Hacker News Reader

Comment by golf1052 | Hacker News Reader

This is taken as a yearly number but looking at week over week numbers [1] Sweden did have excess deaths in March and April.

1. https://www.nytimes.com/interactive/2020/04/21/world/coronav...

View on HN

Other countries already have excess deaths (e.g. US: https://public.tableau.com/views/COVID_excess_mort_withcause...) indicating that not only people have died sooner than they would have (which is irreparable), but more people will have died within this year than in a year without Covid. It could be argued that US & Sweden's policies are not that distant. I guess that US has cities with much higher population density & much more unavoidable commute (as in the higher affected areas in Europe).

Sure, that makes sense, although I think the annualized number is quite interesting too as it appears a risk-off approach has led to a commensurate decrease in other fatalities.

The tail at the end of the reporting period graph looks like lag, not a sudden drop in death rate.

Correct me if I'm wrong, is that not just that this year's total is only 10.25 months of data as compared to prior year's full 12 months?

Don't these death stats have a built in lag because of coroners, medical examiners, and stats authorities?

I’m not sure, but I’d expect Swedish statistics to update the same day, as we have very little if any record keeping on paper, and databases are centralised here.

I'd expect the same. However, reality does not agree. Apparently, the law/rules/whatever state that a death must be reported within 14 days (I have no source for this). I do not know about deaths in general, but for Covid-19 deaths in that Excel sheet on the Arcgis site, the death count takes about 10 days to stabilize. (Why? My guess would be: because public health care in Sweden is managed by the 21(?) regions, each with their on systems for everything.)

I’d be interested in learning this from the source! Thanks for your reply

Digital does not mean instant. There are plenty of batch systems around that process all changes at a regular schedule. A chain of such subsystems (especially when they're not synchronised) can easily delay feedback by days or weeks.

True indeed.

Since King Gustav Vasa who indebted the nation to fund his wars and in order to finance this, forever doomed the swedes to suffer centralisation, in order to establish his taxation regime.

What I mean with this historical tidbit is I’d expect the Swedish state to keep that data in one database, not different databases around the country. But I have no source, and would love to learn, if anyone working in Swedish healthcare IT is on HN.

Reply on news.ycombinator.com