Two-armed response-adaptive clinical trials are modelled as Markov decision problems to pursue two overriding objectives: Firstly, to identify the superior treatment at the end of the trial and, secondly, to keep the number of patients receiving the inferior treatment small. Such clinical trial designs are very important, especially for rare diseases. Thomas Ondra presents the main solution techniques for Markov decision problems and provides a detailed description how to obtain optimal allocation sequences.ContentsIntroduction to Markov Decision Problems and ExamplesFinite and Infinite Horizon Markov Decision ProblemsSolution Algorithms: Backward Induction, Value Iteration and Policy IterationDesigning Response Adaptive Clinical Trials with Markov Decision ProblemsTarget GroupsResearchers and students in the fields of mathematics and statistics Professionals in the pharmaceutical industry
Publication in the field of natural sciences Includes supplementary material: sn.pub/extras
Thomas Ondra
Allocation Sequence Backward Induction Clinical Trial Designs Policy Iteration Value Iteration
“It is a remarkably concise exposition of the use of MDPs in a pharmaceutical setting; as such, the book’s audience is the intersection of researchers and students with a sophisticated mathematical background and professionals in the pharmaceutical industry. … the book is a very sophisticated treatise on MDP models for managing response-adaptive clinical trials and both mathematicians and pharmaceutical professionals would appreciate it.” (James Smith, Interfaces, Vol. 45 (4), July–August, 2015)