Coronavirus thread

[Doug]

Imperial College of London




https://www.imperial.ac.uk/mrc-globa...report-34-ifr/


FWIW ymmv


Report 34 - COVID-19 Infection Fatality Ratio Estimates from Seroprevalence


“The infection fatality ratio (IFR) is a key statistic for estimating the burden of coronavirus disease 2019 (COVID-19) and has been continuously debated throughout the current pandemic....


Using these age-specific estimates, we estimate the overall IFR in a typical low-income country, with a population structure skewed towards younger individuals, to be 0.23% (0.14-0.42 95% prediction interval range). In contrast, in a typical high income country, with a greater concentration of elderly individuals, we estimate the overall IFR to be 1.15% (0.78-1.79 95% prediction interval range).


Go to page 11 on the pdf study for the age chart...
https://www.imperial.ac.uk/media/imp...-Report-34.pdf

[Doug]

[Nephrology]

Their stratification by national wealth is interesting.

[TGS]

Their stratification by national wealth is interesting.

Without further digging, I'd attribute it to the same thing you and I were talking about a week ago in PMs.

Healthcare structures are incapable of effectively tracking it, and the governments' not just incapable but also uninterested. People get sick and die at home, and it's usually attributed to god's will, the river spirits, a hex, or some such. At the very least, mortality is not viewed the same as in western nations and that impacts the tracking and reporting. I would expect reported infections and specific deaths to go down as a country's wealth decreases.

tl;dr: third world gonna third world.

I'd be interested to see what a rigorous peer-reviewed article came up with to explain it.

[Nephrology]

Without further digging, I'd attribute it to the same thing you and I were talking about a week ago in PMs.

Healthcare structures are incapable of effectively tracking it, and the governments' not just incapable but also uninterested. People get sick and die at home, and it's usually attributed to god's will, the river spirits, a hex, or some such. At the very least, mortality is not viewed the same as in western nations and that impacts the tracking and reporting. I would expect reported infections and specific deaths to go down as a country's wealth decreases.

tl;dr: third world gonna third world.

I'd be interested to see what a rigorous peer-reviewed article came up with to explain it.

Yup, exactly my thoughts. That and I'd guess too that poorer countries don't have the ability to keep unhealthier people alive quite in the same way that wealthier nations do. Not a lot of dialysis centers in Djibouti I imagine.

[luckyman]

As are we.

Well, hopefully I’m not speaking too early here, but it seems she has taken a turn for the better in the last 2 days. She never had to get intubated or even get oxygen I believe, and it has been something like 14 days since her initial exposure, so we are hopeful.

A master of understatement, Her quote is “this Covid thing is no joke”.

[Maple Syrup Actual]

Well, sounds like your moniker is apt!

[Glenn E. Meyer]

Good news, hoping for more!

[luckyman]

Well, sounds like your moniker is apt!

Hah, well yes for many reasons. I got that nickname when friends at a restaurant I used to go to met my wife. I definitely “married up” [emoji1]

[whomever]

"It's a case fatality rate...meaning fatalities vs confirmed cases within an age group. It's not fatalities among ALL living people of those age groups. "

Indeed. Let's break it down:

1)Some one says 'the fatality rate for 70+ is .054%'. We want to do a sanity check on whether that can possibly be true.

2)If that was true, then the number of observed fatalities would be .054% times the number of 70+ people who have had covid.

3)We don't know how many 70+ people have had covid, but we can reasonably guess that it is less than the entire population (since some of the population is younger than 70, and some of the population hasn't had covid).

4)So, if the original rate is correct, we can't have experienced more than .054% times 330 million deaths. But we have observed more deaths than that. Therefore the original claimed rate cannot be correct.


Engineers call this kind a sanity check, to see if the values are even remotely possible.

Another example:

There is a guy called John Allen Paulos who has written a couple of great books on innumeracy. He gives an example in the preface of one of his books: he's sitting at a Phd defense, and the candidate makes the comment 'Child deaths due to gun violence have doubled every year since 1950' (this was in the 80's or 90's). Can that claim possibly be true?

Well, assume 1 kid was killed in 1950. Then 2 would have been killed in 1951, 4 in 1952, 8 in 1953 and so on. By 1980 the number would have been well in excess of the total world population - so the claim can't possibly be true.