Chen James J, Moon Hojin, Kodell Ralph L
Division of Biometry and Risk Assessment, National Center for Toxicological Research, Food and Drug Administration, Jefferson, AR 72079, USA.
Regul Toxicol Pharmacol. 2007 Jun;48(1):45-50. doi: 10.1016/j.yrtph.2006.10.008. Epub 2006 Dec 12.
Risk assessment involves an analysis of the relationship between exposure and health related outcomes to derive an allowable exposure level or to estimate a low-dose risk. Acceptable levels of human exposure for non-cancer effects generally are derived by dividing an experimental no-observed-adverse-effect-level or a lower confidence limit benchmark dose by a product of several uncertainty factors. This paper presents a hierarchical modeling framework for a probabilistic approach to non-cancer risk assessment. The hierarchical model integrates the distributions of uncertainty factors and the distribution of the actual exposure level to construct the dose-response model for the proportion of population at risk and the dose-response model for the expected proportion of population at risk for a given exposure distribution. The proposed approach is based on the use of the BMDL (lower confidence limit on the benchmark dose) as a POD (point of departure) for risk assessment of non-cancer effects.
风险评估涉及对暴露与健康相关结果之间的关系进行分析，以得出可允许的暴露水平或估计低剂量风险。非癌症效应的可接受人体暴露水平通常是通过将实验性未观察到不良效应水平或较低置信限基准剂量除以几个不确定性因素的乘积得出的。本文提出了一种用于非癌症风险评估概率方法的分层建模框架。该分层模型整合了不确定性因素的分布和实际暴露水平的分布，以构建处于风险中的人群比例的剂量反应模型以及给定暴露分布下预期处于风险中的人群比例的剂量反应模型。所提出的方法基于使用BMDL（基准剂量的较低置信限）作为非癌症效应风险评估的起始点（POD）。
