be a big disappointment to people who are looking toward these tests to help them return to something resembling a normal life.
First of all, it's not clear whether someone who has antibodies to the coronavirus in their blood is actually immune. Your body produces these antibodies within about a week of infection.
In many other diseases, people do have a period of immunity after they have been exposed to a microbe and recover from illness. But that has not been demonstrated yet with the coronavirus.
Another problem is that test results are wrong much more frequently than you might expect. While tests may truthfully say they are more than 90% accurate, in practical use they can often perform far below that level.
Deborah VanderGaast of Tipton, Iowa, would love to know the results of an antibody test.
"What if we're already immune and we just don't know it and we don't have to be scared?" she asks. VanderGaast runs a day care center for children with developmental and behavioral disabilities. They're a lot like little kids everywhere.
"We laugh about the sanitizing everything, because, it's just reinfected just two seconds later," she says with a laugh.
VanderGaast is eagerly awaiting the rollout of a blood test that would be able to tell if she and her staff have antibodies to the coronavirus. They are now hypervigilant, and stay home for 14 days if they were even potentially exposed to the coronavirus. They wouldn't have to do that if they knew they were immune, she says.
VanderGaast says the test isn't available in her county now. But it is starting to take off nationally.
Dr. Jeremy Gabrysch runs a mobile medical service in Austin, Texas. He got a supply of antibody tests made by a major Chinese manufacturer and says he has tested several hundred people in the last few days.
"We offer the test for people who may have suspected they might have had coronavirus back in February or March when testing with the nasal swab [and PCR diagnostic test] was very limited," he says. The charge: $49 a test.
Gabrysch says he only tests people when he has other evidence they might have been exposed. "If they had an illness that sounds like it could have been coronavirus and they had a positive antibody test, then it's very likely that this is a true positive, that they indeed had COVID-19," he says.
The test he's using, produced by Guangzhou Wondfo Biotech in China, boasts a specificity of 99%, which means it only falsely says a blood sample contains antibodies against the coronavirus 1% of the time. But despite that impressive statistic, a test like that is not 99% correct, and in fact in some circumstances could be much worse.
That's because of this counterintuitive fact: The validity of a test depends not only on the technology, but how common the disease is in the population you're sampling.
"It is kind of a strange thing," admits Dr. H. Gilbert Welch, a scientist at Brigham and Women's Hospital in Boston who studies issues surrounding tests and screening. "An antibody test is much more likely to be wrong in a population with very little COVID exposure."
This is a result of statistics, rather than the technology of any given test.
Here's a simple way to look at it. Say you are running a test that gives five falsely positive results in a group of 100 people. That doesn't sound too bad. But consider this. If 5% of those 100 people were actually infected with the coronavirus, you should get five correct test results (true positives), along with the five false positive results.
While the manufacturer can rightly claim the test is 95% specific, in this population "the test will be wrong half the time," Welch says. "Half the people will be falsely reassured."
And test results can be considerably worse.
The Food and Drug Administration does not regulate these tests, but White House coronavirus task force coordinator Dr. Deborah Birx has said that she expects manufacturers to achieve a standard of 90% specificity (and 90% sensitivity, another measure of test performance that's less important in this context).
Here's what would happen if you used a test with 90% specificity in a population in which only 1% of the people have coronavirus. Nobody knows for sure, but that could be the situation in many parts of the country.
In that instance, more than 90% of the positive results would be false positives, and falsely reassuring. (You can run your own examples on the calculator on this page).
Tests being put to use now vary widely in their specificity.
Lab giant Becton Dickinson says its coronavirus antibody test has a specificity of 91%. Emory University says its test has a specificity of 95%. Stanford University says its test demonstrated 100% specificity, at least in a preliminary study involving 100 samples. The Centers for Disease Control and Prevention has been developing an antibody test since the early days of the epidemic, but has not yet published results about its performance.
One way to limit the problem of false-positive results is to focus on populations where the disease is more common — in other words, in situations where true positives are much more frequent than false positives. That's likely to be the case in hospitals that have cared for coronavirus patients.
"It would be wonderful for health care workers to know their immune status, just to give them peace of mind," says Dr. Jordan Laser, a pathologist at Northwell Health on Long Island, New York.
Even so, Laser says it would still be a mistake to rely on these results.
Jeremy Gabrysch, the doctor in Austin, is also using these tests to satisfy the curiosity of his patients, not to provide actionable guidance about whether it's safe for people to assume they are immune.
"We don't recommend that people interpret the result that way, and we counsel against that," he says. "It's not a get-out-of-jail-free card."
Because of these significant limitations, antibody tests fall short of being the imagined passport that would allow people to get back to business as usual and ignore the coronavirus.
But high-quality antibody tests will still be valuable in instances where individual false-positive results matter less. That would be the case for surveys of entire populations, where errors can be factored out.
In these studies, antibody tests are being used to answer critical questions about where the coronavirus is and how prevalent it is. That information can help officials plan how to let normal lives resume — but across populations as a whole, rather than one person at a time.
You can reach NPR Science Correspondent Richard Harris at firstname.lastname@example.org.
A conversation with California biologist Andrew Cohen inspired this line of inquiry.
Support the news