RISK FACTORS OF LUNG CANCER IN NONSMOKING WOMEN

不吸烟者患肺癌的原因知之甚少。的 这项提案的目的是确定肺癌的危险因素 在非吸烟者中使用多中心研究中收集的现有数据， 非吸烟者患肺癌这项多中心研究的主要目标是 是为了确定被动吸烟是否会导致非吸烟者患肺癌。我们 现在，我想使用本研究中收集的数据来评估：a） 来自各种来源的室内空气污染作为导致 肺癌; B）蔬菜，水果， 和类胡萝卜素和肺癌的风险; c）之间的关系 以前的肺部疾病和肺癌的风险; d）关系 癌症家族史与肺癌风险之间的关系;以及 职业暴露作为肺癌危险因素的作用。数据 可用于约650例肺癌病例，1200例基于人口的 对照组和350名结肠癌对照组， 美国的大都市地区（亚特兰大、休斯顿、洛杉矶县， 路易斯安那州和旧金山弗朗西斯科湾区）。的 与上述每个因素相关的独立效应及其 被动吸烟与肺癌的潜在混杂效应 协会将进行评估。数据分析将包括检查每个 根据相关协变量（例如， 种族、年龄、研究领域）。优势比和相应的置信度 将计算间隔，以估计肺癌相关风险。 暴露在每一个可疑的因素。多元logistic 回归模型将用于确定是否有任何观察到的 风险因素是其他风险因素关系的混杂因素。

Little is known about the causes of lung cancer in nonsmokers. The objective of this proposal is to identify risk factors for lung cancer in nonsmokers using existing data collected in a multicentered study of lung cancer in nonsmokers. The primary goal of this multicentered study was to determine if passive smoking causes lung cancer in nonsmokers. We now would like to use the data collected in this study to evaluate: a) the role of indoor air pollution from a variety of sources as a cause of lung cancer; b) the relationship between intake of vegetables, fruits, and carotenoids and risk of lung cancer; c) the relationship between previous lung diseases and risk of lung cancer; d) The relationship between family history of cancer and risk of lung cancer; and e) The role of occupational exposures as a risk factor of lung cancer. Data are available on about 650 lung cancer cases, 1200 population-based controls, and 350 colon cancer controls who were interviewed in 5 metropolitan areas in the US (Atlanta, Houston, Los Angeles County, Louisiana, and San Francisco Bay Area) between 1985 and 1991. The independent effect associated with each of the above factors and its potential confounding effect on the passive smoking-lung cancer association will be evaluated. Data analyses will include examining each of the suspected risk factor stratified by relevant covariates (e.g. race, age, study area). Odds ratios and corresponding confidence intervals will be calculated to estimate the lung cancer risk associated with exposure to each of the suspected factors. Multiple logistic regression models will be used to determine whether any of the observed risk factors are confounders for other risk-factors relationships.