McClellan R O
Chemical Industry Institute of Toxicology, Research Triangle Park, North Carolina 27709, USA.
Regul Toxicol Pharmacol. 1994 Dec;20(3 Pt 2):S142-68.
This article provides a brief overview of the report "Science and Judgment in Risk Assessment," prepared by a Committee of the National Research Council/National Academy of Science in response to a U.S. Environmental Protection Agency request mandated by the Clean Air Act Amendments of 1990 (CAAA-1990). The report critiques EPA's current approaches for characterizing human cancer risks from exposure to chemicals and offers recommendations for the conduct of future cancer risk assessments, especially those required in implementing the CAAA-1990 provisions which are concerned with hazardous air pollutants. The report offers specific recommendations that address the role of default options, data needs, methods and models, uncertainty, variability, and the aggregation of data. A cross-cutting theme of the report is the use of an iterative approach in which screening assessments with limited data and, of necessity, default options used in the absence of specific scientific data may be performed initially followed by subsequent assessments, as needed, in which increasing amounts of data are developed and incorporated. In some instances, the specific data on a given chemical or pollutant source will replace conservative default options used in earlier assignments. The report includes two authored appendices that address issues related to the use of default options and their replacement by specific scientific information. One appendix by Finkel advocates a principle of "plausible conservatism" for choosing and altering default options and in making cancer risk estimates. A second appendix by McClellan and North advocates the full use of scientific information in the risk assessment process. This article gives major attention to the key aspects of the NRC/NAS report, especially those dealing with the use and replacement of default options. The default options and the extent to which the options are replaced with specific science have major impact on the final quantitation of cancer risk for exposure to chemicals.
本文简要概述了美国国家研究委员会/美国国家科学院的一个委员会应1990年《清洁空气法修正案》（CAAA - 1990）授权的美国环境保护局要求编写的《风险评估中的科学与判断》报告。该报告批评了美国环境保护局目前描述化学品暴露对人类癌症风险的方法，并为未来癌症风险评估的开展提出了建议，特别是那些在实施CAAA - 1990中与有害空气污染物相关条款时所需的评估。该报告提出了具体建议，涉及默认选项的作用、数据需求、方法和模型、不确定性、变异性以及数据汇总。该报告的一个贯穿各领域的主题是采用迭代方法，即最初可使用有限数据进行筛选评估，并且在缺乏具体科学数据时必然要使用默认选项，随后根据需要进行后续评估，在后续评估中会获取并纳入越来越多的数据。在某些情况下，关于特定化学品或污染物源的具体数据将取代早期评估中使用的保守默认选项。该报告包括两篇作者撰写的附录，讨论了与默认选项的使用以及用具体科学信息取代默认选项相关的问题。芬克尔撰写的一篇附录主张在选择和更改默认选项以及进行癌症风险估计时遵循“合理保守主义”原则。麦克莱伦和诺思撰写的第二篇附录主张在风险评估过程中充分利用科学信息。本文主要关注了NRC/NAS报告的关键方面，特别是那些涉及默认选项的使用和取代的方面。默认选项以及这些选项被具体科学信息取代的程度对化学品暴露癌症风险的最终定量有重大影响。
