Questions: Cost-Effectiveness Analysis and Economic Evaluation of Health Interventions

Question 1

A new oncology drug costs $300,000 per year and extends life by 4 months at full quality. Its ICER is calculated at $900,000 per QALY. A country's willingness-to-pay threshold is $50,000 per QALY. Which conclusion is most accurate?

• A. The drug should be approved because any life extension has inherent value regardless of cost
• B. The drug is not cost-effective by this threshold — each QALY it produces costs far more than alternative uses of that health budget
• C. The drug is clinically ineffective because its ICER exceeds the threshold
• D. The ICER is invalid because no threshold can appropriately value human life

Answer: (B)

Cost-effectiveness analysis does not judge whether the drug works — it judges whether it represents good value relative to what else the budget could buy. An ICER of $900,000/QALY against a threshold of $50,000/QALY means that 18 times more health could theoretically be produced by spending that money elsewhere. Option A conflates clinical and economic evaluation. Option C confuses ICER (a value-for-money measure) with efficacy (a clinical measure). Option D is a real philosophical position but doesn't answer the question.

Question 2

A mental health intervention has an ICER of $90,000 per QALY from a healthcare system perspective, but only $12,000 per QALY from a societal perspective. The country's WTP threshold is $50,000 per QALY. What best explains this large discrepancy?

• A. The societal analysis contains errors because perspective should not affect the ratio of costs to health gains
• B. The healthcare perspective only counts medical costs while missing substantial benefits outside that system — reduced productivity losses, lower criminal justice costs, reduced caregiver burden — that the societal perspective captures
• C. The societal perspective artificially inflates QALYs by counting benefits to third parties
• D. The intervention has different clinical effects in different populations, explaining the different cost estimates

Answer: (B)

Perspective is one of the most consequential methodological choices in CEA. A healthcare system perspective counts only direct medical costs and health outcomes measured within the system. A societal perspective adds productivity gains, informal caregiver time, reduced criminal justice costs, education impacts, and other downstream effects. Mental health interventions often look far more cost-effective from a societal perspective because they prevent large non-medical costs. This is not an error — it is a different answer to a different question.

Question 3

A cost-effectiveness analysis that reports a single ICER value without sensitivity analyses is providing an incomplete and potentially misleading result.

• T. True
• F. False

Answer: True

Every input to a CEA model — disease incidence, treatment efficacy, quality-of-life weights, discount rates, time horizon — carries uncertainty. A single ICER is a point estimate that gives a false impression of precision. Sensitivity analysis (one-way or probabilistic via Monte Carlo) reveals how the conclusion changes as assumptions vary. If modest changes in key inputs flip the conclusion from 'cost-effective' to 'not cost-effective,' the policy recommendation is fragile and should not drive decision-making.

Question 4

The willingness-to-pay threshold used in cost-effectiveness analysis is a scientific parameter derived from empirical data about the value of health, making it objective and comparable across countries.

• T. True
• F. False

Answer: False

The WTP threshold is a value judgment — a policy choice about how much a society is willing to spend for an additional unit of health. It is not derived from any natural fact. The WHO's suggestion of 1–3× GDP per capita is a heuristic, not a scientific finding. Different countries use very different thresholds (the UK's NICE uses roughly £20,000–30,000/QALY; some middle-income countries use $2,000–3,000/DALY). Even within countries, thresholds are contested and sometimes violated — rare disease treatments are frequently funded above the stated threshold.

Question 5

Why is sensitivity analysis considered essential rather than optional in cost-effectiveness analysis, and what does it reveal that a single ICER cannot?

Model answer: Sensitivity analysis is essential because every parameter in a CEA model — incidence rates, effectiveness estimates, quality-of-life weights, discount rates, time horizons — rests on assumptions that carry uncertainty. A single ICER gives the false impression of a precise answer. Sensitivity analysis reveals the robustness of the conclusion: if the ICER remains well below the WTP threshold across all plausible parameter values, the recommendation is solid; if it crosses the threshold with modest changes in key inputs, the recommendation is fragile. Probabilistic sensitivity analysis (Monte Carlo simulation) produces a distribution of possible ICERs rather than a point estimate, showing decision-makers the probability that the intervention is cost-effective — which is the information actually needed for policy.

The practical consequence is that a CEA without sensitivity analysis should not drive coverage decisions. The analysis answers a specific question (is this cost-effective under these assumptions?) but policy requires knowing how sensitive that answer is to the assumptions. A drug that appears cost-effective only under optimistic assumptions about efficacy may not justify coverage once uncertainty is fully characterized.