Juan M. Contreras, Beomsoo Kim, and Ignez M. Tristao
In this paper, we use a longitudinal census of laser in situ keratomileusis (LASIK) eye surgeries collected directly from patient charts to examine the learning-by-doing hypothesis in medicine. LASIK surgery has precise measures of presurgical condition and postsurgical outcomes. Unlike other types of surgery, the impact of unobservable underlying patient conditions on outcomes is minimal. Individual learning by doing is identified through observations of surgical outcomes over time, based on the cumulative number of surgeries performed. Collective learning is identified separately, through changes in a group adjustment rule determined jointly by all the surgeons in a structured internal review process. Our unique data set overcomes some of the measurement problems in patient outcomes encountered in other studies and improves the possibility of identifying and separating the impact of learning by doing from other effects. We cannot conclude that the outcome of LASIK surgery improves as an individual surgeon’s experience increases, but we find strong evidence that experience accumulated by surgeons as a group in a clinic significantly improves outcomes.