CFE 2019: Start Registration
View Submission - CMStatistics
Title: Jointly model longitudinal and semicompeting risks data to accommodate informative dropout and death Authors:  Qiuju Li - University College London (United Kingdom) [presenting]
Abstract: In longitudinal studies, both dropout and death can occur during the follow-up and therefore truncate the observation of the longitudinal outcome of interest from a subject. We propose a new likelihood-based approach to accommodating informative dropout and death by jointly modelling the longitudinal outcome and semicompeting event times of dropout and death. Maximum likelihood and Bayesian approaches are used for estimation. Also, since extrapolation beyond death is often not appropriate, it is desirable to obtain the longitudinal outcome profile of a population given being alive. Under the proposed joint modelling framework, the conditional longitudinal profile of being alive can be obtained in a closed form. The proposed methods are illustrated in the application to the HIV Epidemiology Research Study (HERS).