How to implement Statistical Process Control

In this article we’re going to take a look at how you might implement Statistical Process Control (SPC), and how you can use this methodology within your Quality Improvement (QI) projects. The Institute for Healthcare Improvement (IHI) describes Statistical Process Control as: ‘a branch of statistics that combines rigorous time series analysis methods with graphical presentation of data.’ It is well known as a tool that provides an easy way for people to track the impact of improvement projects, so let’s find out more on how to implement Statistical Process Control!

What is Statistical Process Control?

Let’s have a quick recap on SPC and how you might want to use it in your QI project. Described as ‘the scientific, analytical method used in industries such as healthcare and manufacturing to record data and monitor a process over time,’ the NHS describes Statistical Process Control as: ‘an analytical technique that plots data over time. It helps us understand variation and in so doing guides us to take the most appropriate action’.

SPC is widely used in the NHS – who started using the method in the 2000s - as a way to understand whether change results in improvement. If you are able to accurately record and evaluate your QI project data, it will help you make key decision and benefit you and your team.

When to use Statistical Process Control?

In the report by NHS England and NHS Improvement ‘Statistical process control’, the authors recommends that ‘SPC should be used throughout the life cycle of the [QI] project to help you identify a project, get a baseline and evaluate how you are currently operating. SPC will also help you to assess whether your project has made a sustainable difference.’

We recommend that you start to monitor your QI processes using SPC before you implement any change. This will then allow you to measure the success of that change and confirm whether or not any significant improvements have started to surface.

How to use Statistical Process Control?

‘Deploying Statistical Process Control is a process in itself, requiring organizational commitment across functional boundaries’. Let’s find out more about the different stages which will help you learn how to implement SPC and work on SPC or control charts.

As SPC charts identify the difference between normal and unusual data variation, and – in addition - whether that variation is positive or negative, it means that using an SPC chart can help you quickly work out if a change has an impact and if this impact is significant or not.

We’ll drill down and take a look at the various steps you need to take in order to implement Statistical Process Control and in order to make the most of the methodology in your QI project.

Decide on your Method of Measurement

Statistical Process Control methodology is based on data analysis. So, before you start the SPC process, you need to decide exactly what data you are going to collect. The two types of data used in an SPC chart are called ‘Variable data’ or ‘Attribute data’.

‘Variable data is essentially measurement data which you can measure along a continuous scale, such as: temperature, time, distance, weight.

Attribute data is based on upon discrete distinctions such as good/bad, percentage defective, or number defective per hundred. It is also known as “discrete” or “count” data. This type of data relates to attributes that can be classified into categories. For example, a patient is alive or dead, pregnant or not pregnant.’

You might find this article helpful when you decide what type of measurement you can use:

Decide on and qualify your Measurement System

It is really important that you qualify your measurement system. This helps you to ensure that if an error exceeds an acceptable level, it should not be used.

Develop a sampling plan and start data collection

In order to collect subgroup data, you will need to develop a sampling plan. You'll need to train your data collectors to collect data at a determined frequency and in a random fashion.

Decide on what kind of chart to use

You will also need to decide what type of chart to use – a Run Chart or an SPC chart. This will depend on what type of data you collect. SPC charts are different to run charts, where you use an X and Y axis, and are much more detailed. You can read more about the differences and how to decide which chart to use in our blog.

Analyse Data to Identify Root Cause and Correct

As you use control charts as the basis for taking action to improve a process, you need to analyse data. You are looking for the process to be stable in order to continue. And a process is considered stable when there is ‘random distribution of the plotted points within the limits’. If you note an ‘out-of-control’ condition, the next step is to collect and analyse data to identify the root cause.

So, as we’ve seen, there are many stages to consider when you implement Statistical Control Process or SPC. And it is a complex process. We also know that it’s important to start monitoring the process before you implement any changes. There is a lot of work and you need initial resources to set up SPC. And it can be time consuming.

However, there are things you can do to make life easier for yourself and your QI team. Just like using software such as Life QI. This software makes easier and less time consuming to implement Statistical Process Control with automatic alerts to remind you to carry out certain charts. It can give you that peace of mind, as well as help you to improve the analysis of improvement project processes.