Appendix: Estimating utility values for non-alcoholic steatohepatitis health states: a discrete choice experiment

<b>These are peer-reviewed supplementary appendices and figures for the article '</b><b>Estimating utility values for non-alcoholic steatohepatitis health states: a discrete choice experiment</b><b>' published in the</b><b> </b><b><i>Journal of Comparative Effectiveness Research</i></b><b>.</b><br><b>Appendix file 1:</b> General Public DCE Survey<b>Appendix file 2:</b> Sample DCE choice questions<b>Appendix File 3:</b> Estimated marginal utilities among NASH patients by fibrosis stage based on results from conditional logit model<b>Appendix File 4: </b>NASH Check scores and frequency of symptoms by NASH Fibrosis Stage<b>Appendix File 5:</b> Linear regression results exploring associations between background and clinical characteristics with EQ-5D utility values in the patient sample (n=154)<b>Background:</b> This study estimated utility values for non-alcoholic steatohepatitis (NASH). Previous studies have assumed that health-related quality of life does not vary between the early stages of NASH. <b>Materials &amp; Methods: </b>Discrete choice experiment (DCE) surveys estimated the value of avoiding fibrosis progression. Patients also completed the EQ-5D-5L. Marginal rates of substitution estimated utility change associated with fibrosis progression. <b>Results:</b> DCE surveys were completed by the UK general public (n = 520) and patients with NASH (n = 154). The utility decline between fibrosis stages F1 and F4 decompensated was between -0.521 to -0.646 (depending on method). <b>Conclusion:</b> Three methods were used to estimate utilities for NASH, each one showed sensitivity to advancing fibrosis, including in the early stages, which is often considered asymptomatic.