On scary statistics in context.
1/10 patients who have gone into intensive care with Covid-19 had died; another 1/10 had left intensive care alive.

That is a 50% death rate among one-fifth of the cases (but its *not* a 50% death rate among admissions) https://twitter.com/sundersays/status/1247263872286961664

It is certainly serious situation. There is an admirably clear explanation in the quotes attached to this data as to why that 48% rate v.likely to come down once all outcomes are known + why initial sample is unlikely to be representative of later cases
https://www.sciencemediacentre.org/expert-reaction-to-icnarc-report-on-the-first-reported-775-patients-critically-ill-with-covid-19/

"Until we have many more cases the true case fatality rate for patients admitted to an ICU with COVID-19 remains unknown. Strictly ICNARC are not reporting a mortality rate, they are reporting the number of deaths. Commentators are turning this into a mortality rate”

This is a 29th March 2020 release. The numbers are now larger, but in similar proportions. The basic principles & caution in the quotes about projecting from this early sample (especially to any single individual case) appear to remain valid.
https://www.sciencemediacentre.org/expert-reaction-to-icnarc-report-on-the-first-reported-775-patients-critically-ill-with-covid-19/

(Background: this thread offering context on a tweet which appears to have been deleted since. In short, we don't know what the stats are or mean. If knowing the odds matters, its a serious situation but prospect of a full recovery still more likely than not. Fingers crossed).