Have the 2 EE's on here heard that the CDC model thinks up to 100,000,000 US citizens have had Covid, TWICE?
Yeah, just looked through the link posted above. Very interesting. Good if true.
Their multiplier between actual and confirmed cases is about 8x. I've long thought that anything above 10x is starting to peg my BS detector. 8x is within the plausible range IMHO. As I'm not an epidemiologist, I have no argument with their number.
Bear in mind, though, that the CDC model has not said anything about what's happened post Sep 30. Their model only accounts for what happened up until then, and had a total number of ~53M.
It was the NPR journalist who added this:
Since then, the CDC's tally of confirmed infections has increased to 12.5 million. So if the model's ratio still holds, the estimated total would now be greater than 95 million, leaving about 71% of the population uninfected.
That's the sort of assumption a journalist will make and an epidemiologist won't commit to.
For example, if availability and use of testing, both for symptomatic folks and for those who are asymptomatic and want "convenience" tests for travel/etc are more prevalent, the multiplier between actual and confirmed cases would be expected to decrease.
On the opposite end, the higher test positivity rate that we've experienced in many areas in the current spike suggests a higher "missed positive" rate. Low positivity suggests you're over-testing and thus not missing very many cases. High positivity means you're not testing enough so there are probably a lot more missed cases.
So maybe 100M is a good number, or maybe it isn't. But the CDC certainly didn't come up with the 100M number.