Changes in children's exposure as a function of age and the relevance of age definitions for exposure and health risk assessment.

OBJECTIVE: The objective of this study was to review and synthesize the existing exposure information available to support the characterization and estimation of children's environmental health risks as a function of age.
METHOD: This includes a review of the existing peer-reviewed literature and reports from the US Environmental Protection Agency (EPA) up through January 1, 2003 for information about exposure data for American children with a focus on identifying the age categories used and data gaps that limit our ability to estimate children's risks from exposure to environmental hazards.
RESULTS: On the basis of this synthesis, several key data gaps emerge that suggest some areas in which exposure assessors may want to focus attention, including current breast milk consumption by infants and breastfeeding information for children over age 1 year; children's food-handling practices and how these lead to exposure (eg, by eating with dirty hands or by eating food that has dropped onto a contaminated surface); fish-intake rates for young children and for children whose families include sport fishers or whose families rely on self-caught fish for sustenance; incidental and intentional soil intake by children; soil adherence for dermal exposure; relationships between various microactivities, macroactivities, and microenvironments where children spend time; and a correlation between exposure factors and growth (ie, how children's exposure behaviors change over time). In contrast, relatively good exposure information exists for characterizing children's growth and water ingestion, and at least some exposure information exists for the wide ranges of exposures of regulatory interest.
CONCLUSION: Given the currently available data, exposure assessors can estimate children's potential health risks from a number of different types of exposure, but longitudinal data are needed to reduce the significant uncertainties that arise from reliance on currently available data, and a number of dose-response challenges remain.