Good Science, Bad Regulation, and Toxic Risk Assessment

Abstract

Regulation of toxic substances is an extremely complex, uncertain, and controversial enterprise. The regulatory process is customarily divided into two discrete functions: risk assessment ostensibly is a scientific activity that develops estimates of health hazards at varying exposure levels, while risk management is a political activity that balances competing interests and values to determine whether identified toxic risks should be considered unacceptable or tolerable. This sharp distinction between the scientific and social policy dimensions of toxics regulation is embodied in the Environmental Protection Agency's (EPA's) guidelines for estimating carcinogenic hazards, which provide that risk assessments must "use the most scientifically appropriate interpretation" and should "be carried out independently from considerations of the consequences of regulatory action." The requirement for adoption of the "most scientifically appropriate interpretation" reflects EPA's current priority on attaining "good science" in risk-assessment proceedings. In other words, EPA and other federal agencies now stress the need for scientifically credible risk assessments and presume that their analyses should be grounded exclusively on the best available scientific theories and data even if the resulting predictions do not achieve the degree of reliability ordinarily required for valid scientific conclusions.