吸烟、饮酒与胃癌关系的条件logistic回归分析  

Conditional logistic analysis of smoking, alcohol consumption and gastric cancer

作　　者：魏跃红[1] 吕桦[2] 倪进发[2] 叶冬青[2] 臧桐华[2] 

机构地区：[1]广州市疾病预防控制中心,广东广州510080 [2]安徽医科大学公共卫生学院流行病和卫生统计学系,安徽合肥230032

出　　处：《疾病控制杂志》2006年第2期116-119,共4页Chinese Journal of Disease Control and Prevention

基　　金：安徽省教育厅科研基金(2002kj150)

摘　　要：目的探索吸烟、饮酒及其交互作用与胃癌发生的关系。方法采用配对病例对照研究，收集了2003年8月--2004年8月303例胃癌新发病例和303例健康对照。对可能的危险因素进行了单因素、调整混杂因素（10年前人均年收入、文化程度、性格、慢性胃部疾病史）的条件logistic回归分析。估计其OR值及95％置信区间。结果调整前后，随着吸烟年数延长、每天吸烟量增大及吸烟包年数增多，胃癌的OR值呈上升趋势（趋势检验，P=0．0214、P=0．0121、P=0．0009）。吸烟年限〉30年、每天吸烟量〉20支和综合指标吸烟包年数〉20者均显示出与胃癌有较强的正关联，其调整后的OR值分别是4．54（95％CI=1．20～17．13）、5．65（95％CI=1．43-22．29）和7．25（95％CI=1．70～30．84）；未发现饮酒与胃癌有统计学关联（P=0．242）、未发现调整混杂因素后饮酒与吸烟之间有交互作用，其交互项Wald x^2=0．0276，P〉0．05。结论吸烟与胃癌的发病呈正相关；单独饮酒与胃癌发生无明显关系，饮酒并不增加吸烟者患胃癌的风险，两者无协同作用。Objective To explore the relationship between smoking, alcohol consumption and gastric cancer in Lujiang, Anhui province. Methods A total of 303 newly diagnosed cases with gastric cancer and 303 healthy people matched by sex and age as controls were analyzed by a case-control study in Lujiang from Aug 2003 to Aug 2004. Odds ratios （OR） and its 95% confidence interval （CI） were estimated by conditional logistic model analysis. Confounding factors such as year income, education level, character and history of chronic disease were adjusted. Results The results showed that smoking was positive associated with gastric cancer, while alcohol consumption was no statistical signification. No interaction was found between alcohol consumption and smoking （Wald 7.2 = 3. 129, P = 0. 077）. Conelu. sions There was positive relationship between smoking and gastric cancer in rural areas.

关 键 词：胃肿瘤 吸烟 饮酒 

分 类 号：R735.2[医药卫生—肿瘤] R749.62[医药卫生—临床医学]