Magna vulputate netus ultricies pharetra duis sit vitae consectetur.
Contact Support
layer 1 layer 2 layer 3 layer 4 layer 5 abstract shapes

How to use Statistical Process Control charts to Improve Healthcare

Posted by Alfie McDonald on August 18th 2022

Statistical Process Control (SPC) gives healthcare organisations access to complete data sets to visualise the performance of processes. In this article, we dive into what SPC charts are, how they work, and what benefits SPC chart software can bring to healthcare organisations.

Methods of quality control in healthcare

Healthcare systems are complex networks of care processes and pathways. The quality of care delivered in a healthcare system is largely dependent on the how well these complex networks function, and how the people in the networks work together.

There are numerous ways in which healthcare organisations can implement quality improvements to improve the quality of care delivered, whether that’s through tradition trial-and-error processes, or more contemporary methods of data visualisation.

In a trial-and-error process, users would implement a change to combat a difficulty, barrier, or problem, and then would have to wait to see what impact that change had on the process over time. While the trial-and-error process can be very effective, it is too reliant on individual judgements, is highly repetitive, and can be extremely time-consuming.

In today’s digital age we have far more advanced tools at our disposal; tools that allow us to visualise and report on the performance of processes in far more effective and reliable ways.

Rigorous methodologies with strict protocols and rigid data requirements are common occurrences within the Healthcare Industry. However, understanding whether a change has made a positive impact to a process over time can be a difficult aspect to measure.

This is where Statistical Process Control (SPC) comes in.

What is a SPC chart?

SPC charts are quality control charts which provide a visual representation of the performance of a process over time, enabling professionals to determine whether a change to a process has made a positive impact, or whether a process is capable of meeting a target.

SPC is an effective improvement tool as it enables users to easily see whether an implemented change has resulted in improvements. By employing SPC software rules to data, healthcare professionals are able to tailor SPC charts to the specific process they are measuring.

How do SPC charts work?

 There are broadly two kinds of SPC charts used to identify variation, the ‘Run’ and ‘Control’ chart.

Run chart examples in healthcare:

Run charts are a simple yet powerful chart type that feature a central line (CL) of data plotted over time. In order to identify whether a process is stable or unstable, special rules are applied within the chart’s software.

For effective data visualisation, run charts should be used only when a user has a minimum of 10 data points. The y-axis of the chart shows the property under investigation (i.e. % or count), which is plotted over time on the x-axis. The centre line of run charts is generally drawn at the median.

By continually adding new data points when a process is in motion, the run chart enables users to determine whether the central tendency of a process is changing.

An example of a scenario where you could use a run chart to evaluate process performance can be seen – the chart demonstrates how run charts could be used to measure the impact of a smoothing technique (change to reduce variation) made within a hospital to reduce the daily average cycle time for patients.


Control chart examples in healthcare:

Control charts or ‘Shewhart charts’ are considered to be the more comprehensive variant of the run chart. The control chart features a central line drawn at the mean of measurements, as well as features an Upper Control Limit (UCL) and Lower Control Limit (LCL) enabling users to differentiate between common and special cause of variation. Additionally, control charts have their own set of rules for identifying special causes which are different from the rules applied to run charts.

To visualise meaningful patterns, control charts are used when users have a minimum of 25 data points.

Figure 2 is illustrative of a control chart used to measure the impact of a smoothing technique to reduce the amount of infectious waste produced in a hospital.

Special cause variation:

Special Cause Variations are causes that are not part of the process and do not affect everyone working within the process but arise due to specific circumstances. This kind of variation is known to be ‘extrinsic’ to a process.

Variation of this kind results in an unstable process because the variation is not predictable, and is due to assignable causes of variation. Special cause variation can be positive as well as negative, however, in both cases it shows that a process is unstable, and that the variation is unpredictable. 

If a process shows favorable special cause variation, then we would want to take actions that make it a permanent part of future processes.

For example, if a hospital is measuring x-ray scans and the retake rate of scans, and notices favorable special cause variation due to a technician scanning differently to the established method, resulting in lower retake rates, the hospital would want to understand the technician’s approach and adopt it as the new standard method for future x-ray scans. Obviously, the same applies to unfavorable special cause variation whereby steps would need to be taken to prevent it going forward.

Common cause variation:

This kind of variation is inherent to the design of a process and is therefore described as being ‘intrinsic’. Variation of this kind results in a stable process because it is predictable.

Unlike special cause variation, common cause variation affects everyone working in the system and impacts all outcomes of the process. If a process only has common causes affecting its outcomes, then it is defined as a stable process and one that is in statistical control. The process is stable because the processes variation is predictable and falls within the established limits, however this doesn’t necessarily mean that the process is performing optimally, only that it is behaving predictably.

Examples of Common Cause Variation include the impacts of your process components, including poorly defined standard procedures, poor working conditions, or measurement errors.

If a process’s outcomes are affected by both common and special causes it would be defined as being an unstable process. Being unstable however, doesn’t necessarily mean that there is large variation, it just means that the variation is unpredictable.

Using Statistical Process Control to improve the quality of healthcare

There are four major reasons why you would want to use SPC charts in a health-centric environment.

  • Firstly, a healthcare professional can use either a Run or Control chart to visual and therefore understand data over time, fundamentally understanding a process’s performance.
  • Secondly, SPC charts are used for quality control. Professionals can use SPC chart software to differentiate between common and special cause variation. For measuring quality control, control charts are commonly used due to the addition of the Upper Control Limit and Lower Control Limit lines.
  • Thirdly, SPC charts are used to determine whether a process is stable (and therefore predictable) or unstable (and therefore unpredictable), enabling professionals to focus on making effective quality improvements.
  • Finally, SPC chart software is used to predict expected outcomes such as when a user wants to understand whether a process is capable of meeting the desired level of performance.

Statistical Process Control examples in healthcare

To illustrate how the healthcare industry may use control charts, we can use examples of individual measurements. For example, healthcare professionals can use SPC charts to measure the system for administered medicine.

To provide effective medical care, a hospital has to administer medicine on a timely basis (e.g., within 2 hours of the prescribed time). The hospital can create a control chart measuring the medicine administration system, where a professional submits each time they administer medication.

Obviously, we wouldn’t expect to see perfect control due to common cause variation e.g. patients refusing medication or patients being unavailable at the time of prescription. However, the use of SPC charts will help professionals understand if the process of medicine administration is stable or unstable, and therefore in need of quality improvement.

How can BCN Group support you?


BCN Group has got the experience, expertise, and services in SPC which few other tech companies based in the UK have. Being part of a select few digital transformation companies with SPC chart software, we can accommodate any organisations need to visualise process performance and quality improvement.

Cloud2, a BCN Group company is at the forefront of the provision of SPC charts in Power BI. As proud suppliers of SPC charts to numerous NHS trusts across the UK, and with a growing base of prospective international clients, we have hands-on experience in your industry.

Whether you’re a hospital, a trust, a pharmacy or operate a care service, BCN Group have got the services to cater to your needs. Our consultative approach means that whatever organisation needs, we can tailor approach to you.

Find out more information about Statistical Process Control for healthcare, or get in touch with our experts today by clicking the link below.

Magna vulputate netus ultricies pharetra duis sit vitae consectetur.
Download Whitepaper
Magna vulputate netus ultricies pharetra duis sit vitae consectetur.
Contact Support