Numbers They Can t Ignore: Using Data to Drive Improvement

Size: px

Start display at page:

Download "Numbers They Can t Ignore: Using Data to Drive Improvement"

Charity George
5 years ago
Views:

1 Numbers They Can t Ignore: Using Data to Drive Improvement Brian Schnarch, Director of System Planning and Analysis Champlain System Improvement Workshop June 20, 2016

2 Of Course, There s More to Improving Performance Than Data Perceived importance, prioritization Passionate leadership Patient demand/expectation Accountability, Direction Incentives/disincentives Change management Resources Effectiveness of available levers Not so simple! Action Decision Advice Knowledge Information Data 2

3 That Said, Data is Crucial In God we trust. All others must bring data. W. Edwards Deming What gets measured gets managed Peter Drucker War is ninety percent information. Napoleon Bonaparte Information is the oil of the 21st century, and analytics is the combustion engine Peter Sondergaard 3

4 First- Get The Right Metrics All potential Indicators Champlain LHIN Scorecard Filters Ministry Direction Alignment w/ LHIN Mandate, Role, Priorities Good data available regularly Valid, evidence-based indicator Important to patients Makes sense to providers Levers available (or will be) 4 No particular order. All filters must be applied. 4

5 The Right Metrics- Aligned to Directions and Priorities Champlain LHIN Scorecard Indicators Strategic Direction Action Indicator Strat Dir Integration LHIN Organizational Health 5 Integrate community and home care services Evolve primary care neworks Integrate MH & addictions services Operations Personal support services visit within 5 days* Nursing visit within 5 days* Home Care CCAC Visit Wait Community Clients* Patients in acute hospital beds needing other care (%ALC)* Patients in acute/sub-acute needing other care (ALC rate)* Readmissions for certain chronic conds Fall-related ED visit rate among seniors ED visits for conditions best managed elsewhere Physician visit within 7 days of discharge Satisfaction with Community Care Repeat mental health emergency dept visits* Repeat substance abuse emergency depart visits* Status of LHIN Annual Business Plan Initiatives LHIN Enterprise Risk Assessment LHIN Operational Budget Variance LHIN Staff Turnover Access Sustainability Action Provide for culturally and linguistically appropriate care Implement strategies to achieve performance targets Expand enabling tech to bring care close to home Continue implementing funding reform & innovative care models Enhance palliative care in settings of choice Grow Health Links Indicator Health Service Staff Trained in Aboriginal Cultural Competancy % of health service providers designated and identified for French language services Time in the ED, complex pts* Time in the ED, non-complex pts* % of MRI scans completed within target* % of CT scans completed within target* % of hip replacements completed within target* % of knee replacements completed within target* Telemedicine Visit Rate Hospital cost efficiency CCAC home care cost efficiency Working Capital (Current Ratio) - Hospitals CCAC Palliative Patients Who Die in Place of Choice Health Links Patients with a Coordinated Care Plan *Ministry LHIN Accountability Agreement Indicators (MLAA) 5

6 What if they don t care or don t get it? 6

Make it Scream When it Should MRI Wait Time* by Qtr, Champlain Stocktake (Q4 10/11-Q1 11/12) 250 200 MRI Wait Time by LHIN and Month (June 2010-June 2012) 196 194 197 222 218 222

150 146 HNHB Central West 115 117 116 116 119 112 110 104 Mississauga Halton Toronto Central 100 79 92 101 106 106 99 104 Central Central East South East Champlain N Simc Muskoka

7 Make it Scream When it Should MRI Wait Time* by Qtr, Champlain Stocktake (Q4 10/11-Q1 11/12) MRI Wait Time by LHIN and Month (June 2010-June 2012) Erie St.Clair South West Waterloo Well HNHB Central West Mississauga Halton Toronto Central Central Central East South East Champlain N Simc Muskoka North East 50 North West 0 Jun 10 Jul 10 *90 th percentile (days) for completed P2, P3 and P4 adult cases, excluding timed procedures 7 Aug 10 Sep 10 Oct 10 Nov 10 Dec 10 Jan 11 Feb 11 Mar 11 Apr 11 May Jun Jul Aug 11 Sep 11 Oct 11 Nov 11 Dec 11 Jan 12 Feb 12 Mar 12 Apr 12 May Jun Jul

8 Make it Simple and Clear (Even Technical People Like That) 8 8

9 Make it Colourful & Colour-Coded 9 9

10 But Not An Art Project! 1 10

11 Graph Design Matters Generally avoid Chart junk 3-D charts Pie charts Stacked and group bar charts Generally aim for High data-to-ink ratio Data that stands out Proper labels and notes Careful choice of scale, graph type, colours Maybe a table? Lots of tips online (e.g. Stephen Few Blog) 11 1

12 Tell It Show it to different people Show it in different ways Talk about it Write about it Tell the story Repeat, repeat Numbers have an important story to tell. They rely on you to give them a clear and convincing voice Stephen Few 12

13 But Don t Tell It by Yourself Statistics are no substitute for judgment Henry Clay Engage those most knowledgeable and most impacted to interpret the data Multiple perspectives required for the full picture 1 13

Give The Right Type of Information to Each Audience 1

w/in target Apr -Sept 2013 Wait list Apr 1-Sep 30 2013

D 105 98% 98% -10 E 102 88% 100% -6 G 92 93% 100% -16 A

100% -1 R 44 100% 100% 0 I 38 55% 94% -2 L 34 65% 53% 8 Q

14 Give The Right Type of Information to Each Audience 1 Volume (Oct 12- Sept 13) % w/in target Oct 12-Mar 13 % w/in target Apr -Sept 2013 Wait list Apr 1-Sep Doc B % 12% -59 F % 100% 8 O % 9% -41 D % 98% -10 E % 100% -6 G 92 93% 100% -16 A % 93% 0 C % 100% 1 J 69 91% 100% 0 K % 100% -1 R % 100% 0 I 38 55% 94% -2 L 34 65% 53% 8 Q % 92% 7 N 27 71% 90% -14 T % 71% -3 U 10 75% 50% 14 6

15 What if they shoot down the data? (They Will) Acknowledge, assess and address the issues Work to improve data if possible Keep using it unless it is a total mess or there is a better option Use multiple measures if possible Errors using inadequate data are much less than those using no data at all. Charles Babbage 15

16 We ve Nailed the Problem. What about the solution? If you torture the data long enough, it will confess. Ronald Coase 16

Disentangle the Drivers and (try to) Quantify the Impacts Surgery Wait Time Example Appropriate Supply Funding Human resources Education OR hours Beds Capacity He who would search for pearls must

17 Disentangle the Drivers and (try to) Quantify the Impacts Surgery Wait Time Example Appropriate Supply Funding Human resources Education OR hours Beds Capacity He who would search for pearls must dive below John Dryden Intake and Assessment Demand Efficiency Inappropriate Data Quality Intake and Assessment Throughput Cancellations Scheduling Distribution of resources Distribution of volumes 17

18 Develop Strategies Based on the Drivers and Bang for the Buck Champlain MRI Wait Time Example (May 2016) Driver Strategy Cost relative to expected gain Distribution, scheduling, Central intake and triage $$ queue management Demand Demand Promote appropriate scan ordering primary care? Demand Promote appropriate scan ordering- hospital $ Efficiency Improve scheduling to optimize slot utilization $ Supply Fund additional scans $$$$ 18

19 With the data right, you can adjust course and get to your destination 19

Repeat C-Sections (Q3 2015-16) Shorter Wait

20 And Tout the Good News When You Get There Best in Province: Early Elective Low-Risk Repeat C-Sections (Q ) Shorter Wait Times for Hip and Knee Replacements (Feb 2015) 20

21 Thank you Brian Schnarch Director of Performance and Analysis Champlain LHIN All images either paid for or royalty free. 21

MINISTRY/LHIN ACCOUNTABILITY AGREEMENT (MLAA) MLAA Performance Assessment Dashboard /10 Q3

MINISTRY/LHIN ACCOUNTABILITY AGREEMENT (MLAA) MLAA Performance Assessment Dashboard - 29/1 Q3 README The 29/1 MLAA Dashboard has been designed to reflect various reporting fiscal periods as well as the