论文部分内容阅读
本文发展了无条件和条件Ad-logistic回归模型理论,同时实现病例与对照,以及不同发病年龄之病例间的比较分析,所估计的复相对危险度和复归因危险度,包括了危险因素对于发病与否及发病早晚两方面的方差贡献,因而更为客观地评价了其在病因学和发病学中的作用。现有的logistic回归模型,只是本文方法在忽略病例发病年龄差异时的一种特例。为说明方法的应用,文中给出了食管癌和肺癌病例——对照研究的分析实例,并与现有logistic回归结果进行了比较。
In this paper, the unconditional and conditional Ad-logistic regression model was developed to compare the cases and controls, as well as the cases of different ages of onset. The estimated relative risk and the risk of reversion included the risk factors Whether or not the contribution of variance and the incidence of both early and late onset, and thus more objective evaluation of its role in etiology and pathogenesis. The existing logistic regression model is just a special case of this method ignoring the difference in the age of onset. To illustrate the application of the method, an analysis of case-control studies of esophageal and lung cancer is presented and compared with existing logistic regression results.