Monte Carlo modeling of childhood lead exposure: Development of a probabilistic methodology for use with the USEPA IEUBK Model for Lead in Children

An integrated stochastic exposure (ISE) model using microexposure techniques was developed to characterize short-term variability and uncertainty in residential childhood lead exposure at the census tract level. The ISE model was linked to the biokinetic module of the United States Environmental Protection Agency (USEPA) IEUBK Model for Lead in Children (v. 0.99D) to predict the distribution of blood lead concentrations (PbB). We demonstrate an application of the ISE model using probability distributions to characterize variability and uncertainty in age-specific physiology and activity patterns, demographics and housing conditions, concentrations of lead in multiple environmental media, and media-specific uptake rates. Output from the ISE/IEUBK model was compared with output from the IEUBK Model using central tendency point estimates for exposure variables and an assumed geometric standard deviation (GSD). Results of sensitivity analyses suggest that key exposure variables include percent lead in paint, dust lead concentration, and soil and dust ingestion rate; however, the relative importance of each input variable depends on (1) a child's age, (2) the percentile of the PbB distribution, and (3) the time step used to simulate intraindividual variability in parameter values.