Implementation of a National Synthetic Data Infrastructure for Privacy-Preserving Analytical Workflows in the Czech National Health Information System: Development and Evaluation Study
JMIR Medical Informatics · 10.11.2026 · 14 · e80683-e80683
DOI: 10.2196/80683 ↗Kurz gefasst
Fachartikel in der Zeitschrift JMIR Medical Informatics, online erschienen am 10.11.2026. Verfasst von Martin Komenda und 9 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Health Information and Statistics of the Czech Republic, Palackého náměstí 4, P.O. BOX 60, Praha, 128 00, Czech Republic, 420 725174940. Hauptthemen laut MEDLINE: Workflow, Health Information Systems, Privacy, Computer Security. Die Arbeit nennt 41 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.
Abstract (Originalsprache)
Background
Demand for health data to support research, innovation, and evidence-based policymaking is increasing, but access to patient-level data is constrained by privacy and data protection requirements. To address this challenge, the Czech Republic developed a national synthetic data infrastructure within the National Health Information System (NHIS). The infrastructure aims to support privacy-preserving secondary use of health data by enabling the development and validation of analytical workflows without providing direct access to operational patient-level databases.
Objective
This study describes the design, implementation, governance framework, and operational evaluation of the Czech NHIS synthetic data infrastructure as a national platform for privacy-preserving secondary use of health data and for the development of reproducible analytical workflows. The system strongly supports the continuing relevance and importance of national infrastructures during the transition to full European Health Data Space (EHDS) implementation.
Methods
The infrastructure was implemented using selected nationwide registries, including reimbursement, cancer, reproductive health, and mortality databases. Synthetic datasets were generated using registry-specific rule-based procedures designed to preserve database structures, variable formats, and selected longitudinal relationships while preventing reconstruction of real patient records. Researchers developed and tested SQL-based analytical workflows within the synthetic environment, whereas execution on primary NHIS databases and disclosure control review were performed exclusively by authorized personnel. The implementation was evaluated using predefined operational indicators, including user adoption, request volume, processing outcomes, transparency measures, and privacy-related incidents. Implementation (Results) The infrastructure enables the development and validation of SQL-based analytical workflows without direct access to primary patient-level data. Synthetic datasets representing approximately 0.04% to 0.26% of the selected NHIS registries were created while maintaining schema compatibility and selected analytical relationships. Between the platform’s public launch and July 23, 2026, a total of 170 users registered, 199 requests were submitted, and 184 requests have been processed (173 with data delivery). A total of 41 (20.6%) requests included consent to share SQL scripts, thereby supporting public transparency and reproducibility. Practical implementation was demonstrated through operational use cases in oncology, nephrology, cardiology, and vaccination research. No privacy-related incidents were reported during the evaluation period.
Conclusions
The Czech NHIS synthetic data infrastructure demonstrates that privacy-preserving secondary use of national health data can be implemented through a General Data Protection Regulation (GDPR)–aligned governance framework that combines synthetic data, controlled execution on operational databases, disclosure-risk assessment, and transparent analytical workflows. Rather than serving as a statistically valid surrogate for inferential research, the infrastructure supports the reproducible development of workflows and the secure preparation of population-level analyses. As a fully operational national implementation aligned with the principles of the EHDS, the platform may provide a reference model for other countries developing privacy-preserving health data ecosystems.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Institute of Health Information and Statistics of the Czech Republic, Palackého náměstí 4, P.O. BOX 60, Praha, 128 00, Czech Republic, 420 725174940.
- Institute of Biostatistics and Analyses, Faculty of Medicine, Masaryk University, Brno, Czech Republic.
- Faculty of Biomedical Engineering, Czech Technical University in Prague, Prague, Czech Republic.
- Czech Society of Cardiology, Prague, Czech Republic.
- Faculty of Health Studies, Jan Evangelista Purkyně University in Ústí nad Labem, Ústí nad Labem, Czech Republic.
- AstraZeneca, Czech Republic, Prague, Czech Republic.
Zitieren
APA
Komenda, M., Klimeš, D., Tichopad, A., Jarkovský, J., Májek, O., Mužík, J., Táborský, M., Těšitelová, V., Habětínek, J., & Dušek, L. (2026). Implementation of a National Synthetic Data Infrastructure for Privacy-Preserving Analytical Workflows in the Czech National Health Information System: Development and Evaluation Study. JMIR Medical Informatics, 14, e80683-e80683. https://doi.org/10.2196/80683
Vancouver
Komenda M, Klimeš D, Tichopad A, Jarkovský J, Májek O, Mužík J, et al. Implementation of a National Synthetic Data Infrastructure for Privacy-Preserving Analytical Workflows in the Czech National Health Information System: Development and Evaluation Study. JMIR Medical Informatics. 2026;14:e80683-e80683. doi:10.2196/80683
BibTeX
@article{Komenda2026,
author = {Komenda, Martin and Klimeš, Daniel and Tichopad, Ales and Jarkovský, Jiří and Májek, Ondřej and Mužík, Jan and Táborský, Miloš and Těšitelová, Vladimíra and Habětínek, Jan and Dušek, Ladislav},
title = {Implementation of a National Synthetic Data Infrastructure for Privacy-Preserving Analytical Workflows in the Czech National Health Information System: Development and Evaluation Study},
journal = {JMIR Medical Informatics},
year = {2026},
volume = {14},
pages = {e80683--e80683},
doi = {10.2196/80683}
}
EUVIMED