It might be helpful if I explain a little about how clinical knowledge is advanced and compare this to anecdotes.
I'll start with the lowest standard of evidence, the case report, which you can find in medical journals, usually for very unusual things, and also in grand rounds and conferences, and for training students and residents.
A case report starts by describing the patient and symptoms. "A 44-year old African-American male presented complaining of pain in the lower back, etc. etc." A physical description of any unusual appearance (e.g. the patient had purple striations along the abdomen and upper legs and the face appeared swollen with indications of periorbital edema). Various initial measurements must be given (blood pressure was 140/100, heart rate 95 bpm, respiration 18. Pitting edema +2 was detected in lower limbs). Then any initial laboratory measurements (blood measurements that might be relevant, which always include things like blood glucose, electrolytes, certain enzymes and blood biomarkers that might be relevant to the initial presentation or eventual course of the disease, since these are always written after the fact). Then the treatment is described, including drugs, dose, route and frequency of adminitration, responses including adverse ones, and time course of the clinical response. The clinical response must include quantitative responses including lab values (which might include blood gasses, serum electrolytes, inflammatory markers, etc. in the case of a respiratory infection), as well as the results of any imaging that was done. Of course all adverse effects of the treatment are documented in detail, indeed if an unexpected bad response was observed this might be the whole point of the case report. Eventually the final outcome is described: The patient got better, and maybe was released from the hospital (happy ending) or didn't get better and is still dealing with the problem (not so happy) or died (definitely the least happy outcome). And lastly, a brief discussion of what it might all mean in the context of existing consensus.
Case reports are useful when a disease that seemed to usually follow one course suddenly followed another unexpected course in a patient with a set of preexisting conditions, or when a new treatment seemed to work, or didn't work, etc. etc. You get the idea. We use them a lot for teaching medical students and even occasionally undergraduates because it can make learning the basic science a lot more interesting, even to the instructor.
But the thing is, without all of that information about the patient, including the quantitative lab information, a case report is worthless. Compare this to the anecdote in the NY Post (owned by Rupert Murdoch, who also owns Fox News as part of his media empire), where it amounts to "I was really sick and it seemed like I was going to die and they gave me a single dose of this magic drug pill and I got better by the next day, and God Bless Donald Trump". The last few words are implicit because after all, it is the NY Post.
And hence my really skeptical response.
