Survival analysis is the study of time to event outcomes, such as the times from an initiating event (birth, diagnosis, start of treatment) to some “terminal” event (relapse, recovery, death). It is most prominently used in the biomedical sciences, although it has important applications also in the fields of engineering, economics or sociology. Several phenomena are specific to survival data. Most importantly, ‘censoring’ occurs when it is not known when the event of interest takes place, but rather that it has not taken place until a certain time point. This type of data requires special statistical methods, whether the goal is to investigate one type of event, a sequence of events, or more complicated models.
In the Survival Analysis group, we are engaged in research both with departments within the LUMC and with groups outside LUMC. Moreover, we develop new methodology in which we address questions that arise from our clinical collaborations.
For group members and contact information click here.
The methodological research in the field of survival analysis includes various topics, such as:
- Competing risks and multi-state models: these models arise when a patient can have multiple events. These events are mutually exclusive in competing risks models. As an extension, multi-state models describe both sequences of events and competing events.
- Frailty models: these models are used to explain lack of fit of univariate survival analysis models (like deviation from the proportional hazards assumption) or to model dependence of survival times in clustered data.
- Meta-analysis for survival data
- Performance indicators for survival outcomes
- Time-varying covariate effects: these result from violation of the proportional hazards assumption of the Cox model, especially in survival data with long follow-up.
Collaborative and clinical research
The section Medical Statistics of the Department of Medical Statistics and Bioinformatics is involved in a large number of clinical trials conducted by, among others, the Department of Surgery and the Department of Clinical Oncology, such as the Dutch Gastric Cancer Trial, the TME trial, the TEAM trial and the PORTEC trials. We also participate in studies of the European Society for Blood and Marrow Transplantation (EBMT), the Dutch Children Oncology Group (DCOG) and the German Bone Marrow Donor Center (DKMS).
For more information, click here.
The Survival Analysis group teaches statistics courses in the bachelor and master programs of Medicine, Biomedical Science and Clinical Technology within LUMC. We also teach the courses of Survival Analysis and Statistics in the master program Statistical Science for the Life and Behavioural Sciences, and we supervise master theses in this master program. Finally, we teach specialized courses on survival analysis, competing risks and multi-state models and dynamic prediction, throughout the world.
For more details click here.