I'm inclined to agree with you, however the FDA has problems with a lack of transparency. A pharmaceutical company can run 20 studies, and show one result to the FDA that shows safety and efficacy with p < .05. We already have an irreproducibility crisis in science. Medical studies, more than any other type of science, should be pre-registered with the FDA, and the results of those studies should be made public, regardless of their findings.

https://en.wikipedia.org/wiki/Pre-registration_(science)

https://www.xkcd.com/882/