An Event-by-Event Probabilistic Methodology for Assessing the Health Risks of Persistent Chemicals in Fish: A Case Study at the Palos Verdes Shelf

Abstract

A human health risk assessment of recreational anglers who consume fish from the Palos Verdes Shelf was conducted. The uptake of DDT, DDE, and DDD (collectively total DDT or tDDT) and polychlorinated biphenyls (PCBs) due to fish ingestion was characterized using Monte Carlo techniques. This analysis relied upon 176 probability density functions developed from over 300,000 individual pieces of information to represent 17 different exposure factors that influence the human uptake of persistent organic chemicals in fish. The carcinogenic and noncarcinogenic risks were estimated using a microexposure event modeling approach that estimates exposure on an event-by-event basis. This evaluation relied upon several large studies that provided site-specific data on angler behavior and concentratioins of chemicals in 13 fish species. Our results indicate that the median theoretical increased lifetime cancer risk associated with estimated exposure to tDDT and PCBs was 5 x 10(-8) for anglers who fish on commercial passenger fishing vessels (CPFVs) and who catch and eat fish from the Palos Verdes Shelf. The mean risk for these anglers was 2 x 10(-7), and the 95th percentile risk was 8 x 10(-7). At the 9.5th percentile, the hazard quotients for anglers were less than 1, indicating that noncancer effects are unlikely. These results are in contrast with prior risk assessments of this site that suggested that consumption of white croaker alone posed a cancer risk of 2 x 10(-3) and a hazard quotient of 32. Our results were validated by their agreement with several independent local studies regarding fishing and consumption practices. This assessment indicates that the levels of tDDT and PCB in fish at the Palos Verdes Shelf do not pose a significant risk to human health among recreational anglers. Based on the size of the local angler population, no cases of cancer would be expected to result from eating Palos Verdes Shelf fish. The methodology used here should be applicable to characterizing the risks to those who ingest fish from the waterways of most industrialized nations.

Noncommercial Fish Consumption and Anglers at Risk

“…The effective response rate for the on-site survey was 97% (n = 946). To correct for participation bias, consumption rates were weighted by the inverse of fishing frequency (Wilson et al, 2001). A more detailed discussion of the weighting measure is found elsewhere (Sheaffer, Beaman, O'Leary, Williams, & Mason, 2000).…”

“…Once these variables have been measured, they are entered into an equation to calculate consumption rate per angler (Meredith & Malvestuto, 1996). Rates have been reported for different recall periods, ranging from a week to a year (Hanrahan et al, 1999;Wilson, Price, & Paustenback, 2001). Rates measured for a shorter recall period (e.g., three months) are often used to extrapolate to a yearly consumption rate (Cavan & Gibson, 1996).…”

Site‐Specific Applications of Probabilistic Health Risk Assessment: Review of the Literature Since 2000

“…D. Wilson and Associates; AMEC Earth and Environmental; ChemRisk Wilson et al (2001) year-to-year variation in an individual's exposure variability (U.S. EPA, 2005b). Probabilistic risk assessment has also been implemented at other contaminated sites in the United States.…”

“…Nevertheless, in most cases, PRA applications in regional risk assessment in the United States are methodologically similar to site-specific assessments, but often incorporate more parameters and more complex models. Wilson et al (2001) conducted coupled probabilistic risk assessment and micro-exposure modeling to assess human health risks originating from consumption of fish contaminated with DDT (dichloro-diphenyl-trichloroethane) and PCBs at the Palos Verdes Shelf in California. Hope (2003) developed a probabilistic food-chain model to assess human health risks from consumption of mercury-contaminated fish from the Wilamette River basin in Oregon and to develop total maximum daily loads.…”

Immune Systems, Geographic Information Systems (Gis), Environment and Health Impacts

“…The availability of multiple biomarkers of exposure can help in determination of cumulative risk through multiple ecotoxicants (Galloway et al, 2002;Chevre et al, 2003). Most models used in risk assessment are based on determination of the probability of exposure to ecotoxicants in a certain population (Adams et al, 1994;Burger, 1995;Paustenbach, 2000;Wilson et al, 2001;Williams & Paustenbach, 2002). Biomarkers of exposure generally consist of metabolites or other products that result from the interaction between environmental toxicants and receptor molecules in target cells, and that can be measured in some compartment within a living organism (Needham et al, 1995;Diehl-Jones & Bols, 2000;Bouchard et al, 2002).…”