Is Cost-Minimisation Analysis Appropriate?

Test the clinical-evidence gate before comparing costs alone.

Critical rule: “No statistically significant difference” is not evidence of equivalence. CMA is appropriate only when meaningful differences in all decision-relevant outcomes have been credibly excluded.

1. Identify the comparative claim

Evidence claim Was the margin specified and justified before results were known? Suitable comparative design and analysis? Analysis populations appropriate for the design?

2. Test the interval against the accepted margin

Accepted absolute margin Confidence-interval lower bound Confidence-interval upper bound

For equivalence, the full interval must lie within −margin to +margin. For non-inferiority, the relevant one-sided boundary is tested, but passing it does not by itself establish decision equivalence.

3. Confirm outcome completeness and applicability

All decision-relevant outcomes considered? Safety and tolerability sufficiently similar? Evidence applicable to intended population, setting and comparator? Could plausible gains or losses outside the tested endpoint change the decision?

4. Define the costing scope

Perspective Time horizon Currency / price year Discount rate, if required

Costs per patient

Component	Option A	Option B

5. Audit summary

Complete the evidence gate.

The tool will explain whether CMA is supported and why.

Interpretation safeguards

A non-significant superiority test does not demonstrate equivalence.
Non-inferiority can support part of the evidence case, but possible benefits, other endpoints, safety and decision context must still be considered.
A stable cost saving cannot rescue a failed clinical-evidence gate.
The least-cost option is conditional on perspective, time horizon, resource scope, price year and uncertainty.
If meaningful outcome differences remain plausible, retain outcomes explicitly in CEA, CUA, CCA or another suitable evaluation method.