Rowe J N, Springer J A
Risk Anal. 1986 Jun;6(2):171-80. doi: 10.1111/j.1539-6924.1986.tb00205.x.
Using the most comprehensive inhalation study available, (Wagner, et al., 1974), the dose-response effects of the four major types of asbestos fibers (amosite, anthophyllite, crocidolite, and chrysotile: Canadian, Rhodesian) for lung cancer have been determined. From linear regression analysis of the animal data and five human epidemiology studies giving a wide range of risk estimates, slopes of the curves have been determined and lifetime risk estimates made. Projected risks for rats are presented with and without surface area (s.a.) conversion factors. On the basis of cumulative exposure, the geometric mean of the point estimates for the human studies (0.0146) is quite close to the geometric mean of the animal data (0.0179 without s.a.; 0.0122 with s.a. calculations). These values also match quite well if one of the studies (McDonald, et al.) is eliminated (geometric mean = 0.031) due to qualitatively different exposure considerations (mining and milling vs. industrial environments). Animal risks based on a concentration per day basis (assuming an average 70-year lifespan for humans) are below the lowest human estimate but within 5-6 fold (less) of the projected risk from nonsmoking asbestos workers (2.2 X 10(-3) using the Hammond et al. study.
利用现有最全面的吸入研究（瓦格纳等人，1974年），已确定了四种主要类型石棉纤维（铁石棉、直闪石、青石棉和温石棉：加拿大产、罗德西亚产）对肺癌的剂量反应效应。通过对动物数据进行线性回归分析以及五项给出广泛风险估计值的人类流行病学研究，确定了曲线斜率并进行了终生风险估计。给出了有和没有表面积（s.a.）转换因子情况下大鼠的预计风险。基于累积暴露，人类研究点估计值的几何平均值（0.0146）与动物数据的几何平均值（不考虑表面积时为0.0179；考虑表面积计算时为0.0122）相当接近。如果由于暴露考虑因素在性质上不同（采矿和研磨与工业环境）而排除其中一项研究（麦克唐纳等人的研究）（几何平均值 = 0.031），这些值也相当匹配。基于每天浓度的动物风险（假设人类平均寿命为70岁）低于人类的最低估计值，但在不吸烟石棉工人预计风险的5至6倍以内（更低）（使用哈蒙德等人的研究为2.2×10⁻³）。
