Hospital Load Balancing : A Data-Driven Approach to Optimize Ambulance Transports During the COVID-19 Pandemic in New York City

Most emergency medical services (EMS) systems send patients to the closest hospitals. This long-standing principle, while maximizing patient welfare in usual times, can lead to hospital overloading and negative patient outcomes during unusual EMS demand shifts caused by events such as the COVID-19 pandemic. This latter phenomenon was observed in New York City (NYC) during the spring of 2020 and overwhelmed many outer-borough hospitals. To prepare for a future potential incident surge, and to complement existing approaches to relieve congestion in emergency departments, Columbia University and the Fire Department of the City of New York collaborated to design a load-balancing optimization algorithm to transport 9-1-1 patients to hospitals. The algorithm automatically balances patient transport time and hospital capacity city-wide, by integrating daily-run optimization with data-driven prediction to proactively avoid hospital overload in advance of dispatch. We present the algorithm, explain how it enhances the existing system to guard against future unexpected incident surges, and discuss a range of implementation considerations. The load balancing rule was implemented in NYC's Computer Aided Dispatch system on January 13, 2021 through May 20, 2021 and again on December 31, 2021 through January 27, 2022