OBJECTIVES: To explore the relationship between incremental non-healthcare sector societal costs and incremental quality adjusted life years (QALYs) in published ICER economic evaluations.

METHODS: ICER drug reviews conducted between May 2019 and November 2021 were identified. Data including condition, intervention(s), comparator(s), perspective of analysis, and incremental QALYs and costs were extracted for each comparison and perspective undertaken (i.e., healthcare sector and societal). Details regarding the domains included in the societal perspective were summarized descriptively, and the relationship between incremental non-healthcare sector societal costs and incremental QALYs were assessed graphically and for fit.

RESULTS: Twenty ICER reviews were included in the analysis. Of the 17 reviews where a societal perspective analysis was undertaken, 9 were excluded because: a) incremental costs from the societal perspective were higher than incremental costs from the health system perspective (n=1), b) treatments were less costly and equally, or more, effective (n=3), c) QALY gains were marginal (i.e., <0.015; n=1), or d) insufficient data was available due to rounding (n=4). For the 8 remaining reviews (28 pairwise comparisons versus usual care), a linear relationship (R² = 0.8126) was observed between the incremental non-healthcare sector societal cost savings and incremental QALYs. Our results suggest that for every additional lifetime discounted QALY gained, incremental non-healthcare societal cost savings increased by $12,000 (2021 USD) on average (range: $5,600 to $108,000). Our analysis is limited by the small number of reviews included in our analysis.

CONCLUSIONS: Economic modelers could assume this relationship holds for analyses that lack societal level information. This estimate offers a placeholder value to update once additional data are available. Future research should explore this relationship in other reviews and assess the variability in findings according to the characteristics of the condition such as severity and mean age of onset.