We all need data for Quality Improvement, but do you know how to use it properly? The run chart is arguably a QI team’s most valuable tool, a simple yet powerful way to tell your story at a glance — but they are only useful if designed and analyzed correctly. As QI Coaches, we have found teams have difficulty putting data into run charts and analyzing data for random and nonrandom variation to promote learning and improvement. This session will quickly break down the process and provide helpful tips and tricks for correctly building and using effective run charts.
Tool: Run Charts are analytical tools to display and analyze improvement data. These graphical displays include data plotted in order of time and a time series analysis. The median is the centerline and is required to analyze data using probability-based rules. Goal lines and annotations are typically included on run charts
Problem: Data is necessary to guide learning and improvement. Data displayed in run charts provide information and tell us if we are improving. When run charts are constructed or analyzed incorrectly, there is a failure to learn and improve. Run charts can be used for units, departments, and organizations
Tool Selection: Run Charts are one of the tools recommended for graphic display and analysis of data in the Model for Improvement. Run charts can be developed in Microsoft Excel and do not require additional software. The analysis allows for understanding random and nonrandom variation to drive improvement work
Usage: Run charts demonstrate improvement and are used for team improvement efforts as well as communication tools to share improvement work with organizational leaders. We use run charts in all of our improvement projects starting with the first data collection. The first few data points are essentially the baseline data
Results: In 20 years of coaching and teaching, we have observed that many teams still don’t use run charts, don’t understand them, don’t use them correctly, or may not have even seen them at all. Some teams use trend lines erroneously in an attempt to ‘explain away’ the data