Biostatistics Weekly Seminar

Dynamic assessment of the risk of adverse events is essential to informing treatment decisions, with the goal of optimizing patient outcomes and/or avoiding over-treatment. In liver transplantation, to ensure equitable allocation of available deceased-donor organs for end-stage liver disease patients, the risk of death is regularly assessed based on the change in biomarkers or vital signs to evaluate a patient’s medical status and urgency of treatment receipt. In order to broaden the scope of jointly analyzing the longitudinal covariate process and the time-to-event outcome, we propose a landmark method which directly models the restricted mean survival time (RMST) while accommodating dependent censoring. Advantages of RMST models include removing the assumption that the hazard model (e.g., in a Cox regression) is correct at every time point, since RMST models consider a single restriction time as opposed to a process. Moreover, many investigators prefer the area under the survival curve over hazard rate as a clinical endpoint due to interpretability. We derive the asymptotic properties of the proposed estimators, and a simulation study demonstrates their decent performance in finite samples. We also explore model extensions tailored to address diverse scientific purposes, and the proposed methods are illustrated using national registry data on a cohort of patients wait-listed for a liver transplant.

Virtual: Zoom Link to Follow
12 January 2024
1:30pm