Combining environmental information via hierarchical modeling: an example using mutagenic potencies

Abstract

SUMMARYMutagenesis assays of toxic environmental agents often provide only a single snapshot of the agent's potential risk, leading to inferences on a possibly isolated phenomenon. Hierarchical models can enhance inferences on environmental agents and obtain more global perspectives on the potential risk. We investigate the use of such models to combine information across environmental studies. We emphasize combination of potency estimates from bioassays used for environmental hazard identification and risk assessment. The models and methods illustrate hierarchical Bayesian techniques to estimate the parameters and make inferences on the unknown potency values.

Benchmark Dose Estimation Incorporating Multiple Data Sources

“…Piegorsch and Cox (5) reviewed generalized linear models with random effects in the estimation of the effective dose in ordinal and continuous regression. Simmons et al (6) looked at modeling mutagenic potencies from different Salmonella strains, and Coull et al (7) conducted a meta-analysis for methyl-mercury using a Bayesian hierarchical model to combine BMD point estimates from multiple studies. While these studies have focused on risk assessment, most have looked at continuous response regression, with less emphasis on nonnormal data.…”