A Sensitivity Analysis of Recently-Proposed Distributional and Data-Derived Approaches to Noncancer Risk Assessment. A. M. Wilson, P. Williams, S. J. S. Baird, L. R. Rhomberg, and J. S. Evans; The Baird Group, Incorporated, Gradient Corporation, and Harvard School of Public Health
For years, risk assessors, risk managers, and even the general public have complained about the methods used to calculate reference doses (RfDs) for noncancer effects. Their criticisms point to an antiquated methodology that does not accurately determine a safe dose level, does not adequately characterize uncertainty, and offers little scientific basis for the uncertainty factors used. In recent years, there have been numerous attempts in the published literature to address these issues. Some approaches involve the use of distributional methods to characterize all or some of the factors in the RfD calculation and/or the use of data-derived estimates for the uncertainty factors. In turn, these approaches lead to yet another question: Given the currently suggested improvements to RfD methodology, which factor has the greatest effect on the final result? In response, this paper describes a sensitivity analysis of published distributional and data-derived approaches to characterizing dose, interspecies extrapolation, and intraspecies extrapolation. This analysis uses dose-response information from the recent non-cancer risk assessment for ethylene oxide. It employs empirical distributions of the ED50, distributional and data-derived adjustment factors for interspecies extrapolation, and data-derived adjustment factors for intraspecies variability to estimate a population threshold distribution. For the range of currently-proposed adjustment factors, interspecies extrapolation strongly influences the spread and intraspecies variability strongly influences the center (location) of the population threshold distribution. Various methods for modeling the dose-response have significantly less effect on the location and spread of the population threshold distribution.
Go to . . .
2000 SRA Annual Meeting Table of Contents
2000 SRA Annual Meeting Author Index
Main Abstracts Menu Page
RiskWorld Home Page