Concept: Statistical Analysis of Mean Waiting Times

Estimating Waiting Time for Elective Surgery

Administrative data contain a history of contacts in the period before surgery. The preoperative visit to the operating surgeon occurs in a high proportion of cases and makes sense to most clinicians as being the beginning of the wait for surgery. Using the preoperative visit as a marker to monitor waiting times is subject to several limitations including admission status, consideration of multiple procedures, and multiple preoperative visits.

Mean analysis is powerful and statistically sensitive. The mean waiting time can be adjusted for other factors in making comparisons. Mean wait time can be used to assess waiting time beyond the median value and to assess variations in the observed data. A major drawback of using the mean is that it is sensitive to outliers. It is necessary to separate unusual waiting times based on Tukey's robust outlier detection method.

Tukey Method

Under this method, the interquartile range (IQR) is defined as the difference between the 25th and 75th percentile. An outlier is defined as being a value which is greater than [3*IQR] + 75th percentile or less than [3*IQR] - 25th percentile. These upper and lower bounds are created using the conservative [3*IQR] (95% confidence level) to assure robust comparisons of the mean waiting times, uninfluenced by extreme values.

Comparisons

Compare means by formal hypothesis testing at 5% level of significance for differences in mean waiting times based on year, gender, region, and socioeconomic status.