Thread by @sinichol, We've all sorted worldometer's table by deaths/1m, seen the data in the [...]

We've all sorted worldometer's table by deaths/1m, seen the data in the blue bars (to Feb6), and thought, wow

's pretty bad.

But, we're all getting our comparisons so wrong.
i.e. red bars are d/1m just for 50-59yr olds.

Why so different?

Follow me to open your eyes.

1/10

We've all sorted worldometer's table by deaths/1m, seen the data in the blue bars (to Feb6), and thought, wow 's pretty bad.But, we're all getting our comparisons so wrong.i.e. red bars are d/1m just for 50-59yr olds.Why so different?Follow me to open your eyes.1/10

Main cause, 2

factors:

Demographic: a pop's d/1m figure is still wgted by age, so fewer 80+ it's lower

Policy: e.g.

's just managed spread better

Re-ranking most2least aging, a pattern emerges.

For a younger pop to have a high d/1m, it's had more spread & deaths.

2/10

Main cause, 2factors:Demographic: a pop's d/1m figure is still wgted by age, so fewer 80+ it's lowerPolicy: e.g. 's just managed spread betterRe-ranking most2least aging, a pattern emerges.For a younger pop to have a high d/1m, it's had more spread & deaths.2/10

So how much do demographics vary?

Here for:

the countries so far

adding a few more

plus, the world
Are the props. each 10yr age band make up.

50% marks median age, so:
Italy

47
Eng

&Wal

41

30
Nigeria

just 18

So expecting similar deaths overall is silly.

3/10

So how much do demographics vary?Here for:the countries so faradding a few moreplus, the worldAre the props. each 10yr age band make up.50% marks median age, so:Italy 47Eng&Wal 41 30Nigeria just 18So expecting similar deaths overall is silly.3/10

It matters as, per pic,

serology studies show risk goes up 3x with each band.

So, a countries IFR is demographic dependent.

Ignoring healthcare/comorbidity for a sec, it means we can est each countries IFR.

If you infected everyone, Italy will have 10x Nigeria's deaths.

4/10

It matters as, per pic,serology studies show risk goes up 3x with each band.So, a countries IFR is demographic dependent.Ignoring healthcare/comorbidity for a sec, it means we can est each countries IFR.If you infected everyone, Italy will have 10x Nigeria's deaths.4/10

So how do deaths compare?

Here to Feb6 are props. by age in original set of places.

So in terms of death >80:

Eng&Wal: 61%

NY state: 44%

Moldova: 13%

Peru: 18%

Big diffs, but, it is already clear, that there are other factors at play, Norway 64%, Urkaine 20%?

5/10

So how do deaths compare?Here to Feb6 are props. by age in original set of places.So in terms of death >80:Eng&Wal: 61%NY state: 44%Moldova: 13%Peru: 18%Big diffs, but, it is already clear, that there are other factors at play, Norway 64%, Urkaine 20%?5/10

Adding to the pop IFR ests:

Wgted IFR of actual deaths:

than est, skew to old

to young

% dead
and Norway/SK/etc, become clear. Low deaths mean carehome outbreaks look worse.

But, with deaths, protecting old hard:
Austria: worst
Ukraine: best
Eng&Wal: not bad

6/10

Adding to the pop IFR ests:Wgted IFR of actual deaths:than est, skew to oldto young% deadand Norway/SK/etc, become clear. Low deaths mean carehome outbreaks look worse.But, with deaths, protecting old hard:Austria: worstUkraine: bestEng&Wal: not bad6/10

Other factors?

Next biggest reporting.

similar

per http://bit.ly/3a8HZe3 Brazil in places 1/5, Nigeria likely 1/10

15% in UK hosps die, 50% need O2, so outside, 3x.

But, key is that

have not published all-cause yet, so we're only comparing hosp deaths atm.

7/10

Other factors?Next biggest reporting.similarper http://bit.ly/3a8HZe3 Brazil in places 1/5, Nigeria likely 1/1015% in UK hosps die, 50% need O2, so outside, 3x.But, key is thathave not published all-cause yet, so we're only comparing hosp deaths atm.7/10

Where it is about factors like:

healthcare: ICU beds nos, quality, etc

comorbidity rates: e.g. obesity, OECD 4x Pakistan, but we're talking only 15% of pop.

lifestyle impacts: elderly at homeOR in carehomes?

Many cancel out.
e.g. 1st

better healthcare, but fatter.

8/10

Where it is about factors like:healthcare: ICU beds nos, quality, etccomorbidity rates: e.g. obesity, OECD 4x Pakistan, but we're talking only 15% of pop.lifestyle impacts: elderly at homeOR in carehomes?Many cancel out.e.g. 1stbetter healthcare, but fatter.8/10

For sure 2nd

all-cause when pub-ed will change this.

But, till then let's est:

% of poss. deaths

% spread

Shows IFRs in Brazil/Peru likely higher, but they've defo more spread.

100% impossible, but even if IgA/Tcell immunity still real NY shows we could be worse.

9/10