RISK-ADJUSTED CUMULATIVE SUM CONTROL CHARTING PROCEDURES AND STANDARDIZED MORTALITY RATIOS

Abstract

Measuring quality of medical practice is becoming increasingly prominent in quality management because it is a key component in improving quality and efficiency in health care. In this thesis, we develop such a chart and study its performance. The advantages of using more than two outcomes are demonstrated with real data sets. Our chart is shown to be a valuable tool for analyzing and comparing the performances of surgeons.

The Standardized Mortality Ratio (SMR) is now widely used for assessing the performance of hospitals. In this thesis, the asymptotic distribution of the SMR based on an estimated number of deaths are derived. We also use the bootstrap procedure to estimate the finite-sample distribution. In addition, two procedures for comparing surgeons by SMRs are developed. Empirical evidence shows that these procedures are useful and effective for assessing the performance of surgeons.