Professor Shanshan Zhao, Biostatistics and Computational Biology Branch, National Institute of Environmental Health Sciences
Title: Accommodating limit-of-detection in environmental mixture analysis
Abstract: Humans are exposed to a multitude of environmental toxicants daily, and there is a great interest in developing statistical methods for assessing the effects of environmental mixtures on various health outcomes. One difficulty is that multiple chemicals in the mixture can be subject to left-censoring due to varying limits of detection (LOD). Conventional approaches either ignore these measures, dichotomize them at the limits, or replace them with arbitrary values such as LOD/Ö2. Methods have been proposed to handle a single biomarker with limit of detection in such setting, by joint modeling the left-censored biomarker measure with an AFT model and the disease outcome with a generalized linear model. We extend this method to handle multiple correlated biomarkers subject to LOD, through a newly proposed nonparametric estimator of the multivariate survival function and innovative computational approaches. We apply the proposed method to the LIFECODES birth cohort to elucidate the relationship between maternal urinary trace metals and oxidative stress markers.