代谢综合征组分的验证性因子分析

The components of metabolic syndrome analyzed by confirmatory factor analysis

目的用验证性因子分析(CFA)检验代谢综合征(MS)的单因子和四因子模型并比较其拟合优度。方法研究对象1 302例,无已知糖尿病史,通过口服葡萄糖耐量试验(OGTT)检查出糖尿病患者579例。构建四个模型:标准单因子模型包括MS的核心组分(胰岛素抵抗指数、平均动脉压、甘油三酯/高密度脂蛋白胆固醇和腰围);扩展单因子模型包括尿酸和超敏C反应蛋白;四因子模型;原始单因子模型。用CFA比较上述四个模型之间的拟合优度。结果单因子模型的拟和性优于四因子模型,其中标准单因子模型拟合性最好,其比较拟合指数(CFI)0.998,均方根误差的近似值(RMSEA)0.021,赤池信息量准则(AIC)27.2。变量的选择和模型的定义可能是以往的因子分析不支持单因子假说的原因。结论本研究结果支持存在单个潜在致病因子影响MS所有特征表达的假说。

Objective To test the hypothesis weather the components of the metabolic syndrome （MS） could be best described by a single underlying factor using confirmatory factor analysis （CFA）.Methods Atotal of 1 302 patients underwent a 75 g oral glucose tolerance test （OGTT） of which 579 diagnosed as diabetes were included in the study. We formulated four factor models of MS： one-factor standard model, one-factor expanded model, four-factor model and original one-factormodel. CFA was used to compare the goodness of fit of one-factor models with the fit of four-factor models.Results One-factor model was better than four-factor model, of which one-factor standard model was the best goodness-of-fit model with comparative fit index0.998, root mean-square error of approximation 0.021, akaike information criterion 27.2. The different selection of MS variables and model specification explained why previous factor analysis failed to identify a single factor for the MS.Conclusion The study supports the hypothesis that a single underlying factor influences all traits of MS expression.