Evaluating patient flow in the operating theater: An exploratory data analysis of length of stay components

Inefficient scheduling increases the length of stay in an operating theater and the risk of adverse health outcomes for surgical patients, and can cause resource allocation mismatch. We conducted this study to explore the time intervals associated with the operating theater of a general hospital. In a cross-sectional study, data from 622 patients undergoing elective surgery were collected. We measured entrance and exit times into and from the locations in the theater, and also the start and end times for each procedure including surgery and anesthesia. Then we used statistical, machine learning, and visualization techniques to explore the time interval data for seven surgical categories. In our data, women were older and experienced more surgery and recovery time than men. For ophthalmic and urology surgeries the time spent in the recovery room, and for neurosurgery and plastic surgery the duration of surgery, constituted the largest percentage of the stay. A random forest classifier recognized patients with potential short, medium, and long stay in the operating theater (accuracy = 0.91). Our data suggested that the staff habitually assign less priority to the cases with easier surgical procedures and short surgery and recovery time. This caused the formation of an unwritten schedule which interfered with the official one. The differences in surgery types and recovery time can be used to prevent stagnated patient flow in the daily scheduling of an operating theater. Accurate classifiers which include patient characteristics and surgical variables enable staff to recognize patients with a potential long stay in the operating theater.