The most common test for Covid-19 involves inserting a swab up through the nose to collect mucus from the back of the throat next to the nasal cavity.
The resulting mucus sample is then analysed to find traces of viral RNA specific to Covid-19, a process known as reverse transcription polymerase chain reaction, or RT-PCR.
If Covid-19 RNA is detected then it’s highly likely you’re infected, although operational errors such as sample contamination could be to blame. However, if none is found you might still be infected. Results where you are infected but no evidence of infection is found are known as false negatives.
There are a number of reasons why Covid-19 RNA might not show up in your mucus sample. One is timing. Virus particles are more likely to be present in the throat during the beginning of the infection. Another is mild symptoms. Those with mild symptoms are more likely to experience false negatives.
How common are false negatives?
A study in China looked at 1,014 symptomatic patient tests in Wuhan. 601 (59%) of them were RT-PCR positive and 413 (41%) came out negative. However, in addition to the RT-PCR tests, all of the patients were given CT chest scans, another way to test for lung infections such as Covid-19. According to the scans, 308 (75%) of the patients with negative RT-PCR results showed lung symptoms consistent with Covid-19.
This suggests that using only RT-PCR tests to detect Covid-19 could miss 31% (41% x 75%) of infected patients.
Other research from China suggests a false negative RT-PCR test rate of 30%.
Without proper information and advice, there is a risk that people with the disease receiving false negative RT-PCR test results, believe they don’t have it, and behave in ways that risk spreading the infection.
In addition to the effects of test rationing and a failure to establish epidemiological links for newly discovered infections, false negatives are another way that the number of recorded infections could be undercounted.
The high rate of false negatives also helps to support the hypothesis that people testing negative prior to release from hospital, who later test positive, haven’t fully recovered. Rather than catching the virus again, they might instead have received a false negative test result prior to release.
Coronavirus testing research paper (in English)