This is from an exchange between Jerome Groopman and Darshak Sanghavi, discussing Groopman's well-known recent book How Doctors Think. Sanghavi suggested that doctors be graded based on well-measured outcomes so patients can make a well-informed choice when they pick their doctors. Here Groopman points out the problem with this.

This is especially interesting for me, because public schools face a very similar dilemma with accountability. How can we hold doctors and educators accountable in a way that is fair, accurate, and avoids gaming the system?