The purpose of plotting data over time is to identify the variation in what is being measured over time. Understanding how changes you make to the system result in variation enables us to decipher when change is, and is not, an improvement.
There are two forms of variation we must understand in order to determine this:
- Common cause - variation that is expected and natural in the system. Whilst this does not indicate that the system is working well, it does mean that the system is stable. For example, the seasonal increase in the number of A&E visitors over the Christmas holiday period is a variation but it is expected and common.
- Special cause - variation that is unusual and not a consistent part of the system. Special cause variation arises due to unique circumstances and changes, indicating an unstable system. However, this instability is not necessarily a bad thing. For example, the number of pressure sores on a hospital ward may drop dramatically unexpectedly due to increased patient awareness. The aim of most quality improvement projects is to produce a positive special cause variation by introducing changes to a stable system through rapid fire Plan Do Study Act (PDSA) cycles.
Life QI will look for 5 types of special cause variation (SCV) on SPC charts
The five types of SCV used in Life QI are:
- Outlier - A single point above the +3 sigma or below -3 sigma limit.
- Shift - A run of eight points in a row above or below the centreline. Note, a point on the centreline neither adds nor resets the run but is ignored, i.e. if you have seven points above the centreline and the eighth on the centreline then the trigger is dependent on the ninth point, if it is above it makes the run of eight if it is below then it halts the run.
- Trend - Six consecutive points decreasing or increasing. Note, as with the shift rule a point equal to the last point does not add to nor resets the trend.
- High variation - Two out of three points in the outer 1 third of the chart, between the +2 and +3 sigma or between the -2 and -3 sigma limits. Note, a point on the +/-2 sigma limit does not count as in the outer third.
- Low variation - Fifteen consecutive points in the inner third of the chart, between the -1 and +1 sigma limits. Note, a point on the limit does not count.
How to identify variation on Life QI
Life QI automatically identifies and visualises any instances of SCV. All SCV are indicated by pink markers at the top of the chart. Hover your cursor over the marker to clearly show which data points are included in the variation.
You can also click on the marker to see a pop-out explaining exactly which type of SCV it is.