[ExI] England reduces total Covid-19 deaths by 12%

Thu Aug 13 18:11:35 UTC 2020

From: extropy-chat <extropy-chat-bounces at lists.extropy.org> On Behalf Of Henry Rivera via extropy-chat
Sent: Thursday, August 13, 2020 10:15 AM
To: ExI chat list <extropy-chat at lists.extropy.org>
Cc: Henry Rivera <hrivera at alumni.virginia.edu>
Subject: Re: [ExI] England reduces total Covid-19 deaths by 12%

>…I recognize you all are talking about the death count. But I want to bring to your attention an equally slippery number which is the case count. Local hospital COOs I’ve heard from are convinced the positive case counts in the US at least are a vast underestimate due to a lack of available/accessible testing and many people who got sick opting not to get tested. Spike was a case in point for example if I recall correctly. 

-Henry

Hi Henry,

I have never trusted that case number Henry.  It depends on how many people get tested.  I was sick as hell in December and January, the tests at the hospital showed negative for the kinds of flu they are designed to test, so they knew it was something else besides that, led to pneumonia.  My bride and son both were sick about a coupla weeks after I was, but neither had a severe case.  My son’s wasn’t a big deal at all: he didn’t even miss a day of school.

None of the three of us have ever been tested.

I see a clear mechanism whereby the case count could be vastly undercounted (which is good news in a way) while the fatality rate could be systematically overcounted (which is also good news in a way.)  Reasoning: the lethality ratio would be overestimated by a high count on the numerator and a low count on the denominator.

I have had two appointments to be tested, both postponed.  My appointment to do that is now set for November.

We have plenty of reasons to distrust the case count.  Every country in the world has let politics and economics mix with science, we all know what happens in those circumstances.  Every country gets to set their own rules on what counts and what doesn’t.  Various jurisdictions have varying rules on who gets to feed data up to the top levels.  They aren’t carrying along caveats at the ground level (that is a huge indication of a corrupted dataset in itself (sheesh such an embarrassing rookie error too.))

Now we see more and more rookie errors being made, such as that huge biker rally at Sturgis, oh that was a grand opportunity lost.  We could even get past the politics to some extent (what political party are bikers?) and collect a good dataset, but as far as I know, opportunity slipped away, for some of the bikers already returned home, some are just now arriving, and we may lose any semblance of signal in the noise.

Damn.

My intuition tells me that bike rally with a quarter of a million riders should be a super-spreader event.  They cancelled the rally, but it went ahead anyway, so no one is blamed politically.  If that rally doesn’t create a new surge, then I don’t know anything about how the flu spreads.

In any case… I would think the death count is more a reliable number because they do test the corpses.  We know they are including accidents, suicides, suspected cases and all that, well OK then.  We are back to looking at increase in mortality rates for the year and recognizing that that number is a mixture of those who died of covid, with covid and those who died as an indirect result of the consequences of covid for other reasons.

spike  

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20200813/53ec2961/attachment.htm>