From recording health data to reusing it — and improving care.
Based on EIT Health (2024), chapters 4–5, printed pages 70–96 · September 2026
Open with a familiar situation. Introduce the whole story without jargon.
A fictional patient with diabetes. Her doctor uses her records to guide treatment — laboratory results, medication, complications.
Researchers could reuse data from many patients to investigate care. How can those findings improve care for patients like Sofia?
Teaching example invented for this presentation. Conceptual foundation: EIT Health (2024), ch. 5, pp. 84–89.
A common framework — not a single central database.
Common EU framework for electronic health data: access, control, exchange, and secure reuse.
Compatible EHR systems and cross-border platforms that help data flow while preserving meaning.
Controlled access via health data access bodies; builds on and complements the GDPR.
MyHealth@EU — cross-border infrastructure for accessing patient data for care, including across national borders.
HealthData@EU — interoperability platform providing services to support access to data for research, innovation and policymaking.
| Milestone | When | What it means |
|---|---|---|
| Entry into force | March 2025 | Regulation is legally adopted — not yet fully operational. |
| Major provisions | 2029 | Primary use obligations (e.g., cross-border access, EHR system requirements) begin to apply. |
| Secondary use | 2031 | HealthData@EU and secondary-use provisions come into effect. |
European Commission, EHDS overview and implementation timeline.
EIT Health (2024), introduction, pp. 14–17; European Commission, EHDS implementation timeline.
Distinguish the purpose — not the user's profession or the location of the data.
| Primary use | Secondary use | |
|---|---|---|
| Purpose | Care of the individual concerned | Research, innovation, public health or policymaking |
| Example | Sofia's clinician checks her results | Researchers study outcomes across many patients |
| Question | "What does this patient need?" | "What can we learn beyond this individual case?" |
Primary use: Processing of personal electronic health data for the provision of health services to assess, maintain or restore the state of health of the natural person to whom that data relates — including prescription, dispensation, and relevant social security or reimbursement services.
Secondary use: Processing of electronic health data for purposes other than the diagnosis, treatment and care of the patient — including research, innovation, public health, policymaking, and regulatory activities.
Health data access bodyA designated body set up by EU Member States to provide access to electronic health data to third parties for secondary use in a secure way, building on the Data Governance Act. — responsible for assessing and managing access to data for secondary use.
EIT Health (2024), ch. 5, pp. 84–89; EHDS Regulation definitions.
Data recorded for one purpose is not automatically suitable for every future purpose.
Glucose: 7.0 mmol/L
Fasting recorded
Glucose: 126 mg/dL
Fasting status missing
"Glucose elevated"
No numerical result
Invented example. A and B show approximately equivalent concentrations, but not necessarily comparable measurement contexts.
Nurses lack access to the national EHR; many work in unstructured data silos like basic spreadsheets. Suboptimal reporting in national registries (SARI, cancer, death statistics).
Ranks low on integrated care among OECD countries. Paper documentation in nursing and home care keeps entire categories of health data inaccessible for research.
Ireland's reliance on paper records made data quality a low priority. In Spain, only a small percentage of EMR input fields are actually filled out in routine practice; the majority of data is entered as free text.
EIT Health (2024), ch. 4, pp. 74–77: fragmentation, standardisation, costs and underrepresentation.
Keep the focus on why quality matters.
| Requirement | Question to ask |
|---|---|
| Shared structure and meaning | Can we combine and interpret it? |
| Quality suited to the question | Can it answer this question reliably? |
| Metadata and traceability | Do we understand its origin and context? |
| Population coverage | Who is represented — and who is missing? |
Findable — data and metadata are discoverable through standardised identifiers.
Accessible — data can be accessed under defined conditions and terms.
Interoperable — data uses common languages and formats across systems.
R — Reusable Data is well-described so it can be repurposed for new research.
The EHDS legislation builds on the FAIR principles and the concept of data equity to define EU-wide standards for data quality and data utility.
MetadataInformation describing a dataset, including variables, units, collection methods and coverage. — describes a dataset so users can judge its suitability.
TraceabilityThe ability to follow data's origins and processing history across primary and secondary use systems. — follows data's origins and processing history.
Data ontologyA standardised vocabulary that enables sharing of information between disparate systems within the same domain — e.g., terms for medical specialties or diseases. — standardises vocabulary across systems.
TEHDASJoint Action Towards the European Health Data Space — an EU-funded project involving 25 countries that developed a data quality framework for secondary use. — developed a harmonised quality framework enabling trustworthy secondary use.
EIT Health (2024), ch. 4, pp. 70–73 and 78–83.
Make reuse part of good clinical workflows. Care delivery remains the priority.
| Challenge | Direction proposed in the report |
|---|---|
| Extra recording burden | User-friendly structured entry; avoid duplicate entry |
| Fragmented information | Shared standards and appropriate data-transfer infrastructure |
| Limited staff and resources | Training, technical support and funding |
| Little immediate benefit | Return useful insights and demonstrate value for care |
Novel technologies — well-designed EHR forms aligned with clinical practice, entering information once where possible, and carefully implemented automatic capture from devices — can reduce burden while improving data quality.
Natural language processing can expedite the standardisation of legacy free-text data in EHRs — decades of historical records that would otherwise remain inaccessible for secondary use.
Value-based healthcare (VBHC) pays for results rather than services, incentivising keeping people healthy over treating the sick. This aligns incentives with the predictive and preventive models of care that the EHDS is expected to enable. Adoption of VBHC could be a powerful enabler for the digital health innovation the EHDS makes possible.
EIT Health (2024), ch. 5, pp. 84–91; supporting context in ch. 4, pp. 76 and 80–82.
Data access is a step — not the final outcome.
A standardised path is needed for secondary users to recontact patients via their healthcare teams — to request complementary information or report incidental findings. The finding should be reported in a timely manner, defining how, where, and to whom the information on patients must be provided.
Healthcare systems need structured paths to adopt data-driven innovation. Examples: Catalonia's BIOCAT fast-track for digital health technologies; Austria's "digital health pathway" for citizens to manage health data. Value-based healthcare models pay for results, incentivising prevention.
EIT Health (2024), ch. 5, pp. 85–87 and 94–96.
A high-level assessment: one strength and one limitation are enough.
The report draws on expert roundtables and interviews across 10 countries (Poland, Hungary, Italy, Ireland, Portugal, Spain, Sweden, Austria, Germany, BeNeLux, France) during 2022–2023. This is useful for identifying barriers and potential solutions, but it does not establish which proposed solutions will work best or how much they will cost.
The legal discussion should be read historically: it reflects consultations before the final Regulation entered into force in March 2025.
EIT Health (2024), methodological context pp. 15–17; implementation arguments in chapters 4–5.
Name the three Canvas learning goals explicitly.
Explain its purpose and how it supports health-data use and reuse.
Distinguish them by purpose and give an example of each.
Explain why access alone is insufficient for reliable and useful reuse.
Learning goals transcribed from the supplied Canvas screenshot.
Show two questions. Discuss one if time is short.
Would you prefer a very large hospital dataset or a smaller dataset that better represents your target population? What would you need to know before deciding?
Should clinicians record additional information mainly for future research? Under what conditions?
Discussion prompts developed from the implementation tensions in chapters 4–5.
Short, accurate answers for discussion and rehearsal.
No. It is a common framework involving rules, institutions and infrastructures for access, exchange and reuse. One central database is a misleading description.
No. It also includes purposes such as public health, policymaking, innovation and regulatory activities, within the applicable rules.
No. If the purpose is treating the individual, it remains primary use.
No. Standardised records can still contain errors, missing information or unrepresentative populations.
Yes. Suitability depends on the question, the limitations and whether they can be handled appropriately. Imperfect does not mean worthless.
Cleaning helps, but cannot reliably recover information never recorded or automatically correct missing populations. It also costs time and resources.
No. Pseudonymisation replaces identifiers and keeps the information needed for attribution separately. Pseudonymised data remains personal data; anonymisation is different.
No. The EHDS builds on and complements existing data-protection law.
No. Findings need appropriate evaluation, adoption and monitoring. That is why closing the loop is central.
No. It entered into force in March 2025, but major provisions apply progressively, notably in 2029 and 2031.
Do not just list terms. Use this as a practice question, not a prediction of the actual exam.
Primary use involves using health data to provide care to the individual, while secondary use includes purposes such as research and policymaking. Data recorded during care can support secondary use, but it was not necessarily collected with those later purposes in mind.
Differences in coding, units, completeness and population coverage can make records difficult to combine or produce misleading findings. Interoperability helps systems exchange and interpret information, while metadata and traceability help users understand its context and limitations. Data quality must therefore be assessed in relation to the intended question.
Reusable data should be supported through appropriate recording practices, standards and resources without unnecessarily burdening clinical care. For secondary use to improve healthcare, its findings must then be evaluated and incorporated into practice, completing the feedback loop.
Questions & discussion
EIT Health Think Tank (April 2024), Implementing the European Health Data Space Across Europe, chapters 4–5, printed pp. 70–96.
European Commission, EHDS Regulation: overview and implementation timeline. Consulted September 2026.
This guide is educational, not legal advice. The diagrams and fictional patient example are teaching aids, not report findings.