Skip to content

Latest commit

 

History

History
43 lines (28 loc) · 2.04 KB

README.md

File metadata and controls

43 lines (28 loc) · 2.04 KB

Official COVID-19 data from Sundhedsstyrelsen

This is the unofficial repository for the official COVID-19 data covering Denmark from Sundhedsstyrelsen.

Why

Sundhedsstyrelsen is not willing to publish the COVID-19 data in a machine readable format.

For the public to be able to analyse the situation themselves, this repository is created to provide machine readable data for the COVID-19 virus.

How

Most recent data is grabbed from https://www.sst.dk/da/corona/tal-og-overvaagning using a quick-and-dirty script. However...

Some historical data that once was released and later has been removed from the above source has been added manually in certain files. That includes

  • Tested and verified before 16.march 2020

Some data is updated manually from a human reading an image showing the data. That includes

  • Deaths

Semi-automatic

There are some handy scripts to some of the manual work in fetching and updating CSV files, this can be found in sst2csv.

Comments on data

Data sources and quality

Some data comes from the regions. Then it is then sent to Sundhedsstyrelsen (SST) and is then published on the page mentioned above. Some data is collected by Statens Serum Institut (SSI) and then sent to SST and is then published on the page mentioned above. A number of times we have seen that data for older dates have been changed on the page. The reason for this is unknown, but it might be all natural, we get wiser of history as time goes. Finally the information is then copied again to https://politi.dk/coronavirus-i-danmark/foelg-smittespredningen-globalt-regionalt-og-lokalt

Time shift

The Tested, Confirmed and Death count seems to be registered by number at midnight. That is numbers of tested, confirmed and deaths occurred during a specific calendar day. The other numbers are published around 13:00 or 14:00 every day, so that is most likely the correct valid number at noon (12:00) every day. So there is most likely a time shift in the tested, confirmed and death count by approximately 12 hours.