The research conducted at the Section for Biostatistics comprises statistical methodological research primarily of relevance to health science. Furthermore, we cooperate on a current basis with medical doctors and dentists from a number of health science research environments about solving statistical challenges and developing relevant statistical methodology. 

The methodological research can be divided into the following main areas:

• Analysis of register-based epidemiologic studies. This primarily relates to different types of “time-to-event” data – an area continuously calling for development of methods which partly reflect the structure of register data (e.g. truncation, delayed reporting and misclassification), and which can be applied when dealing with complicated sequences of events (e.g. recurrent events, different types of events and multidimensional data).
• Methods for analysis of longitudinal data. In experimental as well as observational studies, repeated measurements of a large number of factors are often made. When analysing such data there is a potentially complicated correlation structure to consider, as variation in data will be due to both intra- and inter-individual factors. This research area also involves methods for analysis of pharmacokinetic and pharmacodynamic data.
• Comparison and evaluation of measurement methods. Diagnostic tests are developed on a current basis, and, consequently, there is a great need for optimal statistical methods for evaluation thereof.