Добавил:
Sekretar
kiopkiopkiop18@yandex.ru
t.me/Prokururor I Вовсе не секретарь, но почту проверяю
Опубликованный материал нарушает ваши авторские права? Сообщите нам.
Вуз:
Предмет:
Файл:Ординатура / Хирургия / Библиотека им академика М.И. Перельмана / Книга_2946_Библиотеки_им_академика_М_И_Перельмана
.pdf
Cost- effectiveness analysis 325
These resources may be owned jointly (by the clients of an insurance company) or individually (by a person who must
make an out-
of- pocket co- payment for a service covered partially by health insurance). Cost- effectiveness analysis can
inform policymakers and individual patients.
15.2 Cost- effectiveness analysis: amethod forcomparing management strategies
Cost- effectiveness analysis is a method for comparing decision alternatives by their relative costs and effectiveness. It
is an analytic tool for comparing the costs and effects of an intervention to at least one alternative. The results are
expressed as a ratio of incremental cost to incremental effect, where “incremental” is the difference between the two
interventions. In this chapter, health outcomes are the measure for comparing interventions, such as cases of a disease-
prevented or quality- adjusted life years (QALYs) gained. Cost- effectiveness analysis always compares strategies. The
incremental cost- effectiveness ratio (ICER) is calculated as the difference in costs between two compared alternatives
(net costs), divided by the difference in their health outcomes (net effectiveness). The ICER estimates the additional
cost of the intervention to buy one more unit of health.
When comparing two treatments, Treatment A and Treatment B, the incremental cost-
effectiveness of Treatment A
relative to Treatment B is:
IncrementalCE
Cost TreatmentA Cost TreatmentB
Effectivene
–
sss TreatmentA EffectivenessTreatment B
The incremental cost in the numerator represents the additional resources required from using Treatment A instead
of Treatment B. The incremental effect in the denominator represents the additional health outcomes by using Treatment
A rather than Treatment B.
Cost-
effectiveness analysis comes into play when decision makers are trying to choose among several interventions,
no intervention is both less expensive and more effective, and resources are limited. These choices should be based on
a comparison of the health benefits, harms, and costs associated with the available alternatives. Cost-
effectiveness
analysis is designed to identify the way to spend on health that gives the most health for our health care dollars– as
individuals or as a society.
15.2.1 Using cost- effectiveness analysis toset institutional policy: anextended example
Cost- effectiveness analysis may be used to set a policy that will affect the actions of others. Consider the problem of a
prepaid group practice administrator who must decide between three strategies for managing the care of a patient
with continuing pain from a kidney stone. A consultant has just presented the administrator with the following table:
Strategy Cost Life expectancy
A $9400 19.60
B $10 000 19.64
C $10 000 19.28
The administrator notes right away that choosing between these strategies will require a compromise: Strategy A is
the least expensive and Strategy B leads to the longest average survival. In deciding whether the additional effective-
ness of Strategy B is worth its extra cost, the administrator calculates the incremental cost- effectiveness of Strategy B.
Strategy B costs $600more per patient than Strategy A and prolongs life 2weeks (0.04 years) longer than Strategy A.
The administrator is frustrated. The cost- effectiveness analysis clarified the problem, but it did not lead to an easy deci-
sion. To understand the administrator ’s dilemma, let us go back to the beginning and find out how the analysis was
performed.
The administrator of a hospital- based group practice must decide whether to accede to the wishes of the chief of
urology who has asked the hospital to purchase an instrument that dissolves kidney stones by ultrasonic waves.
Kidney stones form in the kidney and eventually pass through the ureter to the urinary bladder, causing severe pain
called renal colic. Occasionally, a stone does not pass into the bladder, and renal colic continues until the stone is
removed surgically or dissolved. The urology chief proposes that all patients who do not pass their kidney stone
https://t.me/medicina_free
326 Medical decision making
within 48 hours should undergo treatment with high- energy ultrasonic waves to dissolve the stone. When this
treatment is successful, fewer patients must undergo surgery to remove the kidney stone. Avoiding surgery will be
especially important for patients whose risk of death from surgery is increased because of their poor medical
condition.
The administrator asks the consultant to investigate this request. After investigation, the consultant comes up with
the following facts:
• Fatality rate with surgery:
• Low- risk patients: 2%
• High- risk patients: 10%
• Success rate of treatment:
• Ultrasonic therapy: 80% for all patients
• Surgery: 100% for all patients who survive surgery
• Prevalence of patients at high risk from surgery: 20%
• Cost to the practice of treatments:
• Ultrasonic treatment: $2000 (includes the purchase price of the ultrasound equipment amortized over its
lifetime)
• Surgery: $10 000
After hearing these facts, the administrator is convinced that the ultrasonic treatment may be beneficial to some
patients but worries about the cost of purchasing the machine. Perhaps the surgeons could send patients who are too
sick for surgery to a nearby hospital that has recently purchased an ultrasonic machine. Accordingly, the consultant
investigates a resource-
sharing arrangement whereby low- risk patients would have surgery locally and high- risk
patients would be sent to the nearby hospital.
The administrator negotiates the following arrangement. The neighboring hospital will charge $4000 for an ultra-
sonic treatment. If the ultrasonic treatment fails, the patient will undergo surgery at the neighboring hospital, which
will charge the practice $15
000 to perform the surgery.
The administrator asks the consultant to analyze the choice between:
• Surgery for all patients (the current mode of treatment);
• Ultrasonic treatment for all patients (with surgery if the ultrasonic treatment fails);
• Surgery for low- risk patients; ultrasonic treatments at the neighboring hospital for high surgical- risk patients and
surgery if ultrasonic treatment fails.
The administrator asks the consultant to analyze the problem in terms of costs to the practice and length of life for
the average kidney stone patient, who is 55 years old and has a 20-
year life expectancy. The consultant uses the follow-
ing sequence to perform the analysis.
I. Define the problem to be solved and the objectives of the analysis
The problem: In some patients with renal colic, a kidney stone does not pass spontaneously. Which of the three
strategies for treating such patients should the administrator recommend for adoption by the practice?
The objectives: Predict the consequences of the three strategies by analyzing the expected costs and expected sur-
vival for 55-
year- old patients. If possible, identify a dominant solution which will both maximize survival and
minimize costs. If there is no dominant solution, use incremental cost-
effectiveness analysis to characterize the
decision alternatives.
II. Define the consequences of each decision alternative
The consultant represents the problem by a decision tree in which the probabilities and outcomes are those
identified in the first phase of the investigation. The trees for each decision alternative are displayed sepa-
rately(Figures15.1, 15.2, and15.3). The analyst chooses two outcomes: the patient’s survival and the costs to the
practice.
Treatment strategy 1: Surgery for all patients (the current mode of treatment)
Survival:
• Immediate death due to surgery: Life expectancy 0 years.
• Survive surgery and live to one’s normal life expectancy of 20 years.
Treatment costs per patient:
• Costs: Surgery for all patients: $10 000.
https://t.me/medicina_free
Cost- effectiveness analysis 327
LE = 20, $10 000
LE = 0, $10 000
LE = 20, $10 000
LE = 0, $10 000
High risk
Surgery
Low risk Surgery
Ok
Ok
Die
Die
P = 0.9
p = 0.98
P = 0.1
P = 0.02
P = 0.2
P = 0.8
Maintain status quo; operate on everyone
Figure15.1 Outcomes of surgery within the practice for all patients with persistent renal colic.
Low surgical
risk Do US
risk
Do US
Ok p = 0.98
Die p = 0.02
US successful p = 0.8
LE = 0, $12 000
LE = 20, $12 000
LE = 20, $2000
Ok p = 0.90
Die p = 0.10
US fails
surgery
US fails
surgery
LE = 20, $2000
US successful p = 0.8
LE = 0, $12 000
LE = 20, $12 000
High surgical
Ultrasound for everyone
Figure15.2 Outcomes of a strategy of doing ultrasonic treatment on all patients and performing surgery if ultrasonic treatment fails, both
treatments done within the local practice.
Low risk
Do surgery
High risk
Refer for US
Ok
p = 0
.9
Ok
p = 0
.98
Die
p = 0
.1
Die
p = 0
.02
US successful
US fails
surgery
LE = 0, $10 000
LE = 20, $10 000
LE = 0, $19 000
LE = 20, $19 000
p = 0
.8
p = 0
.2
p = 0
.8
p = 0
.2
LE = 20, $4000
Send high risk patients for US at neighboring
hospital; if US fails, do surgery there
Figure15.3 Outcome of doing surgery within the practice on low- risk patients with persistent renal colic and sending high- risk patients to a
neighboring hospital for ultrasound treatment (and, if it fails, surgery at the neighboring hospital).
https://t.me/medicina_free
328 Medical decision making
Treatment strategy 2: Purchase equipment for ultrasonic treatment. Perform ultrasonic treatment for all patients, with
surgery if the treatment fails. All interventions to be performed within the practice;
Survival:
• Immediate death due to surgery: Life expectancy 0 years.
• Survive surgery and live to one’s normal life expectancy of 20 years.
• Successful ultrasound treatment: Live to one’s normal life expectancy of 20 years.
Treatment costs to the practice:
• Outcome 1: Ultrasound treatment successful: cost $2000.
• Outcome 2: Ultrasound fails. Total cost: cost of ultrasound $2000 + cost of surgery: $10 000.
Treatment strategy 3: Surgery for all low-
risk patients at local hospital; ultrasonic treatment at the neighboring hospi-
tal for patients with high surgical risk; surgery at the neighboring hospital if ultrasound fails.
Survival:
• Immediate death due to surgery: Life expectancy 0 years.
• Survive surgery and live to one’s normal life expectancy of 20 years.
• Successful ultrasound treatment at the neighboring hospital: live to one’s normal life expectancy of 20 years.
Treatment costs to the practice:
• Outcome 1: Surgery at a local hospital for low- risk patients: cost $10 000.
• Outcome 2: Ultrasound at neighboring hospital; surgery at neighboring hospital if it fails. Total cost: cost of ultra-
sound $4000; cost of surgery: $15
000.
III. Average out and fold back the decision tree
The results are shown in the following table:
Strategies Expected cost Life expectancy of patients
Strategy 1: surgery for everyone $10
000 19.28 years
Strategy 2: Buy ultrasound; use it for all patients $4000 19.86 years
Strategy 3: Surgery for low- risk patients; Send high- risk
patients to neighboring hospital for ultrasonic treatment
$9400 19.60 years
What does this analysis mean? To interpret it, examine each of the columns of the table in turn.
Expected cost: The expected cost of each of the three decision alternatives was obtained by averaging out and folding
back, using cost as the measure of outcome.
Operate on everyone
Buy ultrasound
for your practice
Surgery for low- risk patients; high- risk
patients to a nearby hospital for ultrasound
Expected cost $10
000 $4000 $9400
Interpretation: Purchasing the ultrasonic machine is the least costly alternative, presumably because most patients
can be treated with ultrasound and do not require a $10
000 surgical operation. Referral of all patients to the nearby
hospital is expensive because of its high charges for ultrasound and surgery if needed.
Life expectancy: The life expectancy of the patients is obtained by averaging out and folding back the decision trees,
using life expectancy as the measure of outcome.
Operate on everyone Buy ultrasound for your practice High- risk patients are sent to a nearby hospital
Life expectancy 19.28 years 19.86 years 19.60 years
Interpretation: Purchasing the ultrasound machine leads to the longest expected length of life, presumably because
fewer than 10% of the patients are subjected to the risk of death from surgery. In contrast, many more patients would
undergo surgery with the other two decision alternatives.
Purchasing the ultrasonic machine for the practice leads to the lowest costs for the practice and the longest life expectancy for
the patient. Since purchasing the ultrasound machine has the lowest cost and the best survival, it dominates the other choices.
The hospital administrator is relieved to learn that they do not need to analyze the tradeoff between cost and survival.
https://t.me/medicina_free
Cost- effectiveness analysis 329
As the administrator is about to place the order for the equipment for performing ultrasonic dissolution of kidney
stones, they learn about a worldwide shortage of materials for a crucial part of the new version of the ultrasound
machine. The company has shut down its production line and has no idea when this model will become available
again. The alternative is to buy an older, more expensive model, which is less effective at dissolving kidney stones. The
probability of successful treatment is only 0.50with the older machine. The cost to the practice of each procedure will
be $5000 using the older machine, rather than $2000. However, the neighboring hospital, which has the new ultra-
sound machine, can still offer the procedure for $4000. The consultant repeats the analysis after substituting the higher
cost and reduced effectiveness of an ultrasound treatment using the older machine and presents the following table:
Strategies Expected cost Life expectancy of patients
Strategy 1: Surgery for everyone $10
000 19.28 years
Strategy 2: Buy older model ultrasound; use it for all patients $10
000 19.64 years
Strategy 3: Surgery for low-
risk patients; send high- risk patients
to the nearby hospital for ultrasound
$9400 19.60 years
The situation is now quite different. The least expensive alternative is different from the alternative that produces the
longest life expectancy for patients. No strategy dominates. The administrator could choose based on either cost or
survival but decides to use the ICER, which estimates the additional cost of an intervention to achieve one more unit
of clinical outcome. The administrator must first choose which two alternatives to compare.
Strategy 1 is as costly as Strategy 2 and more costly than Strategy 3. Strategy 1 also leads to a shorter life expectancy
than either Strategy 2 or 3. So, it is inferior to both. Therefore, the choice is between Strategy 2 and Strategy 3. Strategy
2 is more costly than Strategy 3 but more effective. So, choosing between them is difficult.
The administrator notes that patients’ life expectancy will be slightly better if the practice buys its own ultrasound
machine (Strategy 2). What will it cost the practice to choose Strategy 2which will avoid the loss of life with Strategy
3, which involves doing surgery on all low-
risk patients and the high- risk patients who fail ultrasonic treatments at the
neighboring hospital? Would purchasing the ultrasonic machine be an extravagant departure from the hospital’s usual
management practices? How will the cost per year of life gained compare to other decisions that the practice has made
recently? The average cost per procedure of buying the ultrasound machine is only $600more than if high-
risk patients
are sent to the other hospital. The incremental cost- effectiveness of Strategy 2 (buying the ultrasonic machine) vs.
Strategy 3 (surgery for low- risk patients; send high- risk patients to nearby hospital for ultrasound) is calculated as
follows:
ncremental CE
Cost Strategy Cost Strategy
LE strategy
23
2
–
LE Strategy 3
ncremental CE
$$
yearsyears
$per y
10000 9400
19 64 19 60
15000
–
..
eear of life saved
The improvement in life expectancy that can be obtained by buying the ultrasound machine will cost an additional
$15 000 per additional year of life gained, which is consistent with past investments by the hospital. In fact, $15 000 per
year of life saved compares very favorably to the cost- effectiveness of services that everyone seems to agree are an
efficient use of resources. Clearly, this expenditure would be defensible in a presentation to the hospital board of
trustees.
Cost- effectiveness analyses often express the gain to the patient as QALYs gained rather than life years saved. QALYs
are life years in a health state multiplied by a measure such as the patient’s utility for the quality of life in the health
state. The topic of QALYs is explored further in Section15.7
15.2.2 Flat- of- the- curve medicine
Often, one must choose from among several ways to use a service. For example, in screening for cancer of the cervix,
one can obtain cervical cytology at any frequency. In practice, the typical frequency ranges from annual testing to test-
ing every 3 years. To help analyze this problem, some investigators plot the cost of adopting each policy against its
benefit (Figure15.4). This analysis shows the small increment in effectiveness relative to the cost of more intensive
https://t.me/medicina_free
330 Medical decision making
screening for this hypothetical cancer. Practicing medicine with policies that provide a relatively small incremental
benefit for the added cost is sometimes called “flat- of- the- curve medicine.” How should one use the information in
Figure15.4 to choose a policy for cancer screening?
In principle, one should adopt the most intensive screening program that has a higher incremental cost- effectiveness
than alternative uses for the resources. Cost- effectiveness analysis is a method for comparing decision alternatives, but
to decide whether any alternative is cost- effective, a decision maker must have a criterion for cost- effectiveness. As
noted in the preceding section, $50
000–$150 000 per QALY is often used. This, or any, threshold for calling a practice
cost- effective has neither a theoretical basis nor a valid empirical rationale. The Second Panel on Cost- Effectiveness
recommends against adopting one specific threshold for declaring a practice to be cost- effective. Instead, researchers
should highlight how clinical or policy recommendations might change across a range of thresholds, which an organi-
zation could use to inform its resource allocation decision making (Neumann etal.,2016). Cost–benefit analysis, our
next subject, provides a direct approach for deciding if a program is worth undertaking.
15.3 Cost–benefit analysis: amethod formeasuring thenet benefit ofmedical services
Definition of cost–benefit analysis: A comparison in which the costs and benefits of a service or services are both expressed
in the same units.
To use cost–benefit analysis to compare different programs or policies, the analyst calculates net benefits by subtract-
ing costs from benefits (B−C). Analysts also use the ratio of benefits to costs to compare policies, although net benefit
provides a more easily explained meaning of the comparison: the gain or loss from an action, measured in units whose
meaning is tangible.
15.3.1 The distinction between cost–benefit analysis andcost- effectiveness analysis
Many people have difficulty distinguishing between cost–benefit analysis and cost- effectiveness analysis. The distinc-
tion is quite subtle but important.
Cost- effectiveness analysis is helpful for comparing alternatives but is otherwise a limited measure for decision
making. It can guide the choice between alternatives (pick the one with the most favorable incremental cost per QALY
gained). However, to decide if either alternative is worth doing, one must choose an arbitrary threshold of cost-
effectiveness. For example, you should have no difficulty choosing between a service whose cost per QALY gained is
$1 000 000 and one whose cost per QALY gained is $100 000. To decide whether either of the services is worth its cost,
you must first establish a definition of a service that is worth its cost (e.g., $50 000 per life year gained). A cost- effectiveness
threshold is arbitrary because the unit of measure of costs (currency) is different than the unit for measuring effective-
ness. Therefore, the units of cost- effectiveness do not have a natural, common- sense meaning that can serve as a
framework for setting a threshold and convincing others that it is a reasonable basis for policy making.
1.0
0.8
310
0.6
15
0.4
6
0.2
30
0
Number of cervical cytology tests
between ages 20 and 50
Fractional reduction in risk of
cervical cancer
Figure15.4 Incremental costs and incremental effectiveness of several methods for screening for hypothetical cancer.
https://t.me/medicina_free
Cost- effectiveness analysis 331
This limitation becomes clearer when we contrast cost- effectiveness analysis with cost–benefit analysis, in which
the measure of benefit and cost is monetary. Because the two measures have the same units (currency), the difference
between them has the common- sense, real- world meaning of profit or loss, which is a natural criterion for deciding if
a service is worth doing. Cost–benefit analysis has an important role to play when the benefits of a policy can be
expressed in units of currency, and there are many valid applications in the policy field.
To summarize, cost–benefit analysis is one of several methods for deciding between alternatives, but its unique role
is as a method for deciding whether a service is worth doing. Another advantage: choosing between policies with dif-
ferent outcomes is possible if the outcomes can be transformed into the same monetary measure, such as dollars.
15.3.2 Placing amonetary value onhuman life
Although cost–benefit analysis is potentially very powerful, it has one major drawback when applied to decisions that
affect human health. Cost is usually expressed in units of currency (dollars, pounds, euro, or yen). Since the outputs of
a service must also be in units of currency, the analyst faces a very difficult problem: how does one place a monetary
value on health outcomes? The measure of the output of many medical policies is additional years of healthy life.
Therefore, to apply cost–benefit analysis to medical problems, one must ask, “How does one place a monetary value
on an additional year of life?” The methods for expressing the value of years of life in monetary terms are far from
satisfactory. Here, we list several approaches.
How were past decisions valued? The size of an investment by society in a program designed to save lives and the
number of lives actually saved is one measure of the value that society places on a life. However, these societal invest-
ments occur in a context that is unique to each occurrence. Generalizing from one such situation to another is risky.
The human capital method: This approach values a policy by its effect on the patient’s lifetime earnings. A life saved at
age 55 years means at least 10more years of gainful employment. The monetary value of that period in the workforce
can be calculated.
The human capital approach is widely used, although it has several deficiencies. By valuing everything by antici-
pated income, the human capital approach implies that the life of a person with a lifetime income of $100 000 is worth
twice as much as a person with a lifetime income of $50 000. This notion may have validity in economic terms, but
counter- examples are easy to imagine. The human capital approach also assumes that people are unwilling to pay
more than their remaining lifetime earnings to postpone death. In fact, people with savings or a home might be willing
to spend heavily from these reserves to postpone death. The next measure of the value of life is the amount that people
are willing to pay to postpone death or avoid disability.
Willingness to pay: The willingness-
to- pay method helps people express a value for life by saying how much they
would be willing to pay for a program that would decrease the probability of a bad outcome by a stated amount.
• The subject might be told:
• Suppose that the probability that you will die from a stroke is 0.01 during your remaining lifetime of 20 years.
What is the most that you would be willing to pay over your lifetime for a program that would reduce this risk
by 50%?
• The subject might reply: “$2000.”
Interpretation: A lifetime probability of 0.01 is equivalent to a risk of 10 stroke deaths per 1000individuals over 20
years. A reduction by 50% would eliminate 5 deaths per 1000individuals over 20 years. Since 5 stroke deaths were
avoided by 1000 people each paying $2000 (total $2
000 000) did not occur during those 20 years, 100life- years would
be gained, or $20 000 per life- year saved.
By asking such questions of a community sample, one can calculate the value that the populace places on saving a
life. However, what one is willing to pay is likely to depend on the exact circumstances. The premise of the willingness-
to- pay approach is that a person who is willing to pay $2000 to reduce the probability of death from a stroke by
5strokes per 1000individuals over 20 years would be willing to pay $400
000 ($2000 divided by 0.005) to avoid certain
death from a stroke. Many would question that premise: reducing the probability of an outcome is not the same cog-
nitively as completely avoiding a certain outcome.
Obtaining a consistent value for human life with the willingness- to- pay approach is difficult for other reasons.
People’s willingness to pay increases with the risk of death. Does this mean that the value that is placed on life depends
on the risk of losing it? Wealthy people are generally willing to pay more for a program to reduce the risk of death than
poor people. People are more willing to pay when they personally benefit from a program than when they are asked
to pay for a program that will benefit others.
Clinicians may not accept the results of cost–benefit analysis because it is so difficult to develop a consensus opinion
on the monetary value of a year of human life. Most clinicians who are interested in analyzing tradeoffs between
https://t.me/medicina_free
332 Medical decision making
clinical strategies use cost- effectiveness analysis. Perhaps policymakers should be paying more attention to the
principal advantage of cost–benefit analysis: because the costs and benefits of a program are both measured in units of
currency, they can express the tradeoff between them with a tangible measure that the public can understand (profit or
loss, in units of currency).
15.3.3 Should clinicians take aninterest incost–benefit analysis?
The problems with applying cost–benefit analysis seem remote from daily practice. The individual clinician is not
concerned with allocating effort between programs to improve the health of the community. The clinician’s primary
responsibility is to look after the interests of the patient. However, the choices open to the clinician may be affected by
resource allocation decisions that are based on cost–benefit analysis. Policymakers must deal with such methods when
they try to decide whether to allocate scarce government resources to screening programs to detect disease in young
people or programs to treat chronic disease in old people. If clinicians want policymakers to listen to them, they should
understand the methods that policymakers depend on and help them to apply these methods to the world of medical
practice. That said, the foundations of cost–benefit analyses in health do not inspire confidence.
15.4 Methodological best practices forcost- effectiveness analysis
As described earlier, cost- effectiveness analysis provides a framework for considering the costs, benefits, and
harms of available alternatives. Benefits can impact both mortality, and morbidity and can accrue differently to
different sectors of society (i.e., to individual patients, caregivers, health systems, or populations). The time during
which the harms, benefits, and costs may accrue can also vary. These complexities challenge both the analyst and
the consumer of cost-
effectiveness analyses. Because of its potential effects on public policy, maintaining public
trust in cost- effectiveness analyses is important. A list of methodological best practices is an important part of the
foundation of a scientific discipline. In 1993, the US Public Health Service convened a Panel of 13nongovernment
scientists and scholars with expertise in economics, clinical medicine, ethics, and statistics. Their charge was to
review the state of cost-
effectiveness analysis and to develop recommendations for the conduct and use of cost-
effectiveness analysis in health and medicine. The resulting Gold Book (Gold 1996) published in 1996 by the origi-
nal Panel on Cost- Effectiveness in Health and Medicine became the point of reference for doing cost- effectiveness
analysis. Its recommendations guided the application of cost- effectiveness analysis for a generation of decision
analysts, economists, and policymakers. Since 1996, the field of cost- effectiveness analysis and our learnings from
the application of its methods have advanced significantly. The original Panel’s efforts needed updating, and in
2016, the Second Panel on Cost-
Effectiveness in Health and Medicine published their expanded recommendations
(Neumann etal.,2016).
The Second Panel on Cost- Effectiveness in Health and Medicine includes recommendations for the conduct, methods,
and reporting of cost- effectiveness analyses. These recommendations seek to ensure the transparency of cost- effectiveness
analyses so that decision makers can understand the tradeoffs of costs, harms, and benefits between strategies. Key
recommendations for cost-
effectiveness analysis include:
• All cost- effectiveness studies report a Reference Case analysis based on a health care sector perspective and
another Reference Case analysis based on a societal perspective. The Reference Cases are defined by recommenda-
tions for components to consider for evaluation, methods to use, and elements for reporting. The Second Panel
recommends that Reference Case analyses measure health effects in terms of QALYs. The purpose of standardiz-
ing methods and the various types of costs and benefits that are included in the Reference Case analyses is to
enhance consistency across analyses performed by different decision analysts and thereby to help policymakers
compare findings across studies.
• The results of the health care sector Reference Case analysis should be expressed as an ICER. The health care
sector perspective should include health care sector (medical) costs reimbursed by third- party payers or paid
for out- of- pocket by patients. Both types of medical costs include current and future costs, related and unrelated
to the condition under consideration. So, for example, an analysis of treatments to prevent heart attacks would
include the costs associated with the treatment itself and the downstream costs associated with heart disease
but also the costs associated with other health conditions (e.g., cancer) that may occur only because the patient
lives longer.
https://t.me/medicina_free
Cost- effectiveness analysis 333
• The analysis should include an Impact Inventory Table that lists the health and nonhealth impacts of interventions
included in the analysis. The main purpose of the Impact Inventory is to ensure systematic attention to all conse-
quences, including those outside of the formal health care sector.
• Analysts should try to identify, quantify, and value nonhealth consequences that may affect the result of the analy-
sis. For example, a treatment for children with autism would not only include clinical costs and benefits associated
with the child’s autism but also the impact of this treatment on the child’s educational and employment
trajectory.
• Analysts should clearly state the perspective of every analysis reported: the health care sector perspective, the
societal perspective, or an additional perspective that the analyst considers to be important to the decisional
context. Analysts should identify the primary decision maker(s) whose deliberations are the target of the
analysis.
• Analysts should describe in clear and understandable language their conduct of the analyses. This directive
applies especially to the assumptions of the model, and how the results change with alternative
assumptions.
Sensitivity analysis should describe the impact of the modeling assumptions that most strongly influence the
results for different perspectives.
15.5 Reference case forcost- effectiveness analysis
As described above, the Second Panel on Cost- Effectiveness in Health and Medicine recommends that a cost-
effectiveness analysis should be performed from the perspective of two Reference Cases. Each Reference Case follows
the same set of standard methodological practices recommended by the Second Panel. One analysis takes a health care
sector perspective. The other takes a societal perspective. From the health care sector perspective, we should include costs
paid by third-
party payers and out- of- pocket costs paid by patients (Table15.1). The societal perspective incorporates
all costs and all health effects regardless of who incurs the costs and who experiences the health effects. These should
include informal health sector costs which are outside of the structured health care system but which may still add
significant financial burden to patients and their families. Examples of such informal health sector costs include patient
time costs (i.e., costs associated with patient time spent traveling to and from care, waiting for, and receiving care),
unpaid caregiver costs, and transportation costs, as well as nonhealth care sector costs such as impacts on productivity,
consumption, and those costs borne by social services, legal, education, housing, or the environment sectors. Health
effects in both Reference Cases should be measured in QALYs.
Table15.1 Cost components included intwo reference case perspectives.
Reference case perspective
Health care Societal
Formal health care sector:*
• Paid for by third- party payers
• Paid for by patients out- of- pocket
Formal health care sector:*
• Paid for by third- party payers
• Paid for by patients out- of- pocket
Informal health care sector:
• Patient time
• Unpaid caregiver time
• Transportation costs
Nonhealth care sectors:
• Productivity
• Consumption
•
Social services
• Legal or criminal justice
• Education
• Housing
• Environment
• Other (e.g., friction costs)
*Includes current and future costs, related and unrelated to the condition under consideration.
https://t.me/medicina_free
334 Medical decision making
15.6 Impact inventory forcataloguing consequences
The Second Panel on Cost- Effectiveness in Health and Medicine recommends the inclusion of an Impact Inventory
Table in all cost-
effectiveness analyses (Table15.2). This Impact Inventory Table lists the health and nonhealth effects
of a health care intervention to ensure that all consequences are considered, including those to patients, caregivers,
social services, and others outside the health care sector. The Impact Inventory provides a framework for the analyst
to organize the types of consequences to be considered in an analysis and to better ensure that all consequences,
including the outcomes of the formal health care sector, are appropriately considered. An end user of the cost-
effectiveness analysis can then use the information in the impact inventory table to better understand the applicability
of the findings to their relevant population and any limitations of the simplifying assumptions on the true potential
impact.
15.7 Measuring thehealth effects ofmedical care
To determine the cost- effectiveness of comparison strategies, we must quantify the impact of the available strategies
on health outcomes. The Second Panel recommends that the Reference Case cost- effectiveness analyses should meas-
ure these health effects in terms of QALYs. QALYs allow an analyst to adjust the estimated life years for the various
Table15.2 Impact inventory template.
Sector
Type of impact (List category within each sector with a
unit of measure if relevant)*
Included in this reference case
analysis from...perspective?
Notes on sources
of evidenceHealth care Sector Societal
FORMAL HEALTH CARE SECTOR
Health Health Outcomes (Effects)
Longevity effects ◻ ◻
Health-
related quality- of- life effects ◻ ◻
Other health effects (e.g., adverse events and secondary
transmissions of infections)
◻ ◻
Medical Costs
Paid for by third- party payers ◻ ◻
Paid for by patients out- of- pocket ◻ ◻
Future related medical costs (payers and patients) ◻ ◻
Future unrelated medical costs (payers and patients) ◻ ◻
INFORMAL HEALTH CARE SECTOR
Health Patient time costs NA ◻
Unpaid caregiver time costs NA ◻
Transportation costs NA ◻
NONHEALTH CARE SECTORS (with examples of possible items)
Productivity Labor market earnings lost NA ◻
Cost of unpaid lost productivity due to illness NA ◻
Cost of uncompensated household production** NA ◻
Consumption Future consumption unrelated to health NA ◻
Social services Cost of social services as part of intervention NA ◻
Legal or criminal
justice
Number of crimes related to intervention NA ◻
Cost of crimes related to intervention NA ◻
Education Impact of intervention on Educational achievement of
population
NA ◻
Housing Cost of intervention on home improvements
(e.g.,removing lead paint)
NA ◻
Environment Production of toxic waste or pollution by intervention NA ◻
Other (specify) Other impacts NA ◻
*Categories listed are intended as examples for analysts.
**Examples include activities such as housework; food preparation, cooking, and clean-
up; household management; shopping; obtaining services; and travel
related to household activity.
NA=not applicable
Published previously in Sanders etal. (2016).
https://t.me/medicina_free
Соседние файлы в папке Библиотека им академика М.И. Перельмана