3 ms·
what suggests to you that there is any notable number of false positives?
by bananapub 5y ago
what suggests to you that there is any notable number of false positives?
- sokoloff 5y agoPrimarily that "all lab tests have errors" and if you test everyone who is hospitalized for any reason (which is sensible), you're inevitably going to get some false positives. Secondarily, in any low-incidence scenario (such as might be the case for "COVID infections among the vaccinated"), even high specificity tests have a high false rate among positive tests.
- a9h74j 5y agoNot an epidemiologist, but prior to Covid I have never seen a professional risk-benefit discussion around screening tests, which did not give serious weight to the various costs of false positives. Patient anxiety. Rounds of invasive biopsies not without risk. Financial costs of the same.
- sokoloff 5y agoI agree (and declined a particular medical screening test [with the advice and consent of my doctor] at my most recent physical because the overall outcomes are worse for testing than not). I think this one as applied to COVID is some mix of fear and human logical frailty. I can only imagine the hue and cries that we'd get if "Hospital XYZ isn't even testing all their patients for COVID!" came to light. That could be used to feed into the "COVID is a hoax!", "Insurance companies are evil; we need single-payer so that costs don't become an obstacle", and many other narratives.