EPA Intends Draft TSCA Systematic Review Protocol to Strengthen Science Used in Chemical Risk Evaluations
On December 20, 2021, the U.S. Environmental Protection Agency (EPA) released the Draft Systematic Review Protocol Supporting TSCA Risk Evaluations for Chemical Substances (Draft Protocol) for public comment. 86 Fed. Reg. 71891. According to EPA’s press release, the Draft Protocol will strengthen EPA’s approach to reviewing and selecting the scientific studies that are used to inform Toxic Substances Control Act (TSCA) chemical risk evaluations and ensure that EPA has the best tools under TSCA to protect human health and the environment. Comments on the Draft Protocol are due February 18, 2022.
On February 16, 2021, the National Academies of Sciences, Engineering, and Medicine (NASEM) announced the availability of a report entitled The Use of Systematic Review in EPA’s Toxic Substances Control Act Risk Evaluations. EPA requested that NASEM review EPA’s 2018 guidance document on Application of Systematic Review in TSCA Risk Evaluations (2018 SR Methodology) and associated materials. In its final report, NASEM states that it “was in strong consensus that the processes used by [the Office of Pollution Prevention and Toxics] OPPT do not meet the evaluation criteria specified in the Statement of Task (i.e., comprehensive, workable, objective, and transparent).” NASEM recognizes the “substantial challenges in implementing review methods on the schedule required by” the Frank R. Lautenberg Chemical Safety for the 21st Century Act (Lautenberg), but concluded that those challenges “have not yet been successfully met.” NASEM recommended that EPA revise its approach, incorporating components from existing methodologies, including EPA’s Integrated Risk Information System (IRIS) program.
In its press release, EPA states that it did not use the 2018 SR Methodology for its 20 ongoing risk evaluations and will not use it again. According to EPA, the Draft Protocol reflects the current approaches for the ongoing TSCA risk evaluations for the 20 high-priority designated chemicals, manufacturer-requested risk evaluations, and supplemental evaluations of asbestos and 1,4-dioxane. The appendices to the Draft Protocol include details specific to the manner in which EPA used the Draft Protocol for each of these chemical risk evaluations.
In the Draft Protocol, EPA streamlined the data evaluation of human health hazards by enabling the use of recent and future IRIS systematic reviews. According to EPA, “[t]his also allows more seamless updates of literature evaluation into the TSCA systematic review evaluations based upon the interoperability of software platforms being utilized in both IRIS and TSCA systematic review evaluation programs.”
The Draft Protocol also includes new methods intended to reduce bias and improve evaluation consistency between reviewers and across chemicals. EPA states that these improvements include coordinated data evaluation training and calibration exercises for reviewers, the development of additional internal evaluation guidance, and improvement of evaluation forms. According to EPA, as recommended by NASEM, data quality evaluation under the Draft Protocol would involve two levels of review, a primary review and a secondary quality-control review for each study. In cases where the two reviewers are not in agreement, an explicit conflict resolution step is included. EPA notes that in response to commenters, the Draft Protocol no longer includes a quantitative/weighted scoring system for data evaluation. Rather, the Draft Protocol applies ordinal rankings to guide the qualitative categorization of high, medium, low, or critically deficient for each data evaluation metric.
The Science Advisory Committee on Chemicals (SACC) will hold a virtual three-day peer review meeting to consider and review the Draft Protocol April 19-21, 2022. Stakeholders must register online to receive the webcast meeting link and audio teleconference information. Registration instructions will be announced on the SACC website by early March 2022. After the public meeting, SACC will prepare meeting minutes summarizing its recommendations to EPA. The meeting minutes will be posted on the SACC website and in Docket No. EPA-HQ-OPPT-2021-0414.
B&C commends EPA’s commitment to address the NASEM recommendations in the Draft Protocol. EPA’s publication of the Draft Protocol represents a significant step forward toward ensuring compliance with the scientific standards under TSCA Section 26 and the implementing regulations under 40 C.F.R. Section 702, that require “a systematic review method … [and] uses a pre-established protocol” as part of the weight of scientific evidence.
EPA acknowledged in the Draft Protocol that “Previously [in the 2018 SR Methodology], EPA did not have a complete clear and documented TSCA systematic review (SR) Protocol.” EPA released the Draft Protocol to correct this deficiency. B&C also notes, however, that EPA did not state in the Draft Protocol how, or if, it intends to evaluate whether application of the 2018 SR Methodology to its completed risk evaluations (e.g., asbestos, citing 2018 SR Methodology) and issued TSCA Section 4 test orders (e.g., trans-1,2-dichloroethylene, citing 2018 SR Methodology) introduced errors, for example, in data collection, data quality rankings, and evidence integration, that would compromise (e.g., under- or overestimate) unreasonable risk determinations or ordered testing for data.
Many of the documents cited as supporting information for updating the Draft Protocol (e.g., Office of Health Assessment and Translation, 2015) were available prior to EPA issuing the 2018 SR Methodology. Rather than utilizing these documents at the time, EPA developed the 2018 SR Methodology de novo. B&C recognizes that in 2018 and now EPA had to consider other evidence streams (e.g., exposure) beyond human health hazard. Regardless, we view this as a potential vulnerability for EPA as it moves forward with risk management on its completed risk evaluations, given that EPA chose to develop its own methodology in 2018 rather than incorporating and adapting existing methodologies that represented the best available science at the time.
We also note that the Draft Protocol states that it “does not include a quantitative/weighted scoring system for data evaluation.” Readers may be surprised, however, to see that EPA retained a quantitative approach for metric rankings (e.g., >1 and <1.7) that is subsequently used to select an ordinal ranking and qualitative categorization for an overall quality level (e.g., High). EPA describes this approach in the appendices of the Draft Protocol for each discipline starting with “Physical and chemical properties” in Appendix K.
Finally, we recognize that the scientific methods used to inform systematic review are not static and that EPA’s Draft Protocol will likely undergo additional updates as the science evolves. B&C encourages readers to consider submitting comments to EPA to assist with ensuring that the “final” Protocol presents the approaches in a manner that is supported by the best available science and that will aid interested persons with utilizing these methods for evaluating their application in EPA’s risk evaluations.