Steenland K, Sieber K, Etzel R A, Pechacek T, Maurer K
National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention, Cincinnati, OH 45226-1998, USA.
Am J Epidemiol. 1998 May 15;147(10):932-9. doi: 10.1093/oxfordjournals.aje.a009383.
The relative risk of coronary artery disease among never smokers exposed to environmental tobacco smoke (ETS) versus never smokers not exposed to ETS is approximately 1.2 based on more than a dozen epidemiologic studies. Most of these studies have controlled for the major heart disease risk factors, but residual or uncontrolled confounding remains a possible explanation for the epidemiologic findings. The authors studied 3,338 never-smoking adults aged 17 years or older, who are representative of all US never smokers, in the 1988-1991 Third National Health and Nutrition Examination Survey (NHANES III) to determine whether selected risk factors for heart disease differ between ETS-exposed and -nonexposed persons. Both self-reported ETS exposure (at home and at work) and serum cotinine levels were available, the latter reflecting recent ETS exposure. After adjustments were made for age, sex, race, and education among adults aged 17 years or older, no significant differences were found between the ETS exposed and the nonexposed for any of 13 cardiovascular risk factors with the exception of dietary carotene, which was lower among the exposed. On the other hand, significant positive linear trends were found between serum cotinine and two risk factors (body mass index and alcohol consumption), and significant inverse trends were found with dietary carotene. There were also few differences between exposed and nonexposed never smokers among adults aged 40 years or older, who are most at risk of heart disease. In this group, however, there was an inverse linear trend between serum cotinine and high density lipoprotein cholesterol (p < 0.001). This finding could result from ETS exposure rather than be an indication of confounding; a similar inverse trend was found for children, confirming other results in the literature. Overall, these data suggest little potential for confounding by the heart disease risk factors studied here when ETS exposure is determined by self-report.
基于十几项流行病学研究，暴露于环境烟草烟雾（ETS）的非吸烟者与未暴露于ETS的非吸烟者相比，患冠状动脉疾病的相对风险约为1.2。这些研究大多对主要的心脏病风险因素进行了控制，但残留或未控制的混杂因素仍是这些流行病学研究结果的一个可能解释。作者在1988 - 1991年第三次全国健康和营养检查调查（NHANES III）中研究了3338名17岁及以上的非吸烟成年人，这些人代表了美国所有非吸烟者，以确定暴露于ETS和未暴露于ETS的人群中所选心脏病风险因素是否存在差异。自我报告的ETS暴露情况（在家中和工作场所）以及血清可替宁水平均可得，后者反映近期的ETS暴露情况。在对17岁及以上成年人的年龄、性别、种族和教育程度进行调整后，除了膳食胡萝卜素（暴露组较低）外，在13种心血管风险因素中，暴露于ETS和未暴露于ETS的人群之间未发现显著差异。另一方面，血清可替宁与两个风险因素（体重指数和饮酒量）之间存在显著的正线性趋势，与膳食胡萝卜素存在显著的负趋势。在40岁及以上、心脏病风险最高的成年人中，暴露和未暴露的非吸烟者之间也几乎没有差异。然而，在这个年龄组中，血清可替宁与高密度脂蛋白胆固醇之间存在负线性趋势（p < 0.001）。这一发现可能是由ETS暴露导致的，而非混杂因素的迹象；在儿童中也发现了类似的负趋势，证实了文献中的其他结果。总体而言，这些数据表明，当通过自我报告确定ETS暴露情况时，此处研究的心脏病风险因素几乎没有混杂的可能性。
