Data Stewards and FAIR Data Training in Biomedical Research

Abstract

The questions in biomedicine are complex and involve a variety of tasks that can only be tackled in a team. Because the data structures and regulatory requirements involved are also complex, it is important to understand how to manage research data properly. This is where the field of tasks of data stewards has emerged. Although the concept of data stewards is not new, their importance is becoming increasingly visible and the characterisation of specific roles of participation in the research process is being refined. 

In Germany, the federal and state governments have been funding the development of a National Research Data Infrastructure (NFDI) since 2020. The NFDI is intended to set standards in data management and, as a digital, regionally distributed and networked knowledge repository, to secure research data in the long term and make it (re-)usable. The NFDI will also ensure connectivity to international developments such as the European Open Science Cloud (EOSC).  

NFDI4Health, the National Research Data Infrastructure for Personal Health Data, aims to build a FAIR data infrastructure for epidemiological, clinical and public health research, focusing on prospectively collected data in trials and studies, while respecting privacy requirements [1]. Interaction with and feedback from user communities is essential to the service development process. A dedicated workstream within the first five years of NFDI4Health is exploring the role of data stewards (Data Steward Pilot) and providing FAIR data management training to users and multipliers of the infrastructure [2,3]. 

The Data Steward Pilot concept takes the perspective of a local interface of infrastructure services to researchers/users in an institution such as a university. Libraries can act as such an interface. Therefore, the pilot embedded two data stewards at two libraries (ZB MED Information Centre for Life Sciences and USB University and City Library Cologne), who help to develop the NFDI4Health services and at the same time support the main use cases of the consortium, i.e. epidemiological studies (ZB MED) and clinical trials (USB). 

The two data stewards are specifically involved in the development of publication guidelines and are committed to supporting users in making their study data FAIR. Direct feedback from this interaction can be used to iteratively improve the NFDI4Health services. 

The data stewards also develop and deliver research data management training to local researchers. These trainings range from introductory level to more advanced, NFDI4Health specific topics (e.g. distributed data analysis). 

The pilot project will produce a practical report with job profiles for data stewards in biomedicine, together with the training materials as open educational resources.  

The results can be seen as a blueprint for other research institutions and can be reused by multipliers such as libraries or research data competence centres. 

In addition to the details of the concept, we will report on our experience in supporting research and providing training to user communities during the first two years of the pilot. We will put our experiences in the context of other data steward initiatives and look forward to the further development and dissemination of the concept. 

References 

[1] https://www.nfdi4health.de/en/ (12.05.2023) 

[2] Shutsko, Aliaksandra, and Birte Lindstädt. 'Nationale Forschungsdateninfrastruktur für personenbezogene Gesundheitsdaten – NFDI4Health: Pilotprojekt zu Bibliotheken und Forschungsdatenkompetenzzentren als Multiplikatoren ("Data Steward")'. GMS Medizin - Bibliothek - Information 20, no. 3 (22 December 2020): Doc27. https://doi.org/10.3205/mbi000484. 

[3] Helling, Patrick, Rau, Felix, Linne, Monika, Dierkes, Jens, Jetten, Mijke, Walek, Anna, and Szuflita-Żurawska, Magdalena. "The Importance of Demand and Environment for Defining and Establishing the Role of Data Stewards". Zenodo, May 2, 2022. https://doi.org/10.5281/zenodo.6511185.