Additional Uses Include:
- Improvement of a process so that it will perform consistently and predictably.
- To achieve higher quality, lower cost and higher effectiveness.
- Set a common language for discussing process performance.
Materials Needed:
- Data points (at least 20-25 data points).
- Graph paper and pencils/pens, or Microsoft Excel/equivilant software that can display data.
How To Use a Control Chart:
Make Some Decisions.
Begin Your Graph.
Y Axis:
X Axis:
Calculate.
UCL, LCL, and the Standard Deviation
Example: Patient Cycle Time (+/- 3 standard deviations)
One standard deviation = 23. Three standard deviations place our UCL at 143.6 and the LCL at 5.6.
When any of the following occur, it is cause for deeper investigation.
- Single point outside of the control limits
- A shift of eight or more consecutive points above or below the centerline
- A trend of at least six consecutive points up or down
- Two out of three consecutive points near a control limit
- At least fifteen consecutive points hugging the centerline
In our example, there are six consecutive points going up after patient 8. This is a reason to do some deeper investigation. Also, consider that it is important to read the chart with customer expectations in mind.
The team may determine that 3 standard deviations are too wide of a parameter to meet customer expectations. You may decide to use 2 standard deviations rather than 3.
Example: Patient Cycle Time (+/- 2 standard deviations)
While none of the data points exceed the UCL or LCL, the data is closer to these limits than in the previous example.
Improving the Control Chart Process:
Inspect Your Data
Firstly, study the data to find patterns and to determine the causes. Ask yourself: is the data within the usual range? Is there a daily, weekly, monthly or yearly pattern? Look at variables that could be affecting your data (types of appointments, number of staff present, walk-ins and no-shows, technology issues, etc). From there, determine areas that need further investigation and/or correction.
Process Improvements
Next, utilize a Process Improvement methodology such as Plan-Do-Study-Act (PDSA) to make improvements. After a correction in the process is made, note that on your Control Chart and continue to collect data to see if an improvement has taken place.
Understand Variation
Apply the rules for detecting special causes. Common-cause variation is where no one or combination of factors has unduly affected the process variation (random variation). Special-cause variation is when one or more factors are affecting the process variation in a non-random way.
Re-plot if Applicable!
If there was a signal, re-plot the chart with a new Mean, UCL and LCL. For example, look at the following chart: The new Mean is 70.7. The standard deviation is now 19.7. The two standard deviations from the mean give a UCL of 110.1 and an LCL of 31.2.
Continue to collect data and monitor the process until the goal is consistently achieved over a period of time.
Key Takeaway
A control chart is a powerful visual tool used to monitor process performance over time, and to set a common language and place to start in discussions with your team.
By tracking data, control charting will support continuous quality improvement, better decision-making, and more consistent outcomes.
