The Cost of Task Switching: Evidence from Emergency Departments

Abstract

Emergency department (ED) physicians treat patients with different symptoms and constantly switch between tasks. Using a comprehensive data set with over 650,000 patient visits to four EDs, we investigate the impact of task switching on physician productivity, quality of care, patient routing, and patient waiting time. To address estimation bias due to measurement errors and endogenous patient routing, we construct an instrumental variable that exploits the exogenous composition of waiting patients. Our estimates indicate that, at different EDs, switching between different types of patients increases the average pick-to-pick time by 3.4 to 16 percent or 0.8 to 3.1 minutes per patient, and reduces patient throughput rates accordingly. Task switching also affects how physicians route patients, although we find little impact on healthcare quality. Our counterfactual analysis further shows that eliminating the switch cost can reduce the average waiting time per patient by 25.3 to 48.3 percent and the average waiting census by 21.6 to 43.1 percent. To mitigate the switch cost, we suggest ED layout designs to facilitate patient sorting and communication between healthcare workers. Accounting for the switch cost in patient routing will also help.