Eliciting Societal Preferences for Non-health Outcomes: A Person Trade-Off Study in the Context of Genomics

Daniel Sheen,Tessa Peasgood,Ilias Goranitis

DOI: https://doi.org/10.1016/j.clinthera.2023.07.004

Abstract:Purpose: Economic evaluations of health technologies traditionally aim to maximize population health outcomes measured by using quality-adjusted life-years (QALYs). Non-health outcomes, however, may have high social value, and their exclusion has the potential to bias decisions regarding allocation of health care resources. This research positions Australian participants as societal decision-makers to explore their willingness to trade-off health gains in adults for non-health benefits in families with a child affected by a rare disease. Methods: To estimate the social value of the different health care interventions, a person trade-off (PTO) method was used. PTOs present participants with groups of beneficiaries that vary in terms of the number of individuals who will benefit, the individuals' characteristics, their expected benefits, or a combination, and ask which group should be prioritized. Each trade-off presented health gains from the treatment of moderate physical and mental health conditions described by the 3-level version of the EuroQol 5-Dimension (EQ-5D-3L) health states. The health gains in these groups were traded-off against non-health gains in families accessing diagnostic genomic testing, and equivalence values were calculated, using median and ratio of means methods, based on the ratio of the group sizes at the point of equivalence. Participants were recruited through Prolific and were stratified according to age, sex, and education. The impact of participant characteristics on equivalence values was assessed using Kruskal-Wallis H tests and ordinary least-squares log-linear regressions. Findings: Participants (N = 434) positioned as societal decision-makers were generally willing to trade-off adult health gains with the familial non-health benefits of genomic testing, showing a preference for valuing both types of outcomes within public health policy. The aggregation of preferences generated 2 weightings for genomic testing against each health treatment, an unadjusted value and a reweighted value to match target demographic characteristics. Converted into QALY value per test, it was found that participants valued the non-health benefits of genomic testing between 0.730 and 0.756 QALY. A minority of participants always prioritized diagnostic genomic testing over the physical (6.0%) or mental (4.6%) health treatments, with a larger minority always prioritizing the physical (15.4%) or mental (14.8%) health treatments. Implications: The findings indicate that participants perceived the non-health parental benefits in children experiencing rare disease to have comparable value to health gains in adults experiencing the moderate physical or mental health conditions described using EQ-5D-3L. These findings suggest that the benefits of genomic tests would be underestimated if only health benefits are included in economic evaluations.