The first challenge with rapid at-home self tests for COVID-19 is getting one. But once you have a test, there are now growing questions about what body part you should swab to get the most accurate and reliable results. Your nose? Throat? Cheek? What about your saliva?
All of the self tests authorized by the U.S. Food and Drug Administration (FDA) are validated only with nasal samples. But recent coverage of the fact that Omicron tends to cause sore throats, and a small study finding that test samples collected in the throat might detect the presence of the virus earlier than samples collected in the nose, have led to growing online questions about whether people should swab their throat instead of their nose when using a rapid self-test. In the U.K., some testing kits include instructions for people to first swab their throat and then the nose to get the most sensitive reading.
Here’s what we know so far about the different ways to swab for COVID-19 rapid tests and how effective they are.
What we know about nasal swabs
In the U.S., the official word is that nasal swabbing works well for rapid tests. The FDA authorized the tests based on studies showing that nasal swab samples are effective for picking up signs of SARS-CoV-2, and manufacturers of the at-home rapid antigen tests also stress that their kits are designed and most accurate for testing with nasal samples. A spokesperson from Abbott, which makes the popular BinaxNOW kits, says the test, using a nasal swab, is 95.6% sensitive in detecting virus in people within seven days of having symptoms, when they are the most infectious. Siemens, which received the most recent authorization for its over-the-counter rapid test, declined to comment on the throat vs. nasal swab issue. BD, another test kit maker, also declined to comment except to say that its test “is not authorized for use with a throat swab or a saliva sample.”
Some recent evidence also points to the effectiveness of this method. Researchers at the University of California, San Francisco (UCSF) reported the results of a study comparing nasal swabs done using at-home test kits against the gold-standard PCR tests that require deeper nasopharyngeal swabs performed by health professionals. They found that the at-home self-swabs of the front part of the nose performed almost as well as the PCR tests.
They published their findings on a pre-print server for scientific papers, so the results have not been peer-reviewed. But the large number of participants and the verification provided by the PCR tests suggest these results are reliable.
The study involved 731 people who went to a free community testing and vaccination site run by UCSF, the Chan-Zuckerberg Biohub, the Latino Task Force and the San Francisco Department of Public Health. People who consented to the study were swabbed four times by a medical professional: once in each nostril using Abbott’s BinaxNOW rapid antigen test swab, and once in each nostril using a longer PCR swab. The results were compared for the two different testing methods. Compared to the PCR tests, the rapid tests were 95% sensitive in detecting Omicron cases when people were highly infectious and had high levels of virus; the tests were less sensitive, as expected, in registering positive with lower levels of virus.
Those results provide reassurance that the at-home tests can indeed detect which people are infected with the Omicron variant. Because the results correlate well with the PCR results, they also provide support that the nose is a good place for collecting samples for the test.
The group is planning to test people by sampling from both their throats and noses using the at-home kits in coming days and compare those results.
Why are people interested in throat swabs?
That might have to do with a changing understanding of how Omicron in particular journeys through the human body when it infects a person, and where it likes to settle in and set up its virus-copying hub. As a respiratory virus, SARS-CoV-2 infects people through the nose and mouth by way of virus-containing droplets that can travel in the air and also transfer from person to person via certain surfaces. While previous versions of the virus focused relatively quickly on the lungs—burrowing into tissues there, replicating and potentially causing serious and life-threatening breathing problems—studies show that Omicron prefers the nose and throat. (In fact, one of Omicron’s hallmark symptoms is a sore throat). Some early studies also show that the virus might start in the throat and move into the nasal passages, meaning that swabbing the throat might provide positive results in someone who is infected earlier than swabbing the nose.
“Based on the pathogenesis of the disease, we are seeing hundreds of people with COVID-19 [in whom sore throats] are so prominent with Omicron,” says Dr. Diane Havilr, chief of the HIV/AIDS Division at UCSF and the study’s senior author. “That’s different from what we’ve seen with all the previous variants, so it would make sense that virus levels are higher in the oral vs. nasal specimens. It’s definitely plausible.”
But the FDA has not authorized any at-home self tests with throat swabs and has warned of the potential dangers of sticking a swab near the throat. “When it comes to at-home rapid antigen COVID-19 tests, those swabs are for your nose and not your throat,” the agency recently posted on Twitter. The warning came as some people started swabbing their throats with their at-home kits after having read about other countries where throat swabs are more common, and about extrapolating from some studies suggesting saliva samples tested using PCR can provide as reliable, or even more accurate results, than deep nasal samples used for PCR.
Do saliva tests work for rapid COVID-19 testing?
Havilr’s study also included results from using cheek swabs, and these “underperformed,” she says, compared to the nose samples. Cheek samples tested using both the rapid tests and PCR missed a significant number of positive cases, with a sensitivity in the single digits.
That doesn’t mean that saliva tests aren’t useful. While cheek samples contain some saliva, saliva tests are designed differently than rapid self-tests, and currently the FDA has not authorized any completely DIY COVID-19 tests using saliva. Available test kits allow people to collect saliva at home, but they must send the sample to a lab. Technicians then use specific procedures to detect the virus and cancel out the other components in saliva that can confound the results.
The bottom line
It’s still an open question about whether throat swabs, especially in the context of Omicron’s dominance, should replace nasal swabs in at-home tests. If throat swabs can detect Omicron earlier, that’s one day sooner that someone who is infectious can isolate and limit the spread of the virus. Yet some experts, including throat specialists, agree with the FDA about the potential dangers and note that people can harm themselves if they don’t swab properly, while others feel the risk isn’t large enough to discourage people from trying it (after watching videos or getting some guidance.)
“I personally swab my throat and my nose to get the best sensitivity when I use the over-the-counter tests at home,” said Dr. Michael Mina, chief science officer at eMed and formerly at the Harvard T. H. Chan School of Public Health, during a recent virtual Q&A session at the Consumer Electronics Show. “But I do recognize that I do that at my own risk when I go off label.”
Gigi Gronvall, senior scholar at the Johns Hopkins Center for Health Security at Hopkins Bloomberg School of Public Health, also sees little harm in swabbing the throat to test. “It’s very low risk experimentation,” she says, as long as people watch online videos that can help.
The data so far, however, seem to suggest that taking that risk isn’t necessary. If the nasal swabs are pretty equivalent in terms of sensitivity to PCR in picking up the virus, then they should be a reliable way to test for COVID-19. A positive rapid test likely indicates that the person does have COVID-19, but the result should be confirmed with a PCR test. A negative test should be corroborated with another rapid test a day or so later.