Title: Optimal design of clinical trials when outcome values are missing
Authors: Robin Mitra - Cardiff University (United Kingdom) [presenting]
Stefanie Biedermann - The Open University (United Kingdom)
Abstract: The presence of missing values complicates analyses. When designing experiments, such as clinical trials, missing values are particularly problematic when constructing optimal designs, as it is not known which values are missing at the design stage. When information about the missing data mechanism is available, it is possible to incorporate this information into the optimality criterion that is used to find designs. The areas of missing data and optimal design are briefly reviewed. Some of the specific challenges are then considered when finding optimal designs for the analysis of clinical trials when outcome values could be missing.