tailgater wrote:So why even test everyone?
20% is 20%.
The real number is the percent who test positive that aren't:
A. symptomatic
B. known to have been near a positive case.
this would give the % of the general population that have it.
You would only get that number with a rigorous and RANDOM sample of the population.
And that number will fluctuate with the spread of the virus. It's not really a meaningful number unless it's close to 60% (which seems to be where we get close to herd immunity).
Right now, 20% of the people who are concerned enough to go out and get a test (I don't know about FL, but in MO if you want a test, you first have to actually go to a Dr, then they'll give you a note saying you should then go and get a test) have tested positive. That's a much, much different sample pool than the general population. At the point 20% of FL has contracted the coronavirus, we should expect possibly 40k-80k dead from that state alone.
I don't have a ton of information about FL. But I find it curious that their testing has remained roughly at 60k.
Let's do a little mental exercise...
If you don't increase your test facility capacity, personnel, hours etc, you will have a fixed limit to the number of tests total you can do in your state.
If your demand for testing begins to exceed the daily capacity, you will always have some people who will not be tested that day. They may go untested, or push out to the next day, which would then push someone else out, who may go untested or push out the next... etc...etc....
At the point in time that happens, you're no longer actually testing the number of new cases a day (you never really are, but for the sake of this argument lets pretend the daily new cases tested is normally closer to reality). What you're now testing is the % of people who think they need a test, who actually do.
So.... what makes up the population of "people who think they need a test"?
... people who are in contact with someone who's tested positive (if one worker tests positive, probably 20 or so coworkers need to go get tested... if that person is in a service industry, you may have an enormous number)
... people who feel ill (with COVID, or with a cold or something else)
... hypochondriacs...
there's probably a few other categories. But as you can see, it's pretty easy to see how 80% of the people going to get a test may not actually be infected. And how that percentage could hold pretty stable for a long while (until the virus actually gets to close to 20% infection of the state, at which point you'll start to see the % positive start going well beyond 20%)