Phase I Study of Surgical Performances with Risk-Adjusted Shewhart Control Charts

Abstract Monitoring a surgical process has gained prominence in recent years. A number of risk-adjusted charting procedures have been developed for this purpose. For the control limits of charting procedures in a phase II monitoring to be accurate, they must be based on data taken from a process that is in control. We propose using a risk-adjusted Shewhart chart for a phase I study because of its ability to detect extreme observations from out of control processes. The statistic used is shown to be a likelihood ratio test statistic. A real data set is used to illustrate the procedure.