If by modern medicine you mean only "the best conclusions derived from the best meta-analyses based on the best evidence" then very little of what any doctor do

If by modern medicine you mean only "the best conclusions derived from the best meta-analyses based on the best evidence" then very little of what any doctor does is grounded in that. I wouldn't call that medicine. I'd call that, maybe, "research biomedical science". I think OP's comment is precisely that the practice of medicine has very little to do with what top-quality research would prescribe, and is more like medieval reasoning (like causes like, etc.). Of course the reason behind this is that we dont have much top-quality research; and what "research" we often have is poorly executed and ambiguous (if honest). What we "need" is 10k "disposable humans" manufactured on-demand, for 100k RCT experiments. What we have is often much closer to astrology.