A0527
Title: Studying mortality in critically ill patients: An analysis of ordinal longitudinal data under case-control sampling
Authors: Ran Tao - Vanderbilt University Medical Center (United States)
Jonathan Schildcrout - Vanderbilt University (United States)
Chiara Di Gravio - Imperial College London (United Kingdom) [presenting]
Abstract: It is common practice in clinical trials to store blood samples at recruitment and analyze them at a later stage to retrospectively obtain information on new exposures. However, high costs limit the number of samples researchers can analyze. The purpose is to discuss the experience of setting up a secondary analysis of the CLOVERS trial where, due to budget and time constraints, stored blood samples could only be analyzed on a third of the original trial population. First, the study design is introduced, and then a semiparametric likelihood approach is described to estimate the parameters of interest. The estimation approach is flexible in that it can be used for any design and exposure of interest. The finite sampling operating characteristics of the proposed approach and the results from the CLOVERS trial are finally presented.
