By Greg Barnes
Ten days after Christmas, a line of vehicles spilled out of a COVID-19 drive-thru testing site in Fayetteville, snarling traffic on a busy street.
That Monday was among the site’s busiest. Its director, Dr. Robert Clinton, said about 900 people were tested that day, so many that he decided to switch from administering rapid antigen tests to polymerase chain reaction tests, commonly known as PCR tests.
Results using rapid tests, which detect protein fragments specific to the coronavirus, can be returned to people in about 15 minutes, with the people who took the tests waiting around for their results.
The PCR tests, quicker to administer, detect the coronavirus’s genetic material. But the tradeoff: They have to be sent to a lab and typically take one to three days to get results back.
Clinton said he didn’t have much choice but to switch to the PCR tests. At 15 minutes for each rapid test result, the line of vehicles extending into the street continued to swell to the point that police had to be called in to direct traffic.
Using the PCR tests, Clinton and his employees could avoid major backups by simply administering the tests, which take only a moment, and sending patients home. The tests get sent on to a lab for diagnosis, and results are emailed to people a day or two later.
The PCR tests not only reduced the wait times, they provided a more accurate form of testing. The PCR tests are referred to as the “gold standard” for detection of SARS-CoV-2, the virus that causes COVID-19, largely because they are more sensitive to viral detection.
But many people and providers like those antigen tests. The rapid tests are significantly less expensive and can be rolled out much quicker. And during the recent holiday season, they provided a quick answer for folks who wanted to visit family members.
The problem is that those results from the antigen tests may have given people a false sense of confidence.
Many false negatives
Shortly before Christmas, many sites, including Clinton’s, continued to rely largely on the rapid tests. Often, people going to those sites did not exhibit any symptoms. They just wanted to get tested before the holiday.
“People wanted to get tested to make sure they were OK to see family,” Clinton said.
For many people, the rapid tests may have provided a false sense of security. Many people probably didn’t know that rapid tests are not always reliable if you haven’t started showing any symptoms of COVID.
A Centers for Disease Control and Prevention report dated Jan. 1 found that one type of rapid antigen test – called Sofia — missed nearly 60 percent of asymptomatic people who had the virus.
Sofia was the first rapid antigen test to get emergency use authorization through the Food and Drug Administration. Many others have since followed. Although some sites in North Carolina use the Sofia tests, Clinton said his site is not among them. Regardless, all of the rapid antigen tests have accuracy problems. They’re definitely not as accurate as the PCR tests.
There are two ways to measure the accuracy of a test. One is how specific a test is to a certain pathogen; is it the right virus you’re detecting. The other measure is sensitivity, namely, whether you can pick up the signal at all.
“You can trust a positive result from an antigen test, but you cannot trust a negative because the test is not sensitive,” said Rachel Roper, an associate professor of microbiology and immunology in the Brody School of Medicine at East Carolina University. “If you test negative, you need to get the PCR test to really know if you have COVID.”
Or to have confidence that you don’t really have COVID.
Rapid tests detect specific protein antigens on the surface of the virus and can identify people at the peak of infection, when virus levels are high, according to the Centers for Disease Control and Prevention. But they aren’t nearly as sensitive to detection as PCR tests, which can pick up tiny amounts of the virus before symptoms appear.
A model developed by the CDC and published last week in the JAMA Network Open found that people with COVID-19 who are not displaying symptoms may be responsible for transmitting as much as 59 percent of all virus cases.
Percent of positive cases double
On Jan. 4, Clinton said, he began using the PCR tests almost exclusively, giving them to about 800 people in a parking lot that once served a K&W Cafeteria. Although some of the results hadn’t been compiled by the next day, Clinton said the number of people testing positive had skyrocketed — from about 15 percent using the antigen rapid tests to 30 percent using the PCR.
“It is hard to know what to make of that,” Clinton said in a private Facebook message. “Are that many more people sick or are we now capturing all of the positives that could have been missing previously?”
Clinton, a disabled veteran who operates the Haymount Urgent Care on Owen Drive in Fayetteville, thinks it’s a combination of both. More people are getting sick, he said, and the PCR tests are more accurate, especially on people who are not exhibiting symptoms of the virus.
Clinton said he and his staff urge asymptomatic people taking a rapid antigen test to follow up with a PCR test if they have been exposed to someone with the virus but have an initial negative result.
“If somebody is in a close family relationship, like living with somebody that’s positive and they have no symptoms and are negative, we still like to do a PCR because those are the ones that are going to be the false negatives,” Clinton said. “So just doing that process captures all those so they don’t get through.”
People aren’t being told
The CDC and the U.S. Food and Drug Administration recommend asymptomatic people who have been exposed to the virus but test negative with a rapid antigen test follow up with the more conclusive PCR test. The N.C. Department of Health and Human Services provides the same guidance to health providers, though it says no further follow-up is needed for asymptomatic people who test negative and have had no close contact with someone who has been exposed.
But people who think they have been exposed and receive a negative rapid test are not always following up with the PCR test, and they are not always being advised that they should.
At least four people, including the author of this article, say they were exposed to the virus, got a rapid test result of negative and were never advised they should also get a PCR test. Two of the four developed COVID-19 symptoms and tested positive a few days later, once the virus had multiplied enough for the rapid tests to detect it.
In the end, the author of this article tested negative.
Because of the limited accuracy of rapid tests and the probability of false negatives, local health departments across the state, including Cumberland County’s, typically won’t use rapid tests unless people are exhibiting symptoms of COVID-19.
“Cumberland County did pilot rapid testing at our test sites only for symptomatic patients and after consultation with a nurse,” Jennifer Green, the county’s health director, said in an email. “Rapid tests that received negative results were sent to the lab for confirmation.”
Clinton didn’t have that luxury. He is paying for rapid tests, which he said can cost a total of about $30,000 a day, out of his own pocket or with money friends have loaned him. He said reimbursements from insurance carriers or the federal government can take a month or two and sometimes don’t happen at all if a person didn’t fill out the required paperwork correctly or didn’t provide a Social Security number. Presently, he said, he is getting about $1 back for every $3 he spends, though he expects that to change soon.
N.C. cases soaring
A big surge in traffic found its way to Clinton’s drive-thru testing site shortly before Christmas. In late November, the doctor said, he and his staff tested about 300 people a day. Shortly before Christmas, that number had jumped to about 800, he said.
The traffic has not let up since. People still wait in line, in lanes that stretch six wide and roughly the length of two football fields.
The percentage of positive cases now being detected at Clinton’s site using the PCR tests remains high, too. Although Clinton believes the infection rate of 30 percent that he saw after switching to the PCR tests has declined, he said it remains significantly high.
COVID-19 infection is soaring throughout North Carolina and the country. According to the N.C. Department of Health and Human Services, 84 of the state’s 100 counties were in the critical red stage for coronavirus cases and spread on Friday. Three days before Christmas, 65 counties were red.
On Saturday, North Carolina reported 11,581 new COVID-19 cases, the highest ever recorded, and two days last week each saw more than 10,000 positive tests. Hospitalizations stand near 3,900, while the percentage of positive test results stood near 14 percent over the past week.
The state’s goal is 5 percent.
To begin the New Year, DHHS Secretary Mandy Cohen issued a dire warning:
“We begin 2021 in our most dangerous position in this pandemic. We have critically high rates of spread in much of our state,” Cohen said. “I encourage you to avoid getting together indoors with anyone who doesn’t live with you. If you plan to see other people keep it outside and very small. Wear a mask the whole time. We must do all that we can to protect one another.”
At about the same time, health officials predicted that another 1,100 people in the state are likely to die from COVID-19 this month.
The 15-minute result is the real game changer here. Besides being 20x more expensive, the PCR tests are taking days to get back results, sometimes weeks. By the time someone learns they’ve got the virus they’ve already infected others. A rapid test, used wisely, can break the chain of transmission.
This is an inappropriate interpretation of these tests. The PCR simply stays positive for too long, making the rapid test look incorrect but it is the PCR test that is remaining positive long after someone is infectious.
Follow me on twitter if you want to understand this more.
I’m a professor at Harvard and study these tests closely. This MMWR report is full of incorrect interpretations.
Comments are closed.