3 ms·
No reliable source at all, saw various people refer to larger numbers and thought that would be conservative. If the false positive rate is actually way lower,
by astrophysician 7y ago
No reliable source at all, saw various people refer to larger numbers and thought that would be conservative. If the false positive rate is actually way lower, then this whole point is totally wrong and I would be happy to just delete this comment rather than misinform someone or discourage someone from getting tested if they should.
FPR: 1%
P(covid|+) = 2.3%
FPR: 0.1%
P(covid|+) = 23%
FPR: 0.01%
P(covid|+) = 75%
FPR: 0.001%
P(covid|+) = 97%
- svara 7y agoI didn't mean to imply that the number was necessarily wrong. I looked around for a reference but couldn't really find anything. The thing is, real time RT-PCR works by detecting specific viral sequences. If those are chosen properly, the FP rate should be super low in principle. However, since the test is so sensitive, even tiny contaminations could lead to FPs. Therefore, I think the FP rate in practice will mostly depend on the sample handling professionalism of the lab doing the test. But I don't have a good intuition for what that could mean in numbers, which is why a reference would have been interesting. Nevertheless, your point is a really generally important one when testing ... pretty much in any situation, so I don't think you should retract it.
- astrophysician 7y agoYea -- that's very interesting and insightful. It sounds like false positive rate is hard to nail down in practice, but I think from what you say there's a good chance it's well below 1%. I won't delete the comment for now, but generally I am trying to be wary of giving actual advice -- I do know stats, but I am not an epidemiologist and I definitely don't want to give out wrong information during a time like this.