We now know that most published medical research findings are false. That likely includes many of the studies that the FDA relied upon to approve certain drugs.

https://doi.org/10.1371/journal.pmed.0020124

My impression is that the situation has been improving in recent years. Between study pre-registration, larger subject groups, and greater statistical rigor I have a lot more confidence that drugs getting approved now actually do what they say on the label. Of course, this is also part of the reason why it now costs >$1B to bring a new drug to market.