The research within our LUMC departments is conducted within departmental research programmes. The research programme below is embedded within the department of Biomedical Data Sciences.
Aim and Focus
The goal of the Medical Statistics research programme is the development and application of statistical models and designs in a broad spectrum of medical research. The main involvements are design, analysis and reporting of experimental as well as observational medical studies aimed at understanding the biological processes and obtaining prognostic models relevant for patient care and therapy. Our research covers a broad spectrum of methodological topics: survival analysis, prediction modelling, statistical genetics, (statistical) bioinformatics, high-dimensional data analysis, family data, epidemiological methods, infectious disease modelling, causal modelling, meta-analysis, patient data encryption, and good research data management. In our statistical bioinformatics research, the main focus is on the analysis of all kinds of -omics data (e.g. genomics, GWAS, transcriptomics, epigenomics, proteomics, metabolomics, and glycomics), as well as on other high-dimensional data (e.g. facial data). We develop and apply methods to integrate -omics data from different sources, statistical models for classification and prognosis, on methods for multiple testing and on computational aspects of statistical models in high-performance computing. In survival analysis the main focus is on dynamic prediction methods and multi-state modelling.
Position in International context
The section of Medical Statistics is one of the leading biostatistical groups in Europe. There is international recognition for the research in the various fields described above. Some key methodological papers from the group are highly cited. Members of the group are frequently invited to other universities, and there are many collaborations in Europe and worldwide. The national and international presence in the fields of statistical -omics data analysis, prediction research, and survival analysis are particularly well established, as also apparent from teaching in international courses.
- Over 150 (co-)authorships per year resulting from collaborations with biomedical research groups
- Several grants have been obtained. Among others: a VIDI by Prof. Goeman, Grants from KIKA and KWF by dr Fiocco, grants from ZonMW and NWO complexity by Prof. Wallinga; EU grants (FP7; IMI) to Prof. Steyerberg.
- The Advanced Data Management group guarantees high quality data acquisition in various large scale projects with external funding.
- ANed/BMS Hans van Houwelingen Award 2016 for best Dutch paper in a refereed journal in the biometrical field in 2014 and 2015, for: Putter H, van Houwelingen HC (2015) “Dynamic frailty models based on compound birth-death processes”, Biostatistics, Vol. 16 Issue 3, pp 550-564
- Journal of Clinical Epidemiology Young Investigator Award 2015 awarded to David van Klaveren for his paper “Biases in Individualized Cost-effectiveness Analysis: Influence of Choices in Modeling Short-Term, Trial-Based, Mortality Risk Reduction and Post-Trial Life Expectancy."
- The book Statistical Analysis of Proteomics, Metabolomics, and Lipidomics Data Using Mass Spectrometry co-edited by Bart Mertens
- Development of novel R packages frailtyEM, hommel, confSAM. The R-packages developed in recent years, mstate, globaltest, penalized, cherry and dynpred, have a large user base and are maintained by our group
- A number of papers/books arisen from this research programme in the last 10 years have become classic papers in the field, having attracted hundreds of citations such as 1) Analyzing gene expression data in terms of gene sets (Bioinf 2007); 2) L1 penalized estimation in the Cox proportional hazards model (Biom J 2010); 3) Competing risks in epidemiology: possibilities and pitfalls (Int J Epid 2012); 4) Dynamic Prediction in Clinical Survival Analysis (CRC Press 2012)
Size and complexity of medical data sets will continue to increase, and new data types will arise. This will add extra complexity to statistical modelling and constantly ask for new methods development. Successful research lines such as the analysis of –omics data and survival analysis will remain relevant and are therefore to be continued. A new research line will be mathematical modelling of infectious diseases, stimulated by the appointment of Prof. Wallinga. Other future research topics will be methodology for multiple testing in large genomics and imaging problems, the combination of several sources of information within and between studies, learning from Big Data, the analysis of microbiome data, statistical methods for quality indicators, methods of incorporating heterogeneity in survival analysis and in family data, prediction modelling, and causal inference in observational data. Our statistical research has always been instigated by statistical problems arising in (LUMC) medical research areas. We will continue to cater to these developments.
Cohesion within LUMC
Traditionally there is a close cooperation with nearly all clinical groups, as reflected in the large number of co-authorships. Because of these intensive collaborations our programme is part of all 7 LUMC research profiles. Prof. le Cessie has a joint appointment at the Dept of Clinical Epidemiology. The cooperation with the molecular (genetic) oriented departments in the field of bioinformatics and statistical genetics is presently well established and still growing, extending from the Molecular Epidemiology programme led by Prof. Slagboom. With the coming of Prof. Steyerberg the collaboration with the Medical Decision Making group will be intensified. Our consultation activities are embedded within the Research Support Desk, allowing even easier access for LUMC researchers to statistical, bioinformatics, and data management advice. We furthermore collaborate in particular with other statistics groups at Mathematics and Psychology. Contacts in this area have intensified because of our joint master’s programme and several joint appointments of faculty members.