estimates of covid spread and mortality

[Rich Ulrich]

There were reports two weeks ago which estimated that the
US prevalence of Covid antibody was about 9% of the
population (30 million people). One report estimated "up to
14%".

Assuming that the US death count is not too far off, the
200 thousand computes out as 0.67 deaths per hundred cases.
The last estimate of case-fatality rate from the CDC (after some
wilder estimates) was 0.65% -- I don't recall what data that
one was based on, but I suspect it overlapped.

I remember that early averages reported from China were
much higher for Wuhan (4.5%), but were under 1% from outlying
areas. And that was before doctors found drugs that helped,
and learned that ventilators should be postponed as long as
possible. (Placing the patients belly-down reportedly helps
their breathing.)

Furthermore, the "average" is an uneven composite, with a
much higher rate among older patients who have other conditions.

Among the things I don't know about fatality rate is the
amount of improvement since March (or since China).
It /seems/ that for those hospitalized in the US, fatalities may
have been cut in half -- where the hospitals are not swamped
with patients. Most of the other "reduction in fataliy rate" is an
illusion that depends on the lower age of cases.

For New York City, fatalities were also a consequence of the
overload on the health system.

Across time and place, fatalities reflect not only age of cases but
also the chosen treatments and the overload on the hospitals.

"Seasonality" is another puzzle to me. Back in the spring, when
case-rates plunged across the world, I took that as confirmation
of a seasonal effect, rather than a consequence of universal masks and
distancing. However, geography (northern vs. southern hemisphere)
never gave good confirmation of that. Australia and New Zealand
have been punctilious in their protection and kept low rates through
their winter. Brazil suffered.

Now, the oncoming winter in the north is greeted by rise in rates.
How much is seasonal? or owes to the tentative opening of schools?
or reflects a fatigue in maintaining standards? If the fatality
rate rises, will that reflect overload or simple seasonality becasue
more people get more /severely/ infected, or drier air worsens
outcomes?

--
Rich Ulrich