Pennington David, Crettaz Pierre, Tauxe Annick, Rhomberg Lorenz, Brand Kevin, Jolliet Olivier
Life Cycle Systems, Swiss Federal Institute of Technology, Lausanne.
Risk Anal. 2002 Oct;22(5):947-63. doi: 10.1111/1539-6924.00263.
In Part 1 of this article we developed an approach for the calculation of cancer effect measures for life cycle assessment (LCA). In this article, we propose and evaluate the method for the screening of noncancer toxicological health effects. This approach draws on the noncancer health risk assessment concept of benchmark dose, while noting important differences with regulatory applications in the objectives of an LCA study. We adopt the centraltendency estimate of the toxicological effect dose inducing a 10% response over background, ED10, to provide a consistent point of departure for default linear low-dose response estimates (betaED10). This explicit estimation of low-dose risks, while necessary in LCA, is in marked contrast to many traditional procedures for noncancer assessments. For pragmatic reasons, mechanistic thresholds and nonlinear low-dose response curves were not implemented in the presented framework. In essence, for the comparative needs of LCA, we propose that one initially screens alternative activities or products on the degree to which the associated chemical emissions erode their margins of exposure, which may or may not be manifested as increases in disease incidence. We illustrate the method here by deriving the betaED10 slope factors from bioassay data for 12 chemicals and outline some of the possibilities for extrapolation from other more readily available measures, such as the no observable adverse effect levels (NOAEL), avoiding uncertainty factors that lead to inconsistent degrees of conservatism from chemical to chemical. These extrapolations facilitated the initial calculation of slope factors for an additional 403 compounds; ranging from 10(-6) to 10(3) (risk per mg/kg-day dose). The potential consequences of the effects are taken into account in a preliminary approach by combining the betaED10 with the severity measure disability adjusted life years (DALY), providing a screening-level estimate of the potential consequences associated with exposures, integrated over time and space, to a given mass of chemical released into the environment for use in LCA.
在本文的第1部分，我们开发了一种用于生命周期评估（LCA）中癌症效应量计算的方法。在本文中，我们提出并评估了非癌症毒理学健康效应的筛选方法。该方法借鉴了基准剂量的非癌症健康风险评估概念，同时注意到与LCA研究目标中监管应用的重要差异。我们采用诱导比背景高10%反应的毒理学效应剂量的中心趋势估计值，即ED10，为默认的线性低剂量反应估计值（βED10）提供一致的出发点。这种对低剂量风险的明确估计在LCA中是必要的，但与许多传统的非癌症评估程序形成了鲜明对比。出于实际原因，在所提出的框架中未实施机制阈值和非线性低剂量反应曲线。本质上，出于LCA的比较需求，我们建议首先根据相关化学排放侵蚀其暴露边际的程度来筛选替代活动或产品，这种暴露边际的侵蚀可能表现为疾病发病率的增加，也可能不表现为疾病发病率的增加。我们在此通过从12种化学品的生物测定数据中推导βED10斜率因子来说明该方法，并概述了从其他更容易获得的测量值（如无可见不良作用水平（NOAEL））进行外推的一些可能性，避免了因化学品而异导致保守程度不一致的不确定性因子。这些外推有助于初步计算另外403种化合物的斜率因子；范围从10^(-⁶)到10³（每毫克/千克 - 天剂量的风险）。通过将βED10与严重程度测量指标伤残调整生命年（DALY）相结合，在一种初步方法中考虑了效应的潜在后果，提供了与暴露相关的潜在后果的筛选水平估计值，该估计值在时间和空间上进行了整合，针对释放到环境中用于LCA的给定质量的化学品。
