Transcript Improvement - Health in Wales

Introduction to Quality and
Quality Improvement
James Mountford
Healthcare Consultant
McKinsey & Company, UK
Pat.O’Connor
National Patient Safety Development
Advisor
Scottish Government
NHS Wales Quality Improvement Academy launch
Cardiff, 25 June 2009
What we will cover
• Why do we need Quality Improvement
• Where are the gaps?
• The model for improvement how does it
work?
• The difference between QI and research
• Skills to Support Improvement
• Starting an improvement project
• Making change stick
What we will cover
• Why do we need Quality Improvement
• Where are the gaps?
• The model for improvement how does it
work?
• The difference between QI and research
• Skills to Support Improvement
• Starting an improvement project
• Making change stick
Play news quiz clip here
McGlynn, et al: The quality of health
care delivered to adults in the United
States. NEJM 2003; 348: 2635-2645
(June 26, 2003)
Conclusion: The “Defect Rate” in the
technical quality of American health
care is approximately
45%
If flying Lufthansa were associated with the same
rate of preventable fatalities that we currently see
in hospitals…
1. Flying would be much safer
2. There would be a crash every week
3. There would be 100,000 passengers killed per year
4. There would be 17 crashes per day and 600,000 deaths per year
Answer Now
Source: McKinsey
If flying Lufthansa were associated with the same
rate of preventable fatalities that we currently see
in hospitals…
1. Flying would be much safer
2. There would be a crash every week
3. There would be 100,000 passengers killed per year
4. There would be 17 crashes per day and 600,000 deaths per year
Answer Now
Source: McKinsey
Routine care has enormous variations e.g. the
average re-intervention rate with hip replacement
surgery in Germany is 2.5%
What is the
reintervention rate
at the worst-inclass hospitals?
1. Twice as high
2. Three times as
high
3. Four times as high
4. More than five
times as high
Answer Now
Source: McKinsey
Routine care has enormous variations e.g. the
average re-intervention rate with hip replacement
surgery in Germany is 2.5%
What is the
reintervention rate
at the worst-inclass hospitals?
1. Twice as high
2. Three times as
high
3. Four times as high
4. More than five
times as high
Answer Now
Source: McKinsey
MRSA rates are higher than
many other European countries
European MRSA rates
European incidences of MRSA per 100,000 bed days
Norway
0.1
Sweden
0.3
Netherlands
0.4
Greece
Spain
Italy
Germany
U.K.
3.3
6.0
6.4
7.4
9.6
France
Source: European Antimicrobial Resistance Surveillance System/National Institute of Public Health
and the Environment (RIVM) in The Netherlands
11.5
Substantial variation at regional level
% of acute myocardial infarction patients receiving primary angioplasty
Anglia
Avon
Beds & Herts
Birmingham
Black Country
C&M
C&W
Dorset
ES & NS
Essex
Manchester
H&W
Kent
Lancashire
LNR
N&E Yorks
NC London
NE London
N England
N Trent
NW London
Peninsular
Shrop & Staff
S Central
SE London
SW London
Sussex
Trent
W Surrey
W Yorkshire
Source: MINAP study, 2008
0
11
15
58
97
2
18
0
0
0
0
18
8
6
17
6
97
100
23
9
0
0
30
97
28
20
14
81
75
93
Quality Healthcare
• Safe avoiding harm to staff and patients from the care that is
intended to help them
• Timely reducing waits and harmful delays for both those who receive
and those who give care
• Effective care based on scientific knowledge to all who could benefit
and refraining from actions to those not likely to benefit
• Efficient avoiding waste, of equipment, supplies, ideas, and energy
• Equitable care that does not vary in quality because of personal
characteristic such as gender, ethnicity, geographic location, and
socio-economic status
• Patient-centered care that is respectful of and responsive to
individual patient preferences, needs and values, and ensuring that
patient values guide all clinical decisions
Crossing the Quality Chasm: A New Health System for the 21st Century, The Institute of Medicine, 2001
What we will cover
• Why do we need Quality Improvement
• Where are the gaps?
• The model for improvement how does it
work?
• The difference between QI and research
• Skills to Support Improvement
• Starting an improvement project
• Making change stick
Risk reduction strategies -Prevention
Detection Mitigation
Reason Systems Model
Reason 1997
The Good? The Bad? The Ugly?
What does your drug trolley look like?
The First Law of Improvement
•“Every system is perfectly
designed to achieve exactly
the results it gets.”
What we will cover
• Why do we need Quality Improvement
• Where are the gaps?
• The model for improvement how does it
work?
• The difference between QI and research
• Skills to Support Improvement
• Starting an improvement project
• Making change stick
AIM
MEASURES
TESTS
Testing
method
The Improvement Guide, API
Fundamental Questions for
Improvement
•
What are we trying to accomplish?
•
How will we know that a change is an
improvement?
•
What changes can we make that will result
in an improvement?
Play basketball video here
A Worthy Aim….
• Improve the reliability and efficiency of
airport security at every major airport in
the US. Over the next six months we will
focus on the Philadelphia Airport and then
spread improvements to all other major
airports by December 2007. Our specific
goals are to:
- Eliminate defects (i.e., number of bags
passing security that should have failed)
- Eliminate waiting times for passengers (i.e.,
no queues at the security stations)
What Are We Trying
to Accomplish?
• What is the aim?
• Needs a clear and concise statement
of what you intend to do
• Needs an unambiguous numerical
target for improvement
• Needs to be aligned with the
organisation’s wider goals
Aims Statements – some examples
Surgery: Achieve 100% compliance with appropriate
selection and timing of prophylactic antibiotic
administration
ICU: Reduce Central Line Infections in the ITU by
75% within 12 months
Medication reconciliation: Medications reconciled
with over 99% accuracy within 6 months
Communication: Safety huddles occur daily on every
ward 95% of days within 14 months
Fundamental Questions for
Improvement
•
What are we trying to accomplish?
•
How will we know that a change is an
improvement?
•
What changes can we make that will result
in an improvement?
How will we know that a change
is an improvement?
Measures! and various types of measurement
• Clinical measures of patients’ health
• What?
• How?
• Documentation of behaviors
• Questionnaires
• Assessments
• When?
• How often?
• Summary of databases
• Chart audits
• Observations
Focus on the Vital Few!
“There are many things in life that are
interesting to know. It far more important,
however, to work on those things that are
essential to quality than to spend time working
on what is merely interesting.
The challenge, therefore, is to be disciplined
enough to focus on the essential (or vital few)
things and set aside those things that might be
interesting but trivial”
R Lloyd, Institute for Healthcare Improvement
Measurement guidelines
• Track those few key measures which assess
progress against the aim (don’t let the
perfect get in the way of the useful)
– Generally 3-6 measures: outcome,
process and balancing
• Don’t wait for IT solutions – use pen and
paper if necessary
• Use qualitative data (feedback) as well as
quantitative data
• Make use of existing datasets
• Integrate measurement into the daily routine
• Plot progress
Outcome, Process, Balancing
measures
Outcome = Voice of the patient. Direct link
to the purpose of care:
• How is the system performing?
• What is the result?
• What is the impact on patients?
Process = Voice of the workings of the
system. What we work on to get to aim:
• Are the parts/steps in the system performing as
planned (inputs)?
• Are key changes being implemented in the
system?
Balancing =Looking at system overall from
different perspectives
• Unanticipated consequences
• Adverse consequences
• Competing explanations for success
Example: Peri-operative Safety
Measures
• Outcome:
– O1. Perioperative Harm Rate and
– O2. Percent Unplanned Surgical Readmissions and
– O3. Percent Unplanned Returns to Theatre
• Process:
– P1. Percent Pre-op Bleeding Risk Assessment (Bleeding)
– P2. Percent Appropriate Beta Blockade (Beta Blockade)
– P3. Percent On-time Prophylactic Antibiotic Admin (SSI)
– P4. Percent Surgery Patients with DVT Prophylaxis (DVT)
– P5. Percent Pre-procedural Checklist Used (Culture)
• Balancing:
•
B1. Percent of patients receiving prophylactic antibiotics
•
discontinued within 24 hours
•
B2. Volume/throughput impact
•
B3. Financial impact
•
B4. Patient / staff satisfaction
Example: Set of Measures for A&E
Service Improvement
Topic: Improve waiting time and patient satisfaction
in A&E
Measure
• % patient receiving discharge materials
• Number of patients
• Total Length of Stay (wait time) in A&E
• Time to registration
• Staff satisfaction
• Patient Satisfaction Scores
• Availability of antibiotics
• “Left without being seen” (LWBS)
• Financials
Type of measure
P
B
O
P
B
O
P
B
B
Minimum Standard for Monthly Reporting in
the Collaborative: Annotated Time Series
Cycle Time in Office
60
Huddles tried
Nurses start
early
Lab
Changes
50
Minutes
Patient
moved into
rooms ASAP
40
30
20
6/12
Goal
7/12
8/11
9/10
10/10
11/9
Measurement, Action, Improvement
Fundamental Questions for
Improvement
•
What are we trying to accomplish?
•
How will we know that a change is an
improvement?
•
What changes can we make that
will result in an improvement?
Peg game
Peg exercise
1
2
4
7
11
3
8
12
6
5
9
13
10
14
15
Why Test Changes?
•
To increase the belief that the change will
result in improvements in your setting
• To learn how to adapt the change to
conditions in your setting
• To evaluate the costs and “side-effects”
of changes
• To minimize resistance when spreading the
change throughout the organization
What is the PDSA Cycle?
Act
• What changes
are to be made?
• Next cycle?
Study
• Complete the
analysis of the data
• Compare data to
predictions
• Summarize what
was learned
Plan
• Objective
• Questions and
predictions (why)
• Plan to carry out
the cycle (who,
what, where, when)
Do
• Carry out the plan
• Document problems
and unexpected
observations
• Begin analysis
of the data
Repeated Use of the Cycle
Changes That
Result in
Improvement
Hunches
Theories
Ideas
Move Quickly
to Testing Changes
•
•
•
•
•
•
Year
Quarter
Month
Week
Day
Hour
“What tests can
we complete by
next Tuesday?”
Example of Testing Multiple Changes:
Airport security queues
Aim: Eliminate
queues at airport
security
Use separate
flows for
people and
bags
Match
capacity &
demand
Use visual
reminders
Use self-scanners
as pre-check
Example of Testing Multiple Changes:
Diabetes
Improving
diabetes using
The Chronic
Care Model
Registry
SelfManagement
Support
Decision
Support
Delivery
System Design
What we will cover
• Why do we need Quality Improvement
• Where are the gaps?
• The model for improvement how does it
work?
• The difference between QI and research
• Skills to Support Improvement
• Starting an improvement project
• Making change stick
How did you learn to ride a bicycle?
Comparative method
• Distance travelled with only
wheels on ground
Improvement method
• Distanced travelled with only
wheels on ground
•
•
•
•
•
• Improving one system,
(systems are not standard)
• 1 child, 1 bike, 1 environment
Standardised
60 children
Height, weight, sex matched
Same environment
Identical bikes.
Method
• ½ started with left foot on
ground
• ½ started with right foot on
ground……
Method
• Lots of encouragement (and a
helmet!)
• Several goes at trying different
things, quickly
R Lloyd, Institute for Healthcare Improvement
Converting research to care
Original research
18%
variabl
Negative
Dickersin, 1987
results
Submission
e
46%
0.5 year Kumar, 1992
Koren, 1989
Acceptance
Negative
17 years to apply
of
Kumar, 1992
0.6 year 14%
results
Publication
17:14
Expert
35%
0.3 year Poyer, 1982
opinion
Balas, 1995
to
patient
care
Lack of
Bibliographic databases
numbers
50
6. 0 - 13.0 yearsAntman, 1992
Poynard, 1985
%
Reviews, guidelines, textbook
9.3 years
Inconsistent
Patient Care
indexing
research knowledge
Balas EA, Boren SA. Managing clinical knowledge for health care improvement. Yrbk of Med Informatics
2000; 65-70
Why are you measuring?
Improvement?
R Lloyd, Institute for Healthcare Improvement
Improvement vs. traditional research –
complementary methodologies
Research
Improvement
Aim
Create new clinical
knowledge
Improve routine practice
(delivery)
Methods
Test blinded
Test observable
Eliminate bias
Stable bias
One variable
Many variables
Context stripped out
Context explicitly in
Fixed hypotheses
Rapid adaptation
One fixed test
Many sequential tests
Assess by statistical
significance
Assess by statistical
significance
“The Three Faces of Performance Measurement:
Improvement, Accountability and Research”
by
Lief Solberg, Gordon Mosser and Sharon McDonald
Journal on Quality Improvement vol. 23, no. 3, (March 1997), 135-147.
“We are increasingly realizing not only
how critical measurement is to the
quality improvement we seek but also
how counterproductive it can be to
mix measurement for accountability
or research with measurement for
improvement.”
What we will cover
•
•
•
•
Why do we need Quality Improvement
The difference between QI and research
Where are the gaps?
The model for improvement how does it
work?
• Skills to Support Improvement
• Starting an improvement project
• Making change stick
Skills for improvement
•Will
•Ideas
•Execution
Improvement Team
System
Leadership
Technical
Expertise
Day-to-day
Leadership
Adopter Categorization:
Speed of Adoption
Our Target
Early
Adopters
2%
13%
Early
Majority
35%
Late
Majority
35%
Tradition-alists
15%
Medicine Reconciliation
Acute Medical Admissions Ward - May 2005
Wrong
dose
Unintended
omission
% of patients reviewed
Wrong
timing
20 patients randomly selected
from 216 admitted on 6 week
days
● 36 admissions per day
● 81% seen by pharmacist
● 85% need intervention
● Mean per patient:
– 3.6 interventions
– 1.6 intervention types
0%
25%
50%
75%
100%
Percent of Unreconciled Medicines on Admission (Standard Project Measure) NHS Tayside Team
70%
60%
Percent Unreconciled on Admission
Testing in 3
wards in
different
specialities
Testing in
medical
admissions
only
50%
Reconciliation form
included in
admissions ward
documentation
Pilot population
altered - 10
medical
admissions, 10
general admissions
New form
being tested
40%
Intake of new
junior doctors
30%
Test of direct
access to
electronic GP
records in 1
patient
- access slow
20%
10%
Test of direct
access to
electronic GP
records in 50
patients on
one day 25% success
rate
Ongoing testing in
wider range of
clinical
specialities - 6
wards
Pilot
population - 20
patients from
across
organisation
Reconciliation
training for
medical
students
Testing /
Implemented in
21/43 wards on
Ninewells site
0%
May-05
Jun-05
Jul-05
Aug-05
Sep-05
Oct-05
Nov-05
Dec-05
Month
Jan-06
Feb-06
Mar-06
Apr-06
May-06
Jun-06
Your to do list
•
•
•
•
•
Start with an existing problem
Build on the existing system
Try and embed in current practice
If it feels to complex …..IT IS
Start and keep your changes small
ICU story
• Historically translation RCTs result into clinical practice in
ICU was difficult
• Uncertainty results were relevant to the case – mix
• Multidisciplinary staff and disparate views of priorities and
• Lack of will to change
• Previous attempts to introduce changes were nurse led or
doctor led
• Solution was education for all staff!
impossible to introduce and sustain uni-professional
led changes
Ventilator Care bundle
•
•
•
•
•
Head of bed elevation 30 degrees
Sedation vacation
PUD prophylaxis
DVT prophylaxis
Multi Disciplinary Rounds Daily goals
Safer Patient Initiative
Learning Set January 2005
• Acceptance of that both the evidence base
and the reasons for change were sound
• We are already doing this!
• Aims are unachievable
• Infections are inevitable in this patient group
• Just another gimmick!
Reality!
• Measurement of the ventilator bundle
elements for 8 patients
– Head of Bed > 30 degrees = 35%
– Sedation Hold = 50%
– PUD prophylaxis = 85%
– DVT prophylaxis = 100%
– Bundle ‘ All or nothing concept’
0% compliance to the ventilator bundle
Aim: Reduce VAP rates by 30% by Oct 2006 by
implementing the VAP care bundle in Intensive care
Measures
Cycle 1e:
Test sedation hold on all
patients with completion of a ventilator
bundle sticker
Cycle 1d:
Test sedation hold guideline with
exclusions five patients with five nurses and one
anaesthetist
Cycle 1c: Test sedation hold element of bundle on three patients
with three nurses and one anaesthetist using feedback from first
test
Cycle 1b:Test repeated with another patient with different nurse and
anesthetist
Cycle 1a: Test sedation hold element of bundle on one patient with
one nurse and one anaesthetist
Process Change: Testing the sedation hold element of the ventilator
bundle with frontline staff
% Compliance with the IHI Ventilator Bundle
ICU, Ninewells Hospital
100
90
80
70
5th element
introduced
50
40
30
20
10
4 element
bundle on
all patients
04/02/05 - 17/02/05 tests of change x 6 to
implement sedation vacation and HOB
on all patients
0
01
-0
205
M
ar
-0
5
M
ay
-0
5
Ju
l-0
5
Se
p05
No
v05
Ja
n06
M
ar
-0
6
M
ay
-0
6
Ju
l-0
6
Se
p06
No
v06
Ja
n07
M
ar
-0
7
M
ay
-0
7
Ju
l-0
7
Se
p07
No
v07
Ja
n08
M
ar
-0
8
M
ay
-0
8
Ju
l-0
8
Se
p08
No
v08
Ja
n09
M
ar
-0
9
M
ay
-0
9
Ju
l-0
9
Se
p09
No
v09
Compliance
60
High % of
agency/bank
nurses
New intake of anaesthetists, high
usage of bank/agency staff.
Education sheet developed.
Month
Au
g05
Oc
t- 0
5
De
c-0
5
Fe
b06
Ap
r- 0
6
Ju
n-0
Au 6
g0
Oc 6
t- 0
6
De
c-0
6
Fe
b07
Ap
r- 0
7
Ju
n-0
Au 7
g07
Oc
t- 0
7
De
c-0
7
Fe
b08
Ap
r- 0
8
Ju
n-0
Au 8
g08
Oc
t- 0
De 8
c-0
8
Fe
b09
Ap
r- 0
9
Ju
n-0
Au 9
g09
Oc
t- 0
9
De
c-0
9
VAP Rate per 1000 patient days
VAP Rate - ICU, Ninewells Hospital
70.00
60.00
50.00
ICU now admitting all
neuro patients
following closure of
neuro ICU
20.00
Month
Implementation of
daily goals
40.00
30.00
Added oral
chlorhexadine
10.00
0.00
Keys and Barriers to Success
Keys
Barriers
PDSA cycles
Small, rapid cycle
Seek usefulness not perfection -stickers
Improve as fast as you test
Multidisciplinary approach
Early adopters
‘having made a difference’
Leadership
Evidenced based
Measurement over time
Outcome & process measures
Run charts - feedback
Monthly review
Resistance to change
‘ not invented here’
‘already doing this’
‘this week’s gimmick’
Culture & behaviour
Educate, educate
Clinician engagement
scepticism
Resources
Data collection
Person dependence
Sustainable process
What we will cover
•
•
•
•
Why do we need Quality Improvement
The difference between QI and research
Where are the gaps?
The model for improvement how does it
work?
• Skills to Support Improvement
• Starting an improvement project
• Making change stick
A “Project”
Moving a Big Dot
Use the PDSA Cycle to:
• Answer the first two questions of the
Model for Improvement
• Develop a change
• Test a change
• Implement a change
Measurement and Data
Collection During PDSA Cycles
• Collect useful data, not perfect data - the
purpose of the data is learning, not evaluation
• Use a pencil and paper until the information
system is ready
• Use sampling as part of the plan to collect the
data
• Use qualitative data (feedback) rather than
wait for quantitative
• Record what went wrong during the data
collection
What we will cover
•
•
•
•
Why do we need Quality Improvement
The difference between QI and research
Where are the gaps?
The model for improvement how does it
work?
• Skills to Support Improvement
• Starting an improvement project
• Making change stick
What would it take for you to come to
work in pyjamas?
Barriers to overcome to build a culture of safety
among junior doctors
Mindsets
•
“Patient safety improvement is
not my responsibility”
•
“I just deal with the patient in
front of me – I don’t worry about
the whole system”
•
“You can’t standardize what we
do – every patient’s different”
•
“Healthcare is
complicated, there will
always be mistakes”
•
“Patient safety is about
targets not patient care”
•
“Patient safety is not a
big problem around here”
“Senior docs know best”
•
•
“I don’t know how to change the
system – I don’t think it would
make a difference even if I tried”
– “I don’t know how to ask for
help”
•
“I often feel scared
because I don’t know
what I’m doing”
•
Poor handover process
•
•
Few opportunities to discuss
harm in blame free setting
Lack of visible safety
performance data
Lack of efficient checks in place
Duplicative
documentation systems,
with a lack of structured
problem lists
•
Poor verbal
communication
Capabilities
Processes/
systems •
•
•
•
•
•
•
•
The ‘Basis’ programme: Six interventions which together
can build a safety culture for junior doctors
…to junior doctors who…
1
Tailored training course
2
Revised audits (Quality
Improvement projects)
3
•
•
•
•
•
Simple tools to encourage
safe practice
4
More opportunities to
discuss harm and near
misses
5
Capabilities
•
•
•
Increased support from
peers and seniors
6
Greater communication on
quality and safety
Improvements
SOURCE: Interviews; team analysis
Mindsets
Are aware of risks to patient safety
Believe that the best doctors clearly prioritise safety
Are excited about the role they can play in improving safety
Are empowered to make changes to the systems around them
Are willing to talk openly about mistakes
Know the major risks to safety and how to deal with them –
including when to ask for help
Are comfortable giving and receiving feedback
Are able to prioritize and structure daily work
Processes/systems
•
•
•
•
•
•
•
•
•
•
Have access to support when required
Are actively involved in improving the system
Are easily able to access clear guidelines
Are regularly exposed to visible safety data
The 100,000 lives campaign covered
the whole of America
79
Source: IHI
Successfully changing practice requires several
dimensions to be addressed simultaneously
Why did IHI’s 100,000 lives campaign excite America (despite there being no National
Health System?)
1 Positive role
modelling
2 Developing
skills
• Leaders in
• Clear, inspiring goal
3 A purpose to
believe in
each hospital
• Voluntary
were
participation
champions
• Professionals
• Nodes created
taking pride in
Taps
into
why
state-wide
improvement
professionals
buzz
went into
4 Reinforced
• Practical
healthcare
• Reporting
by formal
‘how-tos’,
systems and
mechanisms
other tools
‘rhythm’
and coaching
to build
• Various tools
capability
distributed
Source: Interviews with IHI and 100K lives participants; McKinsey
In conclusion
• Safety and quality are big issues
• While they are system/organisation issues, everyone
can make a difference
• Shifts in mindsets & capabilities are more important than
specific tools
• But the model for improvement is a proven tool: start
with the aim, choose measures, run rapid cycles
• Improvement
– Is about implementing research knowledge reliably
– Differs from traditional research in several ways
• Making change stick in a system requires addressing
several dimensions simultaneously