Source: “Variation in False-Negative Rate of Reverse Transcriptase Polymerase Chain Reaction–Based SARS-CoV-2 Tests by Time Since Exposure:, Kucirka, Lauren M, Lauer, Stephen A, Laeyendecker, Oliver, Boon, Denali, Lessler, Justin doi: 10.7326/M20-1495 Annals of Internal Medicine, May 13 2020, https://doi.org/10.7326/M20-1495
The standard test for COVID-19 involves a nasal swab that gets sent to a lab. It’s a test that look for genetic material specific to the COVID-19 virus, and typically is termed a “PCR” test (for the method used to replicate the DNA found on the swab).
Here’s something that I didn’t realize until today: Those PCR tests have an extremely high false negative rate. Something on order of 30 percent.
What does that mean? It means even if you are truly infected with COVID-19, and you get tested, roughly 30% of the time, the PCR test won’t show that you’re infected. You’ll get negative back on the test, but it’s false. The test will come back negative despite the fact that you are infected.
(By contrast, the “false positive” rate of the PCR tests is close to zero. That is, if you really aren’t infected, there’s a very small chance that the test will incorrectly indicate that you are infected.)
When I read that today, I didn’t believe it, so I looked up the reasonably current scholarly literature on it. I’ll present some excerpts from that search below.
The upshot is that when a nasal-swab PCR test registers as positive, you know you’re infected. But when a nasal-swab PCR test registers negative, all that means is that the odds are that you’re not infected. But it is by no means certain.
Near as I can tell, the standard medical advice now seems to be to treat people who have the symptoms of COVID-19, but a negative test, as if they were infected, until such time as a more definitive determination can be made (e.g., via lung CT), or they can be re-tested.
Details follow. In each case, emphasis mine:
"But sensitivity for many available tests appears to be substantially lower: the studies cited above suggest that 70% is probably a reasonable estimate."
N.B., sensitivity is the likelihood that the test will show positive, if you are truly infected. It’s 1 minus the false-negative rate.
Over the 4 days of infection before the typical time of symptom onset (day 5), the probability of a false-negative result in an infected person decreases from 100% ... on day 1 to 67% ... on day 4. On the day of symptom onset, ... 38% . decreased to 20% ... on day 8
Based on this analysis, the false-negative rate for SARS-CoV-2 RT-PCR is shockingly high, even at its lowest on day 8 post-exposure, or 3 days after symptoms. At its best, one out of five people suspected of COVID-19 will test negative. If clinical suspicion is high, infection should not be ruled out on the basis of RT-PCR alone. .