Characterization of Risks to Natives from Consumption of Methylmercury-Contaminated Fish. Ryan A. Hill, Sara M. Hoover, and Tom A. Watson, Triton Environmental Consultants Ltd., #120-13511 Commerce Parkway, Richmond, BC V6V 2L1, Canada
We applied probabilistic risk assessment techniques to analyze the effects on human health due to consumption of methylmercury-contaminated fish from a reservoir and two natural lakes in British Columbia. The target population was children of exposed native women because developmental effects are believed to be the most sensitive endpoint, and natives are believed to consume greater amounts of contaminated fish than most non-natives. We used probability distributions in a Monte Carlo framework to represent exposure variables. We developed an index of neurological effects from the Iraqi methylmercury study using principal components analysis and performed a dose-response assessment, using both the logistic and Weibull models to describe the data. We combined the dose-response analyses with the probabilistic exposure assessment to estimate risks. The selection of the dose-response model strongly influenced both mean risk estimates and distributions of risk. Our results indicate that native women are not placing their infants at significant risk by consuming moderate quantities of fish from two natural lakes and a reservoir in British Columbia.