CALY-SWE value set: An integrated approach for a valuation study based on an online-administered TTO and DCE survey

Purpose: To explore and develop methods to use TTO and DCE data collected in self-administered online surveys to elicit a CALY-SWE value set.

Methods: Building on existing methodological knowledge around DCE and TTO studies, we optimized the web survey in an integrated approach that consisted of a qualitative face validity study, iterative web survey development including a three-stage roll-out, a customized experimental design, a sample size simulation. Based on the inconsistencies of TTO answers per participant, we assessed TTO data quality by calculating a score, and examined the effect of excluding TTO data according to this score on the value set modelling.

Results: Participants in the quality study informed improvements in the survey’s visual design and phrasing. Based on the sample size simulation, we judged a sample size of 1500 with a balance of six DCE and five TTO questions to be appropriate for the valuation study. Change made for the second stage, for example the introduction of a learning state and of color-coding, improved TTO data quality. Excluding TTO answers per-participant based on the score lead to an improved TTO data foundation for the model by several metrics such as no inconsistent coefficients and reduced standard errors relative to the coefficient magnitude.

Conclusion: Developing value sets was feasible with online administered DCE and TTO questions if the web survey is sufficiently optimized and coherent with the experimental design. The severity of inconsistencies could be used to identify and exclude poor quality TTO data to strengthen the value set modelling.