化学品健康风险评估过程中的数据质量评估技术概述

自1938年美国首先颁布管控化学品的法律以来,健康风险评估逐步发展,各个国家和地区相继颁布文件,并已形成较为完善的评估框架.由于在化学品健康风险评估的过程中,存在大量收集引用的数据及信息,因此数据质量评估是保证风险评估结果可信的关键.目前为止美国环保署(environmental protection agency,EPA)和欧盟《化学品的注册、评估、授权和限制》(regulation concerning the registration,evaluation,authorization and restriction of chemicals,REACH)法规均对数据质量评估方法进行了详细的规定.两个地区均采用数据评分(1—4)与证据权重(weight of evidence,WOE)(1/2)相结合的评估方法,不同之处在于欧盟更侧重于对数据整体的相关性、可靠性和充分性进行评估和打分,U.S EPA则更为细致具体,其侧重于不同情境下的数据分组分析,根据不同的评分领域和指标,规定高置信度、中置信度、低置信度和数据不可接受的标准.通过总结对比欧盟及美国的数据质量评估方法,建议我国采用数据打分与证据权重相结合的定量评估方法,并明确规定不同情景下不同数据来源的打分规则和标准,使数据质量评估过程系统化.

Overview of data quality assessment techniques during chemical health risk assessment

Since the United States first promulgated laws and regulations on chemicals in 1938,health risk assessment has gradually developed.Various countries and regions have issued documents and have formed a relatively complete assessment framework.Because of a large number of data and information collected during the human health risk assessment,data quality assessment is the key to ensuring the credibility of risk assessment results.Until now,the U.S.Environmental Protection Agency(EPA)and EU's Regulation Concerning the registration,evaluation,authorization and restriction of Chemicals(REACH)regulations have both carried out data quality assessment methods.Both regions use the method that combine data scoring(1—4)and weight of evidence(WOE)(1/2).The difference is that the EU focuses more on the overall relevance,reliability and adequacy of the data,and the U.S EPA is more detailed and specific.EPA focuses on the analysis of data grouping in different scenarios,and stipulates high confidence,medium confidence,low confidence and unacceptable data.By summarizing and comparing the data quality assessment methods of the European Union and the United States,it is recommended that China adopt a quantitative assessment method that combines data scoring and weight of evidence,and clearly stipulate scoring rules and standards for different data sources in different scenarios to systematically evaluate the data quality.