A NON-ITERATIVE METHOD OF IDENTIFYING THE COST-EFFECTIVENESS FRONTIER USING NET MONETARY BENEFITS AND ITS GRAPHICAL INTERPRETATION

Purpose: Cost-effectiveness analyses aim to identify treatments and policies that maximize benefits subject to resource constraints.  However, the conventional way of identifying the efficient frontier (i.e., the set of non-dominated options) can be algorithmically inefficient, especially when considering many decision alternatives or when performing many sensitivity analyses for which the frontier must be found for each.  Traditional approaches to identify the frontier iteratively remove dominated strategies and do not leverage the relationship between the cost-effectiveness plane and net monetary benefit (NMB).

Method: We use this relationship to design a one-pass algorithm that finds the efficient frontier. Our algorithm is conceptually simple and may better address situations that challenge the conventional approach.  

Result: We prove that identifying the efficient frontier is equivalent to finding policies that maximize NMB for willingness-to-pay levels equal to pairwise incremental cost effectiveness ratios (ICERs). The efficient frontier can be quickly found because evaluation of NMB and comparison of strategies must only be undertaken at these levels and does not require iterative removal of dominated strategies.  Extending our previously published work in Medical Decision Making, we construct a graphical interpretation of our proof that illustrates its intuition and value, highlighting the relationship between the cost-effectiveness and net monetary benefit planes and their interactions.

Conclusion: Our approach offers an alternative to other convex hull identification algorithms which is appropriate to medical decision making because it relies on well-known concepts including the relationship between the cost-effectiveness plane and net monetary benefit.