STATISTICAL METHODOLOGIES FOR EXPOSURE ASSESSMENT

University Of Maryland Baltimore County, Baltimore MD

Investigators

Linked publications & trials

Abstract

The proposed research is on the development of statistical procedures that are appropriate for analyzing exposure data. The statistical models considered are all directly relevant for exposure assessment. The proposed problems are on the following topics: the univariate, bivariate and multivariate lognorma! distributions to describe exposure data, methodology for comparing several test methods or samplers, procedures to analyze samples that include low concentrations of chemicals (in particular, values below the detection limit), regression models for modeling the relationship among exposure variables, statistical techniques for the evaluation of low cost, easy to use and non-invasive test methods, statistical methodology for biological monitoring, in particular, for estimating the concentrations of chemicals and biomarkers in the blood and internal organs based on non-invasive measurements. Difficulties and limitations of some of the currently used statistical techniques will be highlighted, and efficient alternatives will be developed. Novel approaches based on new concepts such as generalized p-values and generalized confidence intervals will be pursued for solving some of these problems. The associated computational details will be addressed in detail, and software codes will be provided.