Game Theory, the Prisoner's Dilemma, and the Residency Match

Eric J. Warm M.D., F.A.C.P.

Richard W. & Sue P. Vilter Professor of Medicine
Program Director, Internal Medicine
University of Cincinnati Academic Health Center

Overview: Medical students are applying to record numbers of programs despite evidence showing that most will match in their first six to eight choices. Programs are adjusting to the increase in applications with varied strategies, but the burden is significant. Many authors offer solutions, but the problem only gets worse.

In this session, we will analyze this behavior and the National Resident Matching Program (NRMP or "the Match") through the lens of Game Theory - the mathematical approach to situations involving two or more decision-makers. Each decision-maker has a number of different actions available, and outcomes depend on what all players choose to do. The Match is a classic "game with incomplete information." In this circumstance, players know their own outcomes (programs know their interns and how skilled they are) but don't know the outcomes of other players (such as the characteristics of interns who matched at other programs) or even what the NRMP knows about all of us.

In this provocative session, we will consider whether "success" in the Match can be predicted or understood through a Game Theory framework. Participants will be asked to consider what might occur if conditions change (e.g. if USMLE becomes pass/fail or if standardized video interviewing becomes mainstream, etc.). Are we in a "Nash Equilibrium" where no player has an incentive to deviate from a chosen strategy after considering an opponent's choice? Game Theory is about anticipating how others will act. Do we want this "Match game" to change? Should we?

Learning Objectives: Upon completion of this session, participants will be able to

1. Recognize elements of Game Theory in the National Resident Matching Program;
2. Describe "the prisoner's dilemma" and its implications for participants in "the Match";