dr gareth kantor 1
Post on 22-Jul-2016
217 Views
Preview:
DESCRIPTION
TRANSCRIPT
Best Care…Always! Campaign
Global Forum on Bacterial Infec3on New Delhi ; October 4, 2011
Dr Gary Kantor
A loose coalition of stakeholders can initiate / sustain system strengthening for patient safety. The process began with antibiotic stewardship, and was driven by a private funder, hospitals and professionals. We are exploring new ways of building will, generating and sharing ideas and filling the execution gap ….changing the system.
Global Epidemic of Harm in Hospitals
3
Adverse events in 9 – 18% of admissions
~ 50% preventable
2.5 – 7.5% are fatal
Qual Safety in Health Care 2008;17:216-223
NEJM Nov 25, 2010
n=795 3 hospitals
Preventable Harm: 1 in 3 Hospital Patients
4
#1. procedures
Health Affairs, 30, no.4 (2011):581-589
Voluntary reports are 1% of events
#3. infection
#2. medications
Hospital-Acquired Infection • World
– 1.4 million patients affected / day
• Developed countries – Hospital incidence up to 10% – USA: 100,000 deaths
• Developing countries – 3 x higher – S Africa 9.7% prevalence; 28.6% ICU
A Duse. SA-HISC study (unpublished) Allegranzi B; Lancet 2010:61458
5
JAMA 2009;301(12):1285-1287 Lancet 2008;372(9651):1719-1720
Pronovost P. NEJM Dec 2006
6
Median rate of infection: ZERO!!
CLABSI rates ↓ by 66% Better than 90% of US ICUs 1,500 lives and $$$ saved Sustained > 3 years
96 ICUs
7
maximal barrier precautions chlorhexidine skin antisepsis optimal catheter site selection daily review of line necessity
EVIDENCE
+ Daily goals sheet
+ Unit-based safety program
ü ü ü ü
CLABSI Bundle
Pronovost P. NEJM Dec 2006
Checklists
8
+ Antibiotic Stewardship
CAUTI CLABSI
SSI VAP
9
Expert Panel
Public Sector
Professionals
Private Hospitals
Sponsors
Improvement Scientists
BUILDING WILL
evidence-based medicine Evidence-based IMPLEMENTATION
Effective changes that lead to an improvement
10
Subject Matter Knowledge
Improvement Science
“what”
“how”
“where”
from
to
11
at least 1 intervention
202 Hospitals enrolled
“The Method for Improvement” = the Scien3fic Method
12
Not protocols
Not “recipes”
Decrease CLABSI to 0 or 300 patient days between by June 2010
AIMS 10 DRIVERS 20 DRIVERS
Reliable implementation of the insertion bundle
Reliable implementation of the maintenance bundle
Reliable identification of a CLABSI
Improve safety culture through multidisciplinary team working and communication
Bundles known by staff
Insertion bundle (HII)
Reliable data collection process – infections & device days
Line insertion documented in notes
Reliable data collection
Reliable Definition – CLABSI
Clinical leadership rounding to view lines
Maintenance bundle (HII)
HAI surveillance data and bundle compliance shared with staff
Reliable data collection
95% compliance with insertion bundle and each element of bundle 95% compliance with maintenance bundle and each element of bundle
30% reduction in antibiotic overuse
Optimal antibiotic use in 80% of patients
AIMS 10 DRIVERS CHANGE CONCEPTS
CAUTI bundle
SSI bundle
CLABSI bundle
VAP bundle
Antibiotic form
Clinical pharmacist review
Path lab hotline
Periodic review for cessation, route, reason for treatment
Prescriber access to knowledge and data
Cost reports
Prompt initiation, for defined reasons
Stable / decreased antibiotic resistance
Resistance reports
Prevention of hospital-acquired infection
*Prevent SSI, CLABSI, VAP and CAUTI
INTERVENTIONS
Day 3 and Day 7 review
Separate AB prescribing from other Rx
Info on how to Rx
Info on what it costs
↑ availability of first dose
Antibiotic ward stock
AB Bundles
*Interventions already associated with the BCA campaign
OUTCOME MEASURE
10 PROCESS MEASURE 20 PROCESS MEASURE
% with compliance to all bundles (“optimal use”)
% receiving timely antibiotics for prevention or treatment – first antibiotic prescribed during hospital course
% compliance with each Inception bundle element:
1. <2 hrs from order → admin (treatment)
2. Prophylaxis within 1 hr of incision
% overall compliance with Day 3 Bundle for the first antibiotic prescribed during hospital course
% compliance with each Maintenance bundle element:
1. Treatment not prophylaxis 2. State antibiotic indication or stop 3. Culture(s) ordered or done 4. Reassess drug choice
% overall compliance with Day 7 Bundle for the first antibiotic prescribed during hospital course
% compliance with each Maintenance bundle element:
1. Stopped or re-ordered 2. Conversion from IV to oral or N/A
Expert and Planning
Group formed
Learning session
1
Learning session
2
Repeated improvement
cycles:
Repeated improvement
cycles:
Learning session
3
18 -24 months
Mentoring and support
16
IDEAS
5.57
3.85
5.66
4.21
7.17
4.95 5.22
2.01 2.10 2.01
3.22 2.93
3.36
2.58
3.12
2.46
3.48
2.90 2.98
2.08 2.33
2.17
1.57
2.05
1.67
0.57 0.85
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
CLABSI -‐ Infec9on Rates Per 1000 Central Line Days Life Healthcare Group -‐ Oct 2008 to Sept 2010
CLABSI-‐Rate Mean
Upper control limit Lower control limit
Repor3ng system and training on BCA in all acute hospitals Cross func3onal workshops in
ICU's to implement bundle compliance ac3ons and increase involvement of Unit Managers
17
n=41 hospitals
HAI : CLABSI RATE
4.95 5.22
2.01 2.10 2.01
3.22
2.93
3.36
2.58
3.12
2.46
3.48
2.90 2.98
2.08 2.33
2.17
1.57
70%
87% 83% 83%
78%
83% 85%
83% 85% 84% 84%
88% 87% 87% 87% 87% 90%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
-‐
1.00
2.00
3.00
4.00
5.00
6.00
Mar-‐09
Apr-‐09
May-‐09
Jun-‐09
Jul-‐0
9
Aug-‐09
Sep-‐09
Oct-‐09
Nov-‐09
Dec-‐09
Jan-‐10
Feb-‐10
Mar-‐10
Apr-‐10
May-‐10
Jun-‐10
Jul-‐1
0
Aug-‐10
Central Line Associated Blood Stream Infec9ons -‐ Bundle Compliance and Infec9on Rate Mar 09 -‐ Aug 10
Infec3on Rate
CLABSI
Central Line Associated Blood Stream Infections - CLABSI
18 n=41 hospitals
BCA : COMPLIANCE : SSI
Bundle Compliance to SSI Period: JUNE-10
57%
89% 89%
96%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
If hair is removed, it is only done withclippers or dipilatory cream
Antibiotics are given within an hour ofincision
Glucose is maintained above 4 andbelow 8 after the initial post operative
assessment in ICU
The patients temperature ismaintained at >36.5 and <37.2 afterthe initial post operative assessment
in ICU
19
313029282726252423222120191817 •161514 •1312 • •11 • •10 •9 • •8 •7 •6 • • • • •5 •4 • • • •3 •2 • •1 • • •0 • •
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 1 2 3 4 5 6 7 8 9 10 11 12
Neurosurgery ICU -‐ Steve Biko Academic Hospital -‐ PretoriaAugust -‐
Infection
9 10 11 12
11 17 4 414 2 1 12 9 4
5
Central Line Associated Bloodstream Infections
Month: April -‐ July 2010
2-‐Oc
t
22-‐Aug
26-‐Aug
7-‐Sep
24-‐Sep
28-‐Sep
9-‐Jul
15-‐Ju
l
29-‐Ju
l
31-‐Ju
l
1-‐Au
g
13-‐Aug
1-‐Jun
10-‐Ju
n
18-‐Ju
n
24-‐Ju
n
1-‐Jul
3-‐Jul
2-‐May
7-‐May
8-‐May
11-‐M
ay
22-‐M
ay
26-‐M
ayDate of infections
4-‐Ap
r
14-‐Apr
15-‐Apr
26-‐Apr
26-‐Apr
Days between infections 10 1 11 0 6 5
212 13
8 6 7 2 6 61 3 11 4 6 9
Days between infections
Data Element 1 2 3 4 5 618 19 20 21 22 16 7 8 9 10 11 7 83 4
Started with
CLABSI Bun
dle
14 15 16 17
Possible contam
inant
Infection 3 days after adm
ission
20
Days Between Infection
No incident
New incident More than 1
incident Poor data
21
Visual Measurement
At the same time every day the Unit manager counts devices in use in the ward
Solving the Denominator Problem
23
www.bestcare.org.za
garyk@discovery.co.za
Not just infections
Not just bundles
24
Everyone in healthcare has 2 jobs
1. Doing the work 2. Improving the work!
All improvement requires change (though not all change is an improvement) Changing:
How and why we measure Methods (of improvement) Our sense of responsibility Leadership Organisations and culture The Health industry
25
How and why we measure
OLD
International data
Counts
Bar graphs
Data for head office / ministry
Individual measures
NEW
Our data
Rates
Run charts (over time)
Measurement for frontline staff
Measures across systems
26
Modified from: D van den Bergh, Netcare Hospital Group
27
BMJ Qual Saf 2011;20:46e51
Rules for iden3fying non-‐random signals
“Run Chart”
27
How we improve
OLD
Audit and inspection (QA)
Checklists for checking
Writing more protocols
“Spray and pray”
NEW
PDSA cycles
Checklists as aids
Focused interventions
Improving critical elements one a time
28
Modified from: D van den Bergh, Netcare Hospital Group
Taking Responsibility OLD
“it doesn’t happen here”
“we already do that”
Can’t do
Accept the inevitable
Victim of limitations
NEW
Knowing the facts
Acknowledging we may not
“if they can so can we”
Persistence
Building skills
29
Modified from: D van den Bergh, Netcare Hospital Group
Clinicians
Skeptical and critical
“this might work”
“worth trying”
“how can we support you”
“I would like to initiate”
30
Leadership OLD
It’s up to the doctors
It’s up to the nurses
It’s up to the Infection
Prevention Practitioners
It’s up to the Infection Control Committee
NEW
Active involvement of senior leadership
“Exco”
31
Modified from: D van den Bergh, Netcare Hospital Group
The Culture OLD
Blaming and punishing
Who (people)
Helping
NEW
Learning and curious
Why (system)
Capacitating (mentors)
32
Modified from: D van den Bergh, Netcare Hospital Group
Health Sector NEW
Collaboration
Sharing
Interconnected systems
Public learns from Private
Private learns from Public
Public-private partnership
OLD
Competition
Secrets
Private vs public
Modified from: D van den Bergh, Netcare Hospital Group
33
Antibiotic Stewardship OLD
Passive observers
No interventions
No measures
Defensiveness
Pilots
NEW
Actively seeking solutions
Identifying opportunities
First level utilisation data
Working together to deal with it
Multiple sites testing change
34
Modified from: D van den Bergh, Netcare Hospital Group
35
www.bestcare.org.za
The Changing View of Quality
We are perfect!
NO ACTION
Get rid of the bad apples
M&M
Quality Assurance
REACTION
Incident reporting
“Standards”
36
The Changing View of Quality
We are perfect!
INACTION
Get rid of the bad apples
System thinking
M&M
Quality Improvement
“Quality Assurance”
REACTION PRO-ACTION
“Quality” Safe
Effective Timely
Equitable Patient-centred
Efficient
Improvement Science
Incident reporting
“Standards”
37
Dashboard of Quality Measures
Generalisable scien9fic evidence
Improvement Science and Knowledge Systems Combine to Produce Improvement
High performance (measured)
Plans for change
Par9cular context
Execu9on
+
+
+
Qual Saf Health Care 2007;16:2–3
Patients get “recommended care” ~ 50% of the time
“control context”
“include time”
“local processes, habits, traditions”
standardisation, forcing functions, education, etc
“drivers of change”
38
top related