+

Basic Methods of Economic

Evaluation

Cost-Effectiveness Analysis (CEA)

Outcomes are measured in natural/clinical/physical
units (e.g. heart attacks avoided, deaths avoided).
The outcome is common to both alternatives but the
effect size and direction may vary;
Costs are measured in monetary units.

Only one domain of outcomes can be explored at a

time.
Result: cost per unit of consequence (e.g. cost/LY
gained)

+
Costeffectiveness Plane
A four-quadrant

figure of cost difference

plotted against effect difference:
quadrant

I, intervention more effective and more

costly than comparator;
quadrant II, intervention more effective and less
costly than comparator;
quadrant III, intervention less effective and less
costly than comparator; and
quadrant IV, intervention less effective and more
costly than comparator. (Culyer, 2005:77-78)

Cost-Effectiveness Analysis (CEA)

The cost-effectiveness acceptability Plane:

Cost-Effectiveness Analysis (CEA)

Decision rule:
Two programmes A (comparator) and B.
If Outcome B = Outcome A => Compare costs (CMA).
If Outcome B > Outcome A and Cost B < Cost A, B is dominant.
If Outcome B > Outcome A and Cost B > Cost A, we have to
make a decision.

In order to make a decision on which intervention to

choose, a cost-effectiveness ratio (CER) should be
calculated.

Cost-Effectiveness Analysis (CEA)

The most commonly CERs used are the:
Average cost-effectiveness ratio (ACER)

Cost B
ACER=
Effectiveness B

Incremental cost-effectiveness ratio (ICER)

(CostB CostA)
ICER=
(EffectivenessB EffectivenessA)

The next question is : Is the intervention cost-effective?

Cost-Effectiveness Analysis (CEA)

There is no magic cut-off number that establishes
whether or not an intervention is cost-effective.
It will depend on what is termed the decision makers
ceiling ratio.
The ceiling ratio can be inferred from the amount that
decision-makers are willing to pay.
To make a decision:
If ICER of the program ceiling ratio adopt the program
If ICER of the program > ceiling ratio do not adopt the
program
7

Cost-Benefit Analysis (CBA)

CBA try to value the outcomes in monetary terms, so
as to make them commensurate with the costs.
Costs are measured in monetary units;

Result: Net benefit or cost-benefit ratio.

CBAs rarely used in health care.
8

Differences Between CEA and CBA

Phelps & Mushlin (1991):
Reporting

style : CBA typically determines in advance the marginal value of a benefit, e.g., QALY or a
life year, and then calculates net benefits. In contrast, CEA typically calculates the price of a QALY or a
life and leaves the decision unstated.
Level

of aggregation : Often CEA is applied at a highly disaggregated level, in contrast to CBA which
could be applied at a much more aggregated level, e.g., society.
measurement

of multi-dimensional benefits: CBA has the advantage that all types of benefits are
converted into a common metric, where in a CEA some of the conversions might be more difficult to
carry through.
Case

of joint production: applying CBA involves adding up all benefits and costs from all dimensions of
a project and comparing them against each other. CEA would look at the marginal CE ratios along the
different dimensions, the difficulty lies then in determining incremental costs in the case of joint
production.

Cost-Utility Analysis (CUA)

The outcomes are measured in healthy years, to which
a value has been attached. Outcomes are measured as
healthy years, typically measured as quality-adjusted
life-years (QALYs).
Costs are measured in monetary units;
CUA is multidimensional and incorporates
considerations of quality of life as well as quantity of life
using a common unit.

Result: Cost per unit of consequence (e.g. cost/QALY).

10

Cost-Utility Analysis (CUA)

The

primary outcome for CUA is expressed as qualityadjusted life-years (QALYs);

Other

generic outcome measures for CUA:

Disability-adjusted life-year (DALY);
Healthy years equivalent (HYE);
Saved-young-life-equivalent.
(Drummond et al., 2005:14)

Differences Between CEA and CUA

Types of
studies

Costs or
measure
s

Benefits or Comments
Consequen
ce
measures

Cost
Effectivenes
s Analysis
(CEA)

Costs
measured in
monetary units
(e.g.. Dollars)

Benefits
measured in
natural units
(e.g.. mmHg,
cholesterol
levels, symptom
free days, years
of life saved)

Results are expressed as dollars

per case or per injury averted.
Different incremental summary
economic measures are
reported (e.g.. Incremental
cost-effectiveness ratio)

Cost Utility
Analysis
(CUA)

Costs
measured in
monetary units
(e.g.. Dollars)

Benefits
expressed in
summary
measures as
combined
quantity and
quality measures
(e.g.. QALY, DALY
etc)

Two dimensions of effects

measured (quality and length of
life); results are expressed for
example as cost per QALY

+
Model-Based CEA
Plus :
the

preferred method of conducting CEAs (Sculpher et al. 2006)

tool for assessing the epidemiologic and economic impact of

diseases and interventions (Goldhaber-Fiebert 2010)
a

way of developing policy and informing current decisions where

trials are incomplete or infeasible (Freedberg et al. 1996)
more

practical method comparing to trial.

Minus :
validation

of models is difficult (especially in chronic disease, due to

lack of long-term data to compare model predictions to)
the

discretionary nature of populating models leads to the

possibility of arbitrariness (McCabe and Dixon 2000, Sheldon 1996)

+
Clincial Trial CEA
Plus :
excellent

tool for validation

opportunity

to collect prospective, high-quality cost data and being

close enough to real practice that provides a credible proxy
effectively

functioning as a gold standard costing method to which

we may compare model estimates
Minus :
larger

sample sizes are needed for high-quality data collection

the

difficulty of amassing participants at a single site. Finally, the

practice of statistical wizardry is no substitute (Graves et al. 2002).
most

trials have short horizons, lasting several months or a few

years at most in course of chronic diseases, it might fails to
capture relevant downstream costs

+
Tesis Mengenai CEA
Judul :
Analisis

Efektivitas Biaya Intervensi terhadap Penderita

Kusta Setelah Selesai Pengobatan Melalui Pengamatan
Semi Aktif dan Pengamatan Pasif : studi kasus di
Kabupaten Pasuruan Tahun 2012 (Medita Ervianti, 2013)
Desain :
Cross

sectional dengan 43 responden.

Menganalisis

efektivitas biaya pada intervensi

pengamatan semi aktif dan pengamatan pasif pada
penderita kusta setelah selesai pengobatan.
Analisis

membandingkan jumlah biaya (langsung dan

tak langsung) sebagai input dan jumlah penderita yang
dapat dikendalikan tingkat cacatnya sebagai output.

Hasil :
Biaya

operasional pengamatan semi aktif 58x lipat lebih tingi dibanding biaya pengamatan pasif.

Analisis

bivariat antara variabel input metode pengamatan terhadap output pengendalian tingkat
cacat menunjukkan hubungan signifikan dengan OR 9,6
Rasio

efektivitas biaya kurang dari 1 kali GDP per capita metode pengamatan semi aktif lebih
efektif biaya dibanding metode pengamatan pasif untuk mengendalikan tingkat cacat pada penderita
yang telah selesai pengobatan.
Diskusi :
Trials

often do not reflect actual clinical practice, given the presence of highly motivated clinicians,
selective patient samples, above average adherence to medication, increased vigilance and the use of
invasive gold standard methods of assessing efficacy (Akhtar, 2003).
Terdapat

variabel lain yang bisa menjadi confounder dan berpotensi bias, misalnya dedikasi tenaga
kesehatan yang melakukan pengamatan yang bisa mempengaruhi keberhasilan dan efektivitas
program (karena sifat dari intervensi kebijakan tersebut menitikberatkan pada pembinaan).
Belum

dijelaskan sejauh mana terkendalinya kecacatan worth the price compared to biaya
operasional 58x lipat lebih besar? --> Quadran I (more effective, more expensive), how much more
effective?
Outcome

pengendalian tingkat cacat hanya dihitung dengan variabel ordinal 0 (tingkat cacat dapat
dikendalikan/tetap/menurun) dan 1 (tingkat cacat tidak dapat dikendalikan/meningkat). Belum
melakukan CUA: perhitungan Life Year gained, mis: QALY/DALY?