www.echo-health.eu
ECHO – European Collaboration for Healthcare Optimization An international project on healthcare performance
Enrique Bernal-Delgado on behalf of the ECHO consortium
www.echo-health.eu
-1.5
-1
-.5
0
.5
1
ENG POR SLV SPNStd Rate/10,000 15.58 17.84 14.12 11.05 EQ5-95 1.88 1.84 2.29 2.34 SCV 0.10 0.18 0.06 0.10
Hysterectomy in benign conditions
www.echo-health.eu
0
0,1
0,2
0,3
0,4
0,5
10
11
12
13
14
15
16
17
18
19
20
2002 2003 2004 2005 2006 2007 2008 2009
ENGLAND Standard Rate (SR) ENGLAND SCV
0
0,1
0,2
0,3
0,4
0,5
10
12
14
16
18
20
22
24
2002 2003 2004 2005 2006 2007 2008 2009
PORTUGAL Standard Rate (SR) PORTUGAL SCV
0
0,1
0,2
0,3
0,4
0,5
10
11
12
13
14
15
16
2002 2003 2004 2005 2006 2007 2008 2009
SLOVENIA Standard Rate (SR) SLOVENIA SCV
0
0,1
0,2
0,3
0,4
0,5
10
11
12
13
14
15
2002 2003 2004 2005 2006 2007 2008 2009
SPAIN Standard Rate (SR) SPAIN SCV
www.echo-health.eu
Challenges in international comparison
• Adapting the rationale to contextual elements – Supply in non existing o strongly intervened markets – Preference-sensitive in a context of limited choice – Maps boundaries do not necessarily mean the same
• Methods – Apples and Pears à the need of a common language – Variation also depends on the basal rates – Size is so different in Europe –
• Geographic representation have to be changed • Some additional statistics are required.
• Explaining variations: the need of profiling institutions
www.echo-health.eu
Challenges in international comparison
• Adapting the rationale to contextual elements – Supply in non existing o strongly intervened markets – Preference-sensitive in a context of limited choice – Maps boundaries do not mean necessarily the same
• Methods – Apples and Pears à the need of a common language – Variation also depends on the basal rates – Size is so different in Europe –
• Geographic representation have to be changed • Some additional statistics are required.
• Explaining variations: the need of profiling institutions
www.echo-health.eu
0
20
40
60
80
100
120
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45
Stan
dardized
rate
CSV
www.echo-health.eu
Challenges in international comparison
• Adapting the rationale to contextual elements – Supply in non existing o strongly intervened markets – Preference-sensitive in a context of limited choice – Maps boundaries do not mean necessarily the same
• Methods – Apples and Pears à the need of a common language – Variation also depends on the basal rates – Size is so different in Europe –
• Geographic representation have to be changed • Some additional statistics are required.
• Explaining variations: the need of profiling institutions
www.echo-health.eu
A heterogeneous reality (lowest level areas )
www.echo-health.eu
Challenges in international comparison
• Adapting the rationale to contextual elements – Supply in non existing o strongly intervened markets – Preference-sensitive in a context of limited choice – Maps boundaries do not mean necessarily the same
• Methods – Apples and Pears à the need of a – Variation also depends on the basal rates – Size is so different in Europe –
• Geographic representation have to be changed • Some additional statistics are required.
• Explaining variations: the need of profiling institutions
www.echo-health.eu Denmark
Former 99 units New 28 areas
www.echo-health.eu
Challenges in international comparison
• Adapting the rationale to contextual elements – Supply in non existing o strongly intervened markets – Preference-sensitive in a context of limited choice – Maps boundaries do not mean necessarily the same
• Methods – Apples and Pears à the need of a common language – Variation also depends on the basal rates – Size is so different in Europe –
• Geographic representation have to be changed • Some additional statistics are required.
• Explaining variations: the need of profiling institutions
www.echo-health.eu
Proper statistics in measuring systematic variation
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Emp
íric
al B
ayes
Coefficient of Variation
Coefficient of Variation vs. EB statistic
www.echo-health.eu
England: knee replacement and prostatectomy
20.6 10.5 5.01 9.3
0.5
11.5
2
SUR EBPGrr EBLNrr EBMarshall
0.5
11.5
2
SUR EBPGrr EBLNrr EBMarshall
16.3 12.2 4.5 9.9
www.echo-health.eu
Estimating SUR: adding spatial correlation SUR vs BYM –areas98
SFV: 0.55
www.echo-health.eu
Challenges in international comparison
• Adapting the rationale to contextual elements – Supply in non existing o strongly intervened markets – Preference-sensitive in a context of limited choice – Maps boundaries do not mean necessarily the same
• Methods – Apples and Pears à the need of a common language – Variation also depends on the basal rates – Size is so different in Europe –
• Geographic representation have to be changed • Some additional statistics are required.
• Explaining variations: the need of profiling institutions
www.echo-health.eu Institutional features in service provision
Joumard I, André C, Nicq C (2010) Healthcare systems efficiency and Institutions. OECD Economics. WP#769 OECD Publishing
www.echo-health.eu
more at www.echo-health.eu
www.echo-health.eu
Country differences in CSV – weighted by average rate
www.echo-health.eu
-5-4.5-4
-3.5-3
-2.5-2
-1.5-1-.50.51
1.52
2.5
ENG POR SLV SPN
Std Rate/10,000 21.45 7.92 22.17 12.05
EQ5-95 3.12 18.81 3.57 22.52
SCV 0.61 0.89 0.60 0.53
C-section low risk deliveries 2009 (women 15-55 yo)
www.echo-health.eu Radical Mastectomy
30
40
50
60
70
80
90
100
2002 2003 2004 2005 2006 2007 2008 200911.21.4
1.61.822.22.4
2.62.83
ENGLAND Standard Rate (SR) ENGLAND SCV
0
2
4
6
8
10
12
14
2002 2003 2004 2005 2006 2007 2008 20090.4
0.6
0.8
1
1.2
1.4
PORTUGAL Standard Rate (SR) PORTUGAL SCV
0
5
10
15
20
25
30
35
40
2002 2003 2004 2005 2006 2007 2008 20090
0.2
0.4
0.6
0.8
1
1.2
1.4
SLOVENIA Standard Rate (SR) SLOVENIA SCV
0
2
4
6
8
10
12
14
16
2002 2003 2004 2005 2006 2007 2008 20090
0.2
0.4
0.6
0.8
1
1.2
1.4
SPAIN Standard Rate (SR) SPAIN SCV
www.echo-health.eu
-1.5
-1
-.5
0
.5
1
1.5
ENG POR SLV SPN
Radical mastectomy in breast cancer 2009
Std Rate/10,000 88.08 13.25 28.99 12.66
EQ5-95 4.29 1.63 4.40 1.82
SCV 1.52 0.60 0.29 0.60
www.echo-health.eu
0,4
0,6
0,8
1
1,2
1,4
20
21
22
23
24
25
26
2002 2003 2004 2005 2006 2007 2008 2009 0,4
0,6
0,8
1
1,2
1,4
0
2
4
6
8
10
12
14
2002 2003 2004 2005 2006 2007 2008 2009
0
0,2
0,4
0,6
0,8
1
1,2
1,4
0
5
10
15
20
25
2002 2003 2004 2005 2006 2007 2008 2009 0
0,2
0,4
0,6
0,8
1
1,2
1,4
0
2
4
6
8
10
12
14
16
18
2002 2003 2004 2005 2006 2007 2008 2009
England Portugal
Slovenia Spain
www.echo-health.eu C-section LR over time
20
21
22
23
24
25
26
2002 2003 2004 2005 2006 2007 2008 20090.4
0.6
0.8
1
1.2
1.4
ENGLAND Standard Rate (SR) ENGLAND SCV
0
2
4
6
8
10
12
14
2002 2003 2004 2005 2006 2007 2008 20090.4
0.6
0.8
1
1.2
1.4
PORTUGAL Standard Rate (SR) PORTUGAL SCV
0
5
10
15
20
25
2002 2003 2004 2005 2006 2007 2008 20090
0.2
0.4
0.6
0.8
1
1.2
1.4
SLOVENIA Standard Rate (SR) SLOVENIA SCV
0
2
4
6
8
10
12
14
16
18
2002 2003 2004 2005 2006 2007 2008 20090
0.2
0.4
0.6
0.8
1
1.2
1.4
SPAIN Standard Rate (SR) SPAIN SCV
www.echo-health.eu
The concept
• ECHO has set about the task of bringing together patient-level data from Austria, Denmark, England, Portugal, Slovenia and Spain, making them comparable.
• ECHO is expanding the usual approach in healthcare performance
international comparison, by adding the variation framework
• Performance dimensions: utilization, equitable access to effective care, quality, and efficiency, in terms of opportunity costs, and provider-level efficiency.
• Healthcare areas or hospital providers will be flagged as good- or poor performers – not a diagnostic tool but a screening tool
www.echo-health.eu
Geographic-perspective Hospital-perspective 1. Effective care (hip fracture,
colectomy in colorectal cancer, breast surgery in breast cancer, etc)
2. Uncertain benefit/harm balance in “non-average” patients (CABG, knee replacement, etc.)
3. Doubtful/ lower value care and opportunity costs (tonsillectomy, spinal fusion, prostatectomy, etc.)
1. In-hospital case fatality rates for a condition (admissions with principal diagnosis of acute myocardial infarction, ischemic stroke);
2. In-hospital case fatality rates after a procedure (in hospital mortality after CABG, PCI, non-ruptured abdominal aortic aneurysm, hip replacement)
3. Patient safety event rates (Trauma after vaginal delivery with and without instrumentation, Catheter-related infection, Pulmonary Thromboembolism or Deep venous thrombosis, Postoperative sepsis)
www.echo-health.eu
Some messages from ECHO … … with potential impact in policy making
Utilization of effective procedures Equity Value - quality Value - efficiency
www.echo-health.eu
Utilization of effective procedures
• At population level, the burden of ischemic disease does barely explain variation in revascularization;
• Might be a symptom of under or overuse
www.echo-health.eu
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75
Q1 Q2 Q3 Q4 Q5 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Q1 Q2 Q3 Q4 Q5
PCI, CABG and burden of ischemic disease
PCI CABG
Pop
ulat
ion
rate
s of
pro
cedu
res
Burden of ischemic disease Burden of ischemic disease
www.echo-health.eu Flagging areas beyond the expected PCI
www.echo-health.eu
Equity
• Revascularization is performed differently across income quintiles, not always coherent with need.
• Differences beyond need might represent inequities in access.
www.echo-health.eu
PCI CABG
0
6
12
18
24
30
2002 2003 2004 2005 2006 2007 2008 2009
Standardise
d Ra
te
Q1 Q2 Q3 Q4 Q5
0
3
6
9
12
15
2002 2003 2004 2005 2006 2007 2008 2009
Standardise
d Ra
te
Q1 Q2 Q3 Q4 Q5
0
1
2
3
4
5
2002 2003 2004 2005 2006 2007 2008 2009
Standardise
d Ra
te
Q1 Q2 Q3 Q4 Q5
AMI
Spain
www.echo-health.eu
Quality
Case-fatality rates varies dramatically across high-volume hospitals, irrespective of the differences in patient case-mix Hospital f lagged as poor (or good) performers are likely to behave consistently overtime.
www.echo-health.eu
0
20
40
60
80
1000 2000 3000 4000
Patient at risk
DenmarkEnglandPortugal
SloveniaSpain
CI-90CI-95
CABG fatality rates in high volume hospitals
B1
C1 D1
B2 B3
A1 A3
A2
C2
C3 D3
D2
www.echo-health.eu
Efficiency (value)
Variation in low-value procedures is huge, within and across countries.
Areas with high number of low-value procedures are facing high opportunity costs
www.echo-health.eu
-5-4.5-4
-3.5-3
-2.5-2
-1.5-1-.50.51
1.52
2.5
ENG POR SLV SPN
Std Rate/10,000 21.45 7.92 22.17 12.05
EQ5-95 3.12 18.81 3.57 22.52
C-section in low-risk deliveries
www.echo-health.eu
Efficiency (value)
Hospitals are getting different outcomes regardless patients’ differences. On the other hand, hospitals are managing differently those resources devoted to treat similar patients. Hospital with poor outcomes and higher use of resources are likely providing lower-value care.
www.echo-health.eu
SHO
RTER
LOS
LESS MORTALITY
www.echo-health.eu
DENMARK
Calculated for female patient, aged 18-55, no comorbidities
www.echo-health.eu
SLOVENJA
Calculated for female patient, aged 18-55, no comorbidities
www.echo-health.eu
SPAIN
Calculated for female patient, aged 18-55, no comorbidities
www.echo-health.eu
PORTUGAL
Calculated for female patient, aged 18-55, no comorbidities
www.echo-health.eu
ENGLAND
Calculated for female patient, aged 18-55, no comorbidities