Prediction of the time to complete a series of surgical cases to avoid cardiac operating room overutilization

Operating room (OR) efficiency could be improved by reducing overutilization. In this article we suggest a methodology to accurately determine the time to complete a series of surgical cases in a single cardiac OR to avoid overutilization. Our methodology includes the basic assumptions that the first case starts on time, there are no add-on cases, and there is a predetermined OR time allocation.

We studied 6,090 cases performed by the department of cardiovascular surgery service at St. Michael's Hospital. Lognormal distributions were fitted to surgical times and turnover times. The distribution of the duration of the schedule was estimated using the Fenton-Wilkinson approximation. A simple model utilizing these distributions was then applied to three months of data to determine if overutilization could be predicted using the model.

The mean difference between the actual schedule duration and the average duration was 0.19 hr (11.64 min). The difference with the second tertile cut-off point was 0.59 hr (35.40 min). Schedules that overran were correctly predicted by the average duration in 86.49% of the cases (with 12 false predictions), and they were correctly predicted by the second tertile cut-off point in 94.59% of the cases (with 26 false predictions).

These results suggest that the sum of the average duration of surgeries and turnover times is indeed a good estimator of the duration of the series of surgical cases. Neither the estimated averages nor the second tertile cut-off points were useful to predict overrun when used alone. The use of the estimated average duration of the schedule, validated by use of the second tertile cut-off point, could reduce overtime when allocating surgeries in a single cardiac OR.