Levine, Hannah

This thesis addresses the level of uncertainty in medical evidence and practice and asks if this level is acceptable. Current medical standards consider randomized controlled trials (RCTs) as the best, and most certain, form of evidence. Similarly, medical practice largely relies on differential diagnosis and diagnostic testing to diagnose patients. I will argue that RCTs as well as differential diagnosis and diagnostic testing contribute to uncertainty in medical practice. In the former, uncertainty stems from lack of representation, elimination of confounding factors, among other issues; while in the latter, uncertainty originates from technological limitations, similarity or variability in presentation, and number of diagnostic possibilities. I will argue that uncertainty in medical evidence and practice can and should be reduced. This can be done by expanding RCT testing and improving it by adding required mechanistic evidence, and by furthering research into better diagnostic technology and training.