CORONAVIRUS Death Thread

[Lojik]

Originally Posted by Lune [You must be logged in to view images. Log in or Register.]

89% accuracy or 'sensitivity' in tests like these is very common and some level of false positive results are inherent in all clinical tests. For example, those flu swabs you get when you go to the doctor have a similar level of sensitivity. They are still useful.

So this is part of the problem. You have laymen like me who fail to properly explain things. The issue (I think) with this stanford study, in addition to selection bias, was that they had two false positives out of 401 tested samples the actual range of the false positive rate could be much higher than 1.2%, meaning that false positives could account for many if not all of the positive cases. Or at least this is my understanding trying to read critiques of it. I'm assuming that had this been done on a much larger scale with the same ratios that it would be much more statistically...robust is maybe the word? It's also pretty contradictory to the fact that we've already had over .1% of the population of NYC die from covid19, and unlikely that the entire population of NYC has been infected already or that the death total will stop right now.