In the Spring and well into the summer, reliance upon the gold standard RT-PCR tests meant a delay in results across the country. At times, results might take up to two weeks, long past the time the individual was contagious. There has been much talk of rapid COVID-19 tests to identify positive individuals more quickly. And we have them now, we call them “Antigen tests.”

There are many different types, but first let’s explain the scientific jargon.

RT-PCR is “real time reverse transcriptase polymerase chain reaction.” This is a molecular test that picks up viral RNA in the person’s’ nostrils. It’s very specific: if you have a positive result, you can be assured you have SARS-CoV-2 viral RNA in your nose.

The RT-PCR is highly sensitive. You’re not likely to get a false negative, which is when you are infected and the test doesn’t find it. That’s good, so you don’t have someone walking around thinking they’re not infected because a test was negative.

From: @Michaelmina_Lab

The flip side to that sensitivity, though, is that the test is so sensitive it picks up trace amounts of viral RNA long after the person has recovered and — as far as we can tell from 8 months of data — no longer contagious. This isn’t technically a “false positive” test. There’s really viral RNA there. But clinically, epidemiologically, it’s not positive for purposes of disease control. New York Times reporter Apoorva Mandavilli explained this better than I can.

San Antonians will remember the Wuhan evacuee who tested negative and was released to North Star Mall. And then subsequent tests came out positive, and everyone panicked. As expected, these later tests picked up non-contagious viral particles. This is not unusual for some RNA viruses.

One can argue that RT-PCR is guilty of “epidemiological false positives.”

The “Antigen” test — and there are many — is a protein test. Instead of looking for RNA, it picks up the protein on the surface of the virus. For this to happen, the virus has to be expressing that protein, which is a pretty good indicator it’s contagious.

For most of the antigen tests out there, specificity is very high. That means that the test only picks up SARS-CoV-2 antigens. If a test is highly specific, you don’t get many false positives. When compared to RT-PCR, the most widely used tests correctly identify SARS-CoV-2 more than 99% of the time, according to the Centers for Disease Control.

For this reason, the CDC gave emergency use authorizations to a number of antigen tests.

But here’s the challenge. The antigen tests are not highly sensitive. If you have a low viral load — i.e. not very many viruses multiplying in your nose or throat, it may not pick them up. Anywhere from 5–15% of people found to have COVID-19 by RT-PCR had negative antigen tests. This is a problem because someone thinks they’re clear, and they’re not.

And, you’re more likely to have a low viral load if you’re asymptomatic. Ok, well, low viral load means probably less contagious, right? Maybe. Last week, in multiple virtual conferences, Dr. Fauci said that up to 50% of infections may be asymptomatic.

Asymptomatic carriers can account for a large proportion — up to 50% — of virus transmissions, Fauci, director of the National Institute of Allergy and Infectious Diseases, told a virtual crowd of critical care clinicians gathered by the Society of Critical Care Medicine (SCCM).”

In July, Dr. Alison Galvani from Yale School of Public Health (my alma mater!) and her colleagues modeled COVID-19 “silent transmission.” Based on data indicating that between 17–30% of infections were asymptomatic, they found that the “presymptomatic phase and asymptomatic infections account for 47% and 6.6% of transmission, respectively.”

Quick diversion: Asymptomatic vs Pre-symptomatic. Asymptomatic means NO symptoms. Ever. Pre-symptomatic means no symptoms at the time of testing, but eventually you develop some symptoms. While COVID-19 can be severe, for some, it’s so mild you barely notice. So you may think that little cough is not a symptom and call yourself “asymptomatic.”

So we need tests that pick up asymptomatic infections. The antigen tests pick up >80% of all infections. A positive is a positive. A negative may not be.

Here’s where it gets tricky. The Council of State and Territorial Epidemiologists, a branch of the CDC, updated their COVID-19 case definition on Aug 5.

There are clinical criteria: basically you have a bunch of symptoms that look like COVID and can’t be explained otherwise. For clinical reasons, a test may not be necessary to treat the individual as a COVID-19 patient (i.e. if tests are not available). Those don’t get counted in overall case counts, but we also have enough tests now so generally laboratory criteria are used.

For laboratory criteria, we start with the gold standard RT-PCR. If that’s positive, you have lab confirmation.

If you have an antigen positive test, it is presumptive.

Now, how to classify?

Case: Confirmed lab, that means RT-PCR positive.


  • Meets clinical criteria AND epidemiologic linkage with no confirmatory laboratory testing performed for SARS-CoV-2. (this means you were a contact to a case and you have symptoms, but we don’t have a lab yet)
  • Meets presumptive laboratory evidence. (means you have an Antigen + test)
  • Meets vital records criteria with no confirmatory laboratory evidence for SARS-CoV-2. (means you died and the doctor writing the death certificate thinks it was due to COVID-19 based on symptoms)

Probable doesn’t mean “maybe, maybe not.” It means likely, we just want to do more to confirm.

Why does this matter? Case definitions are not just to give us daily counts. We need to identify cases — even likely, probable cases — to stop transmission. The goal of antigen testing is to identify infected individuals quickly, so they can isolate. More importantly, so contact tracers can notify their contacts and get them tested, so the virus doesn’t continue spreading. Even 48 hours delay in a PCR test result is too long.

My kids get strep throat all the time. If you get on antibiotics quickly, you’re not contagious after 24 hours. But to get on antibiotics, you need to be tested. Imagine going to the doctor for a horrible sore throat and fever, and then waiting 2 days for a test result to start antibiotics. In those 2 days, you’ve potentially exposed many people.

We don’t have a treatment we can give for COVID. Instead, our advice is to isolate. Sure, we can say everyone should isolate all the time — that’s what we did in March and April. But that is not possible forever.

A positive Rapid Antigen test is presumptive lab confirmation with or without symptoms. Probably the most important people to identify given the likely of asymptomatic transmission are Antigen + and Asymptomatic.

Dr. Rohr-Allegrini is an epidemiologist and tropical disease scientist currently working to prevent diseases through immunizations.