Hacker News Reader

bb2018

This will obviously be studied for years but I imagine plotting taking "total deaths in region X for time period Y - total deaths in region X given comparable previous time period Y" will give a pretty good estimate of the complete effect of Covid. It won't be able to pinpoint individual cases but it seems like a very simple and unbiased way of getting the complete delta in deaths due to Covid.

View on HN

I mentioned this in another thread but check out http://euromomo.eu/ - all-cause mortality vs. baseline for every European country, overall and by age group, going back a decade. There's a time lag because of reporting delay so it will be a few weeks before the current situation in Italy/Spain is well reflected there.

It does only have country-level data, so some amount of excess mortality in one sub-country region (e.g. Lombardy) can get diluted by the rest of the country, although Lombardy is 1/6 of the population of Italy so significant excess should show up.

It's interesting how even the delay adjusted z-score in countries doesn't work very well. Compare the graph of Netherlands to the graph that RIVM presented to the government yesterday (page 6):

https://www.tweedekamer.nl/downloads/document?id=fb38642a-88...

Reply on news.ycombinator.com