With the massive growth in computing capacity and the connectivitythat exists in the world today, medicine has changed drastically.These changes bring opportunity for more change to improvehealth care. Medical information is now available at our fingertipsand can be acquired rapidly. As such, the massive memorizingtasks and learning skills of the past are less relevant. Robust clinicalpathways are available in apps, online medical care tools, andhealth information systems, and in some cases are proven to improvepatient care, with failure to follow them resulting in worsepatient outcomes.1 Thirty years ago, there were only a handful ofclinical prediction algorithms/models to assist physicians in decisionmaking. Now there are hundreds. Despite this, the uptake of algorithmsin clinical practice has been slow, sporadic, and fraught withskepticism.2 This uptake, or lack thereof, has been justified by argumentsthat predictive algorithms were developed in populations ofpatients that were not necessarily applicable to “the patient in frontof me.” In other words, studies were not generalizable. Yet physicianshave become all too quick to order diagnostic tests withoutfollowing algorithms or predictive tools under the assumption thatthis is best, and ignoring the possibility of harm, including radiationexposure, false-positive tests, and the economic burden on society.It is estimated that 5% of the US gross domestic product is spent ondiagnostic tests and procedures that do not result in any improvementin patient outcomes.3.
展开▼
