Abstract
Value sets for the EQ‐5D‐5L are required to facilitate its use in estimating quality‐adjusted life years. An international protocol has been developed to guide the collection of stated preference data for this purpose and has been used to generate EQ‐5D‐5L valuation data for England. The aim of this paper is report the innovative methods used for modelling those data to obtain a value set.Nine hundred and ninety‐six members of the English general public completed time trade‐off (TTO) and discrete choice experiment (DCE) tasks. We estimate models, with and without interactions, using DCE data only, TTO data only, and TTO/DCE data combined. TTO data are interpreted as both left and right censored. Heteroskedasticity and preference heterogeneity between individuals are accounted for. We use Bayesian methods in the econometric analysis. The final model is chosen based on the deviance information criterion (DIC).Censoring and taking account of heteroskedasticity have important effects on parameter estimation. For DCE data only, TTO data only, and DCE/TTO data combined, models with parameters for all dimensions and levels perform best, as judged by the DIC. Taking account of heterogeneity improves fit, and the multinomial model reports the lowest DIC. This paper presents approaches that suit observed characteristics of EQ‐5D‐5L valuation data and recognise respondents' preference heterogeneity. The methods described are potentially relevant to other value set studies.
Highlights
Ben van Hout, School of Health and Related Research, University of Sheffield, Regent Court, 30 Regent Street Sheffield, Sheffield S1 4DA, UK
We describe a variety of models tested: those that use discrete choice experiment (DCE) data only; those that use time trade‐off (TTO) data only; and those that combine the TTO and DCE data
When the TTO data for each respondent were plotted against the predicted TTO values from the 10‐parameter DCE tariff, we found that most respondents' data followed a negative gradient as expected
Summary
Eighty‐six health states were valued using TTO. These were allocated to 10 blocks with 10 health states in each. When respondents completing a TTO task value health state “x” as worse than dead, they may, at the extreme, prefer to die than to live for 10 years in full health (the lead time) followed by 10 years in “x.” In that case the resultant value, given the variant of TTO used in the EQ‐5D‐5L valuation protocol, is −1. Some respondents use zero as the minimum value more than once, including when valuing the worst health state 55555. This suggests that those respondents did not want to go below zero (e.g., do not believe there is such a thing as a health state so bad that experiencing it for 10 years would be worse than dead). All models in the econometric analyses applied these weights to adjust the distribution of the sample
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
