EPIDEMIOLOGIC STUDY OF LUNG CANCER AND WOOD DUST

点击翻译按钮获取中文摘要

A population-based case-control study of lung cancer is proposed to determine if employment in wood-related occupations is associated with increased risk of lung cancer, and to identify specific exposures in these jobs which might explain any increased risks found. Specifically, we will: a) test the hypothesis that persons employed in operational jobs in any of the four major segments of the wood industry are at increased risk of lung cancer; b) test the hypothesis that persons exposed to respirable wood dust in particular, as well as several other potential carcinogens present in wood-related occupations, are at increased risk of lung cancer; and c) determine whether excess risk associated with any of the above factors varies by histologic type, or is modified by other known risk factors such as tobacco use, dietary intake of beta-carotene and vitamin C, or family history of lung cancer. Male cases aged 18-74 years who reside in an 11-county area of western Washington where wood-related occupations are particularly common will be identified from the Cancer Surveillance System of the Fred Hutchinson Cancer Research Center (a SEER registry). Controls of similar age from the same geographic area will be selected by random digit dialing. Approximately 769 cases and an equal number of controls will be interviewed over a three-year period. Subjects will be interviewed by telephone using a structured questionnaire and experienced interviewers. A standard mailed self-administered questionnaire will be used to collect limited dietary information. Exposure to wood dust will be assessed on an individual basis by two industrial hygienists, taking into account job title and typical duties, type of manufactured product, and participation in activities known to produce measurable concentrations of respirable dust. For industries in western Washington in which total and respirable dust levels are not well-characterized, random full-shift sampling will also be accomplished during the first year. Approximately 680 samples will be taken at a minimum of eight industrial sites over two seasons. Stratified analysis and logistic regression will be used to control for the effect of major risk factors and to identify any important effect modification by these factors. Polytomous logistic regression will be used along with pairwise logistic regression to investigate differences in risk among specific histologic types of lung cancer. It is estimated that if wood dust is etiologically related to lung cancer, it may account for 15% of male lung cancer cases in parts of western Washington and 3% nationwide. This would represent approximately 3,000 cases per year in the United States.