A team of experts from the Biobanks and Biomolecular Resources Research Infrastructure Consortium (BBMRI-ERIC) and the University of Luxembourg (UNILU) worked towards the creation of a draft document that describes data access governance frameworks from a sample of data hubs that facilitate sharing of health data for research purposes. This sample includes the European Genome Phenome Archive (EGA), the International Cancer Genome Consortium (ICGC), the US database of Genotypes and Phenotypes (dbGaP), Genomics England, Findata, the Medical Informatics Initiative Germany (MII Germany) and the French Health Data Hub.
Based on the contributors’ pooled experience, these models covered a range of data access governance models used by large-scale data resources. The aim of these frameworks is to provide qualified researchers access to health data while also ensuring protection of the rights and interests of data subjects and data providers.
The draft establishes taxonomy of data access governance models, from centralized – where access to multiple resources is granted by a central data access committee, to fully distributed – where each resource oversees access to its own data. In addition to traditional download mechanisms of providing access, alternatives are being explored where data are made accessible to users remotely in secure processing environments. Federated approaches, enabled by data and technical standards, can allow users to draw insights across networks of these secure databases.
The next phase of the project will be to expand the survey of data access models, including a review of standard data use agreements used to facilitate data sharing. The project will engage data protection authorities and ethics professionals in a dedicated workshop to ensure these models respect legal and ethical principles, before finalizing recommendations for data access governance for health research and innovation clouds (HRICs).