Huang I-Chan, Frangakis Constantine, Atkinson Mark J, Willke Richard J, Leite Walter L, Vogel W Bruce, Wu Albert W
Department of Epidemiology and Health Policy Research, College of Medicine, University of Florida, Gainesville, FL, USA.
Health Serv Res. 2008 Feb;43(1 Pt 1):327-39. doi: 10.1111/j.1475-6773.2007.00745.x.
To compare different approaches to address ceiling effects when predicting EQ-5D index scores from the 10 subscales of the MOS-HIV Health Survey.
Data were collected from an HIV treatment trial. Statistical methods included ordinary least squares (OLS) regression, the censored least absolute deviations (CLAD) approach, a standard two-part model (TPM), a TPM with a log-transformed EQ-5D index, and a latent class model (LCM). Predictive accuracy was evaluated using percentage of absolute error (R(1)) and squared error (R(2)) predicted by statistical methods.
A TPM with a log-transformed EQ-5D index performed best on R(1); a LCM performed best on R(2). In contrast, the CLAD was worst. Performance of the OLS and a standard TPM were intermediate. Values for R(1) ranged from 0.33 (CLAD) to 0.42 (TPM-L); R(2) ranged from 0.37 (CLAD) to 0.53 (LCM).
The LCM and TPM with a log-transformed dependent variable are superior to other approaches in handling data with ceiling effects.
比较在根据MOS-HIV健康调查的10个分量表预测EQ-5D指数得分时处理天花板效应的不同方法。
数据来自一项HIV治疗试验。统计方法包括普通最小二乘法（OLS）回归、截尾最小绝对偏差（CLAD）法、标准两部分模型（TPM）、对EQ-5D指数进行对数转换的TPM以及潜在类别模型（LCM）。使用统计方法预测的绝对误差百分比（R(1)）和平方误差（R(2)）评估预测准确性。
对EQ-5D指数进行对数转换的TPM在R(1)方面表现最佳；LCM在R(2)方面表现最佳。相比之下，CLAD表现最差。OLS和标准TPM的表现处于中间水平。R(1)的值范围为0.33（CLAD）至0.42（TPM-L）；R(2)的值范围为0.37（CLAD）至0.53（LCM）。
具有对数转换后因变量的LCM和TPM在处理存在天花板效应的数据方面优于其他方法。
