Resurrecting this old thread here in case some scroll down just to highlight this excellent video that people really need to see. So many patients get very frustrated or wonder why doctors do things stepwise and don’t just “run all the tests.” Just like in basic science, where “p-hacking” can result in a positive study no matter what, you could just run 30 or so tests and you’ll probably find something “positive.” But does it have anything to do with why the patient actually has symptoms? Every individual is different and no test is perfect, so medicine is probabilistic. It’s why basic bayesian statistics are an essential part of medical school curriculums and tested in medical licensing exams.