We are testing a new beta website for gov.scot go to new site

A Guide to Service Improvement : Measurement, Analysis, Techniques and Solutions - Tools and techniques for the delivery of modern health care


2 Understanding the Patient Journey - Analysis

2.1 Introduction

This section of the Guide to Service Improvement explores a number of tools which provide a structured approach to analysing services and care processes from both the patient and service perspective. Accurate analysis of care processes as well as a clear understanding of demand, activity, capacity and queue is essential to achieve effective and sustainable service change or redesign.

A good definition of a process describes it as a series of connected steps or actions to achieve an outcome.

2.1.1 Using mapping to analyse services:

What is Process Mapping?
Process mapping is a simple technique that can have a significant impact. It is a tool which is used to capture the patients journey of care at every stage. Process maps are an effective way to identify constraints and bottlenecks, rework (activity required to correct situations that could have been avoided) and unnecessary process steps.

Understanding care processes from the patient's perspective is essential if patient focussed service improvements are to be made.

2.1.2 Why is Process Mapping a useful tool to begin to analyse services?

It is important to understand how patients flow through the care delivery system. Process mapping is patient focussed and helps us to understand how patients experience care delivery processes and systems.

It is unlikely that any one member of staff will fully understand the whole service until the process has been mapped. Process mapping is one of the most useful diagnostic tools for determining where problems lie and identifying areas for improvement.

2.1.3 When should Process Mapping be carried out?

Process mapping should underpin all service redesign, demand, capacity, activity and queue management, patient flow modelling and service planning. Process mapping, along with measurement of demand, capacity, activity and backlog provides the evidence base for service improvement. Without a clear understanding of the processes of care there is a risk of changing parts of a process which will not improve the service from the patients perspective and will actually lead to more waits and delays for patients.

2.1.4 How to Process Map

The best way to learn how to process map it is to do it.

The stages in Process Mapping
There are two stages to process mapping. First, understand what happens to the patient, where it happens and who is involved. Second, examine the process map to determine where there are problems such as multiple hand-offs, parts of the process that are unnecessary or do not add value, and where in the process queues occur. These include waste, error and duplication of parts of the process which would flow better if undertaken in a different order.

The process map must always depict the total number of steps taken as well as the number of people involved, the total time taken to perform the process step, and all documents used.

Whilst process mapping is simple, good preparation in advance will ensure that the process map is useful and supports service improvement:

  • Start by defining what you are trying to achieve
  • The start and end point of the process - the scope
  • What measures are you going to use to demonstrate that changes actually do improve the service. The Model For Improvement in Chapter 6 of this guide explains the importance of measurement. Process mapping is the first step in the improvement process and should inform the measures
  • Identify which staff need to be involved in mapping the process
  • Involve patients and or carers
  • Identify the level of detail required

The resources required to Process Map
The process mapping exercise will usually take one day to complete. This may be one whole day or two half days.

A suitable venue is important. Dependant on the level of detail and the size and complexity of the process to be mapped you may have a small or large group of staff involved.

The materials to capture the process map. "Post-it" notes are useful to capture the information. This way they can be rearranged as the exercise progresses and the patient journey emerges. You will also need flip charts, Sellotape and coloured marker pens. The map needs to capture information at the level of one person, one place and one time. Arrange the post-it notes to ensure they capture the journey as the patient experiences it. The first step is to get a group of staff to map the process at a high level by defining key points in the patients' journey. Use the high level map to determine who should be involved in the detailed mapping exercise. It is essential to have representatives of all staffing groups involved in the process at the main mapping exercise. Do not forget to involve patients and or carers. The aim is to map the service as the patient actually experiences it. The patient journey needs to be captured down to the level of "one person, one place and one time". Walk the process with a patient to check that all events are included.

Walk the process with a patient

The learning from service improvement in the NHS suggests there are some key causes of waits and delays, they are:

This is where patient care or information is handed from one individual to another.

Bottlenecks and Constraints
A bottleneck is any part of the system where patient flow is obstructed causing waits and delays. It interrupts the natural flow and hinders movement along the care pathway. However there is usually something that is the actual cause of the bottleneck and this is the constraint. The constraint is that part of the process which is the ultimate restriction on the amount of work that can be done.

There are two different types of bottlenecks; process bottlenecks and functional bottlenecks.

Process bottlenecks are the stage in the process that takes the longest time to complete. Process bottlenecks are often referred to as the rate limiting step or task in a process. In the following example the process bottleneck is step three of the process because it takes the longest time to complete:

Process bottlenecks

Functional bottlenecks are caused by services that have to cope with demand from several sources. Radiology, pathology, radiotherapy and physiotherapy are often functional bottlenecks in healthcare processes. Functional bottlenecks cause waits and delays for patients because they act like a set of traffic lights stopping the flow of patients in one process while allowing the patients to flow in another.

Functional Bottleneck

Functional bottlenecks

The three processes queue for the functional resource and wait their turn.
The functional bottleneck interrupts the flow of the patients in all three processes.

Batching of work
Batching occurs where tasks are placed in a queue before processing. An example of this might be found in the way specimens or samples are sent to the laboratories for analysis and tests.

Parallel Processes
Parallel processes are important and can be the cause of delays for patients and frustration for staff. Mapping, analysing and improving parallel processes can deliver significant improvement. A typical example of a parallel process might be the administrative process between primary care and secondary care for a hospital appointment, which is illustrated below. Parallel processes need to be mapped and understood.

Parallel Processes

2.1.5 Analyse the Process Map

Once the process map is complete the next stage is to analyse it by considering the following:

  • How many times is the patient passed from one person to another (hand off).
  • Where are the delays, queues and waiting built into the process.
  • Where are the bottlenecks.
  • What are the longest delays.
  • What is the approximate time taken for each step (task time).
  • What is approximate time between each step (wait time).
  • What is the approximate time between the first and last step.
  • How many steps are there for the patient.
  • How many steps add no value for the patient.
  • Are there things that are done more than once.
  • Look for re-work loops where activities are taken to correct situations that could be avoided.
  • Is work being batched.
  • Where are the problems for patients.
  • At each step is the action being undertaken by the most appropriate member of staff.
  • Where are the problems for staff.

The purpose of analysing the process map is to identify if the patient is getting the most efficient care at the most appropriate time and place.

Consider the steps which cause the most delays for patients and staff. These steps should then be mapped in more detail. This can be done several times, each time getting a greater level of detail.

2.1.6 An example of a high level and detailed process map

Below is an example of the difference between a high level and detailed process map. It demonstrates the level of detail which is required to truly understand the total patient journey. This example comes from the Centre for Change and Innovation Cancer Service Improvement Programme.

High Level Map

High Level Map

Detailed Level Map

Detailed Level Map

The information from the detailed process map has led to doctors, radiologists and clerical staff in NHS Tayside making a number of changes to their working practices which has reduced the time that patients wait to be seen by a respiratory physician. Previously patients would wait an average of 23 days to be seen. This has now been reduced to an average of 8 days. Radiology staff are now making direct referrals to the respiratory clinic after reporting a highly suspicious chest x-ray. Details of the results and the patient appointment are sent to the GP.

2.1.7 What next?

Generate ideas for change that will improve the patient journey. Identify and test change ideas. Start to use The Model for Improvement in chapter 6 to test and implement changes:

  • Identify the improvement areas from your process mapping and simplify them into manageable tasks Plan
  • Initiate system redesign and carry out the test or change Do
  • Study the data before and after the change Study
  • Plan the next change cycle or plan implementation Act

2.1.7 Where can I find out more information on Process Mapping?

The Centre for Change and Innovation website: www.cci.scot.nhs.uk

2.2 Using flow theory to analyse services

2.2.1 What is Flow?

The term flow describes the progressive movement of products, information and people through a sequence of processes. In the NHS, flow is the movement of patients, information or equipment between departments, staff groups or organisations as part of their pathway of care.

Why is flow modelling useful and how is it different from process mapping?
Process mapping looks at the care process from a patient perspective. Flow analysis builds on the work of process mapping and looks at the care process from a unit or department perspective bringing a number of patient process maps together to look at work flow.

2.2.2 When is flow modelling useful?

Understanding patient flow is vital in complex systems to fully identify repetition, queues, constraints and delays and to balance demand with capacity at each stage of the process. Flow modelling can support service improvement at specific bottlenecks or constraints, in specific clinical areas or across whole health systems. This analysis tool will not show what should be done. The tool will show how well scarce resources are used and how much room there is for improvement.

2.2.3 How to build the flow model

Define the patient group to be analysed and define the start and end points of the flow map.

Building the flow model is simple if all of the steps are followed. In straightforward care processes the process map and the process times will provide sufficient information to examine patient flow. For more complex care pathways the steps to identify patient flow are as follows:

Step 1 Map and agree the process. The process must be mapped to a high level of detail

Step 2 Simultaneously identify and map the communication and information steps around the patient process. The information and communication steps may be complex and cross many departmental and organisational boundaries.

Step 3 Identify who is clinically responsible for the patient. This could be the person performing direct clinical care for the patient. It may change as the patient moves from department to department and maybe within each clinical area.

Step 4 Identify who is responsible for the overall process at each point.

Step 5 Work out the time the whole process takes by identifying how long each stage in the process takes. Use a common unit to record time i.e. hours or minutes.

Using graph paper draw each patient as a horizontal bar one above the other. Set the horizontal axis to represent time with the start of the process on the left and the end of the process on the right. Colour each stage of the process a different colour.

For example in the graph below the constraint is coloured red. It is easy to see over the course of a day that the flow of work to the constraint is inefficient. By examining the flow of patients through the care process in this way it is possible to identify the scope for improvement.

examining the flow of patients through the care process

Where can I find out more on Flow Modelling?
The Unscheduled Care Collaborative sponsored by the Centre for Change and Innovation is using flow modelling to support service improvement in unscheduled care. More information can be found at: www.cci.scot.nhs.uk

The Modernisation Agency www.modern.nhs.uk

The Institute for Healthcare Improvement, USAwww.ihi.org

2.3 Understanding the impact of variation in clinical systems

2.3.1 What is variation?

Variation is a part of everyday life and occurs naturally in most processes. In healthcare measuring, understanding and reducing variation is key to improving patient flow. Variation can be seasonal, monthly, weekly, daily or hourly. Much of the variation in clinical systems is within our control as outlined below:

Sources of variation in a clinical system

Sources of variation in a clinical system

2.3.2 Why is it important to understand variation?

Being able to analyse and understand current system variation is key to being able to make changes that improve overall patient journey times. The variation between demand and capacity is one of the main reasons why queues occur in the NHS, because every time demand exceeds capacity a queue is formed showing itself as a waiting list or backlog. Chapter 3 of this guide explains how to measure demand, capacity, activity and backlog, and chapter 5 explains some practical ways to support better management of queues.

Variation is a part of every day life and it can be Natural Variation or Artificial Variation.

Natural variation is an inevitable characteristic of any healthcare system. Sources of natural variation include:

  • differences in symptoms and diseases that patients present with
  • the times of day that trauma cases arrive
  • the socio-economic or demographic differences between patients

Natural variation cannot be eliminated or reduced it must be understood and managed.

Artificial variation is created by the way the system is set up and managed.

Sources of artificial variation include:

  • the way we schedule elective admissions
  • working hours of staff
  • how staff leave is planned
  • availability of clinical equipment
  • multiple points of entry into services

Artificial variation has much more impact on patient flow than natural variation. It is usually driven by personal preferences and the priorities of staff rather than the needs of the patient.

2.3.4 How to analyse variation - Statistical Process Control

2.3.5 What is statistical process control?

In recent years the benefits of using Statistical Process Control ( SPC) methodology in service improvement have been recognised by a number of improvement agencies. It is a statistical tool based on scientifically robust methodology that uses data to analyse and understand the inherent variation within processes and systems. SPC provides a tool to separate the "information" from the "noise" in data. It does this by providing a mathematical basis for establishing the upper and lower limits of variation in processes that occur normally.

In SPC terminology variation is described as common cause variation, that is variation which is normal and to be expected and special cause variation which produces unusual or unexpected variation.

Two of the simplest SPC techniques are the run chart and the control chart. The purpose of these two techniques is to identify when a process is displaying strange or unusual behaviour. The purpose of the run chart and control chart is to distinguish between the two sorts of variation that a process can exhibit.

2.3.6 Why is statistical process control useful in the NHS?

Statistical process control is a tool to help us to identify and differentiate between the two types of variation that processes exhibit. It also helps to identify trends in data over time. It can also give us early warning signs that a process needs to be redesigned because it is showing unusual patterns of variation.

2.3.7 When can Statistical Process control be used in healthcare?

SPC can help in virtually all aspects of managing healthcare. In the context of service improvement and redesign SPC charts can be used to analyse and monitor changes made as part of the improvement process. As part of a Plan, Do, Study, Act improvement cycle the control chart can be set up to monitor the processes that are being changed. It is then possible to determine whether a change to the process has in fact led to a change in the performance of the process.

2.3.8 How do I create the statistical process control charts?

Run Charts
A run chart is a line graph. It is used when you want to show performance over time or when you want to look for trends or patterns over time. A line graph provides an ideal method of comparing sets of data.

Control Chart
Control charts are basically run charts, but with two distinct differences:

  • a line showing the average or mean values of the measure
  • two lines showing the upper and lower process control limits. These show the typical extent of variation over time.

Using the data from the run chart you can calculate the mean or average and the control limits. The standard formula used in SPC produces the upper control limit and the lower control limit. These are plus and minus three standard deviations from the mean. The mean is obtained by finding the sum of all values divided by the number of values. The mathematical formula to calculate standard deviation is simple to calculate using standard spreadsheet software designed for this purpose.

The value of the control chart is that it enables us to identify if a process is showing common cause variation or special cause variation. They key is to understand the reason for special cause variation and not to react unnecessarily to one off changes in the behaviour of a process. In service redesign you would expect to see changes in the way a process is behaving along the improvement journey as a result of introducing changes using the model for improvement described in chapter 6.

2.3.9 An example of the impact of reducing variation

The following is an example from NHS Borders Colon Service where the variation has been reduced by streamlining the referral pathway, improving the flow of patients through the care process, and improving waiting times:

Referral pathway before Borders Colon Service

Referral pathway before Borders Colon Service

Referral pathway after Borders Colon Service

Referral pathway after Borders Colon Service

2.4 Where can I find out more information

Web sites:

Centre for Change and Innovation www.cci.scot.nhs.uk

The Modernisation Agency: www.modern.nhs.uk

The Institute for Healthcare Improvement, USAwww.ihi.org

Further reading:

Hart MK, Hart R (2002) Statistical Process Control for Healthcare, Duxbury Thompson Learning

Wheeler DJ (1995) Understanding Variation: The Key to Managing Chaos (2nd edition)

Carey RG, Lloyd RC (1995) Measuring quality improvement in healthcare: A guide to statistical process control application, Winsconsin, Quality Press