Scottish Learning Festival The Early Years Collaborative and the Model for Improvement
The Early Years Collaborative
description
Transcript of The Early Years Collaborative
Components of a Learning System
David M. Williams, Ph.D.Improvement Advisor
Institute for Healthcare Improvement
D
D
The Early Years Collaborative • “Scotland will be the best place to grow up.”• Draws upon existing community infrastructure across the country• Focused on age-based workstreams:
• 15% reduction in the rates of still-births and infant mortality by 2015.
• 85% of all children within each CPP will reached all of the expected developmental milestones at the time of the child’s 27-30 month child health review, by end-2016.
• 90% of all children within each CPP have reached all of the expected developmental milestones at the time the child starts primary school, by end-2017.
5
D
Components of a Learning System 1. System level measures2. Explicit theory or rationale for system changes3. Segmentation of the population4. Learn by testing changes sequentially5. Use informative cases: “Act for the individual learn for
the population” 6. Learning during scale-up and spread with a production
plan to go to scale7. Periodic review8. People to manage and oversee the learning system
From Tom Nolan PhD, IHI
Workstream 1 AimTo ensure that women experience positive
pregnancies which result in the birth of morehealthy babies as evidenced by a reduction of
15% in the rates of stillbirths (from 4.9 per 1,000 births in 2010 to 4.3 per 1,000
births in 2015) and infant mortality
(from 3.7 per 1,000 live births in 2010 to 3.1 per 1,000 live births in 2015).
Workstream 2 AimTo ensure that 85% of all children within
each Community Planning Partnership havereached all of the expected developmentalmilestones at the time of the child’s 27 30‐month child health review, by end 2016.‐
Workstream 3 AimTo ensure that 90% of all children within
each Community Planning Partnership havereached all of the expected developmental
milestones at the time the child starts primary school, by end 2017.‐
Source: Brandon Bennett, IA
Big Aims
We are here!
Overall Project Measures vs. PDSA Cycle Measures
AchievingAim
Data for Project Measures: - Overall results related to the project aim (outcome, process, and balancing measures) for the life of the project
Adapting ChangesDuringPDSA Cycles
Data for PDSA Measures: - Just enough data- Quantitative data on the impact of a particular change - Qualitative data to help refine the change - Subsets or stratification of project measures for particular patients or providers- Collect only during cycles
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
Point Number
Pou
nds
of R
ed B
ag W
aste
3.25
3.50
3.75
4.00
4.25
4.50
4.75
5.00
5.25
5.50
5.75
6.00
Median=4.610
Mea
sure
Time
Run Chart - Data for Learning & Improvement
PDSA Measures Guide Learning about our testing.
Process Measures Guide Learning about how our testing is improving reliability of the process.
Outcome Measures Guide Learning about how the reliability of the process is achieving our aim.
Components of a Learning System 1. System level measures2. Explicit theory or rationale for system changes3. Segmentation of the population4. Learn by testing changes sequentially5. Use informative cases: “Act for the individual learn for
the population” 6. Learning during scale-up and spread with a production
plan to go to scale7. Periodic review8. People to manage and oversee the learning system
From Tom Nolan PhD, IHI
Early Years Collaborative
2 day LSExpert
meetingsCluster
meetingsSupports
Expert QI & Early Years faculty
Networking events
Listserv Site Visits
Phone conf Assessments
Monthly Reports via web
2 dayLS
A
P
D
S
A D
P
S
LS1 2 day Kickoff
D
S
P
A
2 dayLS
Key ChangesImprovementMeasures
May 2013 TBC Oct 2013Jan 2013
Pre-work
Oct 2012
Etc…
TBC
R
Reduce infant mortality
Societal Issues
Post-birth actions
Pre-birth actions
Smoking / Alcohol & Drug Misuse
Improved sharing of information
Improved management and quality of care
Improved joint working
Improved identification
Aim 1⁰ 2⁰
Theory of what actions will reduce infant mortality
Health
Improved rate of breastfed babies
Quicker diagnoses of Neonatal Abstinence Syndrome
Nutrition
Mental health & wellbeing
Attachment Improved stability / permanence for LAC
Improved money management
Improved access
Improved family centred response
Improved leadership, culture and planning
Improved teamwork, communication, skills and collaboration
Parenting skills
Identification & reasons for current resilience
WORKSTREAM 1 (conception to 1 year)
Theory of what drives infant mortality
Poverty
Quality Of Home Environment
Domestic Abuse & Violence
Workforce Issues
Transport, Community Capacity & Cultures
Access To Services
EmploymentDetailed aim:To ensure that women experience positive pregnancies which result in the birth of more healthy babies as evidenced by a reduction of 15% in the rates of:• stillbirths (from 4.9
per 1000 births in 2010 to 4.3 per 1000 births in 2015)
• infant mortality (from 3.7 per 1000 live births in 2010 to 3.1 per 1000 live births in 2015)
Version: 06/03/2013
R
Children have all the developmental skills and abilities expected of a 27-30 month old
Societal Issues
Child’s physical & mental health and emotional development
Carer’s physical & mental health and skills
Level of education Improved sharing of information
Improved joint working
Improved early identification
Detailed Aim:85% of all children within each CPP have reached all of the expected developmental milestones at the time of the child’s 27-30 month child health review by end-2016
Misuse of alcohol & drugs
Health
Improved child’s dental health
Improving child nutrition
Poverty
Quality Of Home Environment
Domestic Abuse & Violence
NutritionDisabilities & Mental health
Attachment
Improved stability / permanence for LAC
Improved money management
Improved management, planning and quality of services
Improved family centred response
Improved leadership, culture and planning
Improved teamwork, communication, skills and collaboration
Improving brain development and physical play
Parenting skills & knowledge
WORKSTREAM 2 (1 year to 30 months)
Identification & reasons for current resilience
Aim 1⁰ 2⁰
Theory of what drives developmental milestonesTheory of what actions will
ensure developmental milestones are reached
Workforce Issues
Transport, Community Capacity & Cultures
Early Learning & Play
Access To Services
Employment
Additional Support
Version: 06/03/2013
R
Children have all the developmental skills and abilities expected at the start of primary school
Societal Issues
Child’s physical & mental health and emotional development
Improved sharing of information
Improved management, planning and quality of services
Improved joint working
Improved identification
Detailed Aim:90% of all children within each CPP have reached all of the expected developmental milestones at the time the child starts primary school, by end-2017
Health
Improved child’s dental health
Improving child nutrition
Attachment
Improved stability / permanence for LAC
Improved uptake of benefits
Improved family centred response
Improved leadership, culture & planning`
Improved teamwork, communication and collaboration
Improving brain development and physical play
Early Learning & Play
WORKSTREAM 3 (30 months to start of primary school)
Identification & reasons for current resilience
Aim 1⁰ 2⁰
Theory of what drives developmental milestones
Theory of what actions will ensure developmental
milestones are reached at the start of primary school
Poverty
Quality Of Home Environment
Domestic Abuse & Violence
Workforce Issues
Transport, Community Capacity & Cultures
Access To Services
Employment
Carer’s physical & mental health and skills
Level of education
Misuse of alcohol & drugs
NutritionDisabilities & Mental healthParenting skills & knowledge
Additional Support
Version: 06/03/2013
R
Provide the Leadership System to support quality improvement across the Early Years Collaborative
Early Years Collaborative is a strategic priority & underpins all policy planning and operational activity
Early Years Collaborative values, culture and behaviours are modelled by all leaders at all levels
Infrastructure to support delivery of Early Years Collaborative
Place quality issues at the top of senior leader meeting agendas
Ensure that the senior team participates in Walk-rounds
Display data that depicts progress towards aimDetailed Aim:
Timely delivery of all three workstream “stretch aims”
Early years executive and operational leads are identified
Leaders can describe how they personally maintain early years focus within their working environment
Ensure the development of a measurement system used to understand and drive quality indicators
Build commitment with partners to focus on delivery
CPPs communicate the EYC with enthusiasm and consistency
Leaders illustrate how users are included in design, improvement, and delivery of Early Years
Measurement plan and priorities are established and triangulation with other key data
Leaders demonstrate their ability to set direction and engage and mobilise staff to constantly improve quality of service
Meet regularly with the Implementation Committee to track progress and remove barriers
Ensure a feedback mechanism for issues raised in Walk-rounds
Establish Programme Management and remove barriers
Add Early Years Collaborative and outcomes to the CPP agenda
Establish an EYC Implementation Committee
Assign a senior leader to each improvement area (Workstreams 1-3 and measurement)
Spread plan is in place for core and innovative work
Leaders facilitate change by cultivating innovation from intelligence, insights and wisdom of people working together
WORKSTREAM 4 (Leadership)
Aim 1⁰ 2⁰Theory of what drives leadership support Theory of what actions will
ensure leadership support
Strategy for capturing, celebrating and spreading innovation
Version: 06/03/2013
R
Components of a Learning System 1. System level measures2. Explicit theory or rationale for system changes3. Segmentation of the population4. Learn by testing changes sequentially5. Use informative cases: “Act for the individual learn for
the population” 6. Learning during scale-up and spread with a production
plan to go to scale7. Periodic review8. People to manage and oversee the learning system
From Tom Nolan PhD, IHI
A population, for which claims data exists and achieving Triple Aim results will not result in perverse economic loss.
Cannot be defined by a clinical condition (Diabetes) or issue (readmissions).
An approach to achieving the IHI’s The Triple Aim for a given population - from the perspective of a consumer health plan-less, FFS based, Medicare participating, not for profit, hospital->health system…
Note: The size of the rectangles is meant to be indicative of population size, not cost.
A population, for which claims data exists and achieving Triple Aim results will not result in perverse economic loss.
A sub-population, high cost and or high utilization people from the larger population.
People who have “fallen through the cracks” of our “rescue-care” system.
Cannot be defined by a clinical condition (Diabetes) or issue (readmissions). Note: The size of the rectangles is meant to be indicative of
population size, not cost.
An approach to achieving the IHI’s The Triple Aim for a given population - from the perspective of a consumer health plan-less, FFS based, Medicare participating, not for profit, hospital->health system…
A population, for which claims data exists and achieving Triple Aim results will not result in perverse economic loss.
A sub-population, high cost and or high utilization people from the larger population.
Sub-groups, people from the high cost high utilization sub-population that can be stratified based upon relatively similar needs.
People who have “fallen through the cracks” of our “rescue-care” system.
Cannot be defined by a clinical condition (Diabetes) or issue (readmissions).
Sub-groups based more on needs and less on conditions.
Note: The size of the rectangles is meant to be indicative of population size, not cost.
An approach to achieving the IHI’s The Triple Aim for a given population - from the perspective of a consumer health plan-less, FFS based, Medicare participating, not for profit, hospital->health system…
A population, for which claims data exists and achieving Triple Aim results will not result in perverse economic loss.
A sub-population, high cost and or high utilization people from the larger population.
Sub-groups, people from the high cost high utilization sub-population that can be stratified based upon relatively similar needs.
People who have “fallen through the cracks” of our “rescue-care” system.
Interventions intended to address the needs of high cost high utilization sub-groups, Plan Do Study Act cycles.
Cannot be defined by a clinical condition (Diabetes) or issue (readmissions).
Sub-groups based more on needs and less on conditions.
Some interventions will work and some will not. All should result in learning and start on the smallest practical scale.
Note: The size of the rectangles is meant to be indicative of population size, not cost.
An approach to achieving the IHI’s The Triple Aim for a given population - from the perspective of a consumer health plan-less, FFS based, Medicare participating, not for profit, hospital->health system…
Components of a Learning System 1. System level measures2. Explicit theory or rationale for system changes3. Segmentation of the population4. Learn by testing changes sequentially5. Use informative cases: “Act for the individual learn for
the population” 6. Learning during scale-up and spread with a production
plan to go to scale7. Periodic review8. People to manage and oversee the learning system
From Tom Nolan PhD, IHI
Repeated Use of the PDSA Cycle for TestingChanges That Result in
Improvement
Hunches Theories
Ideas
DATA
Very Small Scale Test
Follow-up Tests
Wide-Scale Tests of Change
Implementation of Change
Sequential building of knowledge under a wide range
of conditions
Spreading
AP D
S
AP
D S
APD
S
A P
DS
Sustaining the gains
48
Percent of Surgeries with Intraoperative Temp Control
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25Obs
Percent
Daily % cases in control
12
3
45
6
Mini-measure tracks improvement cycles
Scottish Borders - Run chart of run charts!R
Scottish Borders - PDSAs completedR
Test! Test! Test!• Burt Sandeman’s Story Post LS1:
– The Challenge To Be Quick• Burt Sandeman’s Story Post LS2:
– The F-Word• Looked after two-year olds:
– My Prediction Was Wrong
30
Components of a Learning System 1. System level measures2. Explicit theory or rationale for system changes3. Segmentation of the population4. Learn by testing changes sequentially5. Use informative cases: “Act for the individual learn for
the population” 6. Learning during scale-up and spread with a production
plan to go to scale7. Periodic review8. People to manage and oversee the learning system
From Tom Nolan PhD, IHI
Act with the Individual
• Dad’s Care• Asset Based Community Development in N. Ay
rshire
32
Components of a Learning System 1. System level measures2. Explicit theory or rationale for system changes3. Segmentation of the population4. Learn by testing changes sequentially5. Use informative cases: “Act for the individual learn for
the population” 6. Learning during scale-up and spread with a production
plan to go to scale7. Periodic review8. People to manage and oversee the learning system
From Tom Nolan PhD, IHI
Things to Consider to “Scale-up”• Determine full scale at project setup and milestones to reach full scale• Different changes may require different scale-up strategies • Consider different dimensions of structure
– Information technology– Physical (e.g. space, equipment, capacity)– Human resources (workforce organization and capabilities)– Financial– Learning system
• Use “5x” (5--25--125--625--3125---) thinking to predict/define the structural issues and set a path forward for testing– (What is working when testing with x that probably won’t work with 5x, ...?)
• Standardize processes (e.g. training, referral)• Understand oversight requirements as the system grows
Components of a Learning System 1. System level measures2. Explicit theory or rationale for system changes3. Segmentation of the population4. Learn by testing changes sequentially5. Use informative cases: “Act for the individual learn for
the population” 6. Learning during scale-up and spread with a production
plan to go to scale7. Periodic review8. People to manage and oversee the learning system
From Tom Nolan PhD, IHI
Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12
0
10
20
30
40
50
60
70
80
Avg. Patient Satisfaction - Clinic B
Last 12 Quarters
Avg
% S
atis
fact
ion
Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12
0
10
20
30
40
50
60
70
80
Avg. Patient Satisfaction - Clinic C
Last 12 Quarters
Avg
% S
atis
fact
ion
Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12
0
10
20
30
40
50
60
70
80
Avg. Patient Satisfaction - Clinic D
Last 12 Quarters
Avg
% S
atis
fact
ion
Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12
0
10
20
30
40
50
60
70
80
Avg. Patient Satisfaction - Clinic E
Last 12 Quarters
Avg
% S
atis
fact
ion
Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12
0
10
20
30
40
50
60
70
80
Avg. Patient Satisfaction - Clinic F
Last 12 Quarters
Avg
% S
atis
fact
ion
Period Review
Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12
0
10
20
30
40
50
60
70
80
Avg. Patient Satisfaction - Clinic G
Last 12 Quarters
Avg
% S
atis
fact
ion
Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12
0
10
20
30
40
50
60
70
80
Avg. Patient Satisfaction - Clinic H
Last 12 Quarters
Avg
% S
atis
fact
ion
Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12
0
10
20
30
40
50
60
70
80
Avg. Patient Satisfaction - Clinic A
Last 12 Quarters
Avg
% S
atis
fact
ion
Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12
0
10
20
30
40
50
60
70
80
Avg. Patient Satisfaction - Clinic I
Last 12 Quarters
Avg
% S
atis
fact
ion
You are here! AimThe Work Remaining to Do
Keeping an eye on the journey
Components of a Learning System 1. System level measures2. Explicit theory or rationale for system changes3. Segmentation of the population4. Learn by testing changes sequentially5. Use informative cases: “Act for the individual learn for
the population” 6. Learning during scale-up and spread with a production
plan to go to scale7. Periodic review8. People to manage and oversee the learning system
From Tom Nolan PhD, IHI
40