Dealing with Missing Data in Vaccine Clinical Trials: from Academics to the Industry
View Presentation View Presentation
*Niel Hens, I-BioStat, Hasselt University 

Keywords: Incomplete data, non-ignorable missingness, sensitivity analyses

In vaccine studies, it is very likely for patients to drop out of the study due to a variety of reasons resulting in monotone and non-monotone missingness processes. Methods to deal with these incomplete longitudinal data have been well developed in the last few decades. The basic question underpinning these methods is whether the mechanism generating the missingness is ignorable or not. In case the missingness mechanism is ignorable, the available data can be analyzed as they are, ignoring the missingness process. In case the missingness mechanism is non-ignorable different modelling approaches, relying on different untestable assumptions should be envisaged to analyze the data. In this talk, a lot of attention will be devoted to model structure and model selection uncertainty from an applied perspective. The focus is on bridging the gap between academics and industry.