• Gordon G Liu, Hongyan Wu, Minghui Li, Chen Gao, and Nan Luo.
• National School of Development, Peking University, Beijing, China.
• Value Health. 2014 Jul 1; 17 (5): 597-604.

ObjectiveTo generate a Chinese general population-based three-level EuroQol five-dimensios (EQ-5D-3L) social value set using the time trade-off method.MethodsThe study sample was drawn from five cities in China: Beijing, Guangzhou, Shenyang, Chengdu, and Nanjing, using a quota sampling method. Utility values for a subset of 97 health states defined by the EQ-5D-3L descriptive system were directly elicited from the study sample using a modified Measurement and Valuation of Health protocol, with each respondent valuing 13 of the health states. The utility values for all 243 EQ-5D-3L health states were estimated on the basis of econometric models at both individual and aggregate levels. Various linear regression models using different model specifications were examined to determine the best model using predefined model selection criteria.ResultsThe N3 model based on ordinary least square regression at the aggregate level yielded the best model fit, with a mean absolute error of 0.020, 7 and 0 states for which prediction errors were greater than 0.05 and 0.10, respectively, in absolute magnitude. This model passed tests for model misspecification (F = 2.7; P = 0.0509, Ramsey Regression Equation Specification Error Test), heteroskedasticity (χ(2) = 0.97; P = 0.3254, Breusch-Pagan/Cook-Weisberg test), and normality of the residuals (χ(2) = 1.285; P = 0.5259, Jarque-Bera test). The range of the predicted values (-0.149 to 0.887) was similar to those estimated in other countries.ConclusionsThe study successfully developed Chinese utility values for EQ-5D-3L health states using the time trade-off method. It is the first attempt ever to develop a standardized instrument for quantifying quality-adjusted life-years in China.Copyright © 2014 International Society for Pharmacoeconomics and Outcomes Research (ISPOR). Published by Elsevier Inc. All rights reserved.

24 keywords...

hide…