Different groups would never have the same variance
Not only do all these age groups have the same variance in this chart, but what *should* happen is that very small groups would have big variance (confidence intervals on this chart).
Well, it appears there are a lot of hypertensive teens with #covid19
Apart from the magical @Surgisphere database, where is the data from?
I'll just say it's weird that 121 US hospitals provide 1499 patients (12 pt/hosp) but 7 Chinese hospitals provide 1507 patients (215 pt/hosp) - for a supposedly US company....
It's a reasonable statistical tool...
If you don't (PSM) match, the groups will be very different (by nature)
The big ones are antihypertensitves (blood pressure meds) and anti-lipid agents (statins).
Neither has been shown to prevent deaths significantly, but keep taking them
So, how convenient that a large study shows.... (guess what)...
And the risk reduction is more pronounced than any other factor (even age) - imagine that!