Description
The construction of the ecosystem for the secondary use of e-health data in the European Health Data Space (EEDS) poses a significant scenario of opportunities for Spanish research, innovation and entrepreneurship. To this end, the European Union is promoting a multitude of strategic projects in which hospitals, health research foundations, universities, research centres and Spanish companies participate. The list of projects is extensive and aims to satisfy at least two objectives: to promote the generation of infrastructures capable of generating quality datasets and to promote conditions for their reuse.
The role of Spain. Strengths in the deployment of the European Health Area
Spain offers significantly favourable conditions not only to participate but also to contribute significantly to the tasks of creating the EEDS:
- First, our public health system is characterized by a high level of integration and structuring. Unlike systems based on reimbursement mechanisms, in which there may be an atomisation in the field of service provision, in our system we have a clear frame of reference in primary care, medical specialities and hospital services.
- On the other hand, the experience deployed by our health environments from the General Data Protection Regulation (GDPR) and, particularly, the lessons learned from the seventeenth additional provision on health data processing of Organic Law 3/2018, of 5 December, on the Protection of Personal Data and guarantee of digital rights (LOPDGDD) they constitute a valuable experience.
- The opening of the National Health Data Space promoted by the Government of Spain and promoted by the Ministry for Digital Transformation and Public Function, the Ministry of Health and the Autonomous Communities allows the deployment of an essential infrastructure for the EEDS.
The National Health Data space was presented on October 29. The event highlighted how this project represents a paradigm shift that revolutionizes the management of health data, promoting a federated, secure and ethical model that preserves the sovereignty and privacy of information while facilitating its use for research, innovation and public policies. Its operation is based on a federated catalog of metadata and a rigorous process of access and analysis in secure environments, which seeks to promote open science and scientific and technological advances, benefiting patients, researchers, managers and industry.
Lessons learned from European Projects
The path taken by Regulation (EU) 2025/327 of the European Parliament and of the Council of 11 February 2025 on the European Health Data Space, amending Directive 2011/24/EU and Regulation (EU) 2024/2847 (EEDSR), poses significant challenges that are addressed in research projects funded by European and national funds. The lessons learned in some of them can be extraordinarily useful for the research and entrepreneurship community in our country. We cannot forget that we start from significant strengths.
1.-Compliance by design
The existence of a new regulation requires a rigorous analysis of the state of the art in our organizations, not only to implement its deployment but also to ensure the preconditions of legal reliability of the datasets and the research that is proposed.
2.-Accountability: proactive responsibility and documentary strength
In our country we come from a long tradition of accountability. The EEDSR will impose on the data requester a set of relevant documentary requirements, such as, for example, having provided safeguards to prevent any misuse of electronic health data. This issue cannot be neglected from the point of view of data holders, who will also have to meet certain requirements. For example, proving that data is legitimate and reusable is an ethical and legally documentable issue; and the simplified procedure for accessing electronic health data through a trusted health data holder requires the latter to document the security of its data space or capabilities to evaluate requests for access to health data.
One of the main obstacles we face in this intermediate period of implementation of the EEDS lies precisely in the organizational culture for the generation of verifiable evidence. As standardization and the set of common rules of the EEDS scale, it will be necessary to deepen the dynamics of proactive responsibility understood as demonstrated responsibility.
3. Secure processing environments
In our country, health environments by their very definition must be safe environments. The deployment of the National Security Scheme (ENS) and the GDPR have allowed the entire health system, public or private, to adopt maturity models that are perfectly consistent with the conditions of the secure processing environments defined by the EEDSR.
Challenges of the Spanish system
Along with the inherent strengths of our system, it is necessary to consider those aspects that present themselves as challenges.
1. Anonymisation and pseudonymisation
In the national context, the aforementioned seventeenth additional provision of Organic Law 3/2018, of 5 December, on the Protection of Personal Data and guarantee of digital rights, defines specific conditions for pseudonymisation. These consist of the functional separation between the teams that pseudonymize and those that reuse data, and the definition of a secure environment that prevents any attempt at re-identification. In addition, there are legal guarantees in terms of individual commitments not to re-identify, deployment of the impact assessment tool related to data protection and supervision by ethics committees. The challenge of anonymization is more demanding, since it implies the impossibility of linking health data with those of the original patient under any conditions.
2. Reeskilling of teams
The European Health Data Space (ESDS) will pose an unprecedented training challenge that will cut across all sectors involved in the health data ecosystem. Research ethics committees should familiarise themselves not only with the permissible secondary uses of health data, but also with the integration of the Artificial Intelligence Regulation and with the ethical principles of the ALTAI (Assessment List for Trustworthy Artificial Intelligence) framework. This need for reeskilling will also extend to health systems and health administration, where Health Data Access Bodies will require highly qualified personnel in these new ethical and regulatory frameworks, as well as reliable data holders who will safeguard sensitive information. Development staff and IT teams will also need to acquire new skills in critical technical areas, such as cataloguing, validation, and curation of data, as well as in interoperability standards that enable effective communication between systems. Perhaps the most sensitive training challenge will fall on new entrants, who will be able to take advantage of opportunities to access datasets for innovative secondary uses. This especially concerns technology startups in the health sector. To face a very demanding regulatory framework (GDPR, Regalmento de AI, EEDSR), the resources and capabilities for legal compliance in Spanish SMEs is notably limited. For this reason, it will be necessary to build a solid culture of data protection and ethical development of reliable artificial intelligence systems from the beginning.
3. Data cataloguing: the challenge of quality and standardization
In the context of the European Health Data Space, deepen the standardization of data through the most functional methodologies – such as OMOP CDM for observational clinical data, HL7 FHIR for dynamic information exchange, DICOM for medical imaging, or reference terminologies such as SNOMED CT, LOINC and RxNorm— is presented as a key strategic element for the creation and re-use of high-quality datasets. However, the adoption of these standards is not enough on its own: the processes of validation, semantic annotation and data enrichment require highly qualified human resources capable of ensuring the coherence, completeness and accuracy of the information, making this training a real precondition for effective participation in the European health data ecosystem. Alignment with the standardized cataloguing of datasets following the HealthDCAT-AP (Health Data Catalog Application Profile) standard, which allows the descriptive metadata of health data resources to be described in a homogeneous way, is presented as one of the immediate challenges, along with the implementation of the work that has been deployed in relation to the data utility quality label, a quality label that assesses the real usefulness of data for secondary uses and is becoming a seal of trust for users and researchers.
If previously in this article the very high capacities of the Spanish health system to generate health data in a systematic way and in significant volumes were highlighted, these aspects of cataloguing, standardization and quality certification will occupy an absolutely central place in designing optimal conditions of European competitiveness in their reuse, transforming the abundance of data into a real strategic advantage that allows Spain to position itself as a relevant player in the research and innovation landscape with electronic health data.
The experience of the EUCAIM project (Cancer Image EU)
The European Health Data Space Regulation aims to enable the secondary use of electronic health data across Europe through harmonised rules in a federated ecosystem. In the cancer arena, fragmented access to high-quality datasets slows down research, limits reproducibility and undermines Europe's ability to develop and validate reliable AI tools for oncology.
EUCAIM demonstrates the viability of an ecosystem for the secondary use of cancer through a federated model that allows cross-border access under harmonized rules guaranteeing adequate control of resources at the local level. And this is deployed through a set of enabling components:
1) A Secure Processing Environment (SPE) federated at European level
EUCAIM is creating a federated PES to enforce data access conditions, control processing, and support secure cross-border analysis under EEDS restrictions. This PES is fully in line with the requirements and measures laid down in Article 73 EEDSR for safe environments.
2) Overcoming the "anonymisation barrier"
EUCAIM promotes a layered anonymization strategy that combines dataholder-autonomous local anonymization processes with platform controls to enable datasets to remain useful for AI research and development. The importance of this approach lies in the fact that it aims to reconcile the protection of privacy with the practical need to have sets with large volumes of data characterized by their diversity.
3) Data cataloguing and standardization
EUCAIM aligns cataloguing with the HealthDCAT-AP principles whose main objective is to apply the FAIR principles, that is, to ensure that data is findable, accessible, interoperable and reusable.
4) Reduction of legal costs
EUCAIM has deployed its own compliance framework aimed at the General Data Protection Regulation and the Artificial Intelligence Regulation. To do this, a robust compliance framework is in place at the platform level that is deployed across complex data ecosystems. This is based on data protection impact assessments (included in the GDPR) with a particular focus on fundamental rights. It also incorporates training and professional retraining of users as a functional requirement, so that compliance capability becomes an essential feature.
5) Support for data users
EUCAIM offers significant advantages to data users, including researchers and AI developers, by establishing a transparent and well-governed environment for data access. The adoption of transparent governance criteria, clearly defined obligations and their technical application by the platform, provide data users with the guarantee that their access is adequate and lawful, fully auditable and remains stable over time. The platform's design ensures that users can leverage powerful data for advanced analytics, including federated processing in a secure environment. Through mandatory training and implementation of standardized procedures, teams benefit from less uncertainty and are better equipped to align with compliance requirements set forth by the EEDSR, GDPR, and AI governance frameworks.
6) Guarantee of patients' rights
EUCAIM's approach is based on data protection by design and by default that unites organisational safeguards with robust technical controls. This framework has been purpose-built to minimise the risk of data misuse, while supporting safe and effective cross-border cancer research and innovation. The result is a system in which the protection of privacy is not an obstacle but a fundamental element that allows the responsible use of data for the benefit of society and science. The model reinforces accountability for the secondary use of health data by combining strong governance oversight, a comprehensive record of actions, and strict and enforceable obligations for all participating entities. All actions taken with patient data are recorded and reviewed, ensuring that all uses are fully auditable. This traceability ensures that the processing of data is kept within the limits of the permitted use and that any deviations can be identified and addressed quickly.
Multi-level governance: the key to sustainable success
The most relevant lesson learned at EUCAIM concerns the imperative need for articulated, coherent and operational multilevel governance. In a broad sense, it is essential to provide effective governance tools and frameworks on three fundamental dimensions:
- Firstly, on the processes for generating datasets and their sharing conditions, establishing clear criteria on what data is generated, how it is standardised, who holds rights over it and under what licences and restrictions it can be shared with third parties.
- Second, on data access request processes, defining transparent and efficient procedures so that researchers, innovators, and policymakers can identify, request, and obtain access to the data needed for their projects, minimizing administrative burdens without compromising ethical and legal guarantees.
- Thirdly, on the processes of validating the correctness of the datasets and adherence to the system, as well as the procedures for authorising access to data, ensuring that only data of certified quality feed the infrastructure and that only authorised users with legitimate purposes access sensitive information.
This procedural governance cannot function without strategic and operational decisions regarding the definition of human resources roles and functions. To do this, it is necessary to have the necessary professional profiles such as data managers, experts in research ethics, cybersecurity specialists, data curators and quality managers. Secondly, it will be essential to define the secure processing environments where analyses are carried out on sensitive data, ensuring that these spaces comply with the highest technical standards of security, traceability, auditing and privacy preservation, and that they are designed to operate under the principle of zero trust) adapted to the health context. Only through this multi-level governance architecture, which integrates technical, organizational, ethical and legal dimensions at all levels of decision-making – from the design of national policies to the day-to-day operational management of platforms – will it be possible to build health data infrastructures that are truly sustainable, reliable and capable of generating long-term social, scientific and economic value. positioning the Spanish healthcare system as a strategic player in the European healthcare innovation ecosystem.
Content prepared by Ricard Martínez Martínez, Director of the Chair of Privacy and Digital Transformation, Department of Constitutional Law, University of Valencia. The contents and views expressed in this publication are the sole responsibility of the author.
