Related Experiment Video
Updated: May 30, 2026

An R-Based Landscape Validation of a Competing Risk Model
Published on: September 16, 2022
Monitoring binary outcomes using risk-adjusted charts: a comparative study
Edit Gombay1, Abdulkadir A Hussein, Stefan H Steiner
1Department of Mathematical and Statistical Sciences, University of Alberta, Edmonton, AB, Canada.
New sequential charts offer improved control over false alarms in healthcare performance monitoring, specifically for 30-day cardiac surgery mortality rates. These methods address limitations of the risk-adjusted cumulative sum (RA-CUSUM) chart.
Area of Science:
- Biostatistics
- Health Services Research
- Medical Quality Improvement
Background:
- Monitoring binary outcomes is crucial for healthcare performance evaluation.
- Cumulative sum (CUSUM) charts, including the risk-adjusted CUSUM (RA-CUSUM), are used for this purpose, particularly for 30-day mortality after cardiac surgery.
- The RA-CUSUM chart, while effective at early change detection, suffers from a type I error probability of 1 and a high rate of false alarms.
Purpose of the Study:
- To propose and evaluate alternative sequential curtailed and risk-adjusted charts.
- To control the type I error rate in monitoring 30-day mortality following cardiac surgery.
- To compare the performance of these new methods against the RA-CUSUM chart.
Main Methods:
- Development of novel sequential curtailed and risk-adjusted charts.
- Evaluation of methodologies based on average run lengths and type I error probabilities.
- Application of the methods to monitor the performance of seven cardiac surgeons.
Main Results:
- The proposed charts demonstrate control over the type I error rate, unlike the RA-CUSUM chart.
- Analysis of average run lengths and type I error probabilities highlights the merits of the new methodologies.
- The study illustrates the practical application of these charts using real-world surgical performance data.
Conclusions:
- The developed sequential charts offer a superior alternative to the RA-CUSUM chart when controlling false alarms is critical.
- These new methods provide a more reliable approach to monitoring healthcare performance, especially in high-stakes scenarios like cardiac surgery outcomes.
- The findings support the adoption of these improved statistical tools for enhancing patient safety and quality of care.
Related Concept Videos
Types of Biopharmaceutical Studies: Controlled and Non-Controlled Approaches
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast, controlled...
Interpreting Run Charts
Relative Risk
Comparing the Survival Analysis of Two or More Groups
The R Chart
R charts are pivotal for pinpointing shifts in process variability. Stability is indicated when all data points remain within the defined upper and lower...
Run Charts

