~5M is the number of people who tested positive by PCR. That's a small share of the total infected, though, since most people don't get sick enough to seek medi

~5M is the number of people who tested positive by PCR. That's a small share of the total infected, though, since most people don't get sick enough to seek medical attention. The total is probably around 25M Americans infected, 174k dead divided by 0.7% IFR[1]. That IFR was calculated from prevalence of IgG antibodies in random samples of the population. The antibody test thresholds were established using blood from patients who tested positive by PCR. Those patients generally wouldn't have sought care if the symptoms were very mild, and milder illness is associated with lower levels of antibodies in the blood. So that estimate of the IFR is probably conservative (i.e., high if anything, so the estimate of the total infected is low if anything). Different populations will see different IFRs, especially with variation in the age distribution (since the IFR is >100x for old vs. young), and also to a lesser extent with other factors. Immunity passports have been endlessly discussed, and have many problems (e.g., the incentive for people to deliberately get infected) that I suspect will prevent wide adoption. It would make sense e.g. to preferentially assign seropositive medical staff to work with vulnerable patients, though. 1. https://www.medrxiv.org/content/10.1101/2020.05.03.20089854v4.full.pdf https://www.medrxiv.org/content/10.1101/2020.05.03.20089854v...