|
February 1997
Statistical Modeling to Predict Elective Surgery Time
Wright I, Koopersburg C, Bonar BA, Bashein G;
Anesthesiology 85:1235-45, 1996
[ see abstract below ]
Accompanying Editorial:
Applications of Information Systems to Operating Room Scheduling.
Dexter F, Macario A
Anesthesiology 85: 1232-4, 1996.
[ editorials do not have abstracts ]
Anesthesiologists have naturally moved into operating room management. Their organizational and computer skills are prized by corporate medicine as a push is made to make operating rooms operate more efficiently. Wright et al. tested a commercial software application (Surgiserver) that used a truncated mean to predict surgical time.
(The program's use of a truncated mean translates into: 10 recorded times for the same operation, throw out the high and the low, like they do in figure skating, and average the remaining 8 surgical times). There are 2 reasons that better prediction of surgical operating room time is important.
First, it will allow OR administrators to decrease under-utilization of OR's (those that end early, remain unused for some portion of the day, but which continue to incur the cost of personnel for the entire time the OR is scheduled). Second is to decrease over-utilization which has real costs in terms of overtime or patient delays. By better predicting surgical times, the theory is that exactly the right number and types of cases can be scheduled for each OR for each surgeon.
Unfortunately, the goal is easier stated than achieved. Surgeon's estimates of their operating times were more accurate than the computer program's estimation. That is, nothing was gained from using this computer analysis of the historical truncated mean operating room time. Errors in time estimation were large (34% of an average 157 minute case).
More detailed modeling using a combination of surgical estimates and historical means via the computer program improved accuracy a little more than 10%, not enough to really matter. Adding some general patient information to the modeling yielded no further improvement.
What does this mean? Surgeons are as good as anything else in predicting their OR times, with some naturally expected variation among surgeons in their accuracy. Combining a surgeon's gestalt for the operation they are about to perform with their historical mean performance enhances accuracy 11%.
Because of the low level of improvement, an investment in a state of the art OR scheduling computer system using a truncated mean is not likely to change our ability to efficiently schedule cases.
Return to the Current Literature Review Front Page, or read the abstract:
ABSTRACT
Background: Accurate estimation of operating times is a prerequisite for the efficient scheduling of the operating suite. The authors, in this study, sought to compare surgeons' time estimates for elective cases with those of commercial scheduling software, and to ascertain whether improvements could be made by regression modeling.
Methods: The study was conducted at the University of Washington Medical Center in three phases. Phase 1 retrospectively reviewed surgeons' time estimates and the scheduling system's estimates throughout 1 yr. In phase 2, data were collected prospectively from participating surgeons by means of a data entry form completed at the time of scheduling elective cases. Data included the procedure code, estimated operating time, estimated case difficulty, and potential factors that might affect the duration. In phase 3, identical data were collected from five selected surgeons by personal interview.
Results: In Phase 1, 26 of 43 surgeons provided significantly better estimates than did the scheduling system (P < 0.01), and no surgeon was significantly worse, although the absolute errors were large (34% of 157 min average case length). In phase 2, modeling improved the accuracy of the surgeons' estimates by 11.5%, compared with the scheduling system. In phase 3, applying the model from phase 2 improved the accuracy of the surgeons' estimates by 18.2%.
Conclusions: Surgeons provide more accurate time estimates than does the scheduling software as it is used in our institution. Regression modeling effects modest improvements in accuracy. Further improvements would be likely if the hospital information system could provide timely historical data and feedback to the surgeons.
|
