Elizabeth Docteur
Independent health policy consultant World Bank
Washington, DC
October 13, 2009
U.S. Health System Performance Viewed Through the Lens of InternaBonal
Comparisons
Overview of presentaBon
• IdenBfy key performance challenges – Access – Cost – Quality of care and populaBon health status
• Point to (hypotheBcal) factors explaining U.S. relaBve performance
• Suggest lessons from OECD experience
Access to care
• U.S. access to care for those with insurance is mixed, relaBve to other countries – RelaBvely short waiBng Bmes – New medicines reach U.S. market quickly; no “4th hurdle”
– Care is foregone due to affordability problems relaBvely oUen
Main U.S. access challenge is coverage: As in Mexico and Turkey, a significant share of US popula?on is
uninsured
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
99.9
99.7
99.5
97.6
97.3
27.3
67.2
50.4
10.2
35.8
59.2
62.1
89.6
98.0
99.0
0 20 40 60 80 100
Australia
Canada
Czech Republic
Denmark
Finland
Greece
Hungary
Iceland
Ireland
Italy
Japan
Korea
New Zealand
Norway
Portugal
Sweden
Switzerland
United Kingdom
France
Germany
Luxembourg
Spain
Belgium
Austria
Netherlands
Slovak Republic
Poland
United States
Turkey
Mexico
Total public coverage Primary private health coverage
Source: OECD Health at a Glance, 2007
Unlike Turkey and Mexico, U.S. rate of uninsured has not improved over last 15 years.
Being uninsured in the United States is associated with ge_ng less care, being less healthy and increased mortality (U.S. InsBtute of Medicine)
Why do coverage shoraalls persist?
• Coverage is voluntary – not automaBc and no mandate to purchase coverage (except in Mass.)
• Problems with availability of insurance – declining share of employers offer health benefits – individual market limits coverage for pre‐exisBng condiBons and insurers can reject applicants based on health risks
• Problems with affordability of insurance – risk raBng, adverse selecBon in voluntary risk pools
Some lessons from OECD experience
• Regulate insurance market to set the playing ground for compeBBon on basis of value in a mulB‐payer system – Dutch and Swiss examples – Risk adjustment
• Make coverage compulsory (or automaBc) – Swiss example
• Subsidize coverage for those who cannot afford it – Dutch and French examples
Cost outlier: U.S. Health spending greatly exceeds other countries’
Per capita spending, 2007 4
763
4 4
17
4 1
62
3 8
95
3 8
37
3 7
63
3 6
01
3 5
95
3 5
88
3 5
12
3 4
24
3 3
23
3 3
19
3 1
37
2 9
92
2 9
84
2 8
40
2 7
27
2 6
86
2 6
71
2 5
81
2 5
10
2 1
50
1 6
88
1 6
26
1 5
55
1 3
88
1 0
35
823
618
0
1 000
2 000
3 000
4 000
5 000
6 000
7 000
Uni
ted
Sta
tes
Nor
way
Sw
itzer
land
Lu
xem
bour
g (2
006)
1 C
anad
a
Net
herla
nds
Aus
tria
Fran
ce
Bel
gium
Ger
man
y
Den
mar
k
Irela
nd
Sw
eden
Icel
and
Aus
tralia
(200
6/07
)
Uni
ted
Kin
gdom
OE
CD
Finl
and
Gre
ece
Italy
Spa
in
Japa
n (2
006)
New
Zea
land
2
Por
tuga
l (20
06)
Kor
ea
Cze
ch R
epub
lic
Slo
vak
Rep
ublic
Hun
gary
Pol
and
Mex
ico
Turk
ey (2
005)
Private expenditure on health Public expenditure on health
Source: OECD Health Data, 2009.
Health expenditure as a share of GDP, 2006 !"
#$
!!#$
!!#!
!%#&
!%#'
!%#(
!%#!
!%#%
)#"
)#$
)#$
)#(
)#!
)#!
)#%
*#*
*#+
*#'
*#'
*#$
*#(
*#(
+#"
+#$
+#!
&#*
&#&
&#'
&#(
"#+
*#)
!
"
#
$%
$&
,-./0123/4/05
36./70894-1
:84-;0
<08=4->
BC8/A@49
DA5/8.4
E4-414
F0-=
48G
H0/I0894-15
H062J0494-1
36010-
<800;0
K;094-1
K/49>
LMEF
DA5/849.42N!O
HC864>
3P4.-
,-./012Q.-@1C=
RA-@48>
:.-94-1
S4P4-2N(O
K8094-1
TAU0=VCA8@
39CW4G2X0PAV9.;
E70;I2X0PAV9.;
Y0U.;C
QC804
BC94-1
ZA8G0>2N(O
[2C\2<FB
'()*+,'-+./01
(1) 2005/2006 (2) 2005 Source: OECD Health Data 2008
• Health expenditure is high share
of final U.S. household
consumpBon
7.1 8.1
8.6 11.2
10.5 16.8
11.0 12.8
11.6 11.8 11.9
10.4 12.0 12.4
11.8 12.9
15.4 13.2 13.3
12.4 11.9
13.9 14.7
12.6 15.0
14.3 14.8 14.9
16.9 14.9
19.8
0 5 10 15 20
United States United Kingdom1
Turkey (2005) Switzerland1
Sweden Spain
Slovak Republic Portugal (2006)
Poland OECD Norway
New Zealand2 Netherlands3
Mexico Luxembourg
(2006)4 Korea
Japan (2006) Italy
Ireland Iceland1 Hungary Greece
Germany France Finland
Denmark Czech Republic
Canada Belgium3 Austria
Australia (2006/07)
Source: OECD Health Data, 2009
Current health expenditure represents a relaBvely high share in U.S. final household consumpBon, 2007
What problems are associated with high U.S. health costs?
• Insurance is increasingly unaffordable – Especially for those who must buy on the individual market, where as
likle as half of the premium intake goes to pay medical claims – Wage increases for employed are dampened by rising insurance cost
• Problems in affordability of health care for the uninsured and underinsured – 62% of bankruptcies in 2007 related to health care costs
• Opportunity cost
• QuesBon of future sustainability
Why is U.S. health care so expensive?
Source: OECD Health Data, 2009.
AUS
AUT BEL CAN
CZE
DNK
FIN
FRA DEU
GRC
HUN
ISL IRL
ITA JPN
KOR
LUX
MEX
NLD
NZL
NOR
POL
PRT
SVK
ESP
SWE
CHE
TUR
GBR
USA
0
1 000
2 000
3 000
4 000
5 000
6 000
7 000
8 000
10 000 15 000 20 000 25 000 30 000 35 000 40 000 45 000 50 000 55 000 60 000
Richer countries spend more on health, although U.S. costs exceed those of countries with comparable
income Health expenditure and GDP per capita, 2007
RelaBvely high administraBve costs in (1) a mulB‐payer system that is (2) characterized by minimal standardizaBon compared to elsewhere (benefits,
payment levels, payment methods)
Share of total health expenditures allocated to administraBve expenses, 2004
(1) 2003 (2) 2002 OECD Health Data October 2006
1
1.7 1.8 1.92.3 2.4
2.7 2.8
3.5 3.5 3.74.1 4.3
4.8
6.2
7.5
10.210.7
7.6
4.4
0
2
4
6
8
10
12
Denm
ark
(2)
Hungary (2
)
Turkey
Portu
gal
Japan (1
)
Poland
Czech
Republic
(2)
Austra
lia
Korea
Spain
Belgi
um (1
)
Canad
a
Netherla
nds
OECD
Switz
erland
Ger
man
y
France
Unite
d Sta
tes
Mex
ico
Luxem
bourg
%
AdministraBve costs and profits account for half of premium for policies purchased in the individual market.
U.S. physicians earn more than counterparts in most countries Physician remunera?on, ra?o to GDP per capita
1.6
2.8
2.5
2.7
2.3
1.7
2.9
4.6
2.8
2.4
4.0
3.7
1.6
3.3
2.5
4.8
4.8
5.3
5.6
7.8
4.9
2.3
4.5
2.4
4.2
8.4
3.7
6.5
0246810
SalariedSelf-employed
Ratio to GDP per capita
Specialists
1.9
1.73.0
1.62.1
2.2
3.8
2.13.4
2.33.3
1.8
2.83.7
4.02.0
3.54.0
3.23.8
4.4
0 2 4 6 8 10
SalariedSelf-employed
Ratio to GDP per capita
General practitioners (GPs)
Australia (2004) Austria (2003)
Belgium (2004) 1 Canada (2004) Czech Republic
Denmark Finland
France (2004) Germany (2004)
Greece 2 Hungary Iceland
Ireland 3 Luxembourg (2003) 3
Mexico Netherlands New Zealand
Norway Portugal
Sweden (2002) Switzerland (2003)
United Kingdom (2004) United States (2001)
Other factors explaining high cost of U.S. health care
• More intensive service mix – Higher share of docs are specialists and U.S. uses more specialist‐intensive care, including elecBve surgery, even though physician consultaBon and hospital discharge rates are relaBvely low
• Physician incenBves to provide excess care to the insured – FFS, defensive medicine to avert malpracBce judgments, ownership of scanners
Some lessons from U.S. experience
• Greater reliance on salary and capitaBon payments helps with cost control, but may come at cost in terms of producBvity
• AcBvity‐based payments appear to encourage efficiency (more service for money), but may not have a posiBve impact on overall health‐system efficiency (less health improvement for money)
• Price controls, budgets and all‐payer rate se_ng can help control rate of growth, but may be an impact on Bmely availability of medicines and services
Quality of care
• U.S. quality of care good in some areas (e.g., cancer care), below average in others (e.g., renal care, asthma care); no parBcular area in which quality of care is excepBonal, relaBve to other countries (Docteur and Berenson, 2009)
• Some evidence that medical errors may be relaBvely more common in the United States
70.8
76.9
72.2
80.5
77.0
76.2
82.6
80.0
82.0
86.1
83.8
85.6
88.6
61.6
75.4
75.5
76.2
77.9
81.1
81.9
82.1
82.4
85.2
86.0
86.1
87.1
88.3
90.5
0 20 40 60 80 100
Poland
Czech Republic
Korea
Ireland
United Kingdom
OECD (14)
Norway
New Zealand
Denmark
France
Netherlands
Finland
Japan
Sweden
Canada
Iceland
United States
Age-standardised rates (%)
2002-2007 1997-2002
Breast cancer 5‐year survival rates, 1997 – 2002 and 2002 – 2007 or nearest available year
Source: OECD Health at a Glance 2009
Mammography, percentage of women aged 50 - 69 screened, 2005
54.7
55.6
60.8
63.0
69.5
70.4
81.9
98.0
0 25 50 75 100
23-country average3
*Australia1
**United States1
*New Zealand2
*United Kingdom
**Canada
*Netherlands
*Norw ay1
PercentageNotes:
* stands for program data whereas ** stands for survey data.
1.2003 2.2002
3. Includes Japan, Poland, the Slovak Republic, Mexico, the Czech Republic, Switzerland, Korea, Hungary, Australia, Belgium, Italy, Portugal, the United States, Iceland, New Zealand, the United Kingdom, Canada, France, Ireland, the Netherlands, Sweden, Finland and Norway.
Source: OECD Health Data 2007
Breast cancer mortality, female, 1995 to 2005
29.5
28.4
27.0
25.8
25.1
24.9
24.2
23.9
23.1
22.4
22.4
21.5
21.3
21.1
20.8
20.7
20.5
20.3
20.0
19.9
19.5
19.5
19.3
19.3
19.2
16.7
11.0
10.4
5.8
0
10
20
30
40
1995 2000 2005 Age-standardised rates per 100 000 females
Source: OECD Health at a Glance 2009
Amenable mortality
• As of 2002‐2003, the US has the highest rate of mortality due to preventable and treatable condiBons (amenable mortality) among 19 countries studied (Nolte and McKee, Health Affairs, 2008)
• This represents a decline in U.S. performance since 1997‐1998, when the U.S. was 15th among 19 countries studied. All countries experienced a decline in rate of mortality amenable to health care, but U.S. achieved a relaBvely small decline.
US health status below OECD average by some measures Life expectancy and infant mortality, 2006
USA to OECD avg. Life expectancy at birth (yrs): Total population Females Males
80.7 < 81.8 75.4 < 76.1 78.1 < 78.9
Life expectancy at age 65 (yrs): Females Males
20.3 > 20.2 17.4 > 16.8
Infant mortality rate (per 1000 live births) 6.7 > 5.1
Source: OECD Health Data 2009.
Life expectancy, Total population at birth, Years
Countries
1995 Total population at birth Years
2006 Total population at birth Years
Increase 1995-2006
Australia 77.9 81.1 3.2 Austria 76.6 79.9 3.3 Belgium 77.0 79.5 2.5 Canada 78.1 80.7 2.6 Czech Republic 73.3 76.7 3.4 Denmark 75.3 78.4 3.1 Finland 76.6 79.5 2.9 France 77.9 80.7 2.8 Germany 76.6 79.8 3.2 Greece 77.7 79.6 1.9 Hungary 69.9 73.2 3.3 Iceland 78.0 81.2 3.2 Ireland 75.6 79.7 4.1 Italy 78.4 81.2 2.8 Japan 79.6 82.4 2.8 Korea 73.5 79.1 5.6 Luxembourg 76.8 79.4 2.6 Mexico 72.5 74.8 2.3 Netherlands 77.5 79.8 2.3 New Zealand 76.8 80.1 3.3 Norway 77.9 80.6 2.7 Poland 72.0 75.3 3.3 Portugal 75.4 78.9 3.5 Slovak Republic 72.4 74.3 1.9 Spain 78.1 81.1 3.0 Sweden 78.8 80.8 2.0 Switzerland 78.7 81.7 3.0 Turkey 67.9 71.6 3.7 United Kingdom 76.7 79.1 2.4 United States 75.7 78.1 2.4 OECD Average 76.0 78.9 3.0 OECD Health Data 2009 - Version: June 09
Life expectancy at birth: US improvement since 1995 falls well short of avg improvement and even improvement among those with greatest longevity
Factors explaining U.S. performance in terms of health and quality
• IncenBves for overuse faced by health care providers (FFS payment, malpracBce encouraged defensive medicine)
• Lack of incenBves for prevenBon (insurers, providers): limited use of P4P, frequent change of coverage over lifeBme
• The uninsured (example: adult asthma admission rates)
• Limited use of health ICT applicaBons (e.g., EHR) that could promote evidence‐based care and help to avert errors
• Lack of integraBon/coordinaBon in the delivery system
• Health status shoraalls also explained by factors not directly in health system purview: violence, teen birth rate, segments of populaBon who are at a great disadvantage in terms of income, educaBon
Some lessons from OECD experience
• SBll at an early stage of research into what structural characterisBcs and policies contribute to top performance in quality of care
• Quality measurement and benchmarking is essenBal
• Improved health data and informaBon systems needed both to track and to improve quality of care – Unique paBent idenBfiers allowing for data linkage
Conclusions
• Every reason to believe that U.S. gets poor value for money, relaBve to other developed countries
• This is the case irrespecBve of whether increased spending over Bme has yielded benefits valued more than they cost
• Lessons from internaBonal experience may be useful to build upon strengths and address weaknesses, although naBonal context (i.e., insBtuBonal factors) and values very important
For more informaBon
• “OECD Health Systems: Lessons from the Reform Experience,” by E. Docteur and H. Oxley, OECD Economics Department Working Paper, 2003.
• “The U.S. Health System: Assessment and ProspecBve DirecBons for Reform,” by E. Docteur, H. Suppanz and J. Woo, OECD Economics Department Working Paper, 2002.
• OECD Health at a Glance, 2009 (forthcoming).
Top Related