Understanding Run Charts for Quality Improvement
Uploaded by
mosatUnderstanding Run Charts for Quality Improvement
Uploaded by
mosat- Why?
- Examples of a Run Chart
- How?
- Interpreting a Run Chart
- Resources
- When should I use a run chart?
Common questions
Powered by AIThe purpose of drawing a centreline, typically the median, in run charts is to provide a reference line against which trends and runs can be identified. It helps visualize deviations and patterns, thereby assisting in the identification of shifts or trends that may indicate changes in performance or quality over time .
The rules for interpreting a run chart to detect non-random patterns include: 1) a run of 6 or more consecutive points above or below the centre line (a shift), 2) a trend of five or more consecutive points all going up or down, 3) too few or too many runs (determined by the number of times the line crosses the centre line plus one, compared to table limits), and 4) an astronomical data point, which is an obviously different value that is acknowledged as unusual .
Trends in run charts can be identified by a sequence of five or more consecutive data points consistently moving in one direction (up or down). These trends typically indicate a persistent change in the process being monitored, which may be due to identifiable causes that need addressing or taking advantage of, depending on the nature of the trend .
A 'run' in a run chart is defined as one or more consecutive data points on the same side of the median. It is calculated by counting the number of times the line connecting the data points crosses the centre line and then adding one, which helps determine if there are too few or too many runs in comparison to expected ranges .
Run charts assist in understanding variation by visually depicting the history and patterns of variation in an indicator or measure over time. They are used to examine data for trends or other patterns and help identify when shifts and changes occur, thus allowing one to see if and when problems appear .
Too few or too many runs in a run chart indicate a non-random pattern because they suggest deviations from what would be expected by chance alone. By comparing the observed number of runs to the limits provided in a standardized table, you can determine if the observed pattern suggests consistent biases or variations that are unlikely to occur randomly, thus indicating potential underlying issues in the process .
A project should transition from using a run chart to a Statistical Process Control (SPC) chart when it progresses to having at least 12 data points and requires a more detailed understanding of the process beyond simple trend analysis. SPC charts offer the benefit of upper and lower control limits, which allow for more robust statistical processes that can identify special cause variations, important for projects where understanding the intricacies of the process is critical .
A run chart is more suitable when you want to look at data over time in a simple way, have limited statistical expertise, or have less than 12 data points with no baseline data. It is ideal for starting projects where in-depth statistical analysis is not required. In contrast, SPC charts are suited for projects with baseline data (at least 12 data points) where more robust statistical processes are needed for identifying special cause variations .
Resources available for creating and learning about run charts include a template on the National Quality Improvement Team's website that allows you to enter raw data into an Excel sheet, which will then automatically create the chart. Additionally, a video explaining run charts is available on the HSE YouTube channel, and a 40-minute talk on run charts can be accessed on the QI Talktime page .
An astronomical data point is significant because it indicates an obviously different value that everyone agrees is highly unusual. Such points hint at special cause variation, which might necessitate further investigation to understand underlying causes and assess the process being analyzed for potential issues .
