Transcript Pharmacoeconomics for Decision-Makers

Introduction to
Pharmacoeconomics
September 15 -16, 2009
Karen L. Rascati, PhD, Professor
The University of Texas College of Pharmacy
Austin, Texas, USA
The Universidade Federal do Rio Grande
do Sul (UFRGS) and the Programa of
Pós-graduação em Economia
(PPGE/UFRGS)
Outline

Day 1
 Part
I - What is pharmacoeconomics
 Part II – Types of pharmacoeconomic studies
 Part III – Costs/Outcomes
 Part IV – Evaluating Studies
 Part V – Evaluation Example 1
Outline

Day 2
VI – Decision Analysis
 Part VII – Markov Modeling
 Part VIII – Evaluation Example 2
 Part IX – Future Issues
 Part
Part I
 What
is Pharmacoeconomics?
Definition
 Pharmacoeconomics
“identifies,
measures, and compares costs and
consequences of pharmacy products
and services”
 Some consider it a sub-set of health
technology assessment (HTA)
Pharmacoeconomic Equation
COSTS  RX  OUTCOMES
 How
much is spent on health care
per year as a percent of GDP?
In
OECD countries?
0
8.9
8.8
8.7
8.4
8.4
8.3
8.2
8.2
OECD
Australia (3)
Norway
Spain
United Kingdom
Hungary
Finland
Japan (4)
6.4
6.2
Korea
Poland
5.7
6.6
Mexico
Turkey (4)
6.8
Czech Republic
7.1
9.0
Italy
Slovak Republic (4)
9.1
Iceland
7.3
9.1
Greece
Luxembourg
9.2
Sweden
7.5
(1) Public and private components are current expenditure,i.e. investments are not separated.
(2) Current expenditure.
(3) Data refer to 2005/06. (4) Data refer to 2005.
Ireland
9.3
New Zealand
10.0
Canada
9.3
10.1
Austria
Netherlands (2)
10.2
Portugal
9.5
10.4
Belgium (1)
Denmark (1)
10.6
Germany
11.1
4
France
8
11.3
12
Switzerland
United States
15.3
Health Expenditure as a Share of GDP, 2006
% of GDP
16
Public
Private
Brazil?
Brazil
About 8% GDP on healthcare
 SUS = tax funded system
 About half is public spending and half is
private spending (for about 20-30% of
population) = much more spent per
person if using private insurance

 What
is the average lifespan for
various OECD countries?
Brazil?
Brazil

Life expectancy is about 72 years
Why is Pharmacoeconomics
important?

Pharmacoeconomics helps assess if
scarce health care resources are being
spent wisely on pharmacy products and
services.
Part II -Types of
Pharmacoeconomic Studies




Cost-minimization analysis (CMA)
Cost-benefit analysis (CBA)
Cost-effectiveness analysis (CEA)
Cost-utility analysis (CUA)
COSTS  RX  OUTCOMES

More than one type may be included in a study
(e.g. CEA and CUA)
Types of Pharmacoeconomic
Studies
Cost-Minimization Costs =
Monetary units
Analysis (CMA)
Outcomes = The
same
CostEffectiveness
Analysis (CEA)
Costs =
Monetary units
Outcomes =
Natural/clinical units
Cost-Utility
Analysis (CUA)
Costs =
Monetary units
Outcomes =
Adjusted by
quality/utility (e.g.,
QALY, DALY)
Cost-Benefit
Analysis (CBA)
Costs =
Monetary units
Outcomes =
Monetary units
Other ‘Cost’ Studies

Cost-consequence analysis (CCA)
 Lists
costs and various outcomes presented
but no calculations or comparisons made

Cost-of-illness (COI)
 Estimate
of total economic burden
(prevention, treatment, losses in productivity)
of a particular condition or disease on society
Part III – COSTS/OUTCOMES
 Cost
analysis :To identify resources
used or consumed in the production
of a good or service and assign
monetary values to these resources.
COSTS  RX  OUTCOMES
Part III – COSTS/OUTCOMES
 PERSPECTIVE
Payer
= Whose Costs?
(third-party private/public and/or
patient)
Provider/ Institution
Employer
Society
Types of Costs

Direct Medical Costs

Direct Non-Medical Costs

Indirect Costs

Intangible Costs
Direct Medical Costs

What is paid for specific health care
services, such as physician services,
hospitalization, and pharmaceuticals

EX: Physical therapy, drugs to tx side
effects, costs of clinic visits
Direct Non-Medical Costs

Costs necessary to enable patients to
receive medical care

EX: Transportation to and from visits,
lodging, baby-sitters (special diet)
Indirect Costs

Measure of the patient’s lost productivity
plus the lost productivity of all unpaid
caregivers

EX: Time off from work, less productive
days, spouses time off from work.
Intangible Costs

Reflect the patient’s level of pain and
suffering. These are the hardest to
measure.

Anxiety, chronic pain, loss of functioning
25
Examples

A daughter takes a week off from work to attend to her ill
father

Inpatient charge of R$268 per day for acute care

Fatigue from chemotherapy

Taxi fare to emergency department

Ambulance service to emergency department
Examples

A daughter takes a week off from work to attend to her ill
father

 INDIRECT COSTS (productivity)
Inpatient charge of R$268 per day for acute care

 DIRECT MEDICAL
Fatigue from chemotherapy

 INTANGIBLE COSTS
Taxi fare to emergency department

 DIRECT NON-MEDICAL COSTS
Ambulance service to emergency department
 DIRECT
COSTS
MEDICAL COSTS
Example – Types of Costs for
Schizophrenia

Direct Medical
 Medications
 Outpatient/profession
al services
 Inpatient services
 Long-term care

Direct Non-Medical
 Law
enforcement
 Shelters

Indirect
 Unemployment
 Reduced
productivity
at work
 Premature mortality
(suicide)
 Caregiver
Incremental Costs
Average costs = total cost / total units
 Incremental = Change in total cost /
change in units

Example: Drug A is R$500 per patient and is
95% effective while Drug B is R$750 per
patient and 97% effective
Incremental Calculation

(R$750 – R$500) / (0.97 – 0.95) =
R$12,500 per extra cure
Adjusting for Time Differences
Two different concepts
 Inflation
 If
data collected over more than one year
 Prices may be adjusted to uniform price

Time Preference
 If
program or therapy extends more than one
year, “discounting” is appropriate
 Used even if inflation rate is zero
Adjustment for Inflation
Can count number of services/ resources
used and multiply by standard costs at one
point in time
OR
Use inflation rate for past years times cost
from past years
Adjustment for Inflation
Example of Standardization: Using Consumer Price Index (CPI) - Brazil
Medical Resources
Cost Estimate for
To Treat Mild
Resource
Infection
Office Visits
R$115.00
Lab to Culture
R$50.00
Organism
Antibiotic
R$28.84
Medication
TOTAL
a: Brazilian CPI for 2005 to 2007 = 11.3%
b: Brazilian CPI for 2006 to 2007 = 7.1 %
Year of Cost Estimate
Cost Adjusted to
2007 R
2005
2006
R$128.00a
R$53.55b
2007
R$28.84
R$210.39
Discounting
A time preference is associated with
money
 Current and future costs are not valued
the same
 If the treatment costs (and outcomes*)
extend for more than one year,
discounting should be conducted to
account for this difference.

Present Value (PV) Formula

PV = Sum of [FC / (1+r)n] for each year in
future
FC = Future Costs (or benefits)
 n = number of years
 r = discount rate per year

Discounting Example
Year Costs are
Estimated Costs
Calculation
Incurred
w/o Discounting
Year 1
R$ 5,000
R$ 5,000 /1.05
Year 2
R$ 3,000
R$ 3,000 / (1.05)2
Year 3
R$ 4,000
R$ 4,000 / (1.05)3
Total Net Present
R$ 12,000
Value (NPV)
Using a 5% discount rate
Present Value
(PV)
R$ 4,762
R$ 2,721
R$ 3,455
R$ 10,938
Sensitivity Analysis
For any costs “estimates” that are
uncertain, a sensitivity or “what if” analysis
should be conducted.
 How do we know the discount rate is 5%?.
Vary the rate from 0% to 10% and see if
decision of “least costly” alternative still
holds.
 Or vary cost of hospitalizations by area

Costs - Summary

When determining costs:
What is the perspective?
Are relevant/realistic costs included?
Is discounting or cost adjustment
appropriate?
Is a sensitivity analysis conducted for
uncertain values?
Types of Pharmacoeconomic
Studies
Cost-Minimization Costs =
Monetary units
Analysis (CMA)
Outcomes = The
same
CostEffectiveness
Analysis (CEA)
Costs =
Monetary units
Outcomes =
Natural/clinical units
Cost-Utility
Analysis (CUA)
Costs =
Monetary units
Outcomes =
Adjusted by
quality/utility (e.g.,
QALY, DALY)
Cost-Benefit
Analysis (CBA)
Costs =
Monetary units
Outcomes =
Monetary units
Cost-Minimization Analysis (CMA)
 Costs
are measured in monetary
units
 Outcomes
are assumed to be
equivalent
 Examples:
compare generics or home
vs. outpatient services.
CMA Research Example
Cost-minimization analysis of erlotinib in
the second-line treatment of non-cell lung
cancer: A Brazilian perspective
 Doral Stephani S; Giorgio Saggia M;
Vicino dos Santos EA.
 Journal of Medical Economics 2008; Vol.
(3), p. 383-96.

Example CMA
Budget impact of erlotinib versus
docetaxol or pemetrexed as second-line
treatment for NSCLC
 Perspective = Private healthcare payer
 Costs = Panel assessed local costs
 Outcomes = from clinical trial that
assessed progression-free survival

Example CMA
Erlotinib was cost saving ($R26,825)
compared to established chemotherapy
(R$40,217 and R$78,911)
 Sensitivity analysis showed robustness

Cost-Effectiveness Analysis (CEA)
 Advantage:
Do not have to place a
dollar value on clinical outcomes
 Disadvantage: Can only compare
options with the same type of
outcome, and only one outcome at a
time can be measured.
Cost-Effectiveness Grid
Cost
Outcome
Lower cost Same Cost
Higher Cost
Less effective
A
B
C
Same
effectiveness
D
E
F
More
effective
G
H
I
Cost-Effectiveness Grid
Cost
Outcome
Lower cost Same Cost
Higher Cost
Less effective
A
B
C
Same
effectiveness
D
E
F
More
effective
G
H
I
Cost-Effectiveness Plane
Cost Differences (+)
Quadrant IV
Dominated
Quadrant I
Trade-off
Effect
Differences
(-)
Effect
Differences
(+)
Quadrant III
Trade-off
Quadrant II
Dominant
Cost Differences (-)
Cost-Effectiveness Plane
Cost Differences (+)
Quadrant IV
Dominated
Quadrant I
Trade-off
Effect
Differences
(-)
Effect
Differences
(+)
Quadrant III
Trade-off
Quadrant II
Dominant
Cost Differences (-)
Examples of Ways to Present Cost and Effectiveness Results
Method
Method 1
CostConsequence
Analysis
(CCA)
Drug A
Drug B
Drug C
Costs
Costs
Costs
R$ 600 per year
R$ 210 per year
R$ 530 per year
Outcomes
Outcomes
Outcomes
GI SFDs = 130
GI SFDs = 200 days
GI SFDs = 250 days
% Healed = 50%
% Healed = 70 %
% Healed = 80 %
GI SFDs = gastro-intestinal symptom-free days
Examples of Ways to Present Cost and Effectiveness Results
Method
Method 2
Average Cost
Effectiveness Ratios
Drug A
Drug B
Drug C
R$ 600 / 130 =
R$ 210 / 200 =
R$ 530 / 250 =
R$ 4.61 per SFD
R$ 1.05 per SFD
R$ 2.12 per SFD
R$ 600 / 0.5 =
R$ 1,200 per cure
R$ 210 / 0.7 =
R$ 300 per cure
R$ 530 / 0.8 =
R$ 662 per cure
GI SFDs = gastro-intestinal symptom-free days
Examples of Ways to Present Cost and Effectiveness Results
Method 3
B compared to A = dominant for both SFDs and % healed;
Incremental
CostEffectiveness
Ratios
C compared to A = dominant for both SFDs and % healed;
C compared to B = R$ 530 – R$ 210 / 250 – 200 SFDs
= R$ 6.40 per extra SFD
C compared to B = R$ 530 – R$ 210 / .8 – 0.7
= R$ 3,200 per extra healed ulcer
GI SFDs = gastro-intestinal symptom-free days
Cost-Utility Analysis
(Some consider this a type of CEA)

Costs measured in dollars

Consequences measured in preferencebased measures, such as QALYs/DALYs

Incorporates mortality and morbidity
(quality and quantity of life)
Steps in Utility Analysis
Describe the health state
 Choose the instrument
 Administer the instrument
 Calculate utility
 Calculate QALYs

Describe the Health State

Example: You often feel tired and sluggish. A
piece of tubing has been inserted into a vein in
your arm or leg. This may restrict your
movement. There is no severe pain, but rather
chronic discomfort. You must go to the hospital
2-3 times per wk (8 hours per visit). You must
follow a strict diet (low salt, little meat, small
amount of fluid, no alcohol). Many people
become depressed because of the nuisances
and restrictions, some feel they are being kept
alive by a machine.
Choose the Instrument
THREE COMMON METHODS

Rating Scales

Time trade-off (TTO)

Standard Gamble (SG)
Rating Scale
Endpoints = Dead / Healthy
 Other health states are explained and
subjects are asked to “rate “ them between
the two endpoints
 May look like a thermometer
 Can compare many health state options
and ask raters to place them on one scale

Rating Scale
Perfect Health
100
Disease state
58
Death
0
Time Trade-off

Subjects are offered two alternatives:
 State
i for time t, followed by death, or
 Healthy time x (less than t) followed by death

Time x is varied until the subject is
indifferent between the two alternatives
Time Trade-off
Alternative 2
1.0
i
0
Alternative 1
x
t
Standard Gamble

Subject is offered two alternatives:
 Alternative
1 is a treatment with 2 possible
alternatives; pt. lives healthy life for x years or
dies immediately
 Alternative 2 is the certain outcome of chronic
state i for the rest of their natural life
Standard Gamble
p
1-p
healthy
dead
i
Comparing the 3 Methods






Rating Scale easiest but time not incorporated
as easily, must transform to QALYs
TTO conceptually easier than SG
SG and TTO give higher values than most using
rating scales
TTO sometimes lower than SG
Some consider SG to be “gold standard”
Much research left to answer “which is best”
Administer the Instrument - to
whom?

The general public
 societal
perspective
 hard to describe to general public

People with the disease
 if
comparing people with the same disease
 may be biased

Health Professionals / Disease Experts
 do
not have to explain or describe
 may be biased
Calculate Utilities

Selected utilities from rating scale
 1.0
Completely healthy
 .84 Kidney transplant
 .58 Hosp. dialysis (pts)
 .56 Hosp dialysis (public)
 .33 Hosp confinement
 0.0 Dead
 <0 ?
Calculate QALYs
For example if dialysis extends a life 10
years at .58 on rating scale = 5.8 QALYs
 If Option A cost R$5000 and extends life
for 6 years at a quality of .8 and Option B
costs R$4000 and extends life for 10 years
at a quality of .3, according to CUA which
would be preferred?

Based on CEA (no adjustment for
quality) which option would you pick?
Option A
2. Option B
3. Need ICER
Option Cost
YLS
1.
A
R$5000 6 years
B
R$4000 10 years
QALYS
0.8*6 =
4.8 QALYS
0.3* 10LYS
3.0 QALYS
Based on CEA (no adjustment for
quality) which option would you pick?
Option A
2. Option B
3. Need ICER
Option Cost
YLS
1.
A
R$5000 6 years
B
R$4000 10 years
QALYS
0.8*6 =
4.8 QALYS
0.3* 10LYS
3.0 QALYS
Based on CUA (QALYS) Which
option would you pick?
Option A
2. Option B
3. Need ICER
Option Cost
YLS
1.
A
R$5000 6 years
B
R$4000 10 years
QALYS
0.8*6 =
4.8 QALYS
0.3* 10LYS
3.0 QALYS
Based on CUA (QALYS) Which
option would you pick?
Option A
2. Option B
3. Need ICER
Option Cost
YLS
1.
A
R$5000 6 years
B
R$4000 10 years
QALYS
0.8*6 =
4.8 QALYS
0.3* 10LYS
3.0 QALYS
DALYS
DALYs = Disability Adjusted Life Years
 Similar to QALYs
 DALYs = The sum of years of potential life
lost due to premature mortality and the
years of productive life lost due to
disability.
 QALYs = Years of healthy life (sum of
quality * years)

Advantages of CUA

Includes patients’ preferences

Provides a single measure to incorporate
morbidity and mortality

Allows comparisons across different
options
Disadvantages of CUA





Time consuming
Results vary depending on who assesses the
conditions and by what instrument is used
Should you discount utilities?
Unanswered questions - Is a 20 QALY gain for
one person = a 1 QALY gain for 20 people?
How much is a QALY/DALY worth?
Cost-Benefit Analysis (CBA)

Costs measured in monetary units

Outcomes measured in monetary units

Calculate Benefit-to-Cost (B:C) ratio
Cost-Benefit Analysis (CBA)

Advantage = can summarize benefits from
many sources into one number (money)
and compare vastly different options

Disadvantage = difficult to place monetary
value on health outcomes
Medical Non-medical
Costs ($)
Benefits ($)
Direct Benefits $
Indirect Benefits $
Intangible Benefits $
Productivity
Patient Preferences
Pain
Suffering
Medical Non-medical
Human Capital (HC)
Willingness-to-pay (WTP)
WTP
Human Capital

Value of health benefits=the economic
productivity they permit
 Cost

of disease=lost productivity
Cost of a sick day=how much you earn that day
Human Capital

Use discounted values of expected
earnings
 Census
estimates (age, gender, education)
 Gather data from individuals
 Labor income is estimated as before-tax
income

Non-labor income is excluded (interest, etc.)
 Use
market values value for non-market
activities (unpaid household work, child care,
etc.)
Human Capital

Problems
 Biased

against specific groups
Age, gender, education
 Earnings

may not equal the value of outputs
Professional athlete versus teacher
 Does
not include values for pain and suffering if the
disease state or condition does not impact
productivity

E.g., Menopause, Impotence vs. Diabetes, Cancer
Willingness-to-Pay

Valuation of goods/services are easier for
marketed vs. non-marketed goods/services
 Health

care vs. coffee or pair of jeans
Valuation of goods/services are based on:
 Need

e.g., health care (pain/suffering, productivity, etc.)
 Resources
 Preference
Willingness-to-Pay

Determines how much people are
willing to pay to reduce the chance of an
adverse health outcome.
 Example:
If a person was willing to pay
R$20 for a ½ hour visit with a pharmacist
to improve their diabetes condition, then
the imputed benefit/person/visit would be
R$20.
Willingness-to-Pay
 Problems
 What
people say vs. what they will
really pay
 Inherent
biases of surveys (e.g.,
starting point bias, income bias)
 Can
the average person answer
questions
HC vs. WTP
HC
WTP

Easier to measure


Only considers
productivity (in
terms of earnings)
Biases against
specific groups



More difficult to
measure
Captures
productivity, patient
preferences
(intangibles)
Biases may not give
accurate responses
CBA Research Example
Costs and Benefits of Influenza
Vaccination and Work Productivity in a
Columbian Company from the Employer’s
Perspective
 Morales A, et al.
 Value in Health, Vol 7, No 4, 2004, p. 433441.

CBA Example
Columbian bank employees volunteered to
be in a prospective study involving
vaccination versus no vaccination for
influenza – 8 monthly questionnaires
 CBA to determine if employer would save
money offering vaccination to employees
(therefore perspective = employer (title)

CBA Example

Fever of at least 2 days with at least one
symptomatioc symptom (fevers, chills,
myalgia) and at least one respiratory
problem (rhinorrhea, sore throat, cough,
hoarseness) = Influenza-like illness (ILI)
CBA Example

Input costs
 Direct
= vaccine and materials, nurse
 Indirect = time lost by employee when getting
vaccinated (20 min) and if any days lost due
to effects from vaccine

Outcome costs (diff vacc vs. no vacc)
 Indirect
= sick leave and reduced efficiency at
work due to ILI
CBA Example
Vaccinated = 14.6% ILI
 Non-vaccinated = 39.4% ILI
 Employer saved $6 to $26 US per
employee vaccinated (depending on
assumed efficiency at work with ILI –
range 70% to 30%)

Part IV - Assessment of
Pharmacoeconomic Studies
1. Is the title appropriate?
2. Is the question (objective) clear?
3. Are the alternatives appropriate?
Assessment
4. Are alternatives described in detail?
5. Is the perspective addressed?
6. Is the type of study stated? What type
was it?
Assessment
7. Are relevant and realistic costs included/
justification for those not included?
8. Are relevant consequences/outcomes
included/ justification for those not
included?
9. Was adjustment or discounting
needed/conducted?
Assessment
10. Are assumptions stated/reasonable?
11. Was a sensitivity analysis conducted for
important estimates/assumptions?
12. Were major limitations addressed?
Assessment
13. Were appropriate generalizations
made? Were extrapolations beyond
population appropriate?
14. Is an unbiased, impartial attitude
portrayed? Was an unbiased
summary of the results presented?
Part V - Evaluate Example 1
Economic Impact of a Rotavirus Vaccine in
Brazil
Journal of Health Population Nutrition, 2008,
Vol 26 (4), p 388-396.
Outline

Day 2
VI – Decision Analysis
 Part VII – Markov Modeling
 Part VIII – Evaluation Example
 Part IX – Future Issues
 Part
Part VI - Decision Analysis

A systematic, quantitative approach for
assessing the relative value of one or
more decision options.
Steps in Decision Analysis

Identify the specific decision
 What
is the perspective?
 What
are the competing options?
 Over
what period of time?
Steps in Decision Analysis

Draw the structure over time
 Boxes
represent choice nodes
(Drug A vs. Drug B)
 Circles represent chance nodes
(S.E. or no S.E.)
 Triangles represent termination nodes
(live vs. die)
Steps in Decision Analysis

Assess the probabilities
 Use
 Use
past literature, experts, judges, panels
reasonable ranges for uncertain
probabilities
Steps in Decision Analysis

Determine the value of each outcome
 Options
must have the same type of outcome
( $ vs. $ or QALY vs. QALY)
 Can
look at costs and effectiveness in the
same model
Steps in Decision Analysis

Conduct a sensitivity analysis
 Choose
those values or probabilities that are
most uncertain or those where a small
difference has a big impact on the results
 Use
reasonable ranges
 Calculate
threshold values
Example

From an article by Alan Baskt, Pharm.D.

“Pharmacoeconomics and the formulary
decision-making process” in Hospital
Formulary, Vol 30, Jan 1995, p.42-50.
Example - ID Decision

Background
 DVT
prophylaxis
 Newer agent Enoxaparin (Lovenox)
 No coagulation monitoring required
 Lower DVT rate than heparin
 26 times more expensive than heparin
Example - ID Decision

Perspective
 Societal

Options
 enoxaparin
 fixed-dose
heparin
 low dose warfarin

Time frame
 about
1 month
Example - Draw Structure
Example - Assess Probabilities
Incidence
Proximal
DVT
Distal
DVT
Pulmonary
Embolism
Major
Bleeding
Minor
Bleeding
Warfarin
Enoxaparin
Heparin
5%
2%
4.8%
19%
2%
5.3%
2.7%
0.1%
1.9%
1.3-3.6%
4.1%
6.2%
6.9%
8.2%
5.7%
Example - Determine Values
Costs
Warfarin
Enoxaparin
Heparin
Drug
0.14
159.88
6.10
PT test x 7 d
8.68
0
0
PTT test x 7 d
8.68
0
8.68
3 home visits
60.00
0
0
CCF nurse
20.87
0
0
7.44
0
0
12.30
0
0
$118.11
$159.88
$14.78
PT and APTT
x 3 visits
Outpt. Rx
TOTAL
Example - Determine Values
Complication
Abbrev.
Cost
Proximal DVT
Comp1
$1,394
Distal DVT
Comp2
$ 860
P. Embolism
Comp3
$6,510
Major Bleed
Comp4
$2,791
Minor Bleed
Comp5
$ 189
Part VII - Markov Modeling





Real health consequences more complex
May need to look at long-term consequences
over multiple years
Patients may “transition” from one health state to
another over time
Basic decision trees get too complex after a few
cycles
Researchers use Markov Modeling to assist with
more complex and chronic disease states
Part VIII Evaluate Example 2
Cost-effectiveness Analysis of Cervical
Cancer Vaccine in Five Latin American
Countries
 Colantonio L, et al.
 Vaccine, Volume 27, 2009, p. 5519-5529

Part IX - Issues
Perspective - Whose costs?
 Appropriate comparators
 Efficacy vs. Effectiveness

 Criteria
 Length
of follow-up
 Switching

Outcomes
 Accuracy
of measurement
 Multiple measures
Issues

Barriers
Does not include budget impact
Lack of expertise in economic
evaluations
Decision-makers mistrust results
Seen as “rationing” – may not want to
acknowledge resources are limited or
that trade-offs are necessary
For More Information (in addition to
my book, of course)…



Methods for the Economic Evaluation of Health
Care Programmes, 3rd ed. Drummond, Sculpher,
Torrance, O’Brien and Stoddart, 2005
Health Care Cost, Quality, and Outcomes:
ISPOR Book of Terms, Berger et al, 2003 –
available soon in Portuguese.
International Society for Pharmacoeconomics
and Outcomes Research http://www.ispor.org