He makes an interesting point that when the FDA approves a drug, it means that Medicare is automatically on the hook to pay for it, and that a more sensible system might decouple approval and mandated reimbursement by having multiple levels of approval (e.g. 1 star = this is probably safe, 2 stars = it's definitely safe, 3 = there's weak/controversial evidence for efficacy, 4 stars = it is safe and it works, etc.).
But it's interesting to me that we hear about FDA scientists resigning when a panel is too permissive, but not when it's too restrictive. Can we chalk this up to something about the kind of people the FDA recruits? I can recall just one instance of an FDA official pushing the institution to go faster, and that was when the doctor's wife died of cancer [1].
[0] https://astralcodexten.substack.com/p/adumbrations-of-aducan...
[1] https://www.nytimes.com/2016/01/03/us/politics/fda-regulator...