Improving prediction of surgery duration using operational and temporal factors

Inherent uncertainties in surgery durations impact many critical metrics about the performance of an operating room (OR) environment. OR schedules that are robust to natural variability in surgery durations require surgery duration estimates that are unbiased, with high accuracy, and with few cases with large absolute errors. Earlier studies have shown that factors such as patient severity, personnel, and procedure type greatly affect the accuracy of such estimations. In this paper we investigate whether operational and temporal factors can be used to improve these estimates further. We present an adjustment method based on a combination of these operational and temporal factors. We validate our method with two years of detailed operational data from an electronic medical record. We conclude that while improving estimates of surgery durations is possible, the inherent variability in such estimates remains high, necessitating caution in their use when optimizing OR schedules.