Some methodological issues with the analysis of preference-based EQ-5D index score

摘要: The EQ-5D index is widely used to assess the preference-based health status. In this paper, we examine analytical issues of regression models for US score. We propose a two-part approach model special features index. first part logistic probability reaching maximum score 1.0. second rest scores that are less than 1.0, which can be least squares with robust standard errors conditional mean, or quantile quantiles such as median. show has some desirable not available in previously published methods illustrate proposed data from Medical Expenditure Panel Survey. method may other utility related quality life similar features.