Introduction
The Federated Data Steward plays a key supporting role in ensuring that health data used in federated learning (FL) is high quality, well-documented, legally compliant, and FAIR — even when it remains distributed across multiple institutions.
Building on the evolving profession of research data stewardship, this role adapts to the specific needs of secure, privacy-preserving, and cross-site data sharing. Federated data stewards act as local or central points of contact who bridge research, infrastructure, and governance efforts.
They help translate policies into practice, ensure datasets are harmonised and well-annotated, and support institutions in aligning with federated learning protocols and standards.
Key Responsibilities
- Assist with mapping and transforming local data into harmonised formats or common data models (e.g. OMOP, FHIR)
- Ensure metadata completeness and consistency to support interoperability
- Guide researchers and IT teams on FAIR principles and legal requirements
- Support the documentation of data flows, wrangling pipelines, and data provenance
- Coordinate with legal teams on data access permissions, pseudonymisation, and retention policies
- Help validate and test data readiness for FL training rounds
- Act as a knowledge hub for tools, standards, and training in FL data stewardship
Common Challenges
- Supporting multiple departments or projects with diverse data formats and quality
- Translating FAIR and legal principles into operational workflows
- Working across silos (research, IT, legal) without formal authority
- Managing uncertainty about roles, responsibilities, and technical expectations in FL
- Ensuring sustainability and documentation beyond the project lifecycle
Recommended Tools & Resources
Data Models & FAIR Alignment
Data Stewardship Tools
Metadata & Validation
Relevant FLKit Sections
- Enhance & Wrangle Data: data cleaning, harmonisation, metadata
- Plan & Govern: local policy support, permissions, documentation
- Enable Infrastructure: aligning local systems with FL pipelines
Training & Further Reading
- ELIXIR Data Stewardship Competency Framework
- Dutch Roadmap to Professionalising Data Stewardship
- RDA Interest Group on Professionalising Data Stewardship
Solution
- European Data Protection Supervisor’s “Preliminary opinion on Data Protection and Scientific Research”
- BBMRI-ERIC ELSI Knowledge Base contains governance templates and guidance for federated learning projects.
- Data Stewardship Wizard (DSW) can help establish governance frameworks for federated learning projects.
- FAIR Cookbook provides step-by-step recipes for data governance tasks.
- TeSS Training Portal offers training materials on data governance and management.
Related pages
More information
Links to FAIR Cookbook
FAIR Cookbook is an online, open and live resource for the Life Sciences with recipes that help you to make and keep data Findable, Accessible, Interoperable and Reusable; in one word FAIR.
Links to DSW
With Data Stewardship Wizard (DSW), you can create, plan, collaborate, and bring your data management plans to life with a tool trusted by thousands of people worldwide — from data management pioneers, to international research institutes.
Training
Contributors
Editor
KU Leuven / University of UHasselt