The Elements of a Run Chart

in Health Care, Measurement, Tools & Methods, Variation, Videos

By David M. Williams, PhD

Harvardx course Practical Improvement Science in Health Care with the Institute for Healthcare Improvement.

A run chart includes several elements. There’s a y-axis, that’s the one that goes up and down on the left hand side, and this shows what you’re actually measuring. This is what your data is about.

Along the bottom is the x-axis, and this shows the time series in which you’re tracking your data. Are you talking about days, weeks, months, sometimes even years.

And I mentioned, in the middle we have a line that’s the median. This is the middle of the data. And it’s easily done by a simple formula, but it tells you where the middle the data is.

In addition, we want to be really, really clear when we’re thinking about our display to label everything. Labeling the y and the x-axis, putting in a very, very clear title.

All of these things help when people are looking at the data to not have to interpret it. They can see it and they can spend their time learning about the data, but not having to try to figure out what it means.

And then finally, one thing that people often worry about when we’re talking about data and talking about run charts, is that the qualitative aspects are going to be lost in the quantitative data that we’re looking at. And annotations help us to tell the story, to actually apply little bits of information that help us talk about the changes that we made or things that were going on the environment, and give us ideas about what’s happening as we’re looking at the behavior of the process.

When we get to 20 data points we can move to even more sophisticated tool, something called the Shewhart statistical process control charts, or the SPC chart. This transforms from a median to a mean and then adds two other lines, an upper control limit and a lower control limit, that help us to apply some additional rules that are more sophisticated and enable us to identify when there are signals of change, or things that are special cause versus common cause variation in our process.

So, of course I introduced you to the sort of progression from a line chart to a run chart to a Shewhart statistical process control chart. The reality is for most of us when we first get started we’re not have enough data yet to use these more sophisticated tools. We’re going to be starting with the basic line chart.

If this was helpful, share and include me @DaveWilliamsATX. Sign up here to receive a monthly email from me that includes all my blog posts and other Improvement Science resources I think you’d appreciate.