Federated Learning and EHDS as Policy Data Infrastructure for Territorial Health Planning in Ageing Europe

Demographic decline, population ageing and the progressive desertification of healthcare services in many European territories pose major challenges for territorial cohesion and equitable access to care. The European Health Data Space (EHDS) establishes a new regulatory framework for the secondary use of health data across Member States, opening opportunities for advanced analytics under strict data protection safeguards. This paper argues that federated learning (FL), when embedded within the EHDS governance framework, can serve as a policy data infrastructure for territorial healthcare planning in ageing and shrinking regions. Rather than focusing on algorithmic performance, we develop a conceptual model linking FL-enabled distributed analytics to policy questions on long-term care demand, service availability and territorial inequalities. We map policy-makers' information needs, analyse how EHDS provisions can underpin multi-territorial FL projects, and propose a framework describing actors, processes and outputs of EHDS-compliant FL initiatives for health and social care planning.