目的:研究生产和生活性青石棉暴露与恶性肿瘤死亡之间的关系。方法:利用1984年在云南省大姚县青石棉污染区建立的含1249例健康人群的回顾性队列,通过COX等比例风险模型计算青石棉接触时间以及生产风炉、原料处理、刷墙、砌灶和浆衣等过程中的青石棉暴露对恶性肿瘤死亡的风险比(hazard ratio,HR)及其95%可信区间(condence interval,CI)。结果:截至2011年4月30日,队列中共有121例恶性肿瘤死亡病例。调整年龄、性别、婚姻、文化程度、吸烟和饮酒等混杂因素后,与接触期限≤5年相比,青石棉接触6~10年的HR=2.898(95%CI:1.496~5.616)、11~15年的HR=4.159(95%CI:2.169~7.975)、〉15年的HR=4.233(95%CI:1.182~9.886),且有显著的剂量-反应关系(Ptrend〈0.05);与不生产风炉相比,零星生产风炉的HR=2.533(95%CI:1.549~4.142),成批生产风炉的HR=2.541(95%CI:1.642~3.935);与不处理原材料相比,机器粉碎原材料HR=1.811(95%CI:1.211~2.709),人工粉碎原材料HR=1.785(95%CI:1.072~2.793);刷墙4次以上HR=1.736(95%CI:1.000~3.012);砌灶HR=2.229(95%CI:0.97~5.122);浆衣HR=3.307(95%CI:1.522~7.184)。结论:生产和生活性青石棉接触均可增加恶性肿瘤的死亡风险。
Objective: To investigate the association between environmental exposure (living environment and production environment) to crocidolite and the cancer mortality risk. Methods: A cohort including 1 249 healthy residents recruited from crocidolite-contaminated area in Dayao County, Yunnan Province in 1984 was retrospectively analyzed. The COX regression model was used to estimate hazard ratio (HR) and 95% confidence interval (CI) for cancer mortality according to time of exposure to crocidolite and the crocidolite exposure in the production and living environments of blast-furnace manufacturing, material processing, whitewashing the wall, masonry stove building, and starching clothing with slurry. Results: A total of one hundred and twenty-one deaths were identified by April 30, 2011. After adjusting for the confounding factors including age, gender, marriage, education, smoking and alcohol consumption, it was found that as compared with population exposed to crocidolite in five years or less, HRs (95% CIs) for cancer mortality in populations exposed to crocidolite in 6-10, 11-15 and more than 15 years were 2.898 (1.496-5.616), 4.159 (2.169-7.975), and 4.233 (1.182-9.886), respectively, and a dose-response relationship (Ptrend〈0.05) was observed; compared with population not engaged in production of blast-furnace, the HRs (95% CIs) for cancer mortality in populations engaged in sporadic production and mass production of blast-furnace were 2.533 (1.549-4.142) and 2.541 (1.642-3.935), respectively; compared with population not engaged in material processing, the HRs (95% C/s) for cancer mortality in populations using crushing machine in material processing and hand- crushing in material processing were 1.811 (1.211-2.709) and 1.785 (1.072-2.793), respectively; the HRs (95% CIs) for cancer mortality in populations whitewashing the wall for 4 times or more, building the masonry stove and starching clothing with slurry were 1.736 (1.000-3.012), 2.22