Data-Sharing for TSCA Risk Evaluations: Context Is Everything
By Ed Rutkowski
When companies receive a request for data from EPA for the purpose of completing a chemical risk evaluation under the Toxic Substances Control Act, they aren’t necessarily aware of how the agency will use the data. At the educational session “Exposure Factors for TSCA Risk Evaluations” held May 21 at AIHA Connect 2025 in Kansas City, three OEHS professionals shared their observations of the EPA risk evaluation process that can prepare companies to meet the agency’s expectations.
The first presenter, Silvia Maberti, PhD, summarized EPA’s approach to exposure characterization, which involves what the agency calls “conditions of use” and “occupational exposure scenarios.” The legislation defines a COU as “the circumstances . . . under which a chemical substance is intended, known, or reasonably foreseen to be manufactured, processed, distributed in commerce, used, or disposed of.” A document on the EPA website (PDF) explains that an OES can be thought of as “a combination of facts, assumptions and inferences that define a discrete situation in which potential exposures might occur.” In the TSCA framework, multiple COUs can be mapped to a single OES, or multiple OESs can be mapped to a single COU. This complexity allows the agency to aggregate data across industries.
In some cases, the framework does not capture the complex reality of workplace exposures. For example, EPA considers “manufacturing” to be a single OES, Maberti said. But manufacturing involves an array of distinct tasks including the activities of maintenance and support personnel in addition to the people who work directly with the chemical of interest. “[EPA has] been doing this rationalization of occupational exposure scenarios and selecting one that represents all the scenarios across the value chain,” Maberti said, although that scenario may not be applicable in all cases.
To perform chemical risk evaluations, which are required by TSCA, EPA has been asking—and occasionally compelling—companies to share their occupational exposure data with the agency. Julie Panko, CIH, FAIHA, explained the importance of providing EPA with proper context for the data companies share. Panko is principal scientist at the consulting firm ToxStrategies and has experience organizing company data for submission to EPA.
“The best outcome is a risk evaluation that accurately represents exposures,” Panko said. For that reason, Panko said, EPA “really value[s] a high quality IH dataset because it reflects actual workplace exposures.”
EPA is primarily interested in air sampling data, Panko said. She emphasized that the agency no longer considers personal protective equipment when evaluating the risks presented by chemical exposures; for example, it will not account for respirators when calculating an inhaled dose.
Panko encouraged companies to provide detailed information regarding the context of the data to help EPA understand its potential comparability—or lack thereof—to specific COUs and OESs. This context includes whether a sample is for a full shift, a partial shift, or a single task, and whether it was collected during routine, non-routine, or turnaround operations. Similarly, it’s not sufficient to tell EPA that a particular sampling and analytical method was used. Companies should also provide details such as laboratory reporting limits and limits of quantification. Most importantly, companies should indicate why the sample was collected, whether it was intended for regulatory compliance or some other purpose.
Panko said that EPA’s risk evaluations tend to group all occupational activities into six broad tasks: loading and unloading, handling of waste, cleaning and maintaining equipment, sampling and analysis, repackaging, and other work activities that occur near the area where a chemical is used. Sometimes, Panko said, the agency uses task data to characterize full-shift exposures even when the task’s duration is less than eight hours. To help ensure EPA uses the data properly, Panko encouraged companies to provide details such as the exposure time and frequency, the number of workers and shifts represented by the data, whether the workers were wearing PPE, any controls that were in place, and whether the task occurred in a part of the facility that meets OSHA’s definition of a regulated area. Because access to regulated areas is restricted, any exposures that occur there will not affect bystanders, which are an important subpopulation for EPA.
EPA’s term for bystanders is “occupational non-users,” or ONUs. This could be a very broad category of individuals for most companies, Panko said, and because ONU exposures are typically very low, they aren’t a priority for OEHS sampling. Unfortunately, Panko said, in the absence of data on ONU exposures, EPA assumes they are exposed at the same level as workers. Companies that don’t have data on ONUs can consider providing area sampling, modeling the exposures, or reporting relevant data from the scientific literature.
The final presenter, Heather Lynch, MPH, DABT, discussed EPA’s concerns related to dermal exposures. Very little data exists on occupational dermal exposures, Lynch said, so EPA has relied heavily on modeling when addressing dermal exposure risk. But the agency’s models include assumptions that don’t reflect workplace realities.
As an example, Lynch explained EPA’s modeling of an exposure resulting from a liquid loading task. The results of the modeling rely to a large degree on the quantity of liquid that remains on the skin after the task is performed. EPA’s default estimate of that amount is very high and is derived from experiments conducted in the 1990s that involved dipping test subjects’ hands in baby oil, wiping off the liquid, and measuring the mass remaining on the skin, Lynch said. Liquid loading in an occupational setting is highly unlikely to involve such a procedure.
Lynch also spoke of her involvement with an effort led by the American Chemistry Council’s exposure assessment working group to assemble a compendium of dermal exposure factors that can help EPA use more realistic assumptions in its risk evaluations. Ideally, though, industry will help fill the data gap by conducting studies of dermal exposures. Lynch suggested that observational studies, interviews, and focus groups can help increase understanding of how many times a worker pours a liquid or touches a surface.
Ed Rutkowski is editor in chief of The Synergist.
Read more coverage of AIHA Connect 2025.
