As a general rule Doctors trust the documentation given to them by the pharmaceutical companies, which in turn is approved after several rounds with the FDA, so

As a general rule Doctors trust the documentation given to them by the pharmaceutical companies, which in turn is approved after several rounds with the FDA, so one would think that the documentation would be pretty reliable. The source of the issue really is just how hard Purdue was fudging their data. And I mean, you can say that a drug company shouldn't be the sole tester of it's own drugs and I'd agree with you, but good luck getting average people to want to pay for it by funding the FDA.