However, individual experiments are not studies in the academic sense. Especially if they are not formally reported so that the experience is not lost in time.
Which is why officially, people have a choice in participating to research studies but aren't allowed free access to whatever drug they want to take. This opinion is also supported by current medical ethics and practitioners have a duty to make their patients clearly understand that.
Fortunately for the rest of the world, american patients have a lot of trouble in understanding that last point.
Eggs are healthy, eggs are unhealthy, eggs are really unhealthy, eggs are healthy, eggs are really healthy.
I had trained medical professionals tell me twenty years ago that I was going to ruin my kidneys with my low-carb dieting. Twenty years later, I'm still keeping the weight off and medical science has started to catch up to what I knew just by experimenting on myself with moonshot claims.
The biggest problem with medical science (just look at how COVID issues were butchered by politics and celebrity), is that they refuse to just say, "We don't really know." Because all too often, they have no clue. But they are trained to come up with answers and sound confident about it.
That is what OP says.
What we open see is that "we don't know" is often equated with "does not work".
(I guess that also very frequently how it works out)
Regarding the "we really don't know" part: one can observe that people here are basing their own shamanic medicine on what they read in scientific papers, suddenly pretending that those papers can be trusted while they'll disparage the theories of clinical practitioners as being old-fashioned and manipulative. Well, it happens that many of those theories are based upon lots and lots of research and certainly more so that the "fad of the day" is. And yes, those theories are often refuted after years of clinical application, which does not bode well for experiments based on "few papers shamanic medicine".
1. people use their own "shamanic" medicine based on published scientific papers,
2. people disparage clinical practioners' theories which are based on research and which are later often refuted,
3. you conclude that the "many of those theories" of (2) are "based on lots of research" but often flawed,
4. which "does not bode well for" experiments based on the papers of (1).
What does (2) have to do with (1), unless the content of the papers and "experiments" in each are closely related?
Perhaps you could clarify what you were trying to say here.
2. They then proceed to ask their doc for metformin, since it's a prescription drug. Unless the doc wants to commit career-seppuku or the patient has a reason for taking metformin according to current guidelines based on years of clinical use and research, the doc will refuse even when presented said Pubmed evidence.
Conclusion: people who want to take metformin will look into alternate sources, because they built a preconceived belief and nothing a sensible professional says will change their mind. People then proceed with metformin and if a change in indication occurs in the future that agrees with their belief, they will feel vindicated while in fact they are just lucky fools.