Do Independent and Parallel Processing Physicians Influence Each Other's Performance? Evidence from the Emergency Department

Abstract

Problem Definition: Understanding potential channels through which physicians impact each other's performance can yield new insights into better management of hospitals' operations. For example, various studies of hospital operations model physicians as parallel servers (e.g., within a queueing system) and assume that they have independent speed and/or quality. In settings where physicians have to work on the same patients or tasks, it is clear that this assumption might be violated as the speed and/or quality of one physician often depends on that of the other. However, it is largely unknown whether and how this interdependency between physicians' performance might exist in settings where physicians do not work together.

Methodology/results: We make use of data on more than 110,000 visits made to our partner hospital's Emergency Department (ED) and investigate whether and how ED physicians who are known to serve their own patients without much interactions with each other affect each other's performance. We find strong empirical evidence that, despite working independently and in parallel, ED physicians (a) affect each other's both speed and quality, and (b) they do so through their shared utilization of resources such as test services or those required for admitting ED patients to the hospital. Thus, we establish evidence of an important resource spillover effect in EDs. Furthermore, we find that this resource spillover effect has an impact which is in the opposite direction to settings in which servers work together. Specifically, we find that a faster peer decreases a focal physician's average speed while a slower peer increases it. Similarly, a higher-quality peer decreases a focal physician's average quality while a lower-quality peer increases it. We further show that during high-volume shifts (i.e., when the shared resources are most constrained), the magnitude of these effects increases.

Managerial implications: Our results indicate that better pairing of ED physicians who work during the same shifts should be viewed as an important lever that can allow hospital administrators to improve their operations without additional investments.