Two years after the foundation of the institute, the Department of Biometry and Data Management has been established in 1983. In the beginning exclusively intended as a service unit supporting the other departments in planning, realising and evaluating epidemiological studies and ensuring an efficient IT infrastructure and database administration, it soon evolved into a department working on questions regarding methods and contents. Besides providing indispensable IT services, the department emphasises on harnessing and developing biometric methods for epidemiological issues.

With the increasing demands and opportunities due to Open Science and Open Data, the department's scope has expanded further. The expertise in this area is bundled and enhanced by the Cross-departmental Center for Research Data Infrastructure and Data Science in Epidemiology. Firstly, the center is dedicated to the further development of research data management and infrastructure in order to fully implement the FAIR principles. In this way, research data will be made findable, accessible, interoperable and reusable. Secondly, the center pools and analyses globally available datasets and develops statistical/computer science methods for such distributed data analyses.

Biometry

The institute has been in the position to assemble biometricians experienced in a wide range of epidemiological studies (case control, cohort, cross-sectional, longitudinal, incidence studies, etc.) as permanent staff. For specific research projects, a number of short-term employees as well as postgraduate students (usually financed by third parties) extend the range of methods employed in the department. In this context the department also offers the opportunity to write a PhD thesis.

Methods development

We are presently working on diverse topics: these include statistical issues related to the pharmacoepidemiologic research database (twophase studies, control for confounding - Statistical Modeling of Secondary and Registry Data Unit), improvement of statistical procedures in the analysis of gene-gene- and gene-environment interactions, the identification of biological pathways involved in the development of diseases and in the statistical modeling of pathways, accumulations of risk or critical time periods in life-course epidemiology (Statistical Methods for Genetics and Molecular Epidemiology Unit), statistical modeling of nutrition data and use of geographical information systems in epidemiologic studies (Statistical Modeling of Primary Data Unit). Furthermore, the department engages in the development of methods for causal inference, e.g. for instrumental variables or time-varying confounding (Statistical Methods for Causal Inference Research Group), and statistical machine learning (Emmy Noether Junior Research Group: Beyond Prediction - Statistical Inference with Machine Learning). With the improved accessibility of research data, development of methods for analysis and combining data, for example by means of record linkage or distributed analyses, is increasingly important.

Statistical support - quantitative methods consulting service

The department also houses a methods consultancy service that is used BIPS-internally as well as externally by other research institutions and scientifically working clinicians, primarily from Bremen and the surrounding region. Support is mainly requested in the planning phase of studies e.g. on the selection of the study design or on sample size calculations. Appropriate analysis methods are proposed and/or developed and support in the application of methods as well as in the interpretation and presentation of results is provided.

Within the inter-departmental working group GeTTCausal (GePaRD & Target Trials for Causal Inference), we combine expertise in secondary data analysis, causal inference, and clinical epidemiology to best analyze important causal questions based on GePaRD using the "target trial emulation" (TTE) approach.

Data Management

The department is responsible for providing an efficient IT infrastructure for the entire institute. This includes the procurement, installation, configuration, and maintenance of required software and hardware, provision of standard services such as email, file, or backup services, as well as the operation of systems necessary for research and research data management. The sensitivity of the data processed at the institute requires a high level of security, making IT security an integral part of the institute's IT infrastructure. By creating tools and methods for data collection and processing, metadata management, data preparation and quality assurance, and software development, the department enables the implementation of FAIR principles, especially in conducting epidemiological studies. This involves the central maintenance, management, and documentation of the institute's data holdings beyond the duration of the actual studies to ensure future usability of the data.

With expertise and the help of advanced tools in research data management and medical documentation, the department supports other departments in planning and conducting studies. Additionally, these tools and expertise are made available to external partners for use in studies. These tasks are consolidated in the IT, Data Management, and Medical Documentation Unit, whose staff have long-standing experience and are comprised of computer science, system administration and medical documentation specialists.

Iris Pigeot
Head of Department (Scientific Director)

Contact:
Prof. Dr. Iris Pigeot
Tel: +49 (0)421 218-56742
Fax: +49 (0)421 218-56741

Email:
pigeot(at)leibniz-bips.de

Vanessa Didelez
Deputy Head of Department
(Head of Research Group)

Contact:
Prof. Dr. Vanessa Didelez
Tel: +49 (0)421 218-56939
Fax: +49 (0)421 218-56941

Email:
didelez(at)leibniz-bips.de

• Statistical Modeling of Primary Data
• Statistical Modeling of Secondary and Registry Data
• IT, Data Management and Medical Documentation
• Statistical Methods for Causal Inference Research Group
• Emmy Noether Junior Research Group: Beyond Prediction - Statistical Inference with Machine Learning

Stefanie Brand
Tel.: +49 (0)421 218-56750
sbrand(at)leibniz-bips.de