Pulling bad science apart is the best teaching gimmick I know for explaining how good science really works.
When you prescribe a new drug, often you are prescribing something that has only been tested in a few thousand people for a very short period of time, perhaps only six months, and that's not long enough to know whether there are any medium- or long-term side effects.
There is this peculiar blind spot in the culture of academic medicine around whether withholding trial results is research misconduct. People who work in any industry can reinforce each others' ideas about what is okay.
If you put me in charge of the medical research budget, I would cancel all primary research, I would cancel all new trials, for just one year, and I would spend the money exclusively on making sure that we make the best possible use of the clinical evidence that we already have.
There is actually quite a lot of crossover between the quacks and drug companies. They use the same tricks and tactics to bamboozle people into buying their pills, but drug firms can afford to use slightly more sophisticated versions.
In general, drug companies are reasonably good at developing new treatments, and there's also a lot of good in the industry.