Introduction to Improvement

The First Law of Improvement “Every system is perfectly designed to achieve exactly the results it gets.” Dr Paul Bataldan

Working harder will not allow us to make the change.

All improvement requires change but not all change will result in an improvement Langley et al, 2009 The Improvement Guide

Applied Science Approach The Typical Approach Applied Science Approach Traditionally we have not been good at making change for improvement reliable – because we have not tested whether it will be delivered appropriately in various conditions and environments before we expect all to implement and make part of every day business. e.g. previous experience of launching new protocols/guidelines – as of Mon am….. expect all to comply fully without really knowing that it’s practical for every system – then we wonder why people don’t comply and the system is unreliable. Reinertsen JL, Bisognano M, Pugh MD. Seven Leadership Leverage Points for Organization-Level Improvement in Health Care (Second Edition). Cambridge, Massachusetts: Institute for Healthcare Improvement; 2008. Available: www.ihi.org p26

The Model for Improvement ‘This model is not magic, but it is probably the most useful single framework I have encountered in twenty years of my own work on quality improvement’ Dr Donald M. Berwick Former Administrator of the Centres for Medicare & Medicaid Services Professor of Paediatrics and Health Care Policy at the Harvard Medical School MFI is made up of 2 parts – the thinking part and the doing part The 3 questions – the thinking part prepares you for the test by asking the ‘so what’ question They need to be thought through, answered and written down by the improver to ensure that they, and therefore their colleagues, are clear on the benefits to them and their work. That message often gets lost in the many things we are expected and asked to do and is one of the reasons why so many interventions are not implemented reliably. If we are not clear on the benefits of something we are much less likely to do it. The Plan Do Study Act cycle is a simple tool that’s used to test out ideas that will improve your systems and processes while learning what works and doesn’t work. It’s a structured approach for making small incremental changes to systems that allow you to test on a small scale, learn from successes and failure and redesign as you go It builds learning and buy in incrementally as tests ramp up and allows clinicians knowledge about their own system is built in to the process

QI Key Factors for Quality Improvement Will Execution Ideas Having the Will (desire) to change the current state to one that is better Having the capacity to apply CQI theories, tools and techniques that enable the Execution of the ideas Developing Ideas that will contribute to making processes and outcome better Will QI Execution Ideas

History – Where has it come from? W.E.Deming (1900-1993) American statistician, author, lecturer, professor Biggest impact on Japanese manufacturing and business 2 types of knowledge – subject matter and profound knowledge Developed the Plan-Do-Study-Act cycle

Model for Improvement Act on the plan - Testing

Aim Measures Changes Implementation The Improvement Guide, API 3 questions and a learning cycle 4 basic aspects: aims, measures, changes and execution The Improvement Guide, API 11 11

Question 1: What are we trying to accomplish? Developing the team’s Aim Statement Do some more work on the thinking part.... First question – the aim!

Check points - developing an Aim Statement AIM Content Explicit over arching description Specific actions or focus Unachievable by hard work alone AIM Characteristics Measurable (How good?) Time specific (By when?) Define participants and customers 13 13

EYC Example Aim Statements To reduce infant mortality, i.e. deaths by 12 months of age (including stillbirths) by 15% by end-2015. 90% of children will receive a bedtime story 3 times a week by the end of the summer term The EYC will be the first agenda item at each CPP meeting by January 2014

Exercise Think about if you were challenged to do a marathon….. Identify an aim (what do you want to achieve, how good, by when?)

Question 2: How do we know that a change is an improvement? Improvement is not just about measurement However… without measurement you will never be able to answer the question!

How will we know a change is an improvement? Measurement of Improvement Work Define measures that will measure the impact of the Improvement work They will monitor and guide your progress This may take a number of different approaches: Percent compliance A count of correct attempts Verbal feedback

This is what you need for a PDSA measure – nothing more

Measures Exercise: Think about if you were challenged to do a marathon… Identify measures (how will you know?)

‘Change’ can mean different things Model for Improvement Q3 What changes can we make that will result in improvement? ‘Change’ can mean different things Change a light bulb, change a nappy, change your car……. reactions to wear and tear or needing replaced - sets things back to where they were Not necessarily a source of improvement Improvement is innovative change for the better

Selecting Changes Copy: use the literature, experience of others, hunches and theories Be strategic: set priorities based on the aim, known problems and feasibility Avoid low impact changes Steal shamelessly and learn from others Use the Driver Diagrams

Innovative thinking....! QI is about making changes that result in lasting improvement not just making changes because something needs replaced or has run its course. Thinking creatively and bringing about change can be a challenge – we are often follow the accepted norms and not try out new ways of doing things. In some organisations change is a challenging concept where tradition and accepted norms are quietly tolerated, even though we may have a good idea that will make a positive change for the future with more effectiveness and or efficiency.

A Practical Need Often Drives Creativity! 23

Change Ideas Exercise: Think about if you were challenged to do a marathon… Identify changes (what changes can you make?)

Why Test Changes? To increase the belief that the change will result in improvements in your setting To learn how to adapt the change to conditions in your setting To evaluate the costs and “side-effects” of changes Overall to minimise the resistance when spreading the change throughout the organisation.

 Do  Study  Act  Plan  Do  Study  Act Northwest Improvement Initiative “What will happen if we try something different?” “What’s next? ” Four parts of the cycle: Plan: Decide what change you will make, who will do it, and when it will be done. Formulate an hypothesis about what you think will happen when you try the change. What do you expect will happen? Identify data that you can collect (either quantitative or qualitative) that will allow you to evaluate the result of the test. Do: Carry out the change. Study: Make sure that you leave time for reflection about your test. Use the data and the experience of those carrying out the test to discuss what happened. Did you get the results you expected? If not, why not? Did anything unexpected happen during the test? Act: Given what you learned during the test, what will your next test be? Will you make refinements to the change? Abandon it? Keep the change and try it on a larger scale? “Did it work?” “Let’s try it!” © R. Scoville & I.H.I.

Cycles of Tests Build Confidence Northwest Improvement Initiative Cycles of Tests Build Confidence Changes that will result in improvement Learning from data Working our way up to implementation is a matter of building belief in the effectiveness of a change. First tests are small scale, under optimal conditions: changes that fail here are discarded quickly. If they succeed, we explore them in more detail: by replication under other circumstances, with different actors, in different locales, etc. If the change looks robust, it will gather the approval it will need for implementation. Along the way, we’re building confidence in the change. Proposals, theories, hunches, intuition © R. Scoville & I.H.I. 31

PDSA Testing - simulation exercise The Skittle Challenge

The Skittle Challenge DO NOT EAT the Skittles – yet… Aim – to be left with one Skittle Measure – number of Skittles left Changes – which one to remove first? – what order to remove them in? – how you work as a team? Execution Put a Skittle on each circle Remove one to start Jump over one at a time and remove it Keep going until you can’t jump over any more How many are left? (Plot your data and annotations) Think about how it went and what you could improve next round. (Theory and prediction based on learning) DO NOT EAT the Skittles – yet…

Number of Skittles Left PDSA # Theory Prediction 1 Start with No. 5 empty Will have 3 skittles left 2 3 4 5 6 Number of Skittles Left 7 Number left 6 5 4 3 2 1 PDSA

Run Chart to Measure Performance D O W N is G Test 2 - XYZABC Test 2 123456 Test 1 ABCXYZ Tests 3-6 987654

The Skittle Challenge Consistency of purpose Prediction Planning Documentation Sharing and stealing! Learning from failure

A solution…. 6-1, 4-6, 1-4, 7-2, 13-4, 2-7, 11-4, 14-5, 10-3, 3-8, 4-13, 12-14, 15-13.

Measurement and Data Collection During PDSA Cycles Collect useful data, not perfect data – data for learning, not evaluation Use a pencil and paper until the information system is ready Use sampling as part of the plan to collect the data to reduce workload Use qualitative data (feedback) rather than wait for quantitative data Record what went well and what didn’t work so well during the test of change

Start Small 1 client 1 day 1 case 1 worker Move to 3,5,7…. as confidence grows

Drop down next “two levels” to plan test cycle! Smaller Scale Tests: Shrink It! rapid cycles of learning Years Quarters Months Weeks Days Hours Minutes Drop down next “two levels” to plan test cycle!

Sequence for improvement Make part of routine operations Sustain & spread a change to other locations Test under a variety of conditions Implement a change Theory & prediction Test a change Study Act Plan Do Support with data and consideration of people Develop a change Based on slide © IHI 2009

PDSA Cycles in EYC Testing approaches to increase attendance at nursery - children with >60% attendance rate Testing out use of a Safety Brief huddle for staff at a multidisciplinary ‘one stop’ substance misuse clinic Testing approaches to establish robust recording and sharing of information in relation to gender based violence cases – Health Visitors and Social Workers Testing an approach to increase uptake of healthy start vitamins