Key ingredients for success are:
- Will – amongst everyone involved
- Ideas- that can be tested out and adapted, adopted or abandoned
- Delivery- of change over an agreed time scale that everyone agrees is an improvement in one or all of these constituents of quality
Safety, Timeliness, Effectiveness, Efficiency, Equity or Patient Centredness
Before you start making changes:
Developing Will requires:
- Formal and informal discussions with all those involved in care about how outcomes, experience for patients and staff are being impacted at the moment and how much better this could be.
- Evidence through local or larger audits or reports of a gap between what should be happening and what is actually happening in the system you work in.
- Building hope that change can happen, that ideas are welcome, that it may take time but starting the journey if the next step.
- Agreeing who needs to be involved and make sure this is wide enough so as to involve all parts of the system that is being considered. It may be some team members are not needed as yet but involve them in the initial conversations so they know what is going on.
- Agreeing a time/ place where regular meetings can happen. This can be a new event or,, more practical, in the beginning is to tag 10-15 minutes onto an existing meeting or reallocate time already spent meeting (for example a multidisciplinary meeting)
The Project Approach:
Having a framework to keep you on task:
- Your Trust may have a commonly used quality improvement methodology it uses. Stick with this if there isn’t one, we suggest using the Model for Improvement as it is simple and well understood once explained.
- The principles are to follow these three questions in this diagram below much like you would if a patient presented to you with symptoms and signs followed by the sequence of testing ideas to select the ones that work and refine how well designed, they are.
- More details on this are available here or short videos are here
Diagrams below describe the Model for Improvement and the Plan, Do, Study, Act process.
Steps to take:
Understand what is happening now by doing one or more of the following:
- Ask patients and families
- Review incidents or thank you letters
- Check a big audit and/or do a small audit to see what the situation is now (25 cases will be adequate)
- Review national best practice guidance for currently unknown gaps in care provided
- Ask colleagues, especially new ones, who may have worked in different ways that could improve your system
- PROCESS MAP what happens now
- FISHBONE DIAGRAMS can help the whole team build a picture of what is needed for a better system
- DRIVER DIAGRAMS will build a one-page map of the main components that need improving and start the bring together
- SHARE these results by doing them together and then sharing them at a meeting, so everyone can add/amend/comment/be part of the baseline analysis
How will you know a change is an improvement?
- Review current outcome measures for your chosen project area
- Review the processes you have identified
- How do those need to improve to make the outcome better?
- Which of these can be measured directly? (time taken, how often for the whole group, did or didn't happen for relevant patients, good or bad experience)
- Which of these are harder to measure directly, but could have a surrogate measure (would did it again like that, difficult but needs practice, was better for me than my last appointment)
Build a Measurement Strategy:
Take your results from your diagrams/process maps and put some measures alongside areas that needs improvement; then agree the main SMART overarching measure of success.
Specific; Measurable; Achievable; Realistic; Timely.
Time ordered data collection is critical to success. Do not wait too long to check if the things you know matter are working better than they were. Read here for why data is key.
Get started testing change ideas:
- Let the whole group decide which to try first; go with the majority choice
- Share out the work
- Test the idea out as soon as you can and before you do agree when to report back
- Use this format and keep it simple, small and achievable within a week
Predicting the outcome of the test makes sure the team are clear on the reason for this being a good idea to improve things.
Meet weekly if you can:
- This ensures everyone is informed, involved, able to report and feed back new or differing insights.
- Keeps the momentum going.
- Adds pace to the project.
- Provides the chance for people to make most meetings but not feel disconnected.
- Provide short action notes swiftly to everyone so all know the progress being made.
- Give a team member the task of connecting with someone in a leadership role on a regular basis to keep them informed and therefore able to help if a problem arises
- Ask for support or help from a QI or improvement team within the Trust if you need it.
- Put the data and progress up in a space where your team can see it informally so everyone is in the loop.
Muddle in the middle:
Problems crop up due to the following:
- The problem has turned out to be more complex than first thought – take a small part of the initial plan and deal with this first and actively agree to do this together.
- The people who need to be involved are not able to join in/ seem less keen than first agreed- take the role of connector and communicator and remind all why this work matters to patients, families, clinical outcomes, the efficiency of clinical teams at work.
- There is a block in the system (technical, human, organizational). The most senior team member or the sponsor of the work (manager, medical director, Matron for example) is key now. They need to know what the plan is, why this issue that is blocked is key to progress, what could happen to get over the barrier, how the team proposes to allay concerns that may be present while still making progress on the set goal. The most senior team member needs to agree with this person next steps, however small and report back to the whole team.
- The progress is taking longer than first thought. Review the initial analysis together. Is there a missing component? Is there a new idea than might now be even better? Have patients been involved to give new insights? Agree to make the steps in the project smaller but come together and review more often.
- Some of the current team are about to move onto other roles/ parts of the rotation. In the few weeks before this is actually happening review the following.
- What have we achieved so far- reviewing the current situation is a great first chapter to have got finished?
- What do we need to measure going forward- how might we do this?
- Who can take on the next steps and develop a handover for them (what, why, who, how)?
- Agree how to either keep them involved or keep them in touch with progress
Any report, poster, presentation will be a combined success for all participants so staying in touch is pretty important for everyone.
Measurement – a step by step approach to knowing your changes are leading to an improvement:
Having a systematic approach to measurement is fundamental to a good and successful improvement plan. Dr Bob Lloyd from IHI has described the steps to take, and I have adapted them below to illustrate how you might work with your team to develop and put into action a good measurement strategy.
|Define the Aim / Goal||Safer care for inpatient on our ward||N/A|
|Concept||To reduce patient falls||Evidence on why shared|
|Measure||Inpatient falls per 1000 bed days||Might be NHS / Trust or new measure|
|Operational Definition||Actual fall by a patient (with or without injury) / number of patient days (28 beds each week is 28 x 7 days)||To make it clear to everyone exactly what the definition is so collecting data is consistent|
|Data Collection Plan||Every month, last day of month calculated and shared, all cases (not a sample) all the medical wards in our hospital||Plan what, when and how before you start. Any changes need to be agreed and not muddy the understanding but clarify the situation|
|Data Collection||Each ward submits data to agreed person. Agreed person gathers all data and sends to Y for immediate review and sharing||Data is what tells us our changes are having the impact we want. Feedback is key to motivation to continue, to avoid needless unintentional harm and to keep us on mission|
|Analysis||Plot each month’s data in sequence on a run chart. One you have 12/ more points one can see what is happening and changing or not and move to a control chart. Use median as measure of data not average to reduce risk of an extreme outlier data point masking real system performance||
If you an shorten the data collection interval you will gain much more speedy feedback on your progress. Assuming:
1. You are testing changes
2. There are enough events to make this practical
Let us look at a real project and move from high level measures, such as those above, to seeing how this is underpinned by PDSA and change measures.
An unexpected and preventable deterioration of a patient on our ward has led to an emergency call to the ITU outreach and a swift transfer to ITU- on reviewing the notes the patients Observations could have been acted on faster and possibly avoided the deterioration that now needs ITU support.
Project is to ensure all patients receive on time observations using the Early Warning Scoring System and that deterioration /lack of progress as expected is detected and acted on. Here are the steps to take then agree on and make sure you discuss and confirm 1-7 above. A good tip is to test the measure and collection plan for a few days then discuss so any glitches are ironed out early.
Step 1 - What should be the sequence of events for our patients?
Step 2 - Last week's audit of all cases on the ward
Step 3 - What does our team have plans to do to improve things in each / all of these areas?
We might divide out the work or do one change at a time. Some changes like never mentally adding up the score again and always using a calculator might be straight forward. Sorting out who calls who when may need changes to the rota display, regularly checking bleep numbers, agreeing a consistent way to communicate the facts, having a back up person if someone really cannot come now. Our current ideas in brief are here below in the far-right column. You may think of more than these.
Remember that this is not just about individuals it is about the WHOLE SYSTEM working as a consistent and highly reliable team and making the right thing very easy to do every time you need to do it. It is about what to do, why to do it and how to design all the things in the system (people, equipment, structures, feedback loops) to make it possible every time it is needed.
Step 4 - What might you measure / how do you know patients are still falling through the improving system?
These are signs of ongoing failure and tell you there is more work to do (either more reliable delivery of existing ideas or some additional changes/ resilience in the system is needed)
Measures can start in your first tests of a change as ‘something did or did not’ happen (attribution Yes/ No) and later be a % of all the possible cases in which it should have happened (82% of all patients on the last shift had NEWS fully done as expected). This is a variable measure that is then plotted on a run chart initially and then once 12 or more points can become a control chart.
Reporting and publishing your project:
- Review the SQUIRE guidelines early in your project so you know what information to collect.
- Review the Bristol Patient Safety Conference guidelines from judge.
- Review some examples of projects that have been published before