As of February 15, Ewout Steyerberg has been appointed Professor of Clinical Biostatistics at the Julius Center of the UMC Utrecht/Utrecht University. Within the Department of Data Science and Biostatistics, he focuses on developing and applying predictive models that help doctors and patients make better medical decisions.
Biostatistics plays an important role in medical research. “Ultimately, it is about deriving reliable answers to important medical questions from data,” says Steyerberg. “It always starts with the question: what exactly do we want to know? Only after that do you look at the data and the methods needed to answer that question.”
How is clinical biostatistics evolving?
The field is evolving rapidly. Classical statistical methods are increasingly being combined with techniques from data science, such as machine learning and artificial intelligence (AI). “Statistics has a long tradition of critically examining data,” he explains. “We consider assumptions, data quality, and uncertainty in the results. In data science, the emphasis is sometimes more on what is technically possible. These two worlds can reinforce each other.”
Due to the increase in available medical data and computing power, analyses are now possible that were unthinkable just a few decades ago. At the same time, caution remains necessary. “More data does not automatically mean better answers. It remains essential to critically assess how reliable a prediction is.”
What are predictive models and how do they work?
The development of predictive models is central to Steyerberg’s research. With these, researchers aim to estimate, for example, the probability that someone will develop a specific disease or how someone will respond to a treatment. These models can help doctors better tailor treatments to individual patients.
“If you can predict the risk of a certain outcome, you can adjust the treatment accordingly. You may be able to spare low-risk patients’ intensive treatment, while treating higher-risk patients more specifically.”
An example is research into dementia. It turns out that simple characteristics such as age and gender already provide a great deal of information about risk. “We can measure a great deal these days, but sometimes a few basic factors prove to be surprisingly informative.”
How reliable are medical predictions?
Although new analytical techniques offer many possibilities, he emphasizes that medical predictions always remain uncertain. “For example, if you say that someone has a ten percent chance of developing a certain disease, that remains an estimate. With a different model, that probability might be five or thirty percent. That is why it is important to be transparent about that uncertainty.” This also applies to more ambitious applications of AI, such as predicting a large number of different diseases simultaneously. “For many conditions, there is still simply too little data to make reliable predictions.”
How are predictive models used in healthcare?
In addition to methodological development, his research explicitly focuses on application in healthcare. Within various research projects, predictive models are being developed to support doctors and patients in making decisions during the treatment process.
In cancer research, for example, an increasing number of patient characteristics are being taken into account in treatment decisions. “By combining data on tumor characteristics, biomarkers, and other factors, we can increasingly better estimate which treatment is most suitable for a patient.”
“Ultimately, it is about deriving reliable answers to important medical questions from data.”
What role does data science play at UMC Utrecht?
The Chair in Clinical Biostatistics aligns closely with the strategy of UMC Utrecht, in which data science and e-health play a key role. Within the Connecting Worlds program, data science is seen as a major driver of research and care.
The chair contributes to various strategic research programs at UMC Utrecht, including the themes of Cancer and Circulatory Health. In addition, digitalization plays an increasingly important role in healthcare. Within the Care for Tomorrow program, the guiding principle is ‘Digital, unless’, in which data-driven applications occupy a prominent place. “Predictive algorithms and AI are being used increasingly often in healthcare,” says Steyerberg. “It is important that we carefully develop and evaluate these applications so that they genuinely contribute to better care.”
How are future doctors and researchers trained in data science?
In addition to research, the chair also makes a significant contribution to education. Students at the bachelor’s, master’s, and PhD levels learn how to analyze and interpret medical data. At a time when AI is becoming increasingly important, it is essential that doctors and researchers continue to think critically about data and models. This applies to both research and the application of algorithms in clinical practice.
Who is Ewout Steyerberg?
Ewout Steyerberg (1967) studied Biomedical Sciences at Leiden University, where he graduated cum laude in 1991. During his studies, he worked as a student assistant in the Department of Medical Statistics, where his interest in biostatistics began. He obtained his PhD from Erasmus University Rotterdam in 1996.
In 2006, he was appointed Professor of Medical Decision Making at Erasmus MC in Rotterdam, with a focus on prognostic modelling and predictive models. He later became Head of the Department of Biomedical Data Sciences at Leiden University Medical Center (LUMC) and Professor of Clinical Biostatistics and Medical Decision Making. In June 2024, he joined UMC Utrecht as Medical Scientific Division Manager at the Julius Center for Health Sciences and Primary Care. As of February 15, 2026, he has been appointed Professor of Clinical Biostatistics. His research focuses on the development and application of predictive models in medicine and on the evaluation of predictive algorithms in clinical practice.
He has published more than 1,000 scientific articles and collaborates with researchers from various medical disciplines, including oncology, cardiovascular medicine, and neurology.
Ewout lives in Gouda with his wife and three children. In his spare time, he enjoys running, often around the lakes near Gouda.
Awards and recognition
For his work, Steyerberg has received various national and international awards, including the Society for Medical Decision Making Student Prize (1995), a TALENT scholarship from NWO for research at Duke University (1996), a KNAW-fellowship (1999–2003), a Marx Research Fellowship from the Dana-Farber/Harvard Cancer Center (2005), and the John M. Eisenberg Award from the Society for Medical Decision Making (2016).
Why biostatistics?
Steyerberg: “For me, biostatistics is about understanding data. It is a kind of puzzle: can we derive an answer to an important medical question from the available data? Nowadays, we have more data at our disposal. At the same time, it is important to remain critical of the quality of that data and the uncertainty in predictions.”
