The Privacy-Preserving Record Linkage (PPRL) Task Force released two post-workshop documents summarizing their technical and ethical analyses of the data linkage landscape, and their reflections and recommendations to enable privacy-preserving record linkage.
Health data research, especially in the case of rare disease patients, would strongly benefit from reliable linkage of patients’ datasets obtained from independent research projects. However, technical difficulties often arise when connecting datasets produced by different organizations and/or investigators, in addition to legal-ethical concerns related to privacy and data protection. Chaired by Dr. Petra Kaufmann, Dr. Dixie Baker and Prof. Bartha Knoppers, the PPRL Task Force set out to discuss and tackle these challenges. The Task Force was set up in collaboration with the Regulatory and Ethics Working Group (REWG) of the Global Alliance for Genomics and Health (GA4GH).
From the technological point of view, the PPRL Task Force concluded that the European patient-identity management solution (EUPID) approach was most aligned with IRDIRC and GA4GH’s global research environment. EUPID was developed by the European Network for Cancer Research in Children and Adolescents (ENCCA) and presents the interesting advantage of using algorithms to generate a context-specific pseudonym for each individual. This approach allows identity linkages to be maintained locally while the re-identification of linked data is done through collaborative processes between separate trusted entities. From the ethical point of view, the PPRL Task Force recommended several features that could be incorporated during the design of a PPRL system such as the use of a trusted cryptographic hashing method to enable linkage or the separation among trusted entities of the linkage and re-identification duties. The Task Force also recognized the existing diversity in methodologies through which patients give consent for their data to enter a linkage system. Importantly, patients should be supported if they want to withdraw their consent, even if this brings certain technical complications.
The PPRL Task Force is now collaborating with the EUPID project to further understand its framework and use. A peer reviewed article on the technical work done to date by the PPRL Task Force will be published in the IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB) Journal shortly.
For more details, read the post-workshop documents: