DETROIT – Health officials continue to emphasize the need for accurate testing to identify people infected by the coronavirus.
But research is calling the accuracy of some tests into question.
Imagine getting a test for COVID-19 because you work at a job where you come in contact with lots of people, or you will be visiting your elderly grandmother in her nursing home, and you’re told it’s negative. You feel like you’re good to go right?
Well according to recent studies it could be falsely negative between 20 to 38 percent of the time.
Two recent papers are highlighting a need to re-examine the accuracy of the nasal swab PCR tests used to identify active infection with COVID-19. The most recent one published in the New England Journal of Medicine is specifically calling on the FDA to better highlight the problem and provide clearer prediction rules that factors in this inaccuracy.
Meanwhile, another paper published in the Annals of Internal Medicine that looked at published information from seven other studies with over 1,300 patients found COVID-19 tests may tell a person they aren’t infected when they really are an average of one out of five times.
Interestingly they found the timing of the test relative to when a person was infected played a significant role in how sensitive the test was.
On the first day of infection after exposure the false negative rate was 100 percent. None of the tests were correctly positive.
By day four, the number decreased to 67 percent of tests incorrectly identifying a person as negative.
On the average day of first symptoms, day 5, the tests did better, 38 percent false negatives and it wasn’t until the 8th day that the false negative rate reached its low of 20 percent.
After that the false negative rate increased again from 21 percent on day 9 to 66 percent on day 21.