Transcript Measuring Health

A Health Report Card
for the Nation
David Cutler
Harvard University
ASHE Presentation, June 6, 2006
The Missing Ingredient in Health
Reform… Health
 Is Medicare Part D worth it?
 How big a problem are medical costs?
 Where should we invest additional money?
 All of these are questions about the balance
between health and money.
Health Measurement
 There is an enormous literature on health
assessment for cost-effectiveness analysis.
This is a good start, but:




It doesn’t add up to any total
Efficacy v. effectiveness
Double counting (medical spending for a
diabetic who has an MI)
Undercounting (surgeries get better over time)
An Analogy
 Suppose we wanted to know how the
economy is doing, but all we know about is:



Frozen custard sales in WI
Auto industry employment in MI
CEO wages
 We need to pull it all together
The Analogy: National Income and
Product Accounts
 World War I exposed fundamental macro questions
 How much could the military take without creating
supply shortages?
 How much could employment grow without creating
inflation?
 After the war, a group of economists started to gather
systematic data on the economy to prepare for future
questions


Led by Wesley Mitchell (Columbia) and Edwin Gay
(Harvard Business School), who founded the National
Bureau of Economic Research
Work fell to Willford King and Simon Kuznets
The Analogy: National Income and
Product Accounts (continued)
 In 1932, the Senate responded to the
Depression by ordering the Commerce
Department to report on the state of the
economy in 1929, 1930, and 1931



Simon Kuznets was loaned to the Commerce
Department to do this.
The first National Income and Product
Accounts took two years to produce
Commerce department decided to
institutionalize this.
The Analogy: National Income and
Product Accounts (continued)
 Simon Kuznets ultimately broke with the
Department of Commerce over the treatment
of non-market activities (Kuznets was in
favor; BEA was opposed).
 Issue of non-market activities is still front and
center.
The Analogy: National Income and
Product Accounts (continued)
 National Health Accounts are the natural
adjunct to National Income and Product
Accounts
 Recommendation 6.1: A health satellite account should be
produced by the Bureau of Economic Analysis in collaboration
with the Centers for Medicare and Medicaid Services of the U.S.
Department of Health and Human Services.
Abraham and Mackie, Beyond the Market: Designing
Nonmarket Accounts for the United States, National
Academy of Sciences, 2005.
Outline
 National Health Accounts: A Conceptual Basis
 The process of constructing accounts
 Health Accounts: A First Pass
 Population health assessment
National Health Accounts: A
Conceptual Basis
The Health Economy
Inputs
 Medical care
 Time investment
 Other consumption
(cigarettes)
 R&D
 Environment
Outputs
 Health (QALE)
 External (income)
effects from being
healthier
An Analogy: National Income and
Product Accounts
Inputs
 Labor
 Capital
 Raw materials
Outputs
 Total sales
Line
2002
2003
1
Gross domestic product
10,481
10,988
2
Personal consumption expenditures
7,385.3
7,757.4
911.3
941.6
3
Durable goods
4
Nondurable goods
2,086.0
2,209.7
5
Services
4,388.0
4,606.2
1,589.2
1,670.6
1,583.9
1,673.0
5.4
-2.4
6
Gross private domestic investment
7
Fixed investment
12
Change in private inventories
13
Net exports of goods and services
-426.3
-495.0
20
Government consumption expend
-itures and gross investment
1,932.5
2,054.8
679.5
757.2
1,253.1
1,297.6
21
Federal
24
State and local
Line
2002
2003
1
National income
9,290.8
9,707.8
2
Compensation of employees
6,019.1
6,203.0
3
Wage and salary accruals
4,974.6
5,100.2
6
Supplements to wages and salaries
1,044.5
1,102.8
9
Proprietors' income with IVA and CCAdj
797.7
846.9
12
Rental income of persons with CCAdj
173.0
164.2
13
Corporate profits with IVA and CCAdj
904.2
1,069.9
14
Taxes on corporate income
195.0
224.9
15
Profits after tax with IVA and CCAdj
709.1
845.0
18
Net interest and miscellaneous payments
582.4
583.2
19
Taxes on production and imports
760.1
788.7
20
Less: Subsidies 1
38.2
48.2
21
Business current transfer payments(net)
89.8
95.2
25
Current surplus of government
enterprises 1
2.8
5.0
Current National Health Accounts
Inputs
 Medical spending by:


Payer
Recipient of funds
Outputs
Issues
 Want outcomes too
 Need to relate inputs to outputs
 The ‘disease’ is the natural way to do this.
Ideal National Health Accounts
Inputs
 Medical spending by:

Disease
 Time investment
 Other consumption
(cigarettes)
 R&D
 Environment
Outputs
 Health by:

Disease
What Do We Need To Do
1. Measure the population’s health
2. Attribute that to particular conditions
3. Measure spending by condition
4. Build models that link conditions and their
treatments to costs and outcomes.
I have been working on this with
 Allison Rosen, U of Michigan
 Susan Stewart, Harvard and NBER
 Rebecca Woodward, Harvard and NBER
 Hsou May, U of Michigan
 Emily Shelton, U of Michigan
 And others
A More Complex Version:
Including Non-Fatal Health
Conceptual Basis
 We operationalize health as QALE.

How many years of quality adjusted life can a
person today expect to live?


The population is healthier today if the average
person has a larger QALE.
In making this comparison, we hold the
population age and gender distribution constant
at the 2000 level.
Conception of non-fatal health
Health
Domain 1:
Domain 2:
Domain 3:
Symptoms / Impairments
Symptoms / Impairments
Symptoms / Impairments
Disease 1
Disease 2
Disease 3
Step 1: Relate self-reported general health
to symptoms and impairments
 Ordered probit regression model.
 Scale each symptom/impairment to a 0-1
QALY metric using the estimated range of the
self-rated health scale
 Hold constant disutility of each symptom/
impairment as calculated in 2000.
 Note: these don’t change much over time.
Source: Stewart, Woodward, Rosen, and Cutler, “A Proposed Method for Monitoring U.S. Population Health:
Linking Symptoms, Impairments, Chronic Conditions, and Health Ratings,” NBER WP 11358.
Step 2: Relate symptoms and impairments
to diseases
 Probit regression model.
 Calculate impact of each disease on each
symptom/impairment using regression
coefficient and prevalence of that disease.

Impact of diseases do change.
Data Sources




NMES (1987) and MEPS (2000)
MCBS (includes institutionalized)
NHANES (disease measurement)
Disease-specific data:



SEER
Framingham Heart Study
…
Comparable Symptoms/Impairments
(NMES 1987 – MEPS 2000)

Primary activity limitations

Social/secondary activity limitations

Walking

Bending/lifting

Self-care

Depressive symptoms

Anxiety symptoms

Vision problems

Hearing problems
Would like to have
cognitive functioning.
g
ea
rin
H
Vi
si
on
us
xi
o
An
ss
iv
e
re
ft
al
ki
ng
e
ar
nd
/L
i
Be
W
lfc
Se
da
ry
Ac
ti v
ity
ec
on
y
D
ep
ci
al
/S
So
Pr
im
ar
Decrement on 0 to 1 scale
Symptoms/impairments with
largest decrements
0.25
0.20
0.15
0.10
0.05
0.00
Change in Health, 1987-2000
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
Life Expectancy
QALE
Overall
Overall
White
Males
LE/QALE
At Birth
Black
White
Black
Females
Weighted average of LE/QALE at each age
QALE increase by impairment
depressive symptoms
hearing
5.000
secondary activity
4.000
anxiety symptoms
3.000
vision
2.000
bending/lifting
1.000
walking
0.000
-1.000
Overall
Males
Females
primary role activity
increased LE
Impairment Question Wording
Primary
Activity
Walking
Bending
Vision
NMES
MEPS
Health keeps you from
working at a job, doing
work around the house,
or going to school
Limited in any way in the
ability to work at a job, do
housework, or go to
school because of an
impairment or a physical or
mental health problem
Any trouble walking one
block because of your
health
Difficulty walking about 3
city blocks or about a
quarter of a mile
Trouble bending, lifting,
or stooping because of
your health
Difficulty bending down or
stooping from a standing
position to pick up an object
from the floor or tie a shoe
Any difficulty seeing (with
glasses if you wear them)
Any difficulty seeing (with
glasses or contacts, if
used)
NMES
MEPS
14%
10%
13%
8%
12%
8%
10%
4%
If health care explains 50% of health
improvement
$140,000
$120,000
$100,000
$80,000
$60,000
$40,000
$20,000
$0
Birth
Age 15
QALE
Age 45
LE Only
Age 65
Conclusions
 The population is healthier in 2000 than in 1987.

Gains larger for men and for Blacks

QALE gains are due more to:


LE increases for men
primary activity, walking improvements for women
 QALE increase is about twice LE increase.
 The productivity of medical care appears to be high.
Summary
 National health measurement is necessary and
possible
 Health has improved immensely, more than enough
to justify the large increase in medical spending

Quality is as important as quantity
 Disease-based models will help evaluate what we
have done and simulate future possibilities.