[生物书合集].Advances.in.Clinical.TBiostatisti

[生物书合集].Advances.in.Clinical.TBiostatisti

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
An alternative formulation describes the dose-toxicity relationship as it applies to the set S ={ x 1 , x 2 , ... , x k } of prespecified dose levels available for use in the trial. For example, Gasparini and Eisele (2000) present a curve-free Bayesian phase I design discussed the possibility of using one prior distribution (the design prior) to determine dose assignments during the phase I trial and a separate prior (the inference prior) to estimate the MTD upon trial completion. Although the use of separate priors for design and inference may appear inconsistent, its usefulness is defended by arguing that analysis occurs later than design (Tsutakawa, 1972). Con- sequently, our beliefs regarding the unknown parameters of the dose- toxicity model may change during the time from design to inference in ways not entirely accounted for by a sequential application of Bayes’ theorem. Since estimation of the MTD is the primary statistical aim of a phase I clinical trial, our subsequent attention will be focused on dose-
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.
Ask a homework question - tutors are online