Accommodation of requests for emergency US and CT: applications of queueing theory to scheduling of urgent studies.

Abstract

The purpose of this study was to determine whether queueing theory would allow prediction of optimal number of schedule slots to be reserved for urgent computed tomography (CT) and ultrasonography (US). Institutional review board approval was obtained; informed consent was exempted. Emergency studies were modeled as a Poisson process; slots were reserved such that rate of rescheduling of routine studies to accommodate emergencies was predicted to be below a certain level. Model was tested with 3 years of emergency US and CT requests. US and CT requests showed Poisson distribution. US rescheduling was near that predicted. CT rescheduling exceeded that predicted, which reflected increasing CT use. By using more recent CT data for prediction, a more concordant rescheduling rate resulted.