Consider the following.

I’m a physician at the end of more than a decade of training. I’ve dissected cadavers in anatomy lab. I’ve pored over tomes on the physiology of disease. I’ve treated thousands of patients with ailments as varied as hemorrhoids and cancer.

And yet the way I care for patients often has less to do with the medical science I’ve spent my career absorbing than with habits, environmental cues and other subtle nudges that I think little about.

I’ll sometimes prescribe a particular brand of medication not because it has proved to be better, but because it happens to be the default option in my hospital’s electronic ordering system. I’m more likely to wash my hands — an activity so essential for safe medical care that it’s arguably malpractice not to do so — if a poster outside your room prompts me to think of your health instead of mine. I’ll more readily change my practice if I’m shown data that my colleagues do something differently than if I’m shown data that a treatment does or doesn’t work.