LEFT: Closed-case Case Fatality Rate 53 days ago (21%).
RIGHT: Closed-case Case Fatality Rate now (21%).
(Source: Worldometers.)
There were 4,376 cases with an outcome 53 days ago. There are 288,082 now. We were told the CFR would drop as time went on and more testing was done.
RIGHT: Closed-case Case Fatality Rate now (21%).
(Source: Worldometers.)
There were 4,376 cases with an outcome 53 days ago. There are 288,082 now. We were told the CFR would drop as time went on and more testing was done.
Here's the thing: the closed-case Case Fatality Rate (CFR) *did* drop. Here's where it stood on March 1: 7%.
Can *anyone* explain why the closed-case CFR dropped from 21% to 7% in February and is back up to 21% now, despite the passage of time and countless newly resolved cases?
Can *anyone* explain why the closed-case CFR dropped from 21% to 7% in February and is back up to 21% now, despite the passage of time and countless newly resolved cases?
(UPSHOT) I am fully aware that "confirmed closed-case CFR" is always going to be *much* higher than "estimated after-action CFR." What I do *not* understand is what it means when confirmed closed-case CFR plunges down to 7% from 21%—due to testing—and then *rises back up* to 21%.
(FURTHER) Anyone who understands CFR understands that *you can't include active cases* because you *can't assume all those cases will resolve with recovery* rather than death. But in looking at that 21% we *do* have to consider testing shortfalls, asymptomatic patients, and more.
(FURTHER2) The data say 25% to 50% of those with coronavirus are asymptomatic. And certainly you have plenty of places that aren't testing enough—including America. But with nearly *300,000* cases closed at a 21% confirmed CFR, it's hard to see how the *final* CFR drops below 3%.
(FURTHER3) So—my question. Is there *really* enough of a testing shortfall; are there *really* enough asymptomatic patients; are recoveries *really* so much more under-reported than deaths...that a 21% CFR at 300,000 closed cases ends up as a post-pandemic estimated CFR under 3%?
(FURTHER4) I expect—that is, I *hope*—someone with expertise will explain how the answer is yes, will say that in fact that 21% with 300,000 cases closed will become an "estimated" 1.5%—or less—with 300 million cases closed. But if so, *when* can we expect to start to see a drop?
(FURTHER5) I ask this because I know many besides me wonder also. Particularly as we know *other* things can cause an *under*-reported CFR: a failure to test post-mortem (which we have); nations lying about deaths (which we have); misattributed deaths (which we have); and *more*.
(FURTHER6) Then there are factors that *don't* involve reporting errors or testing errors that could make that 21% "low" (for now): most notably, the terrifying idea that as hospitals across the world get overwhelmed, whatever CFR we are seeing now is going to *rise*, not drop.
(FURTHER7) So why does this matter? Because right now the *top task* of *everyone worldwide*—as to the coronavirus—is to convince their neighbors and leaders to *take it seriously*. So if experts are *understating* the likely post-pandemic estimated CFR, we need to know it *now*.
