Stijn Vansteelandt



Stijn Vansteelandt

Ghent University

Department of Applied Mathematics, Computer Science and Statistics

Krijgslaan 281, S9 (2nd floor)

9000 Gent, Belgium

Stijn dot Vansteelandt at UGent dot be

How to reach me?

I am Professor of Statistics (80%) in the Department of Applied Mathematics, Computer Science and Statistics at Ghent University (Belgium) and Professor of Statistical Methodology (20%) in the Department of Medical Statistics at the London School of Hygiene and Tropical Medicine (U.K.).

Curriculum Vitae (d.d. March 15, 2017)

Find me on Google Scholar or LinkedIn.


I teach introductory courses in statistics to students in Mathematics and in Pharmacy, as well as advanced courses on Analysis of Clustered and Longitudinal Data, and on Causality and Missing Data in the advanced Master in Statistical Data Analysis.


My research primarily concerns the development of statistical methods for inferring causal effects from observational data. For instance, I develop statistical methods for inferring the attributable mortality due to hospital-acquired infections; key challenges in this context are

- that patients who acquire infection are typically in worse health conditions and thus not comparable with those who do not acquire infection;

  1. -that the survival times are typically unknown for patients who leaving hospital.

In addition, I develop methods for mediation analysis (i.e., inferring direct from indirect exposure effects), for testing statistical or sufficient cause interaction between 2 exposures (e.g., gene-gene or gene-environment interaction), for instrumental variables analysis and for causal inference in outcome-dependent sampling designs.

Besides the previous projects on causal inference, I develop methods for the analysis of incomplete data sets.

My recent research is primarily aimed at making causal inferences less vulnerable to the weaknesses (imprecision, finite-sample bias and susceptibility to model misspecification) of simple inverse probability weighted estimators that dominate causal inference research. I aim to realise this either by improving inverse probability weighted estimation (see my work on bias-reduced double-robust estimation) or by popularising and extending alternative estimation methods, such as g-estimation, that avoid inverse weighting.

A characteristic feature of my research is the use of semi-parametric models.

Editorial boards

I am currently Co-Editor of Biometrics, the flagship journal of the International Biometric Society.

If you wish to send correspondence regarding Biometrics, please do not send it directly to me! Instead, send it to Ms. Ann Hanhart, the Biometrics Editorial Manager, at

Associate Editor of Biometrics (2006-2012)

Associate Editor of Biostatistics (2010-2015),

Associate Editor of Epidemiologic Methods (2011-2015)

Associate Editor of the Journal of Causal Inference (2011-2015)

Associate Editor of Epidemiology (2013-2015)

PhD students and postdocs

I have the pleasure of working with Sjouke Vandenberghe on mediation analysis for dichotomous and survival outcomes, with Haeike Josephy on mediation analysis for cross-over designs, with Oliver Dukes on adjustment for time-varying confounding, and with Vahe Avagyan on covariate and model selection in causal inference.

Previous PhD students

Sylvie Goetgeluk (2008). From data clusters to causal inference: new methodology for the analysis of twin registers.

Guy Baele (2008). Detecting complex substitution patterns in non-coding sequences.

Manoochehr Babanezhad (2009). Measurement error and causal inference with instrumental variables.

Wim Delva (2009). Sexual behavior and the spread of HIV: statistical and epidemiological modeling applications.

Maarten Bekaert (2011). Do patients die from or with infection? Finding the answer through causal analysis of longitudinal intensive care unit data.

Fanghong Zhang (2012). From parametric towards nonparametric mixed modeling of correlated data.

Karel Vermeulen (2015). Bias-reduced Doubly Robust estimation.

Machteld Varewyck (2015). On Quantifying Quality of Care.

Maarten Bijlsma (2016). Age-period-cohort methodology. Confounding by birth cohort in cardiovascular pharmacoepidemiology.

Karl Mertens (2016). Marginal Structural and Structural Nested Models for Causal Inference in Hospital Epidemiology.

Johan Steen (2016). Flexible causal mediation analysis using natural effect models.

Main collaborations

I currently collaborate with Eric Tchetgen Tchetgen (Harvard University) and Vanessa Didelez (University of Bremen) on instrumental variables analysis, with Torben Martinussen (University of Copenhagen) on causal inference for survival outcomes, with Tom Loeys and Beatrijs Moerkerke (Ghent University) on mediation analysis, with Rhian Daniel and Ruth Keogh (London School of Hygiene and Tropical Medicine) on propensity score adjustment and time-varying confounding.

Personal information

I am father of 2 boys, Siebe and Stan. My main hobbies are running, reading (novels) and cooking.