Biostatistics; Data Interpretation, Statistical; Models, Statistical; Regression Analysis; Risk Assessment
- Prognosis research
Research Area: Prognosis research in medicine aims at understanding and improving future outcomes of individuals. We work on aspects of developing, translating and applying statistical methodology in prognosis research. Co-leaders: Daniela Dunkler and Georg Heinze
The general focus of my research is statistical modeling aiming at etiology and prediction.
A statistical model tries to mimic or simplify the reality. It is used to predict or explain an outcome of interest (e.g. the occurance of cardiovascular disease within 5 years) using several predictors or risk factors (e.g. cholesterol level, smoking status, ...).
Techniques, methods & infrastructure
My methodological research concentrates on variable selection methods used for selection of predictors or risk factors in a statistical model. I investigate the advantages and disadvantages of applying data-driven variable selection methods. Whenever applying such methods, the robustness of a model - also called model stability - must be verified as small changes in the data could lead to a different set of selected variables. This topic is a part of my methodological research that I studied in my PhD thesis.
My applied research is mainly in the fields of cardiology and nephrology. One of the projects targets prediction models for the occurance of cardiovascular disease within 5 years. Cardiovascular disease are one of the main causes of death and disability. Therefore, cardiovascular risk is assessed in the Austrian preventive health screening program. In our project, we evaluated existing risk prediction models and updated them to improve their predictive performance in the Austrian population. In another project, we investigate the benefit on survival of undergoing a kidney transplantation instead of remaining on dialysis. In particular, we are interested if this treatment strategy is still beneficial for elderly recipients who are waitlisted for a transplantation over serveral years.
- Wallisch, C. et al. (2021) ‘The roles of predictors in cardiovascular risk models - a question of modeling culture?’, BMC Medical Research Methodology, 21(1). doi:10.1186/s12874-021-01487-4.
- Wallisch, C. et al. (2020) ‘Selection of variables for multivariable models: Opportunities and limitations in quantifying model stability by resampling’, Statistics in Medicine, 40(2), pp. 369–381. doi:10.1002/sim.8779.
- Heinze, G., Wallisch, C. and Dunkler, D. (2018) ‘Variable selection - A review and recommendations for the practicing statistician’, Biometrical Journal, 60(3), pp. 431–449. doi:10.1002/bimj.201700067.
- Wallisch, C. et al. (2022) ‘Review of guidance papers on regression modeling in statistical series of medical journals’, PLOS ONE. Edited by T. Mathes, 17(1), p. e0262918. doi:10.1371/journal.pone.0262918.
- Wallisch, C. et al. (2021) ‘Development and internal validation of an algorithm to predict intraoperative risk of inadvertent hypothermia based on preoperative data’, Scientific Reports, 11(1). doi:10.1038/s41598-021-01743-z.