PT Unknown
AU Palm, J
TI FHIRing up clinical trials: a framework for interoperability-based data analysis
PY 2025
DI 10.22032/dbt.67512
WP https://www.db-thueringen.de/receive/dbt_mods_00067512
LA en
DE Interoperabilität; Klinisches Experiment
AB Background: In 2023, Germany's Federal Ministry of Health launched a digitization strategy supported by the Health Data Usage Act (2024) to improve medical care through secondary use of electronic health record (EHR) data using the HL7 FHIR standard. The Medical Informatics Initiative established 29 Data Integration Centers at university hospitals to harmonize EHR data for cross-site research usage. Methods: We developed the fhircrackr framework, an open-source R package for extracting and transforming FHIR data into analyzable formats. This framework was demonstrated in the HELP trial, a multicenter stepped-wedge cluster randomized clinical trial evaluating a computerized decision support system (CDSS) for Staphylococcus bacteremia management across 5056 patients in 134 wards of five university hospitals. Results: The HELP trial assessed CDSS impact on hospital mortality, 90-day mortality/relapse, and vancomycin use. The study successfully integrated routinely collected EHR data with study-specific data, demonstrating CDSS safety and clinical workflow integration. However, challenges emerged regarding data availability and quality in clinical routine data. Conclusions: This dissertation demonstrates the feasibility of using interoperable EHR data for clinical research while highlighting associated challenges. The fhircrackr framework proved effective for FHIR data transformation, though comprehensive understanding of FHIR and clinical data remains essential. Despite legal, technical, and infrastructure maturity challenges, the interoperable FHIR format enabled effective clinical study evaluation, paving the way for future advances in healthcare data research.
PI Jena
ER