Copyright © 1999-2003 Samuel L. Baker
Sound instructions: Click the sound buttons to play the
voice recordings.
Click again to stop the play.
If you are viewing this over a telephone modem internet connection,
playing the voice recordings involves loading delays of several minutes.
Cost-benefit analysis
I'll discuss:
- General concepts of cost and benefit
- A specific cost-benefit study
Measuring cost
- discussed in slide show on Cost
- Opportunity cost is key cost idea
Measuring benefit
- The economic textbook treatment of benefit
is based on "utility".
- Utility is an imaginary quantity that
goes up when you have more of something you like and down when you have
less.
Utility
- According to standard economic theory,
people go through life trying to maximize their utility.
- This idea pre-dates and ignores 20th-century
theories of psychology and sociology.
Measuring utility
- Strictly speaking, utility is not observable.
- Utility exists only in theory, or in the
thoughts of the person.
- We can observe how a person ranks alternatives,
giving us an ordinal utility scale.
- But we can’t compare or add one person’s
ordinal utility to another’s.
Measuring health
- For health care, we can measure utility,
if we identify it with length of life or health status,
- but that's a judgement call, and peoples'
values differ.
- For example, would you rather look good or feel good?
- Different people answer that question differently.
Cost-benefit and cost-effectiveness
analysis -- What is the difference?
- Cost-benefit analysis
- Both costs and benefits are evaluated
in dollars and compared.
- Cost-effectiveness analysis
- The costs of alternative means of achieving
some benefit are compared.
- The benefit itself is not evaluated in
dollars.
Cost-benefit analysis (CBA):
- Evaluate the objective -- the benefit.
- Evaluate the cost.
- Find if the cost is less than value of
benefit.
Cost-effectiveness analysis
(CEA):
- The objective -- the benefit -- is given.
- Find minimum-cost way to achieve it.
CBA: Business-like public decision-making
- Formal cost-benefit analysis seeks to
apply business-like decision-making to public expenditures.
- First used explicitly for water projects.
- Axnick et al do a cost-benefit analysis
of measles immunization.
- They attach a dollar value to each life
saved or disability averted to arrive at a dollar benefit for
immunizations.
CEA: Comparing alternatives
(Discussed in next lecture:) Neuhauser-Lewicki and Eddy do
cost-effectiveness
analysis.
They compare different applications of
medical services in terms of
- dollars spent per life saved (Neuhauser)
or
- dollars spent per per year of life saved
(Eddy).
Cost-effectiveness analysis:
CEA = Choosing the best way
- Saving lives is the objective, which CEA takes as given.
- CEA studies show where we
should put our money in order to save the most lives.
Both CBA and CEA show that economic decisions are unavoidable
- Neuhauser's stool guaiac CEA study shows:
You cannot decide what the proper course
of treatment is without explicitly considering economic factors.
- This conclusion is disturbing, but you can't get around it.
Axnick’s measles immunization study
implicitly acknowledges the necessity of economic decisions in health
care.
- Otherwise, why do such a cost-benefit analysis?
- What if the analysis had found that the
cost exceeded the benefit?
- Presumably, that would have implied that measles vaccination
is not a good idea!
Political context
At the time of the measles article, Congress
was considering folding Federal health care subsidies, including the
measles
immunization program, into a big block grant for states.
- States could spend block grant
money on whatever they choose.
- Each state could have decided whether to subsidize measles
immunizations. Public health advocates worried that some states might
have decided to spend the money on something else.
Cost-benefit example:
- Axnick, N.W., Shavell, S.M., Witte, J.J.,
"Benefits Due to Immunization Against Measles," PublicHealth Reports,
August
1969, 84(8), pp. 673-680.
Measles immunization benefits
- A classic study of its type.
- Axnick et al do a full cost-benefit analysis,
- by evaluating the lives saved and long-term
disabilities averted in dollar terms.
The value of a life saved
- In Axnick, saved lives are worth what
the person would have earned in the future.
- These future earnings are converted to
present values and added up.
- If the present value calculation is unfamiliar, try
interactive lecture 8 on discounting future income.
Health as a capital stock
Axnick’s method:
- treats health as capital
in the narrow sense of being a financial asset
- evaluates immunization just as you would
a financial investment
Harm from measles
- Measles can be a killer among the undernourished.
- Among the adequately fed, the main danger
is the 1/500 risk of encephalitis (brain infection) or other
complication.
- Most who get measles suffer discomfort
and loss of productivity for about two weeks.
Frequency of harm due to measles
In 1960, before the measles vaccine went into use:
- About 1/10,000 US measles cases resulted
in death.
- 3/10,000 had permanent brain damage requiring long-term care.
- For the rest who got measles, which was pretty much everybody in
those
days, measles caused loss of school
or work time. Axnick calculates a dollar value for this, too.
- Axnick didn’t attempt to put a dollar value on the discomfort.
Measles vaccine history
- Vaccine licensed in 1963.
- Sold or given to patients by private physicians
and state/local health departments.
- 1966: Federal government gave money to
states to subsidize vaccine distribution.
- From the 1970s, states required measles
vaccination for school attendance.
Risks of measles vaccine
- Measles vaccine does carry a risk of causing
illness,
- but the risk of injury from the vaccine
is much smaller than the risk of injury from measles.
- Measles is highly contagious. Before the
vaccine, nearly everybody got measles, which means that
nearly everybody was exposed to the risk of encephalitis.
- Quarantine ineffective.
Comparison of health care costs
with measles immunizations
- The cost of immunizations 1963-68 was
$108 million (Table 1).
- Axnick left out the cost of care of those
made ill by the vaccine.
- The estimated reduction in physician and
hospital services for measles was $77 million (Table 3).
- Immunization added $31 million
to measles care costs during 1963-68.
- $108 million spent minus $77 million saved equals a net health
care cost of $31 million for providing measles vaccinations.
The averted costs of long-term care shift the balance in favor of
immunization.
- Measles immunizations during 1963-1968 prevented thousands of
cases of lifelong disability. The present value of what would have been
the cost of "lifetime care for mentally retarded" was $201 million
(Table 3).
- Including this gives a net health care
cost saving of $170 million.
- $31 million net short-term cost (from above) of measles vaccinations,
minus $201 million in long-term care savings, equals
a net savings of $170 million.
Why calculate the "present value" of long-term
care costs?
Long-term ("institutional") care spending spreads out
over many future years, so Axnick et al used a
present value calculation.
They calculated, in effect, how big a
bank account you'd have needed at that time to be able to pay all the
future
long-term care costs of the encephalitis victims.
The $170 million net health care savings means that ...
There is no health vs. cost tradeoff
for society here
because the measles vaccination program saves money and
improves outcomes.
But Axnick wanted to make the case for
immunization stronger.
"Economic benefits"
- Deaths and disabilities valued at the
present value of earnings not earned.
Valuing life at the present
value of future earnings
- Based on marginal productivity theory
of wages:
- Marginal productivity = The additional production thanks
to hiring one more worker.
- Your contribution to production should,
in theory, be at least equal to what you get paid.
- Otherwise your employer would not hire you.
Valuing life at the present
value of future earnings
- Method used by Dorothy Rice in calculating
the cost of various illnesses in the U.S.
- See Axnick reference 10.
Equity problems
- Gender differences in income
- Men earned (and still earn) more than
women
- So men’s illnesses "cost" more.
- Saving a man’s life is worth more.
- Would tend to skew decisions in favor
of prevention of men’s deaths and disabilities
Equity problems
- Age
- Present value gives greatest weight to
near-future earnings.
- Less weight to distant future earnings.
- No weight to past earnings.
- People aged about 30 are most valuable.
- Children are less valuable, because their
earnings are further in the future.
- Retired elderly have no value.
Equity problems
- Age
- and present value of earnings
- People aged about 30 are most valuable.
- Children are less valuable, because their
earnings are further in the future.
- Retired elderly have no value.
"Economic benefits"
- The reduction in measles saved $216 million
in present value of earnings of people not killed or disabled,
- plus $37 million in work time not lost,
- for a total savings of $253 million.
Cost-benefit analysis therefore
is:
- for measles immunizations in U.S., 1963-1968
| Category |
Benefit + or cost - |
| Immunization costs |
- $108 million |
| Immediate measles care averted |
+ $77 million |
| Long-term care of encephalitis cases averted (present value) |
+ $201 million |
| Economic loss averted (present value) |
+ $253 million |
| Total of above -- the net benefit |
+ $423 million |
- The $423 million figure is used in Table
1 of the article.
Benefit-Cost Ratio
This text was changed after the voice recording was made.
The cost-benefit ratio is the benefit divided by the cost. This is used to make statements like, "This program gains $X for every dollar spent," where X is the benefit-cost ratio.
Even if you put aside all concerns about how you got the benefit-cost ratio, it still may not be correct to say "for every dollar spent" we get X benefit. Immunization programs are much more effective when they are universal than when they are small. A vaccine that is less than 100% effective can nevertheless completely control a disease if nearly everybody gets vaccinated. When a high percentage of a population is immune, the minority who are not immune are safe, because it is unlikely that they will contact an infected person.
In this example, here is the benefit-cost ratio:
$531 million benefit
- $ 77 million short-term care averted
- $201 long-term care averted
- $253 economic loss averted
divided by
$108 million cost of immunizations
= 4.92 benefit-cost ratio
You could say that on average a dollar
spent on measles immunizations returns $4.92 in benefits.
Bear in mind, though, the warning above about external benefit. It implies that the benefit of $100 million spent on immunizations is more than a million times better than spending $1.
Political success:
- Congress continued direct Federal subsidy
for measles immunizations.
- Economic argument was convincing.
Main ideas:
- Utility
- Cost-benefit analysis
- How it’s done
- Present value of future costs and savings
- Valuation of life lost -- equity problems
- Cost-effectiveness analysis
- How it differs from CBA
- Cost-Benefit Analysis -- evaluate the goal and the costs
- Cost-Effectiveness Analysis -- take the goal for granted and
find the least costly way to meet it
E-mail:
The views and opinions expressed in this page are strictly those of the
page author. The contents of this page have not been reviewed or
approved
by the University of South Carolina.