Abstract
Chemical release incidents in the United States involve hazardous chemicals that can harm nearby communities. A historical tracking of these chemical release incidents from 1991 to 2014 across up to 16 states has been conducted by The Agency for Toxic Substances and Disease Registry (ATSDR), utilizing the Hazardous Substances Emergency Events Surveillance (HSEES) and the National Toxic Substance Incidents Program (NTSIP) systems. By analyzing surveillance data, patterns of these different chemical releases can be identified to develop and construct a health-protective course of action. Physiologically Based Pharmacokinetic (PBPK) models can simulate chemical exposures during acute chemical incidents. For a retrospective study of an acute chemical release in 2012, we examined the components necessary to integrate PBPK-modeled exposure assessments in ATSDR's Assessment of Chemical Exposure (ACE) program. We focused on data from a published investigation of vinyl chloride (VC) exposure to assess the utility of PBPK in evaluating exposures among residential populations near the release site. The initial estimate from the real-time air monitoring at the release site revealed that air levels greatly exceeded the Acute Exposure Guideline Levels (AEGL) of 1200 ppm, with PBPK models predicting corresponding VC blood levels of 3.17 mg/L. "Real-time" and "after-action" air modeling estimated VC levels at various distances from the release site over time. PBPK modeling provided insight into possible residential blood levels of VC over several days following the incident. These findings indicate that PBPK modeling could be valuable for reconstructing exposure scenarios associated with acute chemical releases.