The Use of Statistical Process Control for Monitoring Institutional Performance in Trauma Care
- 1 December 2008
- journal article
- research article
- Published by Wolters Kluwer Health in Journal of Trauma: Injury, Infection & Critical Care
- Vol. 65 (6) , 1494-1501
Abstract
Background: In recent years, performance monitoring has gained increasing attention as a tool for evaluating the delivery of health care services and is a topic of increasing importance in trauma systems. The main objective of this article is to illustrate a proactive method for assessing the performance of trauma centers in England and Wales, while taking into account common causes of variation. The aim is to present a methodology that is easily interpretable and avoids the spurious ranking of hospitals which can often lead to the misinterpretation on what is perceived to be the best and worst performing hospital, as measured by a prespecified performance indicator. Methods: The Ws statistic was introduced over 10 years ago to quantify the performance of trauma care systems through definitive outcome based evaluation (DEF) methods. Little advancement on this methodology has been made since its introduction. In this article, we highlight some of the limitations and problems associated with these DEF methods and introduce the funnel plot, a form of control chart to monitor hospital performance. Results: The number of patients included in a Ws statistical analysis can seriously change the ranking of a hospital. These complex issues with ranking means that league tables (or standings charts), which form part of the DEF method are an unsatisfactory method to represent performance indicators. The funnel plot methodology is an alternative graphical method for monitoring hospital performance, which has no emphasis on ranking. We demonstrate the method using mortality data and length of stay as the performance indicators. Conclusion: The funnel plot is a flexible, attractively simple method for comparing hospital performance and avoids spurious ranking of hospitals in league tables. The method can be applied to any number of performance indicators and can help formulate hypotheses about individual hospital characteristics likely to improve performance.Keywords
This publication has 18 references indexed in Scilit:
- An Analysis of Hospital Trauma Care Performance Evaluation2007
- Pay-for-Performance Programs in Family Practices in the United KingdomNew England Journal of Medicine, 2006
- Funnel plots for comparing institutional performanceStatistics in Medicine, 2004
- ‘Online’ monitoring and retrospective analysis of hospital outcomes based on a scan statisticStatistics in Medicine, 2003
- Use of risk-adjusted CUSUM and RSPRTcharts for monitoring in medical contextsStatistical Methods in Medical Research, 2003
- The application of statistical process control charts to the detection and monitoring of hospital‐acquired infectionsJournal of Quality In Clinical Practice, 2001
- Funnel plots for detecting bias in meta-analysisJournal of Clinical Epidemiology, 2001
- Bristol, Shipman, and clinical governance: Shewhart's forgotten lessonsThe Lancet, 2001
- League Tables and Their Limitations: Statistical Issues in Comparisons of Institutional PerformanceJournal of the Royal Statistical Society Series A: Statistics in Society, 1996
- Evaluating Trauma Care1987