Performance of risk-adjusted cumulative sum charts when some assumptions are not met

Abstract

Monitoring health care performance outcomes such as post-operative mortality rates has recently become more common, spurring new statistical methodologies designed for this purpose. One such methodology is the Risk-adjusted Cumulative Sum chart (RA-CUSUM) for monitoring binary outcomes such as mortality after cardiac surgery. When building RA-CUSUMs, independence and model correctness are assumed. We carry out a simulation study to examine the effect of violating these two assumptions on the chart's performance

Modeling the patient mix for risk-adjusted CUSUM charts

“…However, some probability distributions have been utilized. Hussein et al 6 assumed that the risk factor that characterizes the patient mix follows the Poisson distribution. Although this distribution is suitable for integer-based risk scores, its flexibility and fit to the data is often limited.…”

Flexible risk‐adjusted surveillance procedures for autocorrelated binary series

“…The first is the assumption that the binary observations are independent over time, the second that the historically estimated baseline parameters can be treated as known numbers. Concerning the independence assumption it has been shown in a simulation study that, if the monitored sequence of binary observations has some sort of serial dependence, then the theoretical average run lengths (ARL) of RACUSUM charts are greatly affected (Hussein et al, ). Similarly, errors incurred in estimating the baseline rates can affect negatively the performance of the RACUSUM chart (Jones & Steiner, ).…”