饮酒与男性肺癌发病关系的前瞻性队列研究

Alcohol consumption and the risk of lung cancer in males： a prospective cohort study

目的研究我国男性饮酒与肺癌发病风险之间的关联。方法对开滦集团队列（2006--2015年）男性人群每两年随访一次，收集饮酒相关信息及肺癌发病资料；此外，检索开滦附属医院医疗信息系统、开滦集团社保系统、唐山市医疗保险系统，补充收集随访过程中可能遗漏的肺癌新发病例。以不饮酒人群为参比组，利用Cox风险比例回归模型分别分析饮酒频率、饮酒类型与男性肺癌发病风险的关联，计算发病风险比（南艉）及其95％CI。按照吸烟状况对研究人群分层，分别分析从不吸烟与既往吸烟／现在吸烟人群中饮酒与肺癌发病的关联。结果共纳入101751名男性，总随访808146．56人年，中位随访时间8．88年，共收集肺癌新发病例913名。调整年龄、吸烟状态、吸烟量、文化程度、粉尘暴露、BMI等潜在的混杂因素后，既往饮酒、现在饮酒频率〈1次／d、现在饮酒频率≥1次／d人群肺癌发病风险HR值（95％CI）分别为1．30（0．90～1．88）、0．80（0．64～1．01）和1．04（0．85～1．27）。饮用啤酒／红酒（HR=0．91，95％CI：0．69～1．20）和饮用白酒（HR=0．99，95％CI：0．83。1．19）与肺癌发病无统计学关联。依据吸烟状况分层分析，所得结果类似。结论本研究结果尚不支持饮酒与男性肺癌发病风险存在关联。

Objective To investigate the association between alcohol consumption and lung cancer risk in Chinese males. Methods Information on alcohol consumption and outcomes were collected on a biennial basis among males in Kailuan Cohort （2006-2015）. In addition, electronic databases of hospitals affiliated to Kailuan Community, Insurance Systems of Kailuan Community and Tangshan were also used for supplementary information retrieval. Cox proportional hazards regression models were used to evaluate the hazard ratio （HR） and 95%CI of baseline frequency and type of alcohol consumption associated with lung cancer risk in males. Non-drinkers were used as control group. Results A total of 101 751 males were included and 913 new lung cancer cases were identified in the Kailuan male cohort study, with a total follow-up time of 808 146.56 person-years and a median follow-up time of 8.88 years by 31 December 2015. After adjusting for potential confounding factors, the HR of former drinkers, occasional drinkers （〈 1/day） and drinkers （≥ 1/day） were 1.30 （95%CI： 0.90-1.88）, 0.80 （95%CI： 0.64-1.01） and 1.04 （95%CI： 0.85-1.27）, respectively, compared with non-drinkers. In addition, drinking beer/red wine （HR=0.91, 95%CI： 0.69-1.20） and white wine （HR=0.99, 95%CI： 0.83-1.19） showed no significant association with lung cancer. The results were similar when stratified analysis were conducted. Conclusion Our study results don＇ t support the hypothesis that alcohol consumption is significantly associated with the risk of lung cancer in males.