I find the idea to be uninspired and not very feasible, but that chart?! I had no idea it was that bad. Makes you wonder how you get approval for nexium unless (a) that 1 patient is fuckin' cured, and (b) the others are advised to stop the drug immediately when it's clear they're not being helped. From its position at number 2, I doubt that is the case.
I think a much more efficient way to reform clinical trials is to go after informed consent reform. The amount of red tape involved in treatment evaluation is so over the top that you can't even contemplate touching patients for under tens of millions, and you can't contemplate actually making a drug for under hundreds of millions. It's almost as if the ethicists who write informed consent rules contemplate patients who (a) aren't sick, (b) have post-graduate degrees in philosophy, and (c) are physicians who practice law on the side when they write the rules.