1992-2016; 25 years of Mahidol Social and Administrative Pharmacy Graduate Studies

Measuring Cost
in Economic Evaluation
Assoc Prof Arthorn Riewpaiboon
Division of Social and Administrative Pharmacy
Department of Pharmacy
Faculty of Pharmacy Mahidol University
arthorn.rie@mahidol.ac.th
http://www.pharmacy.mahidol.ac.th/staff/arthorn
1
 Learning questions

How does costing affect the EE

results?
What are the costs and (overview)
how to measure?
How to adjust costs in EE?

2
Economic evaluation is the comparative analysis of
alternative courses of action in terms of both their costs
and consequences.
(Drummond et al, 2005)

Outcomes/
Intervention A Consequences A

Cost A
Health
problem

Cost B

Intervention B Outcomes/
Consequences B
3
Effect of costing on EE results
Health
Intervention Outcomes
problem

Effective‐
Costs Burden
ness

EE methods

4
Results
 Cost of intervention
and comparator
= Cost of providing service (cost of health care
program, cost of treatment)
+ Cost of adverse events/complications
(cost of illness)
+ Cost of unsuccessful patients (cost of illness)

5
Cost of vaccination program
= Cost of vaccination program (Cost of program)
+ Cost of the ones who have AEFI (Cost of illness)
+ Cost of the unimmunized ones have the
illness (Cost of illness)
Cost of no vaccination program
= Cost of the (unimmunized) ones who have the
illness (Cost of illness)

6
Cost‐effectiveness of rotavirus vaccination as part of
the national immunization program for Thai children
(Chotivitayatarakorn, 2010)
 expenditure for care according to the WHO CHOICE
 break‐even price of USD6.2 per dose.

Economic analysis for evidence‐based policy‐making

on a national immunization program: a case of
rotavirus vaccine in Thailand (Muangchana et al, 2012)
 patient charge were converted to economic costs.
 break‐even price of USD4.98 per dose (2 dose‐course).

7
8
Cost of rotavirus diarrhea for programmatic
evaluation of vaccination in Vietnam
(Riewpaiboon et al, 2016)
Substantially larger cost burden than those of
Fischer et al.
Several factors might have contributed to this.
 prospective vs retrospective data collection.
 followed up entire episode vs one short
interview.
 actual unit costs vs opinions of health officials.
 multiple study sites in three provinces in
different regions vs one province.
9
What is the costs and how to measure?
Economics is based on three fundamental concepts:
scarcity
– resources are insufficient to support all demands;
choices
– because of resource scarcity we need to choose
between alternative ways of using them;
opportunity cost
– by choosing to use available resources in one way, we
forgo other opportunities to use these same resources. So
cost or economic cost or opportunity cost of engaging in an
activity or producing a product refers to the sum of all other
benefits that can be generated by the same amount of
resources taken away for this activity.

10
The cost (economic or opportunity cost)
of goods or services is the measure of the value of
resources used or consumed to produce the goods or
services.

Financial costs/ accounting cost : measurement of

costs by the historical outlay of funds
(actual monetary flows of the buyer).
(Creese and Parker, 1990)
Source: http://america.pink/payment_3445911.html 11
Opportunity cost; Case of a community pharmacy
‐ Overhead (electricity, water,…) cost =800$/month

‐ Renting fee of office =$300 /month

‐ Furniture given free‐of‐charge from a friend (market price= $500)

capital cost/ month =$9.10 ($109.18/year; useful year = 5 years, discount rate 3%,)

‐ Your salary from working for private hospital = $1,500 /month

or income from direct sale $2,000 / month

‐ Total cost per month = 800+300+9.10+2,000= $3,109.10

12
 Cost of Illness
total economic impact or cost of disease or
health condition on society through the
identification, measurement and valuation
of all direct and indirect costs . (Berger et
al, 2003)

Scope of the illness

Primary illness and consequences;
complications, sequelae/ not include co‐
morbidity
(Kobelt, 2002; Drummond, 2005)13
Prevalence‐based approach
Prevalence‐based approach covers all patients during
time horizon of the study. Time horizon is normally 1 year to
avoid seasonal variation. The patients can start having the
illness before or during the time horizon. Therefore, the
patients included in the study have various levels of disease
progress and severity. Study results are presented as cost per
person per year (or time horizon).

Incidence‐based approach
Incidence‐based approach covers new cases during a
period of time designed (normally 1 year) until end of the
illness (cure or death). This is also called life time cost. Study
results are presented as cost per episode.
14
Prevalence vs incidence approaches;
acute illness
study period

A1.1 A1.2
B1 B2.1 B2.2

C1 C2
D1
F1.1 F1.2
E1
G1
= Duration of episode

Prevalence-based approach includes A1.2 , B1, B2.1, C1, C2, D1, F1.1

Incidence-based approach includes B1, B2.1+B2.2, C1, C2, D1, F1.1+F1.2

15
Prevalence vs incidence approaches;
chronic illness
study period

A1.1 A1.2 A1.3

B1.1 B1.2

C1.1 C1.2 (dead)

D1.1 D1.2 (dead)

E1 (dead) G1
= Duration of episode
Prevalence-based approach includes A1.2 , B1.1, C1.1, D1.2, E1

Incidence-based approach includes B1.1+B1.2, C1.1+C1.2, E1 (life time cost)

16
 Hospitalization, OPD
Medical cost visits, lab, medical
services
Direct cost
Transportation, meal,
Non-Medical cost accommodation, devices
Cost of illness - Caregiver time loss:
treatment/ recovering,
Indirect cost/ disability
Productivity - Patient time loss:
cost/ Time cost treatment/ recovering,
disability, premature death
17
 Direct medical cost
Treatment cost

disabled

Treatment; Hospital Treatment; after/

recovery period
Complete
before treatment recovery

dead
Costs of transportation, meal, hotel, facilities

Time cost of caregivers

Time cost of patient

Direct non-medical cost Indirect cost

Time 18
 Direct medical cost
Treatment cost

disabled

Treatment; Hospital Treatment; after/

recovery period
Complete
before treatment recovery

dead
Costs of transportation, meal, hotel, facilities

Time cost of caregivers

Time cost of patient

Direct non-medical cost Indirect cost

Time 19
 Measurement of resource use/ data collection
Data Source Method

Medical service -Medical record for -Chart review, hospital

utilization for Direct study hospital database/ prospective
Medical Cost -Patient/ family for data collection
treatment from other - Interview, diary
facilities
Meal, transportation, -Patient/ family Interview, diary
hotel for Direct non
Medical Cost
Time of caregivers for Care givers Interview, diary
cost of informal care

Patient time loss for -Patient/ family Interview, diary

Indirect Cost

20
 Valuation of resource used
Valuation of resource used
Total cost = Quantity of resource used x unit cost of the resource

Direct medical cost

• Total outpatient(OP) visit cost:
Number of visit (Q) x Cost per outpatient visit (P)
• Total bed-day cost:
length of stay (Q) x cost per bed-day (P)
(might be: ICU bed + general bed)
• Total medication/medical supplies cost:
Number of units for each item (Q) x unit cost of item (P)
• Total diagnostic tests cost:
Number of units for each item (Q) x unit cost of item (P)
Total direct medical cost:
= Total OP visit cost+ total bed-day cost + total medication cost + total
diagnostic cost + other relevant costs
21
 Bottom-up (Micro costing) approach 1
Average treatment cost

Treatment cost; patient A Treatment cost; patient B Treatment cost; patient C

Reference/standard
unit cost
Number of
services used
or
Hospital
unit cost

OPD; visit IPD; patient day Lab; investigation Pharmacy; dispensing Drugs used

Medical services received Medical services received Medical services received

by patient A by patient B by patient C

22
 Bottom-up (Micro costing) approach 2
Average treatment cost

Treatment cost; Treatment cost; Treatment cost;

patient A patient A patient A

RCC Cost-to-charge ratio (RCC)

= total cost/ total revenue

Revenue Labor + material

(income) + capital costs

Medical service charge of each patient

Medical services received Medical services received Medical services received

by patient A by patient B by patient C

23
Indirect cost: cost of time loss
Cost of morbidity = (N)(E)
N = total number of patient-day loss
E = reference or average earnings per day

Cost of caregiver time = (N)(E)

N = total number of caregiver-day loss
E = reference or average earnings per day

24
Cost of mortality
or permanent severe disablement
n

 Cost of mortality = S  Yi(1  r ) i

1
S = number of lives loss due to illness
Yi = expected GDP/GNP per capita at year i
n = number of years of expected earnings
= age at retirement – average age of patient (at death)
i = time in terms of year (1th - nth year)
eg i = 1 for the first year after the study year
r = discount rate

25
Discounting; reduction of future value
 Opportunity cost of capital; interest rate of long‐term
savings/ financial investment
 Time preference; consumers enjoy near consumption more
than more remote consumption.
‐ People may not live long enough to consume in the
future.
‐ Future is not certain.
‐ People expect to be richer (more consumption) in
the future then decrease in need of money/ goods
(consumption). (A bowl of pho when you are hungry before
lunch is more valuable than another bowl after lunch.)

26
27
 Cost analysis of
health care program
 economic/ financial valuation of resources
used in health care program and
quantification of outputs produced by the
program to explore total cost and cost per
unit of the program outputs.
(Kumaranayake, 2000)

28
Cost of DM screening program

29
(Aroonsiriwattana, 2010)
Types of costs
 Economic costs: Estimates all costs of an intervention,
regardless of the source of funding, so that the opportunity
cost of all resources is accounted for in the analysis,
includes in‐kind and donor contributions.

 Financial cost: Estimates the actual monetary flows of

the buyer such as the Ministry of Health. Does not include
the value of resources already paid for such as personnel
time.

30
 Incremental cost: costs of additional resources required
to add the new program to an existing program or routine
work.
 Marginal cost: (incremental) cost per unit of additional
output.
 Full cost: costs of all resources required to the program.
Some of them are shared with existing program or
routine work, eg., salary of staff who work for both
routine job and the program.
 Average cost: (full) cost per unit of output.

31
 Introduction Costs: Introduction costs are
initial one‐time programmatic activities
(start‐up cost), for instance, micro‐
planning, initial training activities, and
initial sensitization/IEC. These are treated
as capital costs in economic costing.
 Investment Costs: Initial expenditures
used in preparation for an intervention.
These include introduction costs plus
purchase of capital goods such as cold chain
equipment and transport purchases.
32
Sources of costing data
 Direct measurement of COI/ intervention with good
study design using
‐ direct unit cost analysis
‐ standard unit cost
 Service utilization from country database adjusted by
standard unit cost
 Charge from country database adjusted by ratio of cost
to charge (RCC)
 Secondary from study with good quality adjusted by
appropriate CPI
 Expert opinion; resource used x standard unit cost

33
Reference unit costs: Thailand
 Medical services; Standard cost menu/
reimbursement/ prices

 Drugs; Drug and Medical Supply Information

Center (DMSIC)

34
Standard cost menu; unit cost
 Unit cost of medical services at
provincial/ district hospitals
 Unit cost of medical services at health
centers
 Unit cost of hospital pharmaceutical
services
 Cost of transportation and meal for out
patients
35
36
(Riewpaiboon, 2014) 37
Standard unit cost of medical services

Ref: Dengue Thai-July 2015.xlsx 38

(Source: http://dmsic.moph.go.th/dmsic/index.php?&p=1&type=3&t=3&id=1)

39
 How to adjust costs in EE.
Cost component in different EE
methods and perspectives.
Weighted average cost for national
policy
Time adjusted costs

40
Costing in CUA (DALYs/ QALYs)
 Do not include productivity cost (indirect cost) in
nominator (total cost) to avoid double counting.
 This is because full impact of morbidity/ mortality
is included in the calculation of the QALY.
 Time loss due to morbidity, resulting in anxiety, has
caused and been included in loss of quality of life.
 Time loss due to mortality, resulting in zero utility,
has been included in loss of quality of life .
(Luce et al (in Gold et al), 1996, p181‐2)

41
Cost of intervention in societal perspective;
pharmaceutical therapy
 Direct medical cost including drug to be tested
 Direct non‐medical cost including care giver time
cost
 Indirect cost (all time cost of patient)
Or
 Direct medical cost including drug to be tested
 Direct non‐medical cost
 Indirect cost (all time cost of patient and care
giver)

42
Costs
Treatment cost

Travel, meal, hotel costs

Cost
Care giver time loss (informal care) – hours x income per hour

Time loss of patient/ Indirect cost – morbidity – days x income per day

– mortality – years x annual income

Quality of life/ utility

EQ5D – Utility x duration
- Mobility
- Self care
QALY
- Usual activities
- pain/ discomfort
- anxiety/ depression 43
The Erasmus Group/ The Netherlands
 The valuation of health status is not intended to
capture the impact of non health‐related events such
as loss of income.

 Then, individuals may or may not implicitly

incorporate these effects into their health state
valuations.

Pritchard; 2000: p. 37.

44
Riewpaiboon 2014 45
Weighted average cost: level of facilities
To estimate treatment costs at district hospital based
on the relative value of the unit cost of provincial
hospitals and district hospitals (134.95 and 128.67,
respectively).
Cost from a provincial hospital:
3,568.20 THB per admission, 524.68 THB per visit
Estimated cost in district hospital:
 (128.67/134.95)*3,568.20 = 3,402.15 THB per
admission
 (128.67/134.95)*524.68 = 500.26 THB per visit.
(Muangchana et al, 2012)
46
Weighted average cost: country
 Then, the proportion of service utilization
of diarrhea patients at provincial/regional
hospitals (21%) and district hospitals (79%)
was used to calculate the weighted average.
 The country weighted average costs:
 [(21/100)*
3,568.20]+[(79/100)*3,402.15]=3,437.02 THB
per admission
 and [(21/100)*524.68]+
[(79/100)*500.26]=505.39 THB per visit.
(Muangchana et al, 2012)
47
Weighted average cost: level of facilities

(Wang et al, 2009)

48
Time adjusted costs
 All costs were adjusted to 2009 value by
the consumer price index (2002=93.4,
2009=100.9).
 Cost of inpatient service, outpatient
service and drugstore were
 US$108.3 [(100.9/93.4)*3,437=3,713 THB),
 US$13.3 [(100.9/93.4)*505=546 THB]
US$4.7 [(100.9/93.4)*150=162 THB],
respectively.
(Muangchana et al, 2012)
49
Past Present Future
Discount rate

Mt-n Mt Mt+n
Inflation rate Interest rate
(Consumer Price Index; CPI)

• WHO recommended to use the discount

rate = 3% at the base case
•0% and 6% used for sensitivity analysis
(Edejer et al, 2003)
50
 References
 Aroonsiriwattana C. Cost‐Benefit Analysis of Type 2 Diabetes
Screening in Community Pharmacy: A Risk Score Assessment
Method, Case Study at Prachuap Bhesaj Drugstore,
Prachuapkhirikhan (in Thai language). M.Sc. in Pharm: Narasuan
University; 2010.
 Brouwer W, Rutten F, Koopmanschap M. Costing in economic
evaluations. In: Drummond M, McGuire A, eds. Economic
evaluation in health care ‐ merging theory with practice. New
York: Oxford University Press; 2001.
 Chotivitayatarakorn P, Poovorawan Y. Cost‐effectiveness of
rotavirus vaccination as part of the national immunization
program for Thai children. Southeast Asian J Trop Med Public
Health 2010; 41(1): 114‐25.
 Creese A, Parker D. Cost analysis in primary health care: a training
manual for programme managers. Geneva: World Health
Organization; 2000. 51
 Drummond MF, Sculpher MJ, Torrance GW, O'Brien BJ, Stoddart
GL. Methods for the economic evaluation of health care
programmes third edition. Oxford: Oxford University Press; 2005.
 Edejer TT‐T, Baltussen R, Adam T, Hutubessy R, Acharya A, Evans
DB, et al. Making choices in health: WHO guide to cost‐
effectiveness analysis. Geneva: World Health Organization; 2003.
 Fischer T.K., Anh D.D., Antil L., et al. Health care costs of
diarrheal disease and estimates of the cost‐effectiveness of
rotavirus vaccination in Vietnam. J Infect Dis., 2005: 192(10); 1720‐
6.
 Kim S.Y., Goldie S.J. & Salomon J.A. Cost‐effectiveness of
Rotavirus vaccination in Vietnam. BMC Public Health, 2009; 9:
29.
 Kobelt G. Health economics: an introduction to economic
evaluation. Second ed. London: Office of Health Economics; 2002.

52
 Kumaranayake L, Pepperall J, Goodman H, Mills A, Walker
D. Costing guidelines for HIV prevention strategies.
Geneva: UNAIDS; 2000.
 Luce BR, Manning WG, Siegel JE, Lipscomb J. Estimating
costs in cost‐effectiveness analysis. In: Gold MR, Sigel JE,
Russell LB, Weinstein MC, editors. Cost‐effectiveness in
health and medicine. Oxford: Oxford University Press;
1996.
 Pritchard C, Sculpher M. Productivity costs: principles and
practice in economic evaluation. London: Office of Health
Economics; 2000.
 Riewpaiboon A, Shin S, Le TP, Vu DT, Nguyen TH,
Alexander N, et al. Cost of rotavirus diarrhea for
programmatic evaluation of vaccination in Vietnam. BMC
Public Health 2016; 16(1): 777.

53
 Riewpaiboon A. Measurement of costs for health economic
evaluation. J Med Assoc Thai 2014; 97(Suppl. 5): S17‐26.
 Riewpaiboon A. Standard cost lists for health economic
evaluation in Thailand. J Med Assoc Thai 2014; 97(Suppl.5):
S127‐134.
 Tu H.A., Rozenbaum M.H., Coyte P.C., et al. Health
economics of rotavirus immunization in Vietnam:
potentials for favorable cost‐effectiveness in developing
countries. Vaccine, 2012: 30(8): 1521‐8.
 Wang XY, Riewpaiboon A, von Seidlein L, Chen XB, Kilgore
PE, Ma JC, et al. Potential cost‐effectiveness of a rotavirus
immunization program in rural China. Clin. Infect. Dis.
2009; 49(8): 1202‐10.

54