@prieto_alhambra I was thinking about your request for #COVID19 data from UK. The issue seems to be that some of the info from UK has data missing NOT at random (MNAR)
What do you think
@tmorris_mrc and @t_mpham ?

For example, #COVID19 testing depends on the likelihood of the test outcome.

On the other hand, NHS reporting of deaths is more likely to be missing at random (MAR) i.e. missing data relates to where you have died, not whether you have died. However, there seem to be some issues with delays in reporting

I don't know enough about ICU data, but this might be one of the strongest outcome at the moment? However, this may also be MNAR or MAR? as those who end up in ICU may change over time due to capacity?

Having said this, it may still be worth examing time trends in ICU and mortality

Also it would be good to see if the gender differences are universal