Transcript Why measure? - WHA Quality Center

WHA Improvement Forum
For July

“Data Driven Improvement”
 
Presented by Stephanie Sobczak
Courtesy Reminders:
•Please place your phones on MUTE unless you are speaking (or use *6 on your keypad)
•Please do not take calls and place the phone on HOLD during the presentation.
Today’s Webinar
 Data Driven Decision Making
 Data Mining Your Processes
Agenda  From Data to Information
 “Real Time” Improvement
2
Why measure?
The main reason for conducting an
improvement project is to achieve results,
no matter the issue or topic.
And how do we know we have achieve a
desired result that can be proven to others?
We must demonstrate change from a
baseline, or initial measurement, and assess
the degree of change after an intervention.
3
Linking Measures to
Small Tests of Change
AIM: Improve M/S Unit HCAHPS Score for “Patients Always
received requested help” by 10 points by Nov 2013.
Small Test of Change: Anyone within 6 feet of a room will
answer that patient’s need.
Possible Process Measures:
• Track how many call lights occur between the hours of 10am -11am
and 1pm - 2pm
• Hallway observation of day shift staff response to call light by a
volunteer for 1 hour on 3 different days each week
How will we prove this change is effective?
Measurement Best Practices
 Measures are the proof of improvement
 Measures guide improvement by informing the
decisions about which changes to test
 Outcome and Process Data should be plotted
over time on annotated graphs.
 Process measurement should be integrated
into the team’s daily routine, and shared with
staff in ‘real time’.
 Really important skill set for staff to understand
5
Graph Your Data Over Time
6
Time Series Charts
Like the EKG of a process!
7
Data on the Surface
No Improvement over time…
WHY?
8
Taking a Deeper Dive –
Looking for Drivers
What are your Drivers for your Outcomes?
What does the evidence say?
Do you have data for these processes?
If not, can you collect some?
If yes, what does it tell you?
9
Connecting Outcome & Process Measure Data
4.00
3.50
3.00
2.50
2.00
1.50
1.00
0.50
0.00
Bedside protocol STOC started
Median
Baseline (Jan - Mar
2013)
Apr-13
May-13
4West CAUTI
Daily Review of Line Necessity
Process Measure
Jul-13
80%
60%
40%
Jun-13
Jul-13
Reporting Timeframe
Aug-13
Key: Down is better
CAUTI STOC - Bedside Protocol for
Catheter Indication Results
100%
May-13
Jun-13
Reporting Timeframe
(Num/Den)*100
(Numerator/Denominator)*100
(Numerator/Denominator)*1000
4West CAUTI Infection Rates
Aug-13
Key: Up is Better
5.00
4.00
3.00
2.00
1.00
0.00
Jun-13
Jul-13
Aug-13
Reporting Timeframe
10
Data Mining Your Processes
11
6 months, no improvement
30 day readmissions
12
10
8
6
4
2
0
WHY? We do follow-up calls!
Baseline Jul-12 Aug-12 Sep-12 Oct-12 Nov-12 Dec-12 Jan-13
30 day readmissions
12
Follow- up Calls – quarterly data
Readmissions and F/u calls
50
40
30
20
Need to work on improving the number
of calls made
10
0
Baseline Jul-12 Aug-12 Sep-12 Oct-12 Nov-12 Dec-12 Jan-13
30 day readmissions
Follow-up Phone calls
13
Dive deeper into each STOC results
• How soon do you see relationships between
outcome – process – small test of change?
• Ex: Follow-up Calls
– Was the plan executed well?
– Was it the wrong plan?
– Were the results not sustained over time?
– Is more effort required?
14
How to segment data
• Start with theories on ‘why’?
• Dive deeper to demographics such as
days/night, age, Dx, unit, etc.
• Spilt graph into multiple graphs to see driving
forces
Example: Pts. discharged with appointments
15
Age Breakdown
Under 25
5%
Patient Type
25 to 40
9%
80 and
Over
29%
40-65
16%
Cardio
14%
Medical
34%
Neuro
13%
Peds
10%
65 to 80
41%
Surgical
29%
Evidence  INTERACT adoption in LTC
and/or Care Transitions Coaches
16
Some positive trend - is it related?
Transfers to ED from Nursing Homes
60
50
40
30
20
10
0
Adopt INTERACT Toolkit in LTC?
30 day readmissions
Nursing Home to ED transfers
17
Improvement Plan
AIM: Reduce Readmissions by 50% for those
over 65 through implementing Care
Transitions Coaches
Process Measure: Patients receiving coaching
Plan: Implement in March, gather monthly data
18
New Care Transition Coaching Process
16
14
12
10
8
6
4
2
0
30 day readmissions
Pts receiving Care Transitions visits
19
Decisions
Purpose
Improve follow-up call
process
Prevent
Readmissions
INTERACT Toolkit
Adoption
Decrease Transfers
to ED
Care Transitions
Coaching Model
Assistance for older
adults to prevent
readmissions
20
30 day Readmissions & Improvement Effort
12
10
8
6
4
2
0
Jul-13
Jun-13
Start Care
Transitions
Coaching
May-13
Apr-13
Mar-13
Feb-13
Jan-13
Dec-12
Nov-12
Oct-12
Sep-12
Aug-12
Jul-12
Baseline
INTERACT
Initiative in
LTC
Aug-13
New F/u Call
Process
30 day readmissions
21
“Mining” your data
Other analysis:
•
•
•
Does our performance go down with
a higher census?
Is performance the same on every
shift?
Are there variations in the practices of
individuals?
Disclaimer information here…
22
Common Mistakes
• Only one person looks at the process measures
• Staff aren’t aware of the process measures and
how they are directly linked to improvement
• Processes are only measured for a short period
of time
• Processes aren’t measured at all
23
Turn Your Data to Information
•Visual displays of data can provide greater insights
into the systemic knowledge that lives in the data
•Turn “data” into “information”
•Data: raw facts
•Information: data that has been processed and
analyzed so that it is directly useful
•Visual displays of data highlight variation in the
system
•Systems thinking and understanding of variation is
essential for Improvement
24
Telling a Strong Story
25
The Importance of Data Display
• If data is not displayed or interpreted correctly,
incorrect assumptions can be made leading to poor
decisions
• Aggregated or data presented in tabular formats or
with summary statistics, will not help you measure
the impact of process improvement/redesign
efforts.
• Aggregated data can only lead to judgment, not to
improvement.
©Copyright 2012 IHI/R. Lloyd
26
Robut Data Displays Tell a Better Story
Identical Improvement
27
Using Data to Develop an
Improvement Plan
1. Which process do you want to improve or redesign?
2. Does the process contain non-random patterns or
special causes?
3. How do you plan on actually making improvements?
What strategies do you plan to follow to make things
better?
4. What effect (if any) did your plan have on the
process performance?
Run & Control Charts will help you answer questions 2 & 4.
YOU need to figure out the answers to questions 1 & 3.
28
Annotate Your Run Charts
29
Care Transitions Improvement
100
80
60
40
20
0
30 day readmissions
Nursing Home to ED transfers
Follow-up Phone calls
Pts receiving Care Transitions visits
30
Why Annotate?
•
•
•
•
Tells WHY something is changing
Helps show your Improvements work
Improves support/ ‘buy-in’
Documents what you have learned when
revisited in the future.
31
Keep visuals simple – and visible!
32
Provide Users will Real Time
Feedback
Don’t save the reports until meetings
• Staff will better understand improvement work
• Enables faster improvement
• Empowers staff to make their own improvements
• Better shows where the opportunities for
improvements live
33
Summary
• Look further than Outcome Measures
– Measure what you DO / what you TEST
• Create data displays that are simple but
informative
– Measure over time
• Provide real time feedback
– Easy to access, Meaningful measurement
34
Next Month:
Establishing an Accountable Culture
 The “Two Jobs” of Work
August 22  Who’s job is “accountability?”
Noon Strategies to build engagement
35
Thank You!
Questions
Please complete 3 question survey when closing
webinar window.
36