Richard Denison, Ph.D., is a Lead Senior Scientist.
[This post is adapted from comments I provided for the science policy panel at the June 27, 2017, forum TSCA Reform: One Year Later, co-sponsored by Environmental Law Institute, Bergeson & Campbell, P.C., Environmental Defense Fund, and George Washington University Milken Institute School of Public Health.]
I don’t know anyone who opposes EPA using the best science it can and considering all the evidence in making decisions.[pullquote]The irony here is that core features of the final rules – each the result of changes since their proposal made in response to chemical industry comments – actually move us away from any meaningful realization of what best available science means.[/pullquote]
So why is it that this science stuff is so controversial? It’s long been a battleground across all of what EPA does, and the debate over reform of the Toxic Substances Control Act (TSCA) was no exception. I have no doubt this will continue unabated into implementation of the amendments to TSCA made by last year’s Lautenberg Act.
Science policy issues are among the most “cultish” of any policy issues I have ever dealt with. Different camps have formed, each with its own belief system, each seeing a right way and a wrong way of doing science. Each is highly suspicious of the others, including what they mean by each word.
At the risk of appearing cultish myself, I want to briefly discuss my concerns about the final prioritization and risk evaluation rules in relation to the term “best available science.”
Let me first say that section 26(h) of the Lautenberg Act [15 U.S.C. 2625(h)] itself already requires this term to be applied to science decisions EPA makes under the law. Its codification in rules is not needed for it to have effect. Yet this term is clearly important to some who were involved in writing these final rules.
Care to guess how many times this term appear in the final risk evaluation rule? A dozen? Two dozen? Three dozen? Would you believe there are 37 references to “best available science”? That compares to just four references in the proposed rule issued in January.
Somebody really likes this term, and got it into the rules along with a definition. Hmmm.
Whatever the motivation, the irony here is that core features of the final rules – each the result of changes since their proposal made in response to chemical industry comments – actually move us away from any meaningful realization of what this term intends.
First, a major criticism of traditional risk assessment for decades has been its failure to adequately account for the real world: the fact that there are typically multiple uses and sources of exposure to a given chemical, with the potential to affect multiple subpopulations.
I thought we finally had made progress toward addressing this flaw through last year’s TSCA reform: The law requires, and the proposed rules sought to codify the requirement, that EPA must conduct broad reviews of chemicals across their full lifecycles and accounting for their known, intended and reasonably foreseen uses.
Yet the changes made to the final rules represent a renewed effort to move us squarely away from that, by allowing and facilitating EPA to examine only certain use of and exposures to a chemical; little explanation, let alone specific criteria, are provided as to how these use exclusions will be determined.
In reforming TSCA, Congress explicitly required that EPA determine whether or not a chemical substance, not individual uses, presents unreasonable risk, and to do so by conducting comprehensive risk evaluations. This is because, while exposures resulting from certain uses of a chemical viewed in isolation may present low risk to some groups of people, when multiple exposures are combined and when all potentially susceptible subpopulations are considered, such a chemical may well present unreasonable risk and warrant restrictions.
Yet the final risk evaluation rule would not only allow EPA to exclude some uses altogether, but it could make individual risk determinations about specific uses it does examine and set them aside, with no requirement ever to consider them collectively.
How is that best available science?
Second, perhaps my longest-running battle in the chemicals arena has been trying to ensure there is adequate information on chemicals in order to determine their safety. TSCA’s own preamble set that as national policy in 1976, calling for the development of such information and placing the onus of doing so squarely on the companies that make and use chemicals.
Starting with EDF’s 1997 report called Toxic Ignorance, we have drawn attention to how little information is available even on the most widely used chemicals. Some limited progress has been made since then, but the fundamental problem remains. Why?
The chemical industry has always wanted it both ways on this one: It seeks to set an ever-higher bar EPA must meet to take action on a chemical, routinely challenging the science and information used. But it also fights any efforts to require companies to provide the information the agency needs to meet that bar.
Again, I thought we were finally making progress with TSCA reform. Core aspects of that reform addressed the widespread recognition that EPA lacked adequate authority to require testing of chemicals, and the proposed rules sought to ensure that EPA could use its enhanced authority to develop that information where needed far enough ahead of having to make prioritization decisions and risk determinations that it could still meet the law’s aggressive deadlines.
Yet the final rules seem intent on undermining all this: They seek to cut off early information development – and then argue that, given the deadlines, there isn’t time for anything other than very short-term testing. They would require EPA always first to request voluntary submissions, eating up precious time, despite multiple past failures of such voluntary approaches. All of this before EPA could ever invoke the new testing authorities Congress just gave it. Among other problems, this would introduce a strong bias for EPA only to select relatively data-rich chemicals in order to meet its deadlines, rather than dedicating its resources initially to chemicals that present the greatest potential concerns.
How is this best available science?