Plain Language Summary

What is it about?

Traditional cost-effectiveness analysis treats all health benefits equally, regardless of who receives them, but it does not consider whether improvements are distributed fairly across different groups. Distributional cost-effectiveness analysis is a newer method that explicitly incorporates fairness by using a value called an inequality aversion parameter, which measures how much society values equal distribution of health versus maximizing total population health. However, the exact value of this parameter is usually unknown, limiting practical use of this approach. This study proposes calculating threshold inequality aversion parameter values, which represent the minimum level of concern for fairness needed to favor an equity-improving intervention over one that produces more total health but distributes it unequally. These threshold values provide a practical way to interpret results even when the precise inequality aversion parameter is unknown.

How was the research conducted?

The researchers developed mathematical methods to calculate threshold inequality aversion parameter values by finding the point where 2 competing strategies would have equal social value when accounting for both total health and fairness. The methods involved mathematical modeling using the Atkinson Index, an established measure from economics that quantitatively incorporates the inequality of outcomes across a population. The researchers analyzed hypothetical healthcare strategies with different combinations of total health benefits and equity impacts across population groups. The hypothetical examples illustrated how researchers can report and interpret meaningful distributional results without needing to know exact societal preferences for equity in advance.

What were the results?

The central finding is that threshold inequality aversion parameter values can be calculated and interpreted using existing evidence about reasonable ranges for inequality aversion. When comparing 2 strategies, a very low threshold value suggests the equity-improving strategy should be favored, whereas a very high threshold value suggests the more cost-effective strategy should be chosen. The researchers found that interpreting results becomes more complex when comparing 3 or more strategies, requiring careful examination of multiple threshold values. An important finding was that the lowest threshold value does not necessarily indicate the best strategy in multi-strategy comparisons.

Why are the results important?

These results provide a practical solution to a major barrier preventing widespread use of distributional cost-effectiveness analysis in healthcare decision making. Healthcare decision makers can use these threshold values to determine whether interventions that improve equity should be adopted by comparing calculated thresholds against evidence-based ranges. Patients from disadvantaged groups experiencing health disparities may benefit because this approach provides a structured way to justify funding equity-improving interventions. Over time, as more studies report threshold values, accumulated evidence could establish reference databases, further helping equity-informed healthcare decisions and potentially reducing health disparities.

What are the strengths and weaknesses of this study?

The main strength is providing a practical solution that enables distributional cost-effectiveness analysis without requiring precise knowledge of inequality aversion parameters, removing a major barrier to incorporating equity into healthcare evaluations. The primary limitation is that interpreting threshold values still requires understanding plausible parameter ranges, and current evidence is limited and varies across studies and settings. Future research should conduct systematic reviews of existing inequality aversion estimates, develop standardized methods for measuring societal preferences for health equity, and explore how to handle intersecting inequalities when multiple disadvantaged characteristics overlap.

^{Note: This content was created with assistance from artificial intelligence (AI) and has been reviewed and edited by ISPOR staff. For more information or for inquiries on ISPOR’s AI policy, click here or contact us at info@ispor.org.}