Modeling approaches for benchmark dose (R-11834)

According to EFSA's Scientific Committee (SC) the benchmark dose (BMD) approach is a scientifically more advanced method compared to the NOAEL approach when deriving a Reference Point (RP). Model averaging is recommended as the preferred method for calculating the BMD confidence/credible interval. It is recommended to always report the BMD confidence/credible interval rather than the value of the BMD. The lower bound (BMDL) is used to derive the potential RP, and the upper bound (BMDU) is needed for establishing the BMDU/BMDL ratio reflecting the uncertainty in the BMD estimate. In this project the set of candidate models is extended considering several new dose-response models as well as different distributions for continuous endpoints. Model averaging is implemented using a Bayesian paradigm, offering the possibility to use informative prior distributions during model fitting.

Keywords:bayesian modeling, Dose-response analysis, model averaging

Disciplines:Statistics, Bioinformatics data integration and network biology, Development of bioinformatics software, tools and databases