gms | German Medical Science

Workshop organizer: Florian Schwarz, Franziska Jannasch, Sofia Maria Siampani

Inviting organization: NFDI4Health

Content: In this workshop, we will present our work in the National Research Data Infrastructure for Personal Health Data (NFDI4Health) consortium, which is aiming to support the FAIRification of health research data in epidemiological studies across Germany. The participants will be guided through central elements of a federated research project typically involving a multitude of different data holding organizations and their heterogenous datasets. While such large projects include many different facets, we will focus on the practical aspects of data harmonization and distributed analysis.

Initially participants will learn how to make their research data interoperable. To do this, we will go together through a revised step-by-step harmonization protocol and present tools that are developed by Maelstrom-Research and adapted by NFDI4Health to meet the needs of the German scientific community. These tools enable the collection of metadata describing research data, as well as the preparation for the harmonization itself. This information is then used to determine the potential for harmonization and assign specific harmonization rules. Over the past year, the harmonization process underwent some significant changes that are related to the release of the Rmonize package on the Comprehensive R Archive Network (CRAN) by Maelstrom-Research. Together with the participants, we will make a run-through on a few exemplary variables of typical challenges in applying the harmonization rules that we have faced in the NFDI4Health Project thus far. At the end of the first part of the workshop we aim to provide a harmonized data dictionary for these variables, which can be used in the second part of the workshop for federated data analyses.

Following the harmonization of epidemiological research data, the participants will become acquainted with the federated data analysis platform DataSHIELD which has been successfully used in previous national and international projects, such as ENPADASI, InterConnect and INTIMIC. DataSHIELD is an infrastructure and series of R packages that enables the remote and non-disclosive analysis of sensitive research data, thereby mitigating some of the risks other classic cooperative research projects have when it comes to General Data Protection Regulation (GDPR) compliance. Attendees of this workshop will learn how NFDI4Health can support institutes in implementing the DataSHIELD infrastructure. In an interactive session, participants will have the opportunity to see how the central analysis server, hosted at the Max Delbrück Center (MDC) Berlin, works as a gateway to explore DataSHIELD and how it can interact with harmonized datasets that have been uploaded to Opal Servers of the German Institute of Human Nutrition Potsdam-Rehbruecke (DIfE) and MDC. Participants will learn how to assign data from Opal servers, perform data manipulations, subset data, create new variables and perform some simple data analysis.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.