1. What is Your Data Telling You? Clara Ballantine, SIA Ontario Falls VLC Team Call January 11, 2011

2. Objectives In the next 30 Minutes we will:  Differentiate among data for accountability, for research and for improvement  Review how to effectively display Falls data using a run chart  Analyze your run chart data “by the rules”  Consider strategies to share your data effectively with others

3. Data for Improvement, Accountability and Research Source: The Data Guide: Learning from Data to Improve Healthcare. Developed from Solberg, Leif I., Mosser, Gordon and McDonald, Susan. “The Three Faces of Performance Measurement: Improvement, Accountability and Research.” Journal on Quality Improvement. March 1997, Vol.23, No. 3.

4. What Do We Need to Know?  How much variation do we have?  Is this process changing significantly over time?  Have our changes resulted in improvement?  Are we holding the improvement?

5. Run Chart A line graph of data plotted over time Data is kept in time order Can see flow of data Helps answer our improvement questions

6. Made Change Why Use a Run Chart?

8. 8 Run Chart – Falls 4.0 Goal Baseline1MD, 1RN trained Staff Training complete Team Leader resigned New staff started Thanks to: Virginia Flintoft SHN Central Measurement Team

9. Source: National Nursing Home Improvement Collaborative: Pressure Ulcer Prevention and Treatment Handbook, Qualis Health

10. Key Point #1 Annotate your run chart – it tells your story Annotation helps you see the impact of your test of change and can help you decide to “adopt, adapt or abandon” a change.

11. How to Annotate a Run Chart 11 1. Select “Insert” tab 2. Select “Text Box”

12. How to Annotate a Run Chart 12 3. Enter text in Text Box 4. Right click on Text Box and select “Format Shape “ from menu.

13. How to Annotate a Run Chart 13 5. Format text box – use your imagination! fill colour line colour and style etc.

14. % Timely Reperfusion Months percent 1/07234567891011121/08234565789101112 3223323835 402138262227233236293842393650483944 Date Data Run Chart 1/07 23456789 101112 1/08 2345 65 789 101112 15 20 25 30 35 40 45 50 55 60 Change 1 Chg 2,3 Chg 4,5,6 Chg 7 Chg 8,9 Median 35

15. MEDIAN In a series of numbers, the median is physically the middle number. It has the same number of points equal to it or above it as it has equal to it or below it.

16. Finding the Median: Reordering the Data 50 48 44 42 40 39 38 36 35 32 29 27 26 23 22 21 To find the median reorder the numbers from high to low and find the number physically in the middle. If you have two numbers left in the middle, add them together and divide by two. Excel: place cursor in blank cell and type=MEDIAN(A2:A21) where A2 is the first cell you want to include and A21 the last)

17. Key Point # 2 Calculate and show a Median line to see trends and a Goal line Analyze your data “by the rules”

18. Heart Failure Readmission Rates* Source: Residents Satisfaction on Discharge Hand-Off, Peg Bradke, St. Luke’s Hospital, Cedar Rapids, Iowa, Hospital to Home: Optimizing the Transition January 2009 *Percent of heart failure Residents readmitted for exacerbation of their heart failure. Good Aug 06 = Implemented use of new Residents education materials Jan 07 = Initiated complimentary visits

19. Rule 1 JSix or more consecutive POINTS either all above or all below the median. Skip values on the median and continue counting points. Values on the median DO NOT make or break a shift. Median=10Median=11 Median 10 Ott, Ellis, Process Quality Control, McGraw-Hill Book Company, NY, 1975

20. Source: National Nursing Home Improvement Collaborative: Pressure Ulcer Prevention and Treatment Handbook, Qualis Health

21. Rule 2 J Five points all going up or all going down. If the value of two or more successive points is the same count the first one then ignore the identical points when counting; like values do not make or break a trend. Median 11 Olmstead, Pl, “Distribution of Sample Arrangements for Runs Up and Down, Annals of Mathematical Statistics, Vol 17, pp. 24- 33, March, 1945

22. Rule 3 To Determine The Number of Runs Above and Below the Median: – A run is a series of points in a row on one side of the median. Some points fall right on the median, which makes it hard to decide which run these points belong to. – So, an easy way to determine the number of runs is to count the number of times the data line crosses the median and add one. – Statistically significant change signaled by too few or too many runs. Median 11.4 Data line crosses once Too few runs: total 2 runs

23. Rule 3: # of Runs Table is based on about a 5% risk of failing the run test for random patterns of data. Frieda S. Swed and Churchill Eisenhart, (1943). “Tables for Testing Randomness of Grouping in a Sequence of Alternatives. Annals of Mathematical Statistics. Vol. XIV, pp.66 and 87, Tables II and III

24. RULE 4 For detecting unusually large or small numbers: Data that is Blatantly Obvious as a different value Everyone studying the chart agrees that it is unusual Remember: – Every data set will have a high and a low - this does not mean the high or low are astronomical

25. What do “They” Want to Know? Senior leaders:  How are we doing?  How do we compare with others? Falls project team:  What is the impact of our work?  Where do we focus our efforts? Staff  How is this helping our patients/residents/clients?  What can we do?

26. Engage Senior Leaders With Data and Evidence The "whiskers" depict the 95th% CI of the National Mean. Your mean is statisically higher if it is above the "whiskers"; it is the same as the National Mean if it is within the "whiskers"; and it is statistically lower if it is below the "whiskers".

27. Bring Data to Improvement Teams Frequently: Process 1MD, 1RN trained Baseline Staff Training complete Team Leader resigned New staff started Goal Median

28. Bring Data to Improvement Teams Frequently: Outcome

29. Post Process Measure Data for All Staff to See Progress Median

30. Post Process Measure Data for All to See Opportunities for Improvement

31. Key Point #3 Share your data effectively with others at all levels of your organization Consider your audience when choosing which data and how to display

32. For Help With Your Data and Measurement Questions Virginia FlintoftClara Ballantine 416.946.8350613.736.9142 virginia.flintoft@utoronto.cavirginia.flintoft@utoronto.ca clara.ballantine@qhn.caclara.ballantine@qhn.ca Dannie Currie Alexandru Titeu 902.842.0716 416.946.3103 curried@cbdha.nshealth.cacurried@cbdha.nshealth.ca shn.ea@utoronto.cashn.ea@utoronto.ca Chantal Bellerose 514.340.8222 ext 4901 cbellerose@jgh.mcgill.ca