statistical process control (spc) - nhs institute for innovation and improvement

Quick links to popular pages

Print

Statistical Process Control (SPC)

Back to previous page

Statistical Process Control (SPC)

Download

BaseLine invitation & introduction (healthcare).pdf (49.83 KB)

Video instructions

Click here to view video instructions (opens in new window)

Purpose

SPC is a practical statistical approach to resolving problems. If you do any type of

measurement to help gather information and find a solution, this is the tool you should use.

When to use it

You should use SPC throughout the life cycle of the project:

Project identification

Getting a baseline

Where are we now?

Did the project make a difference?

Will the project sustain?

Evaluating the worth of the project

How to use it

An SPC chart looks like this:

SPC is

a type

Fundamentals for quality improvement

Quality and service improvement tools

Facilitation guides

Reducing delays in patient care

Tackling NHS challenges

Organising for Quality and Value

HomeHome About usAbout us CatalogueCatalogue FAQFAQ Contact usContact us EventsEvents Log inLog in Create an accountCreate an account

Safer CareSafer Care Quality & ValueQuality & Value Building CapabilityBuilding Capability CommissioningCommissioning ProductivesProductives NetworksNetworks LeadershipLeadership InnovationInnovation WorldwideWorldwide

Share

ToolsTools

Statistical Process Control (SPC) - NHS Institute for Innovation and Impr... http://www.institute.nhs.uk/quality_and_service_improvement_tools/quali...

1 of 5 5/13/2012 9:20 AM

of

charting that tells us about the variation that exists in the systems that we are looking to

improve.

S - Statistical, because we use some statistical concepts to help us understand processes

P - Process, because we deliver our work through processes. i.e. how we do things

C - Control, by this we mean predictable

When you are interpreting SPC charts there are 4 rules that help you identify what the

system is doing. If one of the rules has been broken, this means that ‘special cause '

variation is present in the system. It is also perfectly normal for a process to show no

signs of special cause. This means that only ‘common cause ' variation is present.

Rule 1 Any point outside one of the control limits:

Rule 2 A run of seven points all above or all below the centre line, or all increasing or

decreasing:

Rule 3 Any unusual pattern or trends within the control limits:

Rule 4 The number of points within the middle third of the region between the control limits

differs markedly from two -thirds of the total number of points:


2 of 5 5/13/2012 9:20 AM

If you want a more

efficient system, you

need to reduce the

variation. Common

causes and special

causes of variation

indicate the need for

two different types of

improvement which can

help you achieve this. If

controlled variation

(common cause) is

displayed in the SPC

chart, the process is

stable and predictable, which means that the variation is inherent in the process.

If you want to improve the process, you will have to change the whole system

l If uncontrolled variation (special cause) is displayed in the SPC chart, the process is

unstable and unpredictable. Variation may be caused by factors outside the process. In

this case, you need to identify these sources and resolve them, rather than change the

system itself.

There are three issues that you should be aware of when using SPC charts to improve a

process:

You should not react to special cause variation by changing the process, as it may

not be the system at fault

You should not ignore special cause variation by assuming that its part of the

process. It is usually caused by outside factors which you need to understand in

order to reduce them

You should ensure that the chart is not comparing more than one process and

displaying false signals.

An example of this would be data covering two hospital sites, or two procedures that

are very different

SPC is a tool that will help you understand the scale of any problem (the degree of

variation) and identify possible causes when used with other investigative tools e.g.

process mapping , spaghetti diagram . You are then able to measure the impact of any

improvement: does it cause more variation - bad, or less variation - good? To understand

more about variation click here.

How to use it

The SPC Generator Tool has now been replaced with the Baseline Tool. For more

information on how to download this, please download the Baseline invitation and

introduction (Healthcare) (49.83 KB) document.

This package constructs all types of SPC charts to help you understand your processes. It

helps you understand what you need to do and how to interpret the results step by step.

Some issues you may encounter when creating your own SPC charts:

Available data - you may need to collect the data for analysis as it may not be

available. To be statistically rigorous, the number of observations (the points you

are measuring) are important. The more frequently you record the observation the

better: daily or weekly is better than monthly

Aggregate data is discouraged (ie the use of percentages, as this often hides the

pattern of the data)

The problem you are observing may be the means by which you are measuring, not

what is really happening to the patient. Sometimes it is better not to act if you aren 't

sure. Investigate further instead

Remember that when you change something in the process, the data points after the

change will be from a new system. When you have a run of points which break a

rule, you will need to recalculate the SPC control limits to show an improvement

(showing the control limits of the new system)

What next?

PDSA cycles: measuring the impact of a PDSA cycle is best done with SPC


3 of 5 5/13/2012 9:20 AM

Variation: essential reading for using SPC charts

Process mapping: useful for understanding variation revealed by SPC charts

Root cause analysis using the five whys: useful for understanding variation revealed by

SPC charts

Sustaining the momentum: to ensure continuous improvement

Additional Resources

Books:

Bicheno and Catherwood ‘Six Sigma and the Total Quality Toolbox' Picsie Books, 2005

Donald Wheeler ‘Understanding Variation ' Knoxville: SPC Press Inc, 1995

Donald Wheeler ‘Making Sense of Data. SPC for the Service Sector' Knoxville, SPC

Press Inc, 2003

Walter A Shewhart ‘Economic Control of Quality of Manufactured Product ' New York, D

Van Nostrand 1931

W.E Deming ‘Out of the Crisis ' Massachusetts: MIT 1986

Donald M Berwick ‘Controlling Variation in Health Care: a Consultation from Walter

Shewhart ' Med Care 1991; 29: 1212 -25.

Esain, A ‘Problem Solving, TQM and Six Sigma ' in Rich, Batemen, Esain, Massey and

Samuel (2006)

‘Lean Evolution: Lessons from the Workplace ' Cambridge University Press

Background

The method of SPC was created by Walter Shewhart during the 1920s. Shewhart was

investigating the production of faulty telephones for Bell Telephones. He found that the

company continually changed the production line when faulty handsets were produced.

This resulted in the continual production of different formats of handset - telephones that

wouldn't talk to each other! Through Shewhart 's work, Bell Telephones understood the

importance of reducing the variation in the manufacturing process to ensure lean

production. Shewhart framed problems as ‘assignable cause ' and ‘chance cause ' and

introduced the SPC chart to differentiate between the two. We refer to these

statements/problems as predictable (normal cause variation) and unpredictable (special

cause variation).

1920s: First control charts developed by the communications industry

1930s: Adopted by other industries

POST WAR: America ignored the principles of SPC

1950s: Japan began to widely use the principles of SPC in industry

1980s: Ford Motor Company adopted SPC principles due to pressure on the US market

from Japanese imports

2000s: SPC adopted in the NHS

Data should always be presented in a way that preserves the evidence. Walter Shewhart

(1931) suggested that displaying data using averages and aggregates loses the richness

of the individual data points. SPC displays the individual data points (in the NHS these are

often individual patients), then provides analysis to interpret what the user sees.

SPC and PDSAs were created by the same person, intended to be used together. PDSA

improvement cycles should only be implemented in a system that displays common cause

variation. If special cause variation is present, then the system is not stable enough to

ensure that the information is a result of the PDSA change.

Recognising variation is a start, but it 's not enough. We must understand its effect and

manage it appropriately.

© Copyright NHS Institute for Innovation and Improvement 2006-2012


4 of 5 5/13/2012 9:20 AM

Accessibility Site map Terms and conditions Copyright Privacy Freedom of Information Media Centre Jobs Events


5 of 5 5/13/2012 9:20 AM

statistical process control (spc) - nhs institute for innovation and improvement

Documents