An exposure assessment for methylmercury from seafood for consumers in the United States.
Level 5 - mechanism / opinion, no new human data
Exposure assessment and pharmacokinetic simulation modeling using secondary survey datasets (by design analogy)
PubMed 12224743 · doi:10.1111/0272-4332.00061
What was done
An exposure model was constructed to estimate methylmercury intake from seafood and predict blood and hair mercury levels in the overall U.S. population, women aged 18–45, and children aged 2–5. Short-term (three-day) seafood consumption survey data were adjusted with long-term purchase diaries to simulate 360-day daily intake distributions (10,000 simulated individuals, 1,000 uncertainty iterations) combined with FDA fish contamination and market-share data. A second simulation (2,000 simulated individuals, 1,000 iterations), parameterized by a 90-day controlled intake study and other published data, converted intake into blood and hair mercury concentrations. Model outputs were evaluated against biomarker data from the National Health and Nutrition Examination Survey (NHANES).
What was found
Predicted blood and hair mercury concentrations for both children and adult women were within a factor of two or less of NHANES biomarker measurements. The simulation systematically underpredicted blood mercury levels for adult women and overpredicted both blood and hair mercury levels for children. No specific numerical biomarker values or intake estimates were reported in the abstract.
Why it matters
This framework demonstrates that population-level dietary surveys combined with pharmacokinetic modeling can reasonably approximate blood and hair methylmercury distributions. It helps evaluate how dietary seafood guidance and contamination variations translate into biological exposure across sensitive demographic groups.
Limits
The study is a mathematical simulation rather than direct human measurement. Short-term 3-day consumption data required statistical adjustments to estimate chronic intake, mercury distributions for non-target species were estimated from market share and assumed distribution shapes, and the model showed systematic bias across subpopulations. Absolute numerical values, confidence intervals, and primary dataset sample sizes were not provided in the abstract.
Cited by
- supports Limiting tuna consumption helps reduce elevated blood mercury levels.