It's kind of strange that Dan didn't even know what a NAAT was some months ago and claimed he was getting results in twenty minutes from a machine that took many hours and even up to three or more days to process, but now he's an expert on it's accuracy/expert on experts of it's accuracy.
PCR molecular testing accuracy highly depends on time since symptoms first appear. Like MAAH and others stated it can be as little as 30%. Pretty much anything after 5 days should be taken as likely not accurate without numerous redundant verifications. Rapid PCR testing is even worse.
Did y'all know that the RT in RT-PCR doesn't stand for Rig-Talk?