Restoring the credibility of risk assessment: A vital need under TSCA reform

Richard Denison, Ph.D., is a Senior Scientist.

The primary means by which chemical risks are to be judged under current legislative proposals for reform of the Toxic Substances Control Act (TSCA), including the Safe Chemicals Act of 2011 (S. 847), is through risk assessment – a key demand of industry.  Yet traditional risk assessments have often fallen short of protecting public health and have sometimes taken decades to identify a “safe” level of exposure to certain chemicals.  As a result, public and health and environmental community confidence in risk assessment is very low.  There are also major technical deficiencies in current risk assessment methodologies that must be addressed if it is to serve as a credible basis for determining chemical risks.  For example, we now know that there are many chemicals for which any level of exposure poses some risk, yet traditional risk assessment assumes a safe level exists for nearly all chemicals.

The Safe Chemicals Act includes provisions to ensure that EPA’s use of risk assessment incorporates the best available science, initially by requiring EPA to rely on the recent recommendations of the nation’s foremost scientific body, the National Academy of Sciences, as to how EPA can improve its practice of risk assessment. Implementing the recommendations is critical to restoring the credibility of and public confidence in risk assessment.

A key recommendation of Science and Decisions, the NAS’ so-called “Silver Book” of risk assessment released in 2009, is to reverse the common practice used by EPA of assuming that there is a level of exposure – a “threshold” – to a chemical below which there is no risk.  Science tells us that many chemicals have effects even at very low doses; moreover, even where such a threshold may be seen for a chemical in a highly controlled study conducted using genetically homogeneous laboratory animals, the extent of variability in the human population as a whole often effectively erases such a threshold.  Hence, NAS recommends that only if EPA has strong affirmative evidence for a threshold in the human population should it assume there is a wholly “safe” level of exposure.

A second key NAS recommendation is that EPA needs to assess risk so as to ensure protection of vulnerable populations, including developing fetuses and children that are typically more susceptible to the effects of chemical exposures. Third, EPA needs to assess “aggregate” exposures to a chemical, accounting for the often multiple sources of exposure to a given chemical.

Finally, a second NAS report urges that, where science allows, EPA risk assessments should identify groups of chemicals to which people may be exposed that act similarly or produce the same or similar health effects, and assess the “cumulative” exposures to all of those chemicals.

Moving the practice of risk assessment into the 21st century through adoption of these recommendations simply must be at the core of any proposal to rely on risk assessment as the primary means to address chemicals under a reformed TSCA.