[conspire] COVID-19: wee bit 'o math, and ...

[Michael Paoli]

> From: Texx <texxgadget at gmail.com>
> Subject: Re: [conspire] (forw) Not quite getting it
> Date: Mon, 23 Mar 2020 16:17:40 -0700

> Some of my previous math came from Johns Hopkins, by the way.
>
> I stand by my estimate that if CA can stay below 45k cases at any given
> time, we should be able to avoid mellting down the health system

So, some not-so-random, and semi-random bits (and some opinion and ...):

I think some of the better bits covering the math, and "exponential"
(actually logistic - at least in theory) curve, etc. are these two
videos (and not heavy on the math, so reasonably understandable probably
by most):

(could be better titled, but regardless):
3Blue1Brown: Exponential growth and epidemics:
https://www.youtube.com/watch?v=Kas0tIxDvrg

Khan Academy: Estimating actual COVID 19 cases (novel corona virus  
infections) in an area based on deaths:
https://www.youtube.com/watch?v=mCa0JXEwDEk

So, taken together, and based (upon above and many other sources) ...
incubation period 2 to 14 (or 12 was it?) days, (very) rarely (but possibly
sometimes) longer, average: 5 days

So, what I'm looking for and hoping to see ...
(first) 6 San Francisco Bay Area Counties, "shelter-in-place"
went in place starting this past Tuesday.  Now, enforcement and
following isn't like a binary switch that went from "before" to "after",
where all the stuff before was consistent with before, and after
consistent with after ... but if we presume to a (rough) first degree
approximation that that was and is the case, then I'd expect to see
decrease in rate for the 6 counties, right around 5 days after the
relevant orders went into effect ... and hopefully then see the curve
go from (approximately) exponential, to linear (or better/less than
linear).  If it doesn't do that, or look like it's heading that way,
we may still be in relatively deep crud ... or maybe not quite (if it's
low/slow enough, and tapering down from exponential soon/fast enough).
Anyway, again, rough order approximation ... when we're down from
exponential, to linear, we're (theoretically) half-way to our
total infections.  But until we get to/below linear growth rate, it may
be relatively difficult to estimate how bad this will get (worst case
estimates are semi-easy, but how close we get to worst case really
depends upon what people do and don't do, and also governmental
(in)actions and how that changes peoples behaviors).
Anyway, those are the numbers I'm most keenly interested in seeing.
And if other areas beyond those first 6 counties do very similar,
I'd guestimate we'd see similar trajectories in their numbers (not
necessarily absolute numbers, but the shapes of, and changes to the
curves ... scaled and shifted appropriately for infected and exposed base
when they started to implement change, and if/as relevant, scaled also for
total population (that mostly changes bounding limits)).

Treatments?  I dunno, haven't been following that closely - lots of bits and
pieces of "reports" out there of varying credibility.  My meta-analysis of
some of what I've seen, seems some of the better/best treatments (and may
not be anywhere near fully vetted/verified ... "yet"), cut in approximately
half, the recovery time, and the lethality.  That's not exactly "great", but
can be a rather significant and meaningful improvement.  I'm guessing also
that if lethality is cut in half, likely also permanent long-term health
impacts (such as chronic lung damage) would likely also be cut (very)
roughly by about half.  Also, I'm thinking, if recovery time is cut in
about half, then in-hospital and other health care services time/resource
consumption is cut in about half ... which conversely means health
care system can handle about twice as many impacted people - with the
same amount of resources (as they're there only about half as long) ...
and also presuming that the needed resources (medication(s), etc.) for
treatment are sufficiently available and don't run out/short.
But I think we'll have much (much!) better information on treatments and
how effective and how much they do/don't help, in the weeks ahead as
much more data is gathered and analyzed, reviewed, (re)tested, vetted,
etc.

And, for those that may be interested/curious, I've got a
COVID-19 playlist I've created on YouTube.
Certainly not intended to cover all, or most current, but some basic
good fundamental information and other random bits ... heck, even a
wee bit 'o humor/entertainment:
https://www.youtube.com/playlist?list=PLIOESHELJOCnqaaUqq7AzTOGp-k-2KzKY
Contents/ordering may change, but thus far I'm going for a "short"
list of what I find more/most interesting(/entertaining) and/or more
probable to be of relatively lasting value (some of the material is some
years old, yet very highly relevant).

One of the videos also sources:
https://ourworldindata.org/
Which is an excellent source - also mentioned earlier by Deirdre.