Application of statistical process control as a tool in research and quality improvement in urology

Back to 77th Annual Meeting
Back to Program Outline

Background: In order to improve the quality and performance of healthcare delivery we must change our way of working. Change does not always mean improvement and in order to determine if change has yielded improvement relevant aspects of performance need to be measured. Traditional statistical analysis methods account for natural variation but require aggregation of measurements over time, which can delay decision making. Statistical process control (SPC) is a branch of statistics developed by Dr Walter Shewhart that combines rigorous time series analysis methods with graphical presentation of data, yielding insights into data more quickly. It is a set of methods for ongoing improvement of systems, processes and outcomes. We provide examples in the use of statistical process control for quality improvement in urology.
Material and Methods: Over the past 2 years we have initiated quality improvement initiatives within the section of pediatric surgery and pediatric urology in order to improve patient access and patient satisfaction. These include a Microsystems analysis of the program, evaluating deficiencies and initiating change to improve the system. In order to measure improvement control charts were used. The charts we used most are XmR and p charts. We use g charts to evaluate adverse events such as postoperative complications and also use ANOM (Analysis of Means) charts to compare physician performance.
Results: Figure 1 shows an XmR control chart that shows the patient satisfaction scores with making the appointment for outpatient services in pediatric urology. This shows that since 2006 the satisfaction scores have been in statistical control with normal cause variation but the scores were low with a mean of 77.5%. In January 2008 we implemented a change in the scheduling process which was more patient friendly. Within 2 months we were able to demonstrate an improvement to a mean score of 93%. The chart was split at that point indicating the new process and new satisfaction score for future monitoring. Similar control charts such as p-chart were used for other improvement projects such as outcomes of endoscopic treatment of vesicoureteral reflux with Deflux. We also used analysis of means chart to evaluate physician performance and evaluated variation in practice and identify areas of improvement.
Discussion: Control charts have provided quick insight and visible data that showed improvement over time. The endoscopic deflux data showed that our success rate has been stable and in statistical control with common cause variation.
Statistical process control is a versatile tool which can help diverse urology practices to manage change in healthcare and improve patients’ care quality, improve satisfaction, health and outcome.

Back to 77th Annual Meeting
Back to Program Outline