LAUSR

Abstract EQ‐5D is a 5‐item questionnaire instrument designed to measure health‐related quality of life. It is extremely important, since it is used to measure health benefits in many studies providing evidence for reimbursement decisions by the National Institute for Health and Care Excellence in England and similar policy bodies in other countries. EQ‐5D has been redesigned in a more detailed form (EQ‐5D‐5L), but much existing cost‐effectiveness evidence is based on the older version (EQ‐5D‐3L). Statistical mapping from one version to another is widely used, exploiting data from multi‐instrument surveys incorporating both variants. However, little is known about the robustness of data from such multi‐instrument surveys. We design a randomized experiment to investigate whether inclusion of both versions at different stages in a single interview gives a reliable picture of the relationship between health measures from the two instruments and embed it in individual interviews from the UK Understanding Society household panel. We find that sequencing of the two versions of EQ‐5D within an interview has a significant impact not only on the resulting data but also on the estimated mapping models. We illustrate the non‐negligible effects in two real‐world cost‐effectiveness examples and discuss the implications for future multi‐instrument survey design.