Advanced statistical analysis and modeling
The Medical Statistics group is involved in a number of areas of advanced statistical analysis and modelling. This includes causal inference and modeling (Prof Saskia le Cessie); advanced statistical modelling, including Bayesian analyses (Dr Erik van Zwet), Prediction modelling (Prof Ewout Steyerberg, Dr Ben Van Calster), and advanced data management (ADM, Dr Karin van der Pal).
One area of expertise is in the development, validation, and updating of clinical prediction models. Methods and guidance is needed to improve model development and validation, updating strategies, impact assessment and implementation.
Consultation
The section Medical Statistics is involved in a large number of studies that are primarily led by other LUMC departments. Staff members of Medical Statistics are open for collaboration or advise on a wide range of statistical methods.
Key Publications
Assessing the performance of prediction models: a framework for some traditional and novel measures
Extensions of net reclassification improvement calculations to measure usefulness of new biomarkers
Prognosis Research Strategy (PROGRESS) 3: Prognostic Model Research
Prediction models need appropriate internal, internal–external, and external validation
Calibration: the Achilles heel of predictive analytics
Our Team members
- Causal models (Prof Saskia le Cessie)
- Advanced statistical modeling (Dr Erik van Zwet)
- Prediction modelling (Prof Ewout Steyerberg, Dr Ben Van Calster)
- Data management (Dr Karin van der Pal)