Make Algorithms Great Again

💬 comments

Why bother learning medicine when there's an easy algorithm for everything?

Ah, if only medicine were so easy that you could run everything through an algorithm. Just upload the patient's age, list of medical problems, current medications, chief complaint, and a photo of the patient into the computer, and voilà! Out comes an assessment and plan, complete with instructions on what to do and when to do it, in a form simple enough for a monkey to execute. If practicing medicine was actually this simple, don't you think robots would have replaced physicians by now? And yet, they haven't. Even the most hypochondriac of patients usually seek actual medical attention after entering their symptoms into WebMD. The simple fact of the matter is, patients and the art of medicine can't be distilled into an "easy algorithm". No two patients even with the same disease are ever the same, and needless to say, patients deserve (and want) more than an algorithm. Algorithms can't account for patient preferences, corner cases, and most importantly of all, clinical gestalt. Apparently this doesn't seem to be a fact that can be appreciated after purchasing pursuing a MSN degree from a 100%-online, for-profit diploma mill that churns out new graduates every eight weeks like E. coli on spoiled beef. This is not to say that algorithms don't have a place in medicine - they certainly do. For example, clinical decision rules such as the HEART score, CURB-65, and Wells criteria are invaluable for risk-stratifying patients and helping to determine appropriate workup and management. But they, as with any other algorithmic approach in medicine, cannot be applied wholesale to patients and diseases. Make no mistake - the medical knowledge and experience required to safely combine algorithms with clinical judgment is something that can only be mastered after years and years of training through medical school, residency, and possibly fellowship.