> A number of doctors should have been raising a number of red flags, because their knowledge contradicted those claims
Purdue committed fraud, but they never made claims that were so outrageous that they would be prima facie false. (If that had been the case, Oxycontin would never have been approved in the first place). The problem wasn't that the claims were outrageously impossible; it's that the claims were wrong, and Purdue either falsified or misrepresented clinical data in order to convince doctors that Oxycontin was safer than it actually was.
> Patient evidence right in front of their eyes
Encouraging people to rely overly on anecdotal observations is incredibly dangerous, especially when most doctors who prescribe opioids don't prescribe them in such numbers that these issues would be readily visible to them in the first place.
Yes, there were a small number of doctors who did aggressively overprescribe and misprescribe Oxycontin. That was the result of a direct and explicit campaign by Purdue to identify and enable those doctors. Some of those doctors were prosecuted and/or did lose their licenses, but those doctors still only amounted to a small part of the opioid crisis.
> I'm obviously not saying they should ignore or distrust all clinical data, but if they see such massive discrepancies then they should be speaking up.
Most doctors weren't seeing massive discrepancies. We know the discrepancies exist now because there were lawsuits and prosecutions which revealed the specific fraudulent behavior that Purdue and others were engaged in. But without the benefit of hindsight, these issues simply weren't reasonably visible to a typical doctor - they were only visible at large scale to Purdue, who went out of their way to mislead everyone else about it.