The European Health Data Space (EHDS) [1] is one of the most ambitious and transformative policy initiatives undertaken by the European Union (EU) in recent decades. Positioned at the intersection of health, digital innovation, and fundamental rights, the EHDS aims to create a trusted framework for the use and sharing of health data across EU Member States and associated countries. Its overarching goal is to empower individuals with secure access to their personal health data, enable its seamless exchange across borders, and facilitate the responsible secondary use of data for research, innovation, policymaking, and public health.

The EHDS is anchored in the broader vision of the EU’s digital and health strategies, particularly the European Data Strategy and the Digital Single Market [2]. It is designed to address key fragmentation challenges in Europe’s health data landscape, where systems, standards, governance models, and legal interpretations vary widely between countries, by providing both a legal and technical infrastructure that guarantees data interoperability, trust, and value creation. The EHDS seeks to ensure that health data flows across Europe as easily and safely as goods and services, enabling better health care delivery, enhanced medical research, and more resilient public health systems [3].

One of the most critical enablers of this vision is the European Electronic Health Record Exchange Format (EEHRxF). Established to facilitate the interoperability of electronic health records (EHRs) across the EU, the EEHRxF sets out a common structure and vocabulary for core health data domains [4]. Its primary objective is to ensure that any European citizen can access and share their health data in a standardized format, no matter where they receive care in the EU. The EEHRxF plays a strategic role in enabling interoperability across cross-border health services, including the MyHealth@EU infrastructure (European Commission [EC]) [5,6], while also supporting the implementation of the EHDS’s primary data use pillar.

The synergies between the EHDS and other European data initiatives, such as the European Open Science Cloud (EOSC) [7], the AI Act [8], and sectoral data spaces (eg, cancer mission and rare diseases) [9], further highlight the strategic role of the EEHRxF. For example, the integration of health data into the EOSC depends on standardized metadata and ontologies; the training of high-risk AI models under the AI Act requires access to high-quality, interoperable datasets; and the development of European Reference Networks for rare diseases hinges on the ability to share structured clinical data across national borders. In all these contexts, the EEHRxF is a key enabler of interoperability, acting as the common reference layer that allows data to flow, be understood, and be reused responsibly.

The EEHRxF currently defines structured formats for several key types of health data: patient summaries; electronic prescriptions; electronic dispensations; medical imaging studies and related imaging reports; medical test results, including laboratory and other diagnostic results and related reports; and discharge reports [1]. These data types were prioritized due to their high impact on continuity of care, especially in cross-border contexts. By harmonizing how this data is represented and exchanged, the EEHRxF supports safer, faster, and more informed clinical decision-making, whether a patient seeks care in another Member State or returns home with medical information that must be integrated into their domestic EHR system.

Technically, the EEHRxF builds on International and European standards such as Health Level Seven International (HL7) Fast Healthcare Interoperability Resources (FHIR) [10,11], Systematized Nomenclature of Medicine Clinical Terms (SNOMED CT) [12,13], Logical Observation Identifiers Names and Codes (LOINC) [14,15], and International Statistical Classification of Diseases (ICD) [16,17], while aligning with European profiles and governance through initiatives like Integrating the Health Care Enterprise (IHE) [18] and the European Committee for Standardization/Technical Committee (CEN/TC) 251 [19]. It defines the structure, content, coding, and semantic consistency required to achieve meaningful interoperability. These standards ensure that both syntactic and semantic interoperability are achieved, enabling not just data exchange, but true data understanding across systems and institutions [20].

In addition to technical interoperability, the success of the EEHRxF and the broader EHDS depends on sustained political will, robust governance, cross-border collaboration, and the engagement of a wide range of stakeholders, including health care providers, IT vendors, patients, researchers, and regulators. Capacity building and change management are essential to ensure that standards are adopted and implemented effectively across diverse national systems [21]. Investments in digital infrastructure, workforce training, and semantic harmonization must accompany regulatory and technical developments to ensure that interoperability is not just defined on paper but realized in practice.

The feasibility demonstrators, while focused on technical interoperability, aimed at getting a clear understanding of all these dimensions [22]. The work conducted relied on numerous interactions with the involved entities, promoting open discussions, leveraging their knowledge and expertise on the way current EHR systems are built and used, and what needs to be done at the technical and organizational level for a successful adoption of the EEHRxF.

Among the data domains prioritized for standardization and exchange under EEHRxF are laboratory reports, hospital discharge reports, and patient summaries. These 3 document types play distinct yet complementary roles in ensuring continuity of care, improving patient safety, and enabling efficient clinical decision-making, especially when care is delivered across institutional or national boundaries. While all 3 are central to the EEHRxF, the empirical approach varies according to their current implementation status. Specifically, laboratory reports and hospital discharge reports were selected for hands-on feasibility demonstrators due to the existing gaps in their technical alignment. In contrast, patient summaries were evaluated primarily through expert workshops and stakeholder surveys, reflecting their higher maturity level in existing European digital health implementations.

Patient summaries serve as concise overviews of a patient’s most essential health information, including current medications, allergies, diagnoses, surgical history, and vaccinations. Designed for use in unplanned care settings, such as emergency rooms or cross-border encounters, they offer clinicians immediate access to critical insights that can inform safe and timely treatment. While the feasibility of this format has already been demonstrated through the MyHealth@EU infrastructure [6], the feasibility demonstration discussed here delved into a more practice-oriented perspective, aiming to understand how particular stakeholders (such as health care providers and professionals, national, regional, and local health infrastructures, and IT vendors) are aligning with it.

Laboratory reports, which include test results such as blood work, microbiology, or pathology findings, are among the most frequently generated and exchanged types of clinical data. The EEHRxF introduces a harmonized structure and coding system (eg, using LOINC for lab test identifiers) to ensure that test results can be interpreted consistently regardless of the country or system where they were produced. This interoperability is particularly important in managing chronic conditions, monitoring infectious disease outbreaks, and coordinating care for mobile populations such as cross-border workers or travelers. In our work, the feasibility demonstration focused on the real-world applicability of the EEHRxF by conducting an in-depth, condition-specific analysis centered on diabetes. The review included a field-by-field mapping of the EEHRxF specification against locally available EHR data, covering field presence, cardinality constraints, semantic encoding, and value set alignment. Special attention was given to the use of standard terminologies (SNOMED CT and LOINC) and their actual implementation within national systems.

Hospital discharge reports are comprehensive clinical documents issued upon a patient’s release from hospital care. They provide a detailed record of diagnoses, interventions, procedures, medications administered, and follow-up recommendations. Standardizing discharge reports under the EEHRxF helps address longstanding fragmentation in how clinical data is documented and shared between hospitals and outpatient care providers. Due to the high level of complexity of such documents, the feasibility demonstration involved a comprehensive analysis of the EEHRxF data structure by conducting a systematic review of required data fields, their cardinality constraints, and the semantic coding systems specified for each data element. The exercise aimed to assess the alignment between EEHRxF specifications and the actual data availability and quality in representative national and regional EHR systems. By identifying structural gaps and mismatches in terminology use, the demonstration provided practical insights into the readiness of health data sources to support cross-border interoperability in line with EHDS objectives.

It should be noted that this nomenclature is derived from the priority categories defined in the “Commission Recommendation on a European Electronic Health Record exchange format (C(2019)800) of February 6, 2019” [4], which was the version available at the beginning of this work and is the one used in this paper. However, the EHDS regulation [1] has since changed the names of these priority categories, and the relationship between the 2 versions is presented in Figure 1.

The methodological approach developed to assess levels of practice and alignment with the EEHRxF is depicted in Figure 2. This 3-step strategy aims to guide the discussion with organizations and stakeholders in order to understand the maturity level, identify adoption obstacles, and determine future directions toward the EEHRxF.

The first step involves the selection of specific priority categories where the discussions and feasibility will focus according to the project objectives and involved organizations. The second step is the assessment methods, where the different participants are engaged in structured exercises to evaluate and discuss the adoption of the EEHRxF. Finally, the third step is the determination of assessment levels that translate into concrete metrics for the vision of the organizations involved in relation to the available guidelines. Throughout all steps, challenges, barriers, and adoption opportunities are always considered to determine the future directions on the EEHRxF adoption.

The involvement of different types of organizations related to and involved in the production, exchange, and consumption of health data was considered from the beginning of this work. The first group was formed with project partners and increased with the entry of new associated partners. In addition to these, through a series of workshops and online surveys, organizations external to the project were also engaged, and their contributions were collected. The organizations involved in these surveys participated on a voluntary basis; hence they were not preselected. We adopted an open participation approach where questionnaires were disseminated via project networks. As such, the resulting data reflects the insights of self-selected stakeholders who were interested in the topic and chose to contribute their professional expertise to this study. The types of organizations involved, and the respective countries, are shown in Figure 3.

Considering the analysis conducted with organizations and their interest and availability, the selected priority categories were the laboratory report and hospital discharge report, for which a feasibility assessment was conducted. In addition to these categories, and during the project’s progress, the patient summary was also considered as a relevant category for analysis. Medical images and reports and electronic prescription (ePrescription) and electronic dispensation (eDispensation) were not considered in this work.

The assessment methods used in this work aim to collect as much information as possible about the adoption of EEHRxF. Considering the different levels of maturity of the organizations involved, as well as the available resources, 3 different methods were considered:

To assess alignment and practice levels and achieve a common understanding regarding the EEHRxF, the eHealth Network (eHN) guidelines serve as the foundation for specifications in this work. The eHN guidelines on electronic exchange of health data aim to facilitate the cross-border interoperability of EHRs in the EU. They do so by supporting Member States in their efforts to ensure that citizens can securely access and exchange their health data wherever they are in the EU. The guidelines provide a list of technical requirements and enablers for implementation [23].

Taking into account the developments of the EEHRxF and the work underway on the technical specifications, the eHN guidelines are used as a basis for consensus regarding the data fields that should be part of each priority category. Within the specific scope of this framework, these 2 dimensions are operationalized as follows:

Furthermore, as future directions, additional fields considered important in each category are collected, as well as the challenges and barriers to the adoption of the EEHRxF in different organizations and possible ways to facilitate this adoption.

To collect more information and have a broader view regarding the possible adoption of the EEHRxF, 3 online surveys related to the assessment of practice for production and exchange of health information in 3 specific priority categories—laboratory report, patient summary, and hospital discharge report—are made available and disseminated through the project’s social media and website. While the surveys are available on the EU Survey platform (EC), they were distributed through different channels, including among partners from the XpanDH and xShare project consortia. The aim is to verify the data currently available in the systems compared to that requested in the available guidelines. Furthermore, the coding systems used and the exchange context considered are also asked.

The laboratory report survey has been available online since April 2024 [29]. Until the submission of this paper, 12 responses to this survey were collected, with another 25 having been started but not submitted or finished. Of the 12 responses submitted, 3 refer to health care providers, 2 to IT or EHR vendors, 4 to national authorities, and 3 to regional authorities. Regarding the countries involved, there are responses from the Netherlands (2), France, Cyprus, Italy (2), Portugal (2), Hungary, Ireland, Spain, and Denmark.

The patient summary survey has been available online since May 2024 [30]. Until the submission of this paper, 29 responses to the survey were initiated, 8 of which were completed. Out of the 8 responses completed, 1 refers to health care providers, 2 refer to IT or EHR vendors, 3 refer to national authorities, and 2 refer to regional authorities. Regarding the countries involved, the 8 responses include responses from Spain, Ireland, France, the United States, Cyprus (2 entities responded), Portugal, and Denmark. The inclusion of a US-based organization reflects an international IT vendor that actively develops and deploys solutions within the European market, making its regulatory readiness highly relevant to the EHDS framework.

The hospital discharge report survey has been available online since July 2024 [31]. Until the submission of this paper, 12 responses to the survey were initiated, 10 of which were completed. From the 10 responses completed, 2 refer to health care providers, 1 refers to IT or EHR vendors, 5 refer to national authorities, and 2 to regional authorities. Regarding the countries involved, there are responses from Spain, France, Cyprus, Italy, Portugal (2 entities responded), Greece, Ireland, Slovakia, and Denmark.

Please note that individual respondent identities and their specific institutional affiliations are kept anonymous to ensure participant privacy and compliance with data protection regulations.

This study involved the collection of primary data through online surveys and expert consultations, alongside the analysis of system-level “TO-BE” architectures. Because the research focused exclusively on institutional, technical, and professional opinions regarding the implementation of the EEHRxF, and did not collect private clinical data, patient identifiers, or sensitive medical histories, it was deemed exempt from formal medical ethics committee review in accordance with national regulations and institutional policies of the participating partners.

Before engaging in any evaluation activity, all survey respondents and workshop participants were explicitly informed about the study’s scope, objectives, and academic purpose. Participation was entirely voluntary and self-selected. Submission of the online questionnaires constituted formal implied consent for the aggregate use of the provided technical and professional insights.

To safeguard the privacy and confidentiality of all human subjects involved, all survey data were fully anonymized at the point of collection. No Personally Identifiable Information (PII) or IP addresses were recorded. Furthermore, individual institutional affiliations of the survey respondents have been kept confidential and presented strictly in an aggregated, nonattributable format.

The results presented in this section reflect the levels of practice and levels of alignment obtained through the online surveys. Importantly, the organizations involved in the execution of the feasibility demonstrators also formally completed these surveys, meaning their institutional data are directly integrated into the quantitative metrics. Additionally, a focused evaluation of the use of coding systems for laboratory tests in the follow-up of patients with diabetes is presented within the context of 2 feasibility demonstrators. These results illustrate how existing EHR systems manage interoperability, particularly using standards, and assess their readiness to align with the EEHRxF specification in real-world clinical workflows.

Table 2 summarizes the level of structured data implementation specifically for fields classified by respondents as relevant (ie, “important” or “very important”) across the three EEHRxF priority categories analyzed: patient summary, laboratory report, and hospital discharge report.

The results reveal a heterogeneous landscape of maturity. Patient summaries show the highest proportion of structured implementation, with 46% (317/682) of relevant fields being structured and coded, reflecting the advanced adoption of standardized formats such as HL7 FHIR and SNOMED CT. Laboratory reports follow, with 35% (204/591) of relevant fields structured, although 40% (237/591) remain unstructured. Hospital discharge reports display the lowest level of structured implementation (438/1549, 28%), with many institutions still relying on free-text documentation. These findings confirm that while conceptual alignment with the EEHRxF is strong, practical readiness varies substantially among data domains, highlighting the need for targeted investment in structuring core clinical and administrative information.

To further quantify the misalignment between strategic prioritization and actual technical readiness, Table 3 isolates the data fields strictly as “very important” and breaks down their current implementation status.

A total of 30 valid responses were analyzed: 8 for the patient summary, 12 for the laboratory report, and 10 for the hospital discharge report. Across the 30 valid responses, “very important” fields represent between 35% (902/2550; hospital discharge report) and 54% (427/784; patient summary) of the total data fields. However, a clear implementation gap is visible, since a substantial portion of these fields remain either unstructured (136/427, 32% to 445/902, 49%) or completely unavailable (47/347, 13% to 113/427, 26%) in current systems, underscoring the practical challenges in achieving full interoperability. Participation spanned 9 Member States and included health care providers, IT or EHR vendors, national authorities, and regional authorities, providing a representative snapshot of differing organizational perspectives on interoperability readiness.

In the context of the feasibility demonstrators focused on laboratory reports, an in-depth analysis was carried out on the use of LOINC codes for laboratory report representation in the follow-up of patients with diabetes. To ensure transparency, the methodology for this microanalysis followed a 3-step process:

The tests considered most important for follow-up of patients with diabetes, as indicated by both demonstrators and for which LOINC codes were extracted, include:

Figure 15 presents the results of the comparative analysis of LOINC code usage between the 2 organizations. The figure displays both the total number of codes used and the degree of overlap between Hungary (yellow) and Portugal (green) for the relevant test categories.

The first set of circles shows the total number of LOINC codes used by each demonstrator for diabetes-related laboratory tests. Hungary reports a total of 57 codes, while Portugal uses 26, with 14 codes shared between the two. This indicates a substantial disparity in coding granularity or implementation coverage, with Hungary adopting a broader coding set.

For liver function tests, both countries show considerable overlap, with 11 codes in common out of 15 used by Hungary and 15 by Portugal, suggesting good alignment in this category.

Regarding hemoglobin, which is a critical test for long-term glucose control in patients with diabetes, Hungary uses 2 codes while Portugal uses only 1, with no overlap. This divergence reflects differences in laboratory workflows, coding practices, or test variants.

In the insulin production (C-peptide) category, Hungary uses 9 LOINC codes, whereas Portugal uses 2, with only 2 codes overlapping. This again illustrates a more extensive use of specific coding in Hungary, offering richer data granularity but also raising questions about semantic alignment.

Overall, the analysis highlights important discrepancies in LOINC code implementation between the 2 countries, even when both recognize the same clinical tests as relevant. While partial overlaps exist, especially in liver function, the divergence in categories like hemoglobin and insulin production underscores the challenges of semantic interoperability. Addressing such differences is crucial to ensure consistent and meaningful data exchange under the EEHRxF, especially for chronic disease monitoring scenarios such as diabetes.

This study demonstrated a strong conceptual alignment with the EEHRxF vision across European health care institutions, alongside a significant gap between strategic importance and practical readiness. Our multimethod assessment highlighted that while a large portion (between 902/2550, 35% and 427/784, 54%) of the defined data fields are perceived as “very important” for clinical care, only a fraction of these (between 283/902, 32% and 178/427, 42%) are currently operationalized in a structured, interoperable format, with a widespread reliance on unstructured free text.

Reflecting these metrics, the alignment of existing health information systems with the EEHRxF reveals a complex and uneven landscape across Member States and health care organizations. While the feasibility demonstrators and supporting surveys conducted confirmed general support for the objectives of the EEHRxF, they also highlighted several critical areas where further harmonization is required to achieve meaningful and sustainable interoperability.

A core dimension of the alignment discussion concerns data structure and availability. Across all 3 priority categories, Patient Summary, Laboratory Report, and Hospital Discharge Report, participants reported varying degrees of adherence to the EEHRxF field requirements. While some data elements are routinely captured and structured, others remain absent or embedded in unstructured free text, making automated exchange and semantic interpretation challenging. For instance, in the case of patient summaries, demographic and medication data are generally well documented, whereas sections such as family medical history or self-reported observations remain underrepresented or inconsistently structured. Similarly, laboratory reports often capture core test results but lack standardized terminologies, and discharge reports tend to exhibit wide heterogeneity in their composition, depending on local workflows and IT capabilities.

Cardinality and optionality of data fields also emerged as an alignment issue. In several demonstrators, it became clear that while required fields were technically present in the systems, their actual population varied across institutions and use cases. This raises questions about the reliability and completeness of the exchanged records, especially in cross-border contexts where local conventions may differ significantly.

From a semantic interoperability standpoint, the discussion uncovered both progress and limitations. The use of internationally recognized code systems, such as SNOMED CT for clinical terms, LOINC for laboratory observations, and ICD for diagnoses, is often patchy. National extensions and proprietary codes are still in use, and in many cases, mappings to the EEHRxF reference terminologies are either incomplete or nonexistent. The feasibility demonstrators also indicated that even when international codes are adopted, the granularity of their application and consistency of use can vary substantially, affecting the interpretability of shared data.

Moreover, participants noted that key contextual metadata, such as data provenance, encryption status, and access permissions, are not always readily integrated into current documentation practices. These attributes are increasingly relevant in the context of secure and compliant data exchange under the EHDS.

Finally, alignment discussions revealed the tension between the standard’s evolving nature and real-world readiness. Several stakeholders expressed uncertainty regarding the maturity and stability of EEHRxF specifications, particularly in areas still under development. This uncertainty affects planning and investment decisions and may deter early adopters.

In the next sections, complete details on the findings are provided, in general and per category, including an identification of (possible) new data fields and of the technical barriers and interoperability challenges.

While previous studies and historical health IT deployments have extensively documented the technical complexities of cross-border data exchange, our findings corroborate and expand upon prior work by demonstrating that the most significant hurdles are often organizational and human-centric. Consistent with existing literature on large-scale health interoperability, the implementation and scaling of the EEHRxF face several sociotechnical challenges that hinder effective adoption across Member States, extending far beyond specific code or architectural barriers.

The implementation of the EEHRxF brings significant promise but also exposes several technical and semantic challenges that can hinder its successful adoption. These challenges stem from the heterogeneity of national health information systems, the uneven maturity of digital infrastructures, and the complexity of harmonizing data standards across diverse clinical, administrative, and regulatory environments. Table 4 summarizes the main technical barriers and interoperability challenges identified through the feasibility demonstrators, stakeholder workshops, and survey responses.

Our study presents some limitations that reflect the early-stage deployment of the EHDS framework. First, the sample sizes obtained through the online surveys (comprising 8 valid responses for patient summaries, 12 for laboratory reports, and 10 for hospital discharge reports) are limited. This small sample is directly tied to the novelty of the EEHRxF specifications, which are currently undergoing their initial phase of institutional adoption and awareness building across Member States. Second, a high concentration of responses is tied to the active partners of the XpanDH and xShare project consortia, including the organizations involved in the 5 feasibility demonstrators. This introduces a potential bias, as these organizations represent early adopters with a higher readiness profile than the broader European average. However, this sample is not exclusive to the projects, and capturing the operational realities of these frontrunner institutions is methodologically vital. To mitigate potential single-source biases and enrich the quantitative metrics, we systematically triangulated the survey outcomes with real-world gap analysis data from the 5 detailed feasibility demonstrators and several qualitative multistakeholder workshops, delivering a realistic evaluation of the current transitional phase of European health data interoperability.

Building on the lessons learned from the feasibility assessment, the recommendations presented in Table 6 distill the key insights into actionable guidance for stakeholders. These recommendations aim to address the systemic challenges identified, such as the instability of EEHRxF specifications, the need for stronger leadership and cocreation mechanisms, and the resource and human capacity constraints faced during implementation.

The feasibility demonstrators have provided concrete, practice-oriented evidence of the potential and challenges in adopting the EEHRxF across real-world health care settings in Europe. Focusing on 3 of the priority domains, laboratory reports, hospital discharge reports, and patient summaries, the demonstrators engaged a diverse set of stakeholders, including health care providers, national authorities, regional authorities, IT vendors, and patients, to assess the technical, organizational, and semantic alignment between existing systems and the EEHRxF specifications.

While based on a limited sample of early adopters, the results indicate a strong conceptual alignment with the EEHRxF vision and a clear recognition of its importance in enabling cross-border health data exchange. However, the demonstrators also revealed a number of barriers that must be addressed to support full-scale adoption. These include the fragmentation of data models, uneven maturity of coding practices (particularly around SNOMED CT, LOINC, and national extensions of ICD), frequent reliance on unstructured or proprietary data formats, and varying degrees of awareness and readiness across Member States and health care institutions.

Importantly, the feasibility assessments conducted did more than map gaps; they identified practical paths toward bridging them. Structured use cases, implementation scenarios, and system mapping exercises highlighted opportunities for transition strategies, from proprietary formats to EEHRxF-compliant structures, and underscored the value of phased adoption that accommodates both current capabilities and future goals. These prospective architectures and design scenarios illustrate potential benefits, particularly in chronic disease management and hospital discharge communication, demonstrating how structured, interoperable data could, upon full implementation, positively impact continuity of care, patient safety, and health system efficiency.

Moreover, the project’s methodology, combining feasibility demonstrators, collaborative workshops, and wide-reaching surveys, proved effective in building shared understanding, raising awareness, and strengthening cocreation across sectors. The structured assessment of practice and alignment provides a replicable framework that can inform ongoing national implementations and contribute to the broader EHDS agenda.

Ultimately, this exploratory work demonstrates that while the EEHRxF is ambitious in scope, its goals appear achievable with the right mix of leadership, investment, stakeholder engagement, and technical support. To move from feasibility to widespread implementation, future efforts must prioritize structured data, semantic interoperability, robust governance, and a strong foundation of trust across health care ecosystems. As the EHDS becomes a reality, the insights and tools here presented serve as valuable assets to accelerate adoption and ensure that the exchange of health data across Europe is secure, meaningful, and beneficial to all.