Download PDF
September 9, 2022

Final Revision to the Risk Determination for PV29 Finds that PV29, as a Whole Chemical Substance, Presents an Unreasonable Risk to Human Health

Bergeson & Campbell, P.C.

On September 6, 2022, the U.S. Environmental Protection Agency (EPA) announced the availability of the final revision to the risk determination for the Colour Index Pigment Violet 29 (PV29) risk evaluation issued under the Toxic Substances Control Act (TSCA). 87 Fed. Reg. 54491. EPA states that the revision to the PV29 risk determination reflects its announced policy changes to ensure the public is protected from unreasonable risks from chemicals in a way that is supported by science and the law. EPA determined that PV29, as a whole chemical substance, presents an unreasonable risk of injury to health when evaluated under its conditions of use (COU).

In its September 6, 2022, press release, EPA states that PV29 is used as an intermediate:

  • To create or adjust color of other perylene pigments;
  • In paints and coatings in the automobile industry;
  • In plastic and rubber products in automobiles and industrial carpeting;
  • In merchant ink for commercial printing; and
  • In consumer watercolors.
     

As reported in our March 8, 2022, memorandum, in its draft revision to the PV29 risk determination, EPA found that PV29 presents unreasonable risk to health of workers and occupational non-users from long-term inhalation exposure. EPA states that “[s]pecifically, long-term exposure to PV29 can cause lung toxicity effects known as alveolar hyperplasia, or an adverse increase in the number of cells in the lungs where oxygen transfer occurs.”

Overall, ten of the 14 COUs EPA evaluated drive the unreasonable risk determination due to risks identified for human health. The unreasonable risk presented by PV29 is driven by exposures from manufacturing (including import), processing, certain industrial/commercial uses, and disposal. Four out of 14 COUs do not drive the unreasonable risk, including distribution in commerce; industrial and commercial uses in finished plastic and rubber products for automobile plastics and industrial carpeting; and consumer use in professional-quality watercolor and acrylic artist paint.

EPA states that it used the whole chemical risk determination approach for PV29 in part because there are risk benchmark exceedances for a significant number of COUs spanning most aspects of the chemical lifecycle, from manufacturing (including import), processing, commercial and industrial use, and disposal. According to EPA, there are health effects associated with PV29 exposures for occupational non-users (ONU) and workers who handle PV29 as part of their jobs. Thus, EPA is better positioned to achieve its TSCA objectives by issuing a whole chemical determination for PV29.

The revised risk determination supersedes the COU-specific no unreasonable risk determinations previously issued by order under TSCA Section 6(i) in the January 2021 PV29 risk evaluation.

In addition, EPA states that the risk determination does not reflect an assumption that workers always or appropriately wear personal protective equipment (PPE), even though some facilities might be using PPE as one means to reduce workers’ exposures. EPA notes that its decision not to assume that PPE is always and appropriately used should not be viewed as an indication that EPA believes there is widespread non-compliance with applicable Occupational Safety and Health Administration (OSHA) standards. According to EPA, it received public comments from industry respondents about occupational safety practices currently in use at their facilities. EPA states that it “will consider these comments, as well as other information on use of PPE and other ways industry protects its workers, as potential ways to address unreasonable risk during the risk management process.”

EPA acknowledges that there could be occupational safety protections in place at workplace locations. According to EPA, not assuming use of PPE reflects its recognition that unreasonable risk may exist for subpopulations of workers that may be highly exposed because they are not covered by OSHA standards, or because OSHA’s chemical-specific permissible exposure limits (PEL) are described by OSHA as being “outdated and inadequate for ensuring protection of worker health,” or because OSHA has not issued a chemical-specific PEL.

EPA states that as it moves forward with a risk management rule for PV29, it will “strive for consistency with existing OSHA requirements or best industry practices when those measures would address the identified unreasonable risk.” EPA will propose occupational safety measures in the risk management process that would meet TSCA’s statutory requirement to eliminate unreasonable risk of injury to health and the environment.

Next Steps for PV29

EPA states that it is now moving forward on risk management to protect workers, including those who handle PV29 as a part of their jobs. EPA expects to focus its risk management action on the COUs that drive the unreasonable risk. EPA notes that it “is not limited to regulating the specific activities found to drive unreasonable risk, and may select from among a wide range of risk management requirements.” As a general example, EPA may regulate upstream activities (e.g., processing, distribution in commerce) to address downstream activities (e.g., consumer uses) driving unreasonable risk, even if the upstream activities do not drive the unreasonable risk. There will be a public comment period for any proposed risk management actions.

Commentary

Bergeson & Campbell, P.C. (B&C®) urges readers to track the progression of the PV29 risk management rule given its potential far-reaching implications for similar chemicals with COUs that include chemical substances that are respirable, poorly soluble, low toxicity (PSLT) particles (e.g., dust). EPA identified PV29 as a PSLT particle. Based on the current risk evaluation, we anticipate that EPA will issue a draft 8-hour existing chemical exposure limit (ECEL) for PV29 of 0.04 mg/m3 and B&C expects that EPA will apply this value to all PSLT substances going forward (e.g., new chemical substances and other existing chemical risk evaluations). This value is lower than the U.S. National Institute for Occupational Safety and Health’s (NIOSH) recommended exposure limit (REL) and OSHA PEL for crystalline silica (i.e., 0.05 mg/m3), a substance EPA acknowledged and differentiated from PSLTs because crystalline silica is a poorly soluble particle, but also has “inherently high cytotoxicity.” EPA nevertheless appears to be poised to set an ECEL lower than the REL and PEL for crystalline silica.

B&C fully supports EPA issuing health protective ECELs when no occupational exposure limits (OEL) or other enforceable limit that is sufficiently protective exist. We are, however, not supportive of EPA making unreasonable risk findings and issuing ECELs that are not supported by the best available science and weight of scientific evidence, as required under TSCA and discussed below.

EPA stated in the Final Risk Evaluation for PV29 that “The [Regional Deposited Dose Ratio (RDDR)] software was utilized for dosimetric adjustment across species instead of the multi-path particle dosimetry (MPPD) model because the MPPD software cannot calculate the DAF for hamsters tested in the Elder et al., (2005) study.” EPA does not explain why it is rejecting a model (i.e., MPPD) that can calculate the appropriate dose metric (i.e.retained mass) for PSLTs simply to use hamster data with a model that can only calculate deposited dose (i.e., RDDR). We also note that EPA ultimately rejected using the hamster data for its calculation of a point of departure (POD). EPA appears to prefer RDDR over MPPD “because the particle size data was not robust enough,” but EPA does not explain why this limitation would not apply equally to both models. EPA’s selection of the RDDR model also conflicts with other recent activities conducted by EPA’s Office of Research and Development (ORD), which support the use of MPPD. For example, ORD used MPPD in support of the Integrated Risk Information System derivation of its inhalation reference concentration for benzo[a]pyrene.

On a separate note, EPA also seems to have ignored the scientific consensus that rats are more sensitive than humans to PSLT exposures. An international workshop that included experts from EPA, OSHA, and NIOSH concluded that the “rat is more sensitive than other species and humans in the lung response to PSLT,” and yet in the PV29 risk evaluation, EPA applies an uncertainty factor that would only be appropriate if humans were more sensitive than rats.

B&C views EPA’s use of the RDDR software as a significant vulnerability as EPA moves forward with drafting the risk management rule for PV29. For example, EPA stated that “The change in model [i.e., RDDR rather than MPPD] resulted in unreasonable risk determinations for all ONUs and industrial and commercial use in automobile paint OEM and refinishing condition of use” (emphasis added). These facts, coupled with conflicting statements within EPA’s analysis, hint that EPA’s model selection might have been based on the preferred outcome (that there is unreasonable risk), rather than an objective scientific evaluation to determine if there is unreasonable risk.

B&C reiterates the importance of tracking the progression of PV29 through the risk management rulemaking process, even if PV29 itself is not in a company’s supply chain. B&C views the underlying science in the Final Risk Evaluation for PV29 as fundamentally flawed and inconsistent with EPA’s statutory requirements for making decisions under TSCA using the best available science and weight of scientific evidence. If EPA moves forward with promulgating a risk management rule on PV29 using deposited dose, the ramifications of this decision, if unchallenged, will likely lead to EPA imposing similar and equally unsupportable limits on all PSLT particles — substances that are ubiquitous across industries.