论文部分内容阅读
本文系刘韵源同志根据IARC于1980年邀请Breslow和Day编写出的《Statistical Methods in Cancer Research Volumel-the analysis of Case-control studies》一书的译稿为基础,并参阅近期文献编写出《病例—对照研究统计方法讲座》,共分十讲。为适应我国读者的要求和建议,本刊将分期连续刊载。原中国医学科学院肿瘤研究所将不再成册出版,特告拟订阅者注意。十讲题目如下: (一) 病例—对照研究中的偏倚、混杂和因果关联概念 (二) 相对危险度和归因危险度 (三) 两暴露水平资料及统计推断 (四) 2×2表的合并信息与混杂控制 (五) 多暴露水平分层资料的分析 (六) 两暴露水平1:1配对数据的分析 (七) 两暴露水平1:M配对数据的分析 (八) 多暴露水平1:1配对数据的统计推断 (九) 多层资料的无条件Logistic同归模型(上),(下) (十) 配对数据的条件Logistic同归分折
This article is based on the translation of the book “Statistical Methods in Cancer Research Volumel-the Analysis of Case-control Studies” written by IARC in 1980 and invited by Breslow and Day, and refer to the recent literature to compile the case-control. Lectures on Statistics Methods, divided into ten lectures. In order to meet the requirements and suggestions of our readers, the journal will be published in stages. The original Institute of Cancer Research of the Chinese Academy of Medical Sciences will no longer be published in a booklet and will be noticed by subscribers. Ten topics are as follows: (a) Case-control study of bias, promiscuity, and causality concepts (ii) Relative risk and attribution risk (c) Data and statistical inference of two exposure levels (iv) 2x2 Combined information and confounding control (v) Analysis of multiple exposure level stratified data (six) Analysis of two exposure levels 1:1 paired data (seven) Two exposure levels 1: Analysis of M paired data (eight) Multiple exposure levels 1: 1 Statistical inference of paired data (9) Unconditional logistic regression model of multi-layer data (top), (bottom) (10) Conditional logistic regression analysis of paired data