Principal investigator: Enrique F. Schisterman, Ph.D.
Evaluation of human health risks, including sensitive endpoints, are difficult and rely on the ability of investigators to detect and quantify compounds in biospecimens. Major methodological challenges underscore the interpretation of data and, ultimately, the meaning of health risks. These challenges have been largely overlooked but remain critical data gaps with relevancy for many disciplines: 

• The large number of observed values below the laboratory LOD especially in young children or couples at risk for pregnancy or study populations needed for assessing reproductive and developmental toxicants;
• Use of different imputation methods for non-detectable values (e.g., imputing zero for values below the LOD, etc.) and the bias such practices introduce;
• The need for causal modeling delineating primary exposures from effect modifiers (e.g., concomitant exposures) from confounders when estimating human health risks; and
• The effect associated with random measurement error (i.e., error associated with the measurement of exposure or health outcome) and biases attributed to laboratory equipment, variation between technicians, temporal changes, and biological variability.

To address this issue, researchers propose methodological research on exposure assessment, especially in the LOD issue (to substitute or not), how best to model serum lipids for non-fasting samples, and how best to model exposure - outcome in the context of known "biologic" determinants of exposure and disease.
 
DESPR Collaborators

· Germaine M. Buck Louis, Ph.D.
· Ondine von Ehrenstein, Ph.D., M.P.H., MSc.


Selected Publications

Bloom MS, Buck Louis GM, Schisterman EF, Liu A, & Kostyniak PJ. (2007). Maternal serum polychlorinated biphenyl concentrations across critical windows of human development. Environmental Health Perspectives, 115(9):1320-1324. [Abstract]

Perkins NJ, Schisterman EF, & Vexler A. (2007). Receiver operating characteristic curve inference from a sample with a limit of detection. American Journal of Epidemiology, 165:325-333. [Abstract]

Vexler A, Liu A, & Schisterman EF. (2006). Efficient design and analysis of biospecimens with measurements subject to detection limit. Biometrical Journal, 48:780-791.  [Abstract]

Mumford SL, Schisterman EF, Vexler A, & Liu A. (2006). Pooling biospecimens and limits of detection: effects on ROC curve analysis. Biostatistics, 7:585-598. [Abstract]

Schisterman EF, Vexler A, Whitcomb BW, & Liu A. (2006). The limitations due to exposure detection limits for regression models. American Journal of Epidemiology, 163:374-383. [Abstract]

Whitcomb BW, Schisterman EF, Buck GM, Weiner JM, Greizerstein H, & Kostyniak PJ. (2005). Relative concentrations of organochlorine pesticides and polychlorinated biphenyls in adipose tissue and serum of women of reproductive age. Environmental Toxicology and Pharmacology, 19:203-213.

Schisterman EF, Whitcomb BW, Louis GMB, & Louis TA. (2005). Lipid adjustment in the analysis of environmental contaminants and human health risks. Environmental Health Perspectives, 113(7):853-857. [Abstract]