Implementation report on pioneering federated data access for the German National Emergency Department Data Registry

Abstract

Continuous access to electronic health records will fuel the digital transformation of medicine. For data-sharing initiatives, the challenge lies in ensuring data access aligns with the interests of data holders. Federated data access authorization, where data remains controlled locally, may offer a solution to balance these interests. This paper reports on a digital health implementation of the federated data access authorization system used in the German National Emergency Department Data Registry. Using data from 2017 to 2024, we analyzed the system's effectiveness in managing data access in a nationwide research network of 58 emergency departments. Facilitating access to more than 7.9 million records, 75% of data access queries were authorized within 15 days. The system also supports periodic queries, enabling recurring real-time access. Query volumes grew from 15 to over 23,000 by 2024, with completion rates of 86%. The system may thus serve as a blueprint for data-sharing initiatives worldwide.

Fig. 1. AKTIN infrastructure processes.

a Electronic health record data are captured in the participating emergency department nodes within an AKTIN data warehouse. After approval from the Data Use and Access Committee (DUAC), the AKTIN IT group translates proposals into R or SQL syntax. From the syntax, a data query is created which is communicated to the nodes using the AKTIN Broker via xml data structure including additional descriptive- and provenance metadata. The local emergency department (ED) node must authorize data access for each query. b The DUAC evaluates proposals and formulates a vote translated into a technical request. c A federated data access authorization process is implemented within the AKTIN DWH software. The AKTIN IT group receives technical requests and then uses the AKTIN Broker middleware to send queries to AKTIN DWHs in the ED nodes. The ED nodes may execute the query, inspect the results in CSV format, and authorize data access. The R logo is used under the terms of the Creative Commons Attribution-ShareAlike 4.0 International license (available from www.r-project.org/logo/). Postgres, PostgreSQL, and the Slonik Logo® are trademarks or registered trademarks of the PostgreSQL Community Association of Canada, and are used with their permission (available from www.postgresql.org/about/policies/trademarks).

Fig. 2. Formal workflow description and translated screenshots of local data access authorization of a query in an AKTIN data warehouse manager within a local node of the AKTIN infrastructure.

Errors in the federated data access authorization process (left) result in failed queries (right). Note: this diagram employs a modified version of BPMN, adapted to clarify specific workflow elements unique to the AKTIN infrastructure. This adaptation may not strictly adhere to formal BPMN guidelines.

Fig. 3. Translated visualization of a query within the dedicated data request view in the AKTIN DWH manager.

Participating emergency department nodes of the AKTIN infrastructure review the query and individually decide if they want to authorize data access.

Fig. 4. Growth of the AKTIN infrastructure and analysis of KPIs.

a Active ED nodes connected throughout the years illustrate the infrastructure growth since 2017, which (b) enables access to 1.6 Million Electronic Health Records in 2023 and 7.9 Million Electronic Health Records in total. Bottom: rate of unanswered queries. Kaplan–Meier survival curves show the proportion of uncompleted individual queries over days after retrieval by the AKTIN DWH. c The behavior of nodes varied slightly across the years; ED nodes replied more swiftly in the first years of the infrastructure. d The behavior differs across nodes. The hexagonal density distribution of different Kaplan–Meier curves illustrates the concentration of query completions for all active nodes.