This is a linkpost for /r/discussion/lw/nzi/80_of_data_in_chinese_clinical_trials_have_been/
This raises some interesting questions.
If the end result is fraud and bad medicine, whether you regulate more or less, is that a reason to regulate less so money isn't wasted on mandatory fraudulent studies?
Regulation raises the the barrier of entry to selling medicine. Does this reduce the amount of fraud because it's harder to to sell completely untested medicine and there's at least some quality control by the regulator? Or does it increase the amount of fraud because once a drug costs huge amounts of money to develop and approve, companies are less willing to take a loss if they discover the drug doesn't really work, and so lie more?
Let's take New England Journal of Medicine. It's a journal that pretends to do peer review. It's the highest ranked journal in medicine. The New England Journal of Medicine endorsed the CONSORT guidelines on best practice in trial reporting.
Does the peer review mean that authors who submit papers to the journal get told to modify their papers to fulfill the standards laid out in the CONSORT guidelines? No, the top rated journal in medicine doesn't use peer review to make sure that the papers in their journal fulfill the standards that the journal endorsed...