胃癌危险因素的病例对照研究──条件Logistic回归分析和通径分析

A CASE-CONTROL STUDY OF RISK FACTORS OF CASTRIC CANCER-CONDITIONAL LOGISTIC REGRESSION ANALYSIS AND PATHOLOGICAL ANALYSIS

该文对２００例胃癌病人，按１：１配比进行了病例对照研究。单因素分析，从２９个研究因素中初筛出１８个可疑危险因素，条件Ｌｏｇｉｓｔｉｃ回归分析，筛选出８个危险因素：快食、饮酒、食蔬菜量少、精神因素、食咸菜、食酸菜、胃癌家族史和文化程度低为胃癌发病的危险因素，其相对危险性分别为４．５２、２．８７、１．６０、３．１７、１．５３、１．８６、３．９４和１．５３。通径分析表明，饮酒、食酸菜、食蔬菜量、快食和精神因素对胃癌的发生有直接影响作用；饮酒、食蔬菜量、食咸菜和文化程度低４因素通过各自的中间环节起间接作用。

A case-control study of 200 pairs was carried out in Shanxi during 1990-1991. 29 variables were analysed with the sing factor analysis. 18 factors were comfirmed as the suspicious risk factors. Conditional Logistic regression analysis of the above 18 factors was used Results showed that the following eight risk factors was used. Results showed that the following eight risk factors were associated with gastric cancer,i. e. ,fast eating habil,drink deficiency of fresh vegetablcs, psychic trauna ,salted foods, sourctout , the family history of gastric cancer and lower education; and the odds ratio were 4.52 , 2.87,1. 60 ,3.17 , 1.53 ,1.86 ,3.94, 1.53 respectively. The causal model with 8 risk factors of gastric cancer waste up and studied with pathological analysis. The result indicated that 7 risk factors had direct and indirect influence on the occurence of gastric cancer respectively.