Hacker News Reader

Comment by bumby | Hacker News Reader

Dr. David Katz was just in Bill Maher’s show indicating the test has a high false negative rate

That doesn't rule out a high false positive rate. The false positive rate can be even higher than the false negative rate.

Totally agree. The difference being a high false negative rate may be more dangerous because it may mean asymptomatic false negative carriers are still spreading the virus. The downside of a false positive is that people are self quarantining needlessly.

For a patient, a high false negative rate is usually worse. For an insurer, a high false positive rate is usually worse. The perspective matters.

> The difference being a high false negative rate may be more dangerous

At an individual level, false negative seems more dangerous. But at a macro level, a high false positive rate could lead to taking dramatically policy decisions

Sorry, I was editing my comment to add that very perspective just as you replied it seems.

I agree with the caveat that false negative rate is much more relevant for society when a disease is very contagious. Take measles, with an R0 in the teens. A high false negative rate can cause explosive growth in the numbers of people catching the disease, which I think is what makes it relevant to the COVID-19 scenario

What appears to be the potential danger of a high false positive is the narrative that it's safe to end stay-at-home because everyone already has/had it. Which seems to be the story being pushed with any talk of a lot of positives.

Good point

Nope, the narrative is "the mortality rate is similar to flu, therefore draconian measures aren't necessary".

Implementation examples: South Dakota and Sweden.

It seems clear that the infection rate has been severely under counted, meaning mortality rates are artificially high. Even better (from the standpoint of restarting) that "flu-like" mortality rate is concentrated in people over 60 years old.

What that all means together is the economy can easily restart, with the most vulnerable (old and sick) taking extra precautions.

This entire thing has been a fascinating exercise in how poorly central planning can work given bad information. The cure has been vastly more damaging than the disease.

It may have both for all we know. I really do think relatively little is known about this virus at the moment. We learn a lot everyday I'm sure, but it's still somewhat of a mystery. We don't like mysteries, which I'm pretty sure is what generates all the fear around the virus.

The evidence shown is a false negative rate of 30% based on a study of 1,000 patients with the virus. So there’s at least some preliminary evidence.

https://www.medpagetoday.com/infectiousdisease/covid19/85717...

The false negative rate isn't that relevant on this context, and has no obvious relation to the false positive rate.

Besides, the article you pointed down in a reply is about a different kind of test.

I was under the impression the PCR test was a better test. Is that incorrect?

I do disagree that a false negative rate is not important in the context of a disease where asymptotic cases may still be contagious, however.

Reply on news.ycombinator.com