Using social network analysis to identify the most central services in an emergency department

In the past decade, emergency department (ED) overcrowding has become an internationally recognized problem that is associated with adverse patient care and safety. ED overcrowding is caused by multi-factorial system issues, many of which are outside the control of ED managers. In this paper we propose service-based approaches to improving ED throughput via social network analysis. We first construct a service network using Current Procedural Terminology codes. Then degree, betweenness, and eigenvector centrality measures are applied to rank each service. Combined with practical knowledge of ED operations, the rankings help us identify the most central services that are likely to help improve throughput, if administered at the point of care or resources are allocated to them appropriately.