Read on Twitter
Prevalence of a disease plays an important role in your probability of having it given you test 
less relevant for testing COVID-19 infection since RT-PCR rarely (if ever) has false
may be relevant for antibody tests, which are less precise
less relevant for testing COVID-19 infection since RT-PCR rarely (if ever) has false may be relevant for antibody tests, which are less precise
The numbers I have here are all made up, the 
general idea is if your probability of having a disease is low, and a test is not *perfect*, your probability of having the disease given you test may be low
general idea is if your probability of having a disease is low, and a test is not *perfect*, your probability of having the disease given you test may be low
The methods used to test for current COVID-19 infection likely don’t suffer from false positives (although false negatives are possible and there are other potential issues that I highlight in this thread 
https://twitter.com/lucystats/status/1248307278404554759?s=21)
But the antibody tests may be less precise
https://twitter.com/lucystats/status/1248307278404554759?s=21) But the antibody tests may be less precise
