Economic Evaluation in Health · Faculty workshop · Session 6
Making decisions
in health care
Drummond et al., Methods for the Economic Evaluation of Health Care
Programmes, 4th ed., 2015.
Chapter 2, pages 19 to 41.
Faculty, residents and research project staff · AIIMS Bhopal, 23 September 2026
By the end of this session
- Explain the budget constraint and health opportunity cost
- Tell a demand-side threshold from a supply-side one
- State India's willingness to pay per QALY, and say what it is not
- Say why the 1 to 3 times GDP rule was abandoned by its own authors
- Describe how HTAIn turns a topic into a recommendation
- Give two reasons a cost-effective intervention might still not be funded
The number we borrowed
In the last session you ranked strategies against λ ≈ ₹2,12,307 per QALY, and took that number on trust.
Where does it come from, and should you believe it?
The same gap sits under this morning's ₹54,881 per QALY for VIA screening. Is that good value? Nobody in this room could say, and neither could the literature, in India, until very recently.
ICER from Chauhan, Prinja et al., PLoS One, 2020. Threshold from Chugh et al., Value in Health, 2026.
The threshold, λ
The most you are willing to pay for one additional unit of health.
Adopt if ICER < λ. Reject if ICER > λ. Arithmetically trivial.
The whole difficulty of this chapter is: where does λ come from?
There are two quite different ways of arriving at it. They give different numbers, and the profession has not settled which is right.
Two answers, from opposite directions
Demand side: willingness to pay
Ask society what a unit of health is worth to it. Survey people, or infer it from what they pay for safety elsewhere.
Answers: what should we be prepared to spend?
Supply side: health opportunity cost
Ask what health the system actually produces at the margin. With a fixed budget, spending here displaces care there.
Answers: what can we afford without doing net harm?
With a fixed budget, only the supply-side number keeps you from doing net harm.
The rule you were probably taught
Cost per DALY averted below 1× GDP per capita is "highly cost-effective". Below 3× is "cost-effective".
From the Commission on Macroeconomics and Health and the WHO-CHOICE programme. It spread because it was simple, universal, and needed no country data.
WHO's own authors have argued it should be retired.
It is a demand-side heuristic with no link to any actual health budget. Applying it approves interventions a country cannot fund without displacing more health than it creates.
Bertram et al., Bulletin of the WHO, 2016. Marseille et al., Bulletin of the WHO, 2015.
What supply-side numbers look like
| Country | Estimated threshold | As % of GDP per capita |
|---|---|---|
| Malawi | $3 to $116 | 1% to 51% |
| Cambodia | $44 to $518 | 4% to 51% |
| El Salvador | $422 to $1,967 | 11% to 51% |
| Kazakhstan | $4,485 to $8,018 | 32% to 59% |
Their conclusion: routinely used thresholds have been too high.
Every upper bound sits below 59% of GDP per capita, well under the 1× that the old rule treated as "highly cost-effective".
Woods, Revill, Sculpher, Claxton, Value in Health, 2016.
India now has its own estimate
Multistage stratified random sample. Twelve hypothetical health states, time trade-off for the health gain and willingness to pay for its value, discounted at 3%. Across health states the estimate ranged from ₹1,70,414 to ₹2,58,985.
Chugh, Jyani, Kar, Rajsekar, Prinja et al., Value in Health, 2026.
So: is ₹54,881 per QALY good value?
| Strategy | ₹ per QALY gained | Against λ ≈ ₹2,12,307 |
|---|---|---|
| HPV vaccination of adolescent girls | 5,693 to 31,511 | well below |
| Screening with VIA every 5 years | 54,881 | below |
Both are cost-effective for India. Vaccination is comfortably the better buy per rupee.
Which fits the decision taken in February 2026. But notice the dates: the analysis was published in 2020 and the threshold estimate in 2026. Policy did not wait for a number.
ICERs from Chauhan et al., 2020. Threshold from Chugh et al., 2026.
One caveat you must carry
₹2,12,307 is a demand-side number: what a sample of Indians said a QALY is worth to them.
It is not an estimate of what the Indian health system gives up at the margin when it spends an extra rupee.
If the supply-side threshold is lower, judging by willingness to pay will approve things the budget cannot absorb.
That is not a criticism of the study, which is careful and says what it measured. It is the open question the field is living with. Teach it as open.
League tables
Rank every intervention by cost per QALY, draw a line where the budget runs out, fund everything above it. India now has one: 79 studies, 2003 to 2019.
Then read the fine print of that same review.
It also needs a comparable evaluation of every competing use of the budget. Nobody has that. What was never studied never appears in the table.
Shah, Singh, Kotwani et al., Frontiers in Public Health, 2022.
A country that publishes its threshold
NICE has used an explicit range since 1999. It is the clearest example anywhere of a threshold you can argue with.
| Until March 2026 | From April 2026 | |
|---|---|---|
| Standard range | £20,000 to £30,000 per QALY | £25,000 to £35,000 per QALY |
Announced 1 December 2025. NICE expects it to allow roughly 3 to 5 additional medicines a year, from a base where about 91% of appraised medicines are already recommended.
NICE, Changes to NICE's cost-effectiveness thresholds confirmed, 1 December 2025.
Why did it move?
Not because anyone re-estimated health opportunity cost in the NHS.
The stated purpose was to unlock innovation and support the sector, improving the UK's operating environment for the life sciences industry.
A threshold raised for industrial policy is no longer measuring what the theory says it measures.
Health economists have said so in print: such changes risk making cost-effectiveness analysis performative rather than informative.
Vallejo-Torres, Edney, Espinosa, Karnon, Longo, Applied Health Economics and Health Policy, 2026.
Who decides in India
Health Technology Assessment in India (HTAIn), now an attached office of the Department of Health Research, Ministry of Health and Family Welfare.
Board
Highest decision-making body. Policy-makers, clinicians and government experts. Endorses recommendations.
Technical Appraisal Committee
Economists, clinicians, researchers, social scientists. Quality assurance and technical stewardship.
Secretariat
In-house scientists and analysts. Prioritises topics, coordinates, monitors progress.
Regional Resource Centres
Government institutes across the country that conduct the studies and build capacity in the states. The network keeps growing.
From a question to a policy
- A user department, a programme or a state, submits a priority topic
- The Secretariat prioritises it and takes it to the Technical Appraisal Committee
- The Committee allocates it to a Regional Resource Centre, which writes a proposal
- The study runs. The report goes back to the Committee for approval
- The Board gives final approval
- Recommendations return to the user department for implementation
Any of you can nominate a topic.
Cost-effective is not the same as funded
| Consideration | The question it asks |
|---|---|
| Budget impact | Cost-effective per patient, but can we afford it for everyone at once? |
| Equity | Who gains and who is left out? Does it widen or narrow a gap? |
| Feasibility | Cold chain, trained staff, supply, referral pathways. Do they exist? |
| Acceptability | Will families consent? Will clinicians adopt it? |
| Timing | Do we act on what we have, or wait for better evidence? |
The economic evaluation is one input to a decision, never the decision.
Extended cost-effectiveness analysis brings equity into the arithmetic. See Dixit, Prinja, Jyani et al., Lancet Global Health, 2023.
A live case worth arguing about
Cost per DALY averted, healthcare perspective, PRIME model, 12-year-old girls in India:
| Vaccine (types covered) | Incremental cost per DALY averted |
|---|---|
| Nonavalent, Gardasil 9 (6, 11, 16, 18 + 31, 33, 45, 52, 58) | US$ 362.78 |
| Quadrivalent, Gardasil 4 (6, 11, 16, 18) | US$ 393.16 |
| Bivalent, Cervarix (16, 18) | US$ 432.24 |
Assuming a constant price per dose, the nonavalent vaccine dominated both others.
India launched in February 2026 with a single dose of the quadrivalent.
Arora and Tuffaha, Asia-Pacific Journal of Clinical Oncology, 2024.
So why not the nonavalent?
Price is not constant
The dominance result assumes every product costs the same per dose. Gardasil 9 does not sell at the Gardasil 4 price. Change that one input and the ranking changes.
16 and 18 do most of the damage
About 83% of invasive cervical cancer in India is attributable to HPV 16 or 18. The five extra types buy less than the count suggests.
Financing
A US$ 250 million GAVI partnership, and a negotiated price, attach to a particular product rather than to a valency.
Supply at national scale
About 1.15 crore girls a year. Guaranteed volume beats a marginally better ICER that cannot be delivered.
Three more that mattered
Domestic manufacturing
Cervavac, from the Serum Institute of India, is also quadrivalent. It was still seeking WHO prequalification at launch, a procurement gate rather than a question of value.
Budget impact
Whether it is affordable this year is a separate test from whether it is cost-effective.
Delivery
Cold chain, school and session logistics, and a single-dose schedule shape what is actually achievable.
Every one of these is legitimate, and none of them is in the ICER.
The threshold, made visible
The origin is current practice. Drag the intervention between quadrants, then drag λ and watch the same point flip from adopt to reject with no number in the analysis changing.
Preset to ₹2,12,307 per QALY · sph-aiimsbhopal.github.io/intro-economic-evaluation-health
Recap
- λ turns an ICER into a decision. Everything hard here is where λ comes from
- Demand-side and supply-side give different numbers. Only the second respects a fixed budget
- India's first estimate is ₹2,12,307 per QALY. Demand-side, so carry the caveat
- League tables still mix QALYs and DALYs, so they cannot be read straight down
- HTAIn advises. User departments decide.
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 2, pp. 19–41.
- Chugh Y, Jyani G, Kar SS, Rajsekar K, Prinja S, et al. Estimating the willingness-to-pay per quality-adjusted life-year to aid health technology assessment in India. Value Health 2026;29(3):457–66. PMID 41176029
- Woods B, Revill P, Sculpher M, Claxton K. Country-level cost-effectiveness thresholds: initial estimates and the need for further research. Value Health 2016;19(8):929–35. PMID 27987642
- Bertram MY, Lauer JA, De Joncheere K, et al. Cost-effectiveness thresholds: pros and cons. Bull World Health Organ 2016;94(12):925–30. PMID 27994285
- Marseille E, Larson B, Kazi DS, Kahn JG. Thresholds for the cost-effectiveness of interventions: alternative approaches. Bull World Health Organ 2015;93(2):118–24. PMID 25883405
- Shah K, Singh M, Kotwani P, et al. Comprehensive league table of cost-utility ratios for health policy decisions in India. Front Public Health 2022;10:831254. PMID 36311623
- NICE. Changes to NICE's cost-effectiveness thresholds confirmed. 1 December 2025. nice.org.uk
- Vallejo-Torres L, Edney LC, Espinosa O, Karnon J, Longo F. Politicised changes to the NICE threshold risk making cost-effectiveness analysis performative, not informative. Appl Health Econ Health Policy 2026;24(2):243–6. PMID 41513890
- Dixit J, Prinja S, Jyani G, et al. Efficiency and equity of prevention and control strategies for rheumatic heart disease in India: an extended cost-effectiveness analysis. Lancet Glob Health 2023;11(3):e445–55. PMID 36796988
- Arora S, Tuffaha H. Cost-effectiveness analysis of human papillomavirus vaccines for the prevention of cervical cancer in India. Asia Pac J Clin Oncol 2024;20(1):55–62. PMID 37132538
- 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
- ICO/IARC HPV Information Centre. Report: India. hpvcentre.net
- Press Information Bureau, Government of India. Cervical cancer vaccination campaign launched. 28 February 2026.
Faculty workshop · Session 7 · modelling
Where the numbers
came from
Every ICER today arrived already calculated, over a lifetime nobody observed. The next session is the machinery that produces them: decision trees and Markov models.
Before the next session: find the figure in your group's paper that shows its model structure. Most papers have one.