Hey @michaelmina_lab, widescale deployment of antigen testing is still being held back in part due to the real costs of false positives. I’ve heard you say to use rapid nucleic acid for confirmation, but those aren’t always available in all settings (eg. UK home use). 1/

A RAT manufacturer has told me they expect false positives to be random events, generally not correlated with anything in the sample. In practice I’ve seen this - people retest 2 or 3 times and get negative every time (then negative PCR). 2/

So why not just suggest confirmation by 2 or 3 repeated antigen tests, where any one is interpreted to mean a true positive? 3/

Some quick and dirty math suggests this turns a 99.8% specific and 80% sensitive test into a 99.9992% specific and 76% sensitive one. Or, for 3x, a 99.9988% specific and 79.2% sensitive one. 4/4