Quality Improvement Toolkit for Diabetes Care

Quality Improvement tools

There are many quality improvement tools available. The attached tools are ones that can help practice teams to improve the care of people living with diabetes.

In the QI training section there is a presentation that gives you slides on these tools and some exercises to use in group work.

Context

Considering context and culture first will allow quality improvements to flourish. The context checklist allows practice teams to consider what needs to be in place to encourage improvement in diabetes care. Changing organisational culture can be a challenge and takes time, but these issues do need to be considered when planning a change.

• Download the context checklist 

Process mapping

Do you need to look at your practice’s processes in order to better care for patients living with diabetes? This tool creates a visual representation of all the steps in a process.

This practice has looked at the map from diagnosis of diabetes to annual review.

The objective is to plan a more efficient process and to identify changes to be made. It was the most popular and obtained the best feedback from the practices who participated in the pilot project. Practices have found it useful to fully explore their review systems for diabetes. Guidance on a process map for reviewing patients can be found here:

• Download process mapping for diabetes 
• There is also a generic guide available: Download process mapping mini guide 

Fishbone diagrams

If you are trying to work out why patients don’t turn up for review or why the treatment target for blood pressure is not being met, then this tool can help you. Fishbone diagrams are also called ‘cause and effect analysis’. They are useful when there are multiple causes of a problem. Practices in the pilot did not choose to use this tool. An illustration can be found here:

• Download the fishbone illustration 

Model for improvement and PDSA cycles

Do you want to introduce a change to improve diabetes care? If so, this model allows you to plan and test the change.

The model for improvement ensures that you and your team are very clear and specific about what you want to improve and how you will know if you have been successful. The model for improvement poses three questions:

• What are we trying to accomplish? 
• How will we know that a change is an improvement?
• What changes can we make that will result in improvement?

Once the third question has been answered, a change can enter a Plan-Do-Study-Act cycle (PDSA). Using this approach allows us to continue the changes that work and stop those that do not. An illustration of the Model for Improvement and PDSA cycles using a change from diabetes care can be found here

• Download the Model for Improvement for diabetes 
• There is also a generic guide for this approach: Download the MFI/PDSA generic guide

QI Ready

To access shared learning networks to assist you in applying practical QI methodologies to better treat this clinical area, join our QI Ready platform.

As a person responsible for improving the care of patients with diabetes across a locality, it is important to have a good understanding of the current performance of your GP practices. This will help you to identify variation in care so that you can prioritise your QI training and interventions.

Nationally-generated benchmarking data can help you with this task. The links below take you to the data you will need, but there are some limitations:

• Data needs to be interpreted carefully, taking into consideration variations in practice context: for example, demographics of the population and list size turnover. As a local clinical lead, you can use your local knowledge to help make sense of variation. 
• Many of the externally-collected data sources are not ‘real-time’ and will not reflect improvements that have already taken place.
• Some of the datasets may be incomplete as practices may have ‘opted out’ of external data collection.
• Not all datasets are applicable across all four nations of the UK.

National Diabetes Audit for England and Wales

The National Diabetes Audit (NDA) continues to provide a comprehensive view of diabetes care in England and Wales and measures the effectiveness of diabetes healthcare against NICE Clinical Guidelines and NICE Quality Standards. Practices have to agree to participate before their data can be extracted.

• NDA Results from 2013/14 and 2014/15 Data Collection

The Excel spreadsheets showing your CCG/LHB’s performance can be downloaded from the site. Ensure macros are enabled so that you can browse the data.

Practice-level data is available for England, but not for Wales. The spreadsheets in audit report one cover participation, registration and demographics; care process completion; structured education: target achievement. Graphs can show the data by audit year and diabetes type. Practices in Wales can access their data via the Audit + NDA practice support module.

On this site there is a power point presentation, along with slides highlighting the national findings there is also space to allow the incorporation of locally produced slides, which can be used to disseminate the results of the audit locally.

Quality and Outcomes Framework (QOF) Data

• QOF Data for the UK up to March 2015

This site is independent of the NHS and pulls its data from published QOF performance data across the four nations of the UK. It is possible to ‘drill down’ to practice-level data for England, Scotland and Northern Ireland. QOF data for Wales is available at Health Board level only. The site allows you to compare your locality with others, as well as with national performance. QOF indicators differ across the four nations.

This data can be helpful in deciding your improvement priorities. In diabetes care it gives information on each indicator and prevalence but not on the number achieving all treatment targets. The treatment targets vary slightly between QOF and NDA. QOF has removed all measurements of care processes apart from foot surveillance so these figures can not distinguish whether a target is not being achieved because the process such as blood pressure measurement has not been done or whether the target level has been reached. Smoking in QOF is now bundled with other diseases. The indicators do not deal with type one and type two diabetes separately which may be important in deciding an intervention when patients with type one diabetes may be cared for mainly in secondary care. QOF data for the preceding April-March is usually published by each nation around October; this website then uploads the data in an easy-to-navigate format over the subsequent few weeks.

Healthier Lives

• Public health data for England

This site pulls data about diabetes (and other long-term conditions) from a variety of sources. It then benchmarks GP practices within their CCG for performance related to both processes of care and the achievement of treatment targets.

The site is very easy to navigate and takes into consideration the level of deprivation of the patients within a practice and CCG area.

There is a time lag between the collection of data and the publication on the website of 12-24 months.

Practice level data sources

Real-time data about performance in diabetes care is not usually visible to locality clinical leads without visiting their local practices. Practices have a number of ways of gathering this data internally in order to assess performance and decide on improvement priorities.

All GP Electronic Medical Record Systems provide a range of tools to give practices real-time data (usually updated every night) about performance against QOF targets, including diabetes care.

GPs and their teams can also run specific searches related to an area of diabetes care they might be interested in and this can be an important step in the ‘diagnosis’ stage of a QI project. An example might be trying to work out how many patients without a diabetes diagnosis have had a raised HbA1c, but no follow-up provided or planned.

Because of the complexity of some of these searches, the University of Nottingham has developed a range of audit tools that can be linked to the practice’s medical record system to give information about performance in a variety of long-term conditions, including diabetes. The PRIMIS Diabetes care audit tool is currently free for practices to download. It has been developed with pharmaceutical company sponsorship, although PRIMIS retain full editorial control.

Eclipse is another extraction tool used in some areas that can be utilised in benchmarking and audit in LTCs. It is composed of three main elements: Eclipse, Eclipse Live and NHSpatient.org. In addition, there is a specific component for diabetes which links to Eclipse Live and NHSpatient.org: Diabetes Manager. It is a commercial product that has been purchased by some primary care organisations.

Measuring and interpreting data

Measuring is an important task in all QI work. You will need to find a way of knowing if your changes have generated an improvement. Some data is easy to interpret, for example a steady month on month increase in the number of patients with optimum HbA1c control.

In dealing with the target for HbA1C, remember that any action can take up to 3 months to have an effect on an individual patient. This means that there will be a delay before you see effect on the chart.

This simple line graph with a goal line is useful when your project involves a gradual improvement in something without much variation and little variation due to chance. It can be used for process measures such as the number of patients who have had their blood pressure, cholesterol, HbA1C, BMI, creatinine, urine albumen, foot surveillance, or smoking status checked as the year progresses. Line graphs are also good for displaying diabetes prevalence, if your project is to find previously undiagnosed individuals.

This cumulative chart was used by one of the pilot practices to show how an intervention had improved their measurement of ACRs

However, some projects can produce data that is highly variable, especially those aimed at improving processes within a practice. Examples of this type of data are:

• The length of time between diagnosis of diabetes and first appointment with the diabetes nurse. 
• The proportion of out-of-target HbA1c results that are actioned each week.

The example below shows the data collected for a project to try different methods of inviting patients to come for their foot check. The measure chosen is the proportion of patients who respond to the invitation each week. The practice was trying a variety of methods for sending for patients (letter, text, phone call) to see which worked best.

This is an example of a ‘run chart’, which makes it possible to see whether the implemented change resulted in an improvement. A run chart is useful when:

• There is variability in the item being measured. 
• It is possible to make frequent measures. It is best if 15 measurements (and at least 10) can be made and recorded before the introduction of a change.

A Run chart is not the best tool to use if:

• The amount of variation is small e.g. only varies by one or two

• The measurement is on 0% or 100% for about half the measurements.

• There is no variability. This may occur when the item being measured is steadily increasing or decreasing

In summary, plot a line graph if you have data from frequent measures. If this shows minimal variation then set a goal, but if there is variation that could be arising from chance, plot a run chart and monitor your change.

Click on the link below to find out how to create and interpret a run chart for variable data. This will show you how to download an Excel spreadsheet that will create your chart. The guide informs you of the rules of analysing the chart and shows examples of their use.

Displaying Data

Data, if displayed well, can be a powerful QI tool as it helps to keep the people involved with your project engaged and motivated. It works best to create large simple graphs and display them on a noticeboard or large electronic screen in a prominent area so people can see them easily, even when the project isn’t directly being discussed.

Example of visual display in action

This magnetic display board has been used by one general practice as a central place for the whole practice team to track their improvements. It shows run charts, a pareto chart, bar graphs and a pie chart that display data relevant to improvement projects.

The board is in the main meeting room, where the staff also eat lunch. Displaying data in a prominent place has helped to:

• Create a practice culture where continuous  improvement is the ‘norm’

• Engage individuals in the team with collecting and displaying the data.

• Engage the whole practice team by regular conversations about the data at practice meetings

• Keep projects on track, and pick up when changes haven’t been achieved or sustained so that something new can be tried 

Guides - for all Quality improvement activities

Quality Improvement for General Practice

This is the RCGP QI Guide, and is not specific to improving diabetes care. The guides contained within this toolkit are derived from this generic guide. If you have tried the methods and tools to improve diabetes care then you may wish to use them in other areas of your practice. It is specific to primary care and can be used by all members of the team. It introduces the Quality Improvement wheel, which is a visual representation providing an overview of your quality improvement journey. It contains more tools than the other pages in this toolkit.

• Download Quality Improvement for General Practice Guide

Mini-guides

The following guides are extracted from the larger guide above and are useful if you just want details of a particular tool:

• Download practical guide on how to build a run chart 
• Download practical guide to process mapping 
• Download a practical guide for Model for Improvement and Plan-Do-Study-Act cycles 

QI and Diabetes Training materials

This section is principally aimed at those who are responsible for a group of practices such as in a federation or cluster. If someone in a practice has an interest or expertise in QI then they will find these materials useful.

Materials on this page can be used to introduce QI tools and methods to the teams who care for patients with diabetes. The materials can be tailored to the time you have to cover each topic. Training is best delivered to as many of the team as possible and it is suitable for both clinicians and non-clinicians. The aim of any session is to leave the attendees with a definition and understanding of QI and to show how it can be used to improve the care of patients with diabetes. It is best to finish the session with an agreement of which area the practice is going to improve.

Presentation

The training on QI in diabetes should take between one and a half and three hours (dependant upon the components chosen) to allow sufficient time for the group work and questions during the presentation. However sometimes clinicians and practice staff cannot spare this amount of time which can especially be the case when the training is delivered in the practice itself. When less time is allowed for the training the time for group work in particular has to be curtailed and the tasks may have to be completed by the delegates at a later date. The context checklist could be a pre- training task or completed away from the training. The timetables give suggested timings for a 3 hour, a 2 hour and a one and a half hour session.

• Download suggested timetables 

Slides in this presentation cover:

• Definition of Quality Improvement 
• Context
• Patient involvement
• Process mapping
• Data
• Fishbone diagrams
• Model for improvement
• Plan-Do-Study-Cycles
• Displaying data
• Run charts

The presentation should be interactive and interspersed with group work. Further information on all these methods and tools are contained in this toolkit.

• Download presentation [PPT]

Group work

During a session it is best to have group work to help to embed the learning. Time may only allow a taster of each of the tools and members will need to complete the work at a later date. Further details on the tools can be found in the Useful QI tools section. This section only includes how to introduce them to a training group.

Context checklist

The context checklist looks at the context that is relevant to the care of patients with diabetes. It may be that the main barrier to improvement is that the team never take the time to meet together to make decisions about practice processes. For group work, take the context checklist and ask the participants to complete it together. Inform the participants that some of these factors will need to be modified prior to commencing any change, and for some others it may not be easy to find solutions. Participants should take it away and add to it after the training session.

• Download context checklist 

Process mapping

In the training session you can ask the participants from one practice to work together to produce a map of the review process of known patients with diabetes. If the meeting is being held in the practice, this is best done on a wall or on a sheet that can be attached to a wall so that others can look at the map later. This group work can also be done with participants from different practices and they can use a table-top to construct the map. Download instructions on how to conduct this group work here.

• Download process mapping instructions 

Model for improvement/Plan-Do-Study-Act

Before doing this exercise it is helpful if the practices have looked at their diabetes performance data (either from the NDA results or any in house data they have, such as QOF). Groups of participants from the same practice can then work together on their priority area. If there is a problem common to many practices, participants could work in mixed-practice groups. Instructions for this group work can be downloaded here.

• Download MFI/PDSA instruction 

Run charts

If you have time available in a training session, this group work on run charts takes about 10 -15 minutes and supplements the guidance contained in the presentation. It is important to stress that run charts are used when the item being measured varies before any intervention has been tried. You can download the instructions for this exercise here and examples of run charts to interpret.

• Download instructions on run charts 
• Download examples of run charts 

Checklist

If you are overseeing a quality improvement project in a practice that is not your own, it can be useful to keep a note of its progress on a word document. This can be completed when you make contact with the practice either by phone, email or a visit.

• Download practice contract record 

Multi practice plan

You may wish to download an Excel spreadsheet to monitor the progress of several practices. There is a version that can be downloaded here.

• Download multi practice plan

The first row shows an example practice. Their intervention was to use one doctor with expertise in diabetes to review all of the out of target HbA1C results. In the third column various interventions are described, and in the fourth column the intended date of action is entered. Those date(s) are then illustrated by colouring the background of the cell blue using the “fill colour” button in the calendar section of the sheet. The status column uses the RGA system where the box is filled red when action is yet to commence; amber when it has started, but not finished; and green when an action is completed.

This report summarises the results and conclusions of a primary care diabetes quality improvement programme commissioned by NHS Digital (formerly the Health and Social Care Information Centre) and conducted by the Clinical Innovation and Research Centre (CIRC) of the Royal College of General Practitioners (RCGP) between June 2015 and September 2016.

• RCGP National Diabetes Audit Quality Improvement Programme Report - September 2016

If you are a CCG or Health Board clinical lead involved in improving the care of patients with diabetes, you may be asked to provide an evaluation of your work. During 2015/16 the RCGP QI clinical leads worked with local clinical leads in England and Wales to carry out similar improvement work. This section includes the evaluation tools used for this project that can be applied to your work.

Reflection template

This template can help you record the context of any reflection you have on contacting a practice, any new information, what has gone well, what could be improved, and the next steps.

• Download reflective sheet 

Interview template

This template allows you to capture what the practice has done, which interventions they have tried, and which quality improvement tools they used. You can record how successful their project was and the barriers to implementation too.

• Download interview template 

Baseline questionnaire

This questionnaire was designed to be used electronically to assess an interviewee’s context, work environment, and their knowledge of QI.

• Download baseline questionnaire 

Follow up questionnaire

Some of the questions in the baseline questionnaire are repeated to assess any change and there are specific questions about the intervention and quality improvement tools they have used.

• Download follow up questionnaire