Quality Improvement Zone

A run chart is a simple graph that helps teams understand how a process is performing, identify patterns and variation, and determine whether changes are leading to improvement.

Run charts are one of the most widely used measurement tools in quality improvement because they allow teams to learn from data over time rather than relying on single data points, averages or snapshots of performance.

A run chart typically includes:

• a measure plotted on the vertical (y) axis
• time or observation number plotted on the horizontal (x) axis
• data points connected by a line to show performance over time
• a median, which acts as a reference point for interpreting patterns in the data
• annotations showing when changes were introduced, or notable events occurred

How to use it

Run charts help teams move beyond simply collecting data to actively learning from it. The chart becomes most useful when data is collected consistently, changes are annotated, and the run chart rules are applied to identify signs of non-random variation.

1. Define the measure

Choose a measure that will help you understand whether your system is improving. This may be an outcome, process or balancing measure. Ensure the measure has a clear operational definition, so it is collected consistently over time.

2. Collect data over time

Collect data at regular intervals and plot each data point in chronological order. Consistent data collection helps create a more reliable picture of performance and variation.

3. Plot the data

Create a line graph showing the measure on the vertical axis and time or observation number on the horizontal axis. Connecting the points makes it easier to see patterns over time.

4. Calculate and add the median

Once around 10 data points have been collected, calculate the median and add it to the chart. The median provides a reference point against which future performance can be compared.

For an initial run chart, especially where historical or baseline data are available, the median is usually calculated using all baseline data points.

5. Annotate changes

Mark when tests of change, interventions or significant events occur. Annotations help connect what happened in the system with what is observed in the data.

6. Apply the run chart rules

Once you have sufficient data, use the run chart rules to identify signals of non-random variation. When one or more of these rules is present, it suggests that something may have changed within the system.

The four run chart rules are:

• a trend: five or more consecutive points moving in the same direction
• a shift: six or more consecutive points on one side of the median
• too many or too few runs compared with what would be expected by chance
• an astronomical data point that stands out clearly from the rest of the data

More detailed guidance on these rules is available in the resources section below.

7. Use the findings to guide improvement

Use the information from the run chart to understand variation, assess the impact of changes and decide what actions to take next. The chart should support learning and decision-making rather than simply reporting performance.

Baseline medians

When carrying out improvement work, it can be helpful to calculate a baseline median and extend it across future data points. This provides a stable reference point against which future performance can be compared.

The example below shows a baseline median calculated from the first 10 data points and extended across the chart. A clear shift can be seen when subsequent data points consistently perform above the baseline median.

Creating a run chart

Run charts can be created in Excel or by using a run chart template. More information on how to produce a run chart in Excel can be downloaded below.