Combining datasets to predict the effects of regulation of environmental lead exposure in housing stock.

Abstract

A model for children's blood lead concentrations as a function of environmental lead exposures was developed by combining two nationally representative sources of data that characterize the marginal distributions of blood lead and environmental lead with a third regional dataset that contains joint measures of blood lead and environmental lead. The complicating factor addressed in this article was the fact that methods for assessing environmental lead were different in the national and regional datasets. Relying on an assumption of transportability (that although the marginal distributions of blood lead and environmental lead may be different between the regional dataset and the nation as a whole, the joint relationship between blood lead and environmental lead is the same), the model makes use of a latent variable approach to estimate the joint distribution of blood lead and environmental lead nationwide.

4 Figures and Tables

Cite this paper

@article{Strauss2001CombiningDT, title={Combining datasets to predict the effects of regulation of environmental lead exposure in housing stock.}, author={Warren J. Strauss and Raymond J. Carroll and Steven M. Bortnick and John Menkedick and Bradley D. Schultz}, journal={Biometrics}, year={2001}, volume={57 1}, pages={203-10} }