179 – Advances in Longitudinal Data Analysis
Using Planned Missing Values in Longitudinal Trials to Relieve Patient Burden and Reduce Costs
Christele Augard
Sanofi Pasteur
Ayca Ozol-Godfrey
Sanofi Pasteur
Robert Small
Sanofi Pasteur
Dominika Wisniewska
Sanofi Pasteur
Some longitudinal trials require subjects to submit to frequent blood draws on visits over time. Often the primary endpoint does not require the observations from every visit. This is true, for example, in vaccine immunogenicity trials and in diabetes trials. Taking samples at every visit can be burdensome to both the subject and the sponsor. Subjects often do not like many blood draws. The cost of assays of every sample can be high. These facts contribute to increased cost and increased subject drop out. In this paper we investigate the idea of bleeding random subsamples at each visit but using the (frequent) high correlation within subjects between visits to build imputation models to implement an MI approach to analyzing the data. We use the observations present as well as other pertinent continuous and categorical variables to build the models. We do the estimation of the imputation models using a method of Raghunathan, Lepkowski et.al. which is very general and can handle many types variables. We give examples using data from some recent vaccine trials. We show how various patterns can reduce cost and possibly drop outs.
