> Above 99%, anyway. That's good enough. You know that if you have a 3% prevalence of the virus in the population, and test 100 people with a test that has 99%

> Above 99%, anyway. That's good enough. You know that if you have a 3% prevalence of the virus in the population, and test 100 people with a test that has 99% specificity and sensitivity, then 3 will be positive, and most likely test positive, but 1 will test positive despite being negative. Thus, of the 4 people testing positive, only 3 will have antibodies and immunity. Not sure that's good enough. (Note that this problem diminishes not only as the tests get better, but as the virus spreads more.)