A slew of tests coming onto the market are designed to help determine whether people have previously been infected with COVID-19. But while many may be eager to know if they have already had the disease and since recovered, there’s a lot of uncertainty about the accuracy of the tests.
These tests are different from the ones used to diagnose a person with a coronavirus infection. Instead, these antibody tests, also called serological tests, are supposed to detect signs from your immune system that indicate your body has already fought off the virus.
In the case of many infections, your immune system develops antibodies to resist infections, which can stick around and make you immune to catching that sickness again. If that proves to be true for the coronavirus, knowing that you’ve already been exposed to and recovered from the illness could mean it’s safe to resume normal activities without risking another infection or passing along the virus.
That’s why officials including National Institute of Allergy and Infectious Diseases Director Anthony Fauci have floated the idea of so-called immunity passports or cards indicating a person has had a positive antibody test. But the World Health Organization has decreed that this would be premature, because like so many things with this virus, more is unknown than known at this point.
For one, scientists have not yet determined whether prior exposure to the coronavirus makes people immune. And the antibody tests are rushing to the market without full approval from the Food and Drug Administration.
The FDA initially took a lenient approach to test manufacturers due to the public health emergency, which led to more than one hundred options coming to market, many of dubious quality. The agency has since tightened its approval process in an effort to ensure more reliable, accurate tests.
The major risks of an inaccurate test are twofold.
First, incorrect test results indicating the presence of coronavirus antibodies in a person who wasn’t actually infected ― called a false positive ― could make a patient wrongly believe they were safe. Second, incorrect test results showing no antibodies in a patient who actually was infected ― called a false negative ― could cause a patient to continue quarantining without a reason. Even a test that’s correct 98% of the time would generate too many false results to be considered truly reliable from both a patient health and a public health perspective.
Part of the problem is that tests designed to be very sensitive to any coronavirus antibodies are more likely to detect them, meaning there would be more false positives. By contrast, tests that are designed to look for very specific antibodies can miss others that are present, leading to false negatives, explained Henry Chambers, an infectious diseases expert at the University of California San Francisco’s School of Medicine.
HuffPost spoke to Chambers about what people should know about coronavirus antibody tests and what they should consider if they want to get one.
What does a coronavirus antibody test measure?
It’s a test on [blood] serum for reactivity of one of these antibody proteins that your human immune response makes in response to, say, a prior infection. So, you know, there’s any number of antibody tests to see if you’ve developed an immune response to some infectious agent. For some diseases, though, the presence of antibodies does not protect against ongoing infection or reinfection.
Who should and who shouldn’t get an antibody test?
I don’t think we know the answer of who should and who shouldn’t get an antibody test in terms of individuals or big populations. But how it’s being deployed is to get a sense of another way of determining how far the virus has spread in the population by seeing the number of people that are tested who react to the virus who have a specific antibody.
So how might it be useful for you as an individual? Let’s say that you were in New York City and maybe you had an illness that was consistent with COVID-19 but your doctor told you to stay at home, monitor yourself, and if you got worse to come in and see him, because he didn’t want you to show up and overburden the system. Then, some time later, if you want to see if that illness that you had was COVID, the test would be useful.
If you had COVID and you wanted to donate plasma for use treating another individual, then you could get an antibody test and that would determine the amount of antibody that you had.
If you’re in a very low-prevalence setting, the test is probably going to be negative because if only one in 1,000 people is infected ― 99.9% are going to be negative. So it doesn’t really provide a lot of assurance unless there is a significant amount of prevalence in the community, in a place like New York, for example. But in the low-prevalence setting, a positive test is more likely to be a false positive. If you want some peace of mind and you’re looking for a negative test, it’s probably going to be negative anyway.
Getting one on your own, unless you’re pretty sophisticated, it’s going to be hard to know what [the result] means, and that’s why I would have a health professional order the test and interpret it for you.
Does a positive antibody test mean you definitely had the coronavirus?
Let’s say you were on a cruise ship, and you didn’t get tested but 70% of the cruise ship population was infected and you had an illness. In that setting, it could be useful because you already know that you were in an exposure situation; you know you had symptoms; and a positive test would have a good predictive value because it was likely the test would be positive. As the likelihood of exposure goes down, then you have to worry about false positives.
Does a negative antibody test mean you definitely have not had the coronavirus?
No. It makes it unlikely, but there are false negatives. Let’s say that you had a very mild infection, for example. We don’t know how robust the antibody response is. Or let’s say that you were asymptomatically infected. That’s going to require further research in populations to tell if the test performs differently in people who had an asymptomatic infection. Because it might be that if you didn’t have a very robust immune response to the infection, your antibody didn’t develop. We just don’t know at this time for sure.
Does a positive antibody test mean you’re immune to contracting the coronavirus again?
That is not yet known, for two reasons. One is that the antibody test is qualitative, not quantitative. It tells you that antibodies have been produced, but not the amount present and not whether the antibodies interact with the virus in such a way that it prevents it from establishing an infection, which is known as “neutralizing.” So all of that also has to be worked out.
I think most experts think that people who have been infected will be immune, and that is the whole basis for this “herd immunity” idea: that individuals are not subject to reinfection once they have cleared the infection. It is a guesstimate about this virus, because one thing we know about this is we don’t know a lot.
What value do these test results have to researchers working on disease surveillance and drug and vaccine development?
The virus test tells you confirmed cases and then that allows you to characterize the death number among known cases, but not the death rate overall. You can get a case death rate ― you know, I had 100 cases and five people died. But if that hundred cases is really 1,000, that’s 900 cases you don’t even know about. Antibody testing allows you to put a better number on the mortality rate to make projections about demands on the health care system, and tells you how well the measures that you’re using to mitigate transmission have performed.
Antibody tests will also be used to monitor the efficacy of a vaccine, at least in early phases, to make sure that you get an antibody response to the vaccine. It will allow you to deploy your vaccine in a strategic manner. Let’s say 20% of the population’s infected; you will be able to decide whether to vaccinate everybody or focus on those who are serologically negative.
Are the tests available today reliable?
Even if a test sounds pretty good ― 98% of the time, it was right ― two out of 100 will give you a false positive.
There’s always a trade-off between sensitivity and specificity. When you start dialing up one, the other gets worse. If you want to not miss somebody who’s been infected, you’re going to have a very sensitive test, but you’re going to misclassify a large number of people who weren’t infected.
The way around that is confirmatory testing. So you do test A and it says you’re positive, and then you say, OK, I’m going to confirm that with an independent test B. If that test is also positive, then it increases the likelihood that you really have been infected. If it’s negative, you’ve called into question the first test and increased the likelihood that you have not been infected. There are testing strategies that can allow you to determine the accuracy of a test, and the only way we’re going to get these answers is to start testing.
This transcript has been edited for length and clarity.
Experts are still learning about the novel coronavirus. The information in this story is what was known or available as of press time, but it’s possible guidance around COVID-19 could change as scientists discover more about the virus. Please check the Centers for Disease Control and Prevention for the most updated recommendations.
A HuffPost Guide To Coronavirus