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 Adverse events in 9 – 18% of admissions ~ 50% preventable 2.5 – 7.5% are fatal
NEJM Nov 25, 2010 Qual Safety in Health Care 2008;17:216-223 3
Preventable Harm: 1 in 3 Hospital Patients
n=795 3 hospitals
#1. procedures #2. medications #3. infection
Voluntary reports are 1% of events Health Affairs, 30, no.4 (2011):581-589 4
Hospital-Acquired Infection • World – 1.4 million patients affected / day
• Developed countries
JAMA 2009;301(12):1285-1287 Lancet 2008;372(9651):1719-1720
– 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 SA-HISC study (unpublished) Allegranzi B; Lancet 2010:61458 5
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
Pronovost P. NEJM Dec 2006 6
EVIDENCE CLABSI Bundle
maximal barrier precautions chlorhexidine skin antisepsis optimal catheter site selection daily review of line necessity
ü ü ü ü
Checklists + Daily goals sheet + Unit-based safety program Pronovost P. NEJM Dec 2006 7
CAUTI
CLABSI + Antibiotic Stewardship
SSI
VAP
Public Sector Improvement Scientists Professionals Private Hospitals Sponsors
Expert Panel
BUILDING WILL
9
from
evidence-based medicine to
Evidence-based IMPLEMENTATION Subject Matter Knowledge
“what”
Effective changes that lead to an improvement
“where”
“how”
Improvement Science 10
202 Hospitals enrolled at least 1 intervention
11
“The Method for Improvement” = the Scien3fic Method
Not protocols Not “recipes” 12
AIMS
Decrease CLABSI to 0 or 300 patient days between by June 2010
10 DRIVERS Reliable identification of a CLABSI Reliable implementation of the insertion bundle
20 DRIVERS Reliable Definition – CLABSI Reliable data collection process – infections & device days Insertion bundle (HII) Line insertion documented in notes
Reliable implementation of the maintenance bundle
Reliable data collection Maintenance bundle (HII) Reliable data collection
Improve safety culture through multidisciplinary team working and communication
Clinical leadership rounding to view lines Bundles known by staff HAI surveillance data and bundle compliance shared with staff
95% compliance with insertion bundle and each element of bundle 95% compliance with maintenance bundle and each element of bundle
AIMS
Optimal antibiotic use in 80% of patients 30% reduction in antibiotic overuse Stable / decreased antibiotic resistance
10 DRIVERS
CHANGE CONCEPTS ↑ availability of first dose
Prompt initiation, for defined reasons Periodic review for cessation, route, reason for treatment Prescriber access to knowledge and data
Prevention of hospital-acquired infection
Separate AB prescribing from other Rx Day 3 and Day 7 review Info on how to Rx Info on what it costs
*Prevent SSI, CLABSI, VAP and CAUTI
*Interventions already associated with the BCA campaign
INTERVENTIONS
Antibiotic ward stock Antibiotic form AB Bundles Clinical pharmacist review Path lab hotline Resistance reports Cost reports SSI bundle CAUTI bundle CLABSI bundle VAP bundle
OUTCOME MEASURE
10 PROCESS MEASURE
20 PROCESS MEASURE
% with compliance to all bundles (“optimal use”)
% receiving timely % compliance with each Inception antibiotics for prevention or bundle element: 1. <2 hrs from order → admin treatment – first antibiotic (treatment) prescribed during hospital 2. Prophylaxis within 1 hr of incision course % overall compliance with Day 3 Bundle for the first antibiotic prescribed during hospital course
% compliance with each Maintenance bundle element:
% overall compliance with Day 7 Bundle for the first antibiotic prescribed during hospital course
% compliance with each Maintenance bundle element:
1. 2. 3. 4.
Treatment not prophylaxis State antibiotic indication or stop Culture(s) ordered or done Reassess drug choice
1. Stopped or re-ordered 2. Conversion from IV to oral or N/A
IDEAS 18 -24 months Repeated improvement cycles:
Expert and Planning Group formed
Learning session 1
Repeated improvement cycles:
Learning session 2
Learning session 3
Mentoring and support
16
8.00
CLABSI -‐ Infec9on Rates Per 1000 Central Line Days 7.17 Life Healthcare Group -‐ Oct 2008 to Sept 2010
7.00
6.00 5.66
5.57
5.22
5.00
4.00
4.95
4.21
Repor3ng system and training on BCA in all acute hospitals
3.85
3.00
3.48
3.36
3.22
3.12
2.93 2.58
2.90 2.98 2.46
2.01 2.10 2.01
2.00
Cross func3onal workshops in ICU's to implement bundle compliance ac3ons and increase involvement of Unit Managers 2.33 2.08
2.17
2.05
1.57 1.00
CLABSI-‐Rate
1.67
Mean 0.85
Upper control limit
0.57
Lower control limit
0.00
n=41 hospitals 17
HAI : CLABSI RATE Central Line Associated Blood Stream Infections - CLABSI Central Line Associated Blood Stream Infec9ons -‐ Bundle Compliance and Infec9on Rate Mar 09 -‐ Aug 10 100%
6.00
87% 5.22 5.00
83%
4.95
83%
83%
85%
83%
85%
88% 84%
84%
87%
87%
87%
87%
90%
80%
78% 70%
4.00
90%
70%
3.22 3.00
3.12
2.93 2.58 2.01
2.00
2.10
60%
3.48
3.36
50%
2.90 2.98 2.46
2.33 2.08
2.01
40%
Infec3on Rate CLABSI
2.17 30% 1.57 20%
1.00
n=41 hospitals
Aug-‐10
Jul-‐10
Jun-‐10
May-‐10
Apr-‐10
Mar-‐10
Feb-‐10
Jan-‐10
Dec-‐09
Nov-‐09
Oct-‐09
Sep-‐09
Aug-‐09
Jul-‐09
Jun-‐09
May-‐09
Apr-‐09
-‐
Mar-‐09
10% 0%
18
BCA : COMPLIANCE : SSI Bundle Compliance to SSI Period: JUNE-10
100%
96% 89%
90%
89%
80% 70% 60%
57%
50% 40% 30% 20% 10% 0% If hair is removed, it is only done with Antibiotics are given within an hour of Glucose is maintained above 4 and The patients temperature is clippers or dipilatory cream incision below 8 after the initial post operative maintained at >36.5 and <37.2 after assessment in ICU the initial post operative assessment in ICU
19
•
•
•
•
• •
•
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
10
1
11
0
6
5
1
3
11
4
6
9
8
6
7
2
6
6
14
2
1
12
9
4
11 17
4
4
3-‐Jul
9-‐Jul
15-‐Jul
29-‐Jul
31-‐Jul
1-‐Aug
13-‐Aug
22-‐Aug
26-‐Aug
24-‐Sep
28-‐Sep
2-‐Oct
7
1-‐Jul
6
24-‐Jun
5
18-‐Jun
4
10-‐Jun
3
1-‐Jun
2
26-‐May
1
22-‐May
9 10 11 12 13 14 15 16 17 18 19 20 21 22 Infection
11-‐May
8
•
8-‐May
7
•
•
7-‐May
6
Infection 3 days after admission
•
2-‐May
5
•
26-‐Apr
4
•
26-‐Apr
3
•
15-‐Apr
1
2
•
•
•
•
• •
•
• •
4-‐Apr
Date of infections
•
•
1 Data Element Days between infections
•
•
Possible contaminant
•
August -‐
7-‐Sep
31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 •
Started with CLABSI Bundle
Neurosurgery I CU -‐ Steve Biko Academic Hospital -‐ Pretoria Month: April -‐ July 2010
14-‐Apr
Days between infections
Central Line Associated Bloodstream Infections
8
8
9 10 11 12
9
10 11 12
Days Between Infection 20
No incident
New incident
More than 1 incident Poor data
Visual Measurement 21
At the same time every day the Unit manager counts devices in use in the ward
Solving the Denominator Problem
www.bestcare.org.za
Not just infections Not just bundles
garyk@discovery.co.za
Everyone in healthcare has 2 jobs 1. Doing the work 2. Improving the work!
24
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
NEW
International data
Our data
Counts
Rates
Bar graphs
Run charts (over time)
Data for head office / ministry
Measurement for frontline staff
Individual measures
Measures across systems
Modified from: D van den Bergh, Netcare Hospital Group 26
BMJ Qual Saf 2011;20:46e51
“Run Chart”
Rules for iden3fying non-‐ random signals
27 27
How we improve OLD
NEW
Audit and inspection (QA)
PDSA cycles
Checklists for checking
Checklists as aids
Writing more protocols
Focused interventions
â&#x20AC;&#x153;Spray and prayâ&#x20AC;?
Improving critical elements one a time
Modified from: D van den Bergh, Netcare Hospital Group 28
Taking Responsibility OLD
NEW
“it doesn’t happen here”
Knowing the facts
“we already do that”
Acknowledging we may not
Can’t do
“if they can so can we”
Accept the inevitable
Persistence
Victim of limitations
Building skills
Modified from: D van den Bergh, Netcare Hospital Group 29
Clinicians Skeptical and critical “this might work” “worth trying” “how can we support you” “I would like to initiate” 30
Leadership OLD
NEW
It’s up to the doctors
Active involvement of senior leadership
It’s up to the nurses
“Exco”
It’s up to the Infection Prevention Practitioners It’s up to the Infection Control Committee
Modified from: D van den Bergh, Netcare Hospital Group 31
The Culture OLD
NEW
Blaming and punishing
Learning and curious
Who (people)
Why (system)
Helping
Capacitating (mentors)
Modified from: D van den Bergh, Netcare Hospital Group 32
Health Sector OLD
NEW
Competition
Collaboration
Secrets
Sharing
Private vs public
Interconnected systems Public learns from Private Private learns from Public Public-private partnership
Modified from: D van den Bergh, Netcare Hospital Group 33
Antibiotic Stewardship OLD
NEW
Passive observers
Actively seeking solutions
No interventions
Identifying opportunities
No measures
First level utilisation data
Defensiveness
Working together to deal with it
Pilots
Multiple sites testing change
Modified from: D van den Bergh, Netcare Hospital Group 34
www.bestcare.org.za
The Changing View of Quality We are perfect!
NO ACTION
Get rid of the bad apples REACTION Quality Assurance “Standards”
M&M
Incident reporting
36
The Changing View of Quality “Quality” We are perfect!
INACTION
Get rid of the bad apples REACTION “Quality Assurance”
M&M
Safe Effective Timely Equitable Patient-centred Efficient
System thinking
PRO-ACTION Quality Improvement
“Standards”
Improvement Science
Incident reporting
Dashboard of Quality Measures
37
Improvement Science and Knowledge Systems Combine to Produce Improvement Generalisable scien9fic evidence “control context”
Patients get “recommended care” ~ 50% of the time standardisation, forcing Qual Saf Health Care + functions, education, etc
2007;16:2–3
Plans for change
+
Par9cular context “local processes, habits, traditions”
+
Execu9on “drivers of change” High performance (measured) “include time” 38