Economic Evaluation in Health · Faculty workshop · Session 2
What economic evaluation is,
and what it is not
Drummond et al., Methods for the Economic Evaluation of Health Care
Programmes, 4th ed., 2015.
Chapter 1, pages 1 to 19.
Faculty, residents and research project staff · AIIMS Bhopal, 23 September 2026
By the end of this session
- Define economic evaluation and state the two features that make an analysis one
- Explain opportunity cost in your own words
- Use Drummond's two questions to sort any study into one of six boxes
- Tell CMA, CEA, CUA and CBA apart
- Say why a costing study is not a cost-effectiveness analysis
28 February 2026, Ajmer
India began vaccinating every 14-year-old girl against HPV.
1.15 croregirls aged 14 every year, in every State and UT
One dosequadrivalent vaccine, Gardasil-4
Freeat government facilities
Press Information Bureau, Government of India, 28 February 2026. Introduced into the routine immunisation schedule with GAVI support.
The problem it addresses
1.2 lakhnew cervical cancer cases each year in India
80,000deaths each year, the second commonest cancer in women
25%of the world's cervical cancer deaths happen here
Burden alone does not justify a programme.
Every disease has a burden. Every programme has a cost. The budget did not grow.
GLOBOCAN 2022 estimates, as cited by the Press Information Bureau, 28 February 2026.
It was not the only option
Screen instead of vaccinate
VIA, cytology or HPV DNA testing for women aged 30 to 65
VIA, cytology or HPV DNA testing for women aged 30 to 65
A different vaccine
bivalent, quadrivalent or nonavalent
bivalent, quadrivalent or nonavalent
A different age band or schedule
9 to 14 years; one dose or two
9 to 14 years; one dose or two
Vaccinate and screen
or spend the money on treatment capacity
or spend the money on treatment capacity
Choosing one of these means not choosing the others.
Start here: resources are scarce
A District Health Society has a fixed envelope this year. Every rupee committed to one thing is a rupee not available for another.
This is not a budgeting inconvenience. It is the reason economic evaluation exists.
Scarcity is not a failure of planning. Even a very rich health system faces it, because demand for health care grows faster than any budget.
Opportunity cost
The value of the benefit forgone because a resource was not used in its next best alternative way.
The cost of the HPV programme is not ₹X crore.
The cost is the health that ₹X crore would have produced somewhere else in the Indian health system.
Money is the measuring rod. Forgone health is the thing being measured.
So the question is never…
Is this intervention worth it?
Is it worth it compared with what we would otherwise do with the same resources?
Which is why an economic evaluation always has at least two arms. And why "cost-effective" is an empty phrase unless you can finish the sentence: cost-effective compared with what, at what threshold, from whose point of view?
Two features define an economic evaluation
1. Inputs and outputs
It deals with both the resources consumed (costs) and the results produced (consequences). Not one or the other.
2. Choice
It compares two or more alternatives. No comparison, no choice. No choice, no economics.
Miss either feature and you have something else. Often something useful, but not an economic evaluation.
Drummond's two questions
Question 1
Are both costs and consequences examined?
Question 2
Is there a comparison of two or more alternatives?
Two yes-or-no questions. Six possible boxes. Only one of them is a full economic evaluation.
The six boxes
No comparison of alternatives
Two or more alternatives compared
Consequences only
Outcome description
Efficacy or effectiveness evaluation
Costs only
Cost description
Cost analysis
Both
Cost-outcome description
FULL ECONOMIC EVALUATION
Five of the six are partial evaluations. Not worthless, but none of them can tell you whether to fund the programme.
One disease, six study designs
No comparison
Alternatives compared
Consequences only
Outcome descriptionDALYs from cervical cancer, national and state level (Ramamoorthy 2024)
Effectiveness evaluationHPV testing vs cytology vs VIA vs none, mortality endpoint (Sankaranarayanan 2009)
Costs only
Cost descriptioncost of treating cervical cancer by stage (Singh 2020)
Cost analysiscost per woman screened by VIA vs cytology vs HPV DNA, no outcomes
Both
Cost-outcome descriptiona programme report giving cost per woman screened and detection rate, with no comparator
FULL ECONOMIC EVALUATIONlifetime costs and QALYs of screening and vaccination (Chauhan 2020)
All six are cervical cancer in India. Four are published studies. The two without a citation are the designs you meet in State and district programme reports.
Partial is not useless
| Partial study | Answers | Cannot answer |
|---|---|---|
| Outcome description | How big is the problem? | What should we do about it? |
| Cost description | What will this cost us? | Is it worth that cost? |
| Effectiveness evaluation | Does it work, and how well? | Can we afford what it displaces? |
| Cost analysis | Which option is cheaper? | Whether the cheaper one is worse |
Every full economic evaluation is built from these.
Your turn
Eight real studies, two questions each: does it measure both costs and consequences, and does it compare alternatives? The widget scores you and says why.
Open it on your phone at sph-aiimsbhopal.github.io/intro-economic-evaluation-health, along with five others we will use today.
Inside the green box: four types
In every full economic evaluation, costs are measured in money.
What differs is how the consequences are valued.
That one distinction gives you CMA, CEA, CUA and CBA.
If you remember nothing else about the four types, remember that the difference lives entirely on the consequences side.
Cost-minimisation analysis
Consequences valued as: nothing. They are shown to be equivalent, then set aside.
The output: which option costs less.
Published example. Ready-made versus custom-made spectacles for school children in India. A non-inferiority trial first established that spectacle wear was equivalent, so the economic analysis reduced to a cost comparison. Cost per child needing spectacles fell from US$26.91 to US$11.15, a saving of 58.6%.
Only legitimate if equivalence was demonstrated, not assumed.
Minakaran et al., Ophthalmic Epidemiology, 2021.
Cost-effectiveness analysis
Consequences valued as: one natural unit. Deaths averted, cases detected,
life-years gained.
The output: cost per unit of effect.
Published example. A national rotavirus vaccination programme for India. At US$1.00 per dose it would cost US$49.8 million and prevent about 44,000 deaths a year, at US$662.94 per life saved. At the GAVI price of $0.15 a dose it was cost-saving.
You can only compare programmes that share the same outcome unit.
Esposito et al., Clinical Infectious Diseases, 2011.
Cost-utility analysis
Consequences valued as: a generic, preference-weighted measure. The QALY,
or in much LMIC work the DALY.
The output: cost per QALY gained.
Published example. A Markov model from a societal perspective over a lifetime, comparing screening strategies and HPV vaccination for India. Screening with VIA every 5 years: ₹54,881 per QALY gained. HPV vaccination, alone or with later screening: ₹5,693 to ₹31,511 per QALY gained.
A generic unit makes a cancer programme and a cardiac programme comparable.
Chauhan, Prinja et al., PLoS One, 2020.
Cost-benefit analysis
Consequences valued as: money, through willingness to pay or lost production.
The output: net benefit in rupees, or a benefit-to-cost ratio.
Why you will read very few of these in health. Valuing a life-year in rupees is technically demanding and ethically contested, and most health ministries do not want their decisions framed that way. CBA remains standard in transport, environment and infrastructure.
Its unique strength: it can compare a hospital with a road.
The four types, side by side
| Consequences valued as | Result | Compares across | |
|---|---|---|---|
| CMA | equivalent, set aside | which is cheaper | identical outcomes only |
| CEA | one natural unit | ₹ per unit of effect | the same outcome unit |
| CUA | QALYs or DALYs | ₹ per QALY gained | any health programme |
| CBA | money | net benefit in ₹ | anything at all |
Read down the second column. That column is the classification.
That "₹ per unit" column has a name
ICER = extra costextra effect
The incremental cost-effectiveness ratio. Extra compared with the next best alternative, not with nothing.
US$662.94 per life saved
Rotavirus, two slides ago
Rotavirus, two slides ago
₹54,881 per QALY gained
VIA screening, last slide
VIA screening, last slide
An ICER is a statement about two options. No single programme has one of its own.
Session 5 computes them, ranks five options, and shows the three ways they go wrong.
A warning: QALYs are not DALYs
QALY, gained
Health gained. Preference-based utility weights, 1 for full health and 0 for dead, usually elicited from the general public. You maximise it.
DALY, averted
Health lost. Disability weights plus years of life lost against a reference life expectancy. You minimise it.
They are not interchangeable, and one is not the arithmetic negative of the other.
You met DALYs first, through burden-of-disease teaching. Most Indian HTA work now reports QALYs, because the Indian Reference Case asks for them. Watch which one a paper uses.
Three errors we will keep catching
- Calling a costing study a CEA. "Cost" in the title does not make it an economic evaluation. Ask Question 1.
- Calling a CEA a CUA. Cost per case detected is not cost per QALY. Look at the denominator.
- Declaring something cost-effective with no threshold and no comparator. The phrase is empty without both.
A systematic review of Indian evaluations found exactly these problems at scale.
Prinja et al., Applied Health Economics and Health Policy, 2015.
Where this lands on your desk
State PIP
Every activity proposed displaces another. Somebody has to justify the choice.
PMJAY package rates
What a procedure is reimbursed at is a costing and value question, revised using these methods.
HTAIn topic nomination
Any of you can nominate a technology for assessment. Knowing what evidence is needed is the first step.
Work you supervise
Label the study honestly. A good cost analysis beats a mislabelled CEA in any thesis or paper.
The same two questions, on a drug
Adjuvant trastuzumab versus chemotherapy alone, non-metastatic breast cancer, India.
| Question 1: both costs and consequences? | Yes. Incremental cost, and QALYs gained |
|---|---|
| Question 2: two or more alternatives? | Yes. Chemotherapy alone, and trastuzumab for a year, six months or nine weeks |
| So which box? | Bottom right: a full economic evaluation, and a cost-utility analysis |
Same two questions, same box. The unit of decision changed, the method did not.
And look at the comparator: four durations of the same drug, not the drug against nothing.
Gupta, Verma, Gupta, Prinja, JCO Global Oncology, 2020. Open access.
Recap
- Opportunity cost, the health forgone elsewhere, is why any of this exists
- An economic evaluation compares costs and consequences across two or more alternatives
- The two questions sort every study into six boxes. Five are partial
- The four full types differ only in how consequences are valued
Nothing so far tells you whether ₹54,881 per QALY is good value.
Hold it until after lunch. First, was that the right question?
Sources
- Drummond MF, Sculpher MJ, Claxton K, Stoddart GL, Torrance GW. Methods for the Economic Evaluation of Health Care Programmes. 4th ed. Oxford University Press; 2015. Chapter 1, pp. 1–19.
- Press Information Bureau, Government of India. Cervical Cancer Vaccination Campaign launched. 28 February 2026. Burden figures from GLOBOCAN 2022.
- Ramamoorthy T, Kulothungan V, Sathishkumar K, et al. Burden of cervical cancer in India: YLL, YLD and DALYs at national and subnational levels. Reprod Health 2024;21:111. PMID 39075548
- Singh MP, Chauhan AS, Rai B, et al. Cost of treatment for cervical cancer in India. Asian Pac J Cancer Prev 2020;21(9):2639–46. PMID 32986363
- Sankaranarayanan R, Nene BM, Shastri SS, et al. HPV screening for cervical cancer in rural India. N Engl J Med 2009;360:1385–94. PMID 19339719
- Chauhan AS, Prinja S, Srinivasan R, et al. Cost effectiveness of strategies for cervical cancer prevention in India. PLoS One 2020;15(9):e0238291. PMID 32870941
- Minakaran N, Morjaria P, Frick KD, Gilbert C. Cost-minimisation analysis from a non-inferiority trial of ready-made versus custom-made spectacles for school children in India. Ophthalmic Epidemiol 2021;28(5):383–91. PMID 33256485
- Esposito DH, Tate JE, Kang G, Parashar UD. Projected impact and cost-effectiveness of a rotavirus vaccination program in India, 2008. Clin Infect Dis 2011;52(2):171–7. PMID 21288839
- Prinja S, Chauhan AS, Angell B, et al. A systematic review of the state of economic evaluation for health care in India. Appl Health Econ Health Policy 2015;13(6):595–613. PMID 26449485
Faculty workshop · Session 3 · Drummond 4th ed., Chapter 4
Framing
the question
The decision problem, the comparator, the perspective, the time horizon. Four choices made before any arithmetic, and each one can decide the answer.
Before the next session: skim the methods of your group's paper and note its perspective, time horizon and discount rate.