1. Quality Improvement Try it – you might like it! Dr Emma Donaldson @E_arnotsmith Emma.donaldson@srft.nhs.uk

2. What is Quality Improvement? Research: –Provides the evidence –Tells us the RIGHT THING TO DO Quality Improvement: –Helps us develop systems to deliver care –Ensures we are DOING THE THING RIGHT Audit: –Provides assurance of excellent care –CONFIRMS WE ARE DOING THE RIGHT THING, RIGHT

3. Every system is perfectly designed to get the results that it gets

5. Model for Improvement A coat hanger for improvement projects!

6. Model for Improvement

7. Start with the aim… The solutions come last Let solutions come from the team If you have a good idea, the team will have it too (then they’ll own it!) If they have a bad idea, let them test it

8. A Good Aim S pecific M easurable A chievable R elevant T ime limited Stre-e-e-e-e-e-e-e-e-e-etch

9. Aim statements 1 – provide adequate pain control to all patients. 2 – 90% of patients who report that they had pain will respond “yes” to “did staff do all they could to control your pain” by June 2012

10. Types of Aim Statement Absolute: 95% of eligible patients should achieve all the measures in the acute stroke care bundle by………… Relative: 50% reduction in delays starting the morning operating list by…………..

11. “The greatest danger for most of us is not that our aim is too high and we miss it, but that it is too low and we reach it.” Michelangelo

12. A Driver Diagram Reinforces the aim statement as the goal Clarifies the big picture Identifies primary system components Aids in development of measurement Most importantly: Helps teams to articulate their contribution to the overall aim and avoid missing important system components

13. 50% reduction in acute central line infections in ICU, MHDU and Renal (G3/Renal Unit) by June 2009 Nominate 2 clinical leads from your ward Introduce systems for: competency training quality assurance encouraging reporting for learning Leadership, Governance & Staff Education Process Standardisation Patient Involvement Introduce system of assessment for most appropriate line Paired insertion of central lines mandatory throughout Mandatory use of care pathways Daily review for removal Consider recruiting patient champion Introduce system of appropriate communications Involve patients in early identification of infection Measurement Present data in ward area Introduce reporting system Celebrate success Develop system for measuring catheter days Central Lines

14. Limitations of drivers Not a perfect science Will require ongoing amendment Interplay between drivers Contribution of each driver is unlikely to be equally distributed

15. Model for Improvement

16. Stroke Measures Aim: To achieve a score of 95% on the Sentinel audit by October 2008 Outcome Measures –Audit score –The mortality rate of stroke patients Process Measures –The % of patients receiving a brain scan within 24hrs Balancing Measures –Time spent by ward staff completing forms –Wait time for other patients requiring brain scan

17. The 3 reasons for measurement Source: Robert Lloyd IHI 2006

18. BeforeAFTER The project Did we achieve anything? Are things better?

19. Ways to display data: Static View…

20. Ways to display data: Dynamic View…

21. Run Charts Viewing TIME ORDERED DATA is a powerful way of detecting change It can tells us when a real change has occurred The pattern contains additional useful information

22. Average Before=8 hours delay Average After=3 hours delay DG 1-11/12

23. So in Quality Improvement we are concerned with plotting data over time in order to understand variation in processes “If I had to reduce my message for management to just a few words, I’d say it all had to do with reducing variation” W.E.Deming

24. Types of Variation Common Cause Is due to natural and regular causes Results in a ‘stable’ process Also known as random variation Special Cause Is due to irregular or unnatural causes that are not inherent to the process Results in an ‘unstable’ process that is not predictable Also know as non-random variation

25. Understanding variation The outcome of every process is affected by lots of little things Each of these little things varies naturally All these little variances add up This makes the process vary over time

26. Min Patients

27. Min

28. Common Cause Variation Min

30. A system can also be affected by a big, unusual influence The size of the change produced is BIG in relation to the common cause variances It happens much less frequently than the common cause variances

31. Patients Min

32. Special Cause Variation Patients Min

34. Statistical Process Control Engineering science uses a robust approach to detect deviations from the usual pattern This can tell you if you have really achieved an improvement, or if a stable process has deteriorated

35. 8 datapoints above the baseline mean = special cause variation

36. Examples – Length of Stay

38. “All improvement will require change, but not all change will result in improvement”

39. Why test change before implementing it? It involves less time, money and risk The process is a powerful tool for learning; from both ideas that work and those that don't It is safer and less disruptive for patients and staff Because people have been involved in testing and developing the ideas, there is often less resistance

40. Hunch Workable solution

41. The PDSA Cycle Plan What are you going to test? What do you predict will happen? Develop the test (Who? What? When? Where? Data?) Do Try out the test on a small scale Observe & document results Study Analyse data Study the results Compare results & predictions Act What will you do next? Adapt Adopt Abandon

42. You already do this every day!! Int J STD AIDS. 2010 Jul;21(7):521-3

43. Successful PDSA Cycles Think ahead Small scale Predict Test with willing staff Don’t ask permission or for consensus Data and documentation

44. What PDSAs Are Not… A radical change to a system /process Full blown trust-wide implementation Mini projects Top down directives ‘PDSA’s ‘test’ a proposed change

45. PDSA principles Initial ideas usually don’t work If a PDSA “fails”, then the idea would not work reliably But lots can be learnt during the process

46. Conclusion - 1 Always start with the aim Spend time working out measures Gather information, set up measurement system Drivers, cause effects, theory of change Only then solutions

47. Conclusion - 2 Test solutions with PDSA cycles Monitor effect with run charts Start small, very small

48. Thank you.