When is a Negative Covid-19 Test Truly Negative?

Amid the concern about the availability of testing for the novel coronavirus 2019 (COVID-19), and the focus on the sometimes lengthy delays in getting results, clinicians and patients should be aware that a subset of tests—no matter how abundant testing is in any particular locale and how quickly results are returned—will generate results that are false.

While both false negatives and false positives are undesirable, false negatives run the risk of increasing community transmission should individuals erroneously believe they’re not infectious and fail to take necessary precautions. This can occur whether people have no symptoms or have symptoms but assume they’re due to something other than COVID-19.

Given the skyrocketing numbers of infection in the U.S., it would be reassuring to know the true sensitivity—or ability to catch true positives—of the slate of tests currently being used. “The figure that a lot of people have been using is 70%, but it’s really fuzzy and hard to hang your hat on it,” Steven Woloshin, MD, MS, a general internist and co-director of the Center for Medicine and Media at The Dartmouth Institute for Health Policy & Clinical Practice in Lebanon, NH, told Contagion®. “The problem is nobody really knows.”

Ideally, any new test given emergency authorization by the U.S. Food and Drug Administration (FDA) can be evaluated against a clinical gold standard that would definitively establish whether or not a person has COVID-19. However, Woloshin explained, such a standard does not currently exist. “[P]reviously authorized PCR tests are used as the gold standard when FDA considers authorizing new tests,” Woloshin said. “That is, the assessment consists of seeing the level of agreement when the old test is positive or negative. This approach overestimates sensitivity because the old test might be [a] false negative.”

According to Woloshin, several recent studies have highlighted a significant rate of false negatives, particularly if the test is administered soon after exposure to the virus. In these cases, tests may not capture viral material because there is not yet enough virus in the person to be picked up. It’s also possible that poor testing technique can miss any virus that is there.

What should clinicians do when a patient tests negative but there is a question as to whether that result is accurate? One solution is multiple applications of even a low-sensitivity test, Woloshin said, the idea being that with more tests comes the likelihood that the virus will be detected. Before recommending repeat tests, clinicians may want to take into account whether the patient has symptoms and whether the patient is in an area that has experienced a lot of infection. A patient with symptoms in a hot spot who tests negative might be reasonably assumed to have the virus, while an asymptomatic patient in an area of low transmission who tests negative probably can take comfort in that negative result.

Woloshin pointed out that it can be difficult to assess a location’s true COVID-19 prevalence without doing random testing within the community, as many people will never seek out a test unless they experience symptoms. Positivity rates at testing centers can give ballpark estimates of viral prevalence, he said.

One potential game changer in the effort to stop transmission would be the availability of inexpensive rapid tests less sensitive ones--that individuals could administer at home, ideally daily.

Researchers at MIT, Harvard, and the Howard Hughes Medical Institute are developing 1 such test that hopefully will allow consumers to do a simple nasal swab or saliva test and get results in an hour, and other labs reportedly are working on similar platforms.

While it’s unquestionably important for people who have Covid-19 to know their positive status as soon as possible, given that some people are much more efficient at spreading the virus than others, Woloshin would like to see testing go even further. “The holy grail would be a test that tells you if you’re infectious,” he said.