Quantitative meta-analytic approaches for the analysis of animal toxicology and epidemiologic data in human health risk assessments
Often, human health risk assessments have relied on qualitative approaches for hazard identification to integrate evidence across multiple studies to conclude whether particular hazards exist. However, quantitative approaches for evidence integration, including the application of meta-analyses to animal and human data have begun to gain traction in human health risk assessments to help summarize the results of systematic reviews. For hazard identification, meta-analysis calculates a pooled effect size, after weighting studies (most commonly by study size) and consideration of multiple potential confounding factors, in order to determine whether the entire database under consideration indicates a chemical is a hazard. Meta-analytical tools have also been developed to assist in dose-response analyses. The U.S. EPA’s Categorical Regression (CatReg) software is one such tool that facilitates meta-analysis through the combination of ordinal dose-response information from different species, strains, sexes, etc. Hypothesis testing features of the software allow users to determine the correct structure of their modeling analysis so that data is combined only when statistically appropriate. For the dose-response of human data, Bayesian methods have been developed that can combine data from cohort and case-control studies in order to calculate pooled logistic slope values. Using these pooled logistic slopes in a lifetable analysis allows risk assessors to calculate lifetime risks for the outcomes of interest. To demonstrate these quantitative methods, case studies will be presented on data from Integrated Risk Information System assessments: TMB (trimethylbenzene) neurotoxicity, asbestos-caused pleural plaques effect on lung function, phosgene-induced lung lesions,and inorganic arsenic-induced bladder cancer.
Disclaimer: The views expressed in this abstract are those of the authors and do not necessarily reflect the views or policies of the U.S. EPA.