By Richard Denison
Richard Denison, Ph.D.,
is a Lead Senior Scientist.
Today, EPA posted on its website
risk determinations for four new chemicals it has reviewed under the new standards prescribed by the Lautenberg Act. While the premanufacture notices (PMNs) for these chemicals were received by EPA prior to the June 22 signing
of the new TSCA, EPA has reviewed them in the context of the new requirements. (Unlike reviews of chemicals already in use, which may take some years to conduct, EPA reviews of new chemicals are generally to be completed within 90 days, which is why we’re already seeing these appear so soon after enactment.)
These decisions are notable in that they are the very first formal decisions EPA has made under the new law. Based on an admittedly quick review of the decisions, I’ll offer a few observations.
A number of positive aspects of EPA’s review and posting are apparent:
- For the first time, an affirmative safety decision has been made by EPA for these new chemicals, reflecting a central reform of new chemical reviews that is required under the new law. In each case released today, EPA’s determination is that the chemical “is not likely to present an unreasonable risk.” (I expect EPA chose relatively “easy” cases for this debut.)
- These decisions have been made public, as required under the new law, which was a rarity under the old law.
- Summaries of the bases for the decisions have also been made public, both in the table provided on the website (see above link) and in 3-page “determination documents” for each substance. These documents provide, in one place, clear descriptions of:
- intended, known and reasonably foreseen uses of the chemical;
- a summary of EPA’s findings;
- descriptions of the criteria used and results of its analysis or consideration of:
- environmental fate
- bioaccumulation potential
- human health hazard
- environmental hazard
- potential exposures
- potentially exposed or susceptible subpopulations
This, in my view, constitutes an unprecedented level of transparency for a program that has often felt like a “black box” in the past.
That’s all good news. But there are other concerning features of the reviews that bear mention:
. Substantial amounts of information were claimed confidential – and kept so by EPA, including:
- the identities of the companies submitting the PMNs
- the identities of the substances
- the specific uses of the substances (with one exception)
This is not atypical with PMNs, but such confidential business information (CBI) claims are now subject to a range of new requirements under the Lautenberg Act. It is not clear the extent to which such claims were scrutinized by EPA in these cases, and as the PMNs were submitted and the claims asserted prior to the signing of the new law, it’s not clear these particular claims would have been subject to the new requirements. Going forward, however, we will be seeking assurance that EPA is applying the new CBI requirements to applicable claims in PMNs and also reflecting them in its reviews and decision documents regarding those PMNs.
Estimated, not measured, data
. Essentially all of the environmental fate, hazard and exposure information on which EPA based its decisions is predicted, not measured, and is derived using its suite of estimation models, structure-activity relationships or read-across. Use of such information is the traditional way in which EPA has conducted reviews in the past when companies fail to provide any or adequate safety data in their PMNs (true in a large majority of cases), and it may well not be adequate going forward as the basis for EPA to make the requisite affirmative safety finding.
While they can be useful, such estimation approaches and models have significant limitations
, don’t work well for some classes of chemicals (e.g., certain heavily halogenated compounds), and aren’t available or reliable for some important health endpoints (especially chronic human health effects). Just one example is that EPA’s persistence models typically only examine a chemical’s persistence in air, water soil, and sediment, and its bioaccumulation model is based on uptake from water. Yet some chemicals (e.g., the “Teflon chemical” PFOA) that are “passed” by those models can nonetheless persist and build up in people’s blood.
In the absence of public knowledge of the identities of the chemicals, it’s difficult if not impossible to independently judge whether sole reliance on such information was adequate in these four specific reviews. And assuming these cases are relatively “easy ones,” they should not be viewed as representative, and the approach EPA deemed adequate here should not be viewed as precedent-setting.
A number of observers have called on companies to provide more safety information up front in their PMNs, in order to increase the likelihood of smooth reviews under the Lautenberg Act. We would echo that call and will be watching this issue closely going forward. In addition, the Lautenberg Act provides EPA with ready authority both to require testing in order to obtain adequate information on which to base its decisions, and to either block or condition commencement of manufacture pending receipt of that information.
Need for more than just summaries
. As much as we welcome the greater transparency and extent of documentation of EPA’s review of these four cases, the determination documents EPA has provided are still only summaries and really only provide EPA’s interpretation of the information and analysis it has conducted. The PMN review process routinely generates more detailed review documents that provide its model inputs, outputs and assumptions. But in the past these have rarely been made public, and when they are available, they are heavily redacted due to CBI claims — in our view, overly so.
EDF urges EPA routinely to make its more detailed PMN review documents public, and to redact only information that is truly confidential.
Cursory consideration of exposure and exposed subpopulations
. EPA’s reviews of exposure potential appear incomplete and insufficient, as well as its identification of potentially exposed or susceptible subpopulations. (The law now includes an explicit requirement that EPA account for and mitigate against unreasonable risks to such subpopulations.) In all four cases, EPA noted that exposure potential was not even estimated due to predicted low bioaccumulation and hazard potential. Among other issues, this amplifies the concern we have over EPA’s heavy reliance on limited estimation models for bioaccumulation and hazard, when that in turn leads EPA not to further consider exposure potential.
In addition, in at least one case the basis of EPA’s identification of workers as the only potentially exposed subpopulation seems wholly inadequate. That chemical’s intended use is listed (generically) as “plastic additive.” This begs many questions, for example: which plastic(s), for what uses? Is there potential for the additive to migrate out of the plastic, which could lead to all kinds of potential exposures to a variety of subpopulations? What about releases in post-use waste management of such plastics?
Going forward, EDF urges EPA to do a more thorough review of the exposure side of the risk equation.
So, these first decisions by EPA under the Lautenberg Act are welcome and appear to represent real progress in a number of respects. We recognize the pressure EPA was under not to unduly extend its reviews of PMNs that had been filed prior to the law’s passage. Going forward, however, and in dealing with less “easy” cases, EPA needs to do more to ensure and communicate that its review is based on information sufficient to make the affirmative finding the new law requires: whether each new chemical is or is not likely to present an unreasonable risk.