Physicians' associations, such as the American Heart Association, can issue treatment guidelines based on available clinical evidence, real-world data, and expert consensus. They already do this anyway, and in most cases their guidelines are the default prescription.

Also: Postmarketing surveillance data, peer-reviewed journals, mechanistic analysis, etc. There are lots of ways to decide which drugs might be of benefit. Leaving the decision to the FDA has, to this point, done far more harm than good.