Coping with interoperability in the development of a federated research infrastructure: achievements, challenges and recommendations from the JA-InfAct

Background Information for Action! is a Joint Action (JA-InfAct) on Health Information promoted by the EU Member States and funded by the European Commission within the Third EU Health Programme (2014–2020) to create and develop solid sustainable infrastructure on EU health information. The main objective of this the JA-InfAct is to build an EU health information system infrastructure and strengthen its core elements by a) establishing a sustainable research infrastructure to support population health and health system performance assessment, b) enhancing the European health information and knowledge bases, as well as health information research capacities to reduce health information inequalities, and c) supporting health information interoperability and innovative health information tools and data sources. Methods Following a federated analysis approach, JA-InfAct developed an ad hoc federated infrastructure based on distributing a well-defined process-mining analysis methodology to be deployed at each participating partners’ systems to reproduce the analysis and pool the aggregated results from the analyses. To overcome the legal interoperability issues on international data sharing, data linkage and management, partners (EU regions) participating in the case studies worked coordinately to query their real-world healthcare data sources complying with a common data model, executed the process-mining analysis pipeline on their premises, and shared the results enabling international comparison and the identification of best practices on stroke care. Results The ad hoc federated infrastructure was designed and built upon open source technologies, providing partners with the capacity to exploit their data and generate dashboards exploring the stroke care pathways. These dashboards can be shared among the participating partners or to a coordination hub without legal issues, enabling the comparative evaluation of the caregiving activities for acute stroke across regions. Nonetheless, the approach is not free of a number of challenges that have been solved, and new challenges that should be addressed in the eventual case of scaling up. For that eventual case, 12 recommendations considering the different layers of interoperability have been provided. Conclusion The proposed approach, when successfully deployed as a federated analysis infrastructure, such as the one developed within the JA-InfAct, can concisely tackle all levels of the interoperability requirements from organisational to technical interoperability, supported by the close collaboration of the partners participating in the study. Any proposal for extension, should require further thinking on how to deal with new challenges on interoperability.

Coping with Interoperability in the Development of A Federated Research Infrastructure: Achievements, Challenges and Recommendations from the JA Infact

Background: Information for Action is a Joint Action (JA InfAct on Health Information) promoted by the EU Member States and funded by the European Commission within the Third EU Health Programme (2014-2020) to create and develop solid sustainable infrastructure on EU health information. The main objective of this the JA-InfAct is to build an EU health information system infrastructure and strengthen its core elements by a) establishing a sustainable research infrastructure to support population health and health system performance assessment, b) strengthening the European health information and knowledge bases, as well as health information research capacities to reduce health information inequalities, and c) supporting health information interoperability and innovative health information tools and data sources.Method: Follsowing a federated analysis approach, JA-InfAct developed an ad hoc federated infrastructure based on distributing a well-defined process-mining analysis methodology to be deployed at each participating partners’ systems to reproduce the analysis and pooling the aggregated results from the analyses. To overcome the legal interoperability issues on international data sharing, data linkage and management, partners (EU regions) participating in the cases study worked coordinately to query their real-world healthcare data sources complying with a common data model, executed the process mining analysis pipeline on their premises, and shared the analysis results enabling international comparison and the identification of best practices on stroke care.Results: the ad hoc federated analysis infrastructure was designed and built upon open source technologies providing partners with the capacity to exploit their data and generate stroke care pathway analysis dashboards. These dashboards can be shared among the participating partners or to a coordination hub without legal issues enabling the comparative evaluation of the caregiving activities for acute stroke across regions.Nonetheless, the approach is not free of a number of challenges that have been solved, and new challenges that should be addressed in the eventual case of scaling up. For that eventual case, 12 recommendations on the different layers of interoperability have been provided. Conclusion: The JA InfAct federated analysis infrastructure has been able to cope with all levels of the interoperability -legal, organisational semantic and technological. Any proposal for extension would require upgrade so as to deal with new challenges on interoperability such as large scale application of GDPR principles, developing high-profile resources capacity across Europe, developing a common strategy to data quality assurance and preparedness for high performance computing and federated learning.

Developing a Transnational Health Record Framework with Level-Specific Interoperability Guidelines Based on a Related Literature Review

With the advent of digital healthcare without borders, enormous amounts of health information are captured and computerized. As healthcare quality largely depends on the reliability of given health information, personal health records should be accessible according to patients’ mobility, even as they travel or migrate to other countries. However, since all the health information is scattered in multiple places, it is an onerous task to carry it whenever people move to other countries. To effectively and efficiently utilize health information, interoperability, which is the ability of various healthcare information technologies to exchange, to interpret, and to use data, is needed. Hence, building a robust transnational health information infrastructure with clear interoperability guidelines considering heterogeneous aspects is necessary. For this purpose, this study proposes a Transnational Health Record framework, which enables access to personal health records anywhere. We review related literature and define level-specific interoperability guidelines, business processes, and requirements for the Transnational Health Record system framework.

Family History Information Exchange Services Using HL7 Clinical Genomics Standard Specifications

A number of family history applications are in use by health care professionals (e.g., CAGENE, Progeny, Partners Health care Family History Program) as well as by patients (e.g., the U.S. Surgeon General’s Family History Program). Each has its own proprietary data format for pedigree drawing and for the maintenance of family history health information. Interoperability between applications is essentially non-existent. To date, disparate family history applications cannot easily exchange patient information. The receiving application should be able to understand the semantics of the incoming family history and enable the user to view and/or to edit it using the receiving applications interface. We envision that any family history application will be able to send and receive an individual’s family history information using the newly created HL7 Clinical Genomics Specifications through the Semantic Web, using services that will transform one format to the other through the HL7 canonical representation.