The Problem of False Negatives

(Comments and reposts are welcome.)

In logic, you can confirm that something exists, but you in fact cannot prove the opposite. Can you “prove” that angels, Martians, ghosts, or Norse gods don’t exist? Think about it? Normally, we don’t try — we relegate these things to the category of “matters of faith” (well, maybe not the Martians) and don’t try to discuss them in either scientific or rational terms.

What about infections?

June 6th, 2020: In honor of the anniversary of D-Day, June 6th, 1944, and to put a context around how little a sacrifice wearing a mask really is

Does a negative test result mean you are actually uninfected? It turns out, it doesn’t.

We know that’s true with Lyme Disease. If it’s not treated within two weeks of infection, the virus can hide in the brain where antibiotics cannot touch it and pop out during periods of high stress to ravage the body. Same with shingles. During the remission periods, testing may not see it.

Unfortunately, the same may be true of the SARS-CoV-2 virus, which manifests itself in the body as Covid-19. We know people get negative tests and then show the symptoms. These “unconfirmed” cases aren’t counted in the infection totals, nor are then isolated from the rest of the population. And they are infectious.

What we don’t know is whether the ability to hide and produce the false negative test result is a characteristic of the virus or a defect of the test. To this point, medical researchers at Dartmouth encourage doctors to doubt negative results when there is reason to do so:

“A negative result on even a highly sensitive test cannot rule out infection if the pretest probability — an estimate before testing of a person’s chance of being infected — is high, so clinicians shouldn’t trust unexpected negative results.”(1)

In statistical terms, we have to apply a Bayesian spin to test results. The risk that a person is infected should play a factor in interpreting test results.

This plays into what I’ve said in other posts: you have no way to know who is infected, so you have to protect yourself with everyone.

Of course, some people say — “just get infected and get it over with.” The problem with that approach is that there is a significant risk of permanent heart, lung or brain damage from the disease. Like Lyme, Chicken Pox or the Energizer Bunny, this is a disease that can keep on giving. For how long? We don’t know. It hasn’t been around long enough to know how long it hangs around or what else it can do.

We have limited testing in the US, and the tests including the one for antibodies can produce 30% false negatives. Since we only count “confirmed” cases (positive test results), both the number of people infected and deaths due to Covid-19 are larger than we are showing today.

Sources:

1. Steven Woloshin, Neeraj Patel, Aaron S. Kesselheim. False Negative Tests for SARS-CoV-2 Infection — Challenges and Implications. New England Journal of Medicine, 2020; DOI: 10.1056/NEJMp2015897
2. The Geisel School of Medicine at Dartmouth. “New report examines challenges and implications of false-negative COVID-19 tests.” ScienceDaily. ScienceDaily, 5 June 2020. <www.sciencedaily.com/releases/2020/06/200605140508.htm>.