Bayesian model-averaged benchmark dose analysis via reparameterized quantal-response models

From MaRDI portal
Publication:2809552




Abstract: An important objective in biomedical risk assessment is estimation of minimum exposure levels that induce a pre-specified adverse response in a target population. The exposure/dose points in such settings are known as Benchmark Doses (BMDs). Recently, parametric Bayesian estimation for finding BMDs has become popular. A large variety of candidate dose-response models is available for applying these methods, however, leading to questions of model adequacy and uncertainty. Here we enhance the Bayesian estimation technique for BMD analysis by applying Bayesian model averaging to produce point estimates and (lower) credible bounds. We include reparameterizations of traditional dose-response models that allow for more-focused use of elicited prior information when building the Bayesian hierarchy. Performance of the method is evaluated via a short simulation study. An example from carcinogenicity testing illustrates the calculations.





Describes a project that uses

Uses Software





This page was built for publication: Bayesian model-averaged benchmark dose analysis via reparameterized quantal-response models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2809552)