Memphis shows 5% positive COVID antibody rate; critics question test accuracy

By , Daily Memphian Updated: May 06, 2020 12:15 PM CT | Published: May 05, 2020 4:00 AM CT

New York City made headlines just over a week ago when 21% of people randomly tested had positive results for COVID-19 antibodies, presumably showing signs of immunity.

In Memphis, 5% of those tested are positive for antibodies, a sign that the disease is in the early stages here and that most people are still susceptible, says Dr. Stephen Threlkeld, the infectious disease expert treating COVID-19 patients at Baptist Memorial Healthcare-Memphis.

“It means you have to watch very carefully as you reopen things, and even before you reopen, to make sure those numbers don’t translate into more cases,” he said.

Monday, Baptist-Memphis had 27 COVID-positive patients, including five in ICU, down from a total of 29 on Friday.

As more states move to reopen their economies, antibody tests are a barometer of how safe reopening may be. But the tests have been riddled with error and in some cases, fraud. Monday, the U.S. Food and Drug Administration announced it was tightening controls set March 16 that allowed more than 100 antibody tests to get to market without review.

The agency said it had to provide regulatory flexibility for the test developers, “given the nature of this public health emergency.”

Editor’s Note: The Daily Memphian is making our coronavirus coverage accessible to all readers — no subscription needed. Our journalists continue to work around the clock to provide you with the extensive coverage you need; if you can subscribe, please do

“However, flexibility never meant we would allow fraud. We unfortunately see unscrupulous actors marketing fraudulent test kits and using the pandemic as an opportunity to take advantage of Americans’ anxiety,” the FDA said.

Antibody tests, in theory, show the community’s state of resistance. The nasal swab test for COVID-19, on the other hand, shows a snapshot of those who are sick on a given day — and where — and kicks off the processes of quarantine and contact tracing.

Both tests are needed because they show different stages of the epidemic.

“Five percent is within the norms of what we would expect,” said Dr. Manoj Jain, the infectious disease adviser to the Memphis/Shelby County COVID-19 Task Force. “When you look at how extensive the epidemic was in New York, it makes sense they have 20% of their people infected with COVID and we have much less. We can see obviously from the number of cases and hospitalizations, that we are not getting inundated at the hospitals.”

The problem here is that it is unknown which 5% received the test.

According to the report the COVID task force received Monday morning, American Esoteric Laboratories here analyzed 4,462 antibody tests and showed a 5% positivity rate.

“The positivity rate depends on several factors,” Jain said.

“The most significant is which population is coming and getting tested. If it is your very high-risk population, such as health care workers and other front-line workers, as you go to the general population, that positivity rate may go down,” Jain said.

“The fact that it was not 20% is a bit reassuring — even if you take the worst-case scenario, people whose doctors felt they were at high risk and had them tested. In New York, it was random.”

New York tested 3,000 people in grocery stores, including 1,300 in New York City. State results showed nearly 14% had antibodies to coronavirus. In the city, where the disease is more prevalent, nearly 21% showed antibodies.

Doctor recovers, gives first serum donation at Methodist University

As more people here survive the disease, the positivity number will go up, but slowly, Jain said, as more people show immunity.

Antibody tests can also show how effectively the disease transfers from asymptomatic people, as researchers at Georgia State University reported May 1 from tests conducted in Atlanta.

When 150 people were tested and showed a 7.1% positivity rate, it was not a surprise, Threlkeld says, “because they had a few more cases then we did.”

What was surprising, he said, is the number of asymptomatic people who tested positive for antibodies and were living with people who had not developed antibodies to the disease.

“The possible shred of good news there is that people who were asymptomatic may not be transmitting the virus as efficiently as people who are full-scale sick and need to eventually go into the hospital,” Threlkeld said. “It’s way too early to hang our hat on that, but at least it is something for optimism.”

The problem, he said, is that every test has a sensitivity and specificity.

“Every test can miss some people who really are positive. It can also tell you you are positive when you really don’t have something. Where you set the cutoff for your test can affect just how sensitive or specific it is,” he said.

UTHSC dean: Current COVID antibody tests aren’t always correct

The FDA “crackdown” he said, came after the agency realized “it has let a lot of tests through.”

“You want to make sure those tests are very accurate because if they are not terribly specific, that is to say, if they call some people positive who really aren’t positive, all of a sudden you have a whole new problem on your hands.”

Antibody tests are being used to determine who may give convalescent serum — blood plasma — as an experimental treatment for people who are sick in the hospital with the disease.

Baptist is running its own convalescent serum program. Methodist Le Bonheur Healthcare recently announced its partnership with the Mayo Clinic on a convalescent serum program.

Threlkeld worked with American Esoteric Laboratories (AEL) on its antibody test and has said he believes it is a good test.

AEL did not respond to a question sent by email about its positivity results. 

Dr. Scott Strome, executive dean of the University of Tennessee Health Science Center College of Medicine, has doubts about the efficacy of any of the current antibody tests. UTHSC is creating its own and has received emergency-use authorization from FDA, although the test has not yet been released.

<strong>Scott Strome</strong>

Scott Strome

Strome says many of the tests identify antibodies from other coronaviruses, including the handful that cause the common cold.

“This is the problem I have with the current antibody tests. We don’t know what it means yet,” he said.

His analogy is footprints on the beach.

“Antibodies can stay for a certain period of time or they can get washed away. A positive antibody test means your footprints are there. A negative test doesn’t mean you weren’t on the beach. It just means it doesn’t know you were there. The tide could have washed over your prints,” he said.

“Similarly, if you walked on hard-packed sand, it’s like you walked on the beach and never were there. It’s the same with antibodies. Some patients may not generate an antibody response.

“People who are positive (in antibody test results) should not presume the antibodies they have are to COVID.”

Bluff City Biz

Sign up to get the latest news and analysis from our Business team.

Manage Your Email Subscriptions


Dr. Scott Strome UTHSC Dr. Stephen Threlkeld Dr. Manor Jain
Jane Roberts

Jane Roberts

Longtime journalist Jane Roberts is a Minnesotan by birth and a Memphian by choice. She's lived and reported in the city more than two decades. She covers healthcare and higher education for The Daily Memphian.


Reading comments and joining the conversation are some of the many benefits of subscribing. Join the conversation by clicking the View Comments button below. Not a subscriber? Click here. 

Our commenting policy can be viewed here

Bluff City Biz

Sign up to get the latest news and analysis from our Business team.

Manage Your Email Subscriptions