Skip to main content

Coronavirus antibody tests aren’t created equal. Here’s what you need to know

The Food and Drug Administration (FDA) recently decided to tighten the regulations on coronavirus antibody tests. These tests are supposed to help people know whether they were previously infected with the virus, but the FDA admitted that many of them aren’t accurate enough to be useful.

Whether you’ve already been administered an antibody test or hope to get one in the future, here’s what you need to know about whether to trust the results.

What are antibodies?

When your body is exposed to an antigen — or outside substance including viruses, bacteria, and certain chemicals — it causes the immune system to attack it. As part of the response, white blood cells create a certain kind of protein called antibodies. These proteins can only bind to that specific type of antigen, and they can help the body more quickly respond to the virus or bacteria, making you less likely to get sick from the exact same strain twice.

Coronavirus
BSIT / Getty

Antibody tests look for these proteins in the blood, after a person has recovered from the illness. Not all antibodies last the same amount of time; some remain in the blood for a lifetime, while others, like those that fight the flu, may only last a matter of months. The individual response can vary as well. “There are people, with any infection, that for whatever reason don’t produce high levels of antibodies,” Dr. Amesh Adalja, a Senior Scholar at the Johns Hopkins University Center for Health Security told Digital Trends.

What is a serology or antibody test? 

Antibody tests differ from polymerase chain reaction (PCR) tests. PCR tests are diagnostic; they’re the ones given to people with symptoms of the virus and use a swab to collect a sample from the nose or throat. There are a couple of types of antibody tests. One uses a drop of blood from a finger prick, while more accurate but less widely available tests are called an ELISA (Enzyme-linked immunoassay). These require larger equipment better suited for a laboratory than a doctor’s office.

For antibody tests, a blood sample is used to determine if a person was previously infected. It can take two weeks for antibodies to develop, but they can stay in the blood for months or years, depending on the disease. Because COVID-19 is so new, it’s not yet clear how long the antibodies remain or whether they’ll protect a person from becoming reinfected. Nevertheless, accurate antibody tests are an important component in plans to ease social distancing restrictions.

What are false positives? 

In order to determine the accuracy of an antibody test, researchers look at sensitivity and specificity. Sensitivity is the percentage of people with the virus who test positive. Specificity is the percentage of people who aren’t infected who test negative. For some diseases, you might have a test with high sensitivity, followed by another test with high specificity. For example, mammograms are widely used but are followed up by a biopsy for a more conclusive diagnosis.

drive-in coronavirus test
NurPhoto / Getty

“It’s less a question of sensitivity with the antibody tests that we’re worried about,” said Dr. Adalja. “It’s that they actually are getting too many false positives, not false negatives.” A false positive is a test that shows a person has antibodies for COVID-19, even though they were never infected. SARS-CoV-2 is a coronavirus, but there a lots of coronaviruses, including many common colds. If an antibody test isn’t specific enough, it might indicate that you have antibodies for a different coronavirus but not for SARS-CoV-2. That would mean you’re still vulnerable to getting infected, but you would be under the impression that you’d already had it. Again, while experts think antibodies will likely provide some immunity, they’re still not sure how much or for how long.

When the FDA first started allowing institutions to perform antibody tests, there weren’t a lot of guidelines around accuracy. Manufacturers were able to validate their own results, leading to a flood of sub-par antibody tests. “Flexibility never meant we would allow fraud,” the FDA said in a statement. “We unfortunately see unscrupulous actors marketing fraudulent test kits and using the pandemic as an opportunity to take advantage of Americans’ anxiety.”

On May 4, the FDA announced makers of antibody tests had to submit requests for emergency use authorization (EUA) within 10 days of validating their tests, and current tests on the market also have to apply. An EUA isn’t as rigorous as full FDA approval, but it does ensure that the organization has reviewed the test and its validation claims. The FDA has authorized 12 antibody tests for emergency use, including a new one from Roche that the company said has 100% sensitivity (its accuracy in detecting coronavirus antibodies) and 99.8% specificity (the accuracy in detecting when no antibodies are present). The FDA is also working on authorizing another 250 serology tests, according to The Washington Post. You can see the list of antibody tests with EAUs on the FDA’s website, including their sensitivity and specificity measures, according to the manufacturers. “There are a handful out there that do have good sensitivity and specificity,” said Dr. Adalja.

Testing the tests

In mid-April, the FDA announced it was collaborating with the National Institutes of Health’s National Cancer Institute (NCI) and National Institute of Allergy and Infectious Diseases (NIAID) and the Centers for Disease Control and Prevention (CDC) to evaluate antibody tests. Outside groups have started examining their accuracy as well.

Getty Images

Recently, several research institutions started verifying which antibody tests show reliable results. Of the 14 tests the researchers examined, only three were consistent. The COVID-19 Testing Project is a collaboration between University of California San Francisco and UC Berkeley scientists, supported by the Chan Zuckerberg Biohub and other donors. The results have not been peer-reviewed, but other experts have been warning about the unreliability of many of these antibody tests for weeks.

How can you find out how accurate your antibody test is?

If you’ve gotten an antibody test but aren’t sure whether or not to trust the results, Dr. Adalja suggests asking your doctor about the brand name and specificity. “I think, for the general public, talk to your doctor who administered the test and then be able to decide whether or not to test that you were administered had sufficient specificity to say for sure whether or not you truly were infected with the novel coronavirus,” he said.

One reason more PCR or diagnostic testing is so important is that there appear to be few false positives with them. If you’re diagnosed with COVID-19 via a swab test, it’s very likely you have it. “I do think we really want to get a handle on the acute infections and make sure that we’re not missing them, so that we can diagnose those cases, isolate them, and trace their contacts,” said Dr. Adalja. “That’s still going to be the core part of this.”

Jenny McGrath
Former Digital Trends Contributor
Jenny McGrath is a senior writer at Digital Trends covering the intersection of tech and the arts and the environment. Before…
FDA approves Roche coronavirus antibody test for emergency use
Coronavirus Testing Labs

Swiss pharmaceutical company Roche says it has received an emergency use authorization from the Food and Drug Administration (FDA) for its new test for coronavirus. The company joins several others such as Abbott Laboratories, Becton Dickinson, and DiaSorin in offering antibody tests for the virus.

“Our best scientists have worked 24/7 over the last few weeks and months to develop a highly reliable antibody test to help fight this pandemic,” Thomas Schinecker, CEO Roche Diagnostics, said in a statement.

Read more
DIY ventilator plans are all over the internet, but is building one a good idea?
Dyson

It’s impossible to say no one saw a pandemic coming. In December 2007, a user of the how-to site Instructables uploaded plans for “The Pandemic Ventilator,” a do-it-yourself machine made out of plywood, plastic bags, threaded pipe, and a programmable logic controller (PLC).
“If a pandemic were to strike, the hospitals could not just go out and buy all the ventilators they need, because there would not be enough parts or manufacturing capability,” warned a user named Panvent, who at the time was expecting an outbreak akin to the avian flu.
Fast forward 13 years and the Pandemic Ventilator seems prescient. The U.S. national stockpile doesn’t have enough vents to support all of the states that need them; many of the ones in it are broken, and as they hunt for a cure, governors are rapidly discussing various ideas for alternative ventilators, including hacking anesthesia machines and hand pumping bag valve masks. One hospital in Italy even transformed scuba masks into personal oxygen machines for several patients.
In America, states like Arkansas and Florida are still trying to figure out of they have enough machines in preparation for a surge and President Donald Trump has blamed numerous governors for not being prepared. Having an extra ventilator on hand seems like a good idea. Should you build or hack your own?

A basic machine with a lot of bells
“At its essence, a ventilator is a pretty basic machine,” explains Randy Cremer, Director of Biomedical Equipment Engineering at New Jersey’s Deborah Heart and Lung Center. “It's just something that will create pressure that's a little higher than the lung pressure, so it will force you to take a breath.”
AmboVent 1690.108
The creator of the Pandemic Ventilator Project (PVP) seems to agree. Now an engineer at a startup developing a wearable dialysis system, this anonymous biomedical technologist has continued to develop schematics for his pet project, posting them on a Blogspot site, along with a discussion of the experiments and tangents the project has taken over the past several years.

Read more
Johns Hopkins opens coronavirus testing data to the public
Coronavirus Testing Labs

Johns Hopkins University has made its coronavirus testing data public — including where tests are available, how many are done, and state-by-state comparisons.

The data set, officially called the Johns Hopkins COVID-19 Testing Insights Initiative, also includes answers to a number of questions about the virus, including details on the testing procedures, tracking of symptoms, who should get tested, the differences between tests, and how long it could take to get results.

Read more