Signature
j* = argmin_(j in J) C_j, conditional on G = 1
| Inputs | Definition | Unit |
|---|---|---|
C_j | Total relevant cost of alternative j under the stated perspective and time horizon | currency per defined population or person |
G | Indicator equal to 1 only when reliable evidence establishes that decision-relevant outcomes do not differ meaningfully | binary indicator |
j* | Feasible alternative with the lowest total relevant cost when G equals 1 | alternative identifier |
|---|
JSet of feasible healthcare alternatives included in the comparison (set of alternatives)
Function
Equivalence-gated least-cost selection function
Selects the lowest-cost feasible healthcare alternative only when reliable evidence shows that all decision-relevant outcomes do not differ meaningfully.
Implementations
Excel
Return the least-cost option
Excel returns the option name associated with the minimum total cost after the equivalence gate is confirmed, and flags a tie when more than one option shares the minimum.
=IF(EquivalenceGate<>1,"CMA not valid",IF(COUNTIF(TotalCosts,MIN(TotalCosts))>1,"Equal lowest cost",INDEX(OptionNames,MATCH(MIN(TotalCosts),TotalCosts,0))))
Assumptions
Equivalence evidence gate satisfied
Reliable comparative evidence shows that plausible differences in every decision-relevant outcome are too small to change the decision.
Complete cost measurement
Total cost includes every material resource difference permitted by the stated analytical perspective.
Consistent costing basis
Alternatives use the same population, time horizon, currency, price year and applicable discounting convention.
Feasible alternatives
Every alternative in the comparison is available and relevant to the healthcare decision.
Worked examples
Selecting the least-cost equivalent option
Reliable equivalence evidence supports a costs-only comparison. After acquisition, administration, monitoring and adverse-event costs are included, Treatment A costs £1,600 per patient and Treatment B costs £1,520 per patient. Treatment B is selected as the least-cost alternative, saving £80 per patient, although its acquisition price alone is £250 lower. Treatment B is alternative 2 in the calculation.
C_j = [1600,1520]; G = 1; j* = 2
Common errors
Treating no significant difference as equivalence
Failure to find a statistically significant difference in a superiority test does not establish that outcomes are equivalent and does not justify cost minimisation.
Sources
NICE Decision Support Unit report
Wailoo A, Dixon S. The use of cost minimisation analysis for the appraisal of health technologies. NICE Decision Support Unit; 2019.
Canonical Identity
Stable URI · Machine-readable · Resolvable · CC BY 4.0