@article{bernhardt_wang_zhang_2014, title={Flexible modeling of survival data with covariates subject to detection limits via multiple imputation}, volume={69}, ISSN={["1872-7352"]}, DOI={10.1016/j.csda.2013.07.027}, abstractNote={Models for survival data generally assume that covariates are fully observed. However, in medical studies it is not uncommon for biomarkers to be censored at known detection limits. A computationally-efficient multiple imputation procedure for modeling survival data with covariates subject to detection limits is proposed. This procedure is developed in the context of an accelerated failure time model with a flexible seminonparametric error distribution. The consistency and asymptotic normality of the multiple imputation estimator are established and a consistent variance estimator is provided. An iterative version of the proposed multiple imputation algorithm that approximates the EM algorithm for maximum likelihood is also suggested. Simulation studies demonstrate that the proposed multiple imputation methods work well while alternative methods lead to estimates that are either biased or more variable. The proposed methods are applied to analyze the dataset from a recently-conducted GenIMS study.}, journal={COMPUTATIONAL STATISTICS & DATA ANALYSIS}, author={Bernhardt, Paul W. and Wang, Huixia Judy and Zhang, Daowen}, year={2014}, month={Jan}, pages={81–91} }
 @article{rogers_wilbur_cole_bernhardt_bupp_lennon_langholz_steiner_2011, title={Quantifying Uncertainty in Predictions of Hepatic Clearance}, volume={3}, ISSN={["1946-6315"]}, DOI={10.1198/sbr.2011.09019}, abstractNote={Preclinical predictions of human pharmacokinetic parameters are routinely used in pharmaceutical research and development. In particular, pharmacokinetic predictions are critical in the decision to advance a potential drug to the clinic, to determine appropriate dosing regimens for first-in-human studies, and as a component of translational pharmacology models. Although the associated biological and mathematical models have been extensively discussed in the pharmacokinetic literature, relatively little work has been done to explicitly relate the estimation error of these methods to the underlying experimental variability. This article proposes and evaluates Bayesian models for this purpose. We apply our methodology to a dataset describing both preclinical and clinical pharmacokinetic experimentation for 12 different anonymized drugs. For each drug and for each preclinical mode of prediction, a credible interval is computed and compared against estimates obtained by direct experimentation with human subjects in the clinic. We conclude that many apparent translational differences may be readily explained as a function of experimental error. We view this problem as representative of a larger class of statistical problems in translational medicine, where the mathematics of translation from one species to another requires multiple experimentally estimated scaling factors.}, number={4}, journal={STATISTICS IN BIOPHARMACEUTICAL RESEARCH}, author={Rogers, James A. and Wilbur, Jayson and Cole, Susan and Bernhardt, Paul W. and Bupp, Jaye Lynn and Lennon, Morgan J. and Langholz, Nathan and Steiner, Christopher Paul}, year={2011}, month={Nov}, pages={515–525} }