Comparison of Four Models for Predicting Blood Lead Levels in Children. J. S. LaKind, S. H. Youngren, and T. B. Piccin, EA Engineering, Science & Technology, Silver Spring, MD; and D. Q. Naiman, Department of Mathematical Sciences, The Johns Hopkins University, Baltimore, MD
According to the Centers for Disease Control and Prevention, there is no threshold blood lead level in children below which no adverse effects occur. Therefore, the traditional risk assessment method of relating dose to an RfD for noncancer effects is not applicable to lead. To assess whether environmental lead concentrations will result in adverse health effects, predicted blood lead levels are compared to a blood lead level of 10 mg/dL, the current CDC level of concern. Children's blood lead levels may be predicted with one of at least four models: USEPA's Integrated Exposure Uptake Biokinetic Model (IEUBK), and models by O'Flaherty (1993), Carlisle and Wade (1992) and Harley and Kneip (1985). This paper utilizes a combination of measured lead media concentrations and default concentrations where data were not available, in conjunction with each of the four models, to predict a distribution of children's blood lead levels. The authors examine the impacts of model default values on the final blood lead level estimations. In addition, the model results are compared to measured blood lead levels, and differences are assessed in terms of model complexity.