A risk assessment for methylmercury in tuna

Abstract

An analysis of the risk of a developmental delay or neurological impairment in children as a result of exposure to MeHg from tuna consumption was conducted. Data used as a basis for the analyses were obtained from surveys of tuna intake, MeHg concentrations in tuna, MeHg levels in the blood of pregnant women, individual relationships between dietary intake and blood levels, individual relationships between blood levels and hair levels, and individual relationships between hair levels and indices of behavioral and neurological performance. Causal or associational relationships, population variability, and uncertainty were modeled for the six data sets. Two predictive models were constructed using the model components. The first was designed to predict the risk of a delay in the onset of talking in a child of a mother who frequently consumes tuna (90 g-day-1). This model uses a one-dimensional Monte-Carlo simulation to assimilate the uncertainties. An average delay of 3-4 days was est imated. The second model was designed to compare the impact of several regulatory options on the risk among the population of tuna-eaters using a scale of central nervous system dysfunction expressly constructed for the assessment. A two-dimensional Monte-Carlo simulation was used to assimilate the distributional components describing population variability and uncertainty.