A threshold inequality aversion parameter (TIAP) approach to interpret distributional cost-effectiveness analysis results.

Abstract

Objectives: Distributional cost-effectiveness analysis (DCEA) requires an inequality aversion parameter to calculate the equally distributed equivalent (EDE). The DCEA decision rule is to choose the strategy with the highest EDE. However, the exact value of the inequality aversion parameter is unknown for most health disparities and settings, hindering the use of DCEA in practice. We therefore propose calculating threshold inequality aversion parameter (TIAP) values in DCEAs that can be interpreted using existing data and conventions.

Methods: We provide the rationale and methods for calculating the TIAP for pairwise and multi-strategy DCEAs. TIAPs can be estimated by finding the inequality aversion parameter that sets the EDEs of two competing strategies equal to each other. The interpretation of TIAPs requires a lower and upper bound of an inequality aversion parameter value range (LBIAR and UBAIR, respectively).

Results: In a pairwise DCEA, a TIAP that is lower than the LBIAR can be interpreted as favoring the equity-improving strategy, whereas a TIAP that is higher than the UBIAR would favor the more cost-effective strategy. TIAPs between the LBIAR and UBIAR require additional context to determine the optimal strategy.

Conclusions: The interpretation of TIAPs is analogous in some ways to how incremental cost-effectiveness ratios (ICERs) are used in conventional cost-effectiveness analysis; ICERs can be calculated without knowing the specific cost-effectiveness threshold, and are then interpreted using empirical estimates or conventions for the setting-specific cost-effectiveness threshold. Although further empirical data and attention toward inequality aversion parameters is needed, reporting TIAPs could enable widespread use of DCEA.