search

dkobak / excess-mortality

Excess mortality during COVID-19 pandemic
GNU General Public License v3.0
331 stars 45 forks source link

Strange "Excess as % of annual baseline" #12

Closed vovodroid closed 2 years ago

vovodroid commented 2 years ago

Hi, it seems that excess percent in file excess-mortality.csv takes as base mortality yearly one, while other data - covid and excess deaths are taken from very beginning pf epidemic.

For instance according to excess-mortality.csv Israel on 12/09/21 has 7416 covid deaths (that is correct) and 6514 total excess, thus yielding 0.88 undercount ratio. But excess is 13.9%, as if baseline were 46863, number corresponds to one year, but not for more than 1.5 year.

The same for Sweden - it looks like base line for 1.5 year is 91762, but it's regular Sweden yearly mortality.

Or did I miss something?

Anyway it would be nice to maintain separately data for 20 (already done), 21, and from very beginning.

dkobak commented 2 years ago

That is exactly what I am computing, yes. That's what "annual" in "% of annual baseline" means. It's done on purpose, so that if the excess deaths stop, then this % value also stops changing (instead of decreasing to zero as time goes by).

vovodroid commented 2 years ago

So what is taken as annual base line in current excess-mortality.csv? In Israel for instance, because of population growth every year die more and more people, while in Russian opposite.

dkobak commented 2 years ago

Baseline is obtained by linear projection to 2020 of the trend observed 2015-2019.

vovodroid commented 2 years ago

Well, it could produce errors in case of more or less steep trend, thus making difficult comparing different countries.

May you at least add table for 2021 only, to see how countries deal with covid in this year?

dkobak commented 2 years ago

Well, it could produce errors in case of more or less steep trend, thus making difficult comparing different countries.

What else do you suggest?

May you at least add table for 2021 only, to see how countries deal with covid in this year?

Yes, I may do it after 2021 is over, to have the entire year in there.

vovodroid commented 2 years ago

What else do you suggest?

At least extrapolate 2015-2019 to 2021 as well, and not use extrapolated 2020.

vovodroid commented 2 years ago

@dkobak For instance for Israel 2015-2019 interpolation gives death growth about 1.54% per year. So usage interpolated 2020 as base line for both excessive death and percent in 2021 brings error. May be not too much, but nevertheless.