Choosing an economic evaluation
Choose an economic evaluation from the decision: health gain for its cost, benefits valued in money, or the payer's budget.
Fit
When it matters, and when it does not
It matters when
- A treatment works, and the question is whether it is worth adopting with a limited budget.
- A payer or committee needs cost per quality-adjusted life year, or the total cost of adopting it.
- You are running a trial and can record costs and quality of life alongside the clinical outcome.
It matters less when
- Whether the treatment works at all is still unknown. Answer that first in a randomized trial. Read: Randomized controlled trial
- The effect estimates you would use are biased or come from very different patients. Fix the evidence first, such as with a systematic review. Read: Systematic review
Before you start
What to check in your study
- Which question: value for money, or affordabilityCost-effectiveness asks whether the benefit is worth its cost. Budget impact asks what adopting it would cost the payer in total.
- The perspectiveWhose costs count. A health system perspective leaves out patients' time and lost work; a societal one includes them.
- A horizon suited to the decisionValue may need long-term health effects. A payer budget usually concerns the next few years. Do not apply one horizon to both.
- The comparatorWhat is done now, not an option nobody uses.
- Where the effect estimates come fromA trial, or a systematic review. Check they fit the patients being modeled.
- The evaluation matches the decision being made.How to check: Ask whether the decision concerns health gain, monetary benefit or annual affordability. Name the evaluation type accordingly.
- The evidence and local costs fit the decision setting.How to check: Review the source population and current care. A model cannot make biased clinical evidence reliable.
- Budget and value questions remain distinct.How to check: The budget-impact page uses payer population and uptake over a short horizon of a few years, without discounting.
Plain language
How it works
Choose the economic question before choosing a model. Cost-effectiveness compares costs with a clinical outcome. Cost-utility uses quality-adjusted life years, which combine life length and quality. Cost-benefit values benefits in money. Budget impact asks what adoption would cost a payer each year.
A good-value treatment can still be unaffordable. Use the linked cost-effectiveness and cost-utility page for value, and budget impact for affordability. State whose costs count, the horizon and the comparator. These choices can change the conclusion.
Worked example
A worked example
- Setting
- A payer considers a new anticoagulant. The local population includes 1,000 eligible patients.
- Question
- Which evaluations would inform value and next year's spending?
- Approach
- Use cost-utility analysis for health gain relative to cost. Use a separate budget impact analysis for affordability under expected uptake.
- Result
- If 200 of 1,000 eligible patients start next year, uptake is 20 in 100.
- What it means
- That uptake helps frame the budget question. It does not answer whether the health gain is worth the cost.
In practice
Recent studies PubMed search · not vetted
Watch out
Common mistakes
The one to remember
The result depends heavily on the perspective, the time horizon and the discount rate, none of which are statistical choices, and changing any of them can reverse the conclusion. Uncertainty must be presented as a distribution rather than a point estimate of the incremental cost-effectiveness ratio.
- Reporting only the cost-effectiveness ratio.Instead: A negative cost and a negative benefit give the same ratio as two positives. Report the incremental cost and benefit, and the plane.
- Assuming a trial sized for its clinical outcome is large enough for costs.Instead: Costs are skewed by a few expensive patients. Say how precisely the trial can estimate the cost difference.
- Analyzing only patients with complete cost data in a trial-based evaluation.Instead: They are rarely a random subset. Name the missing-data assumption, and use multiple imputation under missing at random, with a sensitivity analysis.
Inside Mallard
How Mallard uses this
Plans do not link to this concept yet. Mallard has no sample-size calculation for costs, so a trial sized for its clinical outcome is not thereby sized for its economic result.
concept.choosing_economic_evaluation, the id a plan cue records.Related Learn pages
Related designs and methods
Evidence
Sources & further reading
- Consolidated Health Economic Evaluation Reporting Standards 2022 (CHEERS 2022) Statement: Updated Reporting Guidance for Health Economic Evaluations (opens in a new tab) · Value Health · doi 10.1016/j.jval.2021.11.1351
- Recommendations for Conduct, Methodological Practices, and Reporting of Cost-effectiveness Analyses: Second Panel on Cost-Effectiveness in Health and Medicine (opens in a new tab) · JAMA · doi 10.1001/jama.2016.12195
- Budget impact analysis-principles of good practice: report of the ISPOR 2012 Budget Impact Analysis Good Practice II Task Force (opens in a new tab) · Value Health · doi 10.1016/j.jval.2013.08.2291