How robust are value judgments of health inequality aversion? testing for framing and cognitive effects

Background: Empirical studies have found that members of the public are inequality averse and value health gains for disadvantaged groups with poor health many times more highly than gains for better off groups. However, these studies typically use abstract scenarios that involve unrealistically large reductions in health inequality and face-to-face survey administration. It is not known how robust these findings are to more realistic scenarios or anonymous online survey administration. Methods: This study aimed to test the robustness of questionnaire estimates of inequality aversion by comparing the following: 1) small versus unrealistically large health inequality reductions, 2) population-level versus individual-level descriptions of health inequality reductions, 3) concrete versus abstract intervention scenarios, and 4) online versus face-to-face mode of administration. Fifty-two members of the public participated in face-to-face discussion groups, while 83 members of the public completed an online survey. Participants were given a questionnaire instrument with different scenario descriptions for eliciting aversion to social inequality in health. Results: The median respondent was inequality averse under all scenarios. Scenarios involving small rather than unrealistically large health gains made little difference in terms of inequality aversion, as did population-level rather than individual-level scenarios. However, the proportion expressing extreme inequality aversion fell 19 percentage points when considering a specific health intervention scenario rather than an abstract scenario and was 11 to 21 percentage points lower among online public respondents compared with the discussion group. Conclusions: Our study suggests that both concrete scenarios and online administration reduce the proportion expressing extreme inequality aversion but still yield median responses that imply substantial health inequality aversion.