A Comparison of Detection Probability of CUSUM Procedures under Increasing Population Sizes

Detecting increases of incidence rate of an adverse health event is critical to health surveillance. Motivated by successful applications in industry quality control and productivity improvement, various cumulative sum (CUSUM) procedures have been developed to monitor rate changes of health events when population size changes over time. Instead of using average run length (ARL) as the performance criteria, this paper provides a thorough statistical comparison of CUSUM-type procedures in the health care context based on probability of alarm (PA) evaluation metrics. We show that different methods may favor different parameter settings depending on the corresponding designs.