Day 168: Even the Deer

The world has reached 14 million cases. The US is at 3.7 million cases, while #2 Brazil has hit 2 million cases and #3 India has achieved 1 million. Spain continues in their miraculous defeat of death through bureaucratic incompetence. Massachusetts cases are up a quarter of a percentage point, not to gloat or anything.

Deer have not actually been shown to get coronavirus, but you wouldn't know it from driving around Georgia, where the News & Observer reports that some card has been postering the roadkill with COVID-19 messaging:

A few folks are duly concerned about the specificity of COVID-19 PCR-based tests (having already given up the ghost on the wildly inaccurate serology tests). A June paper in the New England Journal of Medicine estimated their specificity at 70%, based on Chinese clinical studies rather than optimistic theoretical estimates from the test manufacturers themselves. That adds up to a whole lot of false negatives.

The situation of a hospitalized patient with obvious COVID testing negative has been pretty well understood all along, but it's worth looking into the consequences of wide-scale testing with a test that doesn't necessarily pick up a useful number of COVID cases. The paper goes into some numbers on this:

If the test sensitivity were 95% (95% of infected people test positive), the post-test probability of infection with a negative test would be 1%, which might be low enough to consider someone uninfected and may provide them assurance in visiting high-risk relatives. The post-test probability would remain below 5% even if the pretest probability were as high as 50%, a more reasonable estimate for someone with recent exposure and early symptoms in a “hot spot” area.

But sensitivity for many available tests appears to be substantially lower: the studies cited above suggest that 70% is probably a reasonable estimate. At this sensitivity level, with a pretest probability of 50%, the post-test probability with a negative test would be 23% — far too high to safely assume someone is uninfected.

The paper's conclusions aren't particularly notable, and Neil Kurtzman tries to expand on them for the Mises Institute, unfortunately unsuccessfully, because he just pulls a number for specificity (false positives) out of his gluteus and runs with that. PCR tests are, in fact, close to perfect when it comes to not mistaking bits of other things for your target virus, and there are, in fact, no false positives to speak of (unless you're speaking of antibody tests, and we do not speak of those).

It's no good to work through the Bayesean math starting from a gluteal value for specificity. The real reason there are no false positives coming out of Asturias, Spain (the mystery that Kurtzman found so hard to swallow) is that Spain doesn't report serology results anymore. All testing aside, one should not be surprised at zeroes coming out of Spain, where they can't even manage to count their own dead like we do in every other first-world country.

On the lockdown skepticism front, Sweden has returned to their background death rate. Argentina's severe lockdown has not produced the positive results one might have expected, if one thought locking down was any kind of useful medical response, leading to some dissent among the medical ranks:

¿Por qué se instrumentó una cuarentena para individuos sanos cuando no hay registro de tal restricción en la historia de la humanidad? ¿Qué criterios científicos y particularmente epidemiológicos se aplicaron para extender la cuarentena total a cinco provincias sin casos y a otras seis con uno o dos casos? ¿Por qué no se le dio suficiente importancia a la producción natural de anticuerpos por vía del contagio en población no vulnerable, privilegiando la inmunidad adquirida mediante vacunas?

If you enjoy rhetorical questions in Spanish as much as our PlagueBlog staffers do, there are a total of sixteen of them at the link above. El Presidente does not appear to have answered any of them yet.