Evaluation of solutions for data-driven health research networks

Description

Health data research is challenging because of the different data challenges e.g.. governance, security, federation, lack of standardisation, historically poor infrastructure etc. The implementation of the secondary use of healthcare data in Scotland is fragmented, with data providers/processors using different query languages, data storage models and standards of various kinds (e.g. coding, vocabulary). To unite data from geographically decentralised databases, data federation solutions interconnect multiple databases and make data operations interoperable between data controller/processors can improve efficiency and transparency of how data are used (through more unified processes) thus gaining more confidence from the general public.
Several initiatives/software solutions have been launched over the last decade or so to support access to of healthcare data. This work will characterise the existing national and international data-driven health research networks and their software solutions from a methodological, technical and organisational perspective, to meet the needs of the Scottish Safe Haven Network. Our characterisation is based on a framework (compose with principles, architecture and model) that is inspired by the Open Systems Interconnection (OSI) layer model. Existing federation solutions from proprietary and open-source platforms are reviewed and systematically assessed. The work in this paper intends to show the functionalities and responsibilities of all the participating partners. Different modes of operations (cohort discovery, distributed analysis and data extraction) are considered and discussed through the framework.

Health data research faces many challenges e.g. governance, security, federation, lack of standardisation, historically poor infrastructure etc. This work attempt to evaluate the existing software solutions to be adopted by data service industry partners in Scotland. It provides a unique industry as well as academic view on software engineering for health data.

Period	5 Sept 2023