This is just one that was noticed and trackable. How many other quiet half-a-percent-contributors-to-mortality are going unnoticed? I don't think the epidemiolo

This is just one that was noticed and trackable. How many other quiet half-a-percent-contributors-to-mortality are going unnoticed? I don't think the epidemiological surveys for detecting these kinds of thing are well funded, to say the least. There's some sort of Drake equation you can do here: prescription rate times chance of side effect divided by detectable effect size, if the number is below the threshold, it's invisible. It would be interesting to establish something like a holdout group longitudinal study where no new drugs are allowed for 5 years or something and see if their mortality rate differed substantially from the baseline.