Some Challenges with EPA’s TSCA Data Requirements

By Ed Rutkowski

The Toxic Substances Control Act requires EPA to consider several subpopulations when determining whether a chemical presents unreasonable risk to health or the environment. One of those subpopulations is workers, and in recent years EPA has been requiring companies to submit occupational exposure data to the agency. At an AIHA Connect session held May 20 in Kansas City, three OEHS professionals drew on their own experiences to present recommendations for how companies can successfully engage with EPA.

Although working with EPA requires time and effort, companies will likely find that it’s in their best interests to meet the agency’s expectations. “The more you can engage with them, the better it’s going to be for you,” said Heather Lynch, MPH, DABT, principal with Integral Consulting, whose experience includes the submission of occupational exposure data to EPA on behalf of consortia.

As Lynch indicated, fulfilling an EPA request for data can involve significant give-and-take between the company and the agency. Companies must decide whether they want to voluntarily submit data to the agency or wait until they’re requested—or compelled—to do so. In Lynch’s estimation, “there’s a little bit more wiggle room” regarding what the agency requires from companies that voluntarily submit data.

She encouraged companies to provide as much context as possible about their data—not only descriptive statistics such as the mean, the geometric mean, the 95th percentile of a distribution, and the 95th upper confidence limit, but also an assessment of potential outliers and a presentation of methods for left-censored data.

A key challenge is figuring out where a company’s similar exposure groups (SEGs) fit in the EPA’s framework, which involves “conditions of use” (COUs) and “occupational exposure scenarios.” In the context of TSCA, a COU is “the circumstances . . . under which a chemical substance is intended, known, or reasonably foreseen to be manufactured, processed, distributed in commerce, used, or disposed of.” Most SEGs do not fall neatly into these parameters.

Other potential complications are that EPA asks for data on non-routine tasks and on exposures to what TSCA calls “occupational non-users,” or ONUs. Companies that provide data on non-routine tasks should include a description of the exposures’ frequency to help the agency understand what’s happening in their facilities, Lynch said. And some companies may not have ONUs, the term for individuals who work in process areas but do not directly handle the chemical of interest. To meet the requirement for data on ONUs, some companies submit area sampling data as a surrogate, while others submit sampling data for a worker on an adjacent line or a process engineer who supports the operating unit. The worry, Lynch explained, is that if nothing is provided, EPA will use data from a worker such as an operator, which will likely overestimate ONU exposures.

Following Lynch’s presentation, Cathy Sotelo, MPH, CIH, spoke about her company’s experiences sharing data with EPA. Sotelo, the global occupational health leader at Stepan Company, a manufacturer of specialty chemicals, described the agency’s risk evaluation and management process as “a little bit of a black box.” For example, EPA doesn’t share how it uses the data it receives from industry or how it weights the data in its decision-making process.

Companies that choose to voluntarily engage with EPA can do so either alone or as part of a consortium. Working with a consortium, as Stepan did, can have some advantages: for example, the consortium might assume the costs of the sampling and analysis performed to meet EPA’s data requirements, and the data itself will be anonymized—a potential benefit for company leaders reluctant to provide the federal government with specific information about their workplace exposures. As Sotelo explained, a challenge with consortia is that responding to EPA’s requests takes longer and involves aligning your sampling strategy with that of several other companies.

Engaging with EPA requires companies to consider several possibilities for their data. Does the company want to share data only for worst-case scenarios? For all facilities, or just one plant? For which groups of workers, and for which work activities?

Companies that need to modify sampling and analytical methods to meet EPA requirements should be prepared for the time required to implement these changes. “It’s not something that you can just do tomorrow or next week,” Sotelo said.

The final presenter, Sylvia Maberti, PhD, the exposure sciences principal at Exxon Mobil Biomedical Sciences Inc., shared lessons learned from her experiences working with a different consortium that submitted data to EPA on butadiene. Since butadiene is regulated by OSHA, the companies in the consortium had a large amount of data to share, though it didn’t cover every population of interest to EPA. OEHS sampling focuses on exposed groups; workers with no exposure are not typically sampled. But EPA wants proof that specific groups are not exposed. “Proving that negative can be a challenge,” Maberti said.

Another challenge was that the data was collected as part of a targeted sampling strategy; extrapolating it to all workers, as EPA prefers, can lead to an overestimation of exposure. Additionally, the laboratory analysis of the sampled data did not meet EPA’s requirements for the limit of detection.

The session, titled “Case Studies and Lessons Learned of Sharing Occupational Exposure Data with EPA,” is one of several being held as part of AIHA Connect 2025 that address concerns from the OEHS community with the implications of TSCA requirements.

Ed Rutkowski is editor in chief of The Synergist.