[Calendar Icon] Events Calendar

Sarah Ratcliffe, Ph.D.
Director, Division of Biostatistics
Division of Public Health Sciences
University of Virginia

Modeling the Effect of Disease Trajectory on Survival in End-Stage Renal and Liver Disease Patients

Longitudinal biomarkers are commonly measured to follow disease progression. Yet survival predictions are often based on only the last observed measurement. In patients with end-stage renal or liver disease, only “current” values are used to calculate scores that determine one’s place on the transplant waiting list. Even these scores are often completed ignored in predicting post-transplant survival. Further, information on the donor organ is only conveyed at a single time point to potential recipient organizations as part of “matching” and donor longitudinal biomarker values are ignored throughout the process. Finally, depending on when and why a patient dies, informative dropout needs to be incorporated into any post-transplant survival predictions. In this presentation, we will show how joint models have been developed to handle these complex statistical issues, and ongoing research aimed at improving the transplant matching system.