Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

> and I was referring to this data [1]

You were referring to data more than a week old, when the infected cases had neither time to recover nor time to die. The latest figure from yesterday is 1.16% of all cases according to JHU.

> That's not how CFR works

That is exactly how CFR works. If you test more people, C will be closer to I.



No cfr is only I at the end of the epidemic. Either way I slides the data based on those likely exposed.


No. IFR is mortality rate at the end of the epidemic. CFR exactly matches IFR if all infections are diagnosed.


No because you don’t know how many of them will die. In your definition you know the R not the F. Look, I’m quoting the WHO report. You’re welcome to take it up with them.


R is ratio or risk. It is directly computed from C and F just as it is directly computed from I and F.

I don't seen any quotes in your comment.


Case fatality rate is defined as the fatality rate of known cases. You can’t know the final mortality rate until everyone’s either dead or recovered. While anyone has it, it’s preliminary, and it’s called the CFR. Happy to dig the report up for you but this is just a logical conclusion.

CFR matches IFR not when all cases are diagnosed but when all cases are resolved.


As I pointed out in my earlier comment, IFR is not mortality rate. It is the fatality rate to date of those who have been infected. IFR equals mortality rate at the end of an epidemic. CFR will only equal the mortality rate at the end of an epidemic if all infections have been diagnosed. That is, if C=I.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: