EXPRESS: What Drives Physician Efficiency? Evidence from Emergency Department Operations

Emergency department (ED) physicians play a critical role in healthcare, serving on the frontline of acute care delivery. Yet increasing demand, resource constraints, and growing patient complexity pose significant challenges to physician efficiency. In this study, we investigate the determinants of ED physician efficiency using a multi-stage framework. First, we apply Data Envelopment Analysis (DEA) to measure individual physician efficiency based on data from 110,325 patient visits at a leading U.S. hospital. Second, we use a Tobit regression model to analyze how physician experience, shift familiarity, and patient severity influence physician efficiency. Third, we conduct comprehensive robustness checks, including omitted variable bias sensitivity analyses, permutation tests, and machine learning-based classification analyses, to address concerns regarding endogeneity and potential model misspecification. Our results show that physician efficiency increases with experience at a diminishing rate. Efficiency also varies with shift familiarity, with the estimated association weakening as physician experience increases. In addition, higher patient severity is associated with lower physician efficiency, consistent with the greater resource demands of caring for more clinically complex patients. This negative association is attenuated for more experienced physicians, suggesting that experienced physicians are better able to maintain efficiency when managing more demanding clinical conditions. Our findings offer actionable recommendations for improving ED operations. Staffing strategies that assign relatively more experienced physicians to higher-severity patients may help maintain physician efficiency, while shift familiarity may inform a scheduling strategy that emphasizes greater scheduling consistency for relatively less experienced physicians and greater flexibility for relatively more experienced physicians. Simulation results further illustrate the potential reductions in physician resource use associated with these strategies. Together, these findings highlight the value of evidence-based physician staffing and scheduling to improve physician efficiency in EDs.