Presumably you can compare people who die vs those recover and no longer need the ventilator. The number that recover after being on a ventilator (for say, 2 weeks, or something) would seem to provide more information (though I suspect it may be a little harrowing to hear the number).

yeah, I wish i had a study or something, but I happened to be listening to the podcast right when I saw the parent question. the guy was an immunologist iirc, so he has some statistical background. He probably was counting patients who either died or got off the ventilator, but I don't know for sure.