[STAT/AP] Eni Musta: Statistical methods for survival data with a cure fraction

03 juni 2024 15:45 t/m 16:45 - Locatie: Lecture Hall G | Zet in mijn agenda

In survival analysis we are interested in the time until occurrence of a specific event, but often some subjects are immune (cured) and would never experience such an event. For example, thanks to medical advances, many cancer patients get cured and do not experience cancer relapse/death. Time-to-event data with an immune proportion of the population are encountered in other fields as well, e.g. fertility studies, credit scoring, social studies of life events and behaviours etc. Ignoring the cure possibility and analysing such data with standard methods can lead to invalid results and a waste of information. To address this, cure rate models have been developed as an alternative modelling approach for survival analysis that accounts for the presence of immune subjects. What makes the problem statistically challenging is the unobserved cure status for subjects who, because of censoring, have only been followed for a limited period of time. In this talk, I will provide an introduction of cure models and related statistical methodology/theory, focusing on my recent work. Finally, I will touch upon some open problems and challenges in this area.