JMIR Publications

ABSTRACT

Synthetic data represent an approach that may potentially address the longstanding tension between the growing demand for primary health data and the imperative to protect individual privacy. By creating synthetic data sets that mirror the structure of real data but do not disclose personal identifiers, the system enables meaningful secondary use for research, policy analysis and innovation. We present the Czech Republic’s synthetic data infrastructure, developed within the National Health Information System (NHIS), as one of Europe's pioneering models for sharing and analysing primary health data. The synthetic data pool is currently being built on the basis of four large national health registries (health services reimbursements, cancer, reproductive health and mortality), with the potential to be expanded to all 61 NHIS agendas. The main principles are transparency and fairness. We describe the process by which researchers and institutions gain access to the synthetic database and develop scripts that are later used to obtain outputs from real patient-level data. We provide examples of such outputs across medical disciplines and aspects of healthcare. The system has already proven its value in the development of national strategies and has been validated by academic, governmental and commercial stakeholders. By adhering to the principles of the European Health Data Space, this model is becoming transferable to other EU countries.